Previous Page
OCCURRENCE OF ORGANOCHLORINE PESTICIDES IN VEGETABLES GROWN ON UNTREATED SOILS
OCCURRENCE OF ORGANOCHLORINE
PESTICIDES IN VEGETABLES GROWN ON
UNTREATED SOILS FROM AN AGRICULTURAL
WATERSHED
MARIANA GONZALEZ
KARINA S.B. MIGLIORANZA
Universidad Nacional de Mar
del Plata
Mar del Plata, Argentina
and
Consejo Nacional de
Investigaciones Cientı́ficas y
Técnicas (CONICET)
Buenos Aires, Argentina
JULIA E. AIZPÚN DE MORENO
VÍCTOR J. MORENO
Universidad Nacional de Mar
del Plata
Mar del Plata, Argentina
INTRODUCTION
The role of agrochemicals in modern agriculture is
continuously evolving, and their contribution to crop
protection continues to increase. At a global level, the
use of pest control agents (i.e., pesticides) has proved to
assist the solving of many problems facing human health
and food production. Organochlorine pesticides (OCPs)
are a class of compounds of particular concern in the
environment because of their recalcitrance in natural
solids, global transport, distribution, and toxicity, which
leads to include them into the group of persistent organic
pollutants (1). This group had been widely used since early
1940, with an intensified usage during the Second World
War. At that time, crops were duplicated and malaria
incidents were reduced in India from 75 millions to less
than 5 millions persons affected in 10 years as the result
of DDT usage. However, such usage has occasionally been
accompanied with hazards to man and the environment.
Target species became resistant, leading to the application
of higher doses of pesticides. For these reasons, most
developed countries banned or restricted their usage since
1970. However, in the poorer countries, OCP restrictions or
prohibitions date from the later 1990s. In addition, many
of the banned or withdrawn pesticides from the developed
market are still produced and sold in developing country
market (2). Pesticides such as DDT are still in use in India
and Brazil with public sanity purposes (3–5).
When pesticide are agriculturally applied, roughly
85–90% of pesticides never reach the target organism, but
disperse through the air, soil, and water (6,7). Pesticides
that are bound in soils or taken up by plants and animals
are drawn into rivers and lakes and move into the
aquatic food chains inducing severe damage to aquatic
life (8,9). The environmental behavior of OCPs is largely
related to their physical and chemical characteristics.
Some of them are volatile enough to evaporate, reaching
the upper atmosphere and being carried hundreds or
643
thousands of kilometers by a process known as ‘‘Global
Distillation’’ (1,10,11). Thus, the capability of these
pollutants for long-distance transfer, precipitation, and
accumulation in environmental materials results in the
formation of background concentrations in remote regions
from the emission sources (12). Therefore, OCPs can be
found in many places where they had never been applied.
They have been linked to carcinogenicity and endocrine
disruption in mammals, and concerns over toxicity are
exacerbated by their hydrophobicity, which results in
bioaccumulation in fatty tissues and biomagnification
through food chains (13).
The general population can be exposed to low levels of
pesticides in three general ways:
1. vector control for public health and other nonagri-
cultural purposes
2. environmental residues
3. food residues
Regarding food residues, consumption of plants or plant
products, such as vegetables, fruits, and grains, form a
major part of the food consumption of human beings
and cattle (14). Moreover, vegetables are an essential
component of our diet because of their nutritional
value (15).
The impact of soil contaminants on human health and
accumulation of pesticides in food crops is a great concern
because of increased dietary exposure of consumers (16).
Considering that vegetation is the link among the
atmosphere, soil, and human food supply, contamination
of plants will have a great influence on the total daily
intake of a substance (14). Thus, it is important to
understand those processes by which pollutants enter into
this environmental compartment.
ABSORPTION AND ACCUMULATION OF OCPs IN PLANTS
Plants may accumulate OCPs via different pathways:
1. adsorption to the root surface
2. root uptake and transport to the upper plant tissues
3. uptake of airborne vapors by aerial plant parts (stem,
leaves, and fruits)
The relative significance of each route by which OCPs
can be taken up into fruits and vegetables is likely to be
influenced by a wide range of factors (i.e., physicochemical
characteristic of the pesticide, environmental conditions,
and specie-specific characteristics) (17,18) (Fig. 1).
Air to Plant Route
OCPs are emitted to the atmosphere via various routes: by
transfer to the airflow during the treatment of agricultural
crops, evaporation from treated surfaces of plants and
soils, and finally by wind erosion (12).
Simonich and Hites (19) presented a simplified mech-
anism of pollutants absorption by plants from air. Air
to vegetation transfer of gaseous pollutants will be on
the partition coefficient of the pollutant in the octanol
644 OCCURRENCE OF ORGANOCHLORINE PESTICIDES IN VEGETABLES GROWN ON UNTREATED SOILS
Figure 1. Pathways for accumulation of Organochlorine Pesti-
cides by plants. Modified of Lovett et al. (18).
air system, the surface area, and lipid concentration in
plant tissues.
Plant cuticles are the first barrier in the absorption of
airborne and fall out OCPs in all aboveground organs and
parts of higher terrestrial plants (fruit, flowers, stems,
and leaves). Moreover, gaseous contaminants can directly
reach the intercellular aerial medium through the open
stomata (20–22). It was found that OCP residues in plant
tissues decrease very slowly because of their low solubility
in water and high solubility in oils. Thus, sorption of
pesticides by plant surfaces depends significantly on
the morphological features of the leaf, velocity, waxing,
etc. (12,23). Moreover, vapor transfers of OCPs from soil
to the aboveground parts of plants would be expected
to occur to varying extents for all crops. For vegetables
such as lettuce, courggetes, or zucchinis, which develop
in contact with or in close proximity to the ground, it
is possible that uptake of OCPs could also occur by direct
adsorption from the soil and by splash-over of soil particles
during heavy rain (12,23,24).
Soil to Plant Route
The transition of OCPs from soils into plants (35–70%) will
be, in general, higher than into adjacent environments:
water (2–18%) and atmosphere (18%) (12).
The uptake of OCPs by plant roots is affected by the
bioavailability of the pesticide in the soil, leading to a
competition between the organic fraction of the soil and
the root material. Higher soil adsorption will result in a
lower plant bioavailability (25).
After root uptake, plants can translocate pesticides
via xylem to the aboveground plant parts. The extent
of such mobilization will be on pesticide solubility and
plant characteristics, such as root exudates (26). Thus,
pesticides whose kow are higher than 4 will not be
able to translocate from root to aerial tissues in most
plant species (17). However, an exception has been
reported for vegetables from the cucumis and cucurbitacea
genera, which are able to translocate highly hydrophobic
compounds, such as the insecticide DDT, PCBs, and
dioxinas, from roots to fruits (26–29).
OCPs OCCURRENCE IN VEGETABLES
As was previously mentioned, vegetables are essential
components of the human diets. As pesticide residues
remain on fruits and vegetables, it is important to know
the magnitude of such contamination from the point of
view of safety to consumers. Thus, a growing concern
for safer foods has led research into increased pesticides
residues monitoring.
In order to regulate and control such residues, the con-
cept of maximum residues levels (MRL) was established.
This level represents the maximum concentration of the
residue (expressed in mg/Kg) that is legally permitted in
specific food items and animal feeds (31).
OCP residues in vegetables are included in the Food
and Drug Administration (FDA) monitoring program
because they are still present in the environment (32).
A residual content of these compounds in foods today is
still quantifiable, and their avoidance in farming is largely
impossible (33).
Regular surveys and monitoring programs of OCP
residues have been carried out for decades in devel-
oped countries, such as the United States (32,34), the
United Kingdom (35), and Spain (36). Although in some
developing countries, such as India (37), Tunisia (38),
Nigeria (39), Rumania (40), Brazil, the Philippines (41),
Egypt (42–44), Honduras (45), and Argentina (46), reports
about OCP levels in vegetables were made, they were not
made as a part of governmental monitoring programs.
Most of the reports showed that OCPs, such as HCHs,
DDTs, Aldrins, Chlordane, and Heptachlor, were present
in all analyzed samples. In the United States, about 1%
of the samples violated the established MRL (32), but
in developing countries, such as Brazil, this value can
increase up to 3.7% (41). The presence of OCPs in veg-
etable samples is the consequence of either recent and
current use in developing countries and/or the result of
their persistent residues 30 years after their prohibition.
One alternative to synthetic pesticide use is organic
farming. Thus, two decades ago, growing environmental
awareness and concerns about safer foods have lead
people to question modern agricultural practices, which
has been reflected in an increasing demand for organic
production, which is perceived as less damaging to the
environment and to be healthier than conventionally
grown foods (47,48).
ORGANIC AGRICULTURE: WORLD WIDE SITUATION
Organic agriculture has developed rapidly during the
last decade, and governments have encouraged the
 OCCURRENCE OF ORGANOCHLORINE PESTICIDES IN VEGETABLES GROWN ON UNTREATED SOILS
development of this tillage practice, mainly in developed
countries. Chemophobia is the commonest reason for the
public to choose organic food on the assumption that such
food is free of synthetic pesticides (49).
Organic farms distinguish from all other forms of
farming, namely conventional, by a rejection of soluble
minerals as fertilizer and synthetic pesticides in favor
of natural ones (49). However, organic food contains
pesticide traces, although the amounts are lower than
in conventional produce (50,51). Although it is likely that
organically grown foods are lower in pesticide residues,
very little documentation of residues levels exists (52).
Moore et al. (33) studied the residue of eight OCPs in two
matrices (carrots and winter squash) used to elaborate
conventional and organic baby food brands. They found no
detectable residues among all types of food. These results
were the consequence of processing fruits and vegetables
(washing, peeling, and cooking) that largely eliminates
residues, as was largely demonstrated (53–55).
Today, developing countries represent a source of a
wide range of organic produce, especially tea, coffee, cocoa,
herbs, and spices, as well as tropical and subtropical fruits
and vegetables for Europe and the United States (56).
Vegetable diversity produced in Argentina reaches 115
species with field crops covering a surface area of some
600,000 hectares (57), and about 160,000 hectares are
destined to organic production. The European Union,
EE.UU, and Switzerland are the main markets of destiny
for these products (58). Therefore, the study of the factors
that may impact in this kind of production is relevant.
Most organic farms are supposed to remain small and
self-sufficient, whereas conventional farms are increasing
in size as the result of economic pressures.
The environmental fate of OCPs in agricultural
environments of Argentina have been studied, including
the impact of these pollutants in watersheds characterized
by rural and small-case farming land use (59–62). OCPs
were found in vegetables from organic and conventional
farming (46,52,63,64). The presence of these contaminants
in the organic farm was an interesting finding.
OCPs IN VEGETABLES ORGANICALLY CULTIVATED
Previous to the settlement of the organic farm, the
area was characterized by natural vegetation. The farm
is surrounded by important hedgerows and trees, and
neighboring fields on both sides are conventionally farmed
since its establishment about 30 years ago. Agrochemical
use was a common practice in surroundings fields, and
unsprayed field margins were not established.
From the organic farm, several vegetable species were
studied in order to estimate the OCPs levels in their
tissues. The produces include leafy vegetables (lettuce
and chard), tomato, and leek. Most of the studied crops
showed similar OCP levels between root and aboveground
plan parts (stem or leaves). This observation was not
consistent with previously published data where plants
were cultivated in treated soils (28,30,65). Thus, in the
organically treated soils, the air to plant route would be
as important as the soil to plant route in all species.
645
The main OCPs found in vegetables were the same
that were previously reported for conventional produce:
HCHs, DDTs, and Chlordane being these pesticides
forbidden for agricultural purposes. Moreover, high levels
of endosulfans (α- and β-isomers and sulfate metabolite)
were also found as a consequence of their current
usage. All species accumulated OCPs efficiently from
soil in their roots, leading to high root bioconcentration
factors. Common practices in organic tillage constitute the
enrichment of soil with vegetal rests. In this farm, pine
needles collected form surrounding forests are usually
added to the soil at the beginning of the growing periods.
Pine needles, rich in waxes, are known to accumulate
highly lipophilic pollutants from the atmosphere (66,67).
Therefore, those soils enriched with this amendment
showed the highest level of OCPs. Moreover, plants
growing in such soils showed enrichment in the more
lipophilic compounds in their roots.
Aerial tissues showed a great variation in accumulation
regarding species and tissues. In lettuce, differences
between inner and outer leaves were found, with the lowest
concentrations in the outer leaves despite the higher lipid
content, maybe by the growth dilution effect or weathering.
Moreover, plant morphology could be an important factor
affecting OCPs uptake and distribution (23,24). Thus,
lettuce with outer leaves enlarging inner leaves had
higher OCP levels in stem and inner leaves, probably as
a consequence of the accumulation of pollutants between
outer and inner leaves. On the other hand, chard plants
with an inverse wide-high relation and longer stem showed
higher OCP levels in outer leaves (51).
Tomato fruits contained OCPs not only in their peel
but also in the inner flesh tissues, which indicate
absorption rather than adsorption (63). Total OCPs in
flesh were 3-fold greater than in peel, probably as a
consequence of either continuous uptake by aerial tissues
(fruit peel, leaves, and stem) of airborne OCPs, followed
by translocation via phloem throughout the plant and
accumulation in fruit flesh or removal of pesticides
from peel by weather influences as photodegradation,
volatilization, and so on (17).
Acaricide Dicofol is still in use in this country and has
been reported to contain a significant amount of DDT as
manufacturing impurity. Thus, it represents a potential
continuous source of DDT in the studied area. DDT breaks
down to DDE and DDD in soil and generally with time
the parent/metabolite ratio decrease (68). DDT/DDE ratios
found > 1 suggest that besides Dicofol use, new DDT can be
continuously transported to the atmosphere of the organic
farm. The presence of this new DDT can be attributed
to illegal use or to atmospheric transport from those
countries, mostly tropical, where DDT is still in use (61).
Moreover, data suggest that long-range transport could
be the predominant source of chlordane in the farm. The
α-/γ -HCH ratio can be used as diagnostic of HCH sources,
inputs of γ -HCH (lindane) decrease the ratio found in air
and vegetation containing a background of technical HCH
(α-/γ - = 4–6) (68). The low α-/γ -HCH ratio found in all
samples reflect recent inputs of lindane to the atmosphere
of the studied area.
646 OCCURRENCE OF ORGANOCHLORINE PESTICIDES IN VEGETABLES GROWN ON UNTREATED SOILS
Endosulfan is one of the few organochlorine insecticides
still in use; it is primarily used on summer crops such as
tomatoes in the agricultural belt on the northwest area
of the studied watershed. Occurrence of endosulfan in
aerial tissues of tomato agrees with their summer usage
in the surrounding farms, showing the relevance of drift
pesticide during application events. Research has reported
that drift of endosulfans from aerial agricultural spraying
can produce lethal concentration to fish in shallow exposed
water bodies 200 m away from the target spray area (8).
In this watershed, it is also common during summer
months that a massive ‘‘fish kill’’ can occur. One possible
explanation could be that lethal concentration inputs of
endosulfans applied in the surrounding area that reach the
waterbody. Moreover, surface runoff from the organic farm
may constitute a source of endosulfans to the waterbody.
As a whole, it was demonstrated that plants bioaccu-
mulated OCPs efficiently from both routes, air and soil. In
organic tillage, it is common that crop residues are rein-
corporated into soil after harvest. Thus, soil OCP values
may reflect contributions from decomposed plant mate-
rial rather than changes either within soil or via the
atmospheric deposition. Therefore, plant residues consti-
tute a way of entry of atmospheric pollutants into of the
organic farm, leading to the subsequent availability for
root uptake. This fact will be relevant for plants whose
edible parts grow below ground (i.e., carrots, potatoes) or
those that are able to translocate them. Despite OCPs
being detected in all edible tissue (leaves, stem, and fruit
in the studied vegetables), the levels were below the max-
imum residue limits (MRL) considered by both Código
Alimentario Argentino (CAA) (69) and the Codex Alimen-
tarius (CA) (31).
Despite the practice of organic farming management,
OCPs were present in the soil environment and vegetable
tissues confirming the importance of contamination from
adjacent conventional farms. Therefore, the occurrence of
OCP residues in the organic samples may be attributed
to ecosystem factors such as pesticide runoff, drift from
pesticide used by neighboring farming, volatilization,
leaching, and washout. Thus, when organically grown
vegetables are subject to such potential sources of
contamination, the meaning of the term ‘‘organic’’ becomes
obscure and is difficult to interpret especially from the
perspectives of the public consumers.
BIBLIOGRAPHY
1. Wania, F. and Mackay, D. (1996). Environ. Sci. Technol.
30(9): 390–396.
2. Kannan, K., Tanabe, S., Ramesh, A., Subramanian, A., and
Tatsukawa, R. (1992). J. Agric. Food. Chem. 40: 518–525.
3. Minelli, E.V. and Ribeiro, M.L. (1996). Bull. Environ. Con-
tam. Toxicol. 57: 691–696.
4. Mathiessen, P. (1985). Environ. Pollut. 10: 189–211.
5. Ramesh, A., Tanabe, S., Murse, H., Subramanian, A.N., and
Tatsukawa, R. (1991). Environ. Pollut. 74: 293–307.
6. Moses, M. et al. (1993). Toxicol. Ind. Health. 9(5): 913–959.
7. Van der Werf, H.M.G. (1996). Agric. Ecosys. Environ. 60:
81–96.
8. Berril, M., Coulson, D., McGillivray, L., and Pauli, B. (1998).
Environ. Toxicol. Chem. 17: 1738–1744.
9. Hapeman, C.J., Dionigi, C.P., Zimba, P.V., and McConnel,
L.L. (2002). J. Agric. Food Chem. 50: 4382–4384.
10. Iwata, H., Tanabe, S., Sakai, N., and Tatsukawa, R. (1993).
Environ. Sci. Technol. 27(6): 1080–1098.
11. Iwata, H., Tanabe, S., Ueda, K., and Tatsukawa, R. (1995).
Environ. Sci. Technol. 29(3): 792–801.
12. Galiulin, R.V., Bashkin, V.N., and Galiulina, R.A. (2002).
Water Air Soil Poll. 137: 179–191.
13. Patlak, M. (1996). Environ. Sci. Technol. 30: 540–544.
14. McLachlan, M.S. (1996). Environ. Sci. Technol. 30: 252–259.
15. Kumari, B., Madan, V.K., Kumar, R., and Kathpal, T.S.
(2002). Env. Monitor. Asses. 74: 263–270.
16. Torres, C.M., Picó, Y., and Mañes, J. (1996). J. Cromatogr. A.
754: 301–331.
17. Trapp, S. and McFarlane, C. (1995). Plant Contamination:
Modeling and Simulation of Organic Chemical Processes.
Lewis Plublishers, Boca Raton, FL, p. 254.
18. Lovett, A.A., Foxall, C.D., Creaser, C.S., and Chewe, D.
(1997). Chemosphere. 34: 1421–1436.
19. Simonich, S.T. and Hites, R.A. (1995). Env. Sci. Technol.
29(12): 2905–2914.
20. Schreiber, L. and Schonherr, J. (1992). Environ. Sci. Technol.
26(1): 153–159.
21. Riederer, M. (1990). Environ. Sci. Technol. 24(6): 829–836.
22. Bakker, M.I., Bass, W.J., Sijm, D.T.M.M., and Kolloffel, C.
(1998). Phytochemistry. 47(8): 1489–1993.
23. Ripley, B.D., Ritcey, G.M., Ronald Harris, C., Denomme,
M.A., and Lissemore, L.I. (2003). J. Agric. Food Chem. 51:
1328–1335.
24. Trapp, S. and Matthies, M. (1995). Environ. Sci. Technol.
29(9): 2333–2338.
25. Ryan, J.A., Bell, R.M., Damidson, J.M., and O’Connor, G.A.
(1988). Chemosphere. 17(12): 2299–2323.
26. Campanella, B. and Paul, R. (2000). Int. J. Phytoremediation.
2: 145–158.
27. Hülster, A., Müller, J.F., and Marschner, H. (1994). Environ.
Sci. Technol. 28: 1110–1115.
28. Müller, J.F., Hülster, A., Päpke, O., Ball, M., and Marschner,
H. (1994). Chemosphere: 2175–2181.
29. White, J.C. (2000). Inter. J. Phytoremediation. 2: 133–144.
30. Kumari, B. et al. (2003). Env. Monit. Asses. 87: 311–318.
31. Codex Alimentarius. (2001). FAO/WHO Codex Alimentar-
ius Commission, Maximum Residue Limits for Pesticides.
FAO/WHO, Rome, Italy.
32. Food and Drug Administration. (1997). Pesticide program-
residue monitoring. Available at http://www.cfsan.fda.gov.
33. Moore, V.K., Zakib, M.E., and Zakib, M.J. (2000). Food Chem.
71: 443–447.
34. Luke, M.A., Masumoto, H.T., Cairns, T., and Hundley, H.K.
(1988). J. Assoc. Off. Anal. Chem. 71: 415–433.
35. MAFF. (1989). Food Surveillance Paper 25. HMSO, London,
p. 71.
36. Chavarri, M.J., Herrera, A., and Ariño, A. (2004). J. Sci. Food
Agric. 84: 1253–1259.
37. Kaphalia, B.S., Takroo, R., Mehtrota, S., Nigam, U., and
Seth, T.D. (1990). J. Assoc. Off Anal. Chem. 73: 509–512.
38. Driss, M.R. and Boquera, M.L. (1987). Acta Biol. Hungarica.
38: 101–106.

39. Adeyeye, A. and Osibanjo, O. (1999). The Sci. Total Environ.
23: 227–233.
40. Hura, C., Leanca, M., Rusu, L., and Hura, B.A. (1999).
Toxicol. Lett. 107: 103–107.
41. Mansour, S.A. (2004). Toxicol. 198: 91–115.
42. Abou-Arab, A.A.K., Solivan, K.M., El-Tantawy, M.E., Ismail,
B.R., and Naguib, K.H. (1999). Food Chem. 67: 357–363.
43. Abou-Arab, A.A.K., Solivan, K.M., and Naguib, K.H. (1998).
Bull. Nutr. Inst. 18(1): 117–137.
44. Abou-Arab, A.A.K. (1999). Food Chem. 65: 509–514.
45. Kammerbauer, J. and Moncada, J. (1998). Environ. Poll.:
171–181.
46. Miglioranza, K.S.B., Aizpún de Moreno, J.E., Moreno, V.J.,
Osterrieth, M.L., and Escalante, A.H. (1999). Environ. Poll.
105: 91–99.
47. Schifferstein, H.N.J. and Ouete Ophuis, P.A.M. (1998). Food
Qual. Pref. 9(3): 119–133.
48. Williams, P.R.D. and Hammit, J.K. (2001). Risk Analysis.
21(2): 319–330.
49. Trewavas, A. (2004). Crop Prot. 23: 757–781.
50. Baker, B.P., Benbrook, C.M., Groth, E., III, and Lutz Ben-
brook, K. (2002). Food Add. Contam. 19(5): 427–446.
51. Gonzalez, M., Miglioranza, K.S.B., Aizpún de Moreno, J.E.,
and Moreno, V.J. (2004). J. Food Chem. Toxicol. In press.
52. Bourn, D. and Prescott, J. (2002). Cri. Rev. Food Sci. Nutr.
42(1): 1–34.
53. Schattenberg, H.J., III, Geno, P.W., Hsu, J.P., Fry, W.G., and
Parker, R.P. (1996). J. AOAC Int. 79: 1447–1453.
54. Krol, W.J., Arsenault, T.R., Pylpiw, H.M., Jr., and Incorvia
Mattina, M.J. (2000). J. Agric. Food Chem. 48: 4666–4670.
55. Ramesh, A. and Vijayalakshmi, A. (2002). Pest Manag. Sci.
58: 1048–1054.
56. Barret, H.R., Browne, A.W., Harris, P.J.C., and Cadoret, K.
(2002). Food Policy. 27: 301–318.
57. Ministerio de la Producción. (2003). Secretaria de Agri-
cultura Ganaderı́a, Pesca y Alimentos de la República
Argentina. Available at http://www.sagpya.mecon.gov.ar/0-
0/index/agricultura/index agricultura.htm.
58. SENASA. (2003). Dir. Nac. de Fisc. Agroal. Dir. Cal. Agroal.
Coord. Prod. Ecól. ‘‘Situación de la Producción Orgánica en la
Argentina durante el año 2003’’, Bs. As. Argentina.
59. Miglioranza, K.S.B. et al. (2002). Environ. Sci. Poll. Res. 9:
250–256.
60. Miglioranza, K.S.B., Aizpún de Moreno, J.E., and Moreno,
V.J. (2003). J. Soil Sed. 4: 264–265.
61. Miglioranza, K.S.B., Aizpún de Moreno, J.E., and Moreno,
V.J. (2003). Environ. Toxicol. Chem. 22: 712–717.
62. Miglioranza, K.S.B., Aizpún de Moreno, J.E., and Moreno,
V.J. (2004). Water Res. 38: 1765–1772.
63. Gonzalez, M., Miglioranza, K.S.B., Aizpún de Moreno, J.E.,
and Moreno, V.J. (2003). J. Agric. Food Chem. 51: 1353–1359.
64. Gonzalez, M., Miglioranza, K.S.B., Aizpún de Moreno, J.E.,
and Moreno, V.J. (2003). J. Agric. Food Chem. 51: 5024–5029.
65. Mattina, M.J.I., Ianucci-Berger, W., and Dykas, L. (2000). J.
Agric. Food. Chem. 48: 1909–1915.
66. Krauthacker, B., Romanic, S.M., and Reiser, E. (2001). Bull.
Environ. Contam. Toxicol. 66: 334–341.
67. Hank, M., Umlauf, G., and Melchlam, M.S. (1994). Environ.
Sci. Technol. 32: 1546–1543.
68. Bidleman, J.F. and Falconer, R.L. (1999). Environ. Sci.
Technol: 206a–209a.
PESTICIDE CHEMISTRY IN THE ENVIRONMENT 647
69. CAA. (1969). Código Alimentario Argentino: Reglamento
Alimentario, Normas Sobre Higiene. Ley Nacional 18.284.
PESTICIDE CHEMISTRY IN THE ENVIRONMENT
KARINA S.B. MIGLIORANZA
MARIANA GONZALEZ
Universidad Nacional de Mar
del Plata
Mar del Plata, Argentina
and
Consejo Nacional de
Investigaciones Científicas y
Técnicas (CONICET)
Buenos Aires, Argentina
JULIA E. AIZPÚN DE MORENO
VÍCTOR J. MORENO
Universidad Nacional de Mar
del Plata
Mar del Plata, Argentina
INTRODUCTION
Pesticides have greatly enhanced the production and
quality of food, feed, and fiber, as well as the control
of disease vectors and pests adversely affecting the health
and welfare of the world. Specifically, pesticides have
been used in: (a) agriculture to increase productivity and
quality and quantity of food, feed, and fiber; (b) forestry for
pest control in silviculture; (c) the industrial, commercial,
municipal, and military sector for rodent and weed/brush
control around plant sites and in right-of-ways; (d) medical
vector control; and (e) urban environment for termite
control around structures and pest control in gardens (1).
As a consequence, of all the environmental contaminants,
pesticides have probably been the most widely criticized
because of their direct use in natural systems. The nature
of pesticide usage often requires broadcast distribution
over large areas of crops. This wide treatment of the
crop environment often results in treatment of adjacent
areas as well. The use of pesticides is indispensable
for conventional labor and extensive farming systems.
Although the use of pesticides has resulted in increased
crop production and other benefits, it has raised concerns
about potential adverse effects on the environment and
human health.
ENVIRONMENTAL BEHAVIOR OF PESTICIDES
The proportion of applied pesticide reaching the target pest
has been found to be less than 0.3%, thus leaving over 99%
elsewhere in the environment (2). Environmental entry
and transport processes can occur either between or during
rainfall events, including wet and dry deposition, foliar
washoff, and spray drift. Moreover, pesticide dissolved in
atmospheric water vapor may be deposited either on the
soil surface directly or indirectly to the foliar surface, with
subsequent removal during rainfall events via the process
of foliar washoff. Once in the soil, pesticides may volatilize
648 PESTICIDE CHEMISTRY IN THE ENVIRONMENT
into the air, enter the surface and groundwater by runoff
or leaching, be taken up by plants and soil organisms, or
stay in the soil (3–13).
As the use of pesticides in agriculture inevitably
leads to exposure of nontarget organisms (including
humans), undesirable side effects may occur on some
species, communities, or on ecosystems as a whole.
Therefore, the knowledge of pesticide environmental
behavior, like adsorption to soil, leaching to groundwater,
and volatility in the atmosphere, is of primary concern for
an accurate assessment of the risk to the environment and
humans (14).
SORPTION TO SOIL COMPONENTS (ORGANIC AND
INORGANIC)
The soil is a complex medium whose physical, chemical,
and biological properties are highly heterogeneous. The
horizontal and vertical distribution of textural and
structural elements in soil, and the differences in
physicochemical properties, lead to variations in sorption,
desorption, transport, and permeability of soil. The soil
serves as a temporary or permanent sink for any chemical
substance that hits its surface (12).
The sorption of pesticide on soil reduces its mobility;
it is dependent on the physical and chemical properties
of the soil as well as the molecular characteristics of the
pesticide. Soil organic matter is the main soil constituent
responsible for sorption of nonionic pesticides. In order
to assess pesticide mobility, a sorption constant, based
on the soil organic carbon (Koc, dm3 kg−1 ), can be used.
Thus, Koc = Kd/oc, where Kd is a measure of the extent
of pesticide sorption by the soil and oc is the fraction of the
organic carbon presents in the soil (15).
A significant proportion, typically ranging from 20% to
70%, of a pesticide or its metabolites may remain in the soil
as ‘‘persistent residue,’’ sequestered in soil colloids (16).
The sequestration results in a diminution in the rate
and extent of biodegradation by microorganisms in the
field. The decline in toxicity and bioavailability as a
result of the accompanying sequestration may represent a
natural remediation process (17,18).
BIOLOGICAL AND CHEMICAL DEGRADATION
Losses of pesticide in the soil via microbiological and
chemical pathways are collectively termed degradation.
The rates of degradation can be characterized by
a half-life (DT50). In soil, this rate increases with
temperature and water content (19). The products of
pesticide degradation, named metabolites, may have
environmentally undesirable characteristics. Therefore,
when the evaluation of the environmental fate of a
pesticide is carried out, it is recommended to take the
fate of metabolites into consideration.
VOLATILIZATION
Volatilization is a major cause of pesticide loss from target
areas, particularly when pesticides are applied to the
surfaces of soils or plants. Atmospheric transport and
deposition are the major processes for distributing many
pesticides around the world (20). The most appropriate
criterion of the volatilization rate of a pesticide is Henry’s
law constant (Kh, dimensionless), the ratio of the vapor
pressure to the water solubility (21,22). Compounds with
Kh much greater than 2.5 × 10−5 are volatile, with
volatility decreasing with time, whereas compounds with
Kh much smaller than 2.5 × 10−5 are much less volatile,
with volatility increasing with time (21–23).
RUNOFF AND LEACHING
Water may disperse pesticides into the environment via
foliar washoff, surface runoff, and leaching. Runoff may
contribute to pollution of surface water and leaching to
contamination of groundwater.
Runoff is defined as the water and any dissolved or
suspended matter it contains that leaves a plot, field,
or small single-cover watershed in surface drainage (15).
Runoff may include dissolved, suspended particulate, and
sediment-adsorbed pesticides. Pesticides that remain at
the soil surface for longer periods of time because they
are strongly adsorbed and resistant to degradation and
volatilization will be more susceptible to runoff, whereas
incorporation into the soil will reduce runoff risk (24).
Soluble pesticides may be more readily leached into the
soil during the initial rainfall.
Pesticide leaching may cause groundwater pollution.
The extent to which groundwater contamination occurs
will depend on pesticides properties, soil characteristics,
drainage rate, and water table depth. A combination of
mobility and persistence determines whether a compound
will be degraded during its residence time in the
zone above the groundwater (21,25). Gustafson et al. (25)
propose a single numerical index for predicting the water
contamination potential of pesticides, the groundwater
ubiquity score (GUS). It is defined as GUS = log (DT50 ) ×
(4 − log (Koc)).
Pesticides detected in groundwater generally have
GUS values exceeding 2.8, whereas compounds with GUS
values below 1.8 were not detected in groundwater.
Rate of pesticide leaching in soil decreases with
increasing organic matter content and depth of the surface
zone with high biological activity. Moreover, the presence
of macropores (cracks, worm holes, root channels) in
soils enhances the hazard of pesticide leaching to the
groundwater. Via these macropores, water and solutes
can be transported rapidly to the subsoil and groundwater,
bypassing much of the soil matrix (26).
PESTICIDES GROUPS
Over the last 20–25 years, a drastic change in the
general characteristics of pesticides has occurred. They
have changed from those with extremely low water
solubility, strongly sorbed with long half-life to those
that are more water soluble, only slightly sorbed, and
less persistent. According to the chemical structure,
pesticides can be classified as organochlorines, extremely

hydrophobic with very high partition coefficient, and
as organophosphates, carbamates, and pyretroids, which
are more hydrophilic with lower partition coefficient
value (27). Moreover, pesticides can be conveniently
divided into classes, depending on which particular pest
they are directed toward. Thus, the main groups are the
herbicides, insecticides, and fungicides.
Pesticide properties, combined with the amount
applied, determine how many pesticides can be trans-
ported along each route into the environment.
As was mentioned above, the low water solubility
and the high partition coefficient of the organochlorine
pesticides (OCP) confer to these pollutants a high envi-
ronmental persistence. For this reason, the monitoring of
these compounds is highly recommended.
WHAT ARE ‘‘ORGANOCHLORINE PESTICIDES’’?
Organochlorine pesticides are important classes of envi-
ronmental contaminants. Owing to their environmental
persistence, toxic properties, and potential for bioaccumu-
lation, they can be classified as ‘‘persistent organic pollu-
tants’’ (POPs). Of additional concern is the ability of many
POP compounds to undergo long-range atmospheric trans-
port, developing from their semivolatile nature, which has
resulted in their distribution in nearly all environmental
compartments on a global scale, often far removed from
their places of production and application (28–30).
OCP have been widely used throughout the world
since the 1950s, although the use of many of these
chemicals has been subjected to restriction and/or banned
in many countries in recent years. However, the use
of OCP in tropical and subtropical countries has been
increasing in recent years in agricultural and public
health programs (31). In addition to current usage,
significant environmental burdens remain for many of
these chemicals, which can act as secondary sources from
which releases to the environment can occur, even decades
after initial use (32).
OCP enter the soil by deposition from air, drift,
or by washing off from plant surfaces during rainfall
or irrigation. As a result of their strong hydrophobic
(lipophilic) character, these compounds are mainly found
associated with organic matter in soil and lipid tissues of
organisms.
Organochlorine Pesticide Distribution in Relation to Land
Uses: A Study Case
Knowledge of pesticide concentration and spatial distri-
bution is necessary in order to asses the environmental
impact of these compounds. A field study was carried out
as part of an integral research about the dynamics of
OCPs in a shallow lake watershed, Argentina, in relation
to surrounding land uses (12).
The primary objective of this study was to assess the
effect of land uses, such as agricultural, recreational
(used for hiking, camping, and campfires), and natural
(forested area that have never been cultivated) on the
occurrence and distribution of organochlorine pesticides
in soils from Argentina.
PESTICIDE CHEMISTRY IN THE ENVIRONMENT 649
The organochlorine compounds analyzed included α-,
β-, γ -, and δ-HCH; p,p -DDT and its degradation products;
p,p -DDE and p,p -DDD; heptachlor and its metabolite
(heptachlor epoxide); aldrin; dieldrin; α- and γ -chlordane;
and α- and β-endosulfan. Three depths were selected from
each land use (0–15 cm, 15–30 cm, and 45–55 cm).
The results showed that, as was expected, the soil total
organic carbon (TOC) decreased with depth for all three
land uses. For the natural soils, total OCP concentrations
were well correlated with the levels of TOC through the soil
profile, whereas in agricultural and recreational soils, this
correlation was not observed. These results imply that
although TOC is known to be the most important OCP
sorbent in soil surface horizons, additional factors exist,
such as particle size characteristics and organic matter
composition, and physicochemical characteristics of OCPs
involved in pesticide retention.
Natural soil accumulated OCP to a greater extent than
the other soils, even though it has never received direct
OCP application. It is located on a hill that is clothed
with trees and grass cover, which would act as an OCP
trap, avoiding the OCP to leach up through soil profile,
volatilization, and minimizing the risk of soil erosion.
The highest concentrations were found in the surface
layer, decreasing gradually with depth as a result of an
undisturbed soil profile. DDT (472.5 ng/g dry wt), DDE
(99.98 ng/g dry wt), and HCHs (25.44 ng/g dry wt) were the
main pesticides found. Moreover, the levels of OCP in soils
from an organic farm near to these natural soils ranged
between 4–12 ng/g dry wt depending on the enrichment
with pine needles collected from the surrounding natural
areas (33,34). It is known that the waxy surfaces of pine
needles has the ability to bioaccumulate airborne lipophilic
pollutants (35). The soils from the organic farm could be
considered as ‘‘natural soil,’’ because it has never received
direct pesticides application, either the lack of permanent
vegetal cover and the soil disturbation by tillage practice
lead to lower OCP levels. However, in both soils (natural
and organic farm), a similar OCP pattern was found, as
a consequence of the exposure to the same agricultural
environment.
The values of natural soils are similar to those found
by Dimond and Owen (36), (in Maine) for forest soils in
sprayed areas during 1958 to 1967.
DDT was heavily used in Argentina, but it was totally
banned in 1990. Commercial acaricide (Dicofol), still in use
in Argentina, has been reported to contain a significant
amount of DDT as manufacturing impurity and might
represent a continuous source of DDT in the studied area.
However, the high values of DDT found in these soils
suggest that ‘‘new’’ DDT continues to be atmospherically
transported from other countries, mostly tropical, where
DDT is still in use (37–40).
High levels of heptachlor epoxide, a metabolite of hep-
tachlor, were also found possibly because of physicochemi-
cal processes, such as photolysis and photodechlorination,
and because of the abundance of microorganisms and
edaphic biota in the surface layer of these natural soils,
biotransformation processes are also assigned.
Chlordane residues were found in all the samples with
the highest values being those from the surface layer
650 PESTICIDE CHEMISTRY IN THE ENVIRONMENT
of natural soil. The ratios γ -/α-chlordane found in these
soils suggest direct application of technical chlordane in
this zone.
Regarding to HCH group, the surface layer of the
three studied soils showed higher concentration of lindane
than the deeper depth, although there were not marked
differences among depths. Lindane is one of the few
OCPs analyzed in this work that is currently being used
in Argentina.
Those soils dedicated to recreational activities showed
the highest OCP values in the middle depth. During
flooding periods, weathering of the surface layer occurs;
as a result, a clearance of OCPs and predominance of
HCHs are observed. The contribution of these compounds
would come from the lagoon water; significant evidence
exists that HCHs are the main pesticide group found in
the aquatic biota of this waterbody (41).
In the agricultural soil during the last 25 years, a
traditional tillage has been carried out in which a
moldboard plough was used at a depth of 25–30 cm and
in which straw was burned. This kind of tillage practice
enhances organic carbon and fine particles (clay and silt)
loss (30% in relation to natural soils), jointly OCP loss
by volatilization or surface runoff. In addition to this
management practice, the topography enhanced by slopes,
wind, and water erosion during rain and irrigation periods
justify these losses. Despite this, high values of OCPs
in the surface layer were detected, with DDT being the
main pesticide found, supporting the assumption that,
because these pesticides (except lindane and endosulfan)
are forbidden in Argentina, they would reach these soils
via atmospheric transport.
Finally, soil characteristics, topographic position, and
land uses justify the differences in OCP accumulation
among the studied soils. The accumulation pattern
represent a dynamic balance between deposition and
erosion in association with wind-induced circulation,
tillage management practices, and effects by natural
processes, such as rainfall and flooding periods, according
to different soils.
The results of this study suggest that natural areas
should not be overlooked when assessing the sources and
monitoring the occurrence of OCPs.
RELATION TO SURFACE WATER
The use of pesticides on agricultural land may have a neg-
ative effect on nearby aquatic ecosystems (12,13,42,43).
This effect is determined by the specific chemicals used, the
place and methods of application of the chemicals, and the
various mechanisms available for transporting compounds
to the waterbodies, with soil erosion being a major path-
way (44). The extent to which pesticides move from treated
soils depends on the physicochemical properties of the dif-
ferent active ingredients, the soil management practices,
and the combination of environmental variables, such as
frequency, intensity, and duration of precipitation (45).
Hydrophobicity and persistence are the main properties
of pesticides that control their accumulation in sediment
and aquatic biota. Thus, this fact confirms the current
necessity of OCP monitoring.
CONCLUDING REMARKS
It is known that pesticides are released into the
environment as formulated products that are designed
to deliver the active ingredient to the target organisms.
During the last forty years, a transition in the type
of used pesticides has taken place. Thus, the highly
persistent and bioaccumulatives organochlorine pesticides
were replaced by organophosphates, carbamates, and
pyretroids, currently used, with faster degradation in the
environment.
However, despite pesticides undoubtedly contributing
to increase farm productivity and human health, they
also create several problems including widespread accu-
mulation of residues with damage to wildlife, fisheries,
beneficial insects, and even humans as a result of their
dispersion toward nontarget areas. Today, the concern is
the impact of those ‘‘environmental healthy pesticides,’’
because they are being linked to a high fish, wildlife, and
bees toxicity, and some of them, because of their high
water solubility, are potential groundwater contaminants.
For these reasons, the monitoring of pesticide applications
is necessary because it can be helpful in modifying the
application techniques or use of particular chemicals to
avoid adverse effects. Concern over possible impacts of
pesticides on the environment and humans is not confined
just to the highly industrialized nations. Developing coun-
tries are also experiencing the same concern exacerbated
by lack of infrastructure and resources to deal with them.
BIBLIOGRAPHY
1. Himel, C.M., Loats, H., and Bailey, G.W. (1990). Pesticides
in the Soil Environment: Processes, Impact and Modeling.
H.H. Cheng (Ed.). Soil Science Society of America, Inc.,
Madison, WI, pp. 7–52.
2. Pimentel, D. (1995). J. Agric. Environ. Ethics. 8: 17–29.
3. Rapaport, R.A. et al. (1985). Chemosphere. 14: 1167–1173.
4. Rapaport, R.A. and Eisenreich, S.J. (1988). Environ. Sci.
Technol. 22: 931–941.
5. Bidleman, T.F., Widequist, V., Jansson, B., and Soderlund, R.
(1987). Atmosph. Environ. 21: 641–654.
6. Bidleman, T.F. et al. (1989). Arctic. 42: 307–313.
7. Swackhamer, D.C., McVeety, B.D., and Hites, L.A. (1988).
Environ. Sci. Technol. 22: 664–672.
8. Patton, G.W., Walla, M.D., Bidleman, T.F., and Barrie, L.A.
(1991). J. Geoph. Res. 96: 10867–10877.
9. Kolpin, D.W., Thurman, E.M., and Goolsby, D.A. (1996).
Environ. Sci. Technol. 30: 335–340.
10. Miglioranza, K.S.B. et al. (1999). Environ. Poll. 105: 91–99.
11. Miglioranza, K.S.B. et al. (2002). Environ. Sci. Poll. Res. 9:
250–256.
12. Miglioranza, K.S.B., Aizpún de Moreno, J.E., and Moreno,
V.J. (2003). Environ. Toxicol. Chem. 22: 712–717.
13. Miglioranza, K.S.B., Aizpún de Moreno, J.E., and Moreno,
V.J. (2003). J. Soil Sed. 4: 264–265.
14. Van der Werf, H.M.G. (1996). Agric. Ecosys. Environ. 60:
81–96.
15. Leonard, R.A. (1990). Pesticides in the Soil Environment:
Processes, Impacts and Modeling. H.H. Cheng (Ed.). Soil
Society of America, Book Series 2, Madison, WI, pp. 303–349.
 REMEDIATION OF PESTICIDE-CONTAMINATED SOIL AT AGRICHEMICAL FACILITIES
16. Calderbank, A. (1989). Environ. Contam. Toxicol. 108:
71–103.
17. White, J.C., Alexander, M., and Pignatello, J.J. (1999). Envi-
ron. Toxicol. Chem. 18: 182–187.
18. Alexander, M. (1997). Environmentally Acceptable Endpoints
in Soil. D.G. Linz and D.V. Nakles (Eds.). American Academic
of Environmental Engineers, Annapolis, MD, pp. 43–136.
19. Walker, A. (1976). Pestic. Sci. 7: 41–49.
20. Atlas, E.L. and Schauffler, S. (1990). Long-Range Trans-
port of Pesticides. D.A. Kurtz (Ed.). Lewis, Chelsea, MI,
pp. 161–183.
21. Jury, W.A., Spencer, W.F., and Farmer, W.J. (1984). J.
Environ. Qual. 13: 573–579.
22. Spencer, W.F. and Cliath, M.M. (1990). Long-Range Trans-
port of Pesticides. D.A. Kurtz (Ed.). Lewis, Chelsea, MI,
pp. 1–16.
23. Clendening, L.D., Jury, W.A., and Ernst, F.F. (1990). Long-
Range Transport of Pesticides. D.A. Kurtz (Ed.). Lewis,
Chelsea, MI, pp. 47–60.
24. Larsson, S.J. et al. (1995). Chemosphere. 31: 3305–3321.
25. Gustafson, D.I. (1989). Environ. Toxicol. Chem.
339–357.
26. Beven, K. and Germann, P. (1982). Water Resourc. Res. 18:
1311–1325.
27. Carlsen, R.F. and Smith, S.N. (1987). Silent Spring Revis-
ited. G.J. Marco et al. (Eds.). American Chemical Society,
Washington, DC, pp. 71–83.
28. Larsson, P., Jarnmark, C., and Sodergren, A. (1992). Mar.
Poll. Bull. 25: 281–287.
29. Iwata, H. et al. (1994). Environ. Poll. 85: 15–33.
30. Muir, D.C.G. et al. (1996). Environ. Sci. Technol. 30:
3609–3617.
31. Forget, G. (1991). J. Toxicol. Environ. Health. 32: 11–31.
32. Finizio, A., Bidleman, T.F., and Szeto, S.Y. (1998). Chemo-
sphere. 36: 345–355.
33. Gonzalez, M., Miglioranza, K.S.B., Aizpún de Moreno, J.E.,
and Moreno, V.J. (2003). J. Agric. Food Chem. 51: 1353–1359.
34. Gonzalez, M., Miglioranza, K.S.B., Aizpún de Moreno, J.E.,
and Moreno, V.J. (2003). J. Agric. Food Chem. 51: 5024–5029.
35. Krauthacker, B., Romanic, S.M., and Reiser, E. (2001). Bull.
Environ. Contam. Toxicol. 66: 334–341.
36. Dimond, J.B. and Owen, R.B. (1996). Environ. Poll. 92:
227–230.
37. Ostromogil’skii, A.K., Kokorin, A.O., and Afanas’ev, M.L.
(1987). Soviet Meteorol. Hydrol. 28–35.
38. Ramesh, A. et al. (1991). Environ. Poll. 74: 293–307.
39. Matthiessen, P. (1985). Environ. Poll. 10: 189–211.
40. Bidleman, T.F. and Falconer, R.L. (1999). Environ. Sci.
Technol. 33: 206A–209A.
41. González Sagrario, M.A., Aizpún de Moreno, J.E., Moreno,
V.J., and Escalante, A.H. (1998). Environ. Sci. 6: 153–170.
42. Miglioranza, K.S.B., Aizpún de Moreno, J.E., and Moreno,
V.J. (2004). Environ. Sci. Poll. Res. 11: 227–232.
43. Miglioranza, K.S.B., Aizpún de Moreno, J.E., and Moreno,
V.J. (2004). Water Res. 38: 1765–1772.
44. Munn, M.D. and Gruber, S.J. (1997). Environ. Toxicol. Chem.
16: 1877–1887.
45. Cheng, H.H. (1990). Pesticides in the Soil Environment:
Processes, Impacts and Modeling. Soil Society of America,
Book Series 2, Madison, WI, p. 539.
651
REMEDIATION OF PESTICIDE-CONTAMINATED
SOIL AT AGRICHEMICAL FACILITIES
WILLIAM R. ROY
Illinois State Geological Survey
Champaign, Illinois
AGRICHEMICAL FACILITIES
An agrichemical or crop production facility is a commercial
agricultural application business. These facilities store,
sell, mix, load, and apply pesticides and fertilizers to
fields for farmers and livestock producers. There are about
9,250 of these relatively small businesses in the United
States and about 850 registered agrichemical businesses
in Illinois. Pesticide mixing, loading, and handling sites
are diverse and vary in the types of products handled at
the site, the size of the facility, and the construction and
type of containment (1).
8:
Sources of Water and Soil Contamination
Bulk liquid and dry pesticides and fertilizers are typically
mixed, repackaged, or transferred from one container
to another. Accidental and incidental chemical releases
such as spills, tank leaks, hose breaks, and transport
accidents can result in contaminated water and soil at
agrichemical facilities. Moreover, catastrophic events such
as fires and floods can also result in soil, sediment, and
water contamination.
Extent of Water and Soil Contamination
About 25% of the groundwater samples collected from
52 agrichemical facilities in Illinois contained 11 pesti-
cides (2). The four most frequently occurring pesticides
were bentazon, atrazine, trifluralin, and metolachlor.
Nitrate was detected in 16 samples at concentrations rang-
ing from 0.17 to 59 mg N/L. In additional sampling at four
agrichemical facilities, 17 contaminants (nitrate, pesti-
cides, and pesticide degradates) were detected. Atrazine,
for example, detected in 15 of 22 samples, ranged from
0.54 to 61 µg/L. Environmental property audits conducted
at 49 randomly selected agrichemical facilities in Illinois
revealed that pesticides were detected in at least one
soil sample collected at every site and at all sampling
depths (3). The most frequently detected pesticides were
alachlor, atrazine, metolachlor, trifluralin, pendimetha-
line, cyanazine, metibuzin, metribuzin DA, butylate, and
α-BHC. Pesticides were detected to depths of 4.5 m. About
50% of all pesticide detections, however, were found in
the gravel layer that serves as the parking lot/road base
for each facility. In an extreme case, the gravel layer at
an abandoned agrichemcial facility in Illinois contained
4.4% trifluralin by weight (4). Pesticide-contaminated soil
and groundwater at retail agrichemical facilities have
also been documented in Minnesota, California, Florida,
Michigan, Wisconsin (1), and Arkansas (5).
Movement of Pesticides and Fertilizers
A portion of the pesticides and fertilizers in soil may
leach into groundwater. Many of the commonly used
652 REMEDIATION OF PESTICIDE-CONTAMINATED SOIL AT AGRICHEMICAL FACILITIES
pesticides and fertilizers are fairly mobile and can be
leached as surface water infiltrates into the ground. The
contaminated gravel, in which the bulk of pesticides
may occur, poses a risk to groundwater quality (4).
Hence, contaminated soil and gravel can adversely
impact groundwater and possibly drinking water in
rural communities. Because of this potential, it may be
necessary to remediate contaminated soil to minimize its
impact on water quality.
CLEANUP OBJECTIVES
Groundwater Cleanup Objectives
The values for groundwater-cleanup objectives depend on
the specific chemical and the type of groundwater at
the site. For example, in Illinois, Class I groundwater
is called Potable Resource Groundwater, and Class II
groundwater is General Resource Groundwater (6). A
major distinction between the two types of groundwater
is the hydraulic conductivity of water-bearing geologic
materials: groundwater in an aquifer that has hydraulic
conductivity equal to or greater than 1 × 10−4 cm/s is
classified as Class I Groundwater. Therefore, Class
II groundwater would be expected in less permeable
materials. The groundwater quality standards for Class
II groundwater are less stringent than those for Class
I groundwater. Class II groundwater cleanup objectives
are five times greater than those given for Class I
groundwater. For example, the objective for atrazine in
Class I groundwater is 3 µg/L, whereas it is 15 µg/L for
Class II groundwater (6).
Soil Cleanup Objectives
The remediation of contaminated soil also requires
establishing some type of cleanup goals or objectives.
Such objectives are used to determine whether the soil
is ‘‘contaminated’’ in a regulatory context. Depending
on the specific type, agrichemicals are subject to
various physicochemical and biological processes that
can reduce their concentrations in soil and groundwater.
Unlike other common contaminants such as chlorinated
hydrocarbons, agrichemicals are intentionally released
into the environment to promote plant growth and to
limit the spread of weeds. A mixture of application-rate
equivalents and risk-based processes to derive soil cleanup
objectives have been promulgated in the Midwest. For
example, the Illinois Department of Agriculture provides
a default soil-cleanup objective (SCO) for surface soil (0
to 12 inches) and subsurface soil (12 to 60 inches) for
each major pesticide used in Illinois. A facility also has
the option of developing site-specific SCOs. These SCOs
were derived from the application of soil screening levels
(SSLs) developed by the U.S. Environmental Protection
Agency (7,8). The resulting SCOs take into account the
movement and chemical fate of a given chemical in
soil–water systems. In the application of the SSL concepts,
three physicochemical-fate mechanisms are considered:
dilution in groundwater, adsorption by soil organic
matter, and volatilization. Degradation by biotic or abiotic
mechanisms is not considered. The SSL model provides
a dynamic link between groundwater concentrations
and soil concentrations. The default subsurface (12 to
60 inches depth) SCOs were calculated using Class I
groundwater quality standards. The default surface SCOs
were also based on Class I groundwater standards with
the exception of acetochlor, alachlor, aldicarb, atrazine,
cyanazine, dimethoate, disulfoton, phorate, simazine, and
terbufos. For these pesticides, application-rate equivalents
are used as SCOs. As given in (9), each SCO was based
on pesticide label rates to the upper foot of soil or of a
soil/gravel mixture.
EXCAVATION OF CONTAMINATED SOIL
Once the SCOs have been established, the areas
and depths of contamination by agrichemicals at an
agrichemical facility can be estimated. The delineation of
the area and depth of material to be removed is not an exact
science. In practice, the boundaries for excavation are
subjectively interpolated between the locations where soil
cores were collected. The boundaries for excavation may
also, in part, be determined by any physical constraints
such as facility buildings, railroad tracks, or other
permanent structures. Once the contaminated materials
have been excavated from the dealership, they must be
disposed in an environmentally acceptable manner. There
are three options available: land application or spreading,
landfilling, and above-ground treatment.
OPTIONS FOR REMEDIATION
Land Application
Land application for the disposal of soils contaminated
with agrichemicals is a common approach in the Midwest
and may be the least expensive and most readily available
disposal technique (10). It is based on the premise that
if contaminated soil is spread in a thin layer on an
agricultural field, the pesticides and/or fertilizers will
dissipate to smaller and harmless concentrations. In
practice, the pesticides in the contaminated material are
applied at the application-rate equivalent that would
be appropriate for that field and would be expected to
biodegrade with time. If more than one pesticide or
fertilizer is present, the application rate is based on
the most limiting rate for the chemical detected in the
contaminated soil sediment, or fill.
When pesticide-containing soil is land applied, it may
be necessary to minimize the amount of coarse material
in the soil before application. Attempting to spread soil
that contains a significant amount of coarse material such
as gravel can damage most spreading equipment and
be objectionable to the landowner or farmer. A practical
technique to remove the coarse material is to use field-
scale screeners. The excavated, unsorted material is fed
into some type of mechanical screener to separate the
coarse debris from the finer soil. Depending on the size
of the equipment and the mesh size of the screen, a
screener can process between 50 to 150 tons per hour.
The coarse material may be used as a road base, field
roads, or placed back in the original excavation. The fine
 REMEDIATION OF PESTICIDE-CONTAMINATED SOIL AT AGRICHEMICAL FACILITIES
material is then land applied using spreading equipment.
There is a relatively narrow window of opportunity in
which site remediation and landspreading can occur:
postharvest to preplanting. During the planting season,
agrichemical facilities are in their busiest periods and
are reluctant to allow any site excavation because it can
hinder the operation of the facility and service to their
customers. During the winter, it is advisable to avoid
freezing conditions because the soil cannot be handled
effectively if frozen.
Often, there is not enough room at an agrichemical
facility for soil screening. In actual practice, the excavated
soil is transported from the facility to an area where
the soil can be screened and land applied. Compensation
to local farmers or land owners for the use of land to
establish a staging area for the screening process and for
the agricultural fields can add to the costs of remediation.
Landfilling
Historically, pesticide-contaminated soil excavated at
agrichemical facilities has been placed in landfills. When
land application is not possible, using contaminated soil
as a refuse cover or backfill may be an option. However, it
may be more expensive to place the soil in a landfill.
For example, if transportation costs are ignored, the
average cost of landfilling pesticide-contaminated soil in
the Midwest is about $25 per cubic yard compared with
an average of $10 per cubic yard for land spreading.
Moreover, limitations in landfill capacity may be a barrier
to the disposal of contaminated soil in some areas.
Soil Washing
Although land application and landfilling are commonly
used approaches for pesticide-contaminated soil at agri-
chemical facilities, other options are under study. It has
been suggested that soil washing could be applied for
a relatively wide range of pesticides (11). Soil wash-
ing is a treatment process in which contaminants are
extracted from the excavated material using a solution
containing chemical surfactants. For example, under labo-
ratory conditions, a 3% solution of the nonionic surfactant
polyoxethylene-4-lauryl ether was efficacious for remov-
ing atrazine from contaminated fill samples that were
collected at an agrichemical facility (12). After 1 hour
of mixing the fill–liquid suspensions, about 85% of the
atrazine was extracted, compared with 26% removal by
deionized water.
Bioremediation
Bioremediation is another option for remediating con-
taminated soil. Biodegradation is a process in which
microorganisms transform or alter the structure and prop-
erties of chemical compounds by enzymatic biochemical
reactions. Biochemical transformations can break down
toxic, anthropogenic compounds into less toxic metabo-
lites, and/or into inorganic compounds. Bioremediation
is the application of biodegradation to reduce the con-
centrations of contaminants in soil by enhancing the
physicochemical conditions for degradation, the creation
of optimum conditions for remediating contaminated sites.
653
For example, the biodegradation of trifluralin, meto-
lachlor, and pendimethaline in gravel fill samples was
promoted by mixing them with yard-waste compost under
laboratory conditions (13). The initial concentration of tri-
fluralin (2200 µg/kg) was reduced to about 505 µg/kg in a
1:1 fill:compost mixture after 40 days.
The basis for enhanced biodegradation is creating an
environment where microbes are more numerous thereby
increasing the probability of significant degradation.
There are several critical environmental conditions that
affect microbial activity: moisture content, oxygen, redox
potential, pH, nutrients, temperature, and especially the
supply of nitrogen and phosphorous. If the contaminated
soil contains a large amount of carbon but little
nitrogen and phosphorous, biodegradation will cease when
available N and P are depleted. Therefore, fertilization
of the contaminated fill materials may be required as
a management technique to enhance biodegradation.
For example, the addition of cornmeal can enhance the
extent of biodegradation of pesticide-contaminated soil
from an agrichemical facility when the material is land
applied (14).
The addition of a mixture of cornmeal, surfactant,
fertilizer, and clean soil, under laboratory conditions,
increased the rate of biodegradation of atrazine, alachlor,
and metalochor in gravel fill samples that were collected at
three different agrichemical facilities (15). The fill samples
were then either moistened with sterilized deionized
water to 25% moisture content, or amended with the
surfactant-fertilizer solution containing 0.1% oleic acid,
urea, polyoxyethylene-12-octyl phosphate (modified from
the oleophilic fertilizer Inipol EAP 22). After 8 weeks of
incubation, the amount of atrazine recovered decreased
by 55 to 74% in fill–soil samples amended with cornmeal
and the surfactant-fertilizer mixture. Similar reductions
of alachlor and metolachlor were also observed in those
samples (Fig. 1). No significant degradation of atrazine,
alachlor, and metolachlor was observed in the unamended
and sterilized fill samples. During the same incubation
period, however, the rate of reduction of atrazine, alachlor,
and metolachlor was slower in samples amended with
cornmeal alone. The biodegradation half-lives of atrazine
in the fill samples amended with corncob and the
surfactant-fertilizer mixture ranged from 3.9 to 6.8 weeks.
Phytoremediation
Phytoremediation is a technique in which plants, pri-
marily hybrid poplar trees, legumes, and grasses, are
planted in contaminated soil. The plants have the poten-
tial to take up and incorporate contaminants into plant
tissues, enhance biodegradation in the root zone, and
reduce the volume of groundwater available to transport
contaminants off-site. There are few, however, published
examples of field-scale applications of phytoremediation
at agrichemical facilities. Poplar trees were planted at an
agrichemical facility in Illinois to remove nitrogen and
herbicides from shallow surface groundwater (Eric Aitchi-
son, Ecolotree, Inc., personal communication). A total of
440 trees planted grew to a height of 15 feet in 17 months,
and a significant volume of groundwater was apparently
used by the trees. Phytoremediation may have a future,
654 REMEDIATION OF PESTICIDE-CONTAMINATED SOIL AT AGRICHEMICAL FACILITIES

110 THE SELECTION OF THE REMEDIAL APPROACH

100 No treatment
Past handling practices have resulted in pesticide-
Recovery of atrazine (%)

90 contaminated soil and groundwater at retail agrichemical

80 Cornmeal
facilities. During about the last ten years, however, new
containment regulations have been implemented in the
70
Midwest which will hopefully reduce incidental releases
60 of pesticides and fertilizers. For example, the Secondary
50 Containment, Operational Area Containment, and Con-
tainment Management and Operations in Illinois (19)
40 require the use of a containment system designed to
Cornmeal + surfactant +
30 fertilizer
intercept, retain, and recover operational and accidental
spillage, leakage, wash water, and agrichemical residues.
20
0 1 2 3 4 5 6 7 8 When contaminated soil, fill, or sediment is detected
Weeks of incubation at an agrichemical facility during a site assessment,
however, remediation may be required to minimize the
110
adverse impact of the material on surface and groundwa-
ter. Techniques for remediating pesticide-contaminated
100 No treatment soil is a current area of research, and new techniques
Recovery of metolachlor (%)

and technologies are currently under study. However,

90 Cornmeal
land application and landfilling may be the most cost-
80 effective, routine, and available technique to remediate
contaminated soil at retail agrichemical facilities. Experi-
70
ence gained in the Midwest may be very useful for other
60 domestic and international facilities.

50
Cornmeal + surfactant +
40
fertilizer
30
0 1 2 3 4 5 6 7
Weeks of incubation
Figure 1. Amount of atrazine (top) and metolachlor (bot-
tom) in soil–fill samples and the control sample (no
treatment) in the presence of ground cornmeal and the corn-
meal–soil–surfactant–fertilizer combination as a function of
incubation time (15).
but additional research and development are needed.
Phytoremediation, like bioremediation, may be too slow
when rapid remediation is required. Phytoremediation
may be suitable for inactive or abandoned agrichemical
facilities. Moreover, relatively large concentrations of phy-
totoxic chemicals in soil and groundwater can limit plant
growth (16).
OTHER APPROACHES
Other approaches for remediating pesticide-contaminated
soil have been proposed such as incineration, solidifi-
cation/stabilization, thermal desorption, radio-frequency
heating, supercritical carbon dioxide extraction, and
dechlorination by zero-valent iron (10,11,17,18). However,
these approaches may not be cost-effective for a typi-
cal agrichemical facility. For example, incineration of soil
may cost from $800 to $3,000 per cubic yard (17). However,
if the contaminated soil also contains chemicals that are
defined as hazardous under the Resource Conservation
and Recovery Act, the soil may have to be incinerated (1).
BIBLIOGRAPHY
1. Buzicky, G., Liemandt, P., Grow, S., and Read, D. (1992).
8 Agricultural chemical site remediation and regulations.
Pesticide Waste Management. Technology and Regulation.
J.B. Bourke (Ed.). American Chemical Society Symposium
Series 510, Chapter 21.
2. Goetsch, W.D., Kirbach, G.C., and Black, W.F. (1993). Pesti-
cides in well water and groundwater at agrichemical facilities
in Illinois: an initial investigation. In: Agrichemical Facility
Site Contamination Study. Illinois Department of Agricul-
ture, Chap. 4.
3. Krapac, I.G., Roy, W.R., Smyth, C.A., and Barnhardt, M.L.
(1995). Occurrence and distribution of pesticides in soil at
agrichemical facilities in Illinois. J. Soil Contamination 4:
206–226.
4. Roy, W.R., Krapac, I.G., and Chou, S.F.J. (1999). Chemical
fate and transport of atrazine in soil gravel materials at
agrichemical distribution facilities. J. Soil Contamination 8:
365–387.
5. Senseman, S.A., Lavy, T.L., and Daniel, T.C. (1997). Moni-
toring groundwater for pesticides at selected mixing/loading
sites in Arkansas. Environ. Sci. Technol. 31: 283–288.
6. Illinois Administrative Code. Title 35: Environmental Protec-
tion. Chapter I: Pollution Control Board. Part 620. Ground-
water Quality.
7. U.S. Environmental Agency. (1996). Soil Screening Guidance:
User’s Guide. EPA/540/R-963505.
8. U.S. Environmental Agency. (1996). Soil Screening Guidance:
Technical Background Document. EPA/540/R-95/128.
9. Illinois Administrative Code. Title 8: Agriculture and
Animals. Chapter I: Department of Agriculture. Part 259.
Agrichemical Facility Response Action Program.
10. Felsot, A. (1996). Options for cleanup and disposal of pesticide
wastes generated on a small scale. J. Environ. Sci. Health
B31: 365–381.
 PESTICIDE OCCURRENCE AND DISTRIBUTION IN RELATION TO USE
11. Koustas, R.N. and Fischer, D. (1998). Review of separation
technologies for treating pesticide-contaminated soil. J. Air
Waste Manage. Assoc. 48: 434–440.
12. Roy, W.R., Chou, S.F.J., and Krapac, I.G. (1995). Fate and
transport of atrazine in fill materials at agrichemical
facilities: A status report. In: Proc. 5th Annu. Conf. Illinois
Groundwater Consortium, Makanda, IL, March 29–30, 1995,
pp. 117–131.
13. Cole, M.A., Liu, X., and Zhang, L. (1994). Plant and microbial
establishment in pesticide-contaminated soil amended with
compost. In: Enhanced Biodegradation of Pesticides in the
Environment. K.D. Racke and J.R. Coats (Eds.). American
Chemical Society, Washington, DC, pp. 249–268.
14. Felsot, A.S., Mitchell, J.K., and Dzantor, E.K. (1994). Use of
Landfarming to Remediate Soil Contaminated by Pesticide
Waste. Report no. HWRIC RR-070, Hazardous Waste
Research and Information Center, Champaign, IL.
15. Roy, W.R., Krapac, I.G., and Chou, S.F.J. (1996). Fate and
Transport of Atrazine in Fill Materials at Agrichemical
Facilities: An Update. In: Proc. 6th Annu. Conf. Illinois
Groundwater Consortium, Makanda, IL, March 27–28, 1996,
pp. 261–270.
16. Arthur, E.L., Perkovich, B.S., Anderson, T.A., and Coats, J.R.
(1999). Degradation of an atrazine and metolachlor herbicide
mixture in pesticide contaminated soils from two agrochemi-
cal dealerships in Iowa. Water Air, Soil Pollut. 113: 1–17.
17. Risatti, J.B. and Zagula, S. (1993). Potential cost effective
technologies for remediation of Illinois agrichemical facilities.
In: Agrichemical Facility Site Contamination Study. Illinois
Department of Agriculture, Chap. 7.
18. Comfort, S.D., Shea, P.J., Machacek, T.A., Gaber, H., and
Oh, B.T. (2001). Field-scale remediation of a metolachlor-
contaminated spill site using zerovalent iron. J. Environ.
Qual. 30: 1636–1643.
19. Illinois Administrative Code. Title 8: Agriculture and
Animals. Chapter I: Department of Agriculture. Part 255.
Agrichemical Facilities.
PESTICIDE OCCURRENCE AND DISTRIBUTION
IN RELATION TO USE
XUEFENG CHU
Grand Valley State University
Muskegon, Michigan
GENERAL RELATIONSHIP BETWEEN PESTICIDE
OCCURRENCE AND USE
In the United States, about 1 billion pounds of pesticides
are used each year to control weeds, insects, and
other organisms and about 80% of them are used in
agriculture (1). Widespread use of pesticides over the past
several decades has led to their occurrence in groundwater,
surface water, aquatic biota and sediment, and the
atmosphere (2). In 8200 sample analyses conducted
by the National Water Quality Assessment (NAWQA)
program during 1992–1996, 58 pesticides were detected
at least once at or above 0.01 µg/L in both surface and
groundwaters in 20 of the nation’s major hydrologic
basins (3). The NAWQA studies also revealed that
more than 90% of water and fish samples from all
streams contained one or, more often, several pesticides
655
and about 50% of the sampled wells contained one
or more pesticides; the highest detection frequencies
were in shallow groundwater. In particular, the top 15
pesticide compounds found in water are among those with
the highest current use, although some organochlorine
compounds (e.g., DDT) heavily used in the 1960s were
found most often in fish and sediments. The most
frequently used pesticides occurred in geographic and
seasonal patterns that mainly correspond to distributions
of land use and associated pesticide use (4). Clearly, a
significant relation can be expected between the use
of pesticides and their occurrence in both surface and
groundwaters. However, the extent of their occurrence
and spatial and temporal distributions in surface and
groundwater (how, when, and where the pesticides occur)
may vary significantly.
ENVIRONMENTAL PATHWAYS OF PESTICIDES VERSES
THEIR OCCURRENCE AND DISTRIBUTION
The environmental pathways of pesticides determine their
occurrence and distribution. Following release into the
environment, pesticides can be washed off the target
application fields into adjacent surface waterbodies (e.g.,
streams, rivers, lakes) by runoff and erosion, leached by
percolating water down through the vadose zone into the
underlying aquifers, volatized into the atmosphere, taken
up by plants, and accumulated in aquatic organisms
through the food chain and even in human bodies.
Thus, pesticides follow complex pathways and patterns
of occurrence and distribution in the environment (Fig. 1).
TRANSPORT AND TRANSFORMATION PROCESSES OF
PESTICIDES VERSES THEIR OCCURRENCE AND
DISTRIBUTION
Pesticides, existing in three phases (dissolved, adsorbed,
and vapor phases) in the environment, undergo a series
of physical, chemical, and biological transport and trans-
formation processes such as advection, dispersion, degra-
dation, sorption, volatilization, and partitioning between
dissolved and vapor phases in surface and groundwa-
ter (Fig. 1). These essential transport and transformation
processes dominate pesticide occurrence and distribution
in the environment. Due to the complexity of these pro-
cesses, the occurrence and distribution of pesticides in
surface and groundwater can be extremely variable and
are affected by numerous factors such as pesticide use, pes-
ticide properties, land use, soil types, topographic features,
agricultural practices, climatic conditions, hydrology, and
hydrogeologic settings. Whether or not pesticides occur
in a certain part of the environment (a specific river
reach, soil layer, or aquifer) depends on three essential
elements: (1) availability of pesticide, (2) availability of
flowing water, and (3) accessibility/pathways.
VARIABILITY IN THE RELATIONSHIP BETWEEN PESTICIDE
OCCURRENCE AND USE
The relationship between pesticide occurrence and use
varies for different pesticides and geographic regions.
656 PESTICIDE OCCURRENCE AND DISTRIBUTION IN RELATION TO USE
Atmosphere Regional transport
Application
Volatilization
Dissolved-vapor
partitioning
Leaching
Root uptake
Advection and dispersion
Sorption and Vadose zone
desorption
Advection and dispersion
Sorption and Aquifer
desorption
Figure 1. Fate and transport of pesticides and their occurrence in the environment.
Even for a specific pesticide at one site, the relationship
changes with time due to the variability and complexity
of hydrology. In addition, the relationship for surface
water is different from that for groundwater due to their
distinct hydrologic characteristics and pesticide fate and
transport processes. Surface and groundwater systems
have dissimilar timing of pesticide occurrence, temporal
and spatial distributions, and magnitude of pesticide
exposure levels. A pesticide application may result in
a peak concentration in streams and shallow soils in
hours, whereas it may take days, even months, to reach
a peak exposure level in a groundwater system. The peak
concentrations in groundwater can also be much smaller
than those in surface water and shallow soils. A time shift
of arrival of peaks can often be observed in the transport
direction in both surface and subsurface water systems.
FIVE KEY PESTICIDE-USE ISSUES RELATED TO
OCCURRENCE AND DISTRIBUTION
In terms of pesticide use, five key issues that affect pes-
ticide occurrence and distribution need to be addressed:
(1) which pesticide is applied (type and properties of pesti-
cide), (2) how much and how often the pesticide is applied
(amount and frequency of application), (3) when the pesti-
cide is applied (timing of application), (4) how the pesticide
is applied (method of application), and (5) where the pes-
ticide is applied (domain of application). The types of
pesticides that possess specific physical, chemical, and
biological properties affect their occurrence and distribu-
tion directly by altering their environmental persistence,
mobility, and other fundamental transport processes. The
amount of applied pesticides determines the overall level
of pesticides potentially available for occurrence in sur-
face and groundwater, and the application frequency may
determine the possibility and extent of pesticide buildup.
The timing and method of application of pesticides affect
occurrence primarily by changing their environmental
Deposition
Decay
Runoff and erosion
Soil Stream
Decay Sorption and
Decay desorption
Advection and dispersion
Settling
Diffusion Sorption and
Percolation
desorption
Resuspension
Bed
Decay
Sedimentation
pathways and thus spatial and temporal distributions
and exposure levels. Additionally, the overall patterns of
pesticide occurrence and distribution vary with pesticide-
treated crops (application domain) that have different
growing periods and canopy characteristics and require
different soil–water–pesticide management practices.
EFFECTS OF USE OF PESTICIDES ON THEIR OCCURRENCE
AND DISTRIBUTION
Effects of Pesticide Properties on its Occurrence and
Distribution
The capabilities of sorption and volatilization of pesticides,
characterized by distribution coefficient and Henry’s
law constant, determine their existing forms (dissolved,
adsorbed, or vapor phases) and thus their mobility. These
two processes and the degradation of pesticides are critical
to their occurrence and distribution. Pesticides of lower
sorptive capability tend to move quickly with flowing
water. The spread of this type of pesticide throughout
the environment is, to a great extent, determined by
water movement. In contrast, long-lived pesticides of
strong sorption, such as DDT, may last decades in the
environment (4).
Effects of Application Quantity and Frequency on Pesticide
Occurrence and Distribution
Frequent applications of pesticides increase the total
amount of pesticides released into the environment and
hence endanger the vulnerable hydrologic system. In
particular, the accumulative effect of long-lived and
strongly sorbed pesticides is often of great concern.
Effects of Application Timing on Pesticide Occurrence and
Distribution
The combined timing of pesticide application and rain-
fall/irrigation can dominate exposure levels of some
 ASSESSMENT OF POLLUTION OUTFLOW FROM LARGE AGRICULTURAL AREAS
organophosphate insecticides in both surface and sub-
surface environments (5). A pesticide application, immedi-
ately followed by rainfall or overcanopy irrigation, provides
an extremely high potential for the pesticide to trans-
port from target fields to nearby surface waterbodies with
runoff and erosion and to leach into soils and the underly-
ing groundwater.
Effects of Application Methods on Pesticide Occurrence and
Distribution
Pesticides can be applied by different methods, such as
over-canopy spray, under-canopy ground surface spray,
and soil-incorporated application. Application methods
of pesticides affect their occurrence in the environment
by changing their initial distribution and hence altering
their following transport–transformation pathways and
environmental fate. In addition, irrigation methods (e.g.,
over-canopy and under-canopy irrigation) and their
combination with various pesticide application methods
can be critical to pesticide occurrence and distribution.
High variability of concentrations of some pesticides
during irrigation periods can be attributed to differences
in irrigation frequency and water management (6).
SUMMARY
The occurrence and distribution of pesticides in the
environment are closely related to their use and are
determined by a series of pesticide transport and
transformation processes. The occurrence and use of
pesticides follow extremely complex and dynamic patterns
that are affected by numerous factors related to pesticide
use, hydrology, and agricultural activities. Different ways
of pesticide use (application quantity, frequency, timing,
and method) can lead to distinct environmental fates of
pesticides in terms of their occurrence, extent, as well as
spatial and temporal distributions.
BIBLIOGRAPHY
1. U.S. Geological Survey. (2000). Pesticides in stream sediment
and aquatic biota. U.S. Geological Survey Fact Sheet 092-00.
2. Barbash, J.E., Thelin, G.P., Kolpin, D.W., and Gilliom, R.J.
(1999). Distribution of major herbicides in ground water of
the United States, U.S. Geological Survey Water-Resources
Investigations Report 98–4245.
3. U.S. Geological Survey. (1998). Pesticides in surface and
ground water of the United States: Summary of results of
the national water quality assessment program (NAWQA),
Pesticides National Synthesis Project.
Available at: http://ca.water.usgs.gov/pnsp/allsum/index.html
4. U.S. Geological Survey. (1999). The Quality of our nation’s
waters—nutrients and pesticides. U.S. Geological Survey
Circular 1225.
5. Chu, X. and Mariño, M.A. (2004). Semidiscrete pesticide
transport modeling and application. J. Hydrol. 285(1–4):
19–40.
6. Domagalski, J.L. (1997). Results of a prototype surface water
network design for pesticides developed for the San Joaquin
River Basin, California. J. Hydrol. 192: 33–50.
657
ASSESSMENT OF POLLUTION OUTFLOW FROM
LARGE AGRICULTURAL AREAS
RAMAKAR JHA
K.K.S. BHATIA
National Institute of Hydrology
Roorkee, Uttaranchal, India
C.S.P. OJHA
Indian Institute of Technology
Uttaranchal, India
The pollution from large agricultural areas in general
is discussed. The potential of remote sensing and
geographical information system as input for NPS
pollution assessment is emphasized. Mathematical models
commonly used for assessing of nonpoint source (NPS)
pollution are also described. Assessment of nutrient
pollution from three agricultural watersheds of a typical
Indian river basin ranging from 2–5 km2 is considered as
an illustrative example.
INTRODUCTION
The sources of surface water pollution typically fall
into one of two categories: point-source pollution and
nonpoint-source pollution. The term point-source pollu-
tion refers to pollutants discharged from one discrete
location or point, such as an industry or municipal
wastewater treatment plant, to surface water resources.
The term nonpoint-source pollution refers to the dis-
charge of pollutants, which cannot be identified as coming
from one discrete location or point. The pollution out-
flow from agricultural areas is known as nonpoint source
(NPS) pollution or diffuse pollution. NPS pollution occurs
when rainfall, snowmelt, or irrigation water runs over
land or through the ground, picks up pollutants, and
deposits them in rivers, lakes, and coastal waters or
introduces them into the groundwater. Sediment, nutri-
ents, and pesticides are common agricultural pollutants,
which cause surface water and groundwater quality prob-
lems (1–3).
Estimation of NPS pollution from cropland runoff is
an important feature of regional water quality planning.
The primary objective of NPS pollution estimation is to
protect or restore the designated use of a water resource
by reducing pollutant delivery to the water resource.
All nonpoint sources of pollution are not equal. Many
nonpoint sources of pollution are insignificant, whereas
other sources contribute substantially to water resource
impairment. Topographic, hydrologic, and agronomic
factors often combine to make some nonpoint sources
more detrimental to the beneficial use of water resources
than others. Therefore, a method or strategy to identify
and prioritize for treatment NPS areas, which are more
detrimental than others, is desirable. A land treatment
strategy should be developed to guide the selection and
implementation of best management practices (BMPs).
The BMP approach involves identifying and implementing
land use practices in agricultural areas that prevent
658 ASSESSMENT OF POLLUTION OUTFLOW FROM LARGE AGRICULTURAL AREAS
or reduce nonpoint pollution. In the case of erosion-
sedimentation, many such practices are well known for
agricultural activities. However, for some other types
of pollutants, the BMP may not be known. In such
instances, research that relates land use to water quality
is needed.
ROLE OF GEOGRAPHICAL INFORMATION SYSTEM (GIS)
AND REMOTE SENSING FOR NPS POLLUTION
ASSESSMENT
Like any environmental phenomena, there is a spatial
dimension to the management of water quality and
the control of nonpoint-source pollution in agricultural
watersheds. Understanding the spatial relationships
among the various pollution sources in a watershed
is critical to the successful implementation of the
best agricultural management practices. A geographical
information system (GIS) enables the effective integration,
management, and analysis of disparate data sets related
to chemicals, soils, climate, topography, land cover, and
land use. These data sets are important driving variables
for a nonpoint pollution model (4).
Several strategies for integrating distributed water
quality models and GIS have emerged during the past
few years (5–7). In most applications, the GIS was used
primarily to generate model input data and to display
output data from the model—an approach usually referred
to as loose coupling of a model and the GIS (8–10). In loose
coupling, there are two options: (1) loose coupling through
interchange of data files in ASCII format between the
model and the GIS and (2) loose coupling using a common
binary file (10). An interface program (e.g., pre- and
postprocessor) is normally used to convert and organize the
GIS data in the form that the model requires. The loose
coupling can often be time-consuming and problematic,
particularly, for large watersheds. The loose coupling also,
depends on the data structure of the GIS and the data file
format specified by the water quality model (11,12).
The second strategy of coupling the GIS and water
quality models involves close coupling in which the
control programs in the GIS software are slightly modified
to provide an enhanced environment for data transfer
between the model and the GIS database (8–10). The
options for loose and close coupling have significant
overlap, depending on the characteristics of the water
quality model. However, in close coupling, information
is passed between the model and the GIS via memory-
resident data models rather than external files. This
improves the interactive capabilities and performance of
closely coupled modeling environments.
The third strategy of coupling water quality models
and the GIS is tight coupling or full integration (8–10).
This strategy is based on incorporating the functional
components of one system (e.g., the model) within
the other, thereby eliminating the use of interface
programs. The most widely used option of tight coupling
involves developing tightly coupled seamless interfaces
between the model and the GIS. The major benefits, of
tight coupling strategy include improved performance,
enhanced user interface, and increased problem-solving
capabilities through efficient analysis of ‘‘what if?’’
scenarios. There is also little or no redundancy in the
development process because the systems can leverage off
each other.
Data acquired from remote sensors on aircraft and
satellites offer great promise in providing numerous model
inputs. Water quality models used to determine nonpoint
source pollution vary in type and extent of land use for
which these are best suited. Water bodies, urban areas,
forests, cropland, and bare soils or rock can generally be
differentiated very well. Individual species of vegetation
often can be separated, surface soil properties can be
delineated, and the percentage of vegetation or residue
cover on the land can sometimes be estimated. The
spectral, spatial, and temporal characteristics of remotely
sensed data make those data useful for models that
require knowledge of land use, cover, and conditions when
calculating nonpoint-source pollutant loads. The relevant
studies are available in many textbooks and the literature.
MATHEMATICAL MODELS FOR ESTIMATING
POLLUTANTS FROM AGRICULTURAL AREAS
Mathematical models have been used extensively since
the late 1960s to estimate pollutant outflows from
agricultural areas. The term mathematical model may
be new, but all of us have used mathematical mod-
els since we took algebra in school. Such models are
mathematical representations of physical, chemical, bio-
logical, social, economic, and related processes. The
advancement in computers, has made it possible now
to evolve different mathematical approaches (from sim-
ple linear regression to complex differential equations) for
obtaining favorable outcomes. Some of the mathematical
models used for NPS pollution estimation are as fol-
lows: (1) SWMM (13), (2) HSPF (14), (3) ANSWERS (15),
(4) CREAMS (16), (5) SWAM (17), (6) NTRM (18),
(7) SWRRB (19), (8) SHE (20), (9) AGNPS (21), (10) SWAT
(22), and (11) ANSWER-2000 (23).
Most of the mathematical models cited before simulate
hydrologic, chemical, and physical processes involved
in the transport of sediment, nutrients, and pesticides.
In general, mathematical models for estimating NPS
are divided into two classes: (1) a temporal class that
includes event simulation and continuous simulation and
(2) a spatial class that includes lumped parameters and
distributed parameters.
In the temporal class, the continuous simulation
models are preferred because they simulate all NPS
pollution producing events and can be used to estimate
total maximum daily loads and seasonal variations in
NPS loading. However, in the spatial class, distributed
parameter models are preferred. The principal advantage
of distributed parameter models over lumped parameter
models is that distributed parameter models represent the
spatial variability of watershed features (soil, topography,
land use etc.) more accurately. Moreover, the distributed
parameter models can also estimate pollutant losses at
different locations within the agricultural area. Some
of the most commonly used distributed parameter
 ASSESSMENT OF POLLUTION OUTFLOW FROM LARGE AGRICULTURAL AREAS
Channel cell
Overland flow cell
Figure 1. Mixed cell and routing process in overland and
channel flow.
continuous simulation models are AGNPS and ANSWER-
2000. The processes involved in these models are
explained next.
In the beginning of the modeling process, a large
agricultural area is divided into smaller grids or cells.
A typical arrangement of such cells is shown in Fig. 1. The
flow, sediments, and nutrients available at the beginning
of the cell, coupled with those generated within the cell,
are used to assess these values at the outlet of the
cell. Subsequent routing of these values can lead to the
estimation of flow as well as the NPS pollution load at the
outlet of the agricultural watershed.
In a hydrological process, the rainfall added to an
agricultural area in the beginning is intercepted by the
canopy cover, evapotranspirated or infiltrated. Infiltration
capacity is very high initially because of movement of
water to the unsaturated zone, and water within this layer
percolates downward. Water is discharged to the channel
drainage system, when infiltration capacity is exceeded
by the water supply and the depression storage has been
satisfied. The total inflow to the channel drainage system
is a combination of surface runoff, interflow, and base flow,
which is calculated by using a water depletion function to
gradually diminish it. A few specific relevant equations for
the computation of different flow components are shown
in Appendix 1. However, many more equations can be
obtained from the literature.
The sediment transport capacity and potential sedi-
ment are important variables of nonpoint source pollution
from a large agricultural area. As seen from Appendix 1,
the sediment transport capacity is based on a sediment
concentration ratio, c, estimated with a shear–stress rela-
tionship between the dominant flow shear on the soil and
the critical stress based on the Shields criteria (24). The
potential sediment supply due to rainfall can be calculated
from an empirical relationship for rainfall energy rate
adapted from the Universal Soil Loss Equation (USLE)
method. The method accounts for the effects of canopy
cover by introducing cover factors into the calculations
of soil detachment due to rainfall and runoff. Similar
models available in the literature can also be applied for
659
estimating the sediment transport capacity and poten-
tial sediment.
The mathematical approaches used for the NPS water
quality component in this chapter are widely used
NPS models (Appendix 2). These were developed for
the CREAMS model (25). Young et al. (20) modified the
algorithms applied in the CREAMS model and created
the AGNPS model, suitable on watershed scale. The
nutrient simulation is normally divided into two parts
to handle the soluble nutrient in the runoff and in the
sediment separately (21). For the soluble part, the general
assumption is that the rate of change in concentration
of soluble nutrients at the surface is proportional
to the difference between existing concentrations and
concentration in rainfall. The available nutrient content at
the surface is a result of combining the residual nutrient
at the surface with the amount from fertilizer application.
The available nutrient due to rainfall is estimated from
the nutrient concentration in rainfall and given as input
data. The movement rates are evaluated using nutrient
leaching and runoff extraction coefficients given as input
data for the model.
The mathematical equations are used separately for
each grid or cell where the total runoff, sediment, and
nutrient quantities are calculated by adding the surface
runoff contributions from various grids or cells to the
base flow. The routing through the channel system is
achieved using a storage-routing technique based on
continuity. Deposition of sediment and decay of nutrients
are estimated using fractions of the transported mass.
Mixing cell models, like other transport models, is subject
to numerical dispersion affected in part by cell size and the
assumption of complete mixing. Appendix 2 summarizes
some of the commonly used equations.
ASSESSMENT OF NUTRIENTS FROM A TYPICAL INDIAN
RIVER BASIN
To calibrate and test the mathematical equation shown in
Appendixes 1 and 2, the nonpoint-source pollutants from
three agricultural watersheds, ranging from 2–5 km2 ,
of Kali Basin, western Uttar Pradesh, India have
been analyzed. River Kali in western Uttar Pradesh,
India, has a significant socioeconomic value for nearby
areas. It receives many point and nonpoint sources of
pollution (26–29). The study area (Fig. 2) is part of the
Yamuna basin in the Indo-gangetic Plains, composed of
Pleistocene and subrecent alluvium, and lies between 29◦
13 to 30◦ N latitude and 77◦ 35 to 77◦ 45 E longitude. The
mean rainfall across the basin is 1000 mm, which occurs
mainly during the monsoon period. The basin area lies
between elevations from 276 m above mean sea level to
221 m above mean sea level, and the major land use is
agriculture. The soils of the area are loam to silty loam
and are normally free from carbonates. Agricultural waste
is also an important factor contributing to the pollution of
the river water.
The basin boundary, drainage pattern, contour maps,
spot heights and built-up area maps for River Kali, were
digitized using ILWIS—GIS. For the analysis, the spatial
data set available in grids or cells has been transformed
660 ASSESSMENT OF POLLUTION OUTFLOW FROM LARGE AGRICULTURAL AREAS

INDIA
Uttar Pradesh

Saharanpur
Kali
river
1

2
Devband
3
Municipal drain 4
5

Sugar mill drain

6
Muzaffarnagar
1. Raetam
2. Diwalerhi Municipal drain
3. Chetanpuri 7
4. Miragpur Municipal drain
5. Sanpla 8 Industrial drain
6. Maliro bridge
Sugar mill drain
7. Khanjanpur
8. Vihelna 9
9. Mansurpur bridge 10
10. Anchuli River 11
11. Pitlokar Hindon 12

Figure 2. Location of River Kali in India.

to a digital elevation model (DEM), followed by filtering to
arrive at a slope map, or a flow path map. For delineating
land use/land cover maps, classification and quantification
of different crops, and its temporal variation, IRS LISS III
multispectral imageries and IRS-PAN data for the study
period have been used. Ground truth verification was done
to verify the results obtained using remote sensing data.
Further, extensive sampling from different sampling
points giving a total of 576 data sets for nitrate (NO3 )
and orthophosphate (o-PO4 ) was done from March 1999 to
February 2000. The nutrients were analyzed following a
standard method (30). The NPS nutrient (NO3 and o-PO4 )
values obtained using mathematical equation based on
the mixing cell concept, given in Appendixes 1 and 2,
were compared with the observed nutrient values for all
four agricultural watersheds (Fig. 3). It was found that the
coefficient of determination, r2 , between observed and com-
puted values of nutrients was greater than 0.85 in all four
agricultural watersheds. It was also found that the error
estimate viz., standard error, was less than 2 for all cases.
SUMMARY
To estimate pollution from a large agricultural area,
distributed parameter and continuous simulation models
are preferred because they simulate pollution producing
events, total maximum daily pollution load, its seasonal
variations, and represent the pollutant losses at different
locations within the agricultural area. Remote sensing and
GIS are important tools coupled with the mathematical
models to estimate pollution from a large agricultural
area. An illustrative case study from India demonstrates
that NPS pollution can be computed well using the mixing
cell concept with remote sensing and GIS support.
BIBLIOGRAPHY
1. US EPA. (1984). Report to Congress: Non-Point Source
Pollution in the United States, U.S. Government Printing
Office, Washington, DC.
2. US EPA. (1986). Pesticides in ground water: Background
document 8. Office of ground water protection. Washington,
DC.
3. Gould, G.A. (1990). Agriculture, non-point source pollution
and federal law. University of California at Davis Law Review
23(3): 461–471.
4. Joao, E.M. and Walsh, S.J. (1992). GIS implications for
hydrologic modelling: simulation of non-point pollution
generated as a consequence of watershed development
scenarios. Computers, Environment and Urban Systems 16:
43–63.
 ASSESSMENT OF POLLUTION OUTFLOW FROM LARGE AGRICULTURAL AREAS 661

(a) Basin area 2.0 km2 (b) Basin area 4.0 km2
80 80
Observed Observed
70 Computed 70 Computed
NO3 load, kg day−1

NO3 load, kg day−1

60 60
50 50
40 40
30 30
20 20
10 10
0 0
1 3 5 7 9 11 1 3 5 7 9 11
Time, months Time, months
Basin area 2.0 km2 Basin area 4.0 km2
25 25
Observed Observed
Computed Computed
o -PO4 load, kg day−1

o -PO4 load, kg day−1

20 20

15 15

10 10

5 5

0 0
1 3 5 7 9 11 1 3 5 7 9 11
Time, months Time, months

(c) Basin area 4.15 km2

90
80 Observed
Computed
NO3 load, kg day−1

70
60
50
40
30
20
10
0
1 3 5 7 9 11
Time, months

Basin area 4.15 km2

25
Observed
Computed
o -PO4 load, kg day−1

20

15

10

0
1 3 5 7 9 11
Time, months

Figure 3. Observed and computed NPS nutrients from River Kali, India.

5. Maidment, D.R. (1993). GIS and hydrologic modelling. 6. Steyaert, L.T. and Goodchild, M.F. (1994). Integrating geo-
In: Environmental Modelling with GIS. M.F. Goodchild, graphical information systems and environmental simulation
B.O. Parks, and L.T. Steyaert (Eds.). Oxford University models—A status review. In: Environmental Information
Press, New York, pp. 147–167. Management and Analysis. W.K. Michener, J.W. Brunt, and
662 ASSESSMENT OF POLLUTION OUTFLOW FROM LARGE AGRICULTURAL AREAS

APPENDIX 1

Specific Equations for Different Variables Used in NPS Estimation

Variables Author(s) Governing Equationa
1 1/2
Overland flow Manning’s formula Qr = (D1 − DS )5/3 S1 A
R3
Interflow (31) Qint = Rec (Wac − Ret )


dF (m − m0 )(Pot + D1 )
Infiltration and surface detention (32) = k 1+
df F
s(Ta ) 1
(33) Pet = α (Kn + Ln )
s(Ta ) + ζ ρλy
Interception (34) V = (Si + Cp Ea tR )(1 − e−kP )

Surface storage (31) DS = Sd (1 − e−kPe )

1/2
Evapotranspiration (35) Pet = 0.0075Ra C1 δt Tavg
Transport capacity (36) YC = 2.65pcrf


β 60
Shear-stress relationships (36) τd = γ HL So
β + 1 Kf
τc = (σ − 1)γ φD50

Rainfall soil detachment (36) Grf = Erf (1 − GC)CFD

Runoff soil detachment (36) Gro = Ero D

Potential sediment supply (36) YS = (Grf + Gro )t
(NAVS − NAVR )
Soluble nutrient (nitrogen) (21) CRON =
FPOR
∗ [e(−NDMV IEFF ) − e(−NDMV IEFF −NRMV ROFF ) ]
NRNC ROFF
+
PEFF
Sediment attached nutrients (21) NSED = NSCN YSED ER
PSED = PSCN YSED ER
ER = aYSED
b
Tf

V = interception depth (mm), Si = storage capacity (mm), Cp = ratio of vegetated surface area, Ea = evaporation rate (mm h−1 ), tR = duration of the rainfall
(h), k = constant (mm−1 ), P = precipitation (mm), Ds = depression storage (mm), Sd = surface retention value (mm), Pe = cumulative rainfall excess (mm),
F = total depth of infiltrated water (mm), t = time (s), K = saturated conductivity (mm s−1 ), m = average moisture content of the soil, mo = initial soil
moisture content, Pot = capillary potential (mm), D1 = detention storage (mm), Pet = potential evapotranspiration rate (mm d−1 ), α = equilibrium factor,
s(Ta ) = slope of the saturation vapor pressure vs. temperature curve, ξ = psychrometric constant, Kn = short-wave radiation, Ln = long-wave radiation,
ρ = mass density of water, λy = latent heat of vaporization, Ra = total incoming solar radiation (mm), Ct = temperature reduction coefficient, δt = temperature
difference (◦ F ⇐◦ C), Tavg = mean temperature (◦ F ⇐◦ C), Qint = interflow (m3 s−1 ), Rec = coefficient representing depletion, Wac = water accumulation in
the upper zone storage (mm), Ret = retained storage (mm), Qr = channel inflow (m3 s−1 ), D1 − Ds = runoff depth above ponding, R3 = combined roughness
and channel-length parameter optimized for each land class, S1 = average overland slope, A = area of the element (m2 ), YC = sediment transport capacity

B
τd
(kg m−2 ), c = volumetric sediment concentration ratio = A , A = 0.00066, B = 1.61, rf = runoff (mm), τd = dominant flow shear stress, τc = critical
τc
stress, β = discharge parameter (= 5/3), γ = water specific weight (kg m−2 s2 ), HL = average runoff depth (mm), So = average overland slope, Kf = overland
flow friction = 60 + 3140 GC1.65 , GC = ground cover factor, D 50 = median size of soil particles (mm), σ = specific weight of sediment, φ Shields entrainment
0.11 ∗ ∗ τd γ D50
function = + 0.0211 log 10 R , R = Reynolds number = , ν = kinematic viscosity of water (m3 s−1 ), Grf = rate of soil detachment due to rainfall
R∗ ν
(kg m−2 h), CF = canopy factor, D = soil erodibility factor (gJ−1 ), Erf = rate of rainfall energy (Jm−2 h) = i(11.9 + 8.7 log10
i), i = rainfall intensity (mm
60 QL
h−1 ), Gro = rate of soil detachment due to runoff (kg m−2 h), Ero = rate of energy input to the soil by the flow (Jm−2 h) = γ So , QL = unit flow
Kf 2
−1 −1
discharge (m h ), So = element slope, t = time increment (h), CRON = soluble nitrogen concentration in runoff (kg ha ), NAVS = available nitrogen
2

content in the surface (kg ha−1 ), NAVR = available nitrogen in rainfall (kg ha−1 ), NDMV = rate of downward movement of nitrogen into soil, NRMV = rate
of nitrogen movement into runoff, IEFF = effective infiltration (mm), ROFF = total runoff (mm), FPOR = porosity factor, NRNC = nitrogen contribution due to
rain (kg ha−1 ), PEFF = effective precipitation (mm), NSED = overland nitrogen transported by the sediment (kg ha−1 ), NSCN = soil nitrogen concentration
(.001 gNg−1 soil), ER = nutrient enrichment ratio (a = 7.4, b = −0.2), Tf = correction factor for soil texture (0.85 for sand, 1.0 for silt, 1.15 for clay, and 1.50
for peat).

S.G. Stafford (Eds). Taylor and Francis Inc., Bristol, PA, GIS and windows application. Proceedings of Interna-
pp. 333–356. tional Symposium on Water Quality Modelling. Ameri-
7. Chen, C.L., Gomez, E., Chen, C.W., Wu, C.M., Lin, J.J., and can Society of Agricultural Engineers, St. Joseph, MI,
Chen, I.L. (1995). An integrated watershed model with pp. 242–250.
 ASSESSMENT OF POLLUTION OUTFLOW FROM LARGE AGRICULTURAL AREAS
APPENDIX 2
Specific Equations for Routing Component of NPS Assessment
Variables Author(s)
Grid or cell for sediment (25)
Grid or cell for nutrients (25)
a
The subscripts 1 and 2 indicate the beginning and end of the time step; I1,2 = inflow to the reach (m3 s−1 ) consisting of overland flow, interflow base flow
and channel flow from all contributing upstream basin elements; O1,2 = outflow from the reach (m3 s−1 ); S1,2 = storage in the reach (m3 ); t = time step
of the routing (s); R2 = channel roughness parameter; Ax = channel cross-section area (m2 ); So = channel slope; YSEDout = sediment leaving the cell (ppm);
SDep = deposition fraction; YSEDab = sum of all the sediment entering the cell (kg ha−1 ); YSEDin = sediment generated within the element; CCRONout = soluble
nutrients concentration in runoff leaving the cell (ppm); NDec = nutrients decay fraction; CCRONab = sum of all the nutrient entering the cell (kg ha−1 );
CCRONin = nutrient generated within the element.
8. Livingston, D. and Raper, J. (1993). Modelling environmental
systems with GIS: theoretical barriers to progress. In:
Innovations in GIS I. M.F. Worboys (Ed.). Taylor and Francis,
Bristol, PA, pp. 229–240.
9. Nyerges, T.L. (1993). Understanding the scope of GIS: its
relationship to environmental modelling. In: Environmen-
tal Modelling with GIS. M.F. Goodchild, B.O. Parks, and
L.T. Steyaert (Eds.). Oxford University Press, New York,
pp. 75–93.
10. Fedra, K. (1993). GIS and environmental modelling. In: Envi-
ronmental Modelling with GIS. M.F. Goodchild, B.O. Parks,
and L.T. Steyaert (Eds.). Oxford University Press, New York,
pp. 35–50.
11. Olivieri, L.J., Schaal, G.M., Logan, T.J., Elliot, W.J., and
Motch, B. (1991). Generating AGNPS input using remote
sensing and GIS. ASAE paper no. 91–2662. American Society
of Agricultural Engineers, St. Joseph, MI.
12. Donovan, J.K. (1991). The U.S. EPA’s Arizona rangeland
project: integration of GIS and a non-point source pollution
model. Proceedings of the 11 Annual ESRI User Confer-
ence, Environ. Systems Research Institute, Redlands, CA,
pp. 261–272.
13. Metcalf and Eddy, Inc. Univ. of Florida and Water Resources
Engineers, Inc. (1971). Storm Water Management Model
(4 vols.). U.S.Environmental Protection Agency. Washington,
DC.
14. Johanson, R.C., Imhoff, J.C., and Davis, H.H. (1980). User’s
manual for hydrologic simulation program—Fortran (HSPF).
EPA—600/9-80-015, U.S. Environmental Protection Agency,
Athens, GA, p. 684.
15. Beasley, D.B., Huggins, L.F., and Monke, E.J. (1980).
ANSWERS: a model for watershed planning. Trans. ASAE
23(4): 938–944.
16. Knisel, W.G., Jr. (Ed.). (1980). CREAMS: a field scale
model for chemicals, runoff and erosion from agricultural
management systems. Cons. Res. Repsot. No. 26. U.S. Dept.
Agr., Washington, DC, p. 640.
17. DeCoursey, D.G. (1982). ARS small watershed model. Paper
no. 82–2094–Am. Soc. Agr. Eng. St. Joseph, MI, p. 33.
18. Shafter, N.J., Gupta, S.C., Linden, D.R., Molina, J.A.E.,
Clapp, C.E., and Larson, W.E. (1983). Nitrogen-tillage-
residue management model (NTRM). In: Analysis of Ecolog-
ical Systems: State of the Art in Ecological Models. Elsevier,
New York, pp. 525–544.
19. Williams, J.R. and Nicks, A.D. (1985). SWRRB- A simulator
for water resources in rural basins: An overview. Proc.
Natural Resources Modelling Symp., Pingree Park, CO,
October 16–21, 1983. pp. 17–24. USDA-ARS. ARS-30.
663
Governing Equationa
YSEDout = 1000[(1 − SDep )(YSEDab + YSEDin )]
100
CCRONout = [(1 − NDec )(CRONab + CRONin )]
ROFF
20. Abbott, M.B., Bathurst, J.C., Cunge, J.A., Connell, P.E., and
Rasmussen, J. (1986). An introduction to the European
hydrological system—systeme hydrologique european, ‘SHE’
2: structure of physically based, distributed modelling system.
J. Hydrol. 87: 61–77.
21. Young, R.A., Onstad, C.A., Bosch, D.D., and Anderson, W.P.
(1986). AGNPS Agricultural Nonpoint Source Pollution
Model. A Large Watershed Analysis Tool. USDA. ARS
Conserv. Res. Report No. 35, Springfield, VA.
22. Arnold, J.G., Williams, J.R., Srinivasan, R., King, K.W., and
Griggs, R.H. (1995). SWAT—Soil and Water Assessment
Tool: Draft User’s Manual. U.S. Department of Agriculture,
Agricultural Research Service, Grassland, Soil and Water
Research Laboratory, Temple, TX.
23. Bouraoui, F. and Dillaha, T.A. (1996). ANSWERS-2000:
Runoff and sediment transport model. J. Environ. Eng. ASCE
122(6): 493–502.
24. Simons, D.B. and Senturk, F. (1976). Sediment Transport
Technology. Water Resources Publications, Fort Collins, CO .
25. Frere, M.H., Ross, J.D., and Lane, L.J. (1980). The nutrient
submodel. In CREAMS: A Field Scale Model for Chemicals,
Runoff and Erosion from Agricultural Management Systems.
Ed. W.G. Knisel, pp. 65–86, USDA, Cons. Research Report
No. 26.
26. Ghosh, N.C. and McBean, E.A. (1998). Water quality mod-
elling of the Kali River, India. Water, Air, and Soil Pollution
102: 91–103.
27. Jain, C.K., Bhatia, K.K.S., and Seth, S.M. (1998). Assessment
of point and non-point sources of pollution using a
chemical mass balance approach. Hydrol. Sci. J. 43(3): 379–
390.
28. Jha, R., Ojha, C.S.P., and Bhatia, K.K.S. (2000). Development
of deoxygenation and reaeration rate coefficients for a
small tributary of river Hindon, U.P., India. International
Conference on Integrated Water Resources Management
(ICIWRM-2000), New Delhi, India, pp. 464–474.
29. Jha, R., Ojha, C.S.P., and Bhatia, K.K.S. (2001). Refinement
of predictive reaeration equations for a typical indian river.
Hydrol. Process. 15(6): 1047–1060.
30. APHA. (1985). Standard methods for the examination of
water and wastewater (16th). American Public Health
Association, Washington DC, p. 1268.
31. Kouwen, N. (1988). WATFLOOD: a micro-computer based
flood forecasting system based on real-time weather radar.
Ca. Water Res. J. 13(1): 62–67.
32. Philip, J.R. (1954). An infiltration equation with physical
significance. Soil Sci. 77: Jan–Jun.
664 DEEP-WELL TURBINE PUMPS
33. Priestley, C.H. and Taylor, R.J. (1972). On the assessment
of surface heat flux and evaporation using large-scale
parameters. Mon. Weather Rev. 100(2): 81–92.
34. Linsley, R.K., Kohler, M.A., and Paulhus, L. (1949). Applied
Hydrology. McGraw Hill, New York.
35. Hargreaves, G.H. and Samani, Z.A. (1982). Estimating poten-
tial evapo-transpiration. Tech. Note, J. Irrig. and Drain.
Engrg., ASCE 108(3): 225–230.
36. Hartley, D.M. (1987). Simplified process model for water
and sediment yield from single storms. Trans. ASAE 30(3):
710–723.
DEEP-WELL TURBINE PUMPS
BLAINE HANSON
LAWR
Davis, California
Deep-well turbine pumps are used to pump groundwater
to the surface, and in some cases, they may be used to
pump water out of sumps and supply pressure. Deep-
well turbines consist of a housing or bowl, impellers,
and a shaft, all of which are installed in the well
[Figs. 1(a) and 1(b)]. The housing or bowl contains both
the impeller and the diffuser vanes. The shaft connects
the impeller to the electric motor or engine, whereas the
impeller transfers the energy developed by the motor or
engine to the water. Diffuser vanes guide the flow of water
from the impeller to the discharge point of the bowl.
Impellers transfer energy from the motor or engine
to the water either by centrifugal force or by lifting
action. They may be enclosed or semi-open. Enclosed
impellers have a disk or shroud on both sides of the vanes
(Fig. 2). These impellers require a close clearance between
a wearing ring, located at the inlet of the impeller, and
the impeller skirt of the bottom extension of the impeller.
In semi-open impellers, the shroud is only at the top of
the impeller. Semi-open impellers have a shroud only on
the top (Fig. 2). A close clearance (0.076 mm–0.177 mm)
between the bottom of the vanes and the pump housing is
necessary for good performance. Semi-open impellers are
less susceptible to wear from sand because they can be
raised slightly to reduce abrasion. This process, however,
will decrease pump output and efficiency.
Mixed flow impellers are often used in deep-well turbine
pumps. Mixed flow impellers use both centrifugal and
lifting forces to pump water. When the impeller rotates,
centrifugal forces develop in the water inside the impeller
and cause water to flow toward the impeller’s outer edge.
This flow, in turn, results in water flowing into the center
or eye of the impeller. Rotating a mixed flow impeller
also transfers energy to the water by a lifting force.
Impellers that pump water only through centrifugal force
are classified as radial flow impellers. Impellers that pump
water only through lifting forces are classified as axial or
propeller impellers.
Some deep-well turbines use a line-shaft to connect
the impellers to an electric motor or engine. Line shafts
are supported by bearings—either oil-lubricated or water-
lubricated—spaced at intervals along the shaft length.
Oil-lubricated bearings are housed within a tubing inside
the column pipe (the pipe that carries the water to the
surface). An oil reservoir located on the pump head drips
oil down the shaft.
Submersible pumps are deep-well turbines with a
waterproof electric motor installed in the well below the
pumping level. The motor is attached directly beneath the
bowl assembly, and water intake is between the motor
and the pump. A specially designed motor, slimmer and
more compact than a line-shaft motor, is installed in the
well. As both motor and pump are situated inside the well,
submersible pumps do not require a line shaft.
In submersible pumps, it is crucial that enough water
circulate around the motor to keep it cool. As water in
the well should flow past the motor casing into the pump
intake, submersible motors should be located above the
level at which most of the flow enters the well. A shroud
can be installed over the pump intake to force water to
flow past the motor.
As the impellers and bowls must be submerged inside
the well, the diameters of the impeller and bowl are limited
by the diameter of the well, which in turn limits the output
(capacity and pressure) developed by each impeller. This
limitation can be overcome if the bowls are installed in
stages, with each stage adding to the output developed
by the preceding stages. A five-stage pump will, therefore,
have five bowl assemblies, each containing an impeller,
with the water guided from stage to stage by diffuser
vanes built into the bowl assembly. Figure 1 shows a
two-stage pump.
Because deep-well turbine pumps are installed in the
well, pump size is limited by the diameter of the well
casing. If the pump is too large for the casing, rocks
from the gravel pack can become lodged between the
casing, which makes it difficult to remove the pump
and—particularly if the well is crooked—the pump may
be difficult to install. The clearance between the bowls
and the well casing should be at least 25 mm. Narrower
clearances may complicate installation or, in crooked wells
or where gravel becomes lodged between the pump and
casing, may make it difficult for the bowls to be removed
for repair.
The output of the pump must be sufficient to lift the
water to the ground surface and provide the desired
pressure at the desired flow rate or pump capacity.
However, in some cases, deep-well turbines lift the water
only and a centrifugal pump located at the discharge point
of the well, commonly called a booster pump, is placed in
line with the deep-well turbine pump to supply pressure.
The output of a pump is described by its performance
characteristic curves, developed by pump manufacturers.
These curves show the amount of total dynamic head
developed by a single stage at a particular capacity, the
pump efficiency at that capacity, and the horsepower
requirements of the pump. Performance curves show
which pump will provide the desired total head and
capacity and determine what size electric motor or engine
the pump will require.
Figure 3(a) shows the total head and capacity perfor-
mance curve for a typical pump. In a no-flow condition
(shut-off), the total head is maximum, or 22 m for this
 DEEP-WELL TURBINE PUMPS 665

(a)
Top shaft
for adjusting
impellors

Discharge
head

Shaft

(b)

Impeller

Diffuser
vanes

Pump
housing
or bowl
Water intake

Figure 1. Cutaway of a deep-well

turbine pump that shows (a) the
pump head, gear head, and shaft;
and (b) pump housing, impellers,
and shaft. [Source: Hanson (1)].

pump. As the pump capacity increases, the total head
decreases. At low capacities, most of the pump output con-
sists of total head (normally measured as the sum of the
pumping lift and the discharge pressure head), whereas
at high capacities, most of the pump output is in capacity.
Pump efficiency increases as pump capacity increases
until a maximum efficiency is reached [Fig. 3(b)]. There-
after, pump efficiency decreases as capacity continues
to increase.
Metric horsepower (mhp) increases slightly as capacity
increases to a maximum, and then it may decrease slightly
[Fig. 3(c)].
The performance curves are only for one-stage, but
they can develop performance curves for multiple-stage
pumps. For example, Fig. 3(a) shows that in a one-stage
pump, 14.6 m of head is developed at 3028 L/min. The
mhp is 14.2. In a three-stage pump, the head developed
at 3,028 L/min will be three times that of a single stage,
666 DEEP-WELL TURBINE PUMPS

or 43.8 m. The mhp will also be tripled, to 42.6 mhp, (a) 25

and the pump efficiency may change slightly. Usually the

Total dynamic head (m)

manufacturer’s curve will show the efficiency adjustments. 20
The performance curve for a multiple-stage pump can
therefore be estimated by multiplying the head and bhp 15
of the performance curve by the number of stages for each
10
capacity.
Pump performance curves have a variety of shapes. 5
Some head/capacity curves are steep (large decreases in
head occur as capacity increases), whereas others are flat 0
(small decreases in head occur as capacity increases). The 0 500 1000 1500 2000 2500 3000 3500 4000 4500
best curve will depend on changes in pumping conditions.
(b) 100
Where pumping levels fluctuate significantly, a steep curve
90
may allow for significant changes in pumping lift with only

Pump efficiency (%)

80
slight changes in pump capacity. A flat curve may help to 70
maintain a relatively constant pressure over a wide range 60
of flow rates. 50
An initially efficient pump with the necessary head and 40
capacity can become inefficient. Inefficiency is frequently 30
caused by wear because of sand in the well water, but it 20
can also be caused by a mismatch between the pump and 10
0
water system characteristics because of long-term changes 0 500 1000 1500 2000 2500 3000 3500 4000 4500
in groundwater levels or changes to the water supply
system. A mismatched pump is one that is operating (c) 20
properly but at an efficiency much less than the potential
Metric horsepower

maximum efficiency. ‘‘Repairing’’ a mismatched pump will 15

not improve efficiency.
The pump’s performance can be evaluated by first
testing the pump and then comparing that information
with the manufacturer’s performance curves. The pump
test consists of measuring the pumping lift, discharge
pressure, capacity, and input power to the pump and then
calculating overall pumping plant efficiency—that is, the
combined efficiency of the pump and motor or engine.
The overall efficiency will be less than the pump efficiency
reflected in the manufacturer’s performance curve because
of the inefficiency of the motor or engine.
Pump tests are conducted by utility companies, pump
dealers, and consultants. In defining the current status of
the pump, the test may provide any or all of the following
information:
• Pumping lift or level—elevation difference between
the discharge pipe at the pump head and the water
level in the pumping well
Enclosed Semi-open
Figure 2. Enclosed and semi-open impellers. [Source: Hanson
(2)].
10
5
0
0 500 1000 1500 2000 2500 3000 3500 4000 4500
Pump capacity (L/min)
Figure 3. Pump performance curves that show the relationships
between (a) total head and capacity, (b) pump efficiency and
capacity, and (c) metric horsepower and capacity.
• Static or standing water level—elevation difference
between discharge pipe and water level in the well
before pumping
• Drawdown—difference between the pumping level
and the static water level
• Discharge pressure—pressure at the discharge
point of the pumping plant measured with a
pressure gauge
• Discharge pressure head—the height of the column
of water needed to provide the pressure at its
base—calculated by multiplying the pressure by
a conversion constant (0.704, 0.1021, or 10.21 for
discharge pressure in psi, kPa, or bars, respectively)
• Total head—sum of the pumping lift and the
discharge pressure head (note that if the pump is
situated deep in the well, pressure losses caused by
friction in the column pipe may have to be added to
the total head)
• Pump capacity—water flow rate of the pump-
ing plant.
 MICROBIAL QUALITY OF RECLAIMED IRRIGATION: INTERNATIONAL PERSPECTIVE 667

• Well yield—flow rate or capacity divided by the 60

drawdown (a well yield decreasing over time may
indicate plugging in the well) 50
• Input horsepower to electric motor
• Percent motor overload 40

Total head (m)

• Input kilowatts to motor Repaired pump
• Kilowatt-hours per acre-foot of pumped water 30 Pumping lift = 31 m
Capacity = 2032 L/min
• Overall pumping plant efficiency Input horsepower = 28.4
20
Overall efficiency = 50% Worn pump
The pumping lift is measured by lowering an electric Pumping lift = 14 m
10 Capacity = 2361 L/min
water level indicator (two electrodes attached to a wire Input Horsepower = 19.3
connected to an ampere meter) or some other measuring Overall efficiency = 39%
device into the space between the well casing and the 0
0 1000 2000 3000 4000 5000
column pipe. In some cases, an air line is permanently
Pump capacity (L/min)
installed to measure the pumping lift. This measurement
requires access to the inside of the well casing; without Figure 4. Comparison of pump test data with manufacturer’s
such access, the pump test cannot be performed. total head-capacity curve.
The pump capacity is measured with a flow meter
installed in the discharge pipe. To operate properly, a and the operating point, whereas the repaired pump
flow meter requires a straight section of pipe eight to ten showed only a slight difference. However, the efficiency of
pipe diameters long immediately upstream from the meter the repaired pump suggests that the pump is mismatched
to prevent or reduce turbulence in the water. Too much to its operating conditions and probably should be replaced
turbulence from valves, bends, or elbows will prevent by a matched pump.
accurate flow measurements. It is also recommended that Adjusting the impellers can partially restore the
a straight section of pipe two pipe diameters long be capacity and total head of deep-well turbine pumps using
installed immediately downstream from the flow meter. semi-open impellers that are slightly or moderately worn.
The input power of an electric motor or engine must The adjustment consists of slightly lowering the pump
be determined to calculate the overall pumping plant shaft to which the impellers are connected by turning the
efficiency. Input horsepower of electric motors can be easily nut at the top of the motor or gear head. The adjustment
calculated. In some cases, the power meter will provide this must also account for line shaft elongation that occurs
information. Determining the input horsepower of engines when the pump is operating.
is much more complicated and will require measuring the
fuel flow rate to the engine.
The overall pumping plant efficiency can be calculated BIBLIOGRAPHY
by the following equation:
1. Hanson, B.R. (2002). Improving pumping plant efficiency does
not always save energy. California Agriculture 56(4): 123–127.
Eo = (Q × H)/(4634 × mhp)
2. Hanson, B.R. (2000). Irrigation Pumping Plants. University
of California Division of Agriculture and Natural Resources
where Eo is the overall pumping plant efficiency, Q is the Publication 3377.
capacity (L/min), H is the total head (m), and mhp is the
metric input horsepower.
The pump test data should be compared with MICROBIAL QUALITY OF RECLAIMED
the manufacturer’s performance curves to determine IRRIGATION: INTERNATIONAL PERSPECTIVE
how the pump’s actual performance compares with
the manufacturer’s recommended performance. This ABSAR ALUM
comparison step is important because it is possible to Arizona State University
have a pumping plant measuring low efficiency but Tempe, Arizona
nonetheless operating properly. Repairing such a pump
would be of no benefit. If the operating point is close to the
head/capacity performance curve, the pump is operating Wastewater reuse has been practiced in different forms
properly, regardless of the overall efficiency calculated throughout the history of human civilization. Previously,
from the test data. If a significant difference exists between sewage had been dumped in rivers and lakes or used
the performance curve and the operating point, the pump locally to irrigate crops. Explosive population growth and
may be worn and should be repaired or replaced. diminishing fresh water resources around the globe have
The pump test data shown in Fig. 4 illustrate the now made wastewater a valuable resource, especially in
evaluation process. Two pump tests were conducted—the arid and semiarid areas of the world. The introduction
first when the pump was worn, with an overall efficiency of modern wastewater treatment processes has given
of 39% and the second after the pump was repaired, with new dimensions to water reuse. Reclaimed waters are
an overall efficiency of 50%. The worn pump showed a commonly used for landscape or recreational irrigation
significant difference between the manufacturer’s curve (parks, golf courses, athletic fields, school yards, highway
668 MICROBIAL QUALITY OF RECLAIMED IRRIGATION: INTERNATIONAL PERSPECTIVE

medians, and public and private lands) and agricultural
crop irrigation (field crops, vegetables, orchids, and tree
plantations).
Human feces, which may have microbial constituents as
high as 10–30%, are one of the major sources of infectious
and noninfectious agents found in municipal wastewa-
ter. Thus, wastewater can contain high concentrations
of excreta-related microorganisms. The principal microor-
ganisms (infectious agents) that may be present in raw
sewage can be classified into three broad groups: bac-
teria, parasites (protozoa and helminths), and viruses.
During an outbreak, pathogen numbers in local sewage
go up. More than 30 excreta-related diseases of pub-
lic health importance have been documented, and many
of these are of specific importance in wastewater reuse
schemes. Table 1 summarizes the major infectious agents
potentially present in raw domestic wastewater and their
respective diseases.
TYPES OF MICROORGANISMS
Bacteria
Bacteria are microscopic organisms, ranging from approx-
imately 1 to 6 µm in length. Human and animal intestinal
tracts are colonized by different types of bacteria, which
are routinely shed in the feces. The number of bacteria
in feces can be in the range of 108 –1011 per gram of
fecal material, and these bacteria end up in municipal
wastewater (1). Coliforms, such as Escherichia coli, and
Enterobacter constitute a small group of intestinal bac-
teria, which has historically been used as an indication
that an environment is contaminated by human excreta.
In addition to coliforms, some pathogenic bacteria, such
as Salmonella, Shigella, Vibrio, Mycobacterium, Clostrid-
ium, Leptospira, and Yersinia, can also colonize part of a
gastrointestinal tract, resulting in gastroenteritis.
A number of opportunistic bacterial pathogens can also
be found in waste and reclaimed water. These oppor-
tunistic pathogens include Pseudomonas, Streptococcus,
Flavobacterium, and Aeromonas species (2).
Parasites
Parasites are a very diverse group of microorganisms;
they can be further categorized as protozoan, metazoan,
or helminth. Parasites often have a complex life cycle
sometime requiring a stage in intermediate hosts. The
environmental stage of parasites found in wastewater is
normally called cyst/oocyst (protozoans) or egg (helminth).
In general, parasite cysts or eggs are larger than
bacteria. Parasites are present in the feces of infected
persons; however, healthy carriers may also excrete them.
Cysts do not reproduce in the environment but can
survive in soil for months or even years, depending on
environmental conditions.
Entamoeba histolytica, the most dangerous protozoan
parasite, occurs worldwide, although the incidence in the

Table 1. Pathogenic Microorganisms Associated with Wastewater

Group of
microorganisms Examples Disease

Bacteria Campylobacter jejuni Gastroenteritis (campylobacteriosis)

Clostridium Gastroenteritis
Enteropathogenic Gastroenteritis
Escherichia coli
Leptospira
Mycobacterium
Salmonella typhi Typhoid fever
Salmonella sp. Gastroenteritis (salmonellosis)
Shigella Bacillary dysentery
Vibrio cholera Cholera
Yersinia sp. Gastroenteritis (yersiniosis)
Viruses Adenoviruses Gastroenteritis (conjunctivitis, Respiratory)
Caliciviruses Gastroenteritis
Coxackie viruses Flu-like symptoms, gastroenteritis
Echoviruses Flu-like symptoms, aseptic meningitis
Hepatitis A virus Infectious hepatitis
Hepatitis E virus Infectious hepatitis
Polioviruses Poliomyelitis
Reoviruses Respiratory diseases
Rotavirus Acute gastroenteritis
Protozoan Balantidium Balantidiasis (gastroenteritis)
Cryptosporidium Cryptosporidiosis (diarrhea)
Entamoeba coli Diarrhea, ulceration
Entamoeba histolytica Amebiasis (amoebic dysentery)
Giardia lamblia Giardiasis (diarrhea)
Helminth Ascaris sp. Ascariasis (roundworm infection)
Necator americanus Ancylostomiasis (hookworm infection)
Taenia sp. Taeniasis (tapeworm infection)
Trichuis trichuria Trichuriasis (whipworm infection)
 MICROBIAL QUALITY OF RECLAIMED IRRIGATION: INTERNATIONAL PERSPECTIVE
United States is not very well documented. Cryptosporid-
ium parvum oocysts are the most important biological
water contaminants in the United States (3).
Several helminthic parasites may be found in wastew-
ater. Ascaris lumbricoides (roundworm) is one of the most
important human parasites. It is estimated that approx-
imately 25% of the world’s population is infected by this
nematode (4).
Viruses
Viruses are smaller than bacterial disease agents and
range in size from approximately 20 to 200 nm. They are
obligate intracellular parasites that can multiply only
within a host cell. Viruses that can replicate in the
gastrointestinal tract of humans or animals are referred
to as enteric viruses. The presence of enteric viruses
in sewage is a matter of concern because they can be
transmitted through the fecal–oral route. Morethan 120
different human enteric viruses have been reported to
date. An infected person can excrete as many as 106 –1010
plaque-forming units (pfu) of enteric viruses per gram of
fecal material.
The most important human enteric viruses are
enteroviruses (polio, echo, and Coxsackie), caliciviruses,
rotaviruses, reoviruses, parvoviruses, and adenoviruses.
Caliciviruses are the most frequently reported and
documented that are transmitted by contaminated water.
Even though the etiologic agent cannot be successfully
cultivated in the laboratory, there is epidemiological
evidence that caliciviruses are the most common cause
of viral gastroenteritis. Hepatitis A, the virus that
causes infectious hepatitis, is one of the most resistant
to environmental stresses (5). Rotaviruses are the most
infectious of all enteric viruses (6) and thus can be
considered a high risk to health if present in wastewater.
MECHANISM OF DISEASE TRANSMISSION
There are several ways in which an individual can acquire
disease from wastewater use, and those can be broadly
categorized into two modes; direct ingestion and indirect
ingestion. Humans can acquire an etiologic agent either
directly by contact, ingestion, or inhalation of reclaimed
irrigation water or indirectly by contact with the objects
previously contaminated by the reclaimed water. Follow-
ing are some of the conditions for a successful infection or
disease from exposure to reclaimed wastewater.
• Etiologic Agent: The disease-causing agent (bacteria,
virus or a parasite) must be present in wastewater
and hence, in the wastewater from that community.
• Survival of Etiologic Agent: The disease-causing
agent must survive all wastewater treatment pro-
cesses to which it is exposed.
• Exposure: The disease competent person must come
in contact, directly or indirectly, with reclaimed water
carrying the disease agent.
• Infectivity and Infective Dose: The etiologic agent
must be infectious, moreover, the disease-causing
669
agent must be sufficiently numerous at the time of
contact to cause illness.
The exposure to an etiologic agent does not always result
in disease. Disease results from a series of complex
interactions between the host and the infectious agent,
which can be defined in terms of the pathogenicity
and virulence of the disease-causing agent and the
relative susceptibility of the host. The susceptibility of
the host depends on the general health of the subject
and the specific pathogen. Age and the nutritional and
immunological status of subjects are some of the other
factors that determine the outcome of an infection.
MONITORING MICROORGANISMS
The detection, isolation, and identification of all microor-
ganisms in a wastewater sample are very difficult, time-
consuming, and cumbersome. The detection efficacy and
sensitivity are influenced by various physicochemical
parameters of the water sample, which further undermine
the significance of such a strategy. To avoid the necessity
of undertaking such a huge task, which has no equivalent
value, indicator microorganisms are used to estimate the
presence of fecal pathogens in water samples. An indicator
should be present in equal or higher numbers and be more
resistant than pathogens. Historically, fecal coliforms, in
particular, E. coli, have been used as indicators of fecal
contamination (7).
REMOVAL OF MICROORGANISMS BY WASTEWATER
TREATMENT PROCESSES
The purpose of most advanced treatment processes
is to remove either inorganic or organic constituents.
Therefore, the removal of biological contaminants by
these processes is only incidental in many cases and,
generally is not too great. Microbial removal efficiencies of
some treatment processes are given Table 2. Conventional
biological treatment processes (trickling filters, activated
sludge, and oxidation ponds) reduce the quantities of
biological organisms in raw or settled sewage but do not
eliminate them.
The most important treatment process for pathogen
destruction is disinfection. In the United States, the
most common disinfectant for both water and wastewater
is chlorine. The efficiency of disinfection with chlorine
depends on the water temperature, pH, length of contact,
degree of mixing, presence of interfering substances,
concentration and form of the chlorinated species, and
nature and concentration of the organisms to be destroyed.
TYPES OF RECLAIMED IRRIGATION PRACTICES
Reclaimed irrigation can be practiced in the following five
different ways:
• Flood irrigation, which wets almost all the
land surface;
670 MICROBIAL QUALITY OF RECLAIMED IRRIGATION: INTERNATIONAL PERSPECTIVE
Table 2. Concentration of Various Microorganisms in Raw and Treated Wastewatera
Number per 100
Microorganisms Raw Sewage
Fecal coliform (MPN) 109
Salmonella (MPN) 8 × 103
Enteric viruses (PFU) 5 × 104
Helminth ova 8 × 102
Giardia cyst 104
a
Reference 8.
• Furrow irrigation, which wets only part of the
ground surface;
• Sprinkler irrigation, in which the soil is wetted in
much the same way as by rain;
• Surface drip irrigation, in which water is applied at
each individual plant at an adjustable rate;
• Subsurface drip irrigation, in which the topsoil
remains relatively dry but the subsoil is saturated.
CONTROL OF HUMAN EXPOSURE
Four groups of people are at potential risk from
reclaimed irrigation:
• agricultural field workers and their families;
• crop handlers;
• consumers (of crops, meat, and milk);
• population living near the affected fields; and
• recreational area visitors, players, etc.
GUIDELINES AND TREATMENT PRACTICES FOR
WASTEWATER REUSE IN AGRICULTURE
All reuse guidelines or criteria have a single basic
objective, to ensure the safety of public health. Wastewater
is known to harbor a variety of different microorganisms,
influenced by various environmental factors, so it is
possible to address different aspects of this issue.
Reuse guidelines can be based on the level of indicator
microorganisms, epidemiological and/or risk models, and
treatment practices.
Early guidelines for wastewater reuse in agriculture
were concerned only with the pathogen content of
wastewater. A World Health Organization (WHO) meeting
of experts on the reuse of effluents in 1971 reviewed the
information concerning health risks and concluded that
only a limited health risk would result from unrestricted
crop irrigation with wastewater containing no more than
100 coliforms per 100 mL. Following the recommendation
of the WHO meeting in 1971, new epidemiological evidence
was accumulated, and earlier studies were reevaluated. A
meeting in Engelberg in 1985 of public health experts,
environmental scientists, and epidemiologists reviewed
the work that had been done and concluded that the risk of
infection was lower than previously thought, particularly,
from bacterial and viral pathogens (9).
Reduction in Number of Microbes (log10 )
Primary Secondary Tertiary
Effluent Effluent Effluent
2 1 6
1 2 1
0.5 1 5
1 2 1
0.5 0.5 3
The WHO guidelines for wastewater reuse, which were
proposed by the Engelberg meeting in 1985 and reaffirmed
in 1989, were aimed at providing a high degree of health
safety to workers and consumers, but at the same time to
be realistic in the light of resource constraints and limited
technology in many countries.
In the United States the State of California has a
long history of reclaimed water regulation. The relatively
stringent California standards are not economically or
technologically feasible for developing countries.
INTERNATIONAL STANDARDS
The WHO water reuse guidelines (Table 3) have served
as a baseline for the development of reclaimed irrigation
standards throughout the world generally and in develop-
ing countries particularly. On the other hand, California
criteria, which represent the higher end, are extensively
followed by developed countries. Different countries have
adapted both standards to develop their regional microbial
standards for reclaimed irrigation (Table 4).
Italy became the first European country to regulate
reclaimed irrigation when it adopted wastewater reuse
guidelines in 1977 under its 1976 national Water Law.
Although the Italian guidelines were published 1 year
earlier than California guidelines, both essentially follow
the same approach. The region of Sicily has adopted
different regulations, which are much closer to the WHO
approach. As an added precaution, this regulation forbids
irrigating vegetables that are eaten raw with any type of
reclaimed wastewater.
Cyprus produces approximately 25 Mm3 of urban
effluent every year. Government has planned to expand
irrigated agriculture by introducing reclaimed irrigation.
Provisional standards for reclaimed water irrigation were
introduced to regulate reclaimed irrigation, which take
into account the special ecological and environmental
conditions of Cyprus. These standards, which are stricter
than the WHO guidelines, ensure the best possible
application of the effluent for irrigation.
In France, reclaimed irrigation of crops is practiced
mostly around big cities as the easiest way to treat and
dispose of wastewater. In the early 1990s, the growing
demand for irrigation water along with periodic droughts
resulted in increased interest in water reuse. French
health authorities [the Conseil Supérieur díHygiène
Publique de France (CSHPF)] issued the ‘‘health guide-
lines for reuse, after treatment, of wastewater for crop
 Table 3. Microbiological Quality Guidelines for Wastewater Use in Agriculturea by WHO
Intestinal Nematodes Wastewater Treatment
(Mean Arithmetic Faecal Coliforms Expected to Achieve
Number of Viable (Geometric Mean, the Required
Category Reuse Conditions Exposed Group Eggs per liter) Number per 100 mL) Microbiological Quality

A Irrigation of crops likely Workers, consumers, 1 1000 A series of stabilization

to be eaten uncooked, public ponds designed to achieve
sports fields, public the microbiological quality
parks indicated, or equivalent
treatment
B Irrigation of cereal Workers 1 No standard Retention in stabilization
crops, industrial recommended ponds for 8–10 days or
crops, fodder crops, equivalent helminth and
pasture and trees fecal coliform removal
C Localized irrigation of None Not applicable Not applicable Pretreatment as required by
crops in category B if irrigation technology, but
exposure of workers not less than primary
and the public does sedimentation
not occur
a
Reference 10.

Table 4. Guidelines and Criteria for Wastewater Reuse in Irrigation in Various Countriesa
Country Crop Type Treatment Required Microbial Criteria (max.)

Australia Residential use, municipal irrigation,
unrestricted crop irrigation
Ornamental ponds with public access,
restricted crop irrigation, irrigation
of pasture and fodder crops for dairy
animals, fire fighting
Municipal irrigation with restricted
crop irrigation, irrigation of pasture
and fodder crops for grazing animals
Irrigation for turf production,
silviculture, nonfood chain
aquaculture
Israel Cotton, sugar beet, dry fodder seeds,
forest irrigation
Green fodder, nuts, citrus, olives,
peanuts, bananas, almonds, etc.
Deciduous fruits, conserved
vegetables, cooked and peeled
vegetables
Unrestricted crops, including
vegetables eaten raw, parks, and
lawns
Japan Landscape irrigation
Kuwait Fodder, food crops not eaten raw,
forest land
Food crops eaten raw
Saudi Arabia All irrigation purposes (unrestricted)
All irrigation purposes
South Africa Irrigation of dry fodder crops, seed
crops, trees, non-recreational parks,
nurseries (restricted access)
Food crops not eaten raw, cut flowers,
orchards and vineyards, pastures,
parks
Pasture for milking animals
Food crops eaten raw, lawns, nurseries
(unrestricted access)
a
Reference 11.
Secondary, tertiary, filtration and <10 thermotolerant
disinfection coliforms/100 mL
Secondary and disinfection <100 thermotolerant
coliforms/100 mL
Primary sedimentation and lagooning <1000 thermotolerant
or secondary plus disinfection (if coliforms/100 mL
required)
Primary sedimentation and lagooning <10,000 thermotolerant
or secondary coliforms/100 mL
None None
None None
Disinfection with chlorination 250 coliforms/100 mL
(60 minutes contact time)
Sand filtration and disinfection with 12 coliforms/100 mL (80%)
chlorination (60 minutes contact
time)
2.2 coliforms/100 mL (50%)
Disinfection (>0.4 mg/L residual 0.0 E. coli/100 mL
chlorine)
Advanced water treatment (1 mg/L 10,000 coliforms/100 mL
residual chlorine after 12 h@20 ◦ C)
Advanced water treatment (1 mg/L 100 coliforms/100 mL
residual chlorine after 12 h@20 ◦ C)
Advanced water treatment 2.2 coliforms/100 mL
Treatment plants and waste settling <1 intestinal nematode
ponds egg/100 mL
Primary and secondary <1000 fecal coliforms/100 mL
Primary, secondary, and tertiary; <1000 fecal coliforms/100 mL
oxidation pond system
Primary, secondary, and tertiary 0.0 fecal coliforms/100 mL
Advanced (general) drinking water Drinking water standards
standards
671
672 MICROBIAL QUALITY OF RECLAIMED IRRIGATION: INTERNATIONAL PERSPECTIVE

and green spaces irrigation’’(12). The guidelines stipulate Table 6. Crops Regulated by 1991 Mexican Reuse
maintaining of extensive water quality data. Wastewa- Guidelines
ter reuse authorizations are granted on a case-by-case Salad Vegetables Vegetables Herbs Fruits
basis after review of a highly detailed dossier of French
reclaimed irrigation regulations based on WHO guidelines Beet root Beet Coriander Blackberries
but complement them with strict rules of application such Celery Broccoli Epazote Melon
as a minimum distance required between irrigation sites Cucumber Cabbage Mint Strawberries
Lettuce Carrot Parsley Sweet turnip
and residential areas or roadways.
Radish Cauliflower Watermelon
In Mexico, wastewater is extensively used to irrigate Red tomato Courgette
crops. Because a very significant part of Mexican farm Watercress Garlic
produce is sold in U.S. markets, Mexico has developed very Green tomato
comprehensive reclaimed irrigation regulations that are Mushrooms
stricter than WHO suggested standards (Tables 5 and 6). Onion
Israel has developed very comprehensive reclaimed Papalo
water usage regulations because reclaimed water con- Spinach
stitutes a significant part of its national water budget. Wild greens
About 92% of the total wastewater is collected, and 72%
is reclaimed for various purposes. Almost 42% of total
wastewater is reused for irrigation (13). The Israel Min-
istry of Health has established criteria for reclaimed
irrigation and water reuse that are stricter than WHO raw wastewater is directly reused, a widespread practice
standards. According to Israeli standards, settled efflu- in many developing countries, the WHO guidelines are
ent can be used on industrial and fodder crops, pastures already a major step forward.
and hay, vegetables eaten cooked, fruit trees, ornamen- However, other requirements also seem to be necessary
tal plants, or seed plants. Irrigation should be stopped to complement WHO guidelines. Based on an extensive
1 month before the harvest of apples, pears, and plums; analysis of existing guidelines worldwide, the need
broccoli and cauliflower when furrows are irrigated, and for developing health-related chemical criteria for land
tomatoes used for canning if furrows are irrigated. application of reclaimed wastewater has also been
WHO guidelines have been debated for their effec- reported by the WHO (14). This, added to the fact
tiveness to protect public health. A large part of the that the controversy between the ‘WHO camp’ and the
answer probably lies in the treatment requirements of ‘California camp’ has now been raging for a number
limit values. In spite of their safety, the stringency of of years means that the time has come to update the
the California standards is acting as a barrier to their existing guidelines on an international basis. The door
widespread adoption worldwide. The WHO point of view is therefore now open worldwide for a ‘third approach’
is often criticized as being too lax, but the California integrating the epidemiological knowledge base generated
approach is too technologically intensive and expensive since 1989 and the latest technological developments in
for developing countries. One must realize that where wastewater treatment.

Table 5. Conditions for Irrigation of Restricted Crops Under 1991 Mexican Guidelines
Minimum Time Between
Type of Type of the Last Irrigation
Irrigation Wastewatera and the Harvest (days) Crops/Banned Crops

Flood 1 20 All crops listed in Table 6, except garlic, cucumber, sweet turnip, melon, and
watermelon
2 20 All crops listed in Table 6, except melon and watermelon
3 20 All crops listed in Table 6
4 20 All salad crops, vegetables, and fruits in general
Furrow 1 15 All crops listed in Table 6, except garlic, cucumber, sweet turnip, melon,
watermelon, and green tomato
20 No restrictions
2 20 All crops listed in Table 6, except garlic, cucumber, sweet turnip, melon,
watermelon, and green tomato
3 20 All crops listed in Table 6, except melon and watermelon
4 20 All salad crops, vegetables, and fruits in general
Spray 1 20 All crops listed in Table 6, except garlic, cucumber, sweet turnip, melon and
watermelon
2,3,4 20 All salad crops, vegetables and fruits in general
a
Wastewater Type 1: <1000 total coliforms/100 mL and 0 helminth eggs/liter;
Wastewater Type 2: 1–1000 fecal coliforms/100 mL and ≤1 helminth eggs/liter;
Wastewater Type 3: 1001–100,000 fecal coliforms/100 mL, and no restriction on helminth eggs;
Wastewater Type 4: >100,000 fecal coliforms/100 mL and no restriction on helminth eggs.

BIBLIOGRAPHY
1. Talaro, K.P. and Talaro, A. (1999). Foundations in Microbiol-
ogy. McGraw-Hill, New York, pp. 396–397.
2. Ashbolt, N.J. et al. (1995). The identification of human
health significance of environmental aeromonads. Water Sci.
Technol. 31: 263–270.
3. Rose, J.B., Lisle, J.T., and LeChevallier, M. (1997). Water-
borne cryptosporidiosis: Incidence, outbreaks, and treat-
ment strategies. In: Cryptosporidium and Cryptosporidiosis.
R. Fayer (Ed.). CRC Press, Boca Raton, FL, pp. 93–110.
4. Ellis, K.V., Rodrigues, P.C.C., and Gomez, C.L. (1993). Para-
site ova and cysts in waste stabilization ponds. Water Res. 27:
1455–1460.
5. Alum, A. (2001). Control of Viral Contamination of Reclaimed
Irrigated Vegetables by Drip Irrigation. Ph.D. Dissertation,
University of Arizona, Tucson, AZ.
6. Gerba, C.P., Rose, J.B., Haas, C.N., and Crabtree, K.D.
(1996). Waterborne rotavirus: a risk assessment. Water Res.
30: 2929–2940.
7. American Public Health Association. (1989). Standard
Methods for the Examination of Water and Wastewater.
APHA, AWWA, WEF, Washington, DC.
8. National Research Council. (1996). Use of Reclaimed Water
and Sludge in Food Crop Production. National Academy
Press, Washington, DC, p. 91.
9. The Engelberg Report. (1985). Health Aspects of Wastewater
and Excreta Use in Agriculture and Aquaculture. Int. Ref.
Center Waste Disposal (ICRWD) News No. 23, pp. 11–18.
10. World Health Organization. (1989). Health Guidelines for the
Use of Wastewater in Agriculture and Aquaculture. Technical
Report Series No. 776. World Health Organization, Geneva.
p. 74.
11. SAEPA. (1997). South Australian Reclaimed Water Guide-
lines. Draft 10.1.97. South Australian Environmental Protec-
tion Authority.
12. CSHPF. (1991). Recommandations Sanitaires Concernanat
L’Utilisation, Apre’s puration, des Eaux Residuaires Urbaines
Pour L’irrigation des Cultures etdes Espaces Verts. Circulaire
DGS/SD1.D./91/N 51. Conseil Supérieur díHygiène Publique
de France, Paris.
13. Friedler, E. (2001). Water reuse—an integral part of water
resources management: Israel as a case study. Water Policy
3: 29–39.
14. Chang, A.C., Page, A.L., and Asano, T. (1995). Developing
Human Health Related Chemical Guideline for Reclaimed
Wastewater and Sewage Sludge Applications in Agriculture.
World Health Organization, Geneva.
READING LIST
Asano, T. and Levine, A.D. (1996). Wastewater reclamation,
recycling and reuse: past present and future. Water Sci.
Technol. 33: 1–14.
US EPA. (1992). In U.S. EPA, Office of Water and Wastewater
and compliance (Ed.). Guidelines for water reuse. US EPA,
Washington, DC.
World Health Organization. (1973). Reuse of Effluents: Methods
of Wastewater Treatment and Health Safeguards. Technical
Report Series No. 517. A report of Meeting of Experts,
WHO, Geneva.
SOIL SALINITY 673
SOIL SALINITY
DENNIS L. CORWIN
Salinity Laboratory
Riverside, California
DEFINITION, SOURCES, EFFECTS, AND GLOBAL IMPACTS
Soil salinity refers to the presence of major dissolved
inorganic solutes in the soil solution (i.e., aqueous liquid
phase of the soil and its solutes), which consist of soluble
and readily dissolvable salts, including charged species
(e.g., Na+ , K+ , Mg2+ , Ca2+ , Cl− , HCO3 − , NO3 − , SO4 2− ,
and CO3 2− ), nonionic solutes, and ions that combine to
form ion pairs. The primary source of salts in soil and
water is the geochemical weathering of rocks from the
earth’s upper strata, with atmospheric deposition and
anthropogenic activities serving as secondary sources.
The predominant mechanism causing the accumulation
of salt in the root zone of agricultural soils is loss of water
through evapotranspiration (i.e., combined processes of
evaporation from the soil surface and plant transpiration),
which selectively removes water, leaving salts behind.
The accumulation of soil salinity can result in reduced
plant growth, reduced yields, and, in severe cases, crop
failure. Salinity limits plant water uptake by reducing the
osmotic potential, making it more difficult for the plant to
extract water. Salinity may also cause specific ion toxicity
or upset the nutritional balance of plants. In addition,
the salt composition of the soil solution influences the
composition of cations on the exchange complex of soil
particles, which influences soil permeability and tilth.
Irrigated agriculture, which accounts for 35–40% of the
world’s total food and fiber, is adversely affected by soil
salinity on roughly half of all irrigated soils (totaling about
250 million ha), with over 20 million ha severely effected
by salinity worldwide (1). As a result of the potential
detrimental impacts of soil salinity accumulation, it is a
crucial soil chemical property that is routinely measured
and monitored.
LABORATORY METHODS OF SOIL SALINITY
MEASUREMENT
Although a variety of instrumentation has been developed
for the measurement of soil salinity in the field, the most
common technique is still laboratory analysis of aqueous
extracts of soil samples. In the laboratory, soil salinity
is commonly determined from the measurement of the
electrical conductivity (EC) of soil solution extracts. The
current-carrying capacity of soil solution is proportional
to the concentration of ions in the solution. Soil salinity
is quantified in terms of the total concentration of the
soluble salts as measured by the EC of the soil solution in
dS m−1 (2).
To determine EC, measurements are made in a cell
containing two electrodes of constant geometry and
distance of separation (3). The soil solution is placed
between the two electrodes. An electrical potential is
imposed across the electrodes, and the resistance of
the solution between the electrodes is measured. The
measured conductance is a consequence of the solution’s
674 SOIL SALINITY

salt concentration and the electrode geometry whose

effects are embodied in a cell constant. At constant
potential, the current is inversely proportional to the Manifold
solution’s resistance:

ECT = k/RT (1) Vacuum

where ECT is the electrical conductivity of the solution

in dS m−1 at temperature T ( ◦ C), k is the cell constant, Solution
and RT is the measured resistance at temperature T. One sample bottle
dS m−1 is equivalent to one mS cm−1 and one mmho cm−1 ,
where mmho cm−1 are the obsolete units of EC.
Partitioning of solutes over the three soil phases (i.e., Ceramic cup
gas, liquid, solid) is influenced by the soil/water ratio
at which the extract is made, so the ratio must be
standardized to obtain results that can be applied and
interpreted universally. Customarily, soil salinity has Suction cup extractors
been defined in terms of laboratory measurements of the Figure 1. Diagram of a basic suction cup extractor setup for
EC of the saturation extract (ECe ) because it is impractical sampling the soil solution. Courtesy of Corwin (6).
for routine purposes to extract soil water from samples at
typical field water contents. One widely used technique is
to obtain an extract by vacuum filtration of a saturated Cable Release pin
soil paste made with distilled water (4). Commonly used
extract ratios other than a saturated soil paste are 1:1,
Electrolytic
1:2, and 1:5 soil/water mixtures. However, extracts at element
these ratios only provide relative salinity because the Spring
..
soils are adjusted to unnaturally high water contents not ..
..
found in the field. Soil salinity can also be determined Platinum ..
electrodes ..
.. .
from the measurement of the EC of the soil solution at . Inner
some defined field water content (ECw ). An example would assembly
be the EC at field capacity, which represents the water
content of soil 2–3 days after irrigation when free drainage Thermistor
is negligible. Theoretically, ECw is the best index of soil Housing
salinity because it is the salinity actually experienced by Figure 2. Schematic of a porous-matrix salinity sensor. Courtesy
the plant root. Nevertheless, ECw has not been widely used of Corwin (7).
for two reasons: (a) it varies over the irrigation cycle as the
soil water content changes and (b) methods for obtaining Equation 4 is an example of a theoretical approach based
soil solution samples are too labor and cost intensive to be on Kohlrauch’s Law of independent migration of ions,
practical for field-scale applications (5). where each ion contributes to the current-carrying ability
Soil solution extracts can be obtained by various of an electrolyte solution:
means. For disturbed samples, soil solution can be
obtained in the laboratory by displacement, compaction,   ci (λi − β √ci
EC = ECi = 0
(4)
centrifugation, molecular adsorption, and vacuum- or 1000
i i
pressure-extraction methods. For undisturbed samples,
ECw can be determined either in the laboratory on a soil where EC is the specific conductance (dS m−1 ), ECi is the
solution sample collected with a soil-solution extractor (see ionic specific conductance (dS m−1 ), ci is the concentration
Fig. 1) or directly in the field using in situ, imbibing-type of the ith ion (mmolc L−1 ), λi 0 is the ionic equivalent
porous-matrix salinity sensors (see Fig. 2). conductance at infinite dilution (cm2 S molc −1 ), and β is an
Electrolytic conductivity increases at a rate of approxi- empirical interactive parameter (9). Equation 5 shows an
mately 1.9% per degree centigrade increase in temperature empirical equation developed by Marion and Babcock (10):
in the range of 15–35 ◦ C. Customarily, EC is expressed at
a reference temperature of 25 ◦ C for purposes of compar- log TSS = 0.990 + 1.055 log EC (r2 = 0.993) (5)
ison. The EC (i.e., ECe or ECw ) measured at a particular
temperature T ( ◦ C), ECT (dS m−1 ), can be adjusted to a where TSS is the total soluble salt concentration
reference EC at 25 ◦ C, EC25 (8): (mmolc L−1 ).

EC25 = ft · ECT (2) FIELD METHODS OF SOIL SALINITY MEASUREMENT

fT = 0.4470 + 1.4034 exp(−T/26.815) (3)
where fT is a temperature conversion factor.
Theoretical and empirical approaches are available to
predict the EC of a solution from its solute composition (3).
Serious doubts exist about the ability of soil core
samples, soil solution extractors, and porous-matrix
salinity sensors (also known as soil salinity sensors)
to provide representative soil water samples. Small

soil core samples, extractors, and salt sensors do not
adequately integrate spatial variability because of their
small sphere of measurement; consequently, Biggar and
Nielsen (11) suggested that these are ‘‘point samples’’ that
can provide qualitative measurement of soil solutions, but
not quantitative measurements unless their field-scale
variability is established.
As salinity is a highly spatially variable soil property,
the use of ECe to measure salinity at field scales is
generally impractical because of the need for hundreds and
even thousands of soil samples. The use of ECe to measure
salinity at field scales is only practical when soil sampling
is directed using correlated spatial information. Two
potential sources of correlated spatial information used
to direct where soil samples should be taken to measure
ECe are: (a) visual crop observations and (b) geospatial
measurements of apparent soil electrical conductivity
(ECa ) using electrical resistivity (ER), electromagnetic
induction (EMI), or time-domain reflectometry (TDR).
These sources of spatial information are used to direct soil
sampling to characterize field-scale soil salinity variation.
Visual Crop Observation
Visual crop observation is a quick and economical method,
but salinity development is detected after crop damage
has occurred. For this reason, visual observations are
unrealistic because crop yields are reduced to obtain soil
salinity information. However, remote imagery potentially
represents a quantitative approach to visual observation.
Remote imagery may offer a potential for early detection
of the onset of salinity damage to plants.
Apparent Soil Electrical Conductivity
As a result of the time and cost of obtaining soil solution
extracts, developments in the measurement of soil EC
have shifted to the measurement of electrical conductivity
of the bulk soil, referred to as the apparent soil electrical
conductance (ECa ). Apparent soil electrical conductivity
measures the conductance through not only the soil
solution but through the solid soil particles and via
exchangeable cations that exist at the solid-liquid interface
of clay minerals. The techniques of ER, EMI, and TDR
measure ECa .
The interpretation of ECa measurements is not
trivial because of the complexity of current flow in the
bulk soil. Three pathways of current flow contribute
to the ECa measurement: (a) a liquid phase pathway
via salts contained in the soil water occupying the
large pores, (b) a solid-liquid phase pathway primarily
via exchangeable cations associated with clay minerals,
and (c) a solid pathway via soil particles that are in
direct and continuous contact with one another (5). As
a result of the three pathways of conductance, the
ECa measurement is influenced by several soil physical
and chemical properties: (a) soil salinity, (b) saturation
percentage, (c) water content, and (d) bulk density. The
saturation percentage and bulk density are both closely
associated with the clay content. Measurements of ECa as
a measure of soil salinity must be interpreted with these
influencing factors in mind. Traditionally, ECe has been
the standard measure of salinity used in all salt-tolerance
SOIL SALINITY 675
plant studies. As a result, a relation between ECa and
ECe is needed to relate ECa back to ECe , which in turn is
related to crop yield. In spite of these complicating factors,
ECa is the only viable means of spatially characterizing
soil salinity at field scales.
As a result of the ease and quickness of measuring ECa
and the ability to mobilize ECa -measurement equipment,
geospatial measurements of ECa are the most reliable
means of characterizing the spatial variability of soil salin-
ity at field scales. Geospatial measurements of ECa are not
used to directly measure soil salinity but rather to direct
soil sampling from which ECe measurements are subse-
quently made. From the geospatial ECa measurements
and a site-specific relation between ECa and ECe , a map of
soil salinity distribution across the landscape and through
the soil profile can be created. Detailed procedures and
protocols for conducting an ECa survey to characterize the
spatial variability of soil salinity with ECa -directed soil
sampling are provided by Corwin and Lesch (12).
ER methods introduce an electrical current into the
soil through current electrodes at the soil surface. The
difference in current flow potential is measured at
potential electrodes that are placed in the vicinity of
the current flow (Fig. 3). The electrode configuration is
referred to as a Wenner array when four electrodes
are equidistantly spaced in a straight line at the soil
surface. The two outer electrodes serve as the current
or transmission electrodes, and the two inner electrodes
serve as the potential or receiving electrodes. The depth
of penetration of the electrical current and the volume
of measurement increase as the interelectrode spacing,
a, increases. For a homogeneous soil, the soil volume
measured is roughly π a3 , and the depth of penetration is
roughly equal to a. Additional electrode configurations are
discussed by Burger (14).
By mounting the electrodes to ‘‘fix’’ their spacing,
considerable time for a measurement is saved. Veris
Technologies1 has developed a commercial mobile system
for measuring and mapping ECa using the principles of
ER coupled to a global positioning system (GPS). The
Veris Model 3100 consists of a pair of transmission coulter
electrodes and a pair of potential coulter electrodes to
measure the voltage drop. Two sets of arrays permit the
measurement of ECa to depths of 0–30 cm and 0–90 cm.
Resistance meter
Current Current
Potential
electrode
electrodes
electrode
C1 P1 P2 C2
a a a
Figure 3. Schematic of electrical resistivity four electrodes,
where C1 and C2 represent the current electrodes, P1 and
P2 represent the potential electrodes, and a represents the
interelectrode spacing. Courtesy of Corwin and Hendrickx (13).
1
Veris Technologies, Salina, Kansas, USA (www.veristech.com).
Product identification is provided solely for the benefit of the
reader and does not imply the endorsement of the USDA.
676 SOIL SALINITY
EMI instrumentation consists of a transmitter coil and
a receiver coil (Fig. 4). The transmitter coil located at
one end of the instrument induces circular eddy-current
loops in the soil. The magnitude of these loops is directly
proportional to the EC in the vicinity of that loop. Each
current loop generates a secondary electromagnetic field
that is proportional to the value of the current flowing
within the loop. A fraction of the secondary induced
electromagnetic field from each loop is intercepted by the
receiver coil at the opposite end of the instrument, and
the sum of these signals is amplified and formed into an
output voltage, which is related to a depth-weighted ECa .
The amplitude and phase of the secondary field will differ
from those of the primary field as a result of soil properties
(e.g., clay content, water content, salinity), spacing and
orientation of the coils, frequency, and distance from the
soil surface (15).
The two most commonly used EMI conductivity meters
in soil science are the Geonics2 EM-31 and EM-38.
The EM-38 has had considerably greater application for
agricultural purposes because the depth of measurement
corresponds roughly to the root zone (i.e., 1.5 m), when the
instrument is placed in the vertical coil configuration.
In the horizontal coil configuration, the depth of the
measurement is 0.75–1.0 m. The operation of the EM-38
equipment is discussed in Hendrickx and Kachanoski (15).
Mobile EM equipment developed at the USDA-ARS
Salinity Laboratory (16) is available for appraisal of soil
salinity and other soil properties (e.g., water content and
clay content) using an EM-38 or a dual-dipole EM-38 unit.
The dual-dipole EM-38 conductivity meter simultaneously
records data in both dipole orientations (horizontal and
vertical). The mobile EM equipment is suited for the
detailed mapping of ECa and correlated soil properties
at specified depth intervals through the root zone. The
advantage of the mobile EM equipment is that it is
noninvasive, so it can be used in dry, frozen, or stony
soils that are not easily measured with the invasive
‘‘fixed-array’’ approach, which requires good electrode-soil
contact. The disadvantage of the EM approach would be
that the ECa is a depth-weighted value, which is nonlinear
with depth (17).
ER and EM are both well suited for field-scale
applications because their volumes of measurement are
large, which reduces the influence of local-scale variability.
Transmitter Receiver
coil coil
Current loops
Induced current flow in soil
Figure 4. Transmitter coil, receiver coil, and induced current
flow for EMI.
2
Geonics Limited, Mississauga, Ontario, Canada. Product identi-
fication is provided solely for the benefit of the reader and does
not imply the endorsement of the USDA.
However, ER is an invasive technique that requires good
contact between the soil and four electrodes inserted
into the soil; consequently, it produces less reliable
measurements in dry, frozen, or stony soils than the
noninvasive EM measurement. Nevertheless, ER has
a flexibility that has proven advantageous for field
application, i.e., the depth and volume of measurement
can be easily changed by altering the spacing between
the electrodes.
Time-domain reflectometry (TDR) was initially adapted
for use in measuring water content. Later, Dalton
et al. (18) demonstrated the utility of TDR to also measure
ECa , based on the attenuation of the applied signal
voltage as it traverses the medium of interest. Although
TDR has been demonstrated to compare closely with
other accepted methods of ECa measurement, it is still
not sufficiently simple, robust, or fast enough for the
general needs of field-scale soil salinity assessment (5).
Only electrical resistivity and EM have been adapted for
the geo-referenced measurement of ECa at field scales
and larger.
PLANT TOLERANCE TO SOIL SALINITY
The primary effect of soil salinity is on plant growth.
Virtually all plant salt-tolerance studies conducted in the
past have related plant yield response to ECe , which is
another reason why ECe continues to be used as the
primary measure of soil salinity. A salt-affected soil is one
that contains enough soluble salts to hamper the growth
of the crop. The proportions of dissolved ions (primarily
Na+ , Ca2+ , Mg2+ , Cl− , SO4 2− , and HCO3− ) can vary,
and the effects on plants can vary. Generally, however,
plants respond similarly to salinity over a fairly wide
range of combinations of salt. To quantify yield response
to soil salinity, Maas and Hoffman (19) proposed a two-
piece linear response function with a tolerance plateau of
slope zero and a concentration-dependent line whose slope
indicates the yield reduction per unit increase in salinity:
Yr = 100 − b (ECe − a) (6)
where a is the salinity threshold (dS m−1 ), b is the slope
expressed in percent per dS m−1 , and ECe is the mean ECe
taken from the root zone. Values for a and b can be found
in Maas (20).
BIBLIOGRAPHY
1. Rhoades, J.D. and Loveday, J. (1990). Salinity in irrigated
agriculture. In: Irrigation of Agricultural Crops. B.A. Stewart
and D.R. Nielsen (Eds.). Agron. Monogr. No. 30, SSSA,
Madison, WI, pp. 1089–1142.
2. U.S. Salinity Laboratory Staff. (1954). Diagnosis and
Improvement of Saline and Alkali Soils. USDA Handbook
60, U.S. Government Printing Office, Washington, DC.
3. Jurinak, J.J. and Suarez, D.L. (1996). The chemistry of
salt-affected soils and waters. In: Agricultural Salinity
Assessment and Management. K.K. Tanji (Ed.). ASCE, New
York, pp. 42–63.
4. Rhoades, J.D. (1996). Salinity: electrical conductivity and
total dissolved solids. In: Methods of Soil Analysis,

Part 3—Chemical Methods. D.L. Sparks (Ed.). SSSA Book
Series No. 5, SSSA, Madison, WI, pp. 417–435.
5. Rhoades, J.D., Chanduvi, F., and Lesch, S. (1999). Soil
Salinity Assessment: Methods and Interpretation of Electrical
Conductivity Measurements. FAO Irrigation and Drainage
Paper #57, Food and Agriculture Organization of the United
Nations, Rome, Italy.
6. Corwin, D.L. (2002). Solute content and concentration—mea-
surement of solute concentration using soil water extraction:
suction cups. In: Methods of Soil Analysis, Part 4—Physical
Methods. J.H. Dane and G.C. Topp (Eds.). SSSA Book Series
No. 5, SSSA, Madison, WI, pp. 1261–1266.
7. Corwin, D.L. (2002). Solute content and concentration—mea-
surement of solute concentration using soil water extraction:
porous matrix sensors. In: Methods of Soil Analysis,
Part 4—Physical Methods. J.H. Dane and G.C. Topp (Eds.).
SSSA Book Series No. 5, SSSA, Madison, WI, pp. 1269–1273.
8. Sheets, K.R. and Hendrickx, J.M.H. (1995). Non-invasive soil
water content measurement using electromagnetic induction.
Water Resour. Res. 31: 2401–2409.
9. Harned, H.S. and Owen, B.B. (1958). The Physical Chemistry
of Electrolyte Solutions. Reinhold Publ. Corp., New York.
10. Marion, G.M. and Babcock, K.L. (1976). Predicting specific
conductance and salt concentration of dilute aqueous solution.
Soil Sci. 122: 181–187.
11. Biggar, J.W. and Nielsen, D.R. (1976). Spatial variability of
the leaching characteristics of a field soil. Water Resour. Res.
12: 78–84.
12. Corwin, D.L. and Lesch, S.M. (In press). Characterizing soil
spatial variability with apparent soil electrical conductivity:
I. Survey protocols. Comput. Electron. Agric. Agriculture.
13. Corwin, D.L. and Hendrickx, J.M.H. (2002). Solute content
and concentration—Indirect measurement of solute concen-
tration: electrical resistivity—wenner array. In: Methods
of Soil Analysis, Part 4—Physical Methods. J.H. Dane and
G.C. Topp (Eds.). SSSA Book Series No. 5, SSSA, Madison,
WI, 1282–1287.
14. Burger, H.H. (1992). Exploration Geophysics of the Shallow
Subsurface. Prentice Hall PTR, Upper Saddle River, NJ.
15. Hendrickx, J.M.H. and Kachanoski, R.G. (2002). Solute con-
tent and concentration—indirect measurement of solute con-
centration: nonintrusive electromagnetic induction. In: Meth-
ods of Soil Analysis, Part 4—Physical Methods. J.H. Dane
and G.C. Topp (Eds.). SSSA Book Series No. 5, SSSA, Madi-
son, WI, 1297–1306.
16. Rhoades, J.D. (1993). Electrical conductivity methods for
measuring and mapping soil salinity. In: Advances in
Agronomy. Vol. 49. D.L. Sparks (Ed.). Academic Press, San
Diego, CA, pp. 201–251.
17. McNeil, J.D. (1980). Electromagnetic Terrain Conductivity
Measurement at Low Induction Numbers. Tech. Note TN-6,
Geonics Limited, Ontario, Canada, pp. 1–15.
18. Dalton, F.N., Herkelrath, W.N., Rawlins, D.S., and Rhoades,
J.D. (1984). Time-domain reflectometry: simultaneous mea-
surement of soil water content and electrical conductivity
with a single probe. Science. 224: 989–990.
19. Maas, E.V. and Hoffman, G.J. (1977). Crop salt toler-
ance—current assessment. J. Irrig. Drainage Div. ASCE.
103(IR2): 115–134.
20. Maas, E.V. (1996). Crop salt tolerance. In: Agricultural
Salinity Assessment and Management. K.K. Tanji (Ed.).
ASCE, New York, pp. 262–304.
MAINTAINING SALT BALANCE ON IRRIGATED LAND 677
MAINTAINING SALT BALANCE ON IRRIGATED
LAND
JACOB W. KIJNE
Herts, United Kingdom
DEFINITION
The salt balance of an area is described by an equation that
compares the movement of salt into and out of an area.
The salt fluxes are in balance when inflow and outflow
of salt are equal. Salt moves in solution, and the water
and salt balances are therefore linked as in Eq. 1, which
describes the salt balance of an irrigated area:
ICi + RCr + GCg = PCp + S (1)
where: I = irrigation water infiltrating the root zone
R = rainfall entering the root zone
G = capillary flow from the groundwater into the
root zone
P = deep percolation from the root zone
C = salt concentration, and the subscripts referring
to irrigation, rainfall, capillary flow, and
percolation
S = the change in salt content of the soil solution
in the root zone
The water flows can be expressed in m3 /ha/day or any
other convenient set of units, such as mm/month, in which
the unit of time indicates the period over which the salt
balance is measured. The unit of salt concentration, C, is
usually mg/L or part per million (ppm). Apart from areas
close to the sea, it is reasonable to assume that Cr is
zero. It is also often assumed that Cg equals Cp , which is
reasonable for annual periods but not necessarily true for
shorter periods.
For steady-state conditions, the salt concentration
in the soil solution remains the same, and hence, S
equals zero. The ratio of the amounts of drainage water
(percolating from the root zone) and the amount of water
entering the root zone as irrigation and rainfall is called
the leaching fraction, LF. Hence:
LF = Pt/(I + R)t = P/(I + R) (2)
where t equals the time period under consideration.
The ratio of the electrical conductivity (EC) of the
drainage water to that of the applied water (A = I + R)
can be written as ECp /ECa . For steady-state conditions,
this ratio is proportional to the inverse of the leaching
fraction (1/LF). For solutions consisting mainly of chloride
salts, the proportionality constant is 1. For concentrated
solutions, the proportionality constant is less than 1. Often
irrigation water also has a proportionality constant of less
than 1, and its value depends on the LF. For example, the
proportionality constant of Colorado River water dropped
to below 1 when LF became less than about 20% (1).
The lower the value of LF, the lower the proportionality
constant. Leaching fractions of less than 20% are common
in irrigation practice. The relationship between ECp /ECa
and 1/L is also influenced by the solubility of soil lime,
which is a function of the partial pressure of CO2 .
678 MAINTAINING SALT BALANCE ON IRRIGATED LAND
PRESENCE OF SALTS
The salts carried in irrigation water come from salts that
are continuously released from rocks and soils by mineral
weathering and dissolution. Because of their low rainfall,
most arid and semiarid lands are rich in these primary
salts. The salts were deposited over time in sedimentary
rock layers and are dissolved in the waters of aquifers.
Unfavorable salt balances are common in arid lands:
The global extent of salt-affected lands is considerable.
It has been estimated that worldwide 100 million ha of
irrigated land suffer from salinity, 20% of which is severely
affected (2). Annually some 2 to 3 million ha of potentially
productive agricultural land are taken out of production
because of salinization. How much of this land is reclaimed
(to various degrees) and then cultivated again is unknown.
Wide divergence in the salinity figures is reported by
different institutions, probably because of incomplete data
and different definitions of salinity-affected lands; e.g.,
estimates for salt-affected land in India, according to
different sources, range from 7 to 26 million ha, or between
17% and 60% of the irrigated land. For Pakistan, the most
likely figure is some 40%, Israel 13%, Australia 20%, China
15%, Iraq 50%, and Egypt 35% of the irrigated land (3).
In many coastal regions, salt water intrusion has a
strong negative effect on the regional salt balance. The
problem is associated with groundwater extraction from
deep fractured aquifers in the coastal region, often for
drinking water in a coastal city. Sea water intrusion into
the delta of rivers also upsets the regional salt balance.
Here the cause is usually a much diminished outflow from
the river because of upstream extractions for agricultural
and industrial use. Examples include the Colorado River
in the United States, the Murray in South Australia, and
the Rhine in the Netherlands.
SALT BALANCE OF THE ROOT ZONE
Soluble salts added with the irrigation water must be
removed by leaching to maintain the salt balance in the
root zone of irrigated land. If not, crop yields will drop
and—in case of predominantly sodium salts—soil crusting
is likely to occur with an adverse effect on infiltration
and hydraulic conductivity of the soil. The less soluble
salts gradually precipitate when solutions become more
concentrated and are again mobilized when the solutions
become diluted by mixing with freshwater. Depending
on whether precipitation or dissolution of salts (e.g., of
lime or gypsum) occurs in the soil, the amount of salt
leached from the root zone must be less than or more than
that applied by irrigation. Generally, again depending on
the occurrence of precipitation and dissolution reactions,
when too much irrigation water is applied to the soil, more
salts are added with the water, and in consequence, the
drainage water also contains more salts than necessary. If
the drainage water is returned to the river downstream,
water quality becomes worse. Salt concentrations in the
root zone must therefore be controlled to sustain irrigated
agriculture over the long term.
When the quality of irrigation water deteriorates, for
example, by extensive use of (slightly) saline groundwater
when canal water became scarce, nonsteady conditions
occur and cation reactions will change the salt load
of drainage water. Because of exchange of sodium and
magnesium for calcium on the clay particles, mineral
precipitation could double in a soil profile that was initially
high in exchangeable calcium. These nonsteady conditions
could last for many years before a new steady state evolves.
Irrigation thus interferes with the geohydrological and
geochemical regimes of the area. When irrigation was
introduced in areas with low groundwater tables, excessive
irrigation often occurred and deep percolation was the
agent of interference. The main consequences of this
increased groundwater recharge are a high watertable and
the mobilization of natural salt residues of the semiarid
lands. The latter is referred to as salt pickup.
The relationship among the salinity of the applied
water, the leaching fraction, and the resulting soil
salinity is an important one. Crop yields and the
salinity of drainage water could be more accurately
predicted if one could unambiguously calculate the soil
salinity resulting from irrigation applications of known
salinity and at a specified leaching fraction. Table 1
presents various relationships among leaching fraction
(LF), the dimensionless ratio of the average weighted root-
zone salinity (Cs), and the average salinity of applied
water (Ca).
The values in the table are based on steady-state con-
ditions. Because of differences between the underlying
assumptions, considerable variation exists in the calcu-
lated ratios of root zone salinity to irrigation water salinity.
For the four methods of Table 1, the salinity-tolerance
data are from threshold salinity-response functions, and
the leaching equations ignore the effect of sodium salts
on the soil structure. The relation between soil and water
salinity as governed by leaching is a dynamic one and
hard to predict, subject as it is to site-specific feedback
mechanisms among growth of the crop (hence, their water
use through evapotranspiration), weather conditions (5),
and the leaching of salts (6). A contributing factor is the
variability in measured values of the EC of saturated-
soil extracts. The coefficient of variation of the EC of soil
moisture at saturation is about 50% (7).
Below a leaching fraction of 0.1, a small change in the
leaching amount can make a large difference in root zone
salinity. The ratio of root zone salinity to irrigation water
salinity is less sensitive to changes in the leaching amount
at LF values between 0.1 and 0.4, which are most common.
In this range of LF values, root-zone salinity increases
about linearly with the salinity of the applied water. With
the introduction of precision irrigation to reduce water
losses and reduce drainage flows, leaching fractions tend
Table 1. Relationships Between Leaching Fraction and
Ratio of Soil Salinity Over Applied Water Salinity
LF Cs/Ca1 Cs/Ca1 Cs/Ca1 Cs/Ca1
0.05 3.00 7.00 4.0 10.50
0.1 2.00 5.00 2.6 5.50
0.2 1.25 3.00 1.4 3.00
0.3 1.00 2.50 1.3 2.15
0.4 0.83 2.35 1.0 1.75
Source: Kijne (4).

16
14
Mean root zone salinity (mS/cm)
12
10
8 watertables are a problem (waterlogging) or where the
6
4 amount of salt farmers are allowed to discharge in the
2 States indicates that strict adherence to a set of site-
0 flows and recharge to the groundwater is necessary
0 1 2 3 4 5
Salinity of applied water (mS/cm)
Figure 1. Root zone salinity as a function of the applied
water and the leaching fraction. [Source: Hoffman and van
Genuchten (8)].
to be lower with an enhanced risk of salt accumulation in
the root zone.
The relationship of the fourth column in Table 1 uses a
root water uptake function that is exponential with depth.
This function is often used in modeling studies where a
relationship between leaching and root-zone salinity is
required. It is plotted in Fig. 1.
LEACHING OF SALT TO PREVENT YIELD REDUCTION
In irrigation practice, the leaching requirement is usually
linked to the expected yield reduction. During leaching,
soil salinity in the root zone increases with depth. In
planning the desired leaching fraction for crop production
purposes, it is commonly assumed that the EC at the
lower root zone boundary corresponding to 25% to 50%
yield reduction is still acceptable. The weighted average
EC value for the entire root zone (weighted according to
root distribution) would then be less than at the lower
root zone boundary, and the resulting yield reduction is
assumed to be less than 25% to 50%. This result gives
some indication of the required leaching fraction based on
the relation between yield and soil salinity.
The rate of downward flow is highest shortly after
irrigation when the soil water content is still above or
near field capacity. Thereafter leaching continues at a
much-reduced rate. When the soil profile dries between
irrigations, the soil solution becomes more concentrated.
In many soils, the soil solution at field capacity is about
twice as concentrated as when the soil is saturated. A
further complication is that not all downward flow is
equally effective in leaching salts from the root zone.
Leaching is more effective when water moves through the
soil mass, rather than through cracks between aggregates.
Water moving through cracks and wormholes is called
preferential flow. The efficiency of leaching depends on the
fraction of irrigation water moving down the soil profile
mixing with the soil solution. At the start of the leaching
process as much as three-quarter of the applied water
may flow through cracks in clay soils. Once the soil swells
MAINTAINING SALT BALANCE ON IRRIGATED LAND 679
up with moisture, cracks close and the leaching process
LF = 0.05 becomes more efficient.
LF = 0.1
LF = 0.2
LF = 0.3 LEACHING CONTROL
LF = 0.4
Leaching practices influence the quality and quantity of
drainage water. Their control is important when rising
salt load in the drainage water needs to be brought
down. In such cases, restrictions are placed on the
drainage water. Experience in Australia and the United
specific regulations on the amount and quality of drainage
6 7 to maintain the salt balance in irrigated agriculture.
Because of the presence of monitoring systems, farmers
in those countries can make well-informed choices about
the crops to be grown and the cropping intensities they
adopt. Thus, agricultural choices are made within a
strict set of rules and the farmers are made responsible
for the consequences of their decisions. The successful
introduction of regulations with respect to permissible
pumping from and recharge to the groundwater, including
salt loading, in developing countries will be conditional
on the development of adequate monitoring systems and
management institutions.
RIVER SALINITY
All arid zone rivers have natural salt profiles resulting
from mobilization of salts in the catchment area and saline
seeps. Also, irrigation-induced transport contributes to
river salinity as drain water with fossil salts from pumped
groundwater discharges into a river. Increases in the
salinity of rivers and streams in many dry parts of the
world pose an ecological hazard that has largely been
overlooked. This lack of attention for the ecological impact
of increased salinity in inland waters is unfortunate
considering the vulnerability of aquatic ecosystems to
increased salt levels (9).
Salinity profiles for four rivers are shown in Fig. 2,
which illustrates the various degrees to which salts are
returned to the river or remain in the land and the
groundwater (10). The data used in Fig. 2 are averages
that ignore the seasonal changes in salt load as affected
by the flow regime of the rivers. Most drainage water from
agricultural land in Pakistan’s Punjab is being reused,
either from surface drains or pumped up from shallow
groundwater. In fact, in some irrigation systems, one half
to two thirds of the irrigation water is pumped from
the groundwater. According to one estimate, about 116
million tons of salt are added annually to the irrigated
land in Pakistan (11). Because of the absence of extensive
drainage, the leached salts are returned to the land
rather than disposed of in the Indus River system or
in evaporation ponds. The average annual salt load of
the Indus River water measured at rim stations in the
catchment is estimated at 33 million tons, whereas the
outflow to the sea contains only 16.4 million tons. Hence
about half of the annual salt influx remains in the land
680 MAINTAINING SALT BALANCE ON IRRIGATED LAND
Salinity profiles 4 rivers
1200
Murray
Indus
1000 Nile
Colorado
800
Salinity (ppm)
600
400
200
0 of this technique illustrate the limited usefulness of
0 0.2 0.4 0.6 0.8
Relative distance
Figure 2. Salinity profiles of four rivers. [Source: Smedema
(10)].
and in the groundwater (10). Most accumulation takes
place in the Punjab, which is in sharp contrast with Egypt
where a large portion of the irrigated land is underlain
by subsurface drains that take the drainage water back to
the Nile River. The salts do not stay in the Nile Basin but
are discharged into the Mediterranean Sea. During part of
the year, the salt content in the lower Indus is much lower
than in the lower Nile and more salt disposal into the
Indus could be accepted. However, during critically low
flow periods, such disposals would not be possible as the
salinity of the Indus water would exceed acceptable limits.
The only option during those periods would be to store
the drainage water temporarily for release during high
flood periods. Extending the Left Bank Outfall Drain, now
operating in Sindh, into Pakistan’s Punjab may provide a
more permanent, but also more expensive, solution than
the present inadequate number of evaporation ponds.
The salt load of the Murray River shows considerable
seasonal changes. At high flows when most of the river
flow originates in the upper catchment, the salt influx
is low, but at low flows of the river, when more of the
river flow comes from groundwater base flow, the salt
influx is higher. At low flows, salinity levels of the Murray
are sometimes lowest downstream as the salinity of some
tributaries is so much less than that of the main river that
the water is diluted.
The salinity profile of the Colorado River is the result
of large-scale irrigation development in the basin and
concentration of the river water because of evaporative
losses from storage reservoirs along its course. Under
the salinity control projects of the Colorado River basin,
several measures are undertaken to reduce further
increases in river salinity. These measures include control
of canal seepage by lining, restrictions on the quality of
drainage flow back into the river, on-farm water savings by
land leveling and changes in cropping patterns, retirement
of irrigated land, and watershed management.
These examples of analyzing river salinity profiles
clearly demonstrate their potential importance in under-
standing salt balances of irrigated lands at the scale of
river basins.
MEASURING AND MODELING THE SALT BALANCE
Although the methodology for monitoring salinity at
field level has been developed, using it in practice is
problematic. The ability to diagnose and monitor field-
scale salinity has improved considerably during the last
30 years or so. The apparent electrical conductivity of
the soil can be measured by means of four-electrode
sensors, either surface array or insertion probes, remote
electromagnetic induction sensors (e.g., EM31 and EM38),
and time-domain reflectometric sensors. Estimation of
solute concentration from such measurements requires
a calibration model relating soil solution EC to bulk
EC. Many examples of the application and calibration
1 1.2 manufacture-provided calibration curves and the different
degrees of sophistication of the calibration models (12).
Various models have been developed to simulate
transport through soils [e.g., LEACHM (13), SWAP (14)]
that require soil solution EC data for their validation.
These models are based on physical processes of water
infiltration, redistribution, and evaporation. They enhance
the understanding of water flow and chemical transport
behavior and have been applied to simulate the terms of
the water and salt balance (15). Widespread application of
the models is restricted by the need for calibration with
locally collected data and matching adjustment of some
of the parameters. For example, in an Indian study with
LEACHM-C, the root mean square errors (RSME) in the
comparison of observed and predicted values of soil profile
salinity ranged from 28% to 70%. After the water retention
parameters were adjusted, the RMSE values ranged from
13% to 24% (16).
Similar problems of uncertainty in model application
are associated with high watertables developing from
excessive percolation in irrigated areas without adequate
drainage. Plant roots take up water from shallow
watertables as capillary rise meets part of the water
demand of the crop. But if the shallow groundwater
is saline, salt accumulation in the root zone affects
crop production. Several models have been applied to
this situation to ascertain the correct balance between
the potentially positive and negative effects of shallow
groundwater (17).
The spatial variability of soils and the temporal
variability of the hydraulic parameters, especially in clay
soils, as well as the largely unknown quantity of lateral
groundwater flow constrain the applicability of models to
larger areas. Yet, the focus of attention should be changed
to the larger scale because of increased recycling of
drainage water either through pumping from groundwater
or by reusing water from surface drains. It is no longer
sufficient to maintain a favorable salt balance at field
scale; the salt balance should be maintained at system or
river basin scale. As is apparent from the examples of river
salinity mentioned earlier, the need for a holistic approach
to the salt balance of irrigated areas is large and urgent.
BIBLIOGRAPHY
1. Oster, J.D. and Rhoades, J.D. (1990). Steady-state root zone
salt balance. In: Agricultural Salinity Assessment and

Management. K.K. Tanji (Ed.). American Society of Civil
Engineers, New York, pp. 469–503.
2. Ghassemi, F., Jakeman A.J., and Nix, H.A. (1995). Salinisa-
tion of land and water resources: Human causes, management
and case studies. CABI/University of New South Wales Press
Ltd., Melbourne, Australia.
3. Gleick, P.H. (1993). Water in crisis. A guide to the
world’s fresh water resources. Pacific Institute for Studies
in Development, Environment, and Security, Stockholm
Environmental Institute, Oxford University Press, Oxford,
UK.
4. Kijne, J.W. Water productivity under saline conditions. In:
Water Productivity in Agriculture: Limits and Opportunities
for Improvement. J.W. Kijne, R. Barker, and D. Molden
(Eds.). CABI Publishing, Wallingford, UK, pp. 89–102.
5. Shalhevet, J. (1994). Using water of marginal quality for crop
production: major issues. Agricultural Water Management 25:
233–269.
6. Dinar, A., Rhoades, J.P., Nash, P., and Waggoner, B.L.
(1991). Production functions relating crop yield, water quality
and quantity, soil salinity and drainage volume. Agricultural
Water Management 19: 51–66.
7. Tedeschi, A., Beltran, A., and Aragues, R. (2001). Irrigation
management and hydrosalinity balance in a semi-arid
area of the middle Ebro river basin (Spain). Agric. Water
Management 49: 31–50.
8. Hoffman, G.J. and van Genuchten, M.Th. (1983). Water
management for salinity control. In: Limitations to Efficient
Water Use in Crop Production. H. Taylor, W. Jordan, and
T. Sinclair (Eds.). American Society of Agronomy Monograph.
Madison, WI, pp. 73–85.
9. Williams, W.D. (1987). Salinization of rivers and streams: An
important environmental hazard. Ambio 16: 180–185.
10. Smedema, L. (2000). Irrigation-induced river salinization:
Five major irrigated basins in the arid zone. International
Water Management Institute, Colombo, Sri Lanka.
11. Chaudhry, M.R. and Bhutta, M.N. (2000). Soil and water
contaminations from agro-industrial chemical in Pakistan. In:
Proceedings Agroviron 2000: 2nd International Symposium
on New Technologies for Environmental Monitoring and
Agro-Applications. Trakya University, Tekirdag, Turkey,
pp. 499–494.
12. Triantafallis, J., Laslett, G.M., and McBratney, A.B. (2000).
Calibrating an electromagnetic induction instrument to
measure salinity in soil under irrigated cotton. Soil Science
Society of America Journal 64: 10009–1017; G. Amente, J.M.
Baker, C.F. Reece. (2000). Estimation of soil solution
electrical conductivity from bulk soil electrical conductivity
in sandy soils. Soil Science Society of America Journal 64:
1931–1939.
13. Wagenet, R.J. and Hutson, J.L. (1987). LEACHM: leaching
estimation and chemistry model. Version 2. Water Resources
Institute Consortium, Center for Environmental Research,
Cornell University, Ithaca, NY.
14. van Dam, J.C. et al. (1997). Theory of SWAP version
2.0. Technical Document 45. Wageningen Agricultural
university and DLO Winand Staring Centre, Wageningen,
the Netherlands.
15. Trantafallis, J., Huckel, A.I., and Odeh, I.O.A., (2003). Field-
scale assessment of deep drainage. Irrigation Science 21:
183–192.
16. Hagi-Bishow, M. and Bonnell, R.B. (2000). Assessment
of LEACHM-C model for semi-arid saline irrigation.
International Commission on Irrigation and Drainage (ICID)
Journal 49(1): 29–42.
SALT TOLERANCE 681
17. Prathapar, S.A., Meyer, W.S., Bailey, M.A., and Poulton, D.C.
(1996). A soil and groundwater simulation model: SWAGSIM.
Environmental Software 11: 151–158; X. Wang, P.H.J. Hol-
landers, S. Wang, S. Fang. (2004). Effect of field groundwater
table control on water and salinity balance and crop yield
in the Qingtongxia Irrigation District, China. Irrigation and
Drainage 53: 263–275.
READING LIST
Topp, G.C., Reynolds, W.D., and Green, R.E. (Eds.). (1992).
Advances in Measurement of Soil Physical Properties: Bringing
Theory into Practice. SSSA Special Publication 30, Soil Science
Society of America, Madison, WI.
Skaggs, R.W. and van Schilfgaarde, J. (1999). Agricultural
Drainage. Agronomy Monograph 38. American Society of
Agronomy, Madison, WI.
SALT TOLERANCE
SURESH K. GUPTA
CSSRI
Karnal, Haryana, India
Salinity has adversely affected irrigated agriculture for
thousands of years, probably since the first human
civilization in Mesopotamia. The fall of this ancient
civilization is attributed to both waterlogging and soil
salinity (4). David Shedidan in his book ‘‘Desertification
of the United States’’ calls soil salinization of irrigated
lands one of the deadliest forms of desertification (5). No
other region, where irrigated agriculture is practiced, can
be an exception to this kind of desertification. In recent
years, soil salinity has assumed a worldwide dimension
and has become a serious threat to at least one-third
of irrigated lands in the arid and semiarid regions
where lots of irrigation water is used without providing
adequate drainage. Although a long-term solution to the
problem is to carry out improvements in land drainage,
differences in salt tolerance of crops provide opportunities
to continue growing increasingly salt-tolerant crops on
salty lands to buy time before investments are arranged
for reclamation.
SALT TOLERANCE: DEFINITION
The literal meaning of tolerance is to endure, sustain,
or put up with. Thus, the salt tolerance of a plant could
be defined as the plant’s capacity to endure the effects
of excess salt in the medium of root growth (1). That
plants can withstand a fair amount of salts without
adverse effects is implicit in this definition. In fact, a
fair proportion of salts in the growing medium is needed
for good growth. There are instances where plant growth
is stimulated at low concentrations of salts, and only
thereafter, the yield starts declining (2). Apparently, the
problem of soil salinity develops only when the salts
accumulate in the root zone to an extent that it becomes
harmful to the plants. The salt tolerance of a plant is
also expressed in terms of the yield decrease expected
682 SALT TOLERANCE
for a given level of salinity compared to the yield under
nonsaline conditions (3). Many times, salt tolerance and
salt resistance have been used synonymously although,
the two do not convey the same meaning. Salt resistance
is a plant’s ability to decrease or prevent the stress from
penetrating into its tissues, but tolerance is the ability of
the plant to eliminate, reduce, or repair the stress once it
has already penetrated.
Salt or salinity in this article is used as a general term
for salt stress, which could manifest itself from altogether
different causes and compositions of salts under different
situations. The stress could be due to bulk salinity of
the soil water caused by excess soluble salts. Electrical
conductivity (EC) of the saturation extract (ECe ) of a bulk
soil solution has been widely used as a measure of this
kind of stress (6). The adverse effects could also appear
due to high exchangeable sodium, usually expressed in
terms of exchangeable sodium percentage (ESP), due to
excess ions such as chlorides and sodium expressed in
meq L−1 , or due to excess toxic ions such as boron and
fluorine, expressed in mg L−1 . A plant tolerant to one
kind of stress need not necessarily be tolerant to other
kinds of stresses. Therefore, a large number of relative
tolerance tables are now available in the literature (1).
These tables can provide a first approximation of the
relative tolerance of plants to the anticipated kind and
level of stress.
BIOLOGICAL RESPONSE TO SALINITY
The response of plants to salinity has been extensively
studied and widely reported. For the purpose of discussions
in this section, salinity effects are grouped under the
heads (1) the effect of excess soluble salts, (2) the effect of
high exchangeable sodium, and (3) a specific ion effect or
toxicity of specific ions.
Effect of Excess Soluble Salts
Adverse effects appear to be related to the bulk soil salinity
of the soil solution. As the salt concentration increases, the
osmotic pressure of the soil water increases which affects
the availability of water to the plants. When the osmotic
pressure of the soil solution is high, the plants are unable
to extract water as readily as they can from a relatively
nonsaline soil. As such, even if water is present in the
soil, it might not be available to the plants and results in
physiological drought or osmotic desiccation. To survive,
the plants must adjust osmotically and build up high
internal solute concentrations, as is the case with salt
accumulating halophytes. In salt sensitive plants, excess
amounts of salt in the plant cells can be harmful because
of damage to cell membrane, enzymes, and organelles.
The limited uptake of water hinders the uptake of ions
by the plant root system. Thus, plants might encounter
deficiencies of nutrients. Because the soil solution has
high concentrations of certain ionic species, some of these
ions, which are otherwise not harmful, might become
toxic when absorbed in excess amounts. Salt-sensitive
glycophytes have inadequate control over ion uptake, so
such excess absorption of ions cannot be ruled out. Besides,
preferential absorption of one ion may also retard the
absorption of other essential plant nutrients necessary
for the normal growth of plants. Evidence of potassium
deficiencies and inhibition of absorption of nitrate from
saline soils can be cited.
Even glycophytes, which are salt tolerant and possess
the ability to adjust osmotically, exhibit growth reduction
largely on account of energy expenditure for uptake and
synthesis of solutes.
Exchangeable Sodium
High ESP of a soil is associated with structural
deterioration of soils. As a result, the edaphic environment
is adversely affected and results in restricted air and
water movement in the root zone. The factors that
limit the uptake of water in alkali/sodic soils are a
shallow wetting zone, an impermeable layer (crust) on
the surface, low unsaturated hydraulic conductivity, and
low available moisture content. Besides water availability,
sodium toxicity, as a result of high ESP, can also affect
the yield. High ESP can also disturb the mineral nutrition
of the plant because it results in nutritionally inadequate
amounts of Ca, or it might retard the absorption of some
ions (7).
Specific Ion Effects
Some constituents of salts, for example, chlorides, which
are harmless to most crops at moderate concentrations,
may be toxic to some crops at higher concentrations, for
example, citrus and grapes. In addition, some elements
like lithium, and boron are toxic to plants even though
they may be present in only very small quantities.
Under such situations, plants would be subjected to toxic
effects, nutritional imbalance, and deficiencies of certain
essential elements.
In all the cases discussed in this section, it is believed
that the adverse effects are usually due to the cumulative
effects of several factors, although one may be dominating
others in many cases.
DIFFERENCES IN SALT TOLERANCE OF PLANTS
Plants differ greatly in tolerance to salts. A wide range
of salt-tolerant plants that evolved over millions of years
exists in the plant kingdom. Plants with natural salt
tolerance can survive in water of varying degrees of
saltiness, from slightly brackish in the case of barley
to ocean water in some hardy species of mangrove trees.
Plant species called ‘halophytes’ are highly salt tolerant;
some of them withstand salt concentrations in excess of
that of seawater. At least 1500 of them are currently
known (8), many of them have potential for commercial
exploitation (9). These plants contain genes that protect
them from salts in a variety of ways, such as growing
salt-blocking membranes, excreting salt through their
leaves, or sequestering salt in microscopic compartments
called vacuoles. The natural vegetation of arid regions,
xerophytes, also has high salt tolerance. On the other
hand, almost all crop species are glycophytes, plants that
do not grow well in the presence of salts. Most fruit

trees and vegetables are salt-sensitive plants. Generally,
vegetable crops are more tolerant than fruit crops but are
more sensitive to salt than field and forage crops. Within
field and arable crops, beans, pulses, oil-seed crops, and
cereal crops, in that order, show increasing sensitivity to
salts. Many wild relatives of cultivated plants, however,
are reportedly highly tolerant to soil salinity, alkalinity,
and other associated conditions (10).
VISIBLE SYMPTOMS ON PLANTS GROWING UNDER
HIGH SALINITY
Within a narrow range around a plant’s salt tolerance
level, no visible symptoms might appear on plants under
field situations Yet, the crop yield might still be affected.
At high salinity, the plants show signs of wilting, a similar
symptom that appears when a plant is water stressed.
The leaves of salt stressed plants may turn yellow, a
condition called chlorosis. It happens because the excess
sodium blocks iron uptake and the lack of iron inhibits
production of chlorophyll. Stunted growth and deep bluish
green foliage of plants followed by low production are
other common signs in plants grown under high salinity.
The toxic effects that are characteristic of the specific ion
causing toxicity usually show on plant parts.
FACTORS INFLUENCING SALT TOLERANCE
Absolute tolerance to salinity cannot be expressed
explicitly. It varies considerably with edaphic and
environmental factors, which influence salt tolerance in
plants. The most important among the edaphic factors are
the physical condition of the soil, soil fertility, salt species
present in the soil solution, salt distribution in the profile,
salinity level, and irrigation methods. The environmental
factors include temperature, humidity, radiation, and
pollutants in the environment. It has been found that
poor aeration as a result of inadequate drainage may
adversely affect plant response to soil salinity. For several
crops, sensitivity to salts is known to differ markedly
according to the type of salts present and as a result,
isosmotic solutions of different composition might affect
the plants differently. Sorghum plants exhibited greater
growth reduction with sulfate type than chloride type
salinity (11). Therefore, the type of salinity might alter
the order of tolerance in crops. Since salinity stress is
expressed as the total sum of the matric potential and
the osmotic potential, the salt tolerance of crops would
be significantly different in frequently irrigated than
infrequently irrigated conditions. As such, shallow and
frequent irrigations are recommended for crops growing
under a saline/sodic environment. Because of the same
reasoning, a plant response at the same salinity is better
in drip irrigation than with surface irrigation. Gupta
and Dubey (12) reported that irrigation water of much
higher salinity could be used for growing vegetable crops
with pitcher irrigation than with conventional surface
irrigation techniques, as reported by Minhas and Gupta
(13). Apparently, the salt tolerance of most crops under
pressurized modern irrigation techniques needs to be
worked out.
SALT TOLERANCE 683
Plants respond better to salt stress in cool and
humid than in hot and dry weather. Studies on several
crops such as alfalfa, bean, beet, carrot, cotton, onion,
squash, strawberry, clover, salt grass, and tomato have
shown that salinity decreased yields more when these
crops were grown at high temperatures (14–16). Hoffman
et al. (17) observed that alfalfa yields were highest under
ozone concentrations found in many agricultural areas
in California at moderate salinity levels which were
otherwise found to reduce growth in the absence of ozone.
Besides the environmental and edaphic factors dis-
cussed in the foregoing section, biological factors such
as the varieties and rootstock of plants and the stage of
growth also cause significant differences in the response
of plants to salt stress. Evidence is abounds revealing
inter- and intragenic differences in plants’ response to
salt stress. Green alga (Dunaliella vindis) survives in
the highly saline Dead Sea at a salinity level exceed-
ing 35 dS m−1 . A survey of reported data on the response
of crop plants to increasing salinity levels reveals about
eightfold differences among them in tolerance to salinity
stress (18). An evaluation of more than 1000 wheat geno-
types revealed a wide variation in the tolerance to soil
sodicity (given by the pH of the soil solution). These wheat
genotypes could be classified into tolerant, that grew well
up to a pH of 9.6 (ESP and pH are well correlated for
alluvial soils of the Indo-Gangetic plains) and sensitive,
that did not grow well unless the pH was below 8.5. In
between these two classes, varieties could be classified as
medium tolerant and medium sensitive (19). Aswathappa
et al. (20) recorded that some acacia species grew in NaCl
concentrations of 1000 mol m−3 , whereas others stopped
growing at 200 mol m−3 .
Some crops are known to be sensitive to salt stress at
germination and others at emergence and early seedling
growth. In cereal crops, the early productive stage has been
found sensitive. Recent studies indicate that corn (21),
sorghum (22), and cowpea (23) are most sensitive during
the vegetative and early reproductive stages and least
sensitive during the grain filling stage. It has been found
that the tolerance of wheat increased with ontogeny. The
soil saturation extract values at which yield would decline
by 50% is 9.3 dS m−1 for seeding to crown root initiation
stage and increases with age to 13.2 dS m−1 for dough
stage to maturity (24). Though data sets are increasingly
added to the literature to highlight such differences in
response, some criticism is also appearing because of the
limitations of the methods used in such evaluations. It is
highlighted that presently used conventional methods of
soil tests fail to reveal the true stress to which a plant
is subjected at different growth stages. For example, it
is difficult to say whether the young roots at the time
of the emergence and initial establishment stage are
really sensitive to salts or are exposed to a high salt
stress particularly when the salt accumulate at the soil
surface. Similarly, the salinity of the soil water that is
approximately equal to that of the irrigation water at the
top of the root zone is several times more than the salinity
of irrigation water at the bottom. It is natural that at
the later growth stages, plant roots moving deep into the
profile might experience greater salt stress than that given
684 SALT TOLERANCE
by the average soil salinity. Thus, one would tend to agree
to a general statement that in many crops and plants,
increased tolerance with age has been observed, yet the
data are not sufficient to draw conclusions in absolute
terms about the differences in salt tolerance at different
growth stages.
CHARACTERIZATION AND ASSESSMENT OF SALT
TOLERANCE
A number of models have been used to appraise the relative
tolerance of crop plants. In some models, survival of the
plants has been used as a criterion whereas in others,
vegetative growth, absolute yield, and relative yield have
been used to characterize relative salt tolerance. When
comparisons are made among the models, some drawbacks
are apparent in a few of them. For example, survival of
the plants used in many ecological works is very often at
the expense of production. Thus, it seems inappropriate
for agricultural situations. Similarly, vegetative growth
has a limitation. It has been seen that, where vegetative
growth has been affected, many plants pick up later
and yield almost at the optimum level. Absolute yield
is useful agriculturally, as it tells the farmer what to
expect, but has little use for quantifying differences in
tolerance between crops with different levels of yield, for
example, sugar cane and wheat. The use of relative yield
compared to a control (say, non saline environment) allows
comparisons between crops and cultivars, where the yield
is expressed in different units or where the components
of economic yield differ between them. More and more
reports are now appearing where crop production functions
are plotted in terms of relative yield as a continuous
function of increasing salinity or for that matter ESP or
the concentration of a particular ion. At this stage, it seems
to be the most appropriate criterion for evaluating the
relative salt tolerance of plants, although some limitations
are known.
Maas and Hoffman (3), probably for the first time,
proposed a linear response model, containing three
independent parameters, the non saline control yield
(Ymax ), a salinity threshold (ECt ), below which yield
(Y) was unaffected, and a slope (S) representing the
decrease in yield with increase in salinity above this level.
EC0 represents the salinity at which there is no yield.
Mathematically, the response function is written as
Y = Ymax 0 ≤ EC ≤ ECt (1)
Y = Ymax − S(EC − ECt ) ECt < EC ≤ EC0 (2)
Y=0 EC > EC0 (3)
The relative yield of the crop as a ratio of the nonsaline
control could also be used in this model. Then, Y would
become RY, the relative yield and Ymax would be equal
to 1 or 100 depending upon whether the relative yield is
expressed in terms of a fraction or percentage.
The piecewise linear model has been used to produce
tables of salinity response functions for a number of
crops (1,3). Later studies have established that such a
model can effectively describe the production function as
affected by high ESP (25). A model proposed independently
by Oosterbaan is also based on the same premise and
uses segmented regressions defining the threshold as a
breakpoint. The model proposed by Oosterbaan (26) seems
to be quite flexible and could be used for many more
situations besides studying the salt tolerance of crops.
Based on this model, threshold salinity, the salinity at
which a 50% (EC50 ) reduction in yield is expected, and
the salinity at which the yield is zero (EC0 ) have been
used to assess the relative salt tolerance of crops. Both the
threshold salinity and EC0 are not well defined (2), so EC50
seems to be the most appropriate parameter to assess the
relative salt tolerance of crops (18,27), as it integrates
both the slope and the threshold. The data derived
from Maas (28) show that both ECt and slope overlap
between two classes (Table 1). Such overlap gets masked
while calculating the EC50 . van Genuchten (2) proposed a
sigmoid model, defined in terms of EC50 and p where again
EC50 is the salinity where the yield is reduced by 50%, and
p is an empirical constant defining the shape of the curve.
This model also fits well with almost all the data sets of the
Maas and Hoffman model, so it attests to the fact that EC50
could be a good parameter to define relative salt tolerance.
The use of this model in field experimentation, besides
other limitations, may be difficult where differences
between genotypes are small. A limitation of this model
is explained well in Fig. 1. Using EC50 as the criterion, it
would be concluded that the tolerance of four crops is in
the order crop a > crop b = crop c > crop d. On the other
hand, within crops b and c, one would tend to conclude
that crop b is more tolerant to salt at low and medium
salinity levels and crop c is more tolerant at high levels of
salinity. Another problem with this model is that it can be
used to identify tolerance as distinct from yield potential in
many crop cultivars. In wheat, for instance, Kharchia 65, a
wheat variety from Rajasthan (India) showed the highest
degree of salt tolerance though it was not one of the top
yielders amongst the varieties tested at Karnal (India).
Notwithstanding such minor difficulties, this approach
has the potential of application in categorizing the relative
salt tolerance of crops.
RELATIVE SALT TOLERANCE OF CROPS
Extensive data on this issue are part of the literature. The
most comprehensive information on this aspect is compiled
and reported by Maas (1). This includes information on the
relative salt tolerance of herbaceous crops (fiber, grains,
grasses, forage, and special crops), vegetable and fruit
crops, woody crops, ornamental shrubs, trees, and ground
cover crops to the bulk soil salinity of the growing media. It
Table 1. Range of Model Parameters in Describing the
Relative Salt Tolerance of Agricultural Crops
Relative Salt Tolerance of Crops ECt (dS m−1 ) Slope (%)
Sensitive 1–2 15–35
Moderately sensitive 1–3 6–20
Moderately tolerant 3–7 4–20
Tolerant 4–8 4–7

100
80
Relative yield, %
60
40
(a) (b) (c)
20
0 5 10 15 20
Electrical conductivity, dS m−1
also lists the tolerance of agricultural crops and fruit-crop
cultivars and rootstocks to chloride and the tolerance of
agricultural crops and citrus and stone fruit rootstocks to
boron. A number of tables listing the relative tolerance
of crops have also appeared, but most are based on data
reported by Maas and his co-workers. A table showing the
relative tolerance of field, vegetable, forage, and fruit crops
is given in Appendix 1. Gupta and Sharma (25) reported
on the tolerance of a few agricultural crops and grasses to
soil ESP. Porwal et al. (29) compiled the available data in
India on tolerance of crops to saline water and presented
the threshold and slope of the Maas and Hoffman function
in terms of irrigation water quality. Per se, saline water
might not directly affect crop yield, but once the leaching
fractions under a given set of conditions are known, this
data could be effectively used to assess crop tolerance to
soil salinity. Later on, similar data were compiled and
reported by Minhas and Gupta (13).
Although extensive literature is available on the
relative tolerance to salts, considering the different
agroclimatic conditions, different methods employed,
varying farming conditions, and varieties used, the results
obtained do not always agree. As an example, the alkali
tolerance of rice in the literature has been rated as
medium, but in the Indian context, it has been rated
as highly tolerant. This difference arises because of the
fact that the earlier rating was for direct seeded rice and
the India rating is based upon transplanted rice where
35–40 day seedlings have been used for transplanting.
SALT TOLERANCE AND MANAGEMENT OF POOR
QUALITY LAND AND WATER RESOURCES
Several treatments and management practices can reduce
the salt level in soil. However, there are some situations
where it is either not possible or not economically prac-
tical to attain desirably low soil salinity levels. In such
cases, proper plant selection is one way to moderate yield
reductions caused by excessive soil salinity. The ability of a
SALT TOLERANCE 685
(d)
Figure 1. Crop salt tolerance as revealed
25 30
by the piecewise linear model of Maas
and Hoffman.
plant to grow on salty lands or to remove salt from such soil
makes salt-tolerant plants potentially useful for restoring
salt-damaged soils. Salt-tolerant plants can also provide
farmers with greater flexibility in crop selection besides
reducing the leaching requirement. The difference in the
salt tolerance of plants is now being exploited both on
experimental and field scales to use salty water resources
for crop production and to dispose of them in an ecofriendly
manner with reduced investments. It is achieved through
sequential concentration of water. In such a system, fol-
lowing each cycle of use, the volume of water decreases, but
its salinity increases. Thus, after each cycle, a more salt
tolerant crop is chosen before finally disposing of the water
in an evaporation tank or by some other means (Fig. 2).
This kind of practice has also been referred as a multiple
water use system in the literature. The quantity of water to
be disposed of at this stage is much less, so the investment
in an evaporation tank is significantly reduced.
Fresh water
Agricultural land Return flow
Salt-tolerant
with normal
crops
cropping
Return flow
Return flow
Halophytes Trees
Return flow
Evaporation
pond/disposal
Figure 2. A schematic representation of sequential concentra-
tion for managing water resources.
686 SALT TOLERANCE
RECENT DEVELOPMENTS IN SALT TOLERANT CROPS
Conventional breeding for developing salt-tolerant plants
has been attempted for a long time, and the subject
of breeding for salt tolerance has been reviewed by
many workers (30). The lack of success in generating
tolerant varieties [given the low number of varieties
released and their limited salt tolerance (31)] would
suggest that conventional breeding practices are not
enough. Plant breeding from the existing gene pools
could help to achieve small increases in salinity tolerance.
Advances in molecular biology as well do not appear to
have helped greatly in the development of salt-tolerant
genotypes. There is good evidence that salt tolerance
in truly tolerant species (halophytes) is a multifaceted
trait. In halophytes, tolerance depends upon ion exclusion
or compartmentation, synthesis of compatible solutes,
regulation of transpiration, control of ion leakage through
the apoplast, membrane characteristics, and the ability
to tolerate low K:Na in the cytoplasm. It has been
demonstrated that a range of traits is also important in
crops. This suggests that the overall trait of salt tolerance
is determined by a number of genes.
Twenty years ago, Epstein (32) argued for development
of crops in which the consumable portion is botanically
a fruit, such as grain, berries, or pomes, and that
have a truly halophytic response to salinity. In these
plants, Na+ ions would accumulate mainly in leaves, and
because water transport to fruits and seeds is primarily
through the phloem pathway (i.e., the intercellular
connections), much of the salt from these organs would
be screened. Initially, this hypothesis looked beyond
realization. But in the last few years, scientists have
reported increasing success in using genetic engineering
to unravel the secrets of the genes and proteins that
govern salt tolerance. A genetically engineered tomato
that thrives in brackish water was produced in mid-2001
by a team of plant biologists at UC Davis (33). In late
2001, Purdue University scientists discovered a protein
that guides salt into plants (34). Biotechnology specialist
Alan McHughen at UC Riverside plans to begin research
into producing salt-tolerant avocado trees. We also have
other transgenic plants, for example, canola. Up to 10%
of the dry weight of the leaves in this plant is sodium
chloride. Schroeder and his colleagues have also discovered
a gene that helps protect plants from salt damage by
keeping salt in the roots. These results clearly support
Epstein’s argument.
With some success coming our way, it must not be
forgotten that much more basic research is yet to be
conducted. To develop genetically modified plants that
have the triple characteristics of high yield, good quality,
and tolerance to salts will not be an easy task in
the foreseeable future. Besides, the genetically modified
plants must be stable, so their descendants can be
reliably grown with the added traits, and the plants
must be tested for safety. This might take years of
testing. Moreover, recent criticism and experience with
genetically modified plants might hamper the progress
of work. For example, some environmental groups say
that large agricultural companies are rushing too quickly
to introduce genetically modified crops without regard
for potential dangers. Genetically engineered corn to
produce a natural insecticide has been opposed on the
grounds that it could kill insects that are not pests,
such as the monarch butterfly. Even more worrisome,
early genetically modified plants weren’t very useful
to consumers. An earlier genetically engineered tomato
called the Flav-Savr, also produced from research at
UC Davis, flopped in the mid-1990s. To resolve such
issues would require much more effort. Till then,
an integrated approach based on conventional plant
breeding, molecular biology, and genetic engineering
should continue.
BIBLIOGRAPHY
1. Maas, E.V. (1990). Crop salt tolerance. In: Agricul-
tural Salinity Assessment and Management. K.K. Tanji
(Ed.). American Society of Civil Engineers, New York,
pp. 262–304.
2. Van Genuchten, M.Th. (1983). Analyzing Crop Salt Tolerance
Data: Model Description and Users Manual. Research Report
No. 120. US Salinity Laboratory, Riverside, CA, p. 50.
3. Maas, E.V. and Hoffman, G.J. (1977). Crop salt tolerance:
Current assessment. J. Irrig. Drainage Div. 103(IR 2):
115–134.
4. Rhoades, J.D. (1990). Soil salinity-causes and control. In:
Techniques for Desert Reclamation. John Wiley & Sons, New
York, pp. 109–134.
5. Shedidan, D. (1981). Desertification of the United States.
Council on Environment Quality. USA.
6. U.S. Salinity Laboratory Staff. (1954). Diagnosis and
Improvement of Saline and Alkali Soils. USDA Handbook 60.
US Salinity Laboratory, Riverside, CA.
7. Maas, E.V. and Grieve, C.M. (1987). Sodium-induced calcium
deficiency in salt-stressed corn. Plant Cell Environ. 10:
559–564.
8. Aronson, J.A. (1989). HALOPH: A Data Base of Salt tolerant
Plants of the World. Tucson, Arizona. Office of Arid Lands
Studies, the University of Arizona, Tucson, AZ, p. 77.
9. Glenn, E., Brown, J.J., and Blumwald, E. (1999). Salt-
tolerant mechanisms and crop potential of halophytes. Crit.
Rev. Plant Sci. 18: 227–255.
10. Stalker, H.T. (1980). Utilization of wild species for crop
improvement. Adv. Agron. 33: 111–147.
11. Lauchi, A. and Epstein, E. (1984). Mechanism of salt
tolerance in plants. Calif. Agric. 38: 18.
12. Gupta, S.K. and Dubey, S.K. (2001). Pitcher irrigation for
water conservation and use of saline water in vegetable
production. Indian Farming. 51: 33–34,36.
13. Minhas, P.S. and Gupta, R.K. (1992). Quality of Irrigation
Water—Assessment and Management. Indian Council of
Agricultural Research, New Delhi, India.
14. Ahi, S.M. and Powers, W.L. (1938). Salt tolerance of plants at
various temperatures. Plant Physio. 13: 767–789.
15. Magistad, O. et al. (1943). Effect of salt concentration, kind
of salt and climate on plant growth in sand culture. Plant
Physio. 18: 151–166.
16. Hoffman, G.J. and Rawlins, S.L. (1971). Growth and water
potential of root crops as influenced by salinity and relative
humidity. Agron. J. 63: 877–880.
 SALT TOLERANCE 687

17. Hoffman, G.J., Maas, E.V., and Rawlins, S.L. (1975). Salinity APPENDIX A
ozone interactive effects on alfalfa yield and water relations.
Relative Tolerance of Crop/Plants to Soil Salinity (ECe )a
J. Environ. Quality 4: 326–331.
18. Rana, R.S. (1986). Breeding crop varieties for salt affected Tolerant Moderately Tolerant Sensitive
soils. In: Approaches for Incorporating Drought and Salinity
Field Crops
Resistance in Crop Plants. V.L. Chopra and R.S. Paroda
(Eds.). Oxford and IBH, New Delhi, pp. 24–55. Barley Rye Black gram
19. Singh, K.N. (1990). Wheat varieties in alkali soils. Lecture Sugar beet Wheat Bengal gram
in Summer Institute on Innovative Methods and Emerging Rape Oat Green gram
Trends in Plant Breeding Research. CCS Haryana Agricul- Cotton Rice Lentil
tural University. Sorghum
Corn
20. Aswathappa, N., Morcar N.E., and Thomson, L.A.J. (1986).
Sunflower
Salt tolerance of Australian tropical and sub-tropical acacias.
Castor
Proc. Int. Workshop on Aust. Acacias in Developing Countries.
Safflower
J.W. Tumbull (Ed.). Gympie, Australia, pp. 70–73.
Soybean
21. Maas, E.V. et al. (1983). Salt sensitivity of corn at various Pearl millet
growth stages. Irrig. Sci. 4: 45–57. Linseed
22. Maas, E.V., Poss J.A., and Hoffman, G.J. (1986). Salinity Cluster bean
sensitivity of sorghum at three growth stages. Irrig. Sci. Pigeon pea
7: 1–11. Cowpea
23. Maas, E.V. and Poss, J.A. (1989). Sensitivity of cowpea to salt Sesame
stress at three growth stages. Irrig. Sci. 10: 313–320. Groundnut
24. Minhas, P.S., Sharma O.P., and Patil, S.G. (1998). 25 Years Vegetable Crops
of Research on Management of Salt-Affected Soils and Use of
Garden beets Bitter gourd Radish
Saline Water in Agriculture. Central Soil Salinity Research
Asparagus Bottle gourd Carrot
Institute, Karnal, p. 220.
Spinach Brinjal Celery
25. Gupta, S.K. and Sharma, S.K. (1990). Response of crops Amaranths Tomato Green beans
to high exchangeable sodium percentage. Irrig. Sci. 11: Cabbage Coriander
173–179. Pea Cumin
26. Oosterbaan, R.J. (1981). Crop yield, soil salinity and water Lady’s finger Mint
table depths in Pakistan. Annual Report, International Insti- Onion
tute for Land Reclamation and Improvement, Wageningen, Potato
the Netherlands. Carrot
27. Royo, A., Susin, V., and Aragüés, R. (1991). Salt tolerance of Turnip
barley: Analysis of response functions at emergence and at Sweet potato
adult stages. In: Improvement and Management of Winter Dolichos
Cereals under Temperature, Drought and Salinity Stresses. Sponge gourd
A. Acevedo et al. (Eds.). Proceedings of the International Watermelon
Symposium; October 1987; Cordoba, Spain, pp. 373–394. Muskmelon
Chilies
28. Maas, E.V. (1984). Crop tolerance. Calif. Agric. 103: 20–21.
Fenugreek
29. Porwal, A.K., Misra, K.L., and Gupta, S.K. (1990). Assess- Garlic
ment of crop tolerance to saline irrigation waters. Bhu. Jal.
News. 5(4): 35–40. Forage Crops
30. Shannon, M.C. (1990). Genetic approaches for developing Salt grass Perennial rye grass Meadow foxtail
economic salt tolerance crops. In: Agricultural Salinity Bermuda grass Dallis grass Alsike clover
Assessment and Management. K.K. Tanji (Ed.). American Rhodes grass Sudan grass Red clover
Society of Civil Engineers, New York, pp. 161–185. Bridsfoot Alfalfa Burnet
31. Flowers, T.J., Koyama, M., and Yeo, A.R. (1998). Breeding Barley (hay) Rye (hay)
for salt tolerance in crop plants in the age of molecular Oats (hay)
biology. In: Salinity Management in Agriculture. S.K. Gupta Wheat (hay)
et al. (Eds.). Central Soil Salinity Research Institute, Karnal, Orchard grass
India, pp. 303–313. Blue grass
32. Epstein, E. (1983). Crops tolerant to salinity and other Fruit Crops
mineral stresses. In: Better Crops for Food. J. Nugent and
M. O’Connor (Eds.). Ciba Foundation Symposium, London, Date palm Pomegranate Pear
pp. 61–82. Guava Apple
Fig Orange
33. Zhang, H.X. and Blumwald, E. (2001). Transgenic salt
Olive Plum
tolerant tomato plants accumulate salt in the foliage but Grape Peach
not in the fruits. Nat. Biotechnol. 19: 765–768. Ber Mango
34. Shi, H.H., Ishitani, M., Kim, C., and Zhu, J.K. (2000). The Kagzi Lime Avocado
Arabidopsis thaliana salt tolerance gene SOS1 encodes a
The values in the same column under a given category and for the type of
a putative Na+ /H+ antiporter. Proc. Nat. Aca. Sci. 97: crop/plant are in descending order of tolerance. The values are qualitative
6896–6901. and are to be used as a first approximation.
688 GROUNDWATER ASSESSMENT USING SOIL SAMPLING TECHNIQUES
GROUNDWATER ASSESSMENT USING SOIL
SAMPLING TECHNIQUES
JAMES A. JACOBS
Environmental Bio-Systems, Inc.
Mill Valley, California
Groundwater assessment for environmental projects
involves significant soil sampling to evaluate hydrologic
and lithologic conditions. Soil samples in environmental
investigations are typically collected at a minimum of
every one to two meters and at changes of lithology or
obvious signs of contamination, based on soil staining or
discoloration of the soil or in the case of volatile organic
compounds, organic vapor readings. Continuous coring is
recommended for a minimum of at least one soil boring
per site to characterize the subsurface adequately.
VADOSE ZONE
Soil samples collected in the vadose zone above the water
table are in the unsaturated zone. Groundwater is moving
generally vertically, and some horizontal movement is
caused by rootlets, faults, impermeable layers, or other
conduits. Soil samples collected within the capillary fringe
or below the groundwater table are in the saturated zone.
The capillary fringe moves up and down with
seasonal variations of the groundwater table. Petroleum
hydrocarbons move vertically from the source, such as an
underground storage tank, through the vadose zone. As
the hydrocarbons move vertically through the lithologic
section, in general, the finer grained the soil, the greater
the lateral spreading of the contaminant (Fig. 1).
Kp = Conc soil/conc gas
Soil zone
Unsaturated zone
Kp = Conc soil/conc gas
Intermediate zone
Kh = Conc gas/conc water
Capillary fringe
Water table
Saturated zone
Kd = Conc soil/conc water
Groundwater
Figure 1. Equilibrium expressions for partioning volatile organic
compounds among aqueous, vapor, and sorbed phases.
CAPILLARY FRINGE ZONE
The capillary fringe, within the saturated zone, is the zone
immediately above the water table, where water is drawn
upward by capillary attraction. Due to irregularities in
pore sizes among the grains of sediment, the capillary
water does not rise to an even height above the water
table but rather forms an irregular capillary fringe. The
capillary fringe is higher in fine-grained soils such as a
fine silt than in coarse grained soil such as a gravel due
to greater tensions created by the smaller pore openings.
The capillary fringe can be the zone of direct evaporation
of groundwater if the water table is close enough to the
surface (1).
The vertical movement during a complete hydrologic
cycle can be a few centimeters, in other cases, it can be
tens of meters, depending on the recharge of the water,
the proximity of pumping wells, and other factors. This
movement greatly affects the zone above the water table,
which includes the capillary fringe zone, especially when
it is impacted by light nonaqueous phase liquids (LNAPL),
also called floating product or free-product.
When the free-product of a hydrocarbon fuel, such as
gasoline, moves up and down through the hydrologic cycle,
a smear zone develops in the zone above the water table.
Some of this highly impacted soil zone (HISZ) is above the
current capillary fringe zone and may exist below the top of
the current groundwater table. Therefore, a HISZ equals
the cumulative thicknesses of the migrating capillary
fringe zones and frequently exists in areas where free-
product or high concentrations of dissolved hydrocarbons
have existed for a long period (Fig. 2).
For hydrocarbon fuels, which are lighter than water,
the soil sample just above the water table is typically
collected and analyzed. This sample may contain the
highest concentration of contaminants within a soil
boring. Sometimes, buried or submerged contaminants
are trapped beneath the water table when water levels
rise, leaving a former free-product zone below the water
table (2). In these cases, soil samples collected within the
groundwater zone would be appropriate (Fig. 3).
Vadose zone V1
V2
Free phase gasoline
V3
Capillary fringe
Saturated zone
Dissolved phase
Figure 2. Gasoline movement from source (underground storage
tank) to saturated zone (2).
 GROUNDWATER ASSESSMENT USING SOIL SAMPLING TECHNIQUES
Fall
Falling water table
LNAPL
Water table
Low-permeability layer
Direction of groundwater flow
Figure 3. Trapping gasoline contaminants below the water table by fluctuating groundwater elevations (2).
For dense nonaqueous phase liquids (DNAPLs) such as
chlorinated solvents, which have a specific gravity greater
than water, soil samples are collected throughout the
aquifer for hydrogeologic characterization. The highest
concentrations of these dense liquids may be at the base of
the aquifer, above the top of the next underlying aquitard.
SATURATED ZONE
Soil samples collected in the saturated or phreatic zone
below the water table are used to evaluate the aquifer’s
characteristics. Soil samples collected below the ground-
water table are generally not analyzed at a chemical
laboratory, in favor of the more representative ground-
water sample. Nonetheless, these saturated soil samples
can provide a wealth of information about aquifer char-
acteristics, total organic carbon, solid oxygen demand,
porosity and permeability, as well as free-product dis-
tribution. Continuous soil coring of aquifer sediments
allows complete lithologic analysis for potential ground-
water remedial designs. These soil samples help scientists
and engineers understand groundwater dynamics and the
vertical and lateral movement of contaminants in the sub-
surface. Although groundwater sampling is imperative for
groundwater assessments, data from soil samples used
in conjunction with groundwater data provide a more
complete understanding of subsurface conditions.
More detailed soil sampling information can be
obtained from Testa (3), Jacobs (4), American Society for
Testing Materials (5), California Regional Water Quality
Control Board (6), and Morrison et al. (2). There are
numerous minor differences in soil sampling and record
keeping techniques and methods among field samplers,
regional variations, and specific regulatory requirements
that vary from state to state. A generalized summary of
selected environmental soil sampling techniques follows.
689
Summer Spring
Rising water table
LNAPL Trapped LNAPL
SOIL SAMPLING METHODS
Undisturbed or minimally disturbed samples are obtained
by driving a sampler into the soil, either manually, or using
a drilling rig. The soil sampler is retrieved and opened, and
the sample liner containing the soil sample is extracted.
Samples are collected during subsurface investigations in
vertical borings using rigs. Split-spoon samplers and direct
push core samplers are used frequently for shallow soil
sampling. The inner sample liners are typically composed
of polyethylene, brass, or stainless steel. Samples may
also be collected from tank pits and soil piles by driving
a sampling tube into the soil by hand or using a drive-
hammer.
Some analyses or guidelines allow collecting disturbed
samples using a trowel, and then placing them in glass
jars. Photoionization detectors (PID) are used in the
field to measure organic vapors in soil samples. Soil
samples should be collected in accordance with regulatory
guidelines, which may vary among states and locales. The
sample liners, containing soil, are typically sealed with
Teflon tape and plastic caps at both ends. The samples are
then labeled and stored in a refrigerated environment.
Small variations in soil lithology within several inches
can lead to large differences in the results of chemical
analysis. Duplicate soil samples, if taken, are collected
as near as possible to the actual initial soil sample.
Typically, these duplicate samples are collected in the
same lithologic zone as the original sample. Duplicate
samples should be a minimum of 5% of the samples. To
maintain sample integrity, soil samples are delivered to
certified laboratories using chain-of-custody procedures.
Lithologic Description
Soils and unconsolidated deposits for environmental
projects are commonly described according to American
Society for Testing Materials (ASTM) Method D 2488-
84 (5) and the United Soil Classification System (USCS)
690 GROUNDWATER ASSESSMENT USING SOIL SAMPLING TECHNIQUES

Major Group Major Group Description

divisions symbol Description divisions symbol

Inorganic silts and very fine

GW Well-graded gravels, gravel-sand sands, rock flour, silty or

Liquid limit is less than 50

mixtures, little or no fines ML
clayey fine sands or clayey
silts with slight plasticity

Silts and clays

More than half of coarse fraction
is larger than no. 4 sieve size.

Poorly graded gravels or

More than half of materials is smaller than no. 200 sieve size.
Inorganic clays of low to
GP gravel−sand mixtures, little or no
medium plasticity, gravelly
fines
CL clays, sandy clays, silty clays,
lean clays
Gravels
More than half of materials is larger than no. 200 sieve size.

GM Silty gravels, gravel−sand−silt

mixture
Organic silts and organic

Fine-grained soils
OL clays of low plasticity
Coarse-grained soils

GC Clayey gravels, gravel−sand−clay

mixtures

Inorganic silts, micaceous or

MH diatomaceous fine sandy or
silty soils, elastic silts
Liquid limit is greater than 50
Well-graded sands, gravelly sands,
SW
little or no fines
Soils and clays
more than half of coarse fraction
is smaller than no. 4 sieve size.

Inorganic clays of high

Poorly graded sands or gravelly CH plasticity, fat clays
SP sands, little or no fines
Sands

Organic clays and silts of

SM Silty sands, sand−silt mixtures medium to high plasticity
OH

Highly Peat and other highly organic

SC Clayey sands, sand−clay mixtures organic Pt soils
soils

Figure 4. Unified soil classification system (USCS).

for physical description and identification of soils. The using ASTM guidelines. Stratigraphic, genetic, and other
USCS was first used in the geotechnical engineering data and interpretations are usually also recorded in
field and is now commonly used in the environmental the boring log. For consolidated deposits, including
field. Unfortunately, the USCS is not very useful for igneous, metamorphic, and sedimentary rocks, the USCS
describing lithologic characteristics, geologic properties, is combined with other geologic characteristics such as
or the origin or depositional environment of the soil or weathering, sorting, sphericity, and separation.
sediments (Fig. 4). Color is commonly described by comparing the soil
Other soil identification systems include the Burmis- sample with a Munsell Rock Color Chart (7). Each layer
ter Soil Identification System (BS) for unconsolidated of soil is identified by the geologist or engineer using
deposits, which is commonly used with the USCS. The the following items: color, soil type (USCS), classification
other soil description system is the Comprehensive Soil symbol, color, consistency or relative density, moisture,
Classification System (CSCS) developed by the U.S. structure (if any), and modifying information such
as grain sizes, particle shape, cementation, plasticity,
Department of Agriculture. The CSCS describes soils as to
stratification, and other notes.
agricultural productivity potential and best agricultural
land use.
The ASTM Soil Classification Flow Chart and the USCS LITHOLOGIC INTERPRETATION
are generally accepted soil description methods used in
the engineering and environmental fields. Descriptions Once the lithology has been determined, geologists and
of moisture, density, strength, plasticity, etc. are made engineers compare the data to develop a geologic model

for the depositional environment. Knowing the type
and origin of the soil, sediments and rocks provides a
better understanding of groundwater and contaminant
movement and allows predictive modeling of an aquifer.
Once the geochemical data have been evaluated, a site
conceptual model can be developed, and a corrective action
plan and remedial design can begin if the site requires
environmental cleanup.
BIBLIOGRAPHY
1. Fetter, C.W. (1993). Contaminant Hydrogeology. Macmillan,
New York, p. 458.
2. Morrison, R., Beers, R., and Hartman, B. (1998). Petroleum
Hydrocarbon Contamination: Legal and Technical Considera-
tions. Argent Communications Group, Foresthill, CA.
3. Testa, S.M. (1994). Geologic Aspects of Hazardous Waste
Management. CRC Press, Boca Raton, FL, pp. 145–187.
4. Jacobs, J. (2000). Monitoring well construction and sampling
techniques. In: Standard Handbook of Environmental Science,
Health and Technology. J. Lehr (Ed.). McGraw-Hill, New York,
pp. 11.46–11.68.
5. American Society for Testing Materials. (1984). Standard
Practice for Description and Identification of Soils (Visual-
Manual Procedure). Method D 2488-84. ASTM Philadelphia,
PA.
6. California Regional Water Quality Control Board. (1989).
Leaking Underground Fuel Tank Field Manual: Guidelines
for Site Assessment, Cleanup, and Underground Storage Tank
Closure. Sacramento, CA.
7. Munsell Color. (1988). Munsell Soil Color Charts. Munsell
Color. Baltimore, MD.
SKIMMED GROUNDWATER
MUHAMMAD NADEEM ASGHAR
International Water
Management Institute (IWMI)
Lahore, Pakistan
In several unconfined aquifers under irrigated agricultural
areas, the groundwater system consists of a saturated
porous medium containing miscible fluids (fresh ground-
water and saline groundwater). The native groundwater is
deep and saline because of the marine origin of the hydro-
geologic formation. However, fresh groundwater lenses
from deep percolation of the extensive water conveyance
and distribution system, as well as from irrigation and
rainfall, overlie this native saline groundwater.
According to Sufi et al. (1), over 20 BCM of fresh
groundwater is being recharged annually in the Indus
Basin of Pakistan.
The thickness of these fresh groundwater lenses varies
from a few meters to more than 100 meters. For instance,
in the Indus Basin of Pakistan, this is around 30 m thick in
the lower or central parts of the doabs (area between two
rivers), 60 m or more along the margins of the doabs, and
approximately 150 m near the rivers and canals (2). When
groundwater is extracted from these fresh groundwater
lenses, it is called skimmed groundwater. Different types of
SKIMMED GROUNDWATER 691
irrigation wells are available for extracting this skimmed
groundwater (3).
Under natural conditions, the saline groundwater
tends to remain separate from the overlying fresh
groundwater. However, when waterbodies of variable
solute concentrations are in contact, molecular diffusion
causes mixing across the line of contact. Due to
groundwater and saline water movement, a zone of
dispersion, which is also known as the interface, forms
between these two fluids. This interface, which varies in
thickness, is not static but responsive to recharge and
discharge mechanisms. Thus, the processes of diffusion
and dispersion result in a transition zone where the
salinity gradually changes from completely fresh to
fully saline.
Soil surface
Fresh groundwater
Interface
Saline groundwater
Therefore, the use of irrigation wells for skimming
fresh groundwater may cause mixing of the ‘interface’
either within the well or in the aquifer. As a result,
the irrigation well starts supplying skimmed groundwater
of marginal quality. Therefore, the decision regarding
the use of skimmed groundwater must consider the
quality of groundwater with depth, the local hydrogeologic
conditions of the aquifer, the socioeconomics of skimming
well installation and operation, the discharge rate, the
duration of pumping, and above all, the tolerable limits of
skimmed groundwater quality for its intended uses (4).
Farmers are using this skimmed groundwater to fill
the gap between the supply and demand of surface water
in the conjunctive water use environment of the Indus
Basin of Pakistan (5). This irrigation water resource
has emerged as a formidable poverty reduction tool.
Potentially, skimmed groundwater is a huge resource of
irrigation water. In the Indus Basin of Pakistan alone,
it has been estimated that nearly 200 billion cubic
meter (BCM) of fresh groundwater is lying above salty
groundwater (6), and covers about 30% of the irrigated
area of Pakistan.
BIBLIOGRAPHY
1. Sufi, A.B., Latif, M., and Skogerboe, G.V. (1998). Simulating
skimming well techniques for sustainable exploitation of
groundwater. Irrig. Drainage Syst. 12: 203–226.
2. Asghar, M.N., Prathapar, S.A., and Shafique, M.S. (2002).
Extracting relatively fresh groundwater from aquifers under-
lain by salty groundwater. Agric. Water Manage. 52: 119–137.
692 SOIL MOISTURE MEASUREMENT—NEUTRON
3. Saeed, M.M., Asghar, M.N., and Bruen, M. (2003). Options for
skimming fresh groundwater in the Indus Basin of Pakistan:
a review. J. Groundwater Hydrol. 45(3).
4. Asghar, M.N., Saeed, M.M., Ashraf, M., and Qureshi, A.S.
(2002). Potential of Skimming Wells for Extracting Relatively
Fresh Groundwater from Unconfined Aquifers Underlain by
Saline Groundwater. Proc. 2nd Int. Conf. New Trends Water
Environ. Eng. Safety Life: Eco-compatible Solutions Aquatic
Environ., June 24–28, Capri, Italy.
5. Wahaj, R. and Asghar, M.N. (2002). Farmers’ Management
Responses to the gap between Supply and Demand of Canal
Water in a Large-Scale Irrigation System. Proc. 18th Congr.
Int. Comm. Irrig. Drainage, Food Prod. Conditions Water
Scarcity, Increasing Population Environ. Pressure, July 21–28,
Montreal, Canada.
6. NESPAK. (1983). Drainage and Irrigation in Shallow Sweet
Water Areas Overlying Saline Water Using Radial Collector
Wells. Technical Appraisal, NESPAK Ltd, Lahore, Pakistan.
SOIL MOISTURE MEASUREMENT—NEUTRON
MUKAND SINGH BABEL
Asian Institute of Technology
Pathumthani, Thailand
The water (liquid and/or vapor phase) contained in the soil
above the water table is called soil moisture. Soil moisture
is a function of precipitation, soil texture, porosity, the
rate of soil moisture uptake by plants, and the rate of
evapotranspiration. Measurements of soil moisture are
needed for several purposes, including irrigation planning
and scheduling, agronomic studies, flow and transport
processes in the unsaturated zone, and geotechnical
investigations.
Soil moisture can be measured by direct and indirect
methods. Direct methods are those wherein water is
removed from a soil sample by evaporation, leaching, or
chemical reaction, and the amount removed is determined.
Indirect methods involve measuring some property of the
soil that is affected by soil moisture content or measuring
a property of some object placed in the soil, which
comes to water equilibrium with the soil. The indirect
methods generally require calibration. They are suitable
for quick and repetitive measurements and do not require
destructive sampling as needed in direct gravimetric
methods (1).
Although a variety of indirect techniques are avail-
able for measuring soil moisture content, they are neither
equivalent nor applicable in all situations. Thus, choosing
a suitable method and following a well-defined proce-
dure are important for obtaining reproducible and reliable
information. Various in situ recording sensors, water bud-
get models, and experimental approaches are used for
soil moisture measurement, including neutron probe, time
domain reflectometry, electromagnetic induction, electron
resistivity tomography, tensiometers, heat dissipation sen-
sors, and thermocouple psychrometers. Modern methods
of soil moisture measurement such as airborne radar and
satellite imagery are still at the research and develop-
ment stage.
The neutron probe also called the ‘‘neutron moisture
meter’’ is a nondestructive, indirect method commonly
used for repetitive field measurement of soil water content
volumetrically.
PRINCIPLE
The neutron probe is based on the principle of neutron
thermalization. Hydrogen nuclei are nearly the same size
and mass as neutrons and therefore, can scatter and slow
neutrons. High-energy (0.1–10 Mev.), fast (1600 km/s)
neutrons emitted from a radioactive source, such as
americium–beryllium or radium–beryllium are slowed
and changed in direction by elastic collisions with atomic
nuclei of hydrogen atoms in soil. This process is called
‘‘thermalization,’’ and the low-energy (about 0.03 eV)
neutrons, after collision, are called ‘‘thermal/thermalized
neutrons.’’ The density of the thermal neutrons (thermal
cloud of neutrons) depends on the soil–water content;
the vast majority of hydrogen in soil is associated with
water. Higher water content in soil leads to increased
thermalization and, thus, a denser thermal cloud. A
‘‘slow neutron’’ detector is installed adjacent to the source
(emitter) to measure the cloud density. The measurement
is usually in the form of a ‘‘count ratio’’; a higher count
ratio denotes higher water content and vice versa. Thus,
the density of thermal neutrons can be calibrated against
the water content volumetric. (2).
THE EQUIPMENT
Figure 1 shows a typical neutron moisture meter. The
equipment consists of
1. a neutron moisture depth probe containing a
radioactive source that emits high-energy neutrons,
and a detector of thermalized neutrons.
2. protective shielding (for gamma rays) of lead and
polyethylene or paraffin for neutron absorption. The
depth probe is stored within the shield when not
in use.
3. a meter assembly that includes a pulse counter to
register the counts generated by the thermal neutron
detector, a built-in computer, a display screen, and
a keyboard.
4. a thin-walled aluminum or plastic access tube. The
tube diameter is recommended by manufacturers for
optimized equipment performance.
5. a radioisotope license, radioisotope leak test kit,
and radiation monitoring equipment, as required
by the license.
6. calibration curves and parametric data for calibrat-
ing the computer of the equipment, if needed.
CALIBRATION OF EQUIPMENT
The sources of hydrogen other than water (the primary
source of hydrogen) in a given soil are assumed constant
and are accounted for during calibration. Similarly,
 SOIL MOISTURE MEASUREMENT—NEUTRON 693

The sphere of influence of neutrons around the

Neutron probe radiation source varies with water content from about
0.15 m for wet soil (almost saturated) to more than 0.60 m
for very dry soil, depending on how far fast neutrons must
travel to collide with a requisite number of hydrogen
nuclei. An approximate relationship for the radius of
influence, r (in cm), as a function of soil water content
is given by
Sphere of r = 15(θv )−1/3
influence Access tube

PROCEDURE

1. Use the soil auger to form a hole for installing the

Dry rD
Probe
(source and
detector)
Wet rw
Figure 1. Neutron probe device. rD and rW are the radii of
measurements in dry and wet soils (Source: Reference 3).
several nonhydrogen substances including C, Cd, B,
Cl, and Li may be present in some soils and may
COMMENTS AND PRECAUTIONS
also thermalize fast neutrons. However, their effect is
compensated for through soil specific calibration. The
neutron probe is calibrated by paired measurements of
soil moisture content and neutron probe counts. Usually,
the volumetric moisture content, θv is linearly related to
the slow neutron count ratio (CR) as follows:
θv = a + b(CR)
where CR is the ratio of slow neutron counts at a specific
location in the soil and a standard count obtained with the
probe in its shield. A calibration curve can be obtained with
a sufficient number of measurements. This calibration
curve is approximately the same for many soils. Figure 2
depicts a typical calibration curve.
access tube. The access tube is left so as to protrude
about 10 cm above the soil surface and is covered
with an empty can or a stopper to keep water and
debris from entering.
2. Place the probe unit over the access tube to make a
measurement. Select an appropriate counting time,
and make several standard counts while the probe
is in the shield. This will correct for any electronic
changes in the counting circuits.
3. Make one or more counts at each selected measure-
ment depth.
4. Use the calibration curve to convert the count
ratios to volumetric moisture content, or read the
moisture content directly if the equipment has a
built-in computer.
1. In regard to the radiation hazard of devices with
radioactive sources, it is strongly recommended
that the equipment be used according to the
manufacturer’s instructions, which must comply
with safe procedures dictated by the radioactive
source license.
2. Site-specific calibration may be required for a soil
of unusual neutron-absorption properties or if the
diameter or material of the access tube is different
from the manufacturer’s recommendations.
3. As the zone of measurement is spherical and at
least greater than 0.15 m in radius, the probe depth
increment in the access tube should not be less than

Neutron probe calibration

Amsterdam silt loam
0.35
0.3
Volume water content

VWC = 0.056 + 0.329 (CR)

0.25 R2 = 0.982

0.2
0.15

0.1

0.05
0.2 0.4 0.6 0.8 1 1.2 Figure 2. A neutron probe calibration relationship
Count ratio (Source: Reference 3).
694 SOIL N MANAGEMENT IMPACT ON THE QUALITY OF SURFACE AND SUBSURFACE WATER
0.15 m. The approach to the soil surface should be no
closer than 0.15 m because a portion of the neutrons
may escape from the soil.
ADVANTAGES AND DISADVANTAGES
The neutron probe moisture meter has several advantages:
1. It is a widely accepted method of soil moisture
measurement with highly reproducible results.
2. After initial tube installation, no further monitoring
is required.
3. Both horizontal and vertical installation are possi-
ble.
4. It can measure soil water content at multiple depths
and locations.
5. It is possible to couple the equipment with
other devices.
6. The equipment produces reliable results for a
long period.
Following are the disadvantages of the equipment:
1. It contains a source of radioactivity and, therefore,
cannot be automated. It may also pose a health risk
if adequate precautions are not taken.
2. The equipment is expensive and requires calibration.
3. It has relatively limited spatial resolution of
measurement.
BIBLIOGRAPHY
1. Gardner, W.H. (1986). Water content. In: Methods of Soil
Analysis, Part 1. A. Klute (Ed.). Agronomy Monograph No. 9,
American Society of Agronomy, Inc. and Soil Science Society of
America, Inc., Madison, WI.
2. Carter, R.M. (1993). Soil Sampling and Methods of Analysis.
Lewis, Boca Raton, FL.
3. Sumner, E.M. (2000). Handbook of Soil Science. CRC Press,
Boca Raton, FL.
SOIL N MANAGEMENT IMPACT ON THE
QUALITY OF SURFACE AND SUBSURFACE
WATER
DIBYENDU SARKAR
VANDANA VANDANAPU
RUPALI DATTA
University of Texas at San Antonio
San Antonio, Texas
Nitrogen (N) in the forms of nitrate or ammonium is
a nutrient needed for plant growth. When the amount
of soil-supplied N is considered deficient for satisfactory
crop yields, N fertilizer is added to supplement the soil-
supplied N. If N is supplied in excess of crop needs,
a surplus is created that can cause significant water
quality problems. Until recently, the main reason for
investigations into the forms of N in soil systems below
the rooting depth of plants was to evaluate the loss of
nitrate to crops and to estimate the loss of production. In
recent years, much attention has been focused on forms
of N as potential pollutant sources of both groundwater
and surface water. Nitrate concentrations above a certain
level in drinking water may have critical health effects
on animals and human infants. Alternately, N can also
contribute to surface water contamination via surface
water runoff and erosion. Excessive nitrate concentrations
in surface water bodies, such as streams or lakes, can
accelerate algae and plant growth, resulting in oxygen
depletion. Nitrogen must be managed properly to protect
water quality. Best management practices (BMPs) are
methods that are the most effective, practical means of
preventing or reducing nutrient pollution from nonpoint
sources. Nitrogen BMPs primarily involve a change in
fertilizer management practices, such as changing the
timing, placement method, and/or the amount of fertilizer
applied to reduce the possibilities of contaminating water
resources. In addition, conservation tillage practices, cover
crops, and crop rotation all help reduce surface runoff,
erosion, and leaching of soil N. It is imperative to
continue the development and implementation of new and
effective management practices that increase N recovery
and reduce potential N losses to the environment.
INTRODUCTION
Nitrogen is one of the most abundant elements on earth.
It exhibits a number of oxidation states and combines
with a number of other elements to form organic and
inorganic compounds (1). Nitrogen accounts for 78% of
the earth’s atmosphere as elemental nitrogen (N2 ) gas (2).
However, the majority of the earth’s N (98%) is in rock,
sediment, and soils. The amount of N in rocks is about
fifty times more than that in the atmosphere (3). For
mineral soils in the United States, the approximate range
of total N content in the surface six inches is 0.02%
to 0.50% (4). Growing plants, animals, and microbial
populations need a continual source of N. It is an
essential component of the proteins that build cell material
and plant tissue. In addition, it is necessary for the
function of other biochemical agents, including chlorophyll
(which makes photosynthesis possible), enzymes (which
help organisms carry out biochemical processes and
assimilate nutrients, and nucleic acids such as DNA and
RNA which are involved in reproduction) (5). Of all the
major plant nutrients, N is often the most important
determinant of plant growth and crop yield. Without
adequate N, plant growth is restricted. Characteristic
symptoms of N deficiency in plants include yellowing
and death of leaves and stunted growth. Just as too
little N can cause problems, too much N can also cause
problems. These problems can extend to plants, humans,
animals, and the environment. The problems posed to
the environment occur when excess N in soils is carried
away with surface runoff to lakes, rivers, groundwater,
and other aquatic ecosystems. Excess N in rivers, lakes,
and groundwater can be toxic to humans and causes
 SOIL N MANAGEMENT IMPACT ON THE QUALITY OF SURFACE AND SUBSURFACE WATER
water quality problems in natural water systems (6).
Nitrogen, particularly as nitrate, is the most common
contaminant in aquifer systems (7). Hallberg (8) points to
agriculture as the most substantial anthropogenic source
of nitrate, and Keeney (9) suggests that this is caused
by intensive and extensive land use for crop and animal
production. The United States Environmental Protection
Agency (USEPA) has identified eutrophication as the main
problem in U.S. surface waters that have impaired water
quality (10). Excessive amounts of nutrients, especially N
and phosphorus (P), speed up eutrophication (11). Excess
N in the estuaries of the oceans enhances the growth
of aquatic organisms to the point that they affect water
quality and lower dissolved oxygen levels (12–14). This
affects the metabolism and growth of oxygen requiring
species, causing hypoxia (12,13). Implementation of N-
BMPs (best management practices) allows efficient,
profitable production and minimizes N movement to
surface water and groundwater. This article addresses
the concept of N movement in soils and the impacts of N
management on surface and subsurface water quality.
SOURCES OF NITROGEN IN THE ENVIRONMENT
Atmosphere
The earth’s atmosphere is made up of 78% N as a colorless,
odorless, nontoxic gas. The same N2 gas found in the
atmosphere can be found in spaces between soil particles.
However, plants cannot use this form of N (15). A group
of microbes (Rhizobium bacteria) that naturally exists in
the soil converts N to ammonium. In biological fixation,
nodule-forming Rhizobium bacteria inhabit the roots of
leguminous plants and through a symbiotic relationship,
convert atmospheric N2 to a form the plant can use (16).
Atmospheric fixation occurs during lightning storms.
The enormous energy of lightning breaks down N2
molecules and enables their atoms to combine with oxygen
in the air to form nitrogen oxides. These dissolve in rain
forming nitrates that are carried to the earth. Atmospheric
N fixation probably contributes some 5–8% of the total N
fixed (5).
The large amount of fossil fuels used to produce
electricity and in manufacturing industries oxidizes N
because the intense heat given off by their combustion
causes the molecules to react with each other, thereby
adding N to the air as nitrates. These nitrates fall to the
earth during precipitation, adding more N to the soil (17).
Animal Manure
Animal manures are important sources of N in the
terrestrial environment. Manure usually contains both
organic and inorganic N forms. The amount of N in
manure depends on the animal’s species, age, diet, and
bedding materials. Ammonium N is the primary inorganic
form in manure and is readily available to crops. Organic
N becomes available to plants as manure decomposes;
20 to 50% of organic N is available to the first crop
after application. Much of the remaining organic N
becomes available in subsequent years as mineralization
continues (18).
695
Legumes
Leguminous crops, such as alfalfa, beans, and peas,
generally do not require N fertilization because the
Rhizobium bacteria living on their roots take N from the
air and convert it to forms that plants can use. Legumes
provide some of the N for their own growth and often build
up a supply that can be used by crops that follow (19).
Commercial Fertilizers
Commercial fertilizers are derived from the atmospheric N
pool. Industrial processes used to manufacture commercial
fertilizers usually combine atmospheric N with hydrogen
and/or oxygen and/or carbon and then convert the products
to various N fertilizers such as ammonium nitrate,
anhydrous ammonia, and urea (4). These fertilizers are
generally applied directly to soils.
Crop Residues
Crop residues returned to the soil are important sources of
N for succeeding crops if they are properly managed (20).
Nitrogen exists in crop residues in complex organic
forms; the residues are decayed by soil microbes, thereby
converting N and other nutrients to inorganic forms, which
then become available for plant uptake (16).
Organic Soil Matter
Organic soil matter is a major source of N in terrestrial
ecosystems. Organic matter is composed primarily of a
relatively stable material called humus. Even though the
soil has a large pool of N stored in soil humus, this N
is mostly insoluble and is not available for plant uptake.
Organic N must be mineralized to inorganic forms by soil
microbes before plant uptake can occur.
FORMS OF NITROGEN IN SOILS
Nitrogen in soils is found in organic and inorganic forms,
both of which are important sources of N for plant and
microbial uptake. Organic N occurs in crop residues and
organic matter as protein compounds, amino acids, amino
sugars, and complex, largely unidentified substances (4).
More than 90% of all soil N exists as organic N, and for
the most part, is not available for plant uptake. Natural
biological processes carried out by microorganisms in soil
convert organic N to inorganic forms, which plants can
use (15). Inorganic N exists in soil primarily in three
forms: elemental N, ammonium N (NH4 + ) and nitrate N
(NO3 − ). Ammonium N and nitrate N are two forms of soil
N that can be readily used by plants, whereas elemental
N cannot be used directly by plants. However, some N-
fixing bacteria, particularly Rhizobium that coexists with
leguminous plants, use N in the air (elemental N) and
converts it in to a form of N that can be used by plants.
NITROGEN LOSSES FROM SOILS
Nitrogen is lost from soil in several ways: leaching,
denitrification, volatilization, soil erosion and runoff, and
crop removal.
696 SOIL N MANAGEMENT IMPACT ON THE QUALITY OF SURFACE AND SUBSURFACE WATER
Leaching
Leaching is the loss of soluble N. The NO3 − ion is
negatively charged and is not attracted to soil’s clay and
humus, which are also negatively charged. So it stays in
the water between soil particles. As water moves through
soil, the nitrate in the soil solution moves along with the
water (21). This loss most frequently occurs in areas of
heavy rainfall, under excessive irrigation, and in coarsely
textured soils.
Denitrification
Denitrification is the conversion of NO3 − molecules to
gaseous forms that are lost to the environment. When
there is a deficit of oxygen in the soil (i.e., under anaerobic
conditions), certain bacteria meet their energy needs by
reducing NO3 − to dinitrogen gas or to nitrogen oxide (N2 O).
These gases are released to the atmosphere, resulting in a
loss of N from soil (22). Denitrification can be significant
when soils are saturated with water.
Volatilization
Volatilization is the loss of N from the soil as ammonia
(NH3 ) gas. NH3 volatilization commonly takes place when
N is in an organic form called urea. Urea may originate
from animal manure, fertilizers and, to a lesser degree,
from the decay of plant materials. Urea is converted to
ammonium carbonate by urease, an enzyme found in soils.
Ammonium carbonate easily dissociates to ammonium and
carbonate ions. The carbonate ions result in a relatively
short-lived increase in soil pH. Ammonium ions convert
at high pH to NH3 gas, which may escape to the air (23).
This results in a net loss of N from the soil system.
Surface Runoff and Soil Erosion:
Surface runoff and soil erosion can also cause significant
losses of soil N. Runoff and erosion losses include both
inorganic and organic N. Erosion leads to the removal of
organic N in the upper soil profile (24) and NH4 + N, which
is adsorbed to soil particles (25). Nitrate N is water soluble
and is generally not adsorbed to soil particles; hence, it
quickly moves off the soil surface as runoff with increased
rainfall.
Crop Removal/Harvest
Substantial amounts of N are lost from soil through crop
removal (16). When crops are harvested, the N source is
removed. Crop removal accounts for a major portion of the
N that leaves the soil system.
HUMAN HEALTH AND WATER QUALITY CONCERNS
RELATED TO LOADINGS OF N INTO SURFACE AND
GROUNDWATER
Nitrogen is the one element of most concern as it relates
to water quality. Although N can be found in many
forms in the soil, the nitrate form primarily affects
water quality (15). The contamination of surface water
or groundwater by NO3 − N presents a potential health
threat to human population. High NO3 − N levels in water
are a concern because the ingestion of such water can
reduce the oxygen carrying capacity of the blood. Nitrate is
reduced to NO2 − , which oxidizes the Fe of the hemoglobin
in the blood, and methemoglobin, which cannot carry
oxygen, is formed (9). Infants have greater potential than
adults to suffer these ill effects. In the United States, the
maximum NO3 − N level for potable water has been set at
10 mg/L (26). Secondary and related deleterious effects of
exposure to NO3 − include increased respiratory infections,
inhibition of iodine uptake by the thyroid, and possible
reproductive problems (27).
Another area of concern is the toxic effect of ammonia
(NH3 ) on freshwater aquatic life (28). It has been known
since the early 1900s that NH3 is toxic to fish and that
this effect varies with water pH and temperature. A
concentration of 0.02 mg/L as unionized NH3 is the current
limit in freshwater in the United States (29).
The most significant water quality concern with respect
to N is the overenrichment or eutrophication of surface
waters. Nitrates stimulate the growth of algae and
other plankton, which provide food for higher organisms;
however, excess NO3 − N can cause over production of
plankton. As these planktons die and decompose, they
use up the oxygen dissolved in water, which causes an
oxygen shortage for other aquatic organisms. Elevated
N concentrations have altered natural aquatic floral and
faunal population dynamics, exacerbated occurrences of
hypoxia and anoxia, and sped eutrophication in the Gulf
of Mexico (12,30).
NITROGEN BEST MANAGEMENT PRACTICES: IMPACT ON
SURFACE AND GROUNDWATER QUALITY
Nitrogen from fertilizers as well as from other sources
(e.g., biosolids, animal manures) has a direct impact on
water quality. The primary goal of N best management
practices (N-BMPs) is to manage N inputs to crop
production to prevent degradation of water resources while
maintaining profitability. Typically, BMPs are farming
methods that assure optimum plant growth and minimize
adverse environmental effects (31). By putting N-BMPs
into practice, N losses from agricultural soils can be
controlled. There are many BMPs currently in use; some of
the more popular soil N management practices and their
potential effects on surface and subsurface water quality
have been categorized by Owens (32) as follows.
Crop Management
Cover Crops. The use of cover crops in cropping systems
provides several benefits. However, only in recent years
has much of the attention been focused on their potential
to reduce NO3 − leaching losses (32). The key to reducing
NO3 − N losses is to keep soil nitrate concentrations low
during periods of high leaching potential. Nitrate can leach
to groundwater when the soil contains significant NO3 −
N and when water percolates below the root zone. These
conditions often exist during the winter months, when
percolation of winter rainfall can be high and uptake of
water and nitrate by plants is minimal (33).
 SOIL N MANAGEMENT IMPACT ON THE QUALITY OF SURFACE AND SUBSURFACE WATER
Cover crops influence nitrate leaching in two ways:
water budget effects and N uptake effects. Cover
crops can influence the water budget by reducing the
water content of the soil during their active growth
stage (33,34). In humid climates, this water usage can
reduce nitrate leaching significantly. Cover crops function
by accumulating the inorganic soil N between main crop
seasons and by holding it in an organic form, thus
preventing it from leaching. Nonleguminous cover crops
can immobile as much as 70% of the available NO3 − N
in the upper soil profile (35). Generally, nonlegumes are
about three times more efficient than legumes in reducing
nitrate leaching (36). Meisinger et al. (33) reported that
cover crops reduced both the mass of N leached and the
NO3 − -N concentration of leachate by 20–80% compared to
no cover crop control. They also determined that grasses
and Brassicas were two to three times more effective than
legumes in reducing NO3 − -N leaching. Because, grasses
and grains become quickly established in the fall and
develop an extensive root system, they are more efficient
than legumes in capturing soil nitrate and preventing
late fall and winter leaching to groundwater (35,37). In
a western Oregon study of nitrate movement in the soil
profile following harvest of a broccoli crop fertilized with
250 lb N/acre, Hemphill and Hart (38) demonstrated the
effectiveness of a fall-planted rye cover crop in reducing
the movement of soil nitrate to the soil profile. Areas with
cover crop contained 0.18 mg/kg nitrate at 30–40 inches
of soil depth compared to 5.38 mg/kg nitrate at the same
depth where no cover crop was grown.
Controlling erosion is an important component of
surface water quality protection. Cover crops protect
against soil erosion (39,40). Cover crops can be effective in
reducing runoff and erosion, especially during the winter
when there is little crop residue (32). Farmland is most
susceptible to erosion when there is no vegetative ground
cover or plant residue on the soil surface. Cover crops
provide vegetative cover when a crop is not present to
cushion the force of falling raindrops, which otherwise
would detach soil particles and make them more prone to
erosion. The rate of runoff is also slowed, thus improving
moisture infiltration into the soil (41). When legumes are
used as the winter cover crop, fixed N in the legumes
can be supplied to the subsequent crop so that less
fertilizer N has to be applied. This reduces the amount
of inorganic N, which is readily available for removal in
surface runoff (32).
Crop Rotations. Understanding the relationship be-
tween N and crop rotation is very important in N
management decisions (42). Although there are several
benefits to crop rotations (e.g., lower fertilizer N input
because of residual N from legumes, erosion reduction,
minimization of crop diseases, reduction in insect
problems), relatively little attention has been given to
impacts of NO3 − N leaching by most crop sequences (32).
Proper crop sequencing can sometimes be used to improve
fertilizer-N efficiency and reduce nitrate leaching. Johnson
et al. (43) showed that unfertilized soybeans scavenge
large quantities of residual N fertilizer from a previous
corn crop. Monocultures of crops, such as corn, winter
697
wheat, and sorghum, require high N input. It is possible
to reduce N use in a farming system by rotating these
crops with small grains, which require less N, or with
legumes, which may require no added N. This type of crop
diversification should result in less N in the soil profile
and, in general, less leaching.
The combination of crops grown in a rotation also
affects the potential for groundwater contamination by
N. For example, less N moves beyond the rooting zone
with a corn–soybean rotation compared to continuous
corn (44). Including perennial legumes or non leguminous
crops in rotations also decreases NO3 − -N losses. Baker
and Melvin (45) found much lower NO3 − -N concentrations
beneath alfalfa than for corn or soybean.
The key to successful erosion control with crop rotations
is soil cover. Fallow land has the highest erosion potential
in any cropping system. Crop rotations that include small
grains or other cover crops create a vegetative cover, which
protects the soil surface from erosion (46). Grasses and
legumes or small grains provide more protection against
water erosion than row crops (47). Cropping systems with
a higher frequency of sod reduce soil erosion. Growing
cover crops with low residue crops and rotating high
residue crops with low residue crops are also effective
erosion control practices. Some crop rotations do not
reduce erosion unless other practices, such as cover
crops and residue management are followed (48). Crop
rotations that use the land more intensively, such as corn,
wheat, and soybeans grown in 2 years, produce larger
amounts of biomass during the rotation and are more
effective in reducing erosion than a continuous cropping
sequence (49).
Tillage Practices. Soil erosion and runoff contribute
to the degradation of surface water quality. Tillage
management is an important factor to be considered
when attempting to reduce soil erosion and subsequent
sedimentation (50). Tillage practices that leave the soil
surface unprotected greatly increase the potential for
runoff, erosion, and nutrient losses from fields (44).
Conventional tillage buries the protective crop residue
cover and disturbs the soil structure; increased rainfall
may then lead to amplified runoff and erosion (51).
Conservation tillage practices, which reduce runoff
and erosion, reduce N losses (52–54). Such practices
leave greater amounts of crop residue and cause less
soil disturbance. Both factors help reduce runoff and
erosion, reduce transport of soluble and sediment-
attached chemicals, and promote infiltration (55). Baker
and Laflen (56) concluded that conservation tillage should
reduce soil erosion by 75–90% (depending on the amount
of surface residue) compared to conventional tillage.
Conservation tillage also increases infiltration and reduces
average annual runoff volumes by about 25% compared to
conventional tillage (57). The majority of the N in surface
runoff is associated with sediments, so conservation tillage
usually results in a net decrease in N losses.
Several of the factors that reduce surface runoff pro-
mote an increase in subsurface flow (32). Conventional
tillage disrupts surface soil structures, destroying macrop-
ores in the tilled layer. It also enhances the mineralization
698 SOIL N MANAGEMENT IMPACT ON THE QUALITY OF SURFACE AND SUBSURFACE WATER
of organic N from crop residues and soils and increases the
accumulation of residual NO3 − N. In recent years, tech-
nological innovations have led to reduced levels of tillage
in many areas. This trend seems to protect groundwater
in some areas because of reduced N mineralization and
increased N immobilization (44). The effect of tillage on
macropores and the flow of water and chemicals through
macropores are important (58–60). However, using con-
servation tillage, macropores remain intact from the soil
surface to their full depth in the soil profile (61–63).
Macropores potentially increase the amount of water
entering the soil; this increase in infiltration leads to
more leaching. Tillage has both advantages and disad-
vantages with respect to groundwater; information from
studies on local soils should be used to determine if con-
servation tillage would have positive or negative impacts
on subsurface water quality.
Fertilizer Management
Nitrogen Rate. The one most important practice affect-
ing the N contribution to groundwater is the rate of
N application (64). Excessive application of N fertilizer
increases the likelihood that N is leached into surface
water and groundwater (47). The amount of N from other
sources should be given appropriate credit to avoid excess
fertilizer N. The soil can supply N to plants as mineral
N, usually NO3 − N, and as N mineralized from organic
matter. There can be substantial overwinter carryover of
residual NO3 − (65), which would allow reducing of subse-
quent N fertilizer applications.
Nitrogen rate recommendations also include credits
for N fixed by legumes. Winter leguminous cover crops
and legumes in rotation can provide a significant amount
of N for a subsequent nonleguminous crop (66–68). Some
legumes can provide as much as 120 kg N/ha, which shows
strongly that credit needs to be given for legume N to
reduce excessive N fertilizer applications (32). Similarly,
N application rates can be significantly adjusted to account
for manure application. By failing to account for these
sources of N in determining the correct application rate,
a surplus of N may be created; this surplus may leach to
groundwater (69).
Similarly, N credits need to be given for manure appli-
cations. Recent reports of long-term manure applications
indicate that high rates of application can pose a hazard
for NO3 − -N contamination in groundwater (32).
Fertilizer Placement. Correct fertilizer placement is a
BMP that can help protect both surface and groundwater
quality from nutrient pollution. The two most often used
fertilizer placement strategies are banding and broadcast
placement (70).
The efficiency of N fertilizer is generally increased by
band placement that provides many agronomic and envi-
ronmental advantages over broadcast applications (71).
Banding causes an increase in plant N use efficiency. High
crop-use efficiency leaves less N in the soil to leach into
groundwater (70). Banding of N below or with the seed is
by far the best N-BMP to protect surface water quality
because no N is left on the soil surface for erosional loss.
A broadcast-incorporated application leaves a portion of
the N fertilizer on or near the soil (70). A study in Mon-
tana found that spring wheat yields are approximately 6%
higher when urea was banded 1 inch below the seed and
between rows compared to broadcast (72).
Fertilizer Timing. The timing of N fertilizer application
is an important factor that affects the efficiency of fertilizer
N because the interval between application and crop
uptake determines the length of exposure of fertilizer N to
loss processes, such as leaching and denitrification. Timing
N applications to reduce the chance of N losses through
these processes can increase the efficiency of fertilizer N
use (21).
Fertilizer applied in the fall causes groundwater
degradation. Fertilizer-N applications should be managed
very carefully on sandy or gravelly soils due to the
high potential for leaching. Fall applications are not
recommended on these types of soils (44). Fall applications
of anhydrous NH3 or urea on finer textured soils should
be delayed until soil temperature reaches 45 ◦ F or less
when conversion to NO3 − is slow, hence, the potential
for NO3 − -N leaching losses is reduced. Split application
of N fertilizers as opposed to a single application can
reduce potential N losses by up to 30% and thereby reduce
groundwater contamination (73).
Organic Wastes. Animal manure is a good source of
N. It benefits plant growth, improves soil structure,
and provides nutrients. However, manure can become
an environmental problem when it pollutes groundwater
and surface water as a result of improper handling,
storage, and application on erosion-prone soils (74). This
is primarily due to the particulate nature of the material
compared to the more soluble forms of N in inorganic
fertilizers (44).
The other major concern with manure is the movement
of NO3 − N to groundwater. Manure contains mostly
organic and NH4 + N. These forms of N do not move
through soils. However, soil microorganisms can convert
these compounds to soluble NO3 − N that may leach to
groundwater. Nitrogen not mineralized in the first year
after application of manure generally becomes part of
the soil organic matter and releases N relatively slowly.
Constant annual manure applications that supply N to
meet the entire crop demand ultimately cause excessive
fertilization. Potential pollution problems resulted when
manure was applied on a long-term basis at recommended
rates (475 kg/N ha) in Alberta, Canada (75).
To use manures efficiently as a nutrient source for
crop production, the spread pattern must be uniform
throughout the field (76). Nonuniform applications of
manure result in improper nutrient crediting and
can increase the possibility of overfertilization, which
threatens groundwater and surface water quality (77).
Manure should be applied when crops have the highest
need for nutrients. The quicker the nutrients are used,
the less the chances of pollution due to runoff into surface
watercourses or leaching into groundwater sources.
Manure should be stored in properly designed facil-
ities that are protected from excessive runoff, flooding,
or overflow that would allow contamination of surface
 SOIL N MANAGEMENT IMPACT ON THE QUALITY OF SURFACE AND SUBSURFACE WATER
water. Injecting manure is especially important in pre-
venting contamination of surface water supplies. Injection
methods place liquid manure below the soil surface, thus
eliminating surface runoff on sloping soils and volatiliza-
tion of ammonia from the manure on all soils (76).
FUTURE DIRECTIONS
As Owens (32) points out, much is known about using
N for efficient crop production and minimizing its
adverse impacts on surface and subsurface water quality.
Nevertheless, there are areas where further knowledge
is needed to fill in the gaps and to improve upon
what is already known. Following Owens (32), the areas
mentioned here demand particular attention in research:
• Subsurface water quality: Much of the reported
research in this area has been conducted with soil
columns and lysimeters. There is need for more
groundwater quality research on a field or water-
shed scale. A systems approach, in which existing
knowledge is harnessed into quantitative, explana-
tory models, is more attractive and acceptable.
• Urban use: Fertilization of lawns and turf areas and
the subsequent impacts on groundwater quality need
more regulatory attention. In many states, there is no
environmental law protecting water resources from
overfertilization of privately owned lawn or gardens,
although according to a Gallup poll, 22 million
U.S. homeowners spent more than $14.6 billion on
professional landscape/lawn care/tree care services in
1997. Research on this aspect of N management has
been limited and needs to be revisited as a potential
major source of N pollution.
• Soil N Availability Tests: Crops are probably more
deficient in N than in any other element, and yet
there are no widely accepted methods of testing soils
other than testing for residual NO3 − N and NH4 +
N (78). Considerable attention needs to be directed
toward developing soil tests that provide an accurate
in-season assessment of N need for a crop, yet take
into account variation in soil types and kinds of crop
while recommending fertilizer N needs.
• Education: Knowledge about N fertilizer efficiency
and techniques for minimizing nitrate leaching are
available but are not consistently used. The real value
of continued research will be the adaptation and
implementation of these technologies, which require
a major educational/outreach effort.
BIBLIOGRAPHY
1. Rowe, M.L. and Stinnett, S. (1975). Nitrogen in the Sub-
surface Environment. Publication # 600375030. Office of
Research and Development, U.S. Environmental Protection
Agency, Washington, DC.
2. Follett, R.F. (2001). Nitrogen transformation and transport
processes. In: Nitrogen in the Environment: Sources, Problems
and Management. R.F. Follet and J.L. Hatfield (Eds.).
Elsevier, Amsterdam, pp. 17–44.
3. Stevenson, F.J. (1982). Origin and distribution of nitrogen in
soil. In: Nitrogen in Agricultural Soils. F.J. Stevenson (Ed.).
699
Agronomy Monograph No. 22, American Society of Agronomy,
Madison, WI, pp. 1–42.
4. Wells, K.L., Sims, J.L., and Smith, M.S. (1997). Nitrogen in
Kentucky Soils, AGR-43. University of Kentucky Cooperative
Extension Service, University of Kentucky, Lexington, KY.
5. Henry, C., Sullivan, D., Rynk, R., Dorsey, K., and Cogger, C.
(1999). Managing Nitrogen from Biosolids. Publication #
99–508. Northwest Biosolids Management Association and
Washington State Department of Ecology, Olympia, WA.
6. Hallberg, G.R. and Keeney, D.R. (1993). Nitrate. In: Regional
Ground-Water Quality. W.M. Alley (Ed.). Van Nostrand
Reinhold, New York, pp. 297–322.
7. Freeze, R.A. and Cherry, J.A. (1979). Groundwater. Prentice-
Hall, Englewood Cliffs, NJ.
8. Hallberg, G.R. (1989). Nitrate in groundwater in the United
States. In: Nitrogen Management and Ground Water Protec-
tion. R.F. Follett (Ed.). Elsevier, Amsterdam, pp. 35–74.
9. Keeney, D.R. (1982). Nitrogen management for maximum
efficiency and minimum pollution. In: Nitrogen in Agricul-
tural Soils. F.J. Stevenson (Ed.). Agronomy Monograph 22,
American Society of Agronomy, Madison, WI, pp. 605–649.
10. USEPA. (1996). Environmental indicators of water quality in
the United States: EPA 841-R-96-002. EPA Office of Water,
Washington, DC.
11. Baird, J.V. (1990). Soil Facts: Nitrogen Management and
Water Quality. AG-439-2. North Carolina Cooperative Exten-
sion Service. North Carolina State University, Raleigh, NC.
Available: http://www.soil.ncsu.edu/publications/Soilfacts/
AG-439-02/index.htm.
12. Rabalais, N.N., Wiseman, W.J., Turner, R.E., Sengupta,
B.K., and Dortch, Q. (1996). Nutrient changes in the Missis-
sippi River and systems responses on the adjacent continental
shelf. Estuaries 19: 386–407.
13. Downing, J.A., Rabalais, N.N., Diaz, R., Zimmerman, R.,
Baker, J.L., and Duffy, M.D. (1999). Gulf of Mexico Hypoxia:
Land-Sea Interactions. Council for Agricultural Science and
Technology, Task Force Report No. 134.
14. Howarth, R.W. (2000). Clean Coastal Waters: Understanding
and Reducing the Effects of Nutrient Pollution. National
Academy Press, Washington, DC.
15. Killpack, S.C. and Buchholz, D. (1993). Nitrogen in the
Environment: How Nitrogen Enters Groundwater. Water
Quality Initiative publication WQ256. University of Missouri
Extension, University of Missouri-Columbia, Columbia, MO.
16. O’Leary, M., Rehm, G., and Schmitt, M. (1994). Understand-
ing Nitrogen in Soils. FO-03770-GO. University of Minnesota
Cooperative Extension Service, University of Minnesota,
Minneapolis, MN. Available: http://www.extension.umn.edu/
distribution/cropsystems/DC3770.html.
17. Nebel, B. and Wright, R. (2000). Environmental Science.
Prentice-Hall, Upper Saddle River, NJ.
18. Shapiro, C., Wortmann, C.S., and Deloughery, R. (2002).
Manure Testing: What to Request? Nebraska Cooperative
Extension NF02-507. University of Nebraska-Lincoln, Lin-
coln, NE.
19. Peacock, B. (1996). Balancing the Nitrogen Budget. NG2-96.
University of California Cooperative Extension, University of
California, Tulare, CA.
20. Wells, K.L., Thomas, G.W., Sims, J.L., and Smith, M.S.
(2000). Managing Soil Nitrates for Agronomic Efficiency and
Environmental Protection. AGR-147. University of Kentucky
Cooperative Extension Service, University of Kentucky,
Lexington, KY.
700 SOIL N MANAGEMENT IMPACT ON THE QUALITY OF SURFACE AND SUBSURFACE WATER
21. Vitosh, M.L., Johnson, J.W., and Mengel, D.B. (1995). Tri-
state Fertilizer Recommendations for Corn, Soybeans, Wheat,
and Alfalfa. Ext. Bull. E-2567. Michigan State University,
East Lansing, MI.
22. Trautmann, N.M., Porter, K.S., and Wagenet, R.J. (1998).
Nitrogen: The Essential Element. Pesticide Fact Sheets and
tutorials. Cornell Cooperative Extension, Cornell University,
Ithaca, NY.
23. Stecker, J.A. (1993). Nitrogen management for no-tillage sys-
tems in Missouri. Agricultural publication G9175. Univer-
sity of Missouri Extension, University of Missouri-Columbia,
Columbia, MO.
24. Travis, J.W., Halbrendt, N., Hed, B., Rytter, J.L., Ander-
son, E., Jarjour, B., and Griggs, J. (2003). A practical
guide to the application of compost in vineyards. Avail-
able: http://fpath.cas.psu.edu/compostguide.pdf. Accessed 21
March 2004.
25. Khan, M. (1999). Nitrogen Loss from Soils. NDSU crop and
pest report, Issue 3. North Dakota State University, Fargo,
ND.
26. USEPA. (1973). Nitrogenous Compounds in The Environ-
ment. EPA-SAB-73-001. U.S. Government Printing Office,
Washington, DC.
27. Follett, J.R. and Follett, R.F. (2001). Utilization and
metabolism of Nitrogen by humans. In: Nitrogen in
the Environment: Sources, Problems and Management.
R.F. Follett and J.L. Hatfield (Eds.). Elsevier, Amsterdam.
pp. 65–92.
28. USEPA. (1976). Quality Criteria for Water. U.S. Government
Printing Office, Washington, DC, p. 256.
29. Sauer, T.J., Alexander, R.B., Brahana, J.V., and Smith, R.A.
(2001). The importance and role of watersheds in the
transport of Nitrogen. In: Nitrogen in the Environment:
Sources, Problems, and Management. R.F. Follett (Ed.).
Elsevier, Amsterdam, pp. 65–92.
30. Alexander, R.B., Smith, R.A., and Schwarz, G.E. (1995). The
regional transport of point and nonpoint source N to the Gulf
of Mexico. In: Proc. Gulf of Mexico Hypoxia Conf., December
5–6, Kenner, LA.
31. Lilly, J.P. (1991). Best Management Practices for Agricultural
Nutrients. Publication No. AG-439-20. North Carolina Coop-
erative Extension Service, North Carolina State University,
Raleigh, NC.
32. Owens, L.B. (1993). Impact of soil N management on the
quality of surface and subsurface water. In: Soil Processes
and Water Quality. R. Lal and B.A. Stewart (Eds.). Lewis,
Boca Raton, FL, pp. 137–163.
33. Meisinger, J.J., Hargrove, W.L., Mikkelsen, R.L., Williams,
J.R., and Benson, V.W. (1991). Effects of cover crops on
ground water quality. In: Cover Crops for Clean Water.
W.L. Hargrove (Ed.). Soil and Water Conservation Society,
Ankeny, IA, pp. 57–68.
34. Russelle, M.P. and Hargrove, W.L. (1989). Cropping systems:
Ecology and management. In: Nitrogen Management and
Groundwater Protection. R.F. Follett (Ed.). Elsevier, Amster-
dam, pp. 277–317.
35. Muller, J.C., Denys, D., Morlet, G., and Mariotta, A. (1988).
Influence of catch crops on mineral nitrogen leaching and its
subsequent plant use. In: Nitrogen Efficiency in Agricultural
Soils. D.S. Jenkinson and K.A. Smith (Eds.). Elsevier Applied
Science, New York.
36. Roos, D. (1998). Cover crops for sustainable vegetable produc-
tion. Available: http://www.geocities.com/RainForest/Canopy/
1118/lecture.html.
37. Meisinger, J.J., Shipley, P.R., and Decker, A.M. (1990). Using
winter cover crops to recycle nitrogen and reducing leaching.
In: Conservation Tillage for Agriculture in the 1990s. J.P.
Mueller and M.G. Wagger (Eds.). Spec. Bull. 90-1, North
Carolina State University, Raleigh, NC, pp. 3–6.
38. Hemphill, D.D. and Hart, J. (1991). Effect of Nitrogen Rate
and Placement on Broccoli Yield and Nitrogen Uptake. Report
to the Oregon Processed Vegetable Commission.
39. Dabney, S.M. (1998). Cover crop impacts on watershed
hydrology. J. Soil Water Conserv. 53(3): 207–213.
40. Kaspar, T.C., Radke, J.K., and Laflen, J.M. (2001). Small
grain cover crops and wheel traffic effects on infiltration,
runoff, and erosion. J. Soil Water Conserv. 56: 160–164.
41. Hartwig, N.L. (2003). Cover crops and conservation tillage for
soil erosion control on cropland. Pennsylvania State Univer-
sity, University Park, PA. Available: http://www.agronomy.
psu.edu/Extension/LivingMulch/sect9toc.htm.
42. Al-Kaisi, M. (2001). Value of crop rotation in nitrogen man-
agement. Integrated Crop Management Newsletter articles.
Iowa State University, Ames, IA.
43. Johnson, J.W., Welch, L.F., and Kurtz, L.T. (1975). Environ-
mental implications of N fixation by soybeans. J. Environ.
Qual. 4(3): 303–306.
44. Seelig, B.D. and Nowatzki, J. (2001). Working to Avoid
Nitrogen Contamination. AE-1218. NDSU Extension Service,
North Dakota State University, Fargo, ND.
45. Baker, J.L. and Melvin, S.W. (1994). Chemical management,
status, and findings. In: Agricultural Drainage Well Research
and Demonstration Project—Annual Report and Project
Summary. Iowa Department of Agriculture and Land
Stewardship and Iowa State University, Ames, IA, pp. 27–60.
46. Dickey, E., Harlan, P., Vokal, D., and Kisling Crouch, C.J.
(1982). Effects of Agricultural Runoff on Nebraska Water
Quality. Neb Guide G82-586. University of Nebraska
Cooperative Extension Service, University of Nebraska-
Lincoln, Lincoln, NE.
47. Hairston, J.E. (1995). Soil Management to Protect Water
Quality—Cropping and Nonstructural Agronomic Practices
for Soil Management. ANR-790-4.3.3. Alabama Cooperative
Extension Services, Auburn University, Auburn, AL.
48. USDA. (1996). Conservation Crop Rotation Effects on Soil
Quality. Natural Resources Conservation Service, Agronomy
Technical Note No. 2, August 1996.
49. Heath, M.E., Metcalfe, D.S., and Barnes, R.F. (1976). For-
ages—The Science of Grassland Agriculture, 3rd Edn. Iowa
State University Press, Ames, IA.
50. Weston, D. (1994). Water Quality: The Tillage Component.
AE-1072. NDSU Extension Service, North Dakota State
University, Fargo, ND.
51. Vanderwel, D. and Abday, S. (2001). An introduction to
water erosion control. Alberta Agriculture, Food and
Rural Development. Accessed 19 March 2004. Available:
http://www.agric.gov.ab.ca/agdex/500/72000003.html.
52. Legg, J.O. and Meisinger, J.J. (1982). Soil nitrogen budgets.
In: Nitrogen in Agricultural Soils. F.J. Stevenson (Ed.).
Agronomy Monograph 22, ASA-CSSA-SSSA, Madison, WI,
pp. 503–565.
53. Laflen, J.M. and Tabatabai, M.A. (1984). Nitrogen and
phosphorus losses from corn-soybean rotations as affected
by tillage practices. Trans. Am. Soc. Agric. Eng. 27: 58–63.
54. Owens, L.B. and Edwards, W.M. (1993). Tillage studies with
a corn/soybean rotation: Surface runoff chemistry. Soil Sci.
Soc. Am. J. 57(4): 1055–1060.
 SOIL PHOSPHORUS AVAILABILITY AND ITS IMPACT ON SURFACE WATER QUALITY
55. Thomas, G.W., Smith, M.S., and Philips, R.E. (1989). Impact
of soil management practices on nitrogen leaching. In: Nitro-
gen Management and Groundwater Protection. R.F. Follett
(Ed.). Elsevier, Amsterdam, pp. 247–276.
56. Baker, J.L. and Laflen, J.M. (1983). Water quality conse-
quences of conservation tillage. J. Soil Water Conserv. 38(3):
186–193.
57. Baker, J.L. and Johnson, H.P. (1983). Evaluating the effec-
tiveness of BMPs from field studies. In: Agricultural Man-
agement and Water Quality. G.W. Bailey and F.W. Schaller
(Eds.). Iowa State University Press, Ames, IA, pp. 281–304.
58. Thomas, G.W., Blevins, R.L., Phillips, R.E., and McMa-
hon, M.A. (1973). Effect of killed sod mulch on nitrate
movement and corn yield. Agron. J. 65: 736–739.
59. Quisenberry, V.L. and Phillips, R.E. (1976). Percolation of
surface-applied water in the field. Soil Sci. Soc. Am. J. 40:
484–489.
60. Edwards, W.M., Shiptalo, M.J., and Norton, L.D. (1988).
Contribution of macroporosity to infiltration into continuous
corn no-tilled watershed: Implications for contaminant
movement. J. Contam. Hydrol. 3: 193–205.
61. Hauck, R.D. (1984). Nitrogen in Crop Production. ASA-CSSA-
SSSA, Madison, WI.
62. Lal, R. and Stewart, B.A. (1994). Soil Processes and Water
Quality. Lewis, Boca Raton, FL.
63. Pierzynski, G.M., Sims, J.T., and Vance, G.F. (1994). Soils
and Environmental Quality. Lewis, Boca Raton, FL.
64. Smucker, D. (2001). Nitrogen management practices for
potato production. The Vegetable Growers News. July 2001.
65. Bundy, L.G. and Malone, E.S. (1988). Effect of residual profile
nitrate on corn response to applied nitrogen. Soil Sci. Soc.
Am. J. 52: 1377–1383.
66. El-Hout, N.M. and Blackmer, A.M. (1990). Nitrogen status of
corn after alfalfa in 29 Iowa fields. J. Soil Water Conserv. 45:
115–117.
67. McVay, K.A., Radcliffe, D.E., and Hargrove, W.L. (1989).
Winter legume effects on soil properties and nitrogen fertilizer
requirements. Soil Sci. Soc. Am. J. 53: 1856–1862.
68. Smith, S.J., Schepers, J.S., and Porter, L.K. (1990). Assess-
ing and managing agricultural nitrogen losses to the envi-
ronment. In: Advances in Soil Science. B.A. Stewart (Ed.).
Springer-Verlag, New York, pp. 1–43.
69. Randall, G.W. (1996). BMPs For nutrient management. Ill.
Fert. Conf. Proc., January 29–31, 1996.
70. Mahler, R.L. and Mahler, K.A. (1985). Nitrogen BMPs for
Small Farms. WQ-20. University of Idaho Cooperative
Extension system, University of Idaho, Moscow, ID.
71. Swdyky, R. (2003). Nitrogen—split application—A risk man-
agement tool for poor soil moisture conditions. Saskatchewan
Soil Conserv. Assoc. (SSCA) Newslett., Issue 37. Available:
http://ssca.usask.ca/Issue37/Rich1.htm.
72. Jacobsen, J.S., Tanaka, D.L., and Bauder, J.W. (1993). Spring
wheat response to fertilizer placement and nitrogen rate with
limited moisture. Commun. Soil Sci. Plant Anal. 24: 187–195.
73. Hairston, J.E. and Stribling, L. (1995). Fertilizer Manage-
ment to Protect Water Quality—Fertilizer Application Tech-
niques. ANR-790-4.4.3. Alabama Cooperative Extension Ser-
vices, Auburn University, Auburn, AL.
74. McFadden, V.J. and Melvin, S. (1991). Protecting our Water
Quality with Livestock Waste Management. PM 1428b. Iowa
State University Cooperative Extension Service, Iowa State
University, Ames, IA.
701
75. Chang, C., Sommerfeldt, T.G., and Entz, T. (1991). Soil
chemistry after eleven annual applications of cattle feedlot
manure. J. Environ. Qual. 20: 475–480.
76. Randall, G.W. and Schmitt, M.A. (1993). Best Manage-
ment Practices for Nitrogen Use Statewide in Minnesota.
FO-06125-GO.University of Minnesota, Cooperative exten-
sion service, University of Minnesota, MN. Available:
http://www.extension.umn.edu/distribution/cropsystems/
DC6125.html.
77. Hendrickson, S. (2002). BMPs (best management practices)
to Protect Groundwater Quality—Nitrogen Management.
University of Wisconsin Extension, University of Wisconsin,
Madison, WI.
78. Dahnke, W.C. and Johnson, G.V. (1990). Testing soils for
available nitrogen. In: Soil Testing and Plant Analysis, 3rd
Edn. R.L. Westerman (Ed.). Soil Science Society of America,
Madison, WI, pp. 127–139.
SOIL PHOSPHORUS AVAILABILITY AND ITS
IMPACT ON SURFACE WATER QUALITY
DIBYENDU SARKAR
University of Texas at San
Antonio
San Antonio, Texas
Phosphorus (P) is ubiquitous in soils and is an essential
element for plant and animal growth. Without enough
soil P, crop yields will be low. However, excess soil P
may become a part of surface runoff and can accelerate
eutrophication of receiving fresh waters. Eutrophication
has been identified as the main problem in U.S. surface
waters that has impaired water quality. Although P and
nitrogen (N) both influence eutrophication, P is generally
the limiting nutrient in freshwater bodies. The degree of P
input in surface waters depends on the soil availability of P
and the degree of transport of dissolved and particulate P,
especially in agricultural lands that have been fertilized
for a long period of time. The availability of P in soils
is controlled by various factors, such as soil pH, clay
content, soil organic matter, amorphous Fe/Al oxide and
carbonate content, soil moisture content, and degree of
soil compaction. The amount of available P is generally
estimated by a variety of soil-test methods, including
both chemical extractants and nondestructive techniques
simulating plant root action. Phosphorus transport from
the source to surface waters is controlled by the rate
of erosion and surface runoff. Hence, reducing P loss in
agricultural runoff may be brought about by implementing
proper source and transport control strategies. Popular
source control techniques include soil P management,
estimating thresholds for environmental risk assessment,
manure management, and refining feed rations, common
transport management techniques include conservation
practices such as reduced tillage, cover crops, and buffer
strips in critical areas of P export from a watershed.
Phosphorus makes up about 0.12% of the earth’s crust.
It forms complex compounds with a wide variety of
elements—about 150 minerals are known that contain
at least 0.44% P (1). Phosphorus is ubiquitous in soils. It
702 SOIL PHOSPHORUS AVAILABILITY AND ITS IMPACT ON SURFACE WATER QUALITY
is found naturally in all soils (1) and ranges between
less than 0.04% in the sandy soils of the Atlantic
and Gulf coastal plains to more than 0.3% in the
soils of the northwestern United States (2). Phosphorus
is an essential element for plant and animal growth.
Its input has long been recognized as necessary to
maintain profitable crop and livestock production (3).
Without enough soil phosphorus, crop yields will be low (4).
However, many soils in the United States reportedly
contain P well in excess of crop needs, which may enter
lakes and streams in runoff from landscapes that drain
to surface water bodies. Sources of P contamination of
surface waters are numerous and include both nonpoint
and point sources, such as agriculture, municipal sewage
treatment plants, individual septic treatment systems,
decaying plant material, runoff from urban areas and
construction sites, stream bank erosion, and wildlife (5).
Drainage of P-rich runoff to surface water bodies may
increase their biological productivity by accelerating
eutrophication, the natural aging of lakes or streams
brought on by nutrient enrichment. The United States
Environmental Protection Agency (USEPA) has identified
eutrophication as the main problem in U.S. surface
waters that have impaired water quality (6). Because
it causes increased growth of undesirable algae and
aquatic weeds, as well as oxygen shortages, eutrophication
restricts water use for fisheries, recreation, industry,
and drinking (7). In addition, associated periodic surface
blooms of cyanobacteria (blue-green algae) occur in
eutrophic drinking water supplies and may pose a
serious health hazard to livestock and humans. Although
eutrophication is a natural process, it can be hastened
by changes in the land use of a watershed that increase
the amount of nutrients added to an aquatic system (7).
Recent outbreaks of the dinoflagellate Pfiesteria piscicida
in the eastern United States, most notably in the
Chesapeake Bay, have been linked to high nutrient
levels in affected waters (3). Nitrogen (N) and P both
influence eutrophication, but P is the critical element
in most freshwaters (7). Due to low natural levels
of P, biological productivity in surface waters is usually
limited by P availability (5). The identification of P as
a major cause of eutrophication in surface waters led
to early emphasis on point sources delivering P to
surface waters, for example, sewage systems delivering P-
enriched wastewater. According to the National Research
Council (8), overall trends show about equal numbers of
U.S. rivers that have increasing and decreasing P loads.
In general, the decreases are linked to point-source
reductions, and the increases are linked to nonpoint
source increases that result from increased agricultural
land use (9). This review will address the concept of soil
availability of P in agricultural lands and the relationship
between P availability, P transport, and surface water
eutrophication. Popular practices for P management will
also be discussed.
PHOSPHORUS IN SOILS
The total P content of most surface soils is low, averaging
only 0.6%. This compares to an average soil content of
0.14% N and 0.83% potassium (K) (2). In most soils, the
P content of surface horizons is greater than that of the
subsoil. Except in special situations, added P tends to be
fixed by the soils where it is applied and allows little
movement down through the soil profile (7). In addition, P
is cycled from roots to aboveground parts of the plant and
redeposited in crop residues on the soil surface. This builds
up organic material and stimulates biological activity in
surface layers. In reduced tillage systems, fertilizers and
manures are incorporated only to shallow depths, thereby
exacerbating P buildup in the top 5 inches of soil (7).
Many factors influence the P content of soil. Among
these are (1) the type of parent material from which
the soil is derived, (2) the degree of weathering, and
(3) climatic conditions. Among other factors influencing
soil P levels are the rates of crop removal, erosion, and
P fertilization (2).
There are two primary forms of P in soil sys-
tems—organic and inorganic. The organic form of P in
many soils represents more than half of the total P.
Organic P has three different sink/sources (10), namely
nucleic acids (e.g., guanine, adenine, thymine, uracil),
phosphate esters of inositate (e.g., phytin), and phospho-
lipids (e.g., lecithin). Biological processes in the soil, such
as microbial activity, tend to control the mineralization
of organic P. Mineralization is the breakdown or conver-
sion of organic P to inorganic solution P (7). The inorganic
form of P in soils (orthophosphates) exist in various forms:
(1) oxide–phosphate complexes, such as clay mineral edge-
P complexes, and (2) minerals of apatite (Ca phosphates),
Mn phosphates, Fe phosphates, Al phosphates, and to a
lesser degree, pyrophosphate (10).
AVAILABILITY OF SOIL PHOSPHORUS
Loosely defined, P availability refers to the amount of
soil P that is readily available for uptake by plants. If the
amount of available P exceeds plant needs, the excess P
may become available for drainage to surface water
bodies as part of the surface runoff, thereby degrading
surface water quality. As a rule of thumb, soluble P is
considered the most available. However, the solubility
of many inorganic P compounds, as well as organic P, is
generally low, and only small amounts of soil P are in
solution at any one time. In fact, most soils contain less
than a 1 lb acre−1 of soluble P (2). Soluble P in soils,
either from fertilizer or natural weathering, reacts with
clay, iron, and aluminum compounds in the soil and is
converted to less mobile (hence, less ‘‘available’’) forms by
the process of P fixation. Because of the fixation processes
(e.g., sorption and precipitation), P moves very little in
most soils, stays close to its place of application, and
crops seldom absorb more than 20% of fertilizer P during
the first cropping system after application. This fixed,
residual P remains in the rooting zone and becomes slowly
available to succeeding crops (2).
Factors Controlling Availability of Soil Phosphorus
Soil P availability is influenced by the following factors (2):
• Soil pH: Soil pH affects both adsorption and
precipitation of phosphorus. Precipitation of P as
 SOIL PHOSPHORUS AVAILABILITY AND ITS IMPACT ON SURFACE WATER QUALITY
slightly soluble Ca phosphates occurs in calcareous
soils whose pH values are around 8.0. Under acidic
conditions, P is precipitated as Fe or Al phosphates
that have low solubility. From a precipitation
standpoint, the maximum availability of P generally
occurs in the pH range of 6.0 to 7.0 (2). As a typical
oxyanion, phosphate tends to be adsorbed more at
lower pH than at higher pH. Hence, P generally
tends to be more available in high pH soils, unless
precipitated as low-solubility basic salts. Amorphous
Fe and Al oxides that have high specific surface
strongly adsorb P in soil systems (11).
• Soil Organic Matter: High amounts of organic
matter in soils promote mineralization of organic P,
thereby providing available P for plant growth.
Organic matter also acts as a chelating agent,
thereby preventing the formation of insoluble iron
phosphates. Application of organic materials such as
manure, plant residues, or green manure to soils
that have high pH values increases the availability
potential of mineral forms of P in the soil (2).
• Clay Content: Fine-textured soils such as clay and
clay loams have greater P-fixing capacity than
coarse-textured sandy soils. Among soils of similar
clay mineralogy, P fixation increases as clay content
increases and crystallinity decreases (12). Among
clay minerals, 1:1 type clays (kaolinite) have
greater P-fixing capacity than 2:1 type clays (e.g.,
montmorillonite, illite). Soils formed under high
rainfall and high temperatures generally contain
large amounts of kaolinitic clays and therefore have
relatively lower soil P availability.
• Soil Aging: Fixation of soil P increases with time
of contact between soluble P and soil particles.
Consequently, more efficient use of fertilizer P is
generally obtained by applying the fertilizer shortly
before planting the crop (2). As soil ages, the
possibility that P binds to soil particles increases,
thereby decreasing its potential availability.
• Soil Compaction and Soil Moisture: Soil compaction
reduces the soils oxygen supply and decreases the
ability of plant roots to absorb soil P. Compaction
reduces aeration and pore space in the root zone. This
reduces P uptake and plant growth. Compaction also
decreases the soil volume that plant roots penetrate
and limits their access to soil P. Many studies have
been conducted on the effect of soil moisture on P
availability and response to applied P (13). Power
et al. (14) reported that 53% of the variation in the
fertilizer response of spring wheat on medium P soils
resulted from variation in soil moisture. MacKay and
Barber (15) showed that P availability decreased at
soil moisture contents above and below field capacity
because of the effects on both P diffusion and root
growth rates. Uptake of P by corn dropped more than
total plant weight when soil moisture was decreased
below field capacity, indicating that the decrease in
uptake was not likely due only to decreased demand.
In addition to the factors discussed before, several other
factors define soil P availability (13). These factors include,
703
but are not limited to soil solution P concentration, the
amount of solid phase P that can enter the soil solution, the
dissolution/desorption rate, the diffusion rate, Fe content,
carbonate content, carbonate specific surface area, and
organic P content (16–19).
Phosphorus Availability Indices
To evaluate P availability in soils, numerous soil tests
have been developed that extract varying amounts of
P, depending on the types of extractants used. The
extractants can be generally classified into several
categories (20):
• Water or unbuffered salt solutions (e.g., CaCl2 ).
• Dilute concentrations of weak acids (e.g., lactate,
acetate) with or without a complexing agent (e.g.,
F− , EDTA).
• Dilute concentrations of strong acids (e.g., HCl,
H2 SO4 ) with or without a complexing agent (e.g.,
F− , lactate, EDTA).
• Buffered alkaline solutions (e.g., NaHCO3 , NH4 H
CO3 ) with or without a complexing agent (e.g.,
EDTA).
• Anion exchange resin or iron oxide impregnated filter
paper strips.
• Isotopic exchange with 32 P.
Water derives readily soluble P; unbuffered/neutral salt
solutions derive easily exchangeable P (21). Dilute strong
acid solutions solublilize Ca-P, Al-P, and to a lesser extent,
Fe-P (22). Complexing agents are generally introduced
to prevent readsorption of P by Fe oxides (20). Anion
exchange resins and iron oxide impregnated filter papers
(nondestructive to soil components in contrast to chemical
extractants) function as sinks that simulate the action
of plant roots by continuously removing dissolved P from
the soil solution. The quantity of P that is isotopically
exchangeable within a specified time interval gives an
estimate of labile (hence, available) soil P (20).
Criteria for a suitable soil test have been detailed in
several reports (13). Thomas and Peaslee (23) felt that the
extract should (1) remove a reproducible and consistent
proportion of the soil’s available P and (2) reflect the
extent and nature of reactions between soil and added P.
Barrow (24) also felt that a soil test was an integration of
several factors affecting soil P status (e.g., soil P buffering
capacity, P quantity, soil P reaction time) into one
numerical value. Kamprath and Watson (25) noted that
(1) the extractant should extract a proportionate fraction
of available P from soils that differ in other properties,
(2) the procedure should work with reasonable accuracy
and speed, and (3) the soil test P value determined should
be correlatable with P uptake. On the basis of the major
soil types and their properties, various states in the United
States have adopted different soil test methods to predict
available P. A field rating system, the ‘‘P index,’’ has been
developed to assess the potential for soil P to contribute to
surface water pollution (26).
TRANSPORT OF SOIL PHOSPHORUS
Phosphorus enters lakes and streams in runoff from
landscapes that drain to those surface water bodies (5).
704 SOIL PHOSPHORUS AVAILABILITY AND ITS IMPACT ON SURFACE WATER QUALITY
The eutrophication process is clearly accelerated by the
addition of P to surface waters. All forms of P (soluble,
adsorbed, precipitated, and organic) are susceptible to
transport from source soils to water bodies. Transport
of soil P occurs primarily through surface flow, although
the background levels of P entering streams and lakes
via subsurface flow certainly reflect the impacts of
land use. For example, in soils that have considerable
slope, surface or overland flow can infiltrate into the
soil during movement down the slope, move laterally
through a conduit in the soil, and reappear as surface
flow (7). However, in most cases, the concentration of
P in subsurface flow, it has been found is quite low
and well below the eutrophication threshold. In overland
flow (surface runoff), P may be dissolved in runoff water
(soluble/dissolved P or DP). The small quantity of soluble
or readily desorbable P in most soil environments is
due to the low solubility of phosphate minerals and
the considerable adsorption capacities of clays and Fe/Al
oxides for P. These factors result in the majority of total P
transport that occurs as particulate P (PP). Particulate P
enters streams and other surface waters, first undergoing
a solubilization reaction (e.g., desorption) before becoming
available for aquatic biota. The mechanisms involved
in soluble P transport include initial desorption or
dissolution of P bound by soil particles, followed by water
movement from the source soil to a stream or a lake that
later intercepts a sensitive water body (27).
Factors Controlling Surface Transport of Soil Phosphorus
Two major factors influence P losses from the landscape
and movement to surface water bodies (5):
• Erosion: Particulate P is the primary form of
P contributing to degradation of surface water qual-
ity. Erosion largely determines PP movement in the
landscape. Sources of PP in surface waters include
eroding surface soil, stream banks, channel beds,
and plant materials. The contribution of eroded soil
material to P enrichment of surface waters is rather
complex (5). First, part of the sediment eroded from
the landscape deposits at the toe slope in depres-
sional areas, along field boundaries, or in grassy
riparian zones. Thus, sediment yield from the field to
a water body is only a portion of that eroded from the
slope. Second, during the movement of sediment,
the finer fractions are preferentially transported,
whereas larger particles tend to settle out. There-
fore, the P concentration and reactivity of eroded
particulate material is usually greater than that of
the source soil. The enrichment of P may increase
as much as six-fold as the relative movement of fine
to coarser particles increases. For this reason, fine-
textured soils (clay loam, silty clay loam, silty clay, or
clay) have a higher potential of supplying P to surface
waters if erosion occurs (5).
• Runoff: As discussed earlier, runoff water across the
soil surface can contain significant concentrations
of DP. In addition to the readily available P loss
from surface soils via desorption/dissolution during
rainfall, DP can also be lost from standing vegetation
via spring snowmelt because P contained in the tissue
is released due to breakdown of plant cells by freezing
and thawing. The removal of P from plant residue
by overland drainage may account for differences
among watersheds and seasonal fluctuations in
P movement (5). Concentrations of DP in subsurface
flow are low because the P-deficient subsoils sorb
much of soluble P contained in the water percolating
through the soil profile. Exceptions may occur in
organic, highly permeable, peaty, or waterlogged
soils, which tend to fix less P (7).
There is a well-defined relationship between the DP
and PP fractions of total P in runoff as a function
of soil erosion. As erosion increases, the PP fraction
of total P increases, whereas the DP fraction decreases
significantly (5). Irrigation, especially furrow irrigation,
can significantly increase the potential for soil and water
contact and therefore, can increase P loss by both surface
runoff and erosion in return flows. Furrow irrigation
exposes unprotected surface soil to the erosive effect of
water movement. The process of irrigation also has the
potential to increase greatly the land area that can serve as
a possible source for P movement, a fact that is especially
important in the western United States (3) where the
surface water bodies suffer from an increased threat of
eutrophication.
ENVIRONMENTAL MANAGEMENT OF SOIL PHOSPHORUS
When looking at management and/or remedial practices
to minimize the environmental impact of P, several
important factors must be considered. To cause an
environmental problem, there must be a source of P (i.e.,
high soil P levels, manure or fertilizer applications, etc.),
and it must be transported to a sensitive location (i.e.,
by runoff, erosion, or to a lesser extent, leaching). Hence,
reducing P loss in surface runoff (thereby lowering the
eutrophication potential of surface waters) may be brought
about by source and transport control strategies (3).
Source Management
Source management attempts to minimize soil P buildup
by controlling the amount of P applied to a given area.
One of the most highly recommended avenues for source
management is by measuring (and monitoring) soil test P
in agricultural lands and estimating threshold levels for
environmental risk assessment.
Decades of fertilization at rates exceeding those of
crop removal have resulted in widespread increases in
the P status of many agricultural soils, even to levels
considered ‘‘excessive.’’ Environmental concern has forced
many states to consider developing recommendations for
P applications and watershed management based on
the potential for P loss in agricultural runoff (3). As
a result, well-defined ‘‘agronomic’’ and ‘‘environmental’’
thresholds for soil test P have been developed for many
states. The environmental threshold levels range from
two times in Michigan (40 ppm P-agronomic and 75 ppm
P-environmental) to four times in Texas (44 ppm vs.
200 ppm). However, one still cannot accurately project
that a soil test level above an expected crop response level
 SOIL PHOSPHORUS AVAILABILITY AND ITS IMPACT ON SURFACE WATER QUALITY
exceeds crop needs and therefore is potentially polluting.
What will be crucial in terms of managing source P based
in part on soil test levels will be the interval between the
critical soil P value for crop yield and runoff P (3).
Among the other popular source management tech-
niques currently being implemented in farmlands are
manure management (e.g., composting, lime or alum
amendment, separation of solids from liquids); reduction
of off-farm inputs of P in animal feeds (3); and controlling
the rate, method and time of P application (5).
Transport Management
Phosphorus loss via erosion and runoff may be reduced
by increasing residue cover on the soil surface through
conservation tillage (5). Results from a flat, tiled, clay
loam site in Michigan show significantly lower losses
of sediment, total P, and soluble P for chisel tillage
compared to moldboard plow tillage. However, sediment
reductions were much greater than the reductions in total
or soluble P. In other research, DP concentration in the
runoff from no-till practices has often been greater than
that from conventional practices (5).
Strategically placed and properly designed filter strips,
it has been shown, effectively reduce erosion and
P movement, especially PP (5). However, the strips have
had little consistent effect on reducing DP concentrations.
Other common practices for reducing potential P
transport include terracing, contour tillage, riparian
zones, cover crops, and impoundments (settling basins).
Basically, all these methods reduce the impact of rainfall
on the soil surface, reduce runoff volume and velocity,
and increase soil resistance to erosion. Most of these
practices are more efficient in reducing PP than DP, and
none of them should be relied on as the primary or sole
practice to reduce P losses in surface runoff (3). Several
researchers report little decrease in lake productivity (i.e.,
eutrophic plant growth) with reduced P inputs, following
implementation of conservation measures (28,29). Thus,
the effect of remedial measures in the contributing
watershed will be slow for many cases of poor water
quality (7).
BIBLIOGRAPHY
1. Cathcart, J.B. (1980). World phosphate reserves and
resources. In: The Role of Phosphorus in Agriculture.
F.R. Khasawneh, E.C. Sample, and D.J. Kamprath (Eds.).
ASA-CSSA-SSSA, Madison, WI.
2. Griffith, B. (2002). Phosphorus. Efficient Fertilizer Use
Manual. IMC Global Inc.
3. Sharpley, A.N. et al. (1999). Agricultural Phosphorus and
Eutrophication. ARS-149. USDA.
4. Stout, W.L., Sharpley, A.N., and Pionke, H.B. (1998). Reduc-
ing soil phosphorus solubility with coal combustion by-
products. J. Environ. Qual. 27: 111–118.
5. Randall, G. et al. (1998). Phosphorus. Transport to and
Availability in Surface Waters. University of Minnesota
Extension Service Publication, Minneapolis, MN.
6. U.S. Environmental Protection Agency. (1996). Environmen-
tal Indicators of Water Quality in the United States. EPA
841-R-96-002.
705
7. Sharpley, A.N. (1999). Managing Phosphorus for Agriculture
and the Environment. College of Agricultural Sciences Coop-
erative Extension Services Publication, The Pennsylvania
State University, University Park, PA.
8. National Research Council. (1993). Soil and water quality.
An Agenda for Agriculture. National Academy Press,
Washington, DC.
9. Voss, R. and Griffith, B. (1998). Phosphorus and Surface
Water. Illinois Fertilizer and Chemical Association Report,
St. Anne, IL.
10. Evangelou, V.P. (1998). Environmental soil and water
chemistry. Principles and Applications. John Wiley & Sons,
New York.
11. Peters, J.M. and Basta, N.T. (1996). Reduction of excessive
bioavailable phosphorus in soils by using municipal and
industrial wastes. J. Environ. Qual. 25: 1236–1241.
12. Sanchez, P.A. and Uehara, G. (1980). Management consider-
ations for acid soils with high phosphorus fixation capacity.
In: The Role of Phosphorus in Agriculture. F.R. Khasawneh,
E.C. Sample, and D.J. Kamprath. (Eds.). ASA-CSSA-SSSA,
Madison, WI.
13. Fixen, P.E. and Grove, J.N. (1990). Testing soils for phos-
phorus. In: Soil Testing and Plant Analysis, 3rd Edn.
R.L. Westerman (Ed.). SSSA Book Series No. 3. SSSA, Madi-
son, WI.
14. Power, J.F., Brown, P.L., Army, T.J., and Klages, M.G.
(1961). Phosphorus responses by dryland spring wheat as
influenced by moisture supplies. Agron. J. 53: 106–108.
15. MacKay, A.D. and Barber, S.A. (1985). Soil moisture effects
on root growth and phosphorus uptake by corn. Agron. J. 77:
519–523.
16. Soper, R.J. and El Bagouri, I.H.M. (1964). The effect of soil
carbonate level on the availability of added and native
phosphorus in some calcareous soils. Can. J. Soil Sci. 44:
337–344.
17. Fixen, P.E. and Ludwick, A.E. (1982). Residual available
phosphorus in near-neutral and alkaline soils: II. Persistence
and quantitative estimation. Soil Sci. Soc. Am. J. 46:
335–338.
18. Holford, I.C.R. (1977). Soil properties related to phosphate
buffering in calcareous soils. Commun. Soil Sci. Plant Anal.
8: 125–137.
19. Cole, C.V., Innis, G.S., and Stewart, J.W.B. (1977). Simula-
tion of phosphorus cycling in semi-arid grasslands. Ecology
58: 1–15.
20. Kuo, S. (1996). Phosphorus. In: Methods of Soil Analysis.
D.L. Sparks. (Ed.). Part III: Chemical Methods. SSSA Book
Series No. 5. SSSA, Madison, WI.
21. O’Connor, G.A. and Sarkar, D. (1999). Fate of Land Applied
Residuals-Bound Phosphorus. Final Report, DEP WM 661.
22. Nelson, W.L., Mehlich, A., and Winters, E. (1953). The
development, evaluation, and use of soil tests for phosphorus
availability. Agronomy 4: 153–188.
23. Thomas, G.W. and Peaslee, D.W. (1973). Testing soils for
phosphorus. In: Soil Testing and Plant Analysis. L.M. Walsh
and J.D. Beaton. (Eds.). SSSA Book Series No. 3. SSSA,
Madison, WI.
24. Barrow, N.J. (1980). Evaluation and utilization of residual
phosphorus in soils. In: The Role of Phosphorus in Agriculture.
F.R. Khasawneh, E.C. Sample, and D.J. Kamprath (Eds.).
ASA-CSSA-SSSA, Madison, WI.
25. Kamprath, E.J. and Watson, M.E. (1980). Conventional soil
and tissue tests for assessing the phosphorus status of soils.
In: The Role of Phosphorus in Agriculture. F.R. Khasawneh,
706 SOIL WATER ISSUES
E.C. Sample, and D.J. Kamprath. (Eds.). ASA-CSSA-SSSA,
Madison, WI.
26. Sims, J.T. (1993). Environmental soil testing for phosphorus.
J. Prod. Agric. 6: 501–507.
27. Pierzynski, G.M., Sims, J.T., and Vance, G.F. (1994). Soils
and Environmental Quality. Lewis, Boca Raton, FL.
28. Young, T.C. and DePinto, J.V. (1982). Algal-availability of
particulate phosphorus from diffuse and point sources in the
lower Great Lakes basin. Hydrobiologia 91: 111–119.
29. Gray, C.B.J. and Kirkland, R.A. (1986). Suspended sediment
phosphorus composition in tributaries of the Okanagan
Lakes, BC. Water Res. 20: 1193–1196.
SOIL WATER ISSUES
ELIOT C. ROBERTS
Rosehall Associates
Sparta, Tennessee
Landscape plants and turfgrass can eliminate erosion
and conserve water most effectively by considering many
variables in the makeup of the landscape. These variables
include the sizes of mineral components of soil, the
components of the root zone, soil moisture capacity, soil
texture, soil pH, the soil’s biodegradable capacity, and its
mulching capacity. All these factors are explained in this
entry.
Because water and soil are the lifeblood of the planet,
any policy that governs water use or proposes water
conservation must incorporate the dynamic relationship
that takes place between plants, the soil and water.
The soil on which all plants grow is a highly valuable
natural resource. We have a responsibility to protect it
from erosion and to conserve and enrich it for future
generations, just as we have a responsibility to use our
supplies of water judiciously and maintain or improve
its quality. Landscape plants generally and turf-grasses
specifically can help achieve both of these goals.
To realize these benefits, it’s helpful to look at what
scientists have learned about the characteristics of soil
such as texture, size of soil particles, the presence of
organisms, capacity to hold moisture, acidity–alkalinity,
and the presence of pollutants.
Soils are not homogenous, inert materials. They are
composed of mineral particles that include sand, silt, and
clay, as well as living and dead micro/macroflora and
fauna, chemicals, air, and water in various percentages.
SIZES OF THE MINERAL COMPONENTS OF SOIL
From largest to smallest, mineral components of soil
consist of the following:
• Stones = 10 to 100 millimeters in diameter (25
per inch).
• Gravel = 2 to 10 millimeters
• Coarse sand = 0.2 to 2 millimeters
• Fine sand = 0.02 to 0.2 millimeter
• Silt = 0.002 to 0.02 millimeters
• Clay = smaller than 0.002 millimeter. Clay particles
are so small that they are measured in microns (0.002
millimeter is equivalent to 2 microns)
In terms of comparative size, if we enlarge a clay
particle to the size of an apple, then a silt particle on the
same scale would be the size of a limousine and a medium
sand particle would be the size of the White House in
Washington D.C. Because soil particles have relative sizes
this small, there are many of them. For example, a pound
of medium sand contains about 2.5 million particles, a
pound of silt contains more than 2.5 billion particles, and
a pound of clay contains more than 40 trillion particles.
On the basis of total particle surface, one pound of sand
would account for 20 square feet; silt would present 220
square feet of surface, and clay would have 5500 square
feet of particle surface (Fig. 1).
THE MULTIPLE COMPONENTS IN A LANDSCAPE
ROOT ZONE
Also present in most soils are high microbe counts within
the root zone. Often there are more than 900 billion for
each pound of soil. In each 1000-square-foot surface to
a 6-inch root-zone depth, there will be a total of about
45 quadrillion organisms. As these organisms complete
their life cycle and die, they deposit into the soil some
10 pounds of nitrogen, 5 pounds of phosphorus, 2 pounds
of potassium, a half-pound of calcium, a half-pound of
manganese, and one-third of a pound of sulfur for each
100 pounds of dead organisms on a dry-weight basis.
Soil microbiological processes also convert organic
matter into humus. This is an ongoing reaction of great
importance. Humus helps to form and stabilize soil
aggregates that are essential for deep and extensive root
growth. Humus also contributes to the process within the
soil that holds and releases nutrients for plant growth.
In addition, many small animals known as soil fauna
occupy the root zones of plants and contribute to the living
nature of the soil. Depending upon soil conditions that
are favorable for these macroorganisms, from 1 million to
2 million may be present for each 1000 square feet of root
zone. The live weight of these organisms would range from
15 pounds to 30 pounds per 1000 square feet.
Of course, the water molecule associated with the soil
is exceptionally small. One fluid ounce of water con-
tains approximately 1,000,000,000,000,000,000,000,000
(24 zeros, or one trillion trillion) molecules.
GRASS PLANTS HAVE A TREMENDOUS POTENTIAL FOR
ROOT GROWTH
Grasses fit right in with the sizes and numbers of
soil particles found within this fascinating system. For
example, there may be as many as 35 million individual
grass plants per acre, or about 800,000 per 1000 square
feet. No other type of plant culture involves such crowding.
Roots grow down into the soil, and it is there that grass
plants have a tremendous potential for root growth—up to
375 miles of roots from one plant and as many as 14 million

Figure 1. In terms of comparative size, if we enlarge a clay
particle to the size of an apple, then a silt particle on the same
scale would be the size of a limousine, and a medium sand particle
would be the size of the White House in Washington, DC.
individual roots that may have a total surface area of 2500
square feet. Thus, root numbers and surfaces fit well
within the very small spaces surrounding aggregated soil
particles.
It is important to understand soil properties so we
can appreciate how turfgrass needs moisture to grow and
enhance the environment (Fig. 2).
SOIL’S CAPACITY TO HOLD MOISTURE
Soils differ in their capacity to hold moisture. Heavier
clay and silt soils hold more moisture. Sandy soils can lose
moisture through leaching as it runs through the root zone
and down into the subsoil. Grasses with well-developed,
Nonclipped
Clipped
1/4" 3/4" 1.5"
Kentucky blue
Figure 2. Higher mowing of turfgrass promotes a good root
system. For Kentucky bluegrass, 1.5 inches is about right.
SOIL WATER ISSUES 707
deep root systems add sufficient organic matter to help
hold moisture in the soil and thus prevent leaching.
THE INFLUENCE OF SOIL TEXTURE ON WATER
PENETRATION
The texture of the soil (as determined by the amount of
sand, silt and clay) and the amount of thatch (organic
deposit between green leaves and roots) influence the
speed of water penetration into the soil. In general, heavy
soils have many smaller pore spaces and take water in
slowly. Sandy soils with fewer but larger pore spaces
take moisture more rapidly, unless they are inherently
hydrophobic or hard to wet. Soils and thatch that are
hard to wet must be watered slowly with small amounts
of water applied over longer periods of time to prevent
runoff. Sandy soils require less water to penetrate to a
given depth. Loam soils need intermediate amounts of
water, and clay soils require more water to reach the
same depth. As solid particles in the soil decrease on
a percentage basis, moisture-holding capacity increases,
and soil aeration decreases.
THE DEGREE OF ACIDITY AND ALKALINITY IN SOILS
Soil may either be acid, neutral, or alkaline. Soil pH (the
measurement of the degree of acidity and alkalinity) is
influenced by soil properties, biological influences, and
climatic influences. Under acidic soil conditions, silt and
clay particles tend to exist as individual units. Under more
alkaline soil conditions, where calcium and magnesium
are more plentiful, clay and silt particles group together
to form granules. These provide improved soil structure,
which results in more favorable balances of air and water
in the soil. Where soils are acid and have poor structure,
water penetration is much slower.
Soils become acid as carbon dioxide changes to carbonic
acid in the soil, acid-reacting fertilizers are used on a
continuing basis, or acid rainfall lowers soil pH values.
Often a combination of all three of these causes occurs.
In addition to the effect of acidity on physical soil
properties, nutrient fixation and availability also are
modified, depending on the degree of acidity or alkalinity.
For example, phosphates are most available from pH 5.5
to 7.5. Above and below these levels, phosphates are tied
up with other minerals, and their availability is reduced.
Regular soil tests can determine the need for lime or sulfur
or for fertilizer mineral nutrients for specific plant types.
SOIL AS A BIODEGRADABLE AGENT
Biologically healthy soil is the best known medium for
decomposing of all sorts of organic compounds, including
pesticides and pollutants transported by air and water.
These chemicals are known to be biodegradable. This
is an ongoing process, which changes these substances
into harmless compounds plus carbon dioxide and water.
Limited prescribed use of pesticides is not harmful to
beneficial soil organisms and should continue to be an
important, well-accepted part of plant culture.
708 WATER SPREADING
Figure 3. Dr. Thomas Watschke, Pennsylvania State University,
created this highly controlled watershed site documenting that
established turfgrass has a dramatic, positive effect on reducing
nutrient and pesticide pollutants from water runoff.
Water and any pollutants associated with it infiltrate
into the ground more quickly on grass-covered soils than on
any other surface. Thus, runoff is diminished. Infiltration
rates may be as high as 7 inches an hour on sandy soils and
as low as 0.10 inch per hour on clay soils. Thus, recharge
of purified groundwater is an important benefit. An acre
left in open space provides an average of 600,000 gallons
of recharge each year in humid regions. Grasses may use
up to 10% of the water infiltrated, leaving 90% to recharge
the local aquifer (Fig. 3).
GRASS GROUNDCOVER PROVIDES A LIVING MULCH
Good horticultural practice involves using mulches to
conserve soil moisture and increase soil productivity.
Unlike many landscape plants that are either widely
spaced or simply annual in their growth habits, grass
groundcover provides a living mulch over the soil surface.
This is essentially perennial and provides long lasting soil
and water conservation benefits.
Instead of viewing green-lawn groundcovers as static
liabilities, these areas can be seen as dynamic, ever-
changing populations of plants and animals living within
and above the soil. All grasses are natural soil builders.
Particularly in residential areas, lawns and landscapes
help sustain the soil. Within the soil are large populations
of micro- and macroorganisms that are highly competitive.
These create a living, moist soil environment best suited
to sustaining productive landscape soils, while at the same
time purifying our water supply.
WATER SPREADING
PEYMAN DANESHKAR ARASTEH
Soil Conservation and
Watershed Management
Research Institute (SCWMRI)
Tehran, Iran
INTRODUCTION
In arid and semiarid regions, irrigation water shortage
forces farmers to use floodwater for crop production. Water
spreading is a form of runoff farming. It is a form of
flood irrigation accomplished by diverting and spreading
runoff from sloping areas over adjacent flood plains (1).
Water spreading involves the construction of basins, pits,
or barriers in or near natural stream channels to impound
water and cause it to infiltrate the ground surface rather
than leave the basin as surface runoff (2,3).
In other words, diverting or collecting runoff from
natural channels, gullies, or streams with a system of
dams, dikes, ditches, or other means, and spreading it
over relatively flat areas is called water or floodwater
spreading (1,4,5).
In a large part of the world, runoff water harvesting
is best known and practiced in the semiarid areas (6,7).
Runoff farming, which is identical with water harvesting
for irrigation purposes, was already used for more than
2000 years. For example, runoff irrigation facilities in
the Negev Desert in Israel were built 1300 to 2900
years ago (8). Flood irrigation also has been practiced in
other parts of the world, such as Iran, for many years.
Traditional flood irrigated farms, called band-sar, can
be observed in 400,000 ha of Khorasan, the northeast
province of Iran. Band-sar controls about one-fiftieth
of 200–250 mm annual precipitation and one-fourteenth
of 14,000 MCM annual surface runoff of Khorasan
Province, Iran (9). Many sources exist addressing the
floodwater spreading affects on soil, groundwater, wild
habitats, etc. Baghernejad (10) studied physicochemical
and micromorphological changes in topsoil because of
flood spreading in Damghan Playa, Iran. Kowsar and
Yazdian (11) investigated the fate of the dissolved nitrogen
in floodwater that reached the watertable in Bisheh Zard
Basin, Iran, because of flood spread over the land. The
floodwater was from a geological formation containing
NO3 and NH4 . They found that the NO3 exceeded the
N required by plants of the region (rangelands and forest
area), and purification of water must be in order to prevent
the groundwater pollution. Nejabat (12) reported that a
2500 ha floodwater spreading system affected about 7500
ha of adjacent area in Gareh Bygon, Iran. He used a
technique on the base satellite imagery to determine
the increase in cultivated and rangeland areas and the
decrease in wind erosion in water spread area.
Floodwater spreading has several advantages that are
summarized by FAO (13) as follows:
a. Controlling floodwater damage
b. Reducing erosion and sedimentation
c. Increasing groundwater recharge
d. Improving plant cover and vegetation of the land
e. Increasing economic return for people
f. Increasing rangeland productivity
g. Increasing animal and agricultural production
h. Increasing soil fertility
i. Reclamation of renewable resources
j.Land use change by providing an optimized model
for use of flow efficiency land and farming
k. Job creation by fruit trees plantation and increasing
the income of local people
 WATER SPREADING 709

l. Improving the ecological balance of the environment Diversion

m. Reclamation of the pasture and vegetation cover facility Flow direction
n. Providing ground for public participation Natural channel
o. Providing ground for sustainable development

Floodwater spreading also has some disadvan- Delivery

channel
tages (14), listed as follows:

a. Affected by water shortage and climatic risks

b. Affected by farmers’ experiences rather than on the Spreader
channel Drain
scientifically well-established techniques
c. The possible conflicts between upstream and down-
stream users
d. Possible harm to fauna and flora adapted to running
waters and wetlands Figure 1. Schematic layout of a floodwater spreading system.
e. Requirement of relatively large labor inputs
f. Need to a relatively large natural watershed area
g. Need to appropriate investment and policy reforms
to enhance the contribution of rainfed agriculture

In this article, water spreading as an irrigation

technique will be introduced.

FLOODWATER SPREADING

Water harvesting refers to methods used to collect water

from sources where the water is widely dispersed and
quickly changes location or form and becomes unavailable
or that is occurring in quantities and at a location where it Photo by: R. Hasanzadeh
is unusable unless some intervention is practiced to gather
the water to locations where it can provide benefits (15). Figure 2. Diversion dam of Chikhab flood spreading system,
Dehloran, Iran.
A number of water harvesting techniques exist that are
practiced in many of the water scarce areas of the world,
including (15,16):

a. Roof rainwater collection

b. Terracing
c. Small dams
d. Runoff enhancement
e. Runoff collection
f. Water holes and ponds
g. Tanks
h. Floodwater spreading

Floodwater spreading or spate irrigation is one of the

common methods of runoff farming in arid and semiarid
areas with rainfall intensity more than the infiltration
rate of the catchment. USDA emphasized that a major
difference between the water spreading system and other
surface irrigation techniques, such as border and basin
systems, is that the water spreading system is designed
to meet precipitation and runoff conditions of the area
and apply runoff to cropped fields, whereas other systems
are designed to deliver water in accordance with plant
needs (17).
Floodwater spreading has at least an inlet to divert flow
from natural channels to the system, a delivery channel,
and some contour spreader channels to spread water on
Photo by: R. Hasanzadeh
Figure 3. Intake and sediment control facilities of Poldasht flood
spreading system, Makoo, Iran.
the ground as overland flow. Figure 1 shows a schematic
layout of a floodwater spreading system. Figures 2, 3, 4, 5
show a diversion dam and intake system with sediment
control facility, a delivery channel, and the spreaders of
flood spreading system, respectively.
After a flood is diverted from the natural waterway,
it reaches the spreading fields by the diversion or
delivery channel and contour spreader. Sheet flow over
710 WATER SPREADING

Photo by: R. Hasanzadeh

Figure 4. Diversion or delivery earthen canal of Sarchahan flood Photo by: R. Hasanzadeh
spreading system, Hajiabad, Iran. Figure 7. A cemented sandbag-type spillway of Poldasht flood
spreading, Makoo, Iran.

Photo by: R. Hasanzadeh

Figure 5. Contour spreader and downstream cultivated field of Photo by: R. Hasanzadeh
Poldasht flood spreading system, Makoo, Iran.
Figure 8. A gabion-type spillway of Chikhab flood spreading,
Dehloran, Iran.

the spreading area provides sufficient time to infiltrate

more and more water to the soil and completes water
holding capacity of the plant root zone. Excess water
will be drained to a downstream contour spreader from
one or more spillways. Figure 6 shows a schematic flood
spreading system with some spillway, and Figs. 7, 8, 9
show three types of spillways, a cemented sandbag,
a gabion type, and a cemented rockwork spillway,
respectively.
GOVERNING EQUATIONS
Overland flow because of floodwater spreading is on
the base of shallow water hydraulics. The general
characteristic of shallow water flow (SWF) is that the
vertical scale is much smaller than typical horizontal scale,
which is true in many situations. SWF surface is nearly
horizontal, which allows a considerable simplification in
the mathematical formulation and numerical solution by
assuming the pressure distribution to be hydrostatic.
However, they may not exactly be two-dimensional
(2D). The flow exhibits a 3D structure, because of bed
friction, just as in boundary layers. Moreover, density
stratification, because of a difference in temperature or
salinity, causes variations in the third direction. In many
SWFs, these 3D effects are negligible, and it is sufficient
to consider the depth averaged form, which is 2D in the
horizontal plane (18). This restricted form is commonly
indicated by the term shallow water equations (SWEs).

Flow
Direction Spillway

Figure 6. Schematic spillway (16).

 WATER SPREADING 711

growing coefficient matrix, and used it for a typical flood-

water spreading plot in advance phase of flow to the end
of the spreading area.

DESIGN

NRCS (4) introduced the main conditions for applying the

floodwater spreading as:

a. Soils have suitable intake rates and adequate water

holding capacities for the crop to grow
b. Soils are suitable for production of feed, forage, or
grain crops
Photo by: R. Hasanzadeh
c. The topography is suitable for the diversion or
Figure 9. A cemented rock work spillway of Chikhab flood collection and spreading of water to achieve the
spreading, Dehloran, Iran. desired results
d. Runoff or stream flow is available at the time of
the year and in a volume sufficient to increase
In this article, we limit our attention to incompressible
plant growth
fluid (water) and constant temperature and salinity. By
assuming hydrostatic pressure, and with no Coriolis and e. Flows can be collected or delivered and spread
wind effects and negligible effective stresses, the 2D depth and exceed water returned without causing exces-
averaged equations of motion are written as (19,20): sive erosion
f. Fish and wildlife will not be significantly adversely
Continuity equation: affected
g. Grazing of the spreading area can be controlled
∂h ∂(uh) ∂(vh)
+ + = −I (1) NRCS (4) also described the water quantity and
∂t ∂x ∂y
quality requirement of flood spreading. FAO pointed
out that water spreading can be used under 100 to
x-momentum equation:
350 mm annually with a slope of less than 1% and

 even topography (16). USDA divided floodwater spreading
∂(uh) ∂(u2 h) ∂(uvh) ∂ h2
+ + +g systems into flow-type and detention-type systems (17).
∂t ∂x ∂y ∂x 2
The flow-type systems incorporate free drainage from the
uI irrigated area and are divided into spreader-ditch, syrup-
= gh(S0x − Sfx ) + (2)
2 pan, and dike-bleeder types. Detention-type systems
retain the applied water on the irrigated area until it
y-momentum equation: has infiltrated and are divided into manual and automatic
inlet control subsystems. Apart from the type of system,


∂(vh) ∂(v2 h) ∂(uvh) ∂ h2 two parameters must be determined as design criteria of
+ + +g
∂t ∂y ∂x ∂y 2 the system, design application depth (DAD) and divertible
volume of flow rate (DV or DQ). USDA stated that DAD
vI
= gh(S0y − Sfy ) + (3) could be determined by a relationship between intake
2 rate, flow duration, or time of concentration, and DV or
DQ could be determined as a ratio of volume of storm or
where h is flow depth; u and v are flow velocities in x and y
peak discharge of upstream catchment.
directions, respectively; I is the infiltration rate; t is time;
g is gravitational acceleration; S0x and S0y are ground
surface slopes because of x and y directions, respectively; SUITABLE SITES
and Sfx and Sfy are friction slopes in the direction of x and
http://www.fao.org/docrep/u3160E/33160e00.htm
y axes, respectively.
http://www.fao.org/ag/agl/aglw/wharv.htm
Numerical solutions of SWEs with full or simple
http://www.eng.warwick.ac.uk/ircsa/index.htm
forms were described in several articles. Menendez (19),
Akanbi and Katapodes (20), and Vreugdenhil (18) intro-
duced numerical methods of solution of 2D SWEs. Arasteh BIBLIOGRAPHY
and Vahhadj (21) developed a quasi2D finite difference 1. Holtzworth, L. and Lacey, J. (1994). Species Selection,
semianalytical model of kinematic waveform of SWEs for Seeding Techniques and Management of Irrigated Pastures
Bisheh Zard flood spreading system, Iran. Arasteh (22) in Montana and Wyoming. EB 99. Montana State University,
used an implicit finite difference scheme for the men- Bozeman, MT.
tioned quasi2D model. Arasteh and Banihashemi (23–25) 2. OTA. (1983). Water Related Technologies for Sustainable
developed a solution method for implicit four points finite Agriculture in US Arid and Semi-Arid Lands. United States
difference method to solve 1D kinematic wave SWE, called Congress, Office of Technology Assessment, Washington, DC.
712 SPRINKLER IRRIGATION

3. OTT. (2001). Technologies Affecting Groundwater. United of the 8th International Conference on Rainwater Catchment
States Office of Technology Transfer, Washington, DC. System. 173–181.
4. NRCS. (1985). Waterspreading. Natural Resources Conserva- 22. Arasteh, P.D. (2002). Quasi-2D Simulation of Shallow Water
tion Service Practice Standard No. 640. Flow in a Flood Spreading System. Soil Conservation and
5. Prinz, D., Tauer, W., and Vogtle, Th. (1994). Application of Watershed Management Research Institute, Research Report
remote sensing and geographic information systems for No. 77-0510300000-01, p. 80 (in Farsi).
determining potential sites for water harvesting. Proc. 23. Arasteh, P.D. and Banihashemi, M.A. (2003). Numerical
of the XII CIGR Congress on Expert consultation on solution of kinematic wave equation to simulate overland
Water Harvesting for Improved Agricultural Production, flow using four-point method with growing coefficient matrix.
Washington, DC. 79–87. Proc. of the 4th Iranian Hydraulics Conference, Tehran, Iran,
6. Prinz, D. (1994). Water harvesting and sustainable agricul- 753–760 (in Farsi).
ture in arid and semi-arid regions. Proc. of the CIHEAM 24. Arasteh, P.D. and Banihashemi, M.A. (2003). Application of
Conference on Land and Water Resources Management in 1D kinematic wave model to simulate advance phase of
Mediterranean Region, Washington, DC. 745–762. overland flow in a flood spreading plot. Proc. of the 3rd Aquifer
7. Alem, G. (1999). Rainwater harvesting in Ethiopia: an Management Congress, Tehran, Iran, 140–147 (in Farsi).
overview. Proc. of the 25th WEDC Conference on Integrated 25. Arasteh, P.D. and Banihashemi, M.A. (2004). Possibility of
Development for Water Supply and Sanitation, Washington, application of kinematic wave model to simulate overland
DC. 387–390. flow on flood spreading plots. J. Pajouhesh va Sazandegi. 61:
8. Evenari, M., Shanan, L., and Tadmor, N. (1971). The Negev: 14–23 (in Farsi).
The Challenge of a Desert. Harvard University Press,
Cambridge, MA.
9. Arabkhedri, M. and Partovi, A. (1997). Band-sar, a practice SPRINKLER IRRIGATION
in optimum using of water and soil resources in an arid
region. Proc. of the 8th International Conference on Rainwater MUHAMMAD NADEEM ASGHAR
Catchment System. 34–40.
International Water
10. Baghernejad, M. (1999). Artificial recharge system and the Management Institute (IWMI)
fate of dissolved and suspended particles in floodwater, a Lahore, Pakistan
case study in Damghan Playa, Iran. Proc. CD of the 9th
International Conference on Rainwater Catchment System.
11. Kowsar, S.A. and Yazdian, A.R. (1999). Geologic nitrogen In arid and semiarid environments, supplemental irri-
in Agha Jari formation of Bisheh Zard Basin: a dilemma.
gation has become a prerequisite to minimize risks
Proc. CD of the 9th International Conference on Rainwater
Catchment System.
of crop failures in both rain-fed and irrigated agricul-
tural areas. Usually, surface irrigation, in which the
12. Nejabat, M. (1999). Improving environmental characteristics
in a wild area around a floodwater spreading system, a entire or most crop area is flooded, is used for irrigat-
case study. Proc. CD of the 9th International Conference on ing these croplands (Fig. 1). More often, however, water
Rainwater Catchment System. management skills on the farm are lacking, water sup-
13. FAO. (1999). Floodwater utilization: a successful approach plies may be uncertain, and so irrigation efficiency tends
towards sustainable development in arid environment. Proc. to below.
of the Conference on the Multifunctional Character of Agricul- Because of the present scarce global water resources,
ture and Land. Available at http://www.fao.org/mfcal/pdf/ft improving irrigation efficiency—getting more crop per
iran.pdf. drop—and making the best use of water for agriculture
14. Prinz, D. and Malik, A.M. (2002). Runoff farming. Available are prerequisites for the future (1). Shifting from surface
at http://www.wca-infonet.org/id/81755. irrigation to sprinkler irrigation, which is a method of
15. Pereira, L.S., Cordery, I., and Iacovides, I. (2002). Coping irrigation under pressure in which water is sprinkled
with water scarcity. IHP technical documents in hydrology.
UNESCO. 58: 57–63.
16. Critchley, W., Siegert, K., and Chapman, C. (1991). Water
Harvesting. FAO, Rome, Italy.
17. Hart, W.E., Collins, H. G., Woodward, G., and Humpherys,
A.S. (1983). Design and operation of gravity or surface
systems. In: Design and Operation of Farm Irrigation
Systems. M.E. Jensen (Ed.). pp. 550–559.
18. Vreugdenhil, C.B. (1994). Numerical Methods for Shallow
Water Flow. Kluwer Academic Publisher, Amsterdam, the
Netherlands, p. 261.
19. Menendez, A.N. (1987). On the accuracy of numerical
methods for the two dimensional shallow water equation.
Proc. of the XXII Congress of IHAR. 372–377.
20. Akanbi, A.A. and Katapodes, N.D. (1988). Model for flood
propagation in initially dry land. J. Hydraul. Engin. 14(7):
689–706.
21. Arasteh, P.D. and Vahhadj, S.R. (1997). Mathematical mod-
eling of shallow water flow in a flood spreading system. Proc. Figure 1. Surface irrigation.

Sprinkler
Flexible hose
Riser
Lateral pipeline
Pipe connector
Figure 2. Lateral pipeline.
to imitate rainfall, will contribute substantially to these
goals. In contrast to surface irrigation, sprinkler irrigation
can help to irrigate frequently with just the right amount
of water to avoid crop distress and to avoid overirrigation
(Figs. 2 and 3).
In recent decades, various sprinkler irrigation methods
and installations have been developed to meet the
needs of farmers, particularly smallholders. Hand-move
sprinkler irrigation is the most widely adopted and
least expensive system operated with pressure ranges
from 2.0–3.5 bars. The sprinklers are mounted on
lateral pipelines temporarily laid across a field. To
reduce labor requirements, this system was modified to
hose-move sprinkler irrigation (Fig. 4). In this modified
system, lateral pipelines are placed permanently, and
the sprinklers, mounted on tripod stands, are not fitted
directly to the lateral pipelines but are connected to
them via flexible hoses. The hoses and sprinklers can
be moved laterally on either side to cover a number of
lateral positions (2).
Generally, in sprinkler irrigation, the uniformity of
distribution is more sensitive to wind. However, this
irrigation provides ‘air-conditioning’ effects on the crops
if used in anti frost sprinkling and can also reduce
pesticide applications if used as a spray. It is also
fairly well established that a plan of heavy sprinkling
at infrequent intervals will ordinarily produce a deeper
root structure and thus better crop growth and yields
than frequent, light sprinkling. However, light irrigation
Figure 3. Hand-move sprinkler irrigation.
SPRINKLER IRRIGATION 713
Figure 4. Hose-move sprinkler irrigation.
may be beneficial when there is a minimum crop water
requirement. Under shallow water table conditions, when
groundwater contributes significantly to meet the crop
water requirement, light irrigation is also required to
dilute the salinity in the effective root zone. Sprinkler
irrigation may be used to germinate and establish salt-
sensitive crops by keeping the seedbed adequately moist
and salt-free by uniformly applying small depths of
irrigation water.
When sprinkler irrigation is done with saline water, it
brings water into contact with foliage, and salt deposits
on leaves may adversely affect some crops. However,
the susceptibility of plants to leaf injury from saline
sprinkled water depends on leaf characteristics that affect
the rate of absorption and is not generally correlated
with tolerance to soil salinity. The degree of spray injury
varies with weather conditions, especially the water
deficit of the atmosphere. Visible symptoms may appear
suddenly following irrigation when the weather is hot
and dry. Increased frequency of sprinkling, in addition to
increased temperature and evaporation, increases the salt
concentration in the leaves due to adsorption and results in
leaf damage. However, irrigation at night (or any other low
evaporation period) minimizes the salt concentration in
the leaves due to adsorption (3). Furthermore, evaporation
losses during sprinkling can also be minimized by avoiding
midday irrigation.
High capital costs, nonavailability of needed equipment
in the local market, energy dependence, and lack of local
expertise to operate and maintain sophisticated hydraulic
equipment are the main constraints that hinder the
acceptability of the system by the farmers, particularly
smallholders. For the same reasons, surface irrigation is
still likely to be dominant in the future, even though it
wastes water and is a major cause of waterlogging and
salinization.
However, the integration of surface irrigation and
sprinkler irrigation could help in controlling salinity in the
root zone with minimum application of irrigation water by
improving leaching efficiency. In this practice, sprinkler
irrigation helps in uniform distribution of groundwater,
and surface wetness due to sprinkling that helps in
minimizing the application of surface water still can meet
the crop water requirement. This surface wetness also
714 STOMATES

allows salts to dissolve in the root zone, which improves (a)

the leaching efficiency of surface irrigation. Therefore, Guard cells
the value of sprinkler irrigation can still be promised by
promoting it together with surface irrigation (4).

BIBLIOGRAPHY

1. FAO. (2002). Crops and Drops—Making the Best Use of Water Pore
for Agriculture. On-line Publications, Food and Agriculture
Organization of the United Nations, Rome.
2. Phocaides, A. (2000). Technical Handbook on Pressurized Irri-
gation Techniques. On-line Publications, Food and Agriculture
Organization of the United Nations, Rome.
3. Kruse, E.G. (1995). On-farm irrigation and drainage prac-
tices. In: Agricultural Salinity Assessment and Management.
K.K. Tanji (Ed.). ASCE, New York, pp. 349–371.
Epidermal cells
4. Asghar, M.N. et al. (2002). Helping Measures for Farmers
to Mitigate Drought in Canal Water Scarce Areas of the
Indus Basin, Pakistan. Proc. Natl. Symp. Drought Water
Resour. Pakistan, Center of Excellence in Water Resources
Engineering, UET, Lahore, pp. 63–73. (b)

STOMATES
R. ANDRÉS FERREYRA
Ag Connections, Inc.
Murray, Kentucky

Stomates or stomata (singular: stomate or stoma) are

structures on the surfaces of plant leaves that allow gas
exchange (transpiration, i.e., loss of water vapor, uptake
of CO2 , and emission or uptake of O2 ) between the interior
of a leaf and the atmosphere.
Plant leaves are covered by a waxy layer, the cuticle,
that prevents water loss from the cells on the leaf
surface, or epidermis. The cuticle does not completely
cover the leaves, however; it is interrupted by microscopic
pores, surrounded by pairs of specialized guard cells
(Fig. 1a). A stoma is a unit composed of a pore and its
guard cells. Guard cells, unlike regular epidermal cells,
usually contain chloroplasts (photosynthesizing organelles
containing chlorophyll). In some plants, there are also
additional, specialized subsidiary cells that differ in shape
from regular epidermal cells and participate in the osmotic
changes that drive guard cell movement.
Stomata can be found in all aboveground parts of plants
but are more frequent in leaves. They usually appear on
one (the lower) side of leaves but may appear on both.
When stomata occur on the lower (abaxial) leaf surface,
the plant is said to be hypostomatous; hyperstomatous
plants have stomata only on the upper (adaxial) leaf
surface, and plants with stomata on both surfaces
are called amphistomatous. In this latter case, abaxial
stomata are generally more numerous. The number of
stomata per unit area is called the stomatal density.
This figure may vary when plants are grown under
different environmental conditions such as atmospheric
CO2 concentration.
The size and density of stomata vary widely among
species (Table 1). When fully open, stomatal pores are
Open Closed
(high turgor) (low turgor)
Figure 1. (a) Micrograph of a peanut (Arachis hypogaea L.)
epidermis, showing the guard cells, the stomatal pore, and
epidermal cells: (b) Diagram of open and closed stomata, showing
cellulose microfibrils.
3–12 µm wide and from 10 to more than 30 µm long.
Species with smaller stomata tend to have a higher
density that can range from 60–80 per mm2 in corn to
150 in legumes such as alfalfa and clover, 300 in apple,
and more than 1000 in scarlet oak (1). Dicotyledon leaves
with a netted venation pattern generally have a scattered
arrangement of stomata, whereas monocotyledons and
conifers tend to have their stomata arranged parallel to
the main axis of leaves.
Some plants locate their stomata at the bottom of
recessed hollows and sometimes grow hair-like structures
above them, apparently to limit water loss by creating a
high-humidity environment above the pore. Most plant
leaves have small open spaces under the pore, called
substomatal cavities (Fig. 2).
STOMATAL CONDUCTANCE MODEL
Transpiration and its dependence on stomatal control
have been traditionally described by using an electric
analog model based on conductances or their reciprocal,
resistances. The basic assumption of the conductance
 STOMATES 715

Table 1. Occurrence and Dimensions of Stomata for Several Representative Speciesa

Dimensions of
Guard Cell Dimensions, µm, Stomatal Pore,
Stomatal
for each Leaf Surface Abaxial, µm
Density,
Pores/mm2 Adaxial Abaxial
Representative Common
Plant Type Species Name Upper Lower Length Width Length Width Length Depth

Moss Polytrichum commune Hair cap moss 16 16 46 15 46 15 15 12

Fern Osmunda regalis Royal fern 0 67 56 19 30 15
Conifer Pinus silvestris Scotch pine 120 120 28 7 28 7 20 6
Dicot tree Tilia europea European linden 0 370 25 9 10 8
Dicot herb Helianthus annuus Sunflower 120 175 35 13 32 14 17
Dicot herb (xerophyte) Sedum spectabilis Autumn joy 35 56 32 10 33 10 20 18
Monocot herb Allium cepa Onion 175 175 42 19 42 19 24 18
Monocot C3 grass Avena sativa Oats 50 45 52 15 56 13 20 10
Monocot C4 grass Zea mays Corn 98 108 38 10 43 12 20 10
a
Adapted from Reference 1.

model is that the diffusion of gases within plant tissues Cuticle

and in the atmospheric boundary layer is restricted by
resistances and that the resulting system of gas fluxes
Palisade
into and out of leaves can be described by the equations mesophyll
developed for analyzing electrical circuits (Fig. 3).
Nobel (2) described the diffusion of gases along
Epidermis
stomatal pores by applying Fick’s law and a stomatal Spongy
conductance (which we will denote as gst w ). This is a mesophyll
widely used parameter in modeling applications; stomatal
conductance, the rate of flow per unit area of the leaf,
is considerably easier to measure than the flux density
within a stomatal pore. Stomatal conductance is closely Substomatal
cavity
related to stomatal aperture; Jones (3) showed how to
estimate conductance from observed stomatal dimensions.
Given the dynamic behavior of stomatal aperture,
Guard cells Pore Cuticle
however, conductance is highly variable, especially in
natural conditions. Time-averaged values are typically Figure 2. Diagram of a (hypostomatous) leaf cross section,
used in calculations. showing different tissues and cell types.
The cuticular conductance, gcut w , is probably the most
uncertain of all conductances in this model. Traditionally it
has been assumed to be negligible, but experimental data sum of the stomatal and intercellular space resistances.
show that cuticular transpiration varies greatly among According to this scheme, the total leaf resistance to water
species. Larcher (4) generalized that it constitutes 10–33% movement can be described as
of total transpiration, but some researchers presented even
w + rw ) · rw
(rias st cut
higher values (5). rw = + rbl
w
rw + rw + rw
ias st cut
Nobel proposed calculating gias w , the conductance of
intercellular air spaces, based on the average distance
and the total leaf conductance is
between mesophyll cell surfaces within a leaf and the
inner surface of stomatal pores. Consistent application of  −1
gw = rw
the conductance model would thus require measurements
on leaf cross-sectional images. Parkhurst (6) postulated
Nobel (2) provided some representative conductance
that intercellular air space is a very important, usually
values: for crop plants such as beet, spinach, tomato,
ignored, component of the leaf gas exchange pathway,
and pea, gias
w ranges from 800 to 4000 mmol/m /s (20
2
although it is more relevant for photosynthesis than for
to 100 mm/s); for open stomata, gst w ranges from 200 to
transpiration.
800 mmol/m2 /s (5 to 20 mm/s). These values are somewhat
Calculations with the electric analogy are simplified
lower for other agricultural crops.
by replacing all conductances with resistances, taking
 −1
riw = giw , where the superscript ‘‘i’’ stands for ‘‘bl,’’
‘‘st,’’ ‘‘cut,’’ or ‘‘ias’’ (boundary layer, stomata, cuticle, and THE STUDY OF STOMATA
intercellular air space, respectively). Figure 3 presents the
corresponding equivalent electrical circuit. The cuticular Malpighi saw pores on leaf surfaces in 1674. Grew showed
resistance acts as a resistance in parallel with the series them in his anatomical work of 1682. The term ‘‘stomata’’
716 STOMATES
P atm
w
r bl
w impression may invade the pore or stomatal aperture may
st
rw
cut
rw
ias
rw
P sat
w
Figure 3. Conductance model of leaf water vapor movement,
expressed in resistance form using electrical symbols. Patm w and
Psat
w are the water vapor pressure in the atmosphere and on the
surface of mesophyll cells, respectively. The riw are the resistances
to water vapor movement in the boundary layer (i = ‘‘bl’’), in
stomata (‘‘st’’), through the cuticule (‘‘cut’’), and in the intercellular
air spaces (‘‘ias’’).
may have been first used by A. de Candolle in 1827,
and research on stomatal behavior began earnestly in
the mid-1900s (1). Gas exchange through small pores
such as stomata was explained formally in physical
terms by Brown and Escombe (7). Later, Stålfelt (8) and
Bange (9) showed how, in moving air (when boundary
resistance is low), transpiration and stomatal aperture are
closely correlated. More recently, Zeiger (10) reviewed the
multiple environmental influences on stomatal behavior;
the review by Outlaw (11) shows the current state of
the art.
The crucial role of stomata in controlling plant CO2
uptake and water loss has motivated the search for meth-
ods of measuring stomatal aperture and conductance.
Weyers and Meidner (12) extensively reviewed these tech-
niques. A few salient methods are discussed briefly below.
Microscopy
In some plants, pieces of epidermis can be peeled off the
leaf and observed on the microscope after fixing in alcohol.
However, it is very difficult to peel the epidermis from
some plants, and the process can also affect stomatal
aperture. An alternative is to make impressions of the
epidermis while the leaves are still on the plant, using
substances such as collodion, silicone rubber, dental paste,
or nail polish. These impressions are later peeled off and
examined on the microscope (1).
The above mentioned methods involving epidermal
manipulation may not be fully representative of in vivo
conditions because the material used to make the
change during the process. There is evidence that ultrafast
freezing can prevent aperture changes (13). This, coupled
with scanning electron microscopy, can capture stomatal
characteristics in great detail (e.g., 14).
Anatomically Explicit Models
Most implementations of conductance models assume
that diffusion paths for both CO2 and water vapor are
identical, thus making transpiration and photosynthesis
proportional to each other as stomatal aperture changes.
This assumption ignores the intracellular, liquid-phase
components of the CO2 pathway, although Nobel (2)
postulated that liquid-phase conductance values are
significant.
Parkhurst (6) criticized conductance models ‘‘in part
because they are usually one-dimensional representations,
but also because they treat continuously interacting
processes as if they were sequential.’’ The lumped-
parameter conductance approach, as opposed to using
spatially distributed parameters and processes, may
introduce error even in the apparently simple simulation
of diffusion within intercellular spaces, especially when
estimating photosynthesis: assimilation really occurs
throughout the tortuous intercellular space concurrently
with a gradual fall in CO2 concentration, rather than
lumped at the end of the flow path as in typical
conductance models.
However, lumping the properties of the leaf interior
was unavoidable until recent developments in numerical
methods made it possible to solve diffusion equations in
very complex domains. Parkhurst (6) proposed a complex
distributed electrical analog model, and Pachepsky and
Acock (15) created a very sophisticated finite-element
model, 2DLEAF, that explicitly represents leaf tissues in
a two-dimensional domain based on measurements made
on micrographs of epidermal peels and leaf cross sections.
Driven by environmental variables such as photon flux
density, air and leaf temperature, gas concentrations
in the atmosphere, and a novel form of parameterizing
the leaf boundary layer (16), the model then simulates
the fluxes of water vapor, CO2 , and O2 . Calculations
with 2DLEAF have shown that distributions of water
vapor and CO2 concentrations in the leaf interior differ
from each other, that photosynthesis and transpiration
rates are not proportional to each other, and that
stomatal aperture affects transpiration more than it
affects photosynthesis (15); this confirms the statement
by Parkhurst (6) that low stomatal conductance reduces
water loss more than it reduces CO2 uptake.
Infiltration
Another experimental technique for estimating stomatal
aperture involves infiltrating graded solutions of different

viscosity into leaves. The viscosity of the most viscous
solution that can infiltrate the pores serves as a
measure of aperture (17). However, this method does not
allow calculating stomatal resistances, and, due to the
differences in stomatal anatomy among species, it is better
suited to relative measurements of stomatal aperture
within the same species (3).
Mass or Viscous-Flow Porometers
The combined resistance of the stomata and the intercel-
lular air spaces of the mesophyll can be estimated for an
amphistomatous leaf by forcing pressurized air through
the leaf from one surface to the other and then measuring
the airflow rate (or the time needed for a specified drop
in pressure). The flow rate is inversely proportional to the
viscous-flow resistance.
Many such mass- or viscous-flow measuring devices, or
porometers, have been described in the literature (18–20).
However, absolute calibration is difficult because the flow
path inside the leaf is complex and the series circuit
measured by the instrument (air flows in through one
leaf surface, the mesophyll, and out the other surface) is
not representative of the actual water vapor diffusion
paths, where the stomata of the two leaf surfaces
are in parallel (3). Additionally, mass flow porometer
designs have been cumbersome (albeit ingenious), and
contemporary interest in separating the adaxial, abaxial,
and mesophyll resistance components has decreased
interest in mass flow instruments in favor of the
diffusion porometer.
Diffusion Porometers
More advanced porometers generally include small
cuvettes or chambers that can be attached to leaves and
either measure the time necessary for a specified change
in the humidity in the chamber (transit-time diffusion
porometers; 21) or measure the water vapor concentration
differential between the air leaving the chamber and the
air entering it (continuous-flow diffusion porometers; 22).
They can work with one or both leaf surfaces.
Diffusion porometers are powerful, albeit expensive.
They are portable, operate rapidly (seconds to a few
minutes), and store data in an onboard memory for
subsequent downloading to a personal computer. Although
they can work with individual leaf surfaces (unlike mass-
flow porometers), they must still average over a relatively
large leaf area and thus cannot describe the behavior of
individual stomata. They are also affected by any factor
that influences water loss (e.g., cuticle conductance), not
just stomatal conductance.
Jones (3) suggested the following equation to estimate
leaf conductance from a continuous-flow porometer:
ue (xw0 − xwe )[1 − (xws + xw0 )/2]
gw = (rw + raw )−1 =
A(1 − xw0 )(xws − xw0 )
where gw is the total conductance to water vapor
(mol/m2 /s); rw is the leaf (combined stomata, cuticle,
intercellular air space, and cell wall) resistance to water
vapor (m2 s/mol); raw is the boundary layer resistance
to water vapor, typically specified by the porometer
STOMATES 717
manufacturer; ue is the flow rate entering the chamber
(mol/s); xwe is the mole fraction of water vapor in the air
entering the chamber (mol/mol); xw0 is the mole fraction of
water vapor in the air leaving the chamber (mol/mol); xws is
the saturation mole fraction of water vapor at the chamber
temperature (mol/mol), where the relative humidity of the
outlet air is equal to xw0 /xws ; and A is the surface area of
the leaf exposed in the chamber (m2 ).
STOMATAL FUNCTION
Stomata open and close due to changes in the shape of
guard cells. Changes in the concentration of ions inside
guard cells make water move into or out of the cells
osmotically. The corresponding changes in hydrostatic
pressure (turgor pressure, the pressure exerted by the
fluid in the cell that presses the plasma membrane against
the cell wall) in turn make the cells swell or shrink.
Two additional factors influence the change in shape;
first, guard cell walls have a set of cellulose microfibrils
arranged radially around the cell (radial micellation).
When the cell volume tends to increase with turgor
pressure, the microfibrils force a lengthening of the cells
rather than an increase in diameter. Additionally, the
inner walls of guard cells, those adjacent to the pore, are
more rigid than the outer walls. Thus, a lengthening of
the cell translates into an increase in its curvature, and
consequently into an opening of the stoma (Fig. 1b).
The intake of ions for stomatal opening occurs through
the guard cell membrane. A special form of enzyme
called proton ATPase removes protons from the inside
of the cell. This creates an electrical gradient across
the membrane; potassium ions can move into the cell
(through specialized ion channel proteins embedded in the
membrane) driven by this gradient. Chlorine and calcium
ions, as well as starch, play additional roles in this process.
For further reading, Nicholls and Ferguson (23) presented
an excellent introduction to ATPases and active transport
through membranes; Mansfield et al. (24) discussed the
role of calcium.
Stomatal aperture responds to numerous environmen-
tal factors such as light, temperature, relative humidity,
CO2 concentration, and pollutants. It also depends on
internal variables such as water stress, periodic metabolic
oscillations (circadian rhythms), and the concentration
of signaling substances such as abscisic acid (ABA) and
cytokinins. Outlaw (11) and Zeiger (10) explained all of
this in detail; a brief outline is given below.
In general, stomata tend to open in the light, when there
is energy available for photosynthesis. An interesting
exception is that of desert succulent plants of the
crassulacean acid metabolism (CAM) type. They open their
stomata at night (when atmospheric water demand is low),
store CO2 as malic acid, and then convert that into sugars
via photosynthesis during the day with their stomata
closed and the plant safe from water loss.
Stomatal aperture is generally inversely proportional to
CO2 concentration. Some researchers believe that plants
sense the concentration of CO2 in the leaf interior and
respond to that; others believe that the plant responds to
the ratio of internal to external CO2 concentrations.
718 STOMATES
Stomata tend to close at midday, suggesting that plants
can either sense atmospheric humidity, or sense transpi-
ration rate and react to limit water loss. Temperature also
affects stomatal aperture differently for different species;
in general, temperature extremes close the stomata. How-
ever, it is difficult to separate the high temperature
effect from that of a high vapor pressure deficit (i.e.,
high water demand). Wind effects are complex. On one
hand, wind reduces boundary layer resistance and thus
tends to increase transpiration. On the other, there are
reports of plants closing their stomata in response to wind
or vibration.
ABA is a plant hormone believed to be the main sig-
naling mechanism for stomatal control. Roots subjected
to dry soil (or hypoxia due to flooded soil) produce ABA,
which moves upward to the leaves through plant vascu-
lar tissues. ABA interacts with guard cells, allowing the
membranes to leak potassium ions out of the cell, reducing
its concentration gradient, which is accompanied by a loss
of water and turgor pressure and subsequently, stomatal
closure. Tal and Imber (25) presented an interesting exper-
imental confirmation of the central role of ABA in stomatal
control. They showed that mutants that cannot synthesize
ABA show permanent wilting because they cannot close
their stomata to stop water loss. When ABA is applied to
these plants from an external source, however, the stomata
close, turgor pressure is restored, and wilting stops.
STOMATAL REGULATION
Stomatal aperture provides plants with a control mecha-
nism with which they can balance the entry of the CO2
used for photosynthesis and the water lost by diffusion
through the pore from evaporation from the surfaces of
cells inside the leaf (2). Transpiration is usually consid-
ered an inevitable, undesirable side effect of CO2 uptake.
However, transpiration does have important roles: tran-
spiration provides energy to transport water (and hence,
nutrients and signaling substances) in the plant, and it
also dissipates the heat of direct sunlight (by evapora-
tive cooling).
Many ideas have been proposed to explain stomatal
behavior in terms of optimization of cost (transpiration)
and benefits (photosynthesis). Optimally, stomatal aper-
ture would vary throughout the day such that there is
maximum photosynthesis and minimum transpiration.
This implies a complex control system that monitors sev-
eral input variables (several of which were mentioned
in the previous section). Jones (3) and Nobel (2) provided
interesting conceptual discussions in terms of feedback
and feedforward systems. The latter explains the popular
hypothesis that stomata behave so that the sensitivity
of photosynthesis and transpiration to stomatal aperture
(i.e., the corresponding partial derivatives with respect to
stomatal aperture) are a constant multiple of each other.
Cowan (26) and Farquhar and Sharkey (27) discussed this
and other optimization strategies in greater detail.
BIBLIOGRAPHY
1. Kramer, P.J. and Boyer, J.S. (1995). Water Relations of Plants
and Soils. Academic Press, London.
2. Nobel, P.S. (1983). Biophysical Plant Physiology and Ecology.
W.H. Freeman, New York.
3. Jones, H.G. (1992). Plants and Microclimate, 2nd Edn.
Cambridge University Press, Cambridge, UK.
4. Larcher, W. (1995). Physiological Plant Ecology, 3rd Edn.
Springer, New York.
5. Antipov, N.I. (1971). Correlation between cuticular and
stomatal transpiration in woody and bushy plants. Sov. Plant
Physiol. 18(6):1026–1030.
6. Parkhurst, D.F. (1994). Diffusion of CO2 and other gases
inside leaves. Tansley Review No. 65. New Phytol. 120:
449–479.
7. Brown, H.T. and Escombe, F. (1900). Static diffusion of gases
and liquids in relation to the assimilation of carbon and
translocation of plants. Phil. Trans. R. Soc. London Ser. B
193: 223–291.
8. Stålfelt, M. (1932). Der stomatäre regulator in der pflan-
zlichen transpiration. Planta 17: 22–85.
9. Bange, G.G.J. (1953). On the quantitative explanation of
stomatal transpiration. Acta Bot. Neerl. 2: 255–297.
10. Zeiger, E. (1983). The biology of stomatal guard cells. Annu.
Rev. Plant Physiol. Plant. Mol. Biol. 34: 441–475.
11. Outlaw, W.H. (2003). Integration of cellular and physiological
functions of guard cells. Crit. Rev. Plant Sci. 22(5):503–529.
12. Weyers, J.D.B. and Meidner, H. (1990). Methods of Stomatal
Research. Longman, Scientific & Technical, Harlow, UK.
13. van Gardingen, P.R., Jeffree, C.E., and Grace, J. (1989).
Variation in stomatal aperture in leaves of Aven fatua L.
observed by low-temperature scanning electron microscopy.
Plant Cell Environ. 12: 887–898.
14. Shiraishi, M., Hashimoto, Y., and Kuraishi, S. (1978). Cyclic
variations of stomatal aperture observed under the scanning
electron microscope. Plant Cell Physiol. 19: 637–645.
15. Pachepsky, L.B. and Acock, B. (1996). A model 2DLEAF of
leaf gas exchange: Development, validation, and ecological
application. Ecol. Modelling 93: 1–18.
16. Pachepsky, L.B., Ferreyra, R.A., Collino, D., and Acock, B.
(1999). Transpiration rates and leaf boundary layer param-
eters for peanut analyzed with the two-dimensional model
2DLEAF. Biotronics 28: 1–12.
17. Hack, H.R.B. (1974). The selection of an infiltration technique
for estimating the degree of stomatal opening. Ann. Bot. 38:
93–114.
18. Darwin, F. and Pertz, D.F.M. (1911). On a new method of
estimating the aperture of stomata. Proc. R. Soc. London Ser.
B 84: 136–154.
19. Meidner, H. (1992). Developments in mass flow porometry. J.
Exp. Bot. 43: 1309–1314.
20. Shimshi, D. (1977). Fast-reading viscous flow porometer. New
Phytol. 78(3):593–598.
21. Kanemasu, E.T., Thurtell, G.W., and Tanner, C.B. (1969).
Design, calibration, and field use of a stomatal diffusion
porometer. Plant Physiol. 44: 881–885.
22. Beardsell, M.F., Jarvis, P.G., and Davidson, B. (1972). A null-
balance diffusion porometer suitable for use with leaves of
many shapes. J. Appl. Ecol. 9: 677–690.
23. Nicholls, D.G. and Ferguson, S.J. (1992). Bioenergetics 2.
Academic Press, San Diego, CA.
24. Mansfield, T.A, Hetherington, A.M., and Atkinson, C.J.
(1990). Some current aspects of stomatal physiology. Annu.
Rev. Plant Phys. Plant. Mol. Biol. 41: 55–75.
25. Tal, M. and Imber, D. (1971). Abnormal stomatal behavior
and hormonal imbalance in flacca, a wilty mutant of tomato.
 CROP WATER STRESS DETECTION USING REMOTE SENSING
III. Hormonal effects on the water status in the plant. Plant
Physiol. 47: 849–850.
26. Cowan, I.R. (1982). Regulation of water use in relation to
carbon gain in higher plants. In: Encyclopedia of Plant
Physiology. Vol. 12B. O.L. Lange and J.D. Bewdey (Eds.).
Springer-Verlag, Berlin, pp. 535–562.
27. Farquhar, G.D. and Sharkey, T.D. (1982). Stomatal conduc-
tance and photosynthesis. Annu. Rev. Plant Physiol. 33:
317–345.
CROP WATER STRESS DETECTION USING
REMOTE SENSING
M. SUSAN MORAN
USDA ARS Southwest
Watershed Research Center
Tucson, Arizona
PABLO J. ZARCO-TEJADA
Instituto de Agricultura
Sostenible (IAS-CSIC)
Córdoba, Spain
THOMAS R. CLARKE
USDA ARS U.S. Water
Conservation Laboratory
Phoenix, Arizona
Remote sensing in the optical spectrum has potential to
detect plant manifestations of both transient and chronic
crop water stress. Crop water status and evaporation rate
have been directly related to thermal indexes, such as
the crop water stress index (CWSI), the water deficit
index (WDI), and the thermal kinetic window (TKW)
index. Measurements of surface reflectance in narrow
visible and near-infrared wavelengths have potential for
monitoring crop photosynthesis and fluorescence. Surface
reflectances in wide bands of the visible, near-IR, and
short-wave IR spectrum have been used to detect plant
manifestations of chronic crop water stress, such as
phenological adaptations and leaf expansion and loss. To
apply these approaches with satellite-based sensors for
farm-scale irrigation scheduling and crop management,
there are still needs for improved sensor design, timely
image delivery, and reduced image cost.
INTRODUCTION
Numerous studies have shown that crop water stress
has a direct effect on crop growth, development, yield,
and ultimately, on farmers’ profits. New management
techniques in arid and semiarid regions are designed
for controlled deficit irrigation to ensure low water loss
without yield reduction. To achieve this delicate balance
between water use and crop yield, farm managers need
an operational means to quantify plant water deficit and
evaluate the effects of stress on a given crop species at any
stage of development. The following sections present the
plant manifestations of water stress that can be detected
using passive remote sensing methods with sensors that
719
acquire data in the visible (VIS), near-infrared (NIR),
short-wave infrared (SWIR), and thermal infrared (TIR).
CROP ADAPTATIONS TO WATER STRESS
Turner (1) provided a good synthesis of the effects of
water deficit on crop plants and physiological adaptations
to transient stress. He identified crop physiological
adaptations associated with three types of drought
resistance—drought escape, drought tolerance at high
water potential, and drought tolerance at low water
potential—where ‘‘drought’’ was defined as a period
without significant rainfall. The following discussion
focuses on the physiological adaptations that might have
the greatest effect on the reflective and thermal spectral
regions and on the optical properties of plants that would
allow stress detection by remote sensing.
Drought escape is the ability of the plant to complete
its life cycle before serious soil and plant water deficits
develop. For example, studies have shown that wheat
can hasten maturity in response to mild water deficits at
the critical time between flowering and maturity. When
stress occurred between floral initiation and wheat ear
emergence, the number of wheat tillers that produced
ears was less than that under well-watered conditions.
Drought tolerance at high tissue water potential is
sometimes referred to as drought avoidance; it allows
plants to endure drought periods while maintaining
high plant water status. One such crop adaptation is
the reduction of water loss through increased stomatal
and cuticular resistance. This is expressed in increased
epidermal waxes of leaves and a reduction in general plant
productivity. Another adaptation is to reduce the radiation
absorbed by the plant through leaf movement (e.g.,
leaf cupping, paraheliotropism, or wilting) or increased
leaf pubescence and waxiness. Drought tolerance is also
achieved by reduction of leaf area through decreased
leaf expansion, reduced tillering and branching, and
leaf shedding.
Water stress is often expressed by the inability of
some crop species to maintain cell turgor at low water
potentials. The capability of crops to control cell turgor
varies widely by species (1). These variations in cell
turgor under drought conditions can influence the closure
of stomatal apertures and the rates of photosynthesis,
evaporation, and leaf expansion.
Optical Properties of Plant Leaves
A great deal has been written on the optical properties
of crop leaves in the VIS, NIR, SWIR, and TIR
domains (2). In the visible domain, leaf reflectance is
affected mainly by leaf pigments such as chlorophyll,
xanthophyll, carotenoids, and anthocyanins, as well as
by the leaf structure and the dry matter content. In the
NIR, reflectance depends on the anatomic structure of the
leaves and increases with the number of cell layers and the
size of the cells. In the SWIR, reflectance is mainly affected
by the leaf water content; strong water absorption bands
are at 1.45, 1.95, and 2.5 µm. It is generally reported that
leaves under water stress show a decrease in reflectance
720 CROP WATER STRESS DETECTION USING REMOTE SENSING

in the NIR spectrum, a reduced red absorption in the composite temperature, defined by Norman et al.(10) as
chlorophyll active band (0.68 µm), and a consequent blue the ‘‘aggregate temperature of all objects comprising the
shift of the red edge (3). In the TIR, there is a direct link surface,’’ which was shown by Kustas et al. (11) to be a
between the process of plant water evaporation and the linear function of Tc and To .
plant thermal response (i.e., water evaporates and cools
the leaves).
REMOTE SENSING FOR DETECTING CROP WATER STRESS
The theory to explain these leaf optical properties
and the effect of leaf constituents on leaf reflectance
During the past 30 years, progress has been made in
and transmittance was developed into a leaf model by
using remotely sensed data to retrieve information useful
Allen and Richardson (4). The PROSPECT model (5),
for irrigation scheduling and management. The basic
based on a plate model for multiple layers, is widely
approaches have focused on parameters directly related
used to simulate leaf optical properties as function of
to crop water status (e.g., crop water loss, evaporation,
leaf chlorophyll concentration (Cab ), equivalent water
metabolism, photosynthesis, and fluorescence) and plant
thickness, dry matter content, and leaf internal structure;
manifestations of chronic crop water stress (e.g., phenology
it enables simulation of leaf reflectance and transmittance
and leaf expansion and loss).
as a function of different stresses.
Crop Evaporation
Optical Properties of Plant Canopies
An important breakthrough in using remote sensing
Though knowledge of the optical properties of individual
for water stress detection was the development of the
leaves contributes to our understanding of the processes
Idso–Jackson crop water stress index (CWSI) (12,13).
involved, field studies have shown that leaf optical
Jackson et al. (12) derived the theoretical CWSI (CWSIt )
properties alone are not sufficient to understand canopy-
by relating the canopy–air temperature difference (Tc −
level reflectance. Canopy reflectance is a function of the
Ta ) to net radiation (Rn ), vapor pressure deficit (VPD),
leaf optical properties previously described, plus viewing
soil heat flux (G), volumetric heat capacity of air (ρCp ),
and illumination geometry, canopy structure, and soil
psychrometric parameters (γ and ), and canopy and
reflectance. For example, Guyot et al. (6) reported that
aerodynamic resistances (rc and ra ), where
the most important factor that influences the reflectance
of a plant canopy is its geometric structure, not its
(Tc − Ta ) = [ra (Rn − G)/ρCp ]{γ (1 + rc /ra )/
leaf reflectance. Jackson and Ezra (7) concluded that
water stress-induced changes in visible, NIR, and SWIR [ + γ (1 + rc /ra )]} − {VPD/
reflectance of a cotton canopy were due largely to canopy
[ + γ (1 + rc /ra )]} (1)
geometry changes rather than leaf physiological/anatomic
changes in all but the red spectral band.
Taking the ratio of actual (E for any rc ) to potential (Ep for
Furthermore, Myers et al. (8) identified seven param-
rc = rcp ) crop evaporation rates gives
eters (in addition to leaf reflectance) that determined
crop canopy reflectance, of which only the first three
E/Ep = [ + γ ∗ ]/[ + γ (1 + rc /ra )] (2)
could be directly related to crop stress: (1) transmittance
of leaves, (2) amount and arrangement of leaves,
where rcp is the canopy resistance at potential evaporation.
(3) characteristics of other components of the vegetation
Jackson et al. (12) defined the CWSIt , ranging from 0
canopy, (4) characteristics of the background, (5) solar
(ample water) to 1 (maximum stress), as
zenith angle, (6) look angle, and (7) solar azimuth angle.
Canopy reflectance models enable the simulation of these
CWSIt = 1 − E/Ep = [γ (1 + rc /ra ) − γ ∗ ]/
canopy structural characteristics and viewing geometry
on canopy-level reflectance. Closed agricultural canopies [ + γ (1 + rc /ra )] (3)
can easily be assumed to be horizontally homogeneous,
plane-parallel turbid mediums, and therefore the plate Though Jackson et al. (12) provided a thorough theo-
model theory can be applied successfully. On the other retical approach for computing CWSI, the concept is more
hand, open crop canopies, showing heterogeneous struc- universally applied using a semiempirical variation pro-
ture with different shapes, need the development of posed by Idso et al. (13) based on the ‘‘non-water-stressed
models that specifically treat the effects of openings, shad- baseline.’’ This baseline is defined by the relation between
ows, and canopy geometry on the modeled reflectance. (Tc − Ta ) and VPD under nonlimiting soil moisture condi-
A detailed discussion of the different canopy reflectance tions, example, when the plant water is evaporating at the
model approaches by turbid-medium, geometric, hybrid, potential rate (Fig. 1). Such non-water-stressed baselines
and ray-tracing methods is provided by Goel (9). have been determined for many different crops, includ-
Similarly, the TIR emittance of a plant canopy ing aquatic plants and both preheading and postheading
is a function of the temperatures of both the plant growth periods for small grains (14).
components and the soil. Tc is the canopy temperature, Application of CWSI with satellite- or aircraft-
defined by Norman et al. (10) as the TIR temperature based measurements of surface temperature is generally
in which the ‘‘vegetation dominates the [measurement] restricted to almost full-canopy conditions, so that the
field of view minimizing the effect of soil.’’ To is surface temperature sensed is equal to the canopy
the temperature of the soil surface. Ts is the surface temperature (15). To deal with partial plant cover
 CROP WATER STRESS DETECTION USING REMOTE SENSING 721

4 1

Soil-adjusted vegetation index

Maximum
Canopyair temperature, °C

A stress 1. Well-watered vegetation

0 0.8
2. Water stressed vegetation
x C
−4 0.6
B C
A x B
−8 0.4
Minimum 4. Dry bare
stress soil
−12 0.2 3. Saturated
bare soil
−16 0
0 2 4 6 8 −15 −10 −5 0 5 10 15 20 25
Air vapor pressure deficit, kP a Surface −air temperature, °C
Figure 1. Canopy minus air temperature (Tc − Ta ) versus vapor Figure 2. The trapezoidal shape that would result from the
pressure deficit (VPD) for well-watered (line labeled minimum relation between surface temperature minus air temperature
stress) and maximally stressed alfalfa based on measurements at (Ts − Ta ) and the soil-adjusted vegetation index (SAVI). Using a
various sites across the United States. The CWSI for a measured measurement of Ts − Ta at point C, the ratio of actual to potential
alfalfa canopy is computed as the ratio of the distances CB and evaporation (which is equivalent to the Water Deficit Index or
AB (13). WDI) is equal to the ratio of the distances CB and AB (16).

conditions, Moran et al. (16) developed a Water Deficit
Index (WDI) which combined measurements of reflectance
with surface temperature measurements (a composite of
both the soil and plant temperatures), as expressed by
WDI = 1 − E/Ep =[(Ts − Ta )m − (Ts − Ta )r ]/
[(Ts − Ta )m − (Ts − Ta )x ]
where subscripts m, x, and r refer to minimum, maximum,
and actual, respectively. The WDI is operationally
equivalent to the CWSI for full-cover canopies, where
Ts = Tc . Graphically, WDI is equal to the ratio of distances
AC/AB in the trapezoidal shape presented in Fig. 2, where
WDI = 0.0 for well-watered conditions and WDI = 1.0 for
maximum stress conditions. The left edge of the Vegetation
Index/Temperature (VIT) trapezoid corresponds to Ts − Ta
values for surfaces evaporating at the potential rate; the
right edge corresponds to Ts − Ta values for surfaces in
which no evaporation is occurring. In practice, WDI uses
the Penman–Monteith energy balance equation to define
the four vertices of the VIT Trapezoid that encompasses
all possible combinations of a spectral vegetation index
(e.g., Normalized Difference Vegetation Index or NDVI)
and Ts − Ta for one crop type on one day (Fig. 2).
Another promising approach for operational application
is the use of remotely sensed crop coefficients (the ratio
of actual crop evaporation to that of a reference crop) for
window (TKW) is the range of temperatures within which
the plant maintains optimal metabolism. For example,
the TKW for cotton growth is 23.5 to 32 ◦ C; the optimum
temperature is 28 ◦ C, and biomass production is directly
related to the amount of time that canopy temperatures
are within the TKW (Fig. 3), provided that insolation, soil
moisture, and nutrients are nonlimiting. TKWs have been
(4)
identified for several crop species, including cotton, wheat,
soybean, potato, cucumber, and bell pepper. In practice,
the TKW provides a biological indicator of plant health
that is being used for irrigation management.
Crop Photosynthesis
Measurements of leaf chlorosis, which causes diminished
leaf photosynthetic capacity, may be particularly suitable
for early water stress detection. Differences in remote sens-
ing reflectance due to changes in chlorophyll concentration
levels have been detected in the green peak and along the
red edge spectral region (690 to 750 nm) (20). The red edge
occurs between the wavelengths of 0.69 and 0.76 µm due to
the change in reflectance caused by chlorophyll absorption
in the red spectrum and multiple scattering from leaves
in the NIR spectrum (21) (Fig. 4). Several hyperspectral
500
Cotton
400
Number of hours

estimating an actual, site-specific crop evaporation rate

from readily available meteorologic information (17,18). 300
This approach requires only a measure of NDVI and
is simply an improvement of an approach already 200
accepted and used by farmers to manage crops, where
such improvements include increases in the accuracy of 100
evaporation estimates and, by using images, the ability to
0
map within-field and between-field variations. 0 10 20 30 40 50
Canopy temperature, °C
Crop Metabolism
Figure 3. A frequency distribution of seasonal canopy tempera-
Quite distinct from the CWSI and WDI, Burke et al. (19) tures of cotton, where the vertical lines represent the temperature
developed a concept of thermal stress in plants that range that comprises the species-specific thermal kinetic window
linked the biochemical characteristics of a plant with (TKW) as determined from the changes in the apparent Km with
its optimal leaf temperature range. The thermal kinetic temperature (19).
722 CROP WATER STRESS DETECTION USING REMOTE SENSING
indexes proposed in the literature track and quantify
chlorophyll concentration for both leaf and canopy-level
studies (22,23), allowing remote sensing detection meth-
ods to identify vegetation stress through the influence of
chlorophyll content variation. These narrow-band vege-
tation indexes are, among others, the Simple Ratio Pig-
ment Index, SRPI (ρ430 /ρ680 ); Normalized Phaeophytiniza-
tion Index, NPQI (ρ415 − ρ435 )/(ρ415 + ρ435 ); Photochemical
Reflectance Index, PRI (ρ531 − ρ570 )/(ρ531 + ρ570 ); Normal-
ized Pigment Chlorophyll Index, NPCI (ρ680 − ρ430 )/(ρ680 +
ρ430 ); Greenness Index, G (ρ554 /ρ677 ); Structure Inten-
sive Pigment Index, SIPI (ρ800 − ρ450 )/(ρ800 + ρ650 ); and
red edge reflectance-ratio indexes such as (ρ740 /ρ720 ),
(ρ734 − ρ747 )/(ρ715 + ρ726 ), (ρ750 /ρ700 ), and (ρ750 /ρ710 ). See a
review by Zarco-Tejada et al. (23) for more detail.
In agricultural canopies that have large spectral con-
tributions from the soil background and LAI variation in
different growth stages, combined indexes are proposed
to minimize background soil effects while maximizing
the sensitivity to chlorophyll concentration. CARI (chloro-
phyll absorption in reflectance index) (24) and MCARI
(modified chlorophyll absorption in reflectance index) (25)
have been proposed for combination with SAVI (soil-
adjusted vegetation index) (26) and OSAVI (optimized
soil-adjusted vegetation index) (27) to reduce background
reflectance contributions. Successful chlorophyll estima-
tion for corn canopies at different growth stages was
achieved using the TCARI/OSAVI combined index (28)
and with MCARI/OSAVI on open-tree crop canopies (29).
New developments related to vegetation stress, that
have potential application to water stress detection,
focus on monitoring chlorophyll fluorescence (CF). Red
and far-red light is emitted from photosynthetic green
plant tissues in response to photosynthetically active
radiation. Several reviews offer details on theory,
measurement methods, and interpretation, as well as the
relationship to photosynthesis, plant physiological status,
and photosynthetic functioning (30–32). In particular,
relationships between steady-state and dark-adapted
chlorophyll fluorescence features with water stress status
have been investigated for both active and passive
sensors (33–37). This suggests the potential for stress
detection using far-distance remote sensing instruments.
Steady-state CF and photosynthetic rates are inversely
related, such that CF is low when photosynthesis is high,
0.8
High water
0.7
0.6
Reflectance
Low water
0.5
0.4
0.3
0.2
0.1
0 Senescence slope, VI/day
450 550 650 750 850 950 1050
Wavelength, nm
Figure 4. The expected trends of spectral reflectance of a crop
canopy with high and low water treatments (21).
although it can be reversed because of an intensified
protective quenching action on CF production. The total
amount of CF emitted by Photosystems I (PS-I) and II (PS-
II) is typically less than 5% of total light absorbed. Thus,
accurate remote detection of natural CF would enable the
measurement of canopy-level CF without direct contact
with the vegetation canopy. Zarco-Tejeda et al. (38)
demonstrated that CF could be detected at both leaf
and canopy levels using multiple approaches, including
observing its effects on the red edge spectral region.
Crop Phenology
There is evidence that chronic crop water stress can either
hasten (1) or delay (39) crop development, depending on
the species involved and the timing of water stress. The
timing and duration of stress is also of critical importance
to ultimate yield; example, if intense water stress occurs
during anthesis of a wheat crop, the impact on final
grain yield is usually much greater than if this same
stress occurs earlier or later in crop development. Tucker
et al. (40) showed that crop phenological stage could be
determined using a combination of spectral data and
accumulated temperature units (growing degree day). Idso
et al. (39) equated the slope of the spectral vegetation
index (VI) over time with the rate of senescence and
correlated this slope with final grain yield for wheat and
barley under stressed and nonstressed conditions (Fig. 5).
In a similar study, Fernández et al. (41) found that the
hydric stress on wheat could be determined by the slope of
the NDVI with time.
Chronic crop water stress can also manifest itself in
reduced leaf expansion and leaf loss. There have been
several studies that have monitored stress-induced reduc-
tions in biomass and GLAI using remote sensing tech-
niques (42–44). Though these results are encouraging,
this approach is limited by the fact that the VI–GLAI
relation is exponential, leading to saturation of the NIR
response at GLAI values of five to six (2). A better mea-
sure of the effects of water stress on leaf expansion and
biomass production might be the amount of solar radia-
tion intercepted, which is related directly to plant growth.
Kumar and Monteith (45) showed that the fraction of
absorbed photosynthetically active radiation (fAPAR) is
related linearly to NIR/red reflectance. More recent work
800
700 Y = −208 − 41360X
Grain yield g/m2
600 r = 0.78
500
400
300
200
100
0
−0.006 −0.01 −0.014 −0.018 −0.022
Produra wheat Anza wheat Briggs barley
Figure 5. Final grain yield of Produra wheat, Anza wheat, and
Briggs barley versus the slope of the transformed vegetation index
(TVI6) during the senescence period (39).
 CROP WATER STRESS DETECTION USING REMOTE SENSING
by Pinter (46) has shown that this relation between VI
and fAPAR is independent of variations in the solar zenith
angle, thus increasing the usefulness of this remote sens-
ing approach.
CONCLUDING REMARKS
The CWSI, WDI, and TKW are good examples of remote
sensing indexes that link surface temperature measure-
ments to crop and soil evaporation rates. Quantification
of leaf chlorosis based on surface reflectance in narrow
wavelength bands also has potential for detecting tran-
sient water stress. Season-long measurements of surface
reflectance in wide wavelength bands have been used to
detect the effects of chronic plant stress, such as delayed
(or hastened) crop phenology and leaf expansion or loss.
These are only some examples of the many approaches
available for detecting crop water stress using remote
sensors. However, to realize the full potential of satellite-
based estimation of crop water stress, it will be necessary
to continue sensor development, improve image availabil-
ity and timely delivery, and reduce image cost.
BIBLIOGRAPHY
1. Turner, N.C. (1977). Drought resistance and adaptation
to water deficits in crop plants. In: Stress Physiology In
Crop Plants. H. Mussell and R.C. Staples (Eds.). Wiley-
Interscience, New York, pp. 344–372.
2. Bauer, M.E. (1985). Spectral inputs to crop identification and
condition assessment. Proc. IEEE 73: 1071–1085.
3. Horler, D.N.H., Dockray, M., and Barber, J. (1983). The red
edge of plant leaf reflectance. Int. J. Remote Sensing 4:
273–288.
4. Allen, W.A. and Richardson, A.J. (1968). Interaction of light
with a plant canopy. J. Opt. Soc. Am. 58: 1023–1028.
5. Jacquemoud, S. and Baret, F. (1990). Prospect: A model of
leaf optical properties spectra. Remote Sensing Environ. 34:
75–91.
6. Guyot, G., Dupont, O., Joannes, H., Prieur, C., Huet, M., and
Podaire, A. (1984). Investigation into the mid-IR spectral
band best suited to monitoring vegetation water content.
Proc. 18th Int. Symp. Remote Sensing Environ., October 1–5,
1984, Paris, France, pp. 1049–1063.
7. Jackson, R.D. and Ezra, C.E. (1985). Spectral response of
cotton to suddenly induced water stress. Int. J. Remote
Sensing 6: 177–185.
8. Myers, V.I., Bauer, M.E., Gausman, H.W., Hart, W.G.,
Heiman, J.L., MacDonald, R.B., Park, A.B., Ryerson, R.A.,
Schmugge, T.J., and Westin, F.C. (1983). Remote sensing
application in agriculture. In: Manual of Remote Sensing, 2nd
Edn. R.N. Colwell (Ed.). American Society of Photogrammetry
and Remote Sensing, Falls Church, VA, pp. 2111–2228.
9. Goel, N.S. (1988). Models of vegetation canopy reflectance
and their use in estimation of biophysical parameters from
reflectance data. Remote Sensing Rev. 4: 1–212.
10. Norman, J.M., Divakarla, M., and Goel, N.S. (1995). Algo-
rithms for extracting information from remote thermal-IR
observations of the Earth’s surface. Remote Sensing Environ.
51: 157–168.
11. Kustas, W.P., Choudhury, B.J., Inoue, Y., Pinter, P.J.,
Jr., Moran, M.S., Jackson, R.D., and Reginato, R.J. (1990).
723
Ground and aircraft infrared observations over a partially-
vegetated area. Int. J. Remote Sensing 11: 409–427.
12. Jackson, R.D., Idso, S.B., Reginato, R.J., and Pinter, P.J., Jr.
(1981). Canopy temperature as a crop water stress indicator.
Water Resour. Res. 17: 1133–1138.
13. Idso, S.B., Jackson, R.D., Pinter, P.J., Jr., Reginato, R.J.,
and Hatfield, J.L. (1981). Normalizing the stress-degree-day
parameter for environmental variability. Agric. Meteorol. 24:
45–55.
14. Idso, S.B. (1982). Non-water-stressed baselines: A key to
measuring and interpreting plant water stress. Agric.
Meteorol. 27: 59–70.
15. Millard, J.P., Reginato, R.J., Idso, S.B., Jackson, R.D., Goet-
telman, R.C., and Leroy, M.J. (1980). Experimental rela-
tions between airborne and ground measured wheat canopy
temperatures. Photogrammetric Eng. Remote Sensing 46:
221–224.
16. Moran, M.S., Clarke, T.R., Inoue, Y., and Vidal, A. (1994).
Estimating crop water deficit using the relation between
surface-air temperature and spectral vegetation index.
Remote Sensing Environ. 49: 246–263.
17. Bausch, W.C. (1993). Soil background effects on reflectance-
based crop coefficients for corn. Remote Sensing Environ. 46:
213–222.
18. Hunsaker, D.J., Pinter, P.J., Jr., Barnes, E.M., and Kim-
ball, B.A. Estimating cotton evapotranspiration crop coeffi-
cients with a multispectral vegetation index. Irrig. Sci. (in
press).
19. Burke, J.J., Mahan, J.R., and Hatfield, J.L. (1988). Crop-
specific thermal kinetic windows in relation to wheat and
cotton biomass production. Agron. J. 80: 553–556.
20. Rock, B.N., Hoshizaki, T., and Miller, J.R. (1988). Compari-
son of in situ and airborne spectral measurements of the blue
shift associated with forest decline. Remote Sensing Environ.
24: 109–127.
21. Filella, I., and Peñuelas, J. (1994). The red edge position and
shape as indicators of plant chlorophyll content, biomass and
hydric status. Int. J. Remote Sensing 15: 1459–1470.
22. Carter, G.A. (1994). Ratios of leaf reflectances in narrow
wavebands as indicators of plant stress. Int. J. Remote
Sensing 15: 697–704.
23. Zarco-Tejada, P.J., Miller, J.R., Mohammed, G.H., Noland,
T.L., and Sampson, P.H. (2001). Scaling-up and model
inversion methods with narrow-band optical indices for
chlorophyll content estimation in closed forest canopies with
hyperspectral data. IEEE Trans. Geoscience Remote Sensing
39: 1491–1507.
24. Kim, M.S., Daughtry, C.S.T., Chappelle, E.W., McMurtrey,
J.E., III, and Walthall, C.L. (1994). The use of high spectral
resolution bands for estimating absorbed photosynthetically
active radiation (Apar). 6th Symp. Phys. Meas. Signatures
Remote Sensing, Jan. 17–21, 1994, Val D’Isere, France,
pp. 299–306.
25. Daughtry, C.S.T., Walthall, C.L., Kim, M.S., Brown de Col-
stoun, E., and McMurtrey, J.E., III (2000). Estimating
corn leaf chlorophyll concentration from leaf and canopy
reflectance. Remote Sensing Environ. 74: 229–239.
26. Huete, A.R. (1988). A soil-adjusted vegetation index (SAVI).
Remote Sensing Environ. 25: 295–309.
27. Rondeaux, G., Steven, M., and Baret, F. (1996). Optimization
of soil-adjusted vegetation indices. Remote Sensing Environ.
55: 95–107.
28. Haboudane, D., Miller, J.R., Tremblay, N., Zarco-Tejada,
P.J., and Dextraze, L. (2002). Integration of hyperspectral
724 VACUUM GAUGE TENSIOMETER
vegetation indices for prediction of crop chlorophyll content for
application to precision agriculture. Remote Sensing Environ.
81: 416–426.
29. Zarco-Tejada, P.J., Miller, J.R., Morales, A., Berjón, A., and
Agüera, J. (2004). Hyperspectral indices and model simula-
tion for chlorophyll estimation in open-canopy tree crops.
Remote Sensing Environ. 90: 463–476.
30. Papageorgiou, G. (1975). Chlorophyll fluorescence: An intrin-
sic probe of photosynthesis. In: Bioenergetics of Photosynthe-
sis. Govindjee (Ed.). Academic, New York, pp. 319–371.
31. Krause, G.H. and Weis, E. (1984). Chlorophyll fluorescence
as a tool in plant physiology. II. Interpretation of fluorescence
signals. Photosynthesis Res. 5: 139–157.
32. Schreiber, U., Bilger, W., and Neubauer, C. (1994). Chloro-
phyll fluorescence as a nondestructive indicator for rapid
assessment of in vivo photosynthesis. Ecological Stud. 100:
49–70.
33. Evain, S., Flexas, J., and Moya, I. (2004). A new instrument
for passive remote sensing: 2. Measurement of leaf and canopy
reflectance changes at 531 nm and their relationship with
photosynthesis and chlorophyll fluorescence. Remote Sensing
Environ. 91: 175–185.
34. Souza, R.P., Machado, E.C., Silva, J.A.B., Lagôa, A.M.M.A.,
and Silveira, J.A.G. (2004). Photosynthetic gas exchange,
chlorophyll fluorescence and some associated metabolic
changes in cowpea (Vigna unguiculata) during water stress
and recovery. Environ. Exp. Bot. 51: 45–56.
35. Flexas, J., Briantais, J-M., Cerovic, Z., Medrano, H., and
Moya, I. (2000). Steady-state and maximum chlorophyll
fluorescence responses to water stress in grapevine leaves:
A new remote sensing system. Remote Sensing Environ. 73:
283–297.
36. Cerovic, Z.G., Goulas, Y., Gorbunov, M., Briantais, J-M.,
Camenen, L., and Moya, I. (1996). Fluorosensing of water
stress in plants: Diurnal changes of the mean lifetime and
yield of chlorophyll fluorescence, measured simultaneously
and at distance with a [tau]-LIDAR and a modified PAM-
fluorimeter, in maize, sugar beet, and kalanchoë. Remote
Sensing Environ. 58: 311–321.
37. Valentini, R., Cecchi, G., Mazzinghi, P., Scarascia Mug-
nozza, G., Agati, G., Bazzani, M., De Angelis, P., Fusi, F.,
Matteucci, G., and Raimondi, V. (1994). Remote sens-
ing of chlorophyll a fluorescence of vegetation canopies:
2. Physiological significance of fluorescence signal in response
to environmental stresses. Remote Sensing Environ. 47:
29–35.
38. Zarco-Tejada, P.J., Pushnik, J., Dobrowski, S., and Ustin,
S.L. (2003). Steady-state chlorophyll a fluorescence detection
from canopy derivative reflectance and double-peak red-edge
effects. Remote Sensing Environ. 84: 283–294.
39. Idso, S.B., Pinter, P.J., Jr., Jackson, R.D., and Reginato, R.J.
(1980). Estimation of grain yields by remote sensing of crop
senescence rates. Remote Sensing Environ. 9: 87–91.
40. Tucker, C.J., Elgin, J.H., Jr., and McMurtrey, J.E., III.
(1979). Temporal spectral measurements of corn and soybean
crops. Photogrammetric Eng. Remote Sensing 45: 643–653.
41. Fernández, S., Vidal, D., Simón, E., and Solé-Sugrañes, L.
(1994). Radiometric characteristics of Triticum aestivum cv.
Astral under water and nitrogen stress. Int. J. Remote Sensing
15: 1867–1884.
42. Gardner, B.R. and Blad, B.L. (1986). Evaluation of spectral
reflectance models to estimate corn leaf area while minimizing
the influence of soil background effects. Remote Sensing
Environ. 20: 183–193.
43. Pinter, P.J., Jr., Jackson, R.D., Idso, S.B., and Reginato, R.J.
(1981). Multidate spectral reflectances as predictors of yield
in water stressed wheat and barley. Int. J. Remote Sensing 2:
43–48.
44. Mogensen, V.O., Jensen, C.R., Mortensen, G., Thage, J.H.,
Koribidis, J., and Ahmed, A. (1996). Spectral reflectance
index as an indicator of drought of field grown oilseed rape
(Brassica napus L.), Eur. J. Agron. 5: 125–135.
45. Kumar, M. and Monteith, J.L. (1981). Remote sensing of crop
growth. In: Plants and the Daylight Spectrum. H. Smith (Ed.).
Academic, London, pp. 133–144.
46. Pinter, P.J., Jr. (1993). Solar angle independence in the
relationship between absorbed PAR and remotely sensed data
for alfalfa. Remote Sensing Environ. 46: 19–25.
VACUUM GAUGE TENSIOMETER
MAHBUB ALAM
Kansas State University
Garden City, Kansas
Soil water plays an important role in providing water for
plant growth. The vacuum gauge tensiometer is a soil
water measuring device that is sensitive to soil water
change. The quantity of soil water that adheres around
soil particles determines the energy level at which it
is being retained or pulled. In other words, the plant
root has to exert energy to pull water out of a soil
matrix. This energy is also expressed as tension the roots
experience in removing the water. Tensiometer is like a
mechanical root but shows the tension in the vacuum
gauge attached to it. It is an indirect method to determine
soil water status. Direct measure of water content is time-
consuming and labor-intensive, and the result lags behind
the current time. Hence, irrigators depend on indirect
methods. Vacuum gauge tensiometer is inexpensive, easy
to use, and a preferred measurement method, especially
for course-textured soils.
A vacuum gauge tensiometer with its components is
shown in Fig. 1. Variations from this design exist on
the market. A tensiometer is a sealed, water-filled tube
equipped with a vacuum gauge on the upper end and a
porous ceramic tip on the lower end. The basic components
include: a cap, body tube, vacuum gauge with dial, and a
porous ceramic tip (Fig. 1).
The Cap: The cap must seal the tensiometer body tube
airtight or the device will not work. Some models have a
reservoir, like the one shown in Fig. 1, which serves as a
water supply source for the body tube. In these models,
the cap assembly has an extended stem to hold a neoprene
stopper that seals the body tube from outside atmosphere.
Body Tube: The body tube provides support and a liquid
connection between a porous tip and the vacuum gauge.
Tensiometers come in various lengths. Standard lengths
are 6, 12, 18, 24, 36, 48, and 60 inches.
Vacuum Gauge: A hermetically sealed vacuum gauge
(Bourdon1 vacuum gauge) with a round dial is attached
1
A thin-walled flattened tube used in a gauge to read pressure or
vacuum developed by Bourdon, an inventor in France some 150

Cap assembly
Reservoir
Neoprene stopper
Hermetically
sealed
vacuum
gauge
Air free
gauge
chamber
Body tube
Porous ceramic tip/cup
Figure 1. Sketched view of a vacuum gauge tensiometer showing
its components.
at the air-free gauge chamber of the body tube. The round
dial in the gauge is marked with a scale in centibars
to indicate the gauge reading. The scale is calibrated to
read in centibar, or hundredths of one ‘‘bar.’’ A bar is the
unit of pressure, either positive or negative, that has been
adopted for the expression of soil water tension. The bar
is an international unit of pressure in the metric system
and equivalent to 14.5 lb/sq. inch, or 0.987 atmospheres.
One centibar is equal to 1 kPa (kilopascal).
Ceramic Tip/Cup: The ceramic tip is built like a
tubular cup, and it is porous. The openings of the pores
are so small that when saturated with water, air cannot
pass through within the range of soil water tensions to be
measured. As soil water content outside the porous cup
in the soil matrix declines from saturation, water inside
the tube starts moving out through the porous tubular tip,
which causes a partial vacuum inside the tensiometer that
can be read on the vacuum gauge. The needle in the gauge
years ago. It was brought to the United States by Ashcroft Gauge
Company 100+ years ago and still is the standard gauge used
practically worldwide.
VACUUM GAUGE TENSIOMETER 725
starts moving above the face of the dial. The dial reading,
indicated by the needle, shows the suction or tension at
which the water is being pulled by the surrounding soil.
A reading of zero corresponds to a completely saturated
condition, regardless of the type of soil. A reading of 80
kPa (centibar) indicates a dry condition for sandy soils
and would hurt sensitive crops. This number is also the
functional upper limit for tensiometer readings. A tension
higher than 85 kPa (centibar) will cause the water inside
the tube to vaporize, restricting its use to suctions less
than about 85 kPa.
The cup conductance and the gauge sensitivity
determine the response time of the tensiometer. Ceramic
tensiometer tips/cups for field use generally have a
conductance in the order of 3 × 10−5 cm2 /s, which, in a
vacuum gauge system, would have a response time of five
seconds. A response time of 1 minute is adequate for most
field applications.
The tensiometer is sensitive to very small changes
in soil water, making it useful for irrigation scheduling.
Irrigation scheduling is a process to determine when to
irrigate and how much water to apply. Applying water in
an untimely manner with too little or too much water can
result in yield reductions. Overirrigation wastes water,
costs dollars to pump, and may leach nutrients beyond
the root zone. Tensiometers are particularly accurate at
low tensions, which is the wettest part of the soil water
range. Tensiometers are popular with growers of high-
value crops, such as vegetables and fruits, on sandy soils.
A depth label is usually placed on the vacuum gauge or
on the side of the tube to indicate the depth at which the
ceramic tip will be set when installed, which is important
for identification purposes. The soil suction reading on
the vacuum gauge dial is an indication of soil water
availability for plant use and does not need to be calibrated
for salinity or temperature. The readings, however, have
different meanings in terms of their use for irrigation
scheduling depending on soil type. Table 1 and Fig. 2
provide some interpretation of tensiometer readings in
relation to soil texture.
TENSIOMETERS WITH ELECTRONIC READER
The vacuum gauge tensiometer is not suitable for
automated continuous data collection or for controlling
irrigation system operation by switching it on or off.
However, some models have electronic technology added
to the tensiometer that allow them to be remotely read or
automatically switch irrigation on or off.
The switching tensiometer models employ the standard
vacuum gauge, to which a circular magnet is affixed to
the hub of the indicator needle. This magnet is rotated by
the action of the indicator needle as it moves according
to drying/wetting cycle of the soil. The switch is affixed
to the outside of the gauge face and can be adjusted for
various readings where the switch will turn ‘‘on’’ The
switch contains a reed switch that is held open by a small
biasing magnet installed inside the switch. This magnet
is of lesser power than the one mounted on the hub of
the gauge indicator needle. As the indicator needle of
726 VACUUM GAUGE TENSIOMETER

Table 1. Interpretations of Tensiometer Readings

Reading
(Centibars) Status Explanation/Action

0 Saturated Soil is saturated regardless of soil type. If readings persist, there is possible danger of waterlogged soils,
a high water table, poor drainage and soil aeration, or the continuity of the water column in the tube
may have broken.
5–10 Surplus water Indicate a surplus of water for plant growth. Drainage continues, and persistent readings indicate poor
drainage.
10–20 Field Capacity Field capacity for all types of soils. Additional water will drain as deep percolation carrying nutrients
without opportunity for plant use. Sandy soils, however, have very little storage capacity, and suction
values increase rapidly as plants remove water past 15 to 20 centibars. For sensitive crops, like
potato, rapid irrigation may be required before damaging stress can develop.
20–40 Irrigation range Available water and aeration good for plant growth in fine- and medium-textured soils. Irrigation is not
required for these soils at this range. Coarse-textured soils may require irrigation in the 20 to 30
centibar range and the finer sandy soils at 30 to 40 centibar ranges.
40–60 Usual range for starting irrigation. At 40 to 50 centibar, irrigation may need to be started for loamy
soils. On clay soils (silty clay loams, silty clays, etc.), irrigation usually starts from 50 to 60 centibars.
Heavy clay soils still have some available water. Irrigation, however, ensures maintaining readily
available soil water. The stage of growth and type of crop will influence the decision.
70 Dry Stress range. However, crop is not necessarily damaged. Some soil water is available in clay soils, but
may be low for maximum production.
80 Top range of tensiometer accuracy; higher readings are possible, but tension within the water column
inside tensiometer will break between 80 to 85 centibar, which has relationship to elevation of the
area compared with mean sea level. At higher elevation, the continuity of water column inside the
tube may break at a lower reading because of change in vapor pressure according to change in
atmospheric pressure.

partial vacuum reading is converted as current flow

Lim

e
ng signal. The partial vacuum, caused by water moving
ra
it o

n
tio out of tensiometer, causes a deflection in the face of the
f re

tia
ini transducer, which excites the electronic elements of the
ad

n 50
tio
ily

40 device to generate the signal in a certain milliampere

iga 60
av

Im Clay range equivalent to centibar or kPa linear scale. These

aila

30 Silt Clay 70 changing signals coming from the transducer may be

ble

Clay recorded by data logger or computer. The reading

wa

20 Sand 80
is continuous. The data acquisition computer may be
ter

e
ang 90 programmed to start an irrigation system at a certain
ac ity r 10
cap value. The automated versions of tensiometers enable
r

d
ity te

File 0 100
bil me

nge remote operation of the irrigation system.

ra
pa sio

ater
ca ten

ew
Fre Saturated WORKING PRINCIPLE OF A TENSIOMETER
ing of
ad it
re r lim

Soil water exists primarily as thin films around and

pe
Up

between soil particles and is bound to soil particles by

strong molecular forces. As the soil dries, the water
Figure 2. Schematic of the tensiometer vacuum gauge dial to
show interpretation of readings in relation to soil type. films become thinner and become more tightly bound
to soil matrix. This increase in tension within the films
now in contact with the tensiometer causes water to be
the tensiometer rises with the drying of soil to the point drawn from the ceramic tip. The withdrawal of water
where the switch has been set, the magnet on the gauge from the ceramic tip creates a partial vacuum within
overrides the small biasing magnet inside the switch, the tensiometer. Water continues to be drawn until the
which causes the reed switch to close. The switch is wired vacuum created inside the tensiometer equals the tension
into a low-voltage current loop from the control device of the water films outside. At this point, equilibrium is
to the solenoid valve. The switch closure completes the reached and water ceases to flow. The vacuum gauge
circuit and allows the solenoid valve to open and irrigation reading indicates the amount of suction or tension.
to start. When the soil rewets, the indicator needle drops As water is added to the soil from rainfall or
down and the switch opens interrupting current loop from irrigation, the soil suction is reduced. The higher vacuum
the control device that causes the solenoid valve to close, in the tensiometer causes soil water to be drawn
terminating irrigation. into the tensiometer, and the vacuum will be reduced
For automated data collection, the standard vacuum until a balance in tension is reached. The tensiometer
gauge is replaced by a pressure transducer, where the continuously responds and maintains a balance with
 VACUUM GAUGE TENSIOMETER 727

the soil water suction or tension, and the vacuum the soil in the top half to three-quarters of the root zone.
gauge indicates the amount of tension, hence the name If the potential root zone is less than 12 inches, a single
tensiometer. tensiometer may be installed in the center of the zone at
6 inches deep. With deeper-rooted crops, one tensiometer
should be placed at the upper one-quarter of the rooting
PREPARING THE TENSIOMETER FOR FIELD USE
depth and another at the lower quarter point or three-
quarters of the depth (see Fig. 3).
Preparing a tensiometer for use starts with filling the
In deeply rooted crops, or situations where a distinct
reservoir and body tube with distilled water, taking care
break in soil textures exists, three or more tensiometers
that the ceramic tip is wetted from one direction to avoid
may be needed. An example might be 18 inches of sand
air entrapment in the finer pores of the ceramic tip. Allow
overlying a silty soil on which corn is grown and 3 to 4 feet
the tensiometer to stand upright, and soon the tip will
of the root zone is to be managed. One tensiometer might
wet-up and free water will appear like sweat on the
be placed at 6 inches, a second at 18 inches, and a third at 2
outer surface of the ceramic tip. Refill as necessary to
to 3 feet. The differences in tension readings would make it
dispel all the air from the tensiometer. A hand-operated
possible to better assess the soil water conditions. Almost
vacuum pump may be used to help remove air from the
70% of crop water is supplied by the top half and only
tensiometer. Operating the pump when the tensiometer
10% from lower one-fourth of the rooting depth. Irrigators,
is filled with water and the tip is submerged in water
therefore, often manage only the top half or three-fourths
helps remove gases from the pores of the ceramic tip
of the root zone. Tensiometers should be long enough to
and solution. After each pumping, refill the tensiometer
reach the desired depth, and the diaphragm of the vacuum
completely with water and repeat until no more bubbles
gauge must not touch the ground. The tensiometer should
are observed. The tensiometer is then sealed by screwing
never be set in a hole or a depression.
the reservoir cap down securely.
Tensiometers may be installed using a soil auger or
After letting it stand in water in a bucket overnight,
a probe. Manufacturers also provide simple coring tubes.
the tensiometer may be set outside upright in the air. The
Placement should be in a crop row to avoid traffic. Where
air will start drying the tip. The gauge should read 70
furrow irrigation is used, the tensiometers may be angled
when the tip is air-dried. Repeat the wetting and drying
slightly to place the tip under the furrow. The electronic
cycle with the tensiometers that do not respond correctly
tensiometers may require a cover to safeguard the electric
the first time.
connections from sprinkler or rainwater. If a valve cover
Tensiometers require checking prior to installation.
box is used, the tensiometer tips need to be slanted out
As with any measurement device, proper care and
to be in the crop area. The hole should be small enough
maintenance are required. If the tensiometer was used
to create resistance to insertion of the tensiometer and
previously, begin by washing and rinsing it inside and out.
the bottom shaped to form a close soil contact at the tip,
Residues on the porous ceramic tip that were not removed
which may be accomplished by returning a little loosened
by washing may be removed by sanding lightly. Distilled
water treated with three to five drops of chlorine bleach
per gallon of water can be used to inhibit algae growth.
Manufacturers also provide solutions for water treatment,
which may be used according to direction. Distilled water
available in a grocery store has been found to be adequate.
If an excessive amount of air bubbles are noticed, then
boiling may be helpful, but the remaining water must be
stored in an airtight container.
If tensiometers are not to be installed immediately,
cover the tips with a plastic bag to prevent evaporation or
let them stand in a bucket of water until installed.

Zone
TENSIOMETER INSTALLATION of
moisture
Site Selection control
with two
The number of tensiometer installation sites required will tensiometers
depend on the crops grown and field conditions. Fewer
stations of tensiometers are needed when a single crop
is grown in large blocks of uniform soil. If the soils
are variable, different crops are to be grown and more
stations are necessary. Stations need to be selected to
represent an area, and care should be taken not to cause
excessive compaction or destruction of plants during the
installation process.
Except for very shallow-rooted crops, tensiometers are Figure 3. Sketch showing root zone moisture evaluation and
normally installed in groups of two or more to characterize control by two tensiometers.
728 VACUUM GAUGE TENSIOMETER
portion of the soil from the depth of placement back into
the hole and adding a little water. When the tensiometer
is pushed for placement, the soft soil will move around the
tip to conform to the rounded shape of the tip and make a
good contact. A good contact is absolutely essential for the
tensiometer to perform.
Tensiometers must be handled with care, because the
tips may break if handled roughly. The depth label on the
tensiometer will identify the root zone being monitored.
Tensiometer locations need to be marked both in the row
and at the edge of the field. A wooden stake painted
bright or a metal rod with a colored flag attached are
good markers. Locating tensiometers in tall crops can be
a problem. A written log of the station locations is also
recommended.
Servicing
Tensiometers are weatherproof, except for freezing, and
generally require very little service. When first installed,
there may be tiny air bubbles clinging to the sides of the
body tube. However, after one cycle of soil water use, which
creates a high vacuum, the bubbles will rise to the top and
can be eliminated by refilling. The amount of bubbles will
depend on the gas originally present in the vacuum gauge
and the amount dissolved in the water. Servicing is best
done soon after irrigation. Tensions are low, and air that
may have been drawn into the cup at high tensions can be
eliminated by refilling. Tensiometers return to equilibrium
rapidly at low tensions. They also respond quickly to a very
minute withdrawal of water from the system. If much air
is drawn into a tensiometer at low tensions, the porous
cup may be defective, and the tensiometer may need to
be replaced. Some air entry is unavoidable. When using
a number of tensiometers, watch for tensiometers that
accumulate abnormal amounts of air. The colored fluid
concentrate, supplied by the manufacturer for control
of algae, helps to spot air bubbles in the tensiometer
more easily.
Tensiometers should be removed from the field before
freezing. The water can freeze and break the ceramic tip or
the body, or damage the vacuum gauge. Tensiometers need
to be emptied before long-term storage, which prevents
salt deposition in the porous material with evaporation or
rusting of the gauge.
Troubleshooting
A tensiometer that is out of water or leaking will remain
at zero on the gauge, or the reading will fluctuate in the
low suction range. Two or more successive zero readings
may be a sign of a malfunction and should be investigated.
If the gauge remains at zero, refill with water and use a
hand pump, if available, to remove air. The tensiometer
may have been empty because of dry soil. If the tip was
dry, fine air bubbles will rise rapidly for several minutes
and then cease. If larger bubbles rise and continue, a leak
is indicated and the source should be determined. If the
bubbles rise from the bottom, remove the tensiometer and
replace the tip. If the bubbles enter from the side, the
body tube may be cracked and should be fixed. If bubbles
rise from the gauge, the leak may be in the gauge or the
threaded connection. A leaky gauge needs to be replaced,
but a threaded connection can be resealed. If no large
bubbles rise and yet the reading remains at zero then the
reservoir cap may be cracked or the seal may be defective.
Inspect for ‘O’ ring if there needs to be one for a proper
seal. In most cases, the trouble is easily corrected.
A damaged vacuum gauge may stick in one position
or may not respond smoothly with changes in soil water.
Check a suspect gauge against one known to be in good
working order or replace the suspect gauge with a new
one. Readings that are higher than expected, especially
after irrigation, generally do not indicate a tensiometer
failure. The irrigation water may not have penetrated to
the depth of the tensiometer tip.
General Guidelines
• Place two or more tensiometers of different lengths
near one another (a station), usually in the crop row.
Two stations may be enough in a field with uniform
soil and slope. One station should be near the start
of an irrigation set and the other at the end of that
irrigation set. The location of stations will depend on
the type of irrigation system. With furrow irrigation,
this location may be at the upper and lower quarter
points of the first set in the field. For a center pivot,
it may be in the outer and middle spans. If desired,
one set may be established at the starting point of
the pivot and another set at the endpoint.
• Tensiometer stations should be located in represen-
tative areas of the field. Do not position tensiometers
in low spots or on knobs, and place them where the
plant population is representative of the field.
• Tensiometer installation depth is determined by the
active root zone of the crop. For example, for a corn
crop on a deep soil, three tensiometers, installed at
depths of 12, 24, and 36 inches, are recommended at
each station.
• One should wait 24 hours after installing the
tensiometer to obtain reliable readings. If the soil
was dry at installation, irrigation or rainfall may be
needed before obtaining satisfactory readings.
• Tensiometers should be installed as early as possible
during the growing season and left in the field for the
duration of the growing season. The roots of the crop
must grow around the porous tip for reliable readings.
Moving of the tensiometer during the growing season
is not recommended.
LIMITATIONS OF TENSIOMETER
The major criticism of the tensiometer is that it functions
reliably only in the wet range of soil water at readings
of about 80 centibar or less. At higher readings, the
porous tip may leak air and the gases will be drawn
out of the water, and at low pressure, the water will
vaporize causing discontinuation of the tension column or
vacuum. The gauge reading will fall to zero, which is not
as serious as it may seem, because most of the available
water in coarse-textured soils, and about 50 percent or

more in fine-textured soils, have already been used up at
this range.
Another criticism is the price, which ranges from $35
to $60. When used in large quantities, the cost may
seem prohibitive. Irrigation scheduling, however, has been
shown to easily pay off through increased yields or reduced
pumping cost. Proper handling may extend the useful life,
and the cost may be spread over many seasons, making it
cost effective.
Finally, the ceramic tip may gradually fill with
precipitates because of soil water movement through the
pores, which slows water transfer through the tip and
increases the time required for the tensiometer to respond
to a change in soil-water conditions. Some slowing does no
harm, but if the response time becomes too slow, a new tip
should be installed. The response time may be improved
by rubbing the exterior of the tip with fine sand paper
or soaking the tip in a mild acid solution. The amount
of plugging depends on the soil-water chemistry and the
manner of use.
Where tensiometers can be left in the ground, the tip
porosity remains satisfactory for several years in most
soils. However, each time the tensiometer is removed from
the soil, tip life is reduced, which is particularly true if
the soil is calcareous or saline. In extreme cases, where
the tensiometer is installed and removed several times
per season, the tip may need to be replaced after one year
of use. To minimize this damage, a tip that is removed
from the soil should be protected from drying until the
tensiometer has been emptied, cleaned, and dried.
TILE DRAINAGE
TARA REED
University of Wisconsin-Green
Bay
Green Bay, Wisconsin
Tile (or relief) drains are porous pipes buried approxi-
mately 1 m beneath the surface of the soil. These pipes
drain excess water from the land into nearby water bodies,
most typically cisterns, open ditches, or streams. The most
common use of tile drainage is to improve agricultural pro-
duction by draining excess moisture from the plant root
zone and allowing cultivation normally saturated soils.
Subsurface tile drainage also lowers the water table and
reduces soil salinity. Many of the world’s productive agri-
cultural areas, including 25% of the cropland in the United
States, require artificial drainage (1). The area drained by
subsurface drainage systems in the United States has
increased since 1900 from less than 5 to almost 30 million
hectares (2).
TILE DRAINS AND SURFACE WATER QUALITY
The impact of tile drainage on receiving water bodies
varies (3). Site-specific variables such as soil character-
istics, topography, climate, and precipitation patterns as
TILE DRAINAGE 729
well as the configuration of the drainage network and
other land management practices determine the pro-
portion of precipitation resulting in surface runoff and
subsurface drainage and the overall impact on receiv-
ing water bodies. Because rainfall can penetrate further
into the soil, surface water drainage is reduced in tile-
drained fields compared to saturated soils (4). Depending
on local conditions, subsurface drainage can reduce surface
runoff (4,5).
Increased infiltration prolongs the time that water
is in contact with the soil, thereby often reducing the
transport of pollutants that are likely to adsorb to soil
particles and increasing leaching of pollutants that are
easily released (6). The difference between surface and
subsurface water drainage may be minimized when there
are direct surface water inlets into the subsurface drainage
network or during particularly heavy rains, although
Schwab et al. (5) found that good subsurface drainage
decreased storm peak discharge by as much as 30%. On the
other hand, management practices such as riparian buffer
strips that are designed to reduce surface runoff usually
do not target subsurface drainage. In these cases, tile
drains may result in increased pollutant loading during
storms (7).
Pesticides and Herbicides
Agricultural pesticides and herbicides enter surface
waters by both surface runoff and subsurface drainage.
Whether subsurface drainage decreases the amount of
pollutant entering the waters depends on the hydrologic
characteristics of the site and the particular chemi-
cal used. The two pollutant characteristics important
to this discussion are persistence (the time it takes to
degrade to a nonpollutant form) and soil sorption (the
degree to which the pollutant adsorbs to soil parti-
cles). Persistent pesticide/herbicides are most likely to
drain (either by surface runoff or subsurface drainage)
into nearby receiving waters. Subsurface drainage is
most likely to reduce the transport of pesticides and
herbicides with high soil sorption qualities. Soil char-
acteristics can influence sorption, which is greatest
in strongly organic soils and those that have high
clay content.
Pesticides are most commonly detected in subsurface
flow during the first heavy rain following application.
Other peak periods include soil thawing and tillage. How-
ever, the most important factors determining pesticide
transport are precipitation volume and timing and soil
type. Pesticide loss through tile drainage is usually sub-
stantially lower than loss through surface runoff. For an
excellent review of tile drainage and pesticide transport,
see Kladivko et al. (6).
Nutrient Pollution
Nitrogen. Much of the fertilizer applied to a field can
be lost through tile drains. This is particularly true
for nitrogen, which is more mobile than phosphorus in
soil. Different species of nitrogen react differently. Tile
drainage usually decreases losses of organic nitrogen and
increases loss of nitrates and ammonium (3).
730 TILE DRAINAGE
Depending on the form of fertilizer applied, in wet
years, up to 70% of the nitrate applied as fertilizer
can flow off fields through drainage systems (8). Manure
application leads to significantly less nitrate loss to tile
drains than commercial fertilizer application (9). Drain
placement may also be important for reducing nitrogen
losses through drain tiles. For example, shallow drains
reduce nitrate leaching (10), and more frequent drain
tile spacing reduces ammonium losses (11). Placement
of tile drains in areas of preferential flow for manure
or fertilizer applications can elevate effluent nutrient
concentrations (12).
Phosphorus. Even though phosphorus (P) adsorbs stron-
gly to soil particles, artificial drainage is an important
contributor to nonpoint-source phosphorus pollution (13).
Burke (14) reported phosphorus losses through drainage
of approximately 33% of the amount applied in fertilizer.
Other estimates range from 2% (15) to 17% of the applied
phosphorus lost in drain tile effluent (16).
The greatest P loss from agricultural systems occurs
during intensive rains (17). At these times, both surface
flow and tile drain effluent P concentrations are high-
est (18). Tile drain discharge may be nonlinearly related
to rainfall intensity. For example, in a small drainage
system in Sweden, Ulen and Persson (19) found that half
the yearly P export through drain tiles occurred during an
average of 140 hours of episodic rain.
The ability of soils to retain P depends on soil
P concentrations. As soils become P saturated, leaching
becomes more prevalent. The point at which this happens
varies in different soils ranging from 10–119 mg Olsen
P kg−1 soil (20). The mass of phosphorus lost from
agricultural systems via tile drains is small compared
to nitrogen (21), but the impact on phosphorus-limited
aquatic systems may be substantial.
TILE DRAINAGE AND MANAGEMENT
Agriculture in many areas would be difficult, if not impos-
sible, without subsurface drainage. Research is ongoing
into strategies for managing pollutants in drain tile
effluent. Examples include controlled drainage where
subsurface drainage systems are fitted with barriers to
restrict drainage (3), subirrigation systems that both con-
trol drainage and provide increased irrigation to crops
during nonsaturated periods (22), constructed treatment
wetlands (23), and land retirement (24). Finding solu-
tions that are both environmentally and agriculturally
acceptable is critical to the survival of the world’s water-
ways.
BIBLIOGRAPHY
1. Pavelis, G.A. (1987). Farm Drainage in the United States.
In: History, Status, and Prospects. G.A. Pavelis (Ed.). USDA-
ERS, Misc. Publ. 1455, pp. 110–136.
2. Shirmohammadi, A., Wenberg, R.D., Ritter, W.F., and
Wright, F.S. (1993). Management of irrigation and drainage
systems. In: Integrated Perspectives. R.G. Allen, (Ed.). ASCE,
New York, pp. 531–41.
3. Skaggs, R.W., Breve, M.A., and Gilliam, J.W. (1994). Hydro-
logic and water quality impacts of agricultural drainage. Crit.
Rev. Environ. Sci. Technol. 24: 1–32.
4. Bengtson, R.L., Carter, C.E., McDaniel, V., and Halver-
son, B.E. (1984). Corn silage response to subsurface drainage
in alluvial soil. Trans. ASAE. 27: 1391.
5. Schwab, G.O., Fausey, N.R., Desmond, E.D., and Holman,
J.R. (1985). Tile and surface drainage of clay soils. IV.
Hydrologic performance with field crops (1973–80); V. Corn,
Oats and Soybean Yields (1973–80); VI Water quality; Cross
Moling over Tile Drains. Wooster, OH: OARDC, Res. Bull.
1166.
6. Kladivko, E.J., Brown, L.C., and Baker, J.L. (2001). Pesticide
transport to subsurface tile drains in humid regions of North
America. Crit. Rev. Environ. Sci. Technol. 31: 1–62.
7. Reed-Andersen, T. (1999). Flows of phosphorus on an
agricultural landscape: implications for eutrophication and
restoration of Lake Mendota. PhD thesis. University of
Wisconsin - Madison, Madison, WI, pp. 113–138.
8. Randall, G.W. and Iragavarapu, T.K. (1995). Impact of long-
term tillage systems for continuous corn on nitrate leaching
to tile drainage. J. Environ. Qual. 24: 360–366.
9. Mitchell, J.K., McIsaac, G.F., Walker, S.E., and Hirschi, M.C.
(2000). Nitrate in river and subsurface drainage flows from
an east central Illinois watershed. Trans. ASAE. 43: 337–
342.
10. Gordon, R., Madani, T.A., Caldwell, K., Boyd, N., Astatkie, T.,
and Jamieson, R. (2000). Subsurface nitrate-N leaching loss
as affected by drainage size and depth in a shallow slowly-
permeable soil. Can. Water Resourc. J. 25: 331–342.
11. Singh, M.A., Bhattacharya, K., Nair, T.V., and Singh, A.K.
(2001). Ammonium losses through subsurface drainage
effluent from rice fields of coastal saline sodic clay soils.
Water, Air, Soil Pollut. 127: 1–14.
12. Schellinger, G.R. and Clausen, J.C. (1992). Vegetative filter
treatment of dairy barnyard runoff in cold regions. J. Environ.
Qual. 21: 40–45.
13. Sims, J.T., Simard, R.R., and Joern, B.C. (1998). Phosphorus
loss in agricultural drainage: Historical perspective and
current research. J. Environ. Qual. 27: 227–299.
14. Burke, W. (1975). Fertilizer and other chemical losses
in drainage water from Blanket Bog. Irish Journal of
Agricultural Research 14: 163–78.
15. Calvert, D.V. (1975). Nitrate, phosphate, and potassium
movement into drainage lines under three soil management
systems. J. Environ. Qual. 4: 183–86.
16. Johnson, W.R., Ittihadieh, F., Daum, R.M., and Pil, A.F.
(1965). Nitrogen and phosphorus in tile drainage effluent.
Soil Science Society Proceedings, pp. 287–289.
17. Bengtson, R.L., Carter, C.E., Southwick, L.M., Willis, G.H.,
Fouss, J.L., and Rogers, J.S. (1993). Management of irrigation
and drainage systems. In: Integrated Perspectives. R.G. Allen
(Ed.). ASCE, New York, pp. 523–30
18. Phillips, P.A., Culley, J.L.B., Hore, F.R., and Patni, N.K.
(1982). Dissolved inorganic nitrogen and phosphate concen-
trations in discharge from two agricultural catchments in
Eastern Ontario. Agric. Water Manag. 5: 29–40.
19. Ulen, B. and Persson, K. (1999). Field-scale phosphorus losses
from a drained clay soil in Sweden. Hydrological Processes
13: 2801–12.
20. Hesketh, N. and Brookes, P.C. (2000). Development of an
indicator for risk of phosphorus leaching. J. Environ. Qual.
29: 105–110.
 TILE DRAINAGE: IMPACTS, PLANT GROWTH, AND WATER TABLE LEVELS
21. Kladivko, E.J., Van Scoyoc, G.E., Monke, E.J., Oates, K.M.,
and Pask, W. (1991). Pesticide and nutrient movement into
subsurface tile drains on silt loam soil in Indiana. J. Environ.
Qual. 20: 264–270.
22. L.A. Zucker and L.C. Brown (Eds.). (1998). Agricultural
Drainage: Water Quality Impacts and Subsurface Drainage
Studies in the Midwest. Ohio State University Extension,
Bulletin 871.
23. Kovacic, D.A., David, M.B., Gentry, L.E., Starks, K.M., and
Cooke, R.A. (2000). Effectiveness of constructed wetlands in
reducing nitrogen and phosphorus export from agricultural
tile drainage. J. Environ. Qual. 29: 1262–1274.
24. Purkey, D.R. and Wallender, W.W. (2001). Drainage reduc-
tion under land retirement over shallow water table. J. Irrig.
Drain. Eng. 127: 1–7.
TILE DRAINAGE: IMPACTS, PLANT GROWTH,
AND WATER TABLE LEVELS
CHANDRA MADRAMOOTOO
Macdonald Campus of McGill
University
Ste-Anne de Bellevue, Quebec,
Canada
Tile drainage is a method of artificial drainage, which
lowers the water table and reduces water logging on
productive agricultural lands. Improved drainage is an
agricultural, rather than a land conversion, practice. The
objective is to increase production efficiency, crop yields,
and profitability on poorly drained agricultural lands.
In the wetter, humid regions, precipitation will cause
the water table to rise at, or near, the soil surface. If
high amounts of precipitation occur during the growing
season, the water table could rise within the crop root
zone. The same situation occurs in the drier, arid, and
semi-arid regions where irrigation is extensively practiced.
Continual irrigation and canal seepage over the years lead
to an accretion of the water table, and this could also be
injurious to plant growth.
Every soil has some degree of natural drainage.
However, if the rates of precipitation or irrigation exceed
the natural drainage rate of the soil, the high water table
conditions will impair crop growth, because of a lack of
oxygen in the root zone. Artificial, or tile drainage, is
therefore required to remove excess soil water and to
lower the water table, to maintain an optimum soil–air
environment within the crop root zone. Most of the world’s
highly productive agricultural soils require either surface
drainage or tile drainage, or a combination of both, to
maintain stable crop yields. It is estimated that there are
about 200 million hectares of cropland worldwide with
improved drainage. The benefits of drainage are given in
Fig. 1.
Tile drainage (also known as subsurface pipe drainage)
is the installation of horizontal, corrugated, perforated
plastic pipes below the soil surface to enable water table
drawdown. Parallel lateral pipes are normally of 75 to
100 mm in diameter and are installed on grades varying
from 0.1% to 3%. The downstream ends of the laterals
731
• Ability to better manage and control the water table for
enhanced crop productivity
• Reduced flooding
• Better soil water conditions for seedbed preparation and planting
• Earlier and more uniform seed germination
• Improved field machine trafficability
• More favorable air and salt balances in the crop root zone
• Higher crop yields and reduced yield variability
• Ability to plant higher value crops
• Higher farm incomes
• Increased land values
• Less soil erosion
• Better uptake of nutrients
• Improved soil structure
• Enhanced soil microbiological activity
• More timely field machine operations
• Ability to effectively lengthen the growing season, and make
better use of lower heat units in northern regions
• Potential to improve water quality
Figure 1. Benefits of tile drainage.
are connected to a buried collector pipe. The diameter of
the collector increases with the area drained. The spacing
between the lateral pipes depends on the soil type and
the amount of water to be removed. The lateral pipes
can be installed in a parallel layout, herringbone layout,
or at random. Depending on topography, land formation,
and proximity of a water receiving body, the collector may
outlet by gravity to a watercourse, or into a sump. In the
case of the latter, the discharge is then pumped to a lake or
stream. Before the advent of plastic pipe in the late 1970s,
clay tile pipes were used. However, plastic pipes are lighter
and easier and faster to install. The drainage coefficient
(or volume of water to be removed by the drainage system)
in humid regions is about 10–15 mm/day.
One adaptation of tile drainage is the use of surface
inlets to remove water ponded in surface depressions, or
as conservation structures to reduce surface runoff and
erosion on sloping lands. The drainage coefficient for these
systems ranges from 15 to 20 mm/day.
In the irrigated regions, tile drainage not only controls
the rise of the water table, but also reclaims salinized soils.
The salinity of irrigated soils could increase with time,
and this also reduces crop yields. Water in excess of the
plant evapotranspiration requirement is applied during
irrigation. This additional quantity of water is known as
the leaching fraction. Salt is then leached from the root
zone and removed from the field via the tile drains. Deeper
drain installation ensures that salts do not rise to the
root zone because of capillary action, which also prevents
waterlogging of the root zone. The amount of surplus
irrigation water to be removed by the tile drainage system
is less than in humid regions. Consequently, lateral pipe
drainage spacings are wider than in the humid regions.
The drainage coefficient in irrigated regions varies from 2
to 5 mm/day.
Horizontal pipe drainage systems are usually installed
with either a trenchless plow or a chain or wheel type
trencher. The trenchless plow is widely used in Europe
and North America and results in faster installation rates.
The depth and grade of the pipe installation equipment
732 MEASURING AND MODELING TREE AND STAND LEVEL TRANSPIRATION
is achieved with a horizontal laser grade control system.
In cases in which installations are conducted in fine sand
and silty soils, the pipes are wrapped with a synthetic
geotextile, to minimize blockage of the drainage system by
the soil particles.
Another form of subsurface drainage is known as
vertical drainage or tubewell drainage. It primarily con-
trols waterlogging and salinity in the irrigated regions.
The primary purpose of tubewells is to abstract ground-
water for irrigation, and consequently, the water table
is lowered, and salinization caused by capillarity is
minimized. This situation is ideal where the ground-
water is not very brackish, or saline, and therefore
suitable for irrigation. In areas where the groundwa-
ter is saline, then the pumped irrigation water must
first be mixed with fresher, or sweeter, water in irriga-
tion canals.
When subsurface pipe drainage reclaims salinized and
waterlogged lands, it is viewed as an environmentally
beneficial practice, because degraded lands that were once
intensively irrigated and cropped are returned to their full
productive potential.
Drainage installations result in both hydrologic and
land use changes. On poorly drained lands, surplus water
is either ponded on fields or moves as surface runoff to
nearby open water bodies. On tile drained lands, water
percolates below the root zone, which reduces surface
ponding and runoff. Infiltrated water is stored in the
soil profile and will eventually percolate to the deeper
groundwater table.
Some of the most significant technological develop-
ments in tile drainage occurred in the period 1965–1975.
Beginning in the 1980s, environmental concerns about
drainage water quality emerged. High concentrations
of nitrates were being measured at the outlet of tile
systems. Drainage practitioners responded to these con-
cerns with new technologies such as watertable man-
agement, controlled drainage, and subirrigation. Tile
drainage is now regarded as a multifunctional water man-
agement practice, with systems being designed to not
only serve crop needs, but also minimize environmen-
tal impacts.
Controlled drainage and subirrigation make use of the
existing tile drainage system to raise the water table
during certain periods of the year. By either restricting
the outflow of water from the tile outlet or by pumping
water back into the drainage system, particularly during
the growing season, the crop can obtain water from the
water table to meet its evapotranspiration requirement.
This process helps to reduce the effects of drought stress
during the growing season, thus increasing crop yields. An
additional important benefit is that the amount of nitrates
leaving the tile outlet is reduced, because of a process
known as denitrification.
The benefits of tile drainage go beyond the control
of excess soil water and salts in the crop root zone.
Environmental and socioeconomic benefits are associated
with vector control and public health. For example,
drainage of stagnant water eliminates malaria, foot-rot in
large animals, yellow fever, the liver fluke snail, and other
waterborne diseases. The overall benefit is a healthier
human population. Drainage also reduces or eliminates
mildew infections and various root rots of plants.
MEASURING AND MODELING TREE AND
STAND LEVEL TRANSPIRATION
J.M. VOSE
G.J. HARVEY
K.J. ELLIOTT
B.D. CLINTON
(from Phytoremediation:
Transformation and Control of
Contaminants, Wiley 2003)
SUMMARY OF PRACTICAL IMPLICATIONS
Transpiration is a key process in the application of
phytoremediation to soil or groundwater pollutants. To
be successful, vegetation must transpire enough water
from the soil or groundwater to control or take up the
contaminant. Transpiration is driven by a combination of
abiotic (climate, soil water availability, and groundwater
depth) and biotic (leaf area, stomatal functions, root
amount and distribution, and hydraulic characteristics)
that need to be evaluated when considering appropriate
site and species combinations. The protocols are not trivial,
but transpiration can be measured at a variety of scales
using techniques such as direct measurements of sap flow
on individual trees, eddy flux gradient analyses, or gauged
watersheds. Alternatively, models can be used to estimate
transpiration, but these usually require on-site calibration
or parameterization to produce accurate predictions. Case
study analyses across a range of site conditions and
species indicate a maximum transpiration capacity of
approximately 7.5 × 106 liters of water per hectare per
year (8 × 105 gallons of water per acre per year), with
a range of 1.5 × 106 to 7.5 × 106 liters per hectare per
year (1.6 × 105 to 8 × 105 gallons per acre per year).
Variation among sites is related to species, tree size, and
stocking (i.e., vegetation density) differences. Application
of a physiologically based and site-specific parameterized
model suggests reasonable agreement between measured
and predicted transpiration estimates for the Air Force
Plant 4 site in central Texas.
IMPORTANCE OF ACCURATE MEASUREMENTS OF
TRANSPIRATION
Transpiration—the amount of water used by a tree or
stand of trees—is one of the key processes in the appli-
cation of phytoremediation of soil water or groundwater
pollutants. To be successful, native or planted vegetation
must transpire enough water from the soil or groundwa-
ter layer containing the pollutant to control the transport
or decrease the mass of contaminant. Hence, quantify-
ing current and future transpiration and determining the
principal location of water uptake by native and planted
vegetation on the site must be the evaluation criteria
for applying phytoremediation. Quantifying transpiration
 MEASURING AND MODELING TREE AND STAND LEVEL TRANSPIRATION
requires a thorough and accurate assessment of water use
patterns such as, transpiration rates, depth of soil water
uptake, interactions with climate, and soil water avail-
ability. Measuring current transpiration or predicting
future transpiration is not trivial. Because transpiration
is an integrated response of the atmosphere–plant–soil
continuum, measurements and predictions of transpira-
tion capacity must account for (1) variation in climatic
driving variables (i.e., solar radiation, water vapor satura-
tion deficit, precipitation, wind speed, and temperature),
(2) structural and physiological (leaf stomatal function)
characteristics of the vegetation (leaf surface area, and
root area and extent), and (3) soil water dynamics (water-
holding capacity, and permeability).
Evapotranspiration and transpiration are often used
interchangeably, but these processes are different. Evapo-
transpiration includes the amount of water transpired by
the vegetation, and losses due to evaporation of intercepted
precipitation and soil surface evaporation. In forests, inter-
ception evaporation is a function of rainfall intensity
and leaf and branch surface area, ranging from about
10 to 50 percent (1–3). In closed canopied forests, soil
evaporation is a minor component of the overall water
budget (3), but may become increasingly important in open
stands. From a phytoremediation perspective, transpira-
tion is the key factor to consider because interception
evaporation does not involve soil water or groundwa-
ter.
The process of transpiration involves water movement
through the soil, roots, stems, and leaves into the atmo-
sphere in response to water potential gradients—always
moving in the direction of smaller potential or negative
gradients. Water potential is near zero when water is
freely available and decreases to negative values when
water becomes more limiting. The movement of water
from the leaf interior to the atmosphere occurs through
small openings in the leaf called stomata, which open
and close in response to external (e.g., climatic factors)
and internal (e.g., water potentials of leaves) driving vari-
ables. Species vary considerably in stomatal responses
to these driving variables and provide opportunities for
selecting species to optimize transpiration in different
climatic environments.
Five methods are used to quantify transpiration:
(1) precipitation minus runoff on gaged watersheds,
(2) energy balance (e.g., Penman-Monteith equation),
(3) eddy covariance, (4) hydrologic models, and (5) direct
sap-flow measurements. The first three methods are inte-
grated estimates for the entire vegetation–soil complex
and provide estimates of evapotranspiration not tran-
spiration. Hence, those methods do not directly parti-
tion water losses based on transpiration versus evap-
oration and provide no information on the source of
water (i.e., shallow versus deep soil layers) for tran-
spiration. Estimating transpiration with methods 1, 2,
and 3 requires an independent analysis of the contribu-
tion of interception and soil surface evaporation. Hydro-
logic models vary considerably in complexity, ranging
from very simple models [e.g., Thornthwaite (4) indices
of potential evapotranspiration] to detailed physiologi-
cally based models that link vegetation, soils, and the
733
atmosphere (3). In contrast, sap-flow measurements pro-
vide a direct measure of transpiration (after correcting
for time lags) under field conditions at the individ-
ual tree level (5–7). However, modeling or other scaling
approaches are required to extrapolate tree-level measure-
ments to the stand.
In summary, there are numerous approaches to
quantifying transpiration in native or plantation-derived
vegetative ecosystems. However, these methods vary
considerably in accuracy, in data and measurement
requirements, and in the capability to predict future
transpiration rates as stands develop. In this chapter,
we review approaches to quantifying forest transpiration
from the leaf level to the stand and discuss the pros
and cons of different approaches. We then provide
applications of a subset of these approaches from
phytoremediation case studies in Texas, Colorado, and
Florida.
OVERVIEW OF CONTROLS ON TRANSPIRATION
Transpiration rates vary considerably among species and
geographic regions (Fig. 1). Which factors contribute to
this variation? At large scales (i.e., regions), climate is
an overriding control. The strong relationship between
evapotranspiration and precipitation (Fig. 1) suggests that
transpiration is principally limited by soil water supply.
However, other climatic factors such as temperature,
atmospheric vapor pressure deficit, and solar radiation
also play important roles and interact with soil water
availability and physiological status of the plants (Fig. 2).
For example, one of the key effects of temperature is
through the influence on the length of growing season, in
which longer periods with temperatures above freezing
promote longer leaf area duration and hence, surface
area available for transpiration. Frozen or cold soils
also restrict transpiration (9) by limiting the permeability
of cell membranes (10,11). Solar radiation provides the
Evapotranspiration vs. Precipitation
1400
Conifers
Evapotranspiration (millimeters)
1200 Hardwoods
1000
Y = 42.43 + 0.77(X )
800
r 2 = 0.82; p < 0.05
600
400
200
0
200 400 600 800 1000 1200 1400 1600 1800 2000
Precipitation (millimeters)
Figure 1. Relationship between annual evapotranspiration (Y)
and precipitation (X) for hardwood and confier species [redrawn
from (3,8) and data from this chapter]. The regression line
represents the data for the conifer species only. Note that r2
is the correlation coefficient and p is probability.
734 MEASURING AND MODELING TREE AND STAND LEVEL TRANSPIRATION

energy for transpiration and regulates stomatal opening. in the southwestern U.S., because transpired water can
As a result, a strong relationship generally occurs between be derived from groundwater (13,14). Several studies
solar radiation and transpiration, estimated as sap flow have evaluated the influence of phreatophytes on surface
in Fig. 2. Atmospheric vapor pressure deficit provides and groundwater [e.g., (15–17)] from the perspective
the gradient to which leaf-water vapor responds through of negative impacts on streamflow and groundwater
the leaf stomata (Fig. 2), and wind speed has a direct recharge. From a phytoremediation standpoint however,
influence on the leaf boundary layer (12). Optimal climatic the high water consumption of phreatophytes has a
conditions for transpiration include high soil water positive effect to decrease aquifer recharge and influence
availability, high solar radiation, high vapor pressure the movement of contaminated shallow groundwater.
deficits, warm temperatures for extended periods, and The structure, morphology, and physiological charac-
high wind speed. In most cases, these conditions do teristics of the vegetation are also important regulators of
not occur simultaneously because increased soil water transpiration. For example, at equal precipitation inputs,
availability is usually a result of high rainfall that there are large differences in transpiration between conifer
decreases solar radiation (due to increased cloud cover) and and hardwood species (Fig. 1), with hardwoods generally
vapor pressure deficit (due to higher humidity). Species lower than conifers. Causes for these coarse scale differ-
that have the ability to utilize deeper sources of soil or ences are generally well known. The single greatest con-
groundwater [i.e., phreatophytic vegetation such as poplar trolling factor is the quantity of leaves, expressed as leaf
(Populus spp.) and willow (Salix spp.)] are an especially area index (in square meters per square meter). Site water
attractive option in hot, dry, and windy environments availability and leaf area are related in that, sites with the

3.0
May
2.5 July
Sept.
2.0

1.5

1.0

0.5

0.0
0 1 2 3 4
Mean hourly sap flow (kilograms per hours)

Solar radiation (mega joules per squaremeter)

3.0

2.5

2.0

1.5

1.0

0.5

0.0
0 500 1000 1500 2000 2500 3000
Vapour pressure deficit (megapascals)

3.0

2.5
2.0

1.5

1.0
0.5

0.0
Figure 2. Mean hourly sap flows versus climatic 5 10 15 20 25 30
driving variables for three seasonal measurement (May,
July, and September) periods in central Colorado. Temperature (°C)
 MEASURING AND MODELING TREE AND STAND LEVEL TRANSPIRATION
greatest water availability typically have the highest leaf
area index (18,19), although nutrient availability (20,21)
and temperature also play a role (22,23) in determining
the maximum leaf area index. Watershed studies have
documented strong relationships between leaf area and
streamflow, with streamflow increasing exponentially as
leaf area decreases (24,25). Because precipitation minus
streamflow is an estimate of evapotranspiration at water-
shed scales, the implication is a direct control of stand-level
transpiration by leaf area. Other structural and physiolog-
ical factors regulating transpiration include the amount
and permeability of sapwood and stomatal characteristics
such as conductance and responsiveness to climatic varia-
tion and overall plant water status (i.e., water potential of
leaves). Differences among species in leaf area, the rate of
attainment of maximum leaf area, and physiological char-
acteristics regulating the rate of water movement through
the plant (sapwood amount and permeability, and stom-
atal conductance) provide opportunities for manipulating
vegetation composition and structure to optimize tran-
spiration. Optimal structural and physiological conditions
for high transpiration amounts include rapid development
of high leaf area, high stomatal conductance and sap-
wood permeability, and physiological characteristics that
facilitate rapid responses to climatic conditions promoting
transpiration.
Because transpiration is a function of root uptake
from the soil and groundwater, soil characteristics are an
important factor determining transpiration. Root growth
and volume of soil occupied by roots are also important
because water movement is slow when soils are drier
than field capacity. Several factors determine soil water
availability. First, the amount of precipitation entering
the soil is a function of infiltration rate. Soils with low
infiltration rates due to factors such as compaction or fine
texture will have lower soil water availability because
some precipitation may move across the soil surface in
overland flow. Once in the soil, soil water availability
is a function of water holding capacity and unsaturated
hydraulic conductivity, both of which are determined
by soil texture. Texture impacts water availability in
different ways. Heavy clay soils (e.g., pore size less
than 0.2 micrometer) have limited soil water availability
because of very low rates of movement in the soil (i.e.,
conductivity) due to the fine pore space. In contrast,
coarse textured sandy soils (e.g., pore size greater than
50 micrometers) have low water availability because of
rapid drainage. Rooting volume and the presence or
absence of restrictive layers are also important soil factors
determining transpiration. For example, compacted soils
provide a physical barrier to root growth, limiting root
extension (26).
QUANTIFYING TRANSPIRATION
Leaf Level
Because water exits the plant primarily through leaf stom-
ata (a small amount of cuticular transpiration may also
occur in stems of some species), leaf–water relations are
a key factor determining whole-plant transpiration (27).
735
The concentration gradient of water vapor between the
interior of the leaf and the atmosphere at the leaf bound-
ary layer defines the maximum transpiration rate. Vapor
exchange is also determined by the opening size of the
stomata. When stomata are wide open, transpiration
occurs at about 20 percent to 40 percent of the rate of
evaporation of open water (28), whereas closed stomata
limit transpiration to less than 1 percent of open water.
Stomatal opening is controlled by guard cell turgor, which
responds to light, temperature, vapor pressure, and water
potential of the leaves. The rate of movement of water
through the stomata is the stomatal conductance. The
rate of stomatal response to climatic conditions varies by
species, but generally reflects responses to current con-
ditions, whereas stomatal responses to water potential
in leaves may reflect previous climatic and environmen-
tal conditions.
Because of the tight linkage between transpiration,
leaf stomatal conductance (hereafter referred to as leaf
conductance), and water potential of leaves, knowledge of
all three parameters is useful for evaluating transpiration
capacity. For example, species that exhibit high leaf-
level transpiration and conductance, and maintain high
water potential in the leaves have the capacity to
transpire large quantities of water. Similarly, the
relationship between water potential in the leaves and
conductance is often threshold dependent; i.e., species
that maintain high leaf conductance at low water
potential have the capacity to transpire more water
under dry conditions (29). Because of the importance
of factors such as leaf area index and distribution,
sapwood amount and permeability, and the difficulty in
extrapolating spatially and temporally from the leaf to
stand level, there may be no direct correspondence between
leaf-level transpiration and overall stand transpiration.
Typically, leaf conductance and tree and stand-level
transpiration are most highly related in young stands
with simple canopy architecture (7), such as closely
spaced, even-aged monocultures. However, as stands
develop, the linkage between leaf conductance and tree
or stand-level sap flow declines due to shifts in the
importance of stomatal versus boundary layer conductance
to total vapor phase conductance (30,31). Hence, leaf-level
measurements should only be used as an indicator of
transpiration capacity.
Tree Level
Transpiration at the whole-tree level represents the
integrated movement of water vapor from all the leaves
in the crown of the tree. As mentioned in the previous
section, spatial and temporal variation severely limits
extrapolation of individual leaf measurements to the
tree, so more direct measurements at the tree level are
required. Two approaches have typically been used. In
a few instances, entire trees have been enclosed in a
cuvette and the flux of water vapor calculated based on
the rate of increase in humidity within the enclosure.
This approach is severely limited by methodological
constraints such as the size of trees, heat buildup within
the cuvette, and alterations in the boundary layer and
vapor pressure gradients.
736 MEASURING AND MODELING TREE AND STAND LEVEL TRANSPIRATION
Sap-flow rate and volume have also been used as an
estimate of transpiration (32). Because of lags between
water movement in the stem and leaf-level transpiration,
sap flow is not a direct measure of transpiration,
but can be corrected after accounting for lags (33,34).
Typically, a 1- to 2-hour lag correction is applied to
real time sap-flow data to account for this temporal
difference (7,34).
Two sap-flow techniques have been utilized; heat
balance and heat pulse. For the heat-balance approach,
collars consisting of a heating element and thermocouples
above and below the heating element are placed around
the stem and the entire stem section is heated. Sap flow is
calculated using the heat-balance principle based on the
difference in temperature between thermocouples above
and below the heated stem section, after subtracting
for heat loss due to conduction by stemwood (35). An
advantage of this approach is that it integrates sap
flow along the entire stem and does not require an
independent estimate of sapwood area. For larger trees,
paired probes are inserted vertically into the sapwood (36).
The upper probe is heated and both contain thermocouples.
The probes measure heat dissipation, which increases
with sap flow and the resultant cooling of the heat
source, as the apparent thermal conductance of sapwood
increases with sap velocity. To convert sap velocity
to sap flow rate, the cross-sectional area of sapwood
must also be determined. Typically, trees are cored
and sapwood to heartwood ratios quantified. Because
sap-flow probes measure sap flow velocity at only one
location, multiple probes are required to adjust for the
variation in sapwood thickness and permeability in the
stem section. Despite this, unaccounted for variation in
horizontal and vertical variation in sapwood thickness and
permeability introduces some error into sap-flow estimates
obtained with probes. The magnitude of error can be
determined experimentally and corrected for in small trees
by comparing sap flow with actual transpiration using
procedures such as weighing lysimeters. In large trees,
corrections are much more difficult and hence, predictions
have more uncertainty. In contrast to the heat-balance
method, the heat pulse method estimates sap flow based
on the time lag between pulses of heat and the distance
between the sensors (37).
Stand Level
While it is informative to understand transpiration
at the leaf and tree level to help evaluate species
and environments suitable for phytoremediation, stand-
level transpiration ultimately determines how much
soil water and groundwater are removed. However,
unlike leaf and tree measurements, no methods directly
measure stand transpiration. Instead, three indirect
measurement approaches have been utilized. These
approaches involve gaged watersheds, extrapolation of
individual tree measurements, and eddy flux estimates.
Gaged watersheds require a combination of well-defined
watershed boundaries, tight bedrock, and well-constructed
weirs or gages to provide accurate transpiration estimates.
If these criteria are met, then evapotranspiration (ET) is
estimated by the equation
ET = P − RO ± soil water storage (1)
where P = precipitation and RO = runoff, determined
from weirs or gages. Because P is a component of the
equation, the accuracy of precipitation measurements
will also influence evapotranspiration estimates. Changes
in soil water storage are usually assumed to be neg-
ligible at annual time steps, although this is clearly
not the case over shorter intervals. Hence, using this
approach at time steps less than a year requires deter-
mining changes in soil water storage. Because evapo-
transpiration is estimated, interception evaporation must
be determined and subtracted to estimate transpira-
tion.
Extrapolating individual tree measurements to the
stand can be done in a number of ways. For example,
instruments that measure sap flow can be installed on
trees representing the averaged sized tree and mean sap
flow multiplied times the number of trees in the stand
(i.e., a ‘‘mean-tree’’ approach). Considerable uncertainty in
stand-level estimates can accompany this approach where
sites are variable. Alternatively, relationships between
tree diameter, sapwood area, or basal area and sap flow at
the individual tree level can be applied to all trees. In both
approaches, repeated sampling is required to account for
seasonal variability.
The eddy flux method uses water vapor gradients
at fixed intervals above and below the canopy to
calculate evapotranspiration. The technique is based on
the assumption that water vapor flux is proportional
to the vertical gradient of water vapor between two
measurement points (averaged over several minutes).
Typically, measurements are conducted from towers
extending through the canopy. To be useful for estimating
transpiration of a particular stand, the stand must be
large enough to encompass most of the footprint measured
by the sensors. In many phytoremediation applications
conceived as of 2003, the stands are too small for an eddy
flux approach to be appropriate.
Modeling
The use of modeling provides a potentially powerful tool
for predicting current transpiration of native or planted
vegetation and for projecting future transpiration capacity
as a function of stand development. At the coarsest level
of forecasting, gross measures of plant water demand and
use can be derived from empirical estimates of poten-
tial evapotranspiration (4,38). These approaches usually
consider climate and soils to some extent, but do not con-
sider vegetation effects such as leaf area index, rooting
depth, or leaf-level physiological characteristics. Hence,
empirical approaches are useful for gross estimates of tran-
spiration, but have limited utility for evaluating actual
effects on the groundwater. At the other extreme, detailed
physiological models that link the soil–plant–atmosphere
continuum provide much more accurate estimates of tran-
spiration. Depending upon the structure, models may
also provide estimates of specific uptake locations within
the soil profile (3,39,40). Using detailed physiologically
 MEASURING AND MODELING TREE AND STAND LEVEL TRANSPIRATION
based models results in significantly greater data require-
ments. The most accurate application of these models
requires site-specific estimates of soils, climate, and phys-
iological characteristics of the major species on the site.
However, large-scale application of detailed models with
generalized parameters may provide estimates sufficiently
accurate to be used in evaluating phytoremediation appli-
cations.
MEASURING AND MODELING TRANSPIRATION: CASE
STUDY APPLICATIONS
Study Site Descriptions
Sap flow was measured at sites in Texas, Florida, and
Colorado as components of larger studies evaluating the
efficacy of using phytoremediation technology to clean up
shallow groundwater contaminants. The north–central
Texas study site was located, about 15 kilometers
west of Fort Worth. The climate of this area is
characterized as subhumid, with mild winters and hot,
humid summers. The average annual precipitation is
80 centimeters per year with most rainfall occurring
between May and October. Average annual temperature
is 18.6 ◦ C. Study plots were located on the U.S. Naval
Air Station, which adjoins U.S. Air Force Plant 4.
A plume containing trichloroethylene was detected in
the terrace alluvial aquifer in 1985. To demonstrate
phytoremediation potential, eastern cottonwood (Populus
deltoides Marsh.) trees were planted in two plantations
over the TCE plume. One plantation was planted
with vegetative cuttings (whips) and the other with 1-
year-old nursery grown seedlings. Each plantation was
approximately 80 by 20 meters and located perpendicular
to groundwater flow in the alluvial aquifer. Sap-flow
measurements were conducted using the heat-balance
method (collars) in the first and second year after
plantation establishment.
The eastern Florida site was located in the city of
Orlando. The climate of the area is humid, with mild
winters and hot, humid summers. The average annual
temperature is 22.6 ◦ C and the average annual rainfall is
123 centimeters. Native vegetation of interest was located
on the U.S. Naval Training Center. Trichloroethylene and
tetrachloroethylene, which originated from a dry-cleaning
facility that is no longer in operation, contaminate shallow
groundwater. The plume extends under a 2-hectare forest
and seepage wetland before reaching Lake Druid that
borders the forest. A dense and diverse mix of overstory
and understory species occur in the forest (density of
107 trees per hectare), with red bay [Persea borbonia
(L.) Spreng.], camphor [Cinnamomum camphora (L.) Nees
& Eberm.], slash pine and longleaf pine (Pinus spp.),
sweet bay (Magnolia virginiana L.), and live oak and
laurel oak (Quercus spp.) most abundant in the overstory.
The most abundant understory species are skunk vine
(Paederia foetida L.), saw palmetto [Serenoa repens
(Bartr.) Small], cinnamon fern (Osmunda cinnamomea
L.), and Christmas fern [Polystichum acrostichoides
(Michx.) Schott.].
The central Colorado site is located approximately 20
kilometers southwest of Denver. The climate of the area
737
is dry, with warm summers and cold winters. Annual
precipitation averages approximately 44 centimeters,
with 30 percent of this amount received in April and
May. The average annual temperature is 12 ◦ C. Study
plots were located on the U.S. Air Force Plant PJKS.
Trichloroethylene and dichloroethylene from a variety
of sources contaminate the site. Measurements were
conducted in two existing stands of natural vegetation:
cottonwood–willow (Populus spp.–Salix spp.) and Gambel
oak (Quercus gambelii Nutt.) The cottonwood–willow
(Populus spp.–Salix spp.) stand is restricted to riparian
areas (approximately 1 percent of the total land area of
the site), while the Gambel oak (Quercus gambelii) stand
is on more midslope locations (approximately 30 percent
of the total land area of the site).
Methods
The sampling approach and methods varied among the
three studies based on study objectives, species com-
position, and tree sizes. For the Texas study, sap
flow from saplings in the plantation was estimated
using sap-flow gauges (Dynamax Inc., Houston, TX) on
14 to 16 trees (divided equally among whips and 1-
year-old trees) in May, June, July, August, and Octo-
ber over a 2-year period. During each measurement
period, sap-flow measurements were taken every minute
for 2 to 3 consecutive days. Data presented in this
chapter represent averages of both plantations. In addi-
tion, sap flow was measured on nine larger native
trees growing near the plantations using thermal dis-
sipation probes (Dynamax, Inc., Houston, TX). Species
sampled were: eastern cottonwood (Populus deltoides
Marsh.), American elm (Ulmus americana L.), black wil-
low (Salix nigra Marsh.), sugarberry [or large hackberry,
(Celtis laevigata Willd.)], Eastern red cedar (Junipe-
rus virginiana L.), and mesquite (Prosopis pubescens
Benth.). At the end of sampling, increment cores were
taken from the nine large trees for determining sap-
wood area.
For the Orlando study, sap flow was estimated
using thermal dissipation probes installed on nine trees
representative of major canopy species. Species sampled
were: slash pine (Pinus elliottii Engelm.), longleaf pine
(Pinus palustris Mill.), live oak (Quercus virginiana
Mill.), laurel oak (Quercus hemisphaerica Bartram ex.
Willd.), sweet bay (Magnolia virginiana L.), and camphor
[Cinnamomum camphora (L.) Nees & Eberm.]. Two probe
sets were installed into the sapwood on the north
and south sides of sample trees, and sampling was
conducted in November, March, and July for 2 to 3
consecutive days over a 1-year period. At the end of
sampling, increment cores were taken and sapwood area
determined.
For the Colorado study, sap flow was estimated using
thermal dissipation probes on eight trees representing
three species: eastern cottonwood (Populus deltoides
Marsh.), narrow-leaf cottonwood (Populus angustifolia
James.), and Gambel oak (Quercus gambelii Nutt.). Two
probe sets were installed into the sapwood on the north and
south side of sample trees and sampling was conducted in
May, July, and September over a 1-year-period. At the end
738 MEASURING AND MODELING TREE AND STAND LEVEL TRANSPIRATION

of sampling, increment cores were collected and sapwood Transpiration Estimates

area determined.
For all three studies, data were summarized to provide
average hourly sap flow rates (kilograms per hour) or
daily totals (kilograms per day). In addition, climate
was measured at all three studies with climate stations
located on-site. Measurements included: hourly rainfall
(centimeters), wind speed (meters per second), solar
radiation (Watts per square meter), temperature ( ◦ C), and
relative humidity (percentage). Relative humidity and air
temperature were used to calculate vapor pressure deficit
(megapascals).
For the Texas plantation site, we parameterized and
applied a mechanistic model of sap flow (PROSPER) and
compared the results to sap flow measurements. Evapo-
transpiration at the Texas site was simulated because data
were available to parameterize the model (7). The PROS-
PER model has been described in detail elsewhere (39,41),
so only a general description is provided here. The
PROSPER code is a phenomenological, one-dimensional
model that links the atmosphere, vegetation, and soils.
Plant and soil characteristics are combined into a sin-
gle evapotranspiration surface that is characterized by a
resistance to water vapor loss. This resistance is anal-
ogous to the relationship between stomatal resistance
and water potential of the leaves and is a function of
the water potential of the evapotranspiration surface.
Evapotranspiration is predicted by a combined energy bal-
Maximum transpiration rates for the study sites indicate
large variation in transpiration potentials among sites
(Table 1). On a per tree basis, rates ranged from 8 to 120
kilograms per tree per day. Much of this variation was
related to differences in tree size that reflects differences
in leaf area and sapwood area. For example, when pooling
the data across the sites, a significant proportion of the
variation in transpiration rates among and within sites
can be explained by tree diameter (Fig. 3). Larger trees
typically have greater sapwood volume resulting in more
water transporting vessels (angiosperms) and tracheids
(conifers) for sap-flow movement in the stems. Because
leaf area is also related to sapwood area, larger trees will
typically have greater leaf area index; and hence, greater
surface area for transpiration.
When sap-flow rates are adjusted based on sapwood
area (i.e., kilograms per day per square meter of
sapwood), the variation in transpiration reflects species
related differences in physiology (leaf, stem, and root),
leaf area to sapwood area ratios, and site-dependent
factors such as soil water availability and climate
driving variables. Because species composition varies
among sites and physiological and physical factors
influence transpiration simultaneously, these studies
Mean sap flow
350
Sap flow (kilograms per tree per day)

ance–aerodynamic method [Penman-Monteith equation Colorado

modified as described in (42)] that is a function of the 300 Texas
surface resistance to vapor loss described previously. The Florida
PROSPER model uses electrical network equations (41) to 250
balance water allocation among vegetation and soil hori-
200
zons. The flow of water within and between soil and plant is Y = −88.00 + 4.41(X )
a function of soil hydraulic conductivity, soil water poten- 150 r 2 = 0.73; p < 0.05
tial, root characteristics in each soil layer, and surface
water potential. The PROSPER model predicts evapo- 100
transpiration, transpiration, and soil water distribution
50
between soil layers daily, but monthly data are most
accurate. The PROSPER model requires the following 0
climatic data: solar radiation, precipitation, wind speed, 0 10 20 30 40 50 60 70 80
air temperature, and vapor pressure. Initial model param- Tree diameter (centimeters)
eters include surface resistance to vapor loss, leaf area Figure 3. Total daily sap flows (Y) during peak transpiration
index, root distribution and surface area, soil moisture periods (midsummer) versus tree diameter (X) across a range of
release, and several other parameters listed in Goldstein species and site. Note that r2 is the correlation coefficient and p
et al. (41). is probability.

Table 1. Midsummer Peak Sap-Flow Rates Averaged Across Species and Measurement Days
Sap-Flow Rates

Sapwood Kilograms
Area (Square Per Day Per Kilograms Liters Gallons
Centimeters Square Meter Per Tree Per Hectare Per Acre
Site Per Tree) of Sapwood Per Day Per Year Per Year

Texas
Plantation 30 2600 8 3 620 000 387 200
Native trees 820 1463 120 7 551 000 807 600
Colorado 234 1043 24 1 510 000 161 500
Florida 710 1535 109 6 859 000 733 600

cannot separate physiological and climatological effects;
to do so requires an evaluation of transpiration rates of
the same species and genotype in differing climatic and
soil water availability conditions. For example, species
sampled at the three sites represent a mixture of conifers,
and ring porous and diffuse porous hardwood species,
resulting in large differences in sapwood permeability
and specific conductivity among sites and among species
within sites (Fig. 4). In general, sap-flow velocity is lower
in conifers and diffuse-porous species because sap flow
moves through a number of annual rings, whereas water
moves through only one or two annual rings in ring-porous
species (11,43). Despite the limitations of the current
approach, some notable patterns emerge when evaluating
transpiration after adjusting for differences in sapwood
area. For example, the cottonwood (Populus deltoides)
plantation in Texas had the highest transpiration rate
per unit of sapwood area, followed by the Florida stand,
large trees in Texas, and the Colorado stand (Table 1).
The high transpiration rate for cottonwood (Populus
deltoides) in the plantation is a function of species
characteristics that promote high transpiration, high leaf
area per unit sapwood in the developing canopy, and
access to shallow groundwater. In contrast, transpiration
rates per unit of sapwood area were lowest in Colorado,
even though the site contained cottonwood (Populus
deltoides) and several of the measured trees occurred in
the riparian zone. The combination of species composition
and climate characteristics were not as conducive to high
sap-flow rates per unit sapwood area relative to the
other sites.
Using sap-flow techniques to predict actual stand
transpiration requires frequent sampling to account
for seasonal variation. Ideally, sap flow should be
measured continuously for the entire growing season on a
Sap flow (kilograms per squrare meter sapwood)
Sap flow per unit sapwood
2500
2000
1500
1000
500
0 measurements at every phytoremediation site may not be
Elm
Hackberry
Figure 4. Mean growing season sap-flow rate per unit sapwood
for six species in north–central Texas. Black willow (Salix nigra
March.), eastern cottonwood (Populus deltoides Marsh.), and
eastern red cedar (Juniperus virginiana, L.) are diffuse porous
species, while American elm (Ulmus americana L.), hackberry
(Celtis laevigata Willd.), and mesquite (Prosopis pubescens
Benth.) are ring porous.
MEASURING AND MODELING TREE AND STAND LEVEL TRANSPIRATION 739
large number of trees. Because this approach is often
impractical, an alternative is to measure sap flow at
shorter frequencies and calculate bounds or maximum
values as a tool to evaluate phytoremediation potentials.
Because the sampling frequency varied among the case
studies described here, we focused only on measurements
during the highest transpiration period (midsummer).
These estimates can be used as a ‘‘best-case scenario’’
approach—that is, if these rates occurred on the site,
would transpiration be sufficient to control the plume? To
estimate maximum potential transpiration at the stand
level (i.e., kilograms per hectare or gallons per acre),
we extrapolated the tree transpiration data (kilogram
per tree per day) assuming a 180 days transpiration
period and a stem density of 350 stems per hectare,
except for the plantation where actual tree density
was used (Table 1). We emphasize that these data
provide estimates of maximum transpiration capacity
under in situ climate conditions because the peak sap-
flow rates were used in the extrapolation and previous
studies have shown considerable seasonal variation in sap
flow (7). The 350 stems per hectare is representative of
a fully stocked stand under most forest conditions and
is consistent with full canopy closure and maximum leaf
area index.
When comparing results from the sites with mature
trees, the variation in maximum transpiration capacity is
considerable. The Texas site has a maximum transpiration
capacity of approximately 7.5 × 106 liters of water per
hectare per year (8 × 105 gallons of water per acre per
year) if the site was fully stocked with the sampled
species. By contrast, the Colorado site has a maximum
transpiration capacity of approximately 1.5 × 106 liters
per hectare per year (1.6 × 105 gallons per acre per year).
The plantation site in Texas currently has a maximum
transpiration capacity of approximately 3.7 × 106 liters
per hectare per year (4 × 105 gallons per acre per year).
However, we anticipate that transpiration will equal or
exceed the estimate from mature trees on the site (versus
7.5 × 106 liters per hectare per year or 8.0 × 105 gallons
per acre per year) once the canopy develops and achieves
the maximum leaf area.
Comparison of Measured versus Modeled Transpiration
A critical need for phytoremediation is the development
and application of a tool to provide species and site-based
estimates of transpiration. While a powerful tool for mea-
suring transpiration from vegetation already on-site or
quantifying transpiration of planted vegetation, sap-flow
Willow Cottonwood practical. One potential tool for application across sites is
Mesquite Cedar the development or application of models. In most cases,
Species however, models need to be calibrated or parameterized
for specific site and species conditions. To evaluate the
use of such a tool, we parameterized PROSPER for the
Texas plantation using intensive site (soils, climate, and
root distribution) and leaf-level measurements (stomatal
characteristics and leaf area index) [see (7)]. We compared
monthly transpiration estimates obtained with PROSPER
to transpiration estimated from sap-flow measurements
740 MEASURING AND MODELING TREE AND STAND LEVEL TRANSPIRATION
over a 2-year period (Fig. 5). Comparisons indicated gen-
erally good agreement between predicted and measured
values, except during the late summer that was coin-
cident with some of driest and hottest periods (August
and September 1998) in our study. During this period,
PROSPER predicted a considerable decline in transpi-
ration, while measured values showed an increase. We
attribute this discrepancy to an inability of PROSPER
to adequately simulate root uptake from shallow ground-
water during drought conditions, since the original for-
mulation of PROSPER was designed to only simulate
surface and soil water dynamics (39,41). The results of
this comparison are consistent with other studies that
have shown that PROSPER provides reasonable esti-
mates of either evapotranspiration or transpiration (3,40).
However, refinements in the subsurface water and ground-
water hydrology and subsequent availability to tree roots
might improve the predictive capability and usefulness as
a phytoremediation evaluation tool.
CONCLUSIONS AND RECOMMENDATIONS
The importance of transpiration to the success of phytore-
mediation applications suggests that accurate estimates
of current and potential transpiration should be a high
priority when considering this approach for site manage-
ment. Both measuring and modeling transpiration are
important. Assessments require a substantial sampling
commitment for direct measurements or parameteriz-
ing physiologically based models. This requires detailed
knowledge of local site conditions and physiological param-
eters for the major species. For screening assessments,
we recommend that published estimates be used to set
the bounds for maximum transpiration capacity based on
general climate and vegetation characteristics of the loca-
tion. If these general transpiration rates are great enough
to influence groundwater hydrology, then evaluations of
current, enhanced (e.g., manipulating the structure and
species composition of current vegetation), or new vege-
tation transpiration capacity should proceed. Technology
Prosper vs. sap flow
12
Transpiration (centimeters per month)
Predicted transpiration
10 Measured sap flow
8
6 ecosystems. Ecol. Bull. (Copenhagen) 43: 9–19.
4 tion of Pinus contorta sp. latifolia. Oecologia 14: 483–490.
2 effects on water relations of Pinus radiata. Can. J. Bot. 55:
0 11. Kozlowski, T.T., Kramer, P.J., and Pallardy, S.G. (1991). The
January April July October January April July October
1997 1998
Year and month
Figure 5. Comparison of measured sap flow and predicted
transpiration from PROSPER for cottonwood (Populus deltoides
Marsh.) plantations in north–central Texas.
and models exist to provide reasonable estimates and pre-
dictions of transpiration. However, the accuracy of the
estimates depends on the investment in accounting for
the spatial and temporal variation or in providing site
and species-specific estimates for physiologically based
transpiration models.
Acknowledgments
This work was supported in part by the Department of
Defense Environmental Security Technology Certification Pro-
gram Project E95BRAB004, Gregory J. Harvey, Program Man-
ager and in part by the Department of Defense, U.S. Navy
(SOUTH-NAVFACENGCOM), Charleston, South Carolina. We
thank Chris Sobek for her assistance in collecting sap-flow data,
summarizing data, and generating graphs. Several other mem-
bers of the Coweeta staff assisted with field sampling at various
times, including: Bob McCollum, Nathan Stover, Patsy Clinton,
Laura Phillips, and Josh Ward.
BIBLIOGRAPHY
1. Helvey, J.D. (1971). A summary of rainfall interception by
certain conifers of North America. In: Biological Effects of the
Hydrologic Cycle. E.J. Monke (Ed.). Proc. Third Int. Seminar
for Hydrology Professors, Dept. of Agricultural Engineering.
Purdue University, West Lafayette, Indiana, pp. 103–113.
2. Meyers, B.J. and Talsma, T. (1992). Site water balance and
tree water status in irrigated and fertilized stands of Pinus
radiata. Forest Ecol. Manag. 50: 17–42.
3. Vose, J.M. and Swank, W.T. (1992). Water balances. In:
Atmospheric Deposition and Forest Nutrient Cycling: A
Synthesis of the Integrated Forest Study. D.W. Johnson and
S.E. Lindberg (Eds.). Ecological Studies 91. Springer-Verlag,
New York, pp. 27–49.
4. Thornthwaite, C.W. (1948). An approach towards a rationale
classification of climate. Geogr. Rev. 38: 55–94.
5. Hinckley, T.M., Brooks, J.R., Cermak, J., Ceulemans, R.,
Kucera, J., Meinzer, F.C., and Roberts, D.A. (1994). Water
flux in a hybrid poplar stand. Tree Physiol. 14: 1005–1018.
6. Martin, T.A., Brown, K.J., Cermak, J., Ceulemans, R.,
Kucera, J., Meinzer, F.C., Rombold, J.S., Sprugel, D.G., and
Hinckley, T.M. (1997). Crown conductance and tree and stand
transpiration in second-growth Abies amabilis forest. Can. J.
Forest Res. 27: 797–808.
7. Vose, J.M., Swank, W.T., Harvey, G.J., Clinton, B.D., and
Sobek, C. (2000). Leaf water relations and sapflow in
eastern cottonwood (Populus deltoides Bartr.) trees planted
for phytoremediation of a groundwater pollutant. Int. J.
Phytorem. 2(1): 53–73.
8. Knight, D.H., Vose, J.M., Baldwin, V.C., Ewel, K.C., and
Grodzinska, K. (1994). Contrasting patterns in pine forest
9. Fahey, T.J. (1979). The effect of night frost on the transpira-
10. Kaufmann, M.R. (1977). Soil temperature and the drying
2413–2418.
Physiological Ecology of Woody Plants. Academic Press, New
York, p. 657.
12. Gates, D.M. (1980). Biophysical Ecology. Springer-Verlag,
New York, p. 611.
13. Dawson, T.E. and Ehleringer, J.R. (1991). Streamside trees
do not use stream water. Nature 350(6316): 335–337.
 WATER LOGGING: TOPOGRAPHIC AND AGRICULTURAL IMPACTS
14. Busch, D.E., Ingraham, N.L., and Smith, S.D. (1992). Water
uptake in woody riparian phreatophytes of the southwestern
United States: a stable isotope study. Ecol. Appl. 2(4):
450–459.
15. Robinson, T.W. (1970). Evapotranspiration by woody phreato-
phytes in the Humbolt River Valley near Winnemucca,
Nevada. U.S. Geological Survey Professional Paper 491–D,
U.S. Government Printing Office, Washington, DC.
16. Van Hylckama, T.E.A. (1980). Weather and evapotranspira-
tion studies in a saltcedar thicket in Arizona. U.S. Geological
Survey Professional Paper 491–494, U.S. Government Print-
ing Office, Washington, DC.
17. Allen, S.J., Hall, R.L., and Rossier, P.T. (1999) Transpiration
by two poplar varieties grown as coppice for biomass
production. Tree Physiol. 19: 493–501.
18. Gholz, H.L. (1982). Environmental limits on aboveground net
primary production, leaf area and biomass in vegetation zones
of the Pacific Northwest. Ecology 53: 469–481.
19. Long, J.N. and Smith, F.W. (1990). Determinants of stem-
wood production in Pinus contorta var. latifolia forests: the
influence of site quality and stand structure. J. Appl. Ecol.
27: 847–856.
20. Vose, J.M. and Allen, H.L. (1988). Leaf area, stemwood
growth, and nutrition relationships in loblolly pine. Forest
Sci. 34: 546–563.
21. Colbert, S.R., Jokela, E.J., and Neary, D.G. (1990). Effects of
annual fertilization and sustained weed control on dry matter
partitioning, leaf area, and growth efficiency of juvenile
loblolly and slash pine. Forest Sci. 36: 995–1014.
22. Gholz, H.L. (1986). Canopy development and dynamics in
relation to primary production. In: Crown and Canopy
Structure in Relation to Primary Production. T. Fujimori and
D. Whitehead (Eds.). Forest and Forest Products Research
Institute, Ibaraki, Japan. pp. 224–242.
23. Cropper, W.P., Jr. and Gholz, H.L. (1994). Evaluating poten-
tial response mechanisms of a forest stand to fertilization and
night temperature: a case study using Pinus elliottii. Ecol.
Bull. (Copenhagen). 43: 165–160.
24. Douglass, J.E. and Swank, W.T. (1975). Effects of man-
agement practices on water quality and quantity: Coweeta
Hydrologic Laboratory, North Carolina. USDA Forest Service
Gen. Tech. Rep. SE-13, Asheville, North Carolina.
25. Swank, W.T., Vose, J.M., and Elliott, K.J. (2000). Long-term
hydrologic and water quality responses following commercial
clearcutting of mixed hardwoods on a southern Appalachian
catchment. Forest Ecol. Manag. 5267: 1–16.
26. Heilman, P. (1981). Root penetration of Douglas-fir seedlings
into compacted soil. Forest Sci. 27: 660–666.
27. Schulze, E.-D. (1991). Water and nutrient interactions with
plant water status. In: Responses of Plants to Multiple
Stresses. H.A. Mooney, W.E. Winner, and E.J. Pell (Eds.).
Academic Press, New York, pp. 89–101.
28. Waring, R.H. and Schlesinger, W.H. (1985). Forest Ecosys-
tems: Concepts and Management. Academic Press, New York,
p. 340.
29. Zhang, H., Simmonds, L.P., Morison, J.I.L., and Payne,
D. (1997). Estimation of transpiration by single trees:
comparison of sap flow measurements with a combination
equation. Agric. For. Meteorol. 87: 155–169.
30. Heilman, P.E., Hinckley, T.M., Roberts, D.A., and Ceule-
mans, R. (1996). Production physiology. In: Biology of Pop-
ulus and Its Implications for Management and Conserva-
tion. R.F. Stettler, H.D. Bradshaw, Jr., P.E. Hielman, and
741
T.M. Hinckley (Eds.). NRC Research Press, Ottawa, Canada,
pp. 459–489.
31. Martin, T.A., Hinckley, T.M., Meinzer, F.C., and Sprugel,
D.G. (1999). Boundary layer conductance, leaf temperature
and transpiration in Abies amabilis branches. Tree Physiol.
19: 435–443.
32. Steinberg, S.L., van Bavel, C.H.M., and McFarlan, M.J.
(1989). A gauge to measure mass of flow of sap in stems and
trunks of woody plants. J. Am. Soc. Hort. Sci. 114: 466–472.
33. Schulze, E.-D., Cermak, J., Matyessek, R., Penka, M.,
Zimmerman, R., Vasicek, F., Gries, W., and Kucera, J.
(1985). Canopy transpiration and water fluxes in the xylem
of the trunk of Larix and Picea—a comparison of xylem
flow, prorometer, and cuvette measurements. Oecologia 66:
475–483.
34. Phillips, N., Nagchaudhuri, A., Oren, R., and Katul, G. (1997).
Time constant for water transport in loblolly pine trees
estimated from time series of evaporative demand and stem
sapflow. Trees 11: 412–419.
35. Baker, J.M. and van Bavel, C.H.M. (1987). Measurement of
mass flow of water in stems of herbaceous plants. Plant Cell
Environ. 10: 777–782.
36. Granier, A. (1987). Evaluation of transpiration in a Douglas
fir stand by means of sap flow measurements. Tree Physiol.
3: 309–320.
37. Swanson, R.H. (1962). An instrument for detecting sap move-
ment in woody plants. U.S. Department of Agriculture Forest
Service Rocky Mountain Forest and Range Experiment. Sta-
tion Paper. 68.
38. Mongaran, C. (1973). Economic feasibility of irrigating
southern pines. Water Resources Res. 9: 1485–1495.
39. Huff, D.W. and Swank, W.T. (1985). Modeling changes in
forest evapotranspiration. In: Hydrological Forecasting. M.G.
Anderson and T.P. Burt (Eds.). John Wiley & Sons, New York,
pp. 121–151.
40. Vose, J.M. and Swank, W.T. (1994). Effects of long-term
drought on the hydrology and growth of a white pine
plantation in the southern Appalachians. Forest Ecol. Manag.
64: 25–39.
41. Goldstein, R.A., Mankin, J.B., and Luxmoore, R.J. (1974).
Documentation of PROSPER: a model of atmosphere-plant-
soil water flow. EDFB/IBP-73/9, U.S. Department of Energy
Oak Ridge National Laboratory, Oak Ridge, Tennessee.
42. Swift, L.W., Jr., Swank, W.T., Mankin, J.B. Luxmoore, R.J.
and Golstein, R.A. (1975). Simulation of evapotranspiration
and drainage from mature and clear-cut forest and young
pine plantations. Water Resources Res. 11(5): 667–673.
43. Kramer, P.J. and Kozlowski, T.T. (1979). Physiology of Woody
Plants. Academic Press, New York, p. 811.
WATER LOGGING: TOPOGRAPHIC AND
AGRICULTURAL IMPACTS
CHANDRA MADRAMOOTOO
Macdonald Campus of McGill
University
Ste-Anne de Bellevue, Quebec,
Canada
Water logging occurs on poorly drained soils. Under
periods of heavy precipitation or snowmelt, water
percolates very slowly, and soils become wet in short
742 WEED CONTROL STRATEGIES
periods of time. Some soils have inherently low ability to
infiltrate and hydraulic conductivity, especially heavier
clay soils. Soil compaction because of heavy farm
equipment further reduces bulk density, which in turn
reduces the ability to infiltrate soil. The result is that
the water table accretes and may eventually rise to the
land surface. High-intensity rainfalls also lead to ponding
of water on the soil surface. These conditions of surface
ponding, a saturated soil profile, and gradual rise of the
water table to the soil surface all lead to water logging.
One very specific hydrogeologic condition that contributes
to water logging is groundwater artesian pressure.
Land topographic features have an influence on
water logging. On very flat lands, with low hydraulic
gradients to rivers and watercourses, surface runoff is not
easily drained. Water is also easily trapped in surface
depressions on undulating lands. Low-lying coastal and
riverine areas are very susceptible to flooding in the
wet rainy periods. Towns, villages, and agricultural
lands suffer from backwater effects. Infrastructural works
sometimes exacerbates water logging. For example, during
heavy rainfalls, water becomes trapped behind road and
railway embankments. The situation could be extremely
damaging under monsoon conditions in Asia, where
widespread flooding occurs with loss of life and property.
Water logging also occurs in the arid and semiarid
regions on poorly drained soils, because of excess irrigation
water applications and canal seepage. It is estimated that
about 25–30 million hectares of irrigated land suffer from
water logging.
Globally, some areas of permanently wet or water
logged regions exist, which are known ecologically as
wetlands, and they serve very important environmental
and hydrologic functions. Wetlands provide habitats for
waterfowl, flora, fish, and animals. They also replenish
groundwater, provide flood control, and filter pollutants.
Organic soil deposits, or peat bogs, are also characterized
by naturally high water tables.
It is estimated that over 200 million hectares of
agricultural land worldwide suffer from poor drainage.
Crops need an optimum soil-air environment in the root
zone in which to thrive. Under high water table conditions
or poor drainage, crop growth is retarded because of a
lack of oxygen in the root zone. Root and shoot growth
are affected by high water table conditions, as are
other plant physiological conditions. For example, stem
elongation is slowed, senescence and abscission of older
leaves occur, rapid stomatal closure exists, and leaves
begin to yellow. Crop failure is inevitable under prolonged
flooding. Anaerobic conditions lead to reduced oxygen, and
an accumulation of carbon dioxide, ethylene, other toxic
substances, and byproducts in the soil profile, all of which
are harmful to plant growth.
Some plants have the ability to withstand high water
tables better than others. For example, rice is grown
under flooded conditions. Some berry crops, e.g., highbush
blueberries and cranberries grow under saturated soil
conditions. Native vegetation and trees can be found in
swamps, marshlands, and peat bogs. Crop tolerance to
high water tables not only depends on crop type, but
also on stage of crop growth. For example, most crops are
likely to be killed by flooding during germination and early
vegetative growth. As plants mature, they are slightly
more tolerant to high water tables. However, vegetable
crops are very sensitive to high water tables during their
entire growth period. Deeper rooted crops, such as grasses,
sugarcane, corn, shrubs, and trees, are somewhat hardier.
Apart from rice and some berry crops, it is impossible
to undertake large-scale commercial plant and animal
production without removal of excess soil water. Drainage
is therefore essential for lowering the water table on poorly
drained soils. The most common form of drainage is surface
drainage. Open ditches are dug in fields to evacuate excess
water. Lands are also shaped either by forming bedded
fields or by a ridge and furrow system. In this way, crops
can be grown on the raised beds or ridges. Other forms
of land shaping include land leveling and smoothing, or
grading the fields with a slope toward open drains. Surface
depressions in fields can also be filled in by land leveling,
and this will reduce ponded water on the land surface. Tile
drainage, or subsurface pipe drainage, is another form of
drainage for improving the productivity of poorly drained
agricultural lands.
Tile drainage or tubewells are also used to lower the
water table in irrigated lands. If seepage from irrigation
canals is determined to be a major contributor to the water
logging of adjacent lands, the canals can be lined, or buried
pipe interceptor drains are installed adjacent to the canal.
Relief wells are normally installed in lands affected by
artesian uplift.
In the lowing coastal and riverine areas, flood
control structures are built, which include levees, dykes,
walls, and berms. Sluice gates or pumping stations
are constructed to evacuate water from behind these
structures to either the sea or rivers.
WEED CONTROL STRATEGIES
J.R. NEWMAN
IACR-Centre for Aquatic Plant
Management
Reading Berkshire, United
Kingdom
The control of nuisance vegetation in watercourses is
often necessary to ensure adequate flood defense for
surrounding land, to provide facilities for recreational
sporting activity, to provide better angling environments,
to aid navigation, on public health grounds, for industrial
uses, and to maintain adequate water supply and quality.
In recent years, weed control for conservation has also
become necessary to protect native habitats from invasion
by alien, introduced, or nonnative species.
Aquatic weeds are different from terrestrial weeds
in that they are usually a normal constituent of the
aquatic ecosystem. Aquatic plants become weeds when
they grow to excess and become a nuisance by blocking
channels, reducing flow, or preventing other normal uses
of the waterbody. Increased urbanization has resulted
in an increase in the number of nonnative species now
considered aquatic weeds, due primarily to escape of plants

from domestic situations. The increased recreational
pressure on enclosed inland waterways has also resulted
in greater demands for aquatic plant management.
During the last 50 years, there has been a dramatic
change in the type of aquatic weed problem. Emergent
weeds, such as reeds and rushes, and floating leaved
weeds, such as waterlilies, were the major problem, but
eutrophication since the Second World War, as a result of
intensification of agriculture, increased urbanization, and
consequent sewage treatment requirements, has resulted
in a shift to problems caused by filamentous and blue-green
(cyanobacterial) algal blooms.
TYPES OF AQUATIC WEEDS
Although there are a large number of individual aquatic
weed species, for the purposes of control, they can be
divided into four major groups.
Emergent Plants
Emergent plants are rooted in the sediment at the margins
of watercourses. They have stems and leaves that protrude
above the water’s surface. Growth is often limited by
increasing water depth, making this type of weed common
along the edges of water. Some common characteristics of
these plants are long narrow leaves, and heights between 1
and 3 m. Species include Phragmites communis (common
reed), Typha latifolia (bulrush), Schoenoplectus lacustris
(common club rush), Glyceria maxima (reed sweet grass)
and Sparganium erectum (common bur reed).
Smaller emergent and bankside plants, except grasses,
tend to be broad-leaved weeds. These include Ror-
ippa nasturtium-aquaticum (watercress) Apium nodiflo-
rum (fool’s watercress), Berula erecta (water parsnip),
and others.
Other plants in this group are predominantly sub-
merged but may form different types of leaves when they
become emergent due to lowered water levels or grow
above the water surface. These include some Ranunculus
spp. (water crowfoots) and Hippuris vulgaris (mare’s tail).
Plants that grow on banks at or above the water
line should be included in this group. Common weeds in
this class include Lythrum salicaria (purple loosestrife),
Epilobium hirsutum (greater willow herb) and Phalaris
arundinacea (reed canary grass).
A number of riparian weeds associated with water-
courses can be included in this group. The dispersal of
these species is furthered by flowing water. The alien
invasive species Fallopia japonica (Japanese knotweed),
Heracleum mantegazzianum (giant hogweed) and Impa-
tiens glandulifera (Himalayan balsam) are included here.
Floating Plants
Free-Floating Plants. This group includes many of the
worst types of aquatic weeds, such as Azolla spp. (water
fern), Eichhornia crassipes (water hyacinth), Lemna spp.
(duckweeds), Pistia stratiotes (water lettuce) and Salvinia
molesta (giant water fern), all of which have been found
in Britain. They are characterized primarily by very rapid
vegetative reproduction and spread.
WEED CONTROL STRATEGIES 743
Floating-Leaved Plants. This type of plant is character-
ized by rooting in the sediment and forming floating leaves.
They may be rooted at the margins and form floating mats
over the surface of the water, such as Hydrocotyle ranun-
culoides (floating pennywort). This species has leaves held
above the water surface. Plants rooted in deep water that
have floating leaves held flat on the surface include Pota-
mogeton natans (broad-leaved pondweed), Nuphar lutea
(yellow water lily) and Nymphaea alba (white water lily).
Other plants in this group are predominantly sub-
merged, but may form different types of leaves when they
reach the water surface, such as the rosette type of leaf
produced by Callitriche spp. (water starworts).
Submerged Plants
This group includes all submerged plants except algae.
It can be divided into two sub-groups, those rooted
in the sediment, such as Elodea spp. (waterweeds),
Myriophyllum spicatum (spiked water milfoil), and
Potamogeton pectinatus (fennel pondweed), or those
free-floating below the water, such as Ceratophyllum
demersum (rigid hornwort) and Lemna trisulca (ivy-
leaved duckweed).
Algae
Algae can be divided into two types for the purpose
of control, macrophytic types, including filamentous
algae and charophytes, and unicellular types, including
pea-soup algae and cyanobacteria or blue-green algae.
Nuisance algae tend to grow best in nutrient-rich still or
slow flowing water.
THE BIOLOGY AND ECOLOGY OF AQUATIC WEEDS
Seasonal Growth and Dispersal
Aquatic plants are usually a problem only during the
summer months. Various forms of dieback or washout
during autumn and winter ensure that very little aquatic
vegetation is left to overwinter. However, nuisance aquatic
weeds use several strategies to survive winter. These
include seed production by annual riparian weeds and
as an insurance policy by perennial submerged aquatic
macrophytes. Others produce turions (specialized leaf
buds) tolerant of anoxic conditions, for example, P.
pectinatus, fennel pondweed. The majority of emergent
macrophytes have dormant roots and rhizomes. Algae use
spore production and Azolla and others can overwinter as
intact plants, for example, Callitriche spp. and Crassula
helmsii.
Aquatic weed growth usually starts when mean diel
water temperature rises above 6◦ C; this is normally a
week to a few weeks behind terrestrial plant growth.
However, growth is rapid once started and stem
elongation in Ranunculus spp. can produce 6 m of
growth before flowering. Shortening day length usually
initiates flowering so the majority of aquatic plants flower
in or after the last week in June in the temperate
Northern Hemisphere.
Stem fragmentation and vegetative reproduction is the
most common form of dispersal within systems, and seed
744 WEED CONTROL STRATEGIES
production is not considered important for maintaining
weed populations. However, establishment of new weed
populations is most likely to derive from establishment of
seeds. Birds, animals, machinery, boats, and man can also
spread viable fragments of weeds to new sites.
Characteristics of Nuisance Plants
Growth Rate. Species that have extremely fast growth
rates tend to become greater problems than other species.
For example, duckweed and Azolla can double their
biomass in less than 4 days in good conditions.
Morphology. Species that produce dense masses of
vegetation impede the flow of water. Others have rigid
stems that may impede boat traffic, interfere with access
to the water, and trap detritus, forming temporary dams
that may cause flooding.
Dispersal Strategies. Plants that produce new plants
by fragmentation are also very effective aquatic weeds.
Often, mechanical control is responsible for increases in
the distribution of these species in watercourses.
Toxicity and Taint. Some plants are toxic to animals,
and others taint potable waters. Blue-green algae produce
toxins that can kill fish, and other animals and also impart
an unpleasant taste to drinking water.
CONTROL METHODS
Factors Affecting Choice
A number of factors influence the choice of weed
management technique, including the cost of control.
These factors are related to the uses of the water and
the specific landscape situation of the waterbody. Often,
a compromise between weed control efficacy and other
considerations must be made.
Type of Control. In some cases, it is not desirable to
attempt complete eradication of the weed population, and
localized control is preferable. Selective control can be
either by type of species or by, for example, clearing
swims for anglers. Total control may be required where
flood defense is an issue or where nonindigenous species
are present. Temporary alleviation of a weed problem is
sometimes possible by frequent mechanical control during
the season.
Use of the Water. Certain weed control practices may
not be compatible with the use of the water. The most
obvious of these is when water is abstracted for drinking
water supplies downstream of an intended herbicide
application site. The limit of 0.1 µg L−1 for individual
pesticides set by the EU Drinking Water Directive imposes
strict requirements on the distance between the site of
application and the abstraction point. Dilution and flow
rate affect this distance.
Human and Livestock Safety. This includes direct
toxicity of herbicides to operators and water users,
increases in turbidity due to stirring up of sediments,
and unpleasant odors from rotting vegetation or herbicide
residues in water. The safety of farm and other animals
can be compromised by the direct toxicity of the herbicide,
or, as is more likely, by increasing the palatability of
treated poisonous plants.
Irrigation. The hazards to adjacent crops should be
considered through irrigation with water containing
pesticides or spray drift at the time of application and
the spread of weeds such as P. australis into cropland
when dredged spoil is deposited onto land adjacent to
watercourses.
Industry. Weeds blocking intake sluices or the presence
of pesticides in water may affect industrial processes
requiring water.
Environmental Impacts. In general, the more effective a
weed clearance operation, the greater the environmental
impact. Deoxygenation of water, which affects fish,
can occur when large volumes of cut or herbicide-
treated weed are allowed to rot in situ. The number of
invertebrates removed by mechanical control is estimated
at approximately 1 million per m3 of cut weed, and the
number of vertebrates, mainly small fish, is estimated at
about 40 per m3 of weed removed. Care should be taken to
avoid removing marginal vegetation during bird nesting
times. Other indirect effects include destabilization of
banks by removal of plant cover or by deposition of thick
layers of cut vegetation on sloping banks.
MANAGEMENT OPTIONS
Mechanical Control
The following points should be considered when using
mechanical weed control. Rafts of cut weed can drift
downstream and block sluices, pumps, weirs, and other
water control structures. Deoxygenation can occur if weed
is left to rot in water, posing a risk to fish and invertebrate
life. Rapid regrowth of cut weeds often necessitates a late
season cut. In practice, the cost of mechanical control
limits the frequency of the operation, thus a compromise
between weed control and cost is usually necessary.
Cutting. Aquatic weeds can be cut manually or by using
machines; the choice of technique depends on the scale of
the problem to be tackled. Cutting by hand and raking cut
material onto banks is not used widely now, although the
majority of small drainage ditches used to be managed in
this way. Now a wide range of mechanical cutting boats
and weed bucket attachments for bank-based machines
is available, but the principles of cutting and removing
the cut weed remain unchanged. In general, the deeper
the cut, the longer the weed control will last. Timing of
the cut is critical. Good control can be achieved by regular
cutting throughout the growing season, but this may not be
possible for operational reasons. Cutting submerged weeds
before the end of June will require a further cut toward the
end of the season, and regrowth during summer tends to

be faster than initial growth rates. The timing of cutting
emergent and bankside weeds is less critical, but care
should be taken to avoid bird nesting seasons.
Weed boats are used to cut submerged weeds in
relatively deep and wide watercourses where large areas of
weed require cutting. They usually have a front mounted
reciprocating cutting bar which can cut to a depth of about
1.5 m. The work rate is estimated at approximately 1 km
per day on a 10-m wide river. Cut weed must be removed
from the water, either by a separate boat or by the action
of a front rake on the cutting boat after cutting.
Bank-mounted equipment includes draglines, flail
mowers, and dredging equipment. The work rate for
cutting buckets and weed rakes is, on average, about
500 m day−1 on wide rivers.
Harvesting. The use of aquatic weed harvesters com-
bines the cutting and collecting operation, therefore reduc-
ing time and increasing efficiency. They can and/or collect
all types of weed, including filamentous algae and duck-
weeds and can deposit them at a collection point on the
bank. The work rates for this type of equipment vary
considerably with the density of weed and the distance
traveled to offload the collected weed.
Uses for Cut Weed. There are a limited number of uses
for cut weed; these include composting, reed for thatching,
mulches, soil conditioners, and cattle forage. The material
has low value, and processing should be done locally to
minimize costs.
Chemical Control
Herbicides can be used to control a wide range of aquatic
plants. They may be applied to emergent or surface-
floating weeds by a foliar spray in much the same way
as recommended for land plants. Submerged weeds and
algae are treated by adding the herbicide to the water to
build up an effective concentration in the water. This
method will also control some floating and emergent
species. Chemicals currently used to control aquatic plants
are listed in Table 1.
Table 1. Herbicides Suitable for Controlling Major Weed Groups
Asulam
Copper
Trees and shrubs on banks
Bracken and docks ž Ž
Broad-leaved weeds
Grasses
Reeds and sedges
Floating-leaved plants
Free-floating plants
Submerged plants
Submerged plants (flowing water)
Algae ž ž
ž suitable for control; Ž short-term control or some species within groups not susceptible
WEED CONTROL STRATEGIES 745
Consideration should be given to the effects of herbicide
treatment on the function of the waterbody. This not only
involves the direct toxicity of the chemical but also indirect
effects caused by decomposition of dead weeds. Retaining a
proportion of the weed growth is recommended to maintain
a habitat for invertebrates, fish, and other wildlife and to
stabilize the ecosystem. Vegetation left in situ will take
up nutrients and help to prevent the growth of algae.
To minimize the effects of decomposing weed, herbicide
labels have recommendations on treatment intervals that
should be observed. It is normal to treat only 25% of
an infested waterbody at one time, separated by the
application interval. Consideration should be given to
the irrigation interval when using herbicides in water.
This is the time between application and the time when
water is safe to use for irrigating of crops or watering
livestock.
Application Methods
Application to Foliage. Recommendations for the correct
dose are made in the form of weight of product or active
ingredient per hectare. Herbicides are usually applied
by knapsack or boat-mounted sprayer equipment. Using
high-volume and low-pressure equipment to deliver a
coarse spray with a minimum of small droplets is advisable
to reduce the risk of spray drift. Localized or selective
control can be achieved by the appropriate choice of
herbicides and by spraying only those plants required.
Water velocity and quality do not affect the treatment, but
to avoid a buildup of chemical in the water, it is normal to
apply herbicides in an upstream direction.
Application to Water. Recommendations usually refer
to the theoretical concentration of active ingredient that
would be achieved when the chemical has been evenly
distributed throughout the water body but before any
adsorption or degradation has occurred. This is usually
expressed as parts per million (ppm or mg l− 1 or g m−3 ).
Some formulations may be applied on a rate per surface
area basis because they sink onto the weeds or mud.
Maleic Hydrazide
Diquat Alginate
2,4-D amine
Dichlobenil
Glyphosate
Terbutryn
Endothall
Fluridone
Diquat
ž
ž
ž Ž Ž Ž ž
Ž Ž ž ž ž
Ž ž ž ž
ž Ž ž ž Ž ž Ž
Ž ž Ž ž ž
Ž ž ž ž ž ž ž
ž Ž
ž Ž ž
746 WEED CONTROL STRATEGIES

The methods of application depend on the formulation.
Herbicides should be spread as evenly as possible over
the surface of the water. Granular formulations can
be spread by hand using a suitable container, by a
mechanical spreader, or by using an air-assisted blower.
Liquid formulations should be diluted and applied by
subsurface injection. This is usually achieved by trailing
nozzles below the water as close to the top of the
weed bed as possible, without disturbing the bottom
sediment. Viscous gel formulations should be applied with
specialized equipment.
Timing. Emergent and floating leaved plants should be
treated from midsummer, when the leaves have formed
fully. Annual weeds should be treated before flowers have
set to prevent production of seeds.
Treatment of submerged weeds and algae is normally
recommended in spring and early summer when weeds
are actively growing. Treatment of floating-leaved plants
is usually more successful before the floating leaves have
formed. Later in the season, when large biomasses of weed
have developed and particularly when the water is warm,
there is a severe risk of deoxygenation of the water, after
control, caused by decomposition of the dead weed. This
effect is particularly acute when using terbutryn, as the
mode of action blocks photosynthesis very quickly, but
respiration continues.
An indirect hazard to fish may arise through possible
effects on fauna that provide food for fish. These may be
due to direct toxicity to invertebrates or more likely to
loss of habitat after weed control operations. However,
loss of habitat can lead to increased predation by fish.
These effects are temporary and should be set against the
disadvantages of dominant weed species reducing habitat
structure for all aquatic fauna.
Biological Control
The decline in public acceptability of herbicide use in
water, coupled with an increase in the costs of mechanical
control has led to the development of several biological
control agents for aquatic weeds.
Livestock. Horses, cows, goats, and sheep can be used to
graze marginal vegetation. Care should be taken to avoid
excessive poaching of the banks (although some poached
banks are a valuable habitat), and fencing is required to
keep animals from straying.
Chinese Grass Carp and Other Fish. Herbivorous fish are
most common in Asia and South America. One species, the
Chinese grass carp, Ctenopharyngodon idella, is available
for weed control in enclosed situations. Triploid grass carp
are used in the United State to ensure that the fish cannot
breed. Licenses are usually required to introduce of these
fish. Stocking densities vary depending on the type and
severity of the weed problem, but usually a density of
between 75 and 150 kg ha−1 will achieve adequate weed
control. Bottom feeding fish stir up sediment and create
turbid conditions in which submerged macrophytes cannot
grow. These fish include primarily carp and bream.
Waterfowl. Waterfowl graze on some species of floating
and submerged weed. When present in large numbers
(e.g., swans on Ranunculus), they can cause significant
losses of aquatic vegetation, which may not be desirable.
Pathogens and Insects. Classical biological control is not
widely practiced in Europe, and not at all on aquatic
and riparian weeds. Biological control agents survive only
on their host species, meaning they cannot spread to
ornamentals or related crop species. Most native plant
species have a number of native host-specific pathogens
or insect predators. There are effective biological control
agents for many aquatic weed species (Table 2) established
in other parts of the world, but none is used in Europe.
Biological control has been successful in about 45% of
the attempts made on aquatic weeds. It is essentially a
management tool designed to maintain the level of a pest

Table 2. Established Biological Control Agents for Aquatic Weeds

Weed Species Biological Control Agent Type Region

Alternanthera philoxeroides Agasicles hygrophila Flea beetle America

Azolla filiculoides Stenopelmus rufinasus Weevil S. Africa
Eichhornia crassipes Neochetina eichhorniae Weevil America Africa, India
Neochetina bruchi Weevil America, Africa, India
Cercospora piapori Fungus Africa
Altenaria eichhorniae Fungus Africa
Hydrilla verticillata Hydrilla pakistanae Insect America
Hydrelli sarahae Insect
Hydrellia pakistanae Insect
Hydrellia balciunasi Insect
Mycoleptodiscus terrestris Fungus
Plectosporium tabacinum Fungus
Cricotopus lebetis Midge
Myriophyllum aquaticum Lysathia sp Beetle S. Africa
Myriophyllum spicatum Bagous subvittatus Insect N. America
Bagous myriophylli Insect
Phytobius Insect
Salvinia molesta Cyrtobagous salviniae Weevil Africa
Neohydronomus affinis Weevil

at a tolerable level in the community without necessarily
eradicating the weed species. It is a useful tool to employ
in combination with other techniques such as mechanical
or chemical control.
Environmental Control
Environmental control is the technique of changing the
environment, either temporarily or permanently, to reduce
the suitability of the habitat for the target weed. This
technique tends to shift the weed species to other plants,
which may or may not be desirable depending on the
ultimate use of the waterbody in question.
Flow. Aquatic plants are divided into riverine or
lacustrine species. Although not mutually exclusive,
plants that tend to favor slow flowing environments do not
grow in faster flowing water and vice versa. Alterations
to flow can have marked effects on the composition of the
aquatic plant community.
Water Chemistry. Reduction of the quantity of major
plant nutrients entering water tends to reduce the biomass
development of weed species but not necessarily the
species composition of the weed community. Diversion
or further treatment of sewage effluent to remove excess
P has been effective in stabilizing river communities by
removing the dominance of filamentous algae. Nutrient
release from sediments is a major source of supply for
lakes and other slow flowing watercourses, such as canals
and drainage ditches. Even if all nutrients could be
prevented from entering such systems, there would be
sufficient nutrient supply in the sediment to provide for
continued aquatic plant growth for more than 100 years
in most cases.
Shade. Shading submerged aquatic macrophytes can
be an effective method of control, especially in sensitive
areas where chemical control is inappropriate. Shade
can be produced by planting trees or shrubs on the
banks of watercourses. However, this often interferes
with access to the water and may not be appropriate
in all situations. Opaque floating material can be used
in still waters to achieve increased shading, although
it must remain in place for at least 4 months to have
a lasting effect. The use of dyes that absorb light at
photosynthetic wavelengths is also widely used in amenity
lakes and ponds. Increased turbidity from bottom feeding
fish also excludes light, as does the action of powered
boats, and these techniques are most effective in sluggish
water where the sediment can remain suspended for
long periods.
Burning. Emergent weeds, particularly the stiffer
stemmed reeds, do not always collapse in the autumn
and can form large masses of dead, standing material. In
dry ditches and on banks, this material can sometimes
be burned. This can be a useful way of reducing the
bulk of plant material which might otherwise collapse
during the winter and block channels. Burning can
also be used to destroy cut material after drying. This
is good way of disposing of poisonous plants, some of
WEED CONTROL STRATEGIES 747
which remain toxic after death but become palatable
to livestock.
Alterations in Water Level. Exposure of part, or all, of the
bed of a water body by lowering the water level has been
used successfully to manage aquatic vegetation. Control
is achieved either by dehydration of the vegetation or by
exposure to low temperatures. The process is sometimes
termed drawdown or dewatering. It can also alter the
character of the sediment which may reduce weed growth.
However, in deeper water bodies, drawdown can allow
weeds to establish in depths below their normal limit.
Once established, they can grow up toward the surface,
as the level is again raised, to remain in the higher light
intensity. In these situations, drawdown can spread a
weed problem into areas which would normally remain
weed-free.
Some indication of the relative effectiveness of treat-
ments can be see in Table 3. This is not a definitive guide
for every situation, and expert advice should be sought
before applying any technique to a particular weed prob-
lem to optimize efficacy and minimize costs.
The benefits assigned to each technique are for general
guidance only, and the results obtained may be modified
by specific site conditions.
Other Methods of Control
Many other techniques for controlling aquatic vegetation
have been tried, which do not fit closely into any of the
previous categories. These include the following: (1) the
use of lasers to control emergent or floating weeds and
ultrasound vibrations that can disrupt cells of submerged
plants; (2) floating oil films that cause floating weeds
to sink, and (3) increasing wave action that submerges
floating weeds and increases turbidity thus suppressing
submerged vegetation. There are also several biological
control agents, including crayfish, other invertebrates,
and wildfowl, that have been effective against some
weeds. Most of these have been tested in tropical or
subtropical conditions and, at present, the grass carp is the
most effective biological control agent in more temperate
regions. Some of the more effective alternative treatments
are outlined below.
Magnetic Treatment of Flowing Water. The use of
magnetic water treatment devices, it has been shown,
affects the growth of algae. This is thought to be due
to interference with the uptake and storage of calcium
in algal cells. This treatment is most effective in closed
recirculating systems.
Ultrasound. Ultrasonic devices are used widely in
Europe to disrupt cells mainly of algae and biofilm
bacteria in water storage reservoirs where herbicides
cannot be used.
Miscellaneous Treatments. There are a number of
alternative treatments for water that affect the nutrient
concentration and, therefore availability, to aquatic plants.
Most of these involve precipitation of phosphate in an
insoluble form.
748 SCREEN FILTERS FOR MICROIRRIGATION

Table 3. The Relative Efficacy of Management Options

Weed Type

Emergent Floating-leaved Submerged

Management Option Narrow-leaved Broad-leaved Rooted Free-floating Rooted Algae

Mechanical
Hand-cutting Ž Ž Ž X Ž X
Flail mower Ž Ž X X X X
Weed bucket Ž Ž Ž Ž Ž Ž
Weed boat Ž Ž Ž Ž Ž X
Harvester Ž√ Ž√ Ž√ Ž Ž√ Ž
Dredger X Ž
Chemical
√ √ √
Foliar application
√
Ž√ X
√
X
√
Applied to water X X
Biological
Livestock Ž Ž X Ž√ Ž X
Waterfowl X X X
√ √
Ž√ Ž√
Fish X X
√ √
Insects and pathogens Ž Ž Ž Ž√
Biomanipulation X X X X X
Environmental
Shade Ž Ž Ž Ž Ž Ž
Nutrient manipulation X X X Ž Ž Ž
√
Effective control lasting at least one season
Ž Moderate benefit only, or control lasting less than one season
X No useful effect

SCREEN FILTERS FOR MICROIRRIGATION

DOROTA Z. HAMAN Screen filter
University of Florida
Gainesville, Florida

Water quality is a major concern in the manage-

ment of microirrigation systems (drip, microsprinkler,
microspray). Emitters plugged by physical, chemical, or
biological contaminants may create significant problems
in everyday maintenance. Effective and reliable filtration
is required for successful operation of microirrigation sys-
tems. There are different types of filters available for the
removal of physical contaminants from irrigation water.
Selection of a filter depends on the type and amount of
contaminants in the irrigation water, the type of emitters,
the size of the irrigation system, and the desired manage-
ment practices. Screen filters should be a primary choice
when water is pumped from a well where the only filtra-
tion requirement is to remove mineral particulate matter
(Fig. 1). They are sufficient in the absence of organic con-
tamination and in addition they are usually inexpensive
and easy to maintain. To remove both particulate matter
and organic growths, media filters (also called sand filters)
may also be used. They are usually recommended when
large amounts of algae or other organic contaminants
are present.
Screen filters are simple and economical filtration
devices of various shapes and sizes that are frequently
Figure 1. Screen filter installed in the main line of irriga-
tion system.
used for the removal of physical contaminants. They can
be made of metal, plastic, or synthetic cloth enclosed in
a special housing. Screen filters are recommended for the
removal of very fine sand or larger-sized inorganic debris
from irrigation water. It is normally not effective to use
screen filters for the removal of heavy loads of algae or
other organic material typical in surface water supply,
since screens clog rapidly, requiring too frequent cleaning
to be practical.
When surface water is used for microirrigation, screens
are often used as secondary filters, after organic matter
has been removed with media (sand) filters. In this
 SCREEN FILTERS FOR MICROIRRIGATION 749

capacity they prevent washed out media from entering the
irrigation system. When well water is used for irrigation,
a screen filter may be used as a primary filter, or it may
be secondary to a vortex sand separator, depending on the
mineral particle load in the water.
Filtering screens are classified according to the number
of openings per inch, with a standard wire size given
for each screen size (Table 1). The minimum size of
particle retained by a screen filter with a certain mesh
can be determined from Table 1. A 200-mesh screen
is usually recommended for drip type microirrigation
where 150 mesh is often sufficient for microsprinklers
and microsprays. Water filtered with a 200-mesh screen
will contain only smaller particles of very fine sand, silt,
and clay. These particles should freely move through
the emitter openings unless flocculation, aggregation, and
bridging are taking place due to unusual water chemistry.
If clogging problems persist with 200-mesh filtration,
chemical treatment of water may be necessary.
Generally, it is recommended to remove particles down
to a size ten times smaller than the emitter’s passageway
so that grouping and bridging of particles will not cause
clogging. The maximum particle size that can safely enter
the irrigation system with a given emitting device should
be provided by the manufacturer. Organic particles with a
density approaching the density of water tend to group and
bridge more easily. Particles heavier than water, typically
mineral particles, may settle and collect in the low flow
zones of the irrigation system.
The size of the filter is specified by its effective area,
which is the area of the openings in the screen. It is
specified in relation to cross-sectional area of the main
pipe. A desirable ratio is 2 or more (area of openings much
larger than the cross-sectional area of the pipe). The mesh
size of the filter (opening size) will depend on the smallest
particle size to be removed from the irrigation water.
Table 1. Representative Screen Mesh Numbers and the
Corresponding Standard Opening Size Equivalents
Screen Filters Removed Particles
Screen Opening
Size, Size, Soil
mesh µm Classification
4 4760 Very coarse sand
10 2000
20 711 Coarse and medium sand
40 420
80 180 Fine sand
100 152
120 125
150 105
When estimating the appropriate size of filter for a
specific application, one should consider the quality of
water needed, volume of water required to be passed
through the filter between consecutive cleanings, filtration
area of the filter screen, and allowable pressure drop
through the filter. A screen filter can handle a large
range of discharges. However, discharges that are large in
relation to the filtering surfaces will result in greater
pressure losses, shorter life of the filter, and the
requirement for frequent cleaning.
If the irrigation water contains large amounts of sand
or other inorganic contaminants, it is advisable to use
vortex sand separators before the screen filters, which
use centrifugal force to separate heavier particles away
from the fine mesh cartridge. Heavy particles collect at
the bottom of the filter, where they can be periodically
flushed away.
Simple gravity screen filters are often used as primary
filtration where an elevated water source from a canal or
reservoir is available. Since the pressure loss through a
gravity screen filter is very small, these filters are often
used in systems where such losses must be minimized.
They operate on the principle of gravity force, not water
pressure. Because only atmospheric pressure is used,
soft organic material is not forced through the screen.
Gravity screens are easy to install and maintain. Usually,
these filters are constructed as two chambers separated
by a fine mesh screen, which can be cleaned manually
or automatically. Most of them incorporate self-cleaning
water jets.
If water is pumped from a surface source (lake, pond,
stream, river, irrigation canal, or pit), aquatic plants,
fish, and animals should be kept away from the intake
pipe since the efficiency of pumping may be reduced by
debris in the water entering the suction pipe. This can
be accomplished with an intake screen. The design of the
screen must allow for the separation of the debris from
the water without pulling the debris against the screen.
A horizontal screen placed below the water surface or an
intake basket-type filter are often used for this purpose.
Some of these screens are self-cleaning and require very
little maintenance.
Particle Small amounts of larger particles, such as lime rock
Size, µm flakes from the aquifer, can be removed using a Y-
type strainer with replaceable screens and with clean-
1000–2000 out faucets or valves. They can be flushed while in
use if the required high flow for flushing is provided.
250–1000 Blowdown filters operate in the same way as other types
of pressure screen filters. They are designed for easy
cleaning, which is done by opening a valve that diverts the
100–250
water flow through the screen, releasing trapped particles.

180 89 Very fine sand 50–100

200 74
270 53

325 44 Silt 2–50

a a Filter body Screen Cap
— — Clay 2 and less
a
Screen filters are not normally used to remove this size particles. Figure 2. Schematics of a typical screen filter.
750 XERISCAPE
Occasionally, blowdown filters require manual cleaning to
eliminate particles that have become lodged in the screen.
Screen filters can be cleaned manually by removing
the screen and washing it with clean water (Fig. 2).
However, dismantling is not very practical for use with
waters containing high levels of contaminants, since it
must be done very often. Blowdown screen filters greatly
simplify the cleaning process. They are cleaned with the
water diverted through the screen by opening a valve.
This operation does not require disassembling. The screen
does not have to be removed for routine cleaning of the
filter, but it must be removed for occasional cleaning of the
particles that become lodged in the screen.
The most convenient method for cleaning screen filters
is automatic screen flushing triggered by critical pressure
differential across the filter, which increases with buildup
of debris on the screen. Automatic cleaning can also be set
on a time schedule. Filtered water is always recommended
for washing and backflushing. Additional safety devices
such as secondary screen filters at the entrance to each
irrigation system subunit are highly recommended.
XERISCAPE
DOUGLAS F. WELSH
Texas A&M University
Collage Station, Texas
Much attention and controversy have surrounded the
xeriscape concept of landscaping since its inception in
1981. The proper definition of xeriscape is ‘‘quality
landscaping that conserves water and protects the
environment.’’ Above all things, it must be a quality
design that balances the lawn area, shrub and flower
plantings, and the hardscape (i.e., decks, patios and
sidewalks). Landscapes composed of rocks or plastic
flowers alone are not xeriscapes. Xeriscapes are in tune
with the environment; therefore, xeriscape applies to the
desert in southwestern United States as well as the
semitropical southeast.
Xeriscape has seven basic principles:
• planning and design
• soil improvement
• appropriate plant selection
• practical turf areas
• efficient irrigation
• mulching
• appropriate maintenance
USE OF PRACTICAL TURF AREAS IN A XERISCAPE DESIGN
Of the seven principles, none has received more attention
than practical turf areas. This principle, which concerns
turfgrass in the landscape, has been shrouded in mis-
information that has been touted as fact by ‘‘experts’’ in
xeriscape, water supply, and turf culture.
The original turf-related principle established by the
Denver originators of xeriscape was ‘‘limited turf use.’’ For
Denver and much of the arid West, the seemingly logical
approach to reducing landscape water consumption was
simply to reduce the use of turf. However, as the Xeriscape
concept has matured and spread, the principle of limited
turf use was increasingly scrutinized by horticulturists
and turf experts. Today’s xeriscape movement incorporates
a more holistic approach to reducing turf irrigation, fully
recognizing that the type of plant materials or irrigation
in the landscape has as much of an effect on water
consumption as the human factor and good landscape
water management (Fig. 1).
THE NEED TO CHANGE ATTITUDES AND HABITS
Throughout the xeriscape movement, the evident truth is
that plants do not waste water; people do. Another fact
is that irrigation systems do not waste or save water;
people do. The mission of xeriscape is clear: Change
the attitudes and irrigation habits of professional and
amateur landscape managers. Proper water management
provides the greatest opportunity for water conservation
in the landscape.
Xeriscape focuses on the use of turfgrass in the
landscape because of the tremendous potential for irri-
gation water abuses in the name of maintaining green
turfgrass. Within the traditional landscape, turfgrass has
received the major share of total landscape irrigation
because grass often makes up a large percentage of
the total landscape. Through the principles of xeriscape,
turf irrigation can be reduced and the many benefits of
turfgrass can still be derived.
BENEFITS OF TURFGRASS IN THE LANDSCAPE
Turfgrass is an integral component of most landscapes. It
is certainly the best recreational surface for children and
athletes. Furthermore, it has a tremendous mitigating
effect on the environment, reducing heat loads, noise and
water and air pollution. A turfgrass lawn is second only to
a virgin forest in the ability to harvest water and recharge
groundwater resources. As a design component, turfgrass
provides the landscape with unity and simplicity while
inviting participation in it.
However, the fact remains that turfgrass is the highest
user of irrigation water in the traditional landscape. This
Figure 1. The proper definition of xeriscape is ‘‘quality landscap-
ing that conserves water and protects the environment.’’ Above
all things, it must be a quality design that balances the lawn
area, shrub and flower plantings and the hardscape (i.e., decks,
patios and sidewalks).
 XERISCAPE 751

is significantly different from saying that turfgrass is the

highest water-using plant in the landscape—which is not
the case. The discrepancy between these two statements
yields the most common misconception and misrepre-
sentation in xeriscape, and it is therefore the basis of
controversy and unproductive efforts. To resolve this con-
troversy, some scientific and practical fundamentals of
turfgrass are explained using actual xeriscape principles.

XERISCAPE PRINCIPLES FOR REDUCING

TURFGRASS IRRIGATION

Specifically, xeriscape principles promote the following

strategies to reduce turfgrass irrigation:

• Prepare soils for turf areas as carefully as any

other planting area to use all the moisture available,
promoting the plant’s vigor and water-use efficiency.
• Place turf species in landscape zones based on water Figure 2. In residential landscapes, turf areas are important for
requirements. recreation and entertainment.

• Select adapted turf species and varieties that have

lower water demands. landscapes, but in the case of parklands, industrial sites
and rights-of-way, brown turf may be acceptable.
• Irrigate turf in areas that provide function (i.e.,
recreational, aesthetic, foot traffic, dust and noise
SELECTING THE PROPER TURFGRASS SPECIES
abatement, glare reduction, temperature mitigation).
AND VARIETIES
• Use nonirrigated turf areas where appropriate.
• Irrigate turf based on true water needs. Wherever the landscape, selection of turfgrass species
• Decrease fertilization rates and properly schedule and varieties is of utmost importance. Extensive research
fertilization. has shown that there are significant differences in water
requirements among turf species and even among varieties
within species. The capacity of different turf species to
FINE-TUNING TURFGRASS XERISCAPE PRINCIPLES
avoid and resist drought also varies significantly. To
help reduce landscape water requirements, xeriscape
In traditional landscape design, turfgrass makes up the
recommends selecting turfgrass varieties (and other
major portion of landscapes. The tremendous square
landscape plants) that are both adapted to the area and
footage of turfgrass in a landscape accounts for the
have the lowest practical water requirements.
reason that turfgrass irrigation, as a percentage of total
Landscape managers should be keenly aware of
landscape irrigation, is so high. The ‘‘practical turf areas’’
drought-stress indicators shown by turfgrass and other
guideline promotes the use of turf only in those areas
plants in the landscape, including a range of color
of the landscape that provide function. In residential
changes, leaf curl, and wilting, and they should strive
landscapes, a turf area is usually a necessity for recreation
and entertainment. But turf should not be irrigated on
narrow strips of land or other areas that are difficult to
water (Fig. 2).
Good landscape water management begins with
planning and design. By designing the landscape as zones
based on plant water needs, turf can be appropriately
placed for function, benefit and water efficiency. Zoning the
landscape and irrigation system allows watering turfgrass
on a more frequent schedule than shrubs. For established
trees and shrubs, the irrigation strategy should use
deep soil moisture and depend on natural rainfall to
replenish soil moisture. When sufficient rainfall does not
occur, supplemental irrigation of trees and shrubs may
be required.
Another way to incorporate turf into the landscape and
conserve water is simply not to irrigate. Many turfgrass
species are drought-tolerant and can survive extreme Figure 3. Wherever the landscape, selection of turfgrass species
drought conditions. The grass may turn brown for a and varieties is of utmost importance. Extensive research has
while, but rainfall will green it up again. This approach shown that there are significant differences in water requirements
may be unacceptable for many residential and commercial among turf species and even among varieties within species.
752 MEDIA FILTERS FOR MICROIRRIGATION
Figure 4. Attitudes and habits about turf are changing. For
much of the arid West, the seemingly logical approach to reducing
landscape water consumption was simply to reduce the use of
turf. However, as the xeriscape concept has matured and spread,
the principle of limited turf use was increasingly scrutinized
by horticulturists and turf experts. Today’s xeriscape movement
incorporates a more holistic approach to reducing turf irrigation,
fully recognizing that the type of plant materials or irrigation in
the landscape has as much of an effect on water consumption as
the human factor and good landscape water management.
to meet the water needs of each group of plants. By
irrigating only when the plants require water versus by the
calendar, the manager can dramatically reduce landscape
water use.
The water requirements of turfgrasses can be mini-
mized through specific horticultural practices. Decreasing
fertilizer application rates and timely applications of slow-
release fertilizers tend to reduce flushes of growth that
can increase water requirements (Fig. 3).
THE XERISCAPE CHALLENGE
Xeriscape is a challenge and an opportunity for the
‘‘green’’ (landscape, turf and nursery) and ‘‘blue’’ (water
utilities and agencies) industries. Through xeriscape,
these two industries have been brought together to focus
on landscape water use. Although this marriage has not
always been easy, the best minds are prevailing in efforts
to perfect and implement the xeriscape concept.
By embracing the xeriscape concept, including the
principle of practical turf areas, the green and blue
industries can continue to be recognized as good stewards
of the environment (Fig. 4).
MEDIA FILTERS FOR MICROIRRIGATION
DOROTA Z. HAMAN
University of Florida
Gainesville, Florida
To avoid plugging of microirrigation emitters, it is
generally recommended to remove particles down to a size
ten times smaller than the emitter’s passageway so that
grouping and bridging of particles will not cause clogging.
The manufacturer should provide the maximum particle
size that can safely enter the irrigation system with a
given emitting device. Organic particles with a density
approaching the density of water tend to group and bridge
more easily.
Media (also called ‘‘sand’’) filters are well suited for
removal of either organic or inorganic particles. Due to
their three-dimensional nature, media filters have the
ability to entrap large amounts of contaminants. They
do not seal off as easily and therefore will not clog as
often as filters that trap particles on their surface, such as
screen filters.
Media filters used in microirrigation systems operate
under pressure. They consist of fine gravel and ‘‘sand’’
(that actually is crushed granite or silica) of selected
sizes placed in pressurized tanks (Fig. 1). The main body
of the tank contains sand, which is the active filtering
ingredient. The sand is placed on top of a thin layer of
gravel that separates it from an outlet screen. Sharp-
edged sand or crushed rocks are recommended for the
filtering media since the sharp edges catch soft algal
tissue. Crushed granite or silica graded into specific sizes
for a particular system are commonly used porous media.
It is not recommended to use natural sand as a filtering
media since the particles are usually rounded and smooth.
The size of media particles is very important. Too coarse
particles result in poor filtration and possible clogging
of the emitting devices, while too fine particles trigger
unnecessary frequent backwashing of the filter.
Two factors describe the media used in the filter:
uniformity coefficient and mean effective size. Uniformity
coefficient reflects the range of sand sizes within the grade.
It is desirable to keep sand particles as uniform in size
as possible. The uniform size of filtering media assures
better control of the filtration. Grading in size from fine
to coarse causes premature clogging of the filter, since the
nonuniform pores of the media also retain particles small
enough to be tolerated by the emitters. The uniformity
coefficient is represented by the ratio of the size of the
screen opening that would pass 60% of the filter sand to
the screen opening that will pass 10% of the same sand.
For irrigation purposes a uniformity coefficient of 1.5 is
considered adequate.
The mean effective sand size is the size of the screen
opening that will pass 10% of the sand sample. It is
a measure of the minimum sand size in the grade and
therefore an indicator of the particle size that will be
removed by the media. Some examples of sand media are
shown in Table 1. It can also be seen that the quality
Inflow
Discharge
Filtering media
Coarse media
supporting
filtering media
To the irrigation system
Figure 1. Schematic of media filters in filtration mode.
 Table 1. Relationship Between Designated Media Numbers, Filtration Quality, and Equivalent Screen Sizes
Removed Particles Media Filters
Corresponding Opening Filtration
Screen Size, Size, Soil Particle Media Quality,
mesh µm Classification Size, µm Number Material mesh

4 4760 Very coarse 1000–2000 8 granite 100–140

10 2000 sand

20 711 Coarse and 250–1000

40 420 medium sand

80 180 Fine sand 100–250

100 152

120 125 11 Granite 140–200

150 105 16 Silica 140–200

180 89 Very fine 50–100 20 Silica 200–230

200 74 sand

270 53 30 Silica 230–400

325 44 Silt 2–50

—a —a Clay 2 and less

a
Screens are not normally used to remove this size particles.

Figure 2. Media filters installed in a field.

753
754 MEDIA FILTERS FOR MICROIRRIGATION
of filtration increases with a smaller effective size of
filtering media.
The design flow capacity of a media filter is expressed
in liters per second of flow per square meter of surface
area of the media normal to the direction of flow. The
filter should be sized so it can handle the poorest water
quality at a given site and provide the required flow for the
functioning of the irrigation system (Fig. 2). The quality
of irrigation water before filtration may vary with the
seasons and weather conditions. This is especially true
with regard to surface water but can also apply to well
water in some cases.
The effectiveness of a filter is a measure of its ability
to remove particles of a certain size. The effectiveness
of filtration increases with a decreasing size of filtering
media grain size (larger designation number). However,
smaller media size will require more frequent cleaning.
Filtration effectiveness is also inversely proportional
to the flow rate through the filter. The higher the flow
rate, the lower the effectiveness of the media filters.
Therefore, filter manufacturer’s specifications and water
quality samples should be used to select a filter for a
specific application.
Media filters are cleaned by backwashing (Fig. 3). This
operation consists of reversing the direction of water flow
in the tank. Clean water is usually supplied from the
second tank. The upward flow fluidizes the media and
flushes collected contaminants. The backflush water is
discharged and does not enter the irrigation system. The
backwash flow must be carefully adjusted in order to
provide sufficient cleaning without accidental removal of
the media. If flows larger than those recommended are
used to backflush the media, it is very likely that the media
will be flushed from the filter by the backflow water. Only
recommended flow rates for a given media should be used.
Most media filters are backflushed at prescheduled
time intervals or by using automatic devices based on the
pressure loss across the filter. For systems in which low
Inflow
Discharge
Filtering media
Coarse media
supporting
filtering media
To the irrigation system
Figure 3. Schematic of the media filters in backflush mode.
quality water requires frequent backflushing, automatic
cleaning is necessary to avoid problems.
As the filter is backflushed every time the pressure
differential exceeds a predetermined value, large pressure
drops in the irrigation system are avoided, maintaining
the system’s uniformity and efficiency. The pressure
differential triggering backflushing depends on the
pressure required for the proper functioning of the
irrigation system. The irrigation system pump should be
able to supply enough pressure to compensate for the
pressure drop through the filter just before the filter is
washed. Pump capacity must also be adequate to supply
enough pressure and flow rate to flush the filter. Also, it
is advisable in a large irrigation system to use a number
of smaller tanks instead of a few large ones because of
the higher backwash flow that large tanks require for
good cleaning. Automatic backflush systems will eliminate
sudden changes in water quality that can create problems
if a filter is washed only at regular intervals. This
is especially important in systems using surface water
supplies with changing contamination levels.