Aquatic Botany 126 (2015) 38–47

Contents lists available at ScienceDirect

Aquatic Botany
journal homepage: www.elsevier.com/locate/aquabot

Responses to water depth and clipping of twenty-three plant species

in an Indian monsoonal wetland
Beth A. Middleton a,b,∗ , Arnold G. van der Valk b , Craig B. Davis c
a
U.S. Geological Survey, National Wetlands Research Center, 700 Cajundome Boulevard, Lafayette, LA 70506, USA
b
Ecology and Evolutionary Biology, Iowa State University, Ames, IA 50011, USA
c
School of Environment and Natural Resources, Ohio State University, Columbus, OH 43210, USA

a r t i c l e i n f o a b s t r a c t

Article history: Responses of species to disturbances give insights into how species might respond to future wetland
Received 12 November 2014 changes. In this study, species of monsoonal wetlands belonging to various functional types (graminoid
Received in revised form 10 June 2015 and non-graminoid emergents, submersed aquatic, floating-leaved aquatic) varied in their growth
Accepted 15 June 2015
responses to water depth and harvesting. We tested the effects of water depth (moist soil and flooded)
Available online 25 June 2015
and clipping (unclipped and clipped) on the biomass and longevity of twenty-three dominant plant
species of monsoonal wetlands in the Keoladeo National Park, India in a controlled experiment. With
Keywords:
respect to total biomass and survival, six species responded positively to flooding and twelve species
Cattle grazing
Floodplain wetland
responded negatively to clipping. Responses to flooding and clipping, however, sometimes interacted.
Land use change Individualistic responses of species to water levels and clipping regimes were apparent; species within a
Protected area management functional group did not always respond similarly. Therefore, detailed information on the individualistic
Seasonal drought responses of species may be needed to predict the vegetation composition of post-disturbance wetlands.
Supra-seasonal drought In particular, as demands for fresh water increase around the world, studies of life history constraints
Vegetation removal and responses to hydrological changes will aid wetland managers in developing strategies to conserve
Water level tolerance biodiversity.
Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://
creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction
Disturbances such as vegetation harvesting by farmers, grazing
by domestic livestock, and seasonal water fluctuations play a major
role in regulating plant species diversity in tropical seasonal wet-
lands (Osland et al., 2011). Thus, an understanding of the responses
of wetland species to water regimes, and to harvesting/grazing
can provide insights into the maintenance of biodiversity in mon-
soonal wetlands, which in recent years have been affected by both
(Middleton, 1999a). In droughts, species can only survive if they
can tolerate or exploit altered seasonal/inter-annual water level
fluctuations and harvesting/grazing pressure (Bond et al., 2008).
Severe supra-seasonal droughts can eliminate species lacking traits
that allow them to survive prolonged drawdown (e.g., terrestrial
forms or long-lived seed; Bond et al., 2008). At the same time, sub-
∗ Corresponding author. Current Address: U.S. Geological Survey, National Wet-
lands Research Center, 700 Cajundome Boulevard, Lafayette, LA 70506, USA.
Fax: +1 337 266 8586.
E-mail addresses: middletonb@usgs.gov (B.A. Middleton), valk@iastate.edu
(A.G. van der Valk), craigbdavis.1970@gmail.com (C.B. Davis).
mersed species with high root growth rates may be able to recover
quickly from such stresses, in contrast to species lacking these traits
(Engelhardt, 2006). In particular, as wetlands become increasingly
subjected to land-use and climate change (Bates et al., 2008), an
understanding of how aquatic species respond to and cope with
disturbances can help us better predict the impacts of these distur-
bances and even what will be required to restore these wetlands.
Species of monsoonal wetlands are well adapted to seasonal
water level fluctuation, with certain species dominating dur-
ing summer drawdowns and others after post-monsoon flooding
(Finlayson, 2005; Hejný et al., 1998; Middleton, 1999a, 2009).
Seasonal and long-term water level fluctuations help to main-
tain spatial and temporal heterogeneity of wetland zones (Květ
et al., 1988; van der Valk et al., 2015), and thus the high biodi-
versity of many wetland types including arid Australian wetlands,
prairie glacial marshes, and Indian monsoonal wetlands (Finlayson,
2005; van der Valk and Davis, 1978; Middleton, 1999a,b, respec-
tively).
Despite their adaptations to disturbances, current levels of land-
use change may be overwhelming Indian wetland species’ abilities
to persist, which could lead to permanent compositional changes
in these wetlands. Along the Brahmaputra–Ganges–Meghna and

http://dx.doi.org/10.1016/j.aquabot.2015.06.004
0304-3770/Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
 B.A. Middleton et al. / Aquatic Botany 126 (2015) 38–47 39

Indus River in India, floodplain wetlands receive less water as Casanova and Brock, 2000). If this were true, it would greatly sim-
impoundments store water upstream for human usage. These plify predicting the composition of post-disturbance vegetation.
activities may increase with climate change (Chauhan and Gopal, The objective of this study was to determine if the responses
2005), which has been linked to increased variability in monsoon of these wetland species to disturbances were consistent within
precipitation, i.e., both longer droughts and larger floods (Mani their functional type. Specifically, we examined the responses of
et al., 2009). In addition, anthropogenic changes to river floodplains all the dominant wetland species in an Indian monsoonal wetland
could alter their seasonal water-level fluctuations by impounding in the Keoladeo National Park, Rajasthan during the 1980s, to two
wetlands during the dry season, drying them during the wet sea- different water (saturated soil and flooded) and clipping (unclipped
son, or altering inter-annual patterns of drought (Bond et al., 2008). and clipped) regimes. We hypothesized that the responses of the
Virtually all of the original wetlands in this region have been lost species in each functional type would be similar e.g., submersed
due to conversion for agricultural and other development activities species would die or have lower biomass when clipped under moist
(Scott, 1993). conditions.
As predicted by the Intermediate disturbance hypothesis, mod-
erate levels of grazing or cutting in wetland ecosystems can reduce
2. Materials and methods
the growth of dominant species and increase biodiversity by pro-
viding opportunities for sub-dominant species to grow (Middleton,
2.1. Study site description
2013). Without some level of grazing or cutting, biodiversity may
decline in many types of wetlands (Osland et al., 2011; Wesche
Keoladeo National Park is a 29 km2 monsoonal wetland com-
et al., 2012; Middleton, 2013). For example, cutting of the above-
plex of seasonally and temporarily flooded grassland and thorn
ground portions of emergent species can cause a reduction in
woodland in the floodplain of the Gambhir and Banganga Rivers
oxygen levels in the roots, an effect that is particularly pronounced
(27◦ 13 N, 77◦ 32 E; Fig. 1; Middleton et al., 1991). The climate
in deep water, which can lead to the death of individuals of
is semi-arid and characterized by three seasons including a dry
some species (Mathis and Middleton, 1999). Grazing and flooding
summer from March–June (mean maximum to minimum tem-
together may affect the growth of individuals more than expo-
peratures: 41–28 ◦ C), monsoon from July–September, and winter
sure to only one of these factors (Oesterheld and McNaughton,
from October–February with low temperatures above 0 ◦ C (mean
1991). Reduced growth and mortality due to only cutting/grazing
maximum to low minimum temperatures: 19–6 ◦ C; Chauhan and
can occur in wetland species undergoing severe continuous grazing
Gopal, 2005). Before 1990, Park wetlands filled with water depend-
(Middleton et al., 1991), even though, terrestrial grassland species
ing on the amount of monsoon rain and river water discharge
tend to tolerate moderate levels of grazing (Hickman et al., 2004).
(Fig. 2), stayed flooded to varying extents during the cool win-
Drier conditions with accompanying lower water levels can result
ter season, and then rapidly dried during the hot summer season
in overgrazing in semi-arid ecosystems (Fisher et al., 2004). The
(Middleton et al., 1991). Mean annual precipitation in this region
main underlying reason for survival after cutting may be whether
is quite variable from year to year (Irrigation Department, Bharat-
or not cut individuals are able to reallocate resources from roots to
pur, Rajasthan; Fig. 2), averaging about 69 cm year−1 ; the majority
shoots as based on the amount of stored carbon in the roots (Del-Val
of annual rainfall falls during the summer monsoon season (96%
and Crawley, 2005).
during June–September; Koteswaram, 1978). Rainfall may be less
The Keoladeo National Park has undergone a number of different
predictable in this region because of atmospheric warming in
water and grazing management strategies in past decades. Cattle
recent decades (Mani et al., 2009). Perhaps a more important fac-
grazing and harvesting of grasses were common in the Keoladeo
tor in water supply for the Keoladeo National Park is that demands
National Park during the 1980s (Middleton et al., 1991), and in
for water have been increasing in north-central India (Chauhan
other wetlands in north-central India (Chauhan and Gopal, 2005;
and Gopal, 2005), so that discharge into Park has been decreasing
Middleton, 2013). Water discharge to the Keoladeo National Park
(Fig. 2).
(Fig. 1) has been decreasing since the early 1990s (Fig. 1) so that
Floating marsh characterized the deepest zones of the Park
concerns have been raised about the drying of these wetlands
wetlands during the 1980s and included submersed species such
(Sebastion, 2012). Nevertheless, a counter-trend has developed in
as Hydrilla verticillata, floating-leaved species such as Nymphoides
recent years in Asia, so that wetland restoration programs have
cristatum and grasses such as Paspalum distichum, all species with
emerged, particularly to create wildlife corridors and recreational
very low root biomasses (van der Valk et al., 1993). Shallower
parks (Wang et al., 2013).
seasonally flooded zones were dominated by graminoids such as
Many attempts have been made to group wetland species into
Cynodon dactylon, P. distichum, and Scirpus tuberosus. Drier grass-
various kinds of growth or life form groups (see Hutchinson, 1975,
land and woodland were dominated by Acacia nilotica and Vetiveria
pp. 118–132), and to review classifications of wetland species
zizanioides (van der Valk et al., 1993). Although, water buffalo graz-
by their morphological characteristics and functional groups (see
ing stopped after the designation of the national park in 1981, feral
Casanova and Brock, 2000 and Boutin and Keddy, 1993). Casanova
cattle grazing and hand-cutting of vegetation were common after
and Brock (2000) classified wetland species based on their hydro-
this time (Middleton et al., 1991).
logical tolerances (submersed, amphibious fluctuations tolerators,
In recent years, after a number of dry years following upstream
amphibious fluctuations responders, etc.) while Boutin and Keddy
diversion of river water and drought, the Park reflooded via pre-
(1993) recognized seven guilds (functionally similar species) based
cipitation in 2008. To restore seasonal winter flooding, water was
on their growth characteristics (e.g., ruderals, matrix species, inter-
diverted to the Park via the Govardhan drain pipeline in 2012, so
stitial). We have adopted a classification of the wetland macrophyte
that longer term seasonal flooding may become more common in
species of the Keoladeo National Park following Hutchinson (1975)
the future (Sebastian, 2012).
and Rejmánková (2011) based primarily on their morphologi-
cal characteristics. These macrophyte types include free-floating,
floating-leaved, and emergents, with emergents further differenti- 2.2. Experimental design of water depth-clipping study
ated as creeping emergents, and erect emergents (graminoid and
non-graminoid). As with other attempts to group species into func- Twenty individuals of each species were established in October
tional types, it is assumed that all the species of a certain type 1986 from root or rhizome cuttings in cylindrical clay pots (12 cm
will respond in a similar way to disturbances (Hutchinson, 1975; circumference by 25 cm depth) filled with clay soils from the
40 B.A. Middleton et al. / Aquatic Botany 126 (2015) 38–47

Fig. 1. Keoladeo National Park, Bharatpur, Rajasthan, India. Letters designate “blocks”, which are used as a location designator in India. Modified from Hockings and Mathur
(2003).

Fig. 2. Annual fluctuation before and after major river water diversion activities (∼1990) in (A) water discharge (mean m3 per season) released into the Keoladeo National
Park from Ajunbund (temporary reservoir) after the monsoon (September–February) via Ghana Canal on the Gambhir river and in (B) regional annual mean precipitation
(mm). Linear regression lines indicate time trends in discharge: 1965–1989: y = 34.3–0.01x, R2 = <0.01, p = 0.9289; 1990–2008: y = 1255.4–0.62x, R2 = 0.40, p = 0.0035; and in
precipitation: 1971–1989: y = 28,428.7–14.1x, R2 = 0.17, p = 0.0822; 1990–2008: y = 6083.4–2.73x, R2 = <0.01, p = 0.7542. Data are given by monsoonal year (July–September)
from Bharatpur and Rajasthan, which is directly adjacent to the Park (Irrigation Department, Bharatpur, Rajasthan India).

wetland. Plants were grown in well watered conditions for two
months before beginning the experiment in an experimental
pond within the forest nursery of the Keoladeo National Park.
The twenty-three species used in the experiment were all of
the dominant herbaceous and graminoid species present in the
Park in the 1980s have been classified into four functional types:
graminoid emergent (C. dactylon, Cyperus alopecuroides, Cyperus
bulbosus, Desmostachya bipinnata, Eleocharis palustris, Oryza sativa,
Panicum paludosum, Paspalidium punctatum, P. distichum, Sporobo-
lus helvolus, Scirpus littoralis, S. tuberosus, and Vetiveria zizanioides),
 B.A. Middleton et al. / Aquatic Botany 126 (2015) 38–47
non-graminoid emergent/creeping emergent (Ipomoea aquatica,
Typha angustata, Hemiadelphus polysperma), submersed aquatic
(Hydrilla verticillata, Potamogeton crispus, Potamogeton indicus,
Vallisneria spiralis), and floating-leaved/free-floating species (N.
cristata, Nymphoides indica, Eichhornia crassipes). Nomenclature fol-
lows EFloras.org, 2014 and Prasad et al., 1998. Voucher specimens
are stored in the herbarium of Iowa State University.
This study was a 2 × 2 factorial experiment with two water
regime treatments including saturated (well watered, but no stand-
ing water over soil) or inundated (to 12 cm depth) and two clipping
treatments including unclipped and clipped bi-weekly to the sur-
face of the soil (above the apical meristem of the plant). Water
regimes were maintained by lowering the pot into the experimen-
tal pond so that the soil in the pot was either 3 cm emerged or 12 cm
submerged in the water (saturated vs. flooded treatment, respec-
tively). There were five replicates of every treatment combination
(water regime × clipping treatment) for each species (20 pots of
each species, 460 total individuals), which were arranged within
the experimental pond, and the experiment run from December
1986–February 1987.
Heights were measured, numbers of leaves counted, and flow-
ering noted of all culms in the pots every two weeks. Any individual
that appeared to have died was monitored for new growth during
the experiment, and the time of death assessed at the end of the
experiment by examining shoot and root integrity (i.e., solid plant
material was alive vs. soft was decomposing and not alive). Roots
and leaves were placed in a drying oven at 70 ◦ C for several days
until completely dry and then weighed.
2.3. Data analysis
Measured plant responses to water regime and clipping
included the relative percent survival time i.e., the number of days
an individual survived divided by the total number of days in the
experiment expressed as a percent. Total stem length per pot was
assessed as the accumulated total length of stems including the
cumulative length of clippings during the experiment. Total num-
ber of leaves was the total number of leaves accumulated by the
plant during the experiment. Total shoot biomass was the biomass
of the final shoot material plus accumulated clippings. Final root
biomass was the weight of below-ground portions (roots and rhi-
zomes) at the end of the experiment. Total biomass was the total
above and below-ground biomass. In addition, root to shoot ratio
(Květ et al., 1988) was calculated for each individual in the experi-
ment by dividing total root biomass by total shoot biomass. This
method of calculating root to shoot ratio had a shortcoming, in
that the roots and shoots were not directly comparable; total shoot
biomass included the accumulated cuttings from the experimen-
tal treatments, while the root biomass was based only on the final
weight of below-ground materials. For percent survival, accumu-
lated total length of shoots, and total biomass (weight), differences
in the responses of inundated vs. saturated individuals were eval-
uated by dividing the mean of inundated individuals by the mean
of saturated individuals (inundated treatment). Differences in the
responses of clipped vs. unclipped individuals were evaluated in a
similar fashion.
Plant species responses to water regime and clipping were
tested using three-way ANOVA using the main effects of species,
water regime, clipping level and their interactions. Data were
appropriately transformed to meet assumptions of ANOVA. One
degree-of-freedom contrasts were made on mean comparisons
of interest as based on significant interactions of species × water
regime × clipping interactions. All statistical analyses were con-
ducted using JMP SAS (2012).
41
3. Results
3.1. General findings
Table 1 summarizes the overall results of our study and indicates
whether cutting/grazing or water regime had a significant impact
on the longevity or biomass of each species. Of the twenty-three
species, clipping had a major impact on ten species (43%), water
regime (grow better when flooded) on six species (26%), and both
clipping and water regime on six species (26%). Other species (43%)
were not significantly affected by either clipping or water regime.
3.2. Graminoid emergent macrophyte
Percent survival time of eight graminoid emergents did not
differ significantly in the water and/or cutting treatments (per-
cent survival time: 0.86–1.0; Table 1); these species included C.
dactylon, C. bulbosus, E. palustris, O. rufipogon, P. distichum, S. helvo-
lus, and V. zizanioides, (p > 0.05; Tables 1 and 2; Appendix 1A).
Differences in percent survival time between inundated and sat-
urated treatments were variable with P. punctatum in inundated
treatments having shorter percent survival times than in satu-
rated treatments (p < 0.0001; Fig. 3A(i); Appendix 1A). S. tuberosus
had shorter percent survival times when clipped under saturated
rather than inundated treatments (percent survival time: 0.10 vs.
0.89, respectively; p < 0.05); others under inundated treatments (C.
alopecuroides and P. paludosum; p < 0.0001). Desmostachya bipin-
nata and S. littoralis had much shorter percent survival times when
clipped under either saturated or inundated treatments (p < 0.05;
Fig. 3B(i), Tables 1 and 2; Appendix 1A).
Accumulated shoot length and leaf numbers differed for some
graminoid emergents in response to cutting and water regime
(Table 2). Many species had large differences in accumulated stem
length of individuals in both inundated and clipped treatments
(Table 2). Species responses to clipping treatments varied with
water depth (i.e., interactions; Table 2) for above-ground biomass
(total accumulated biomass of stems and leaves during experi-
ment) and below-ground biomass. Total biomass (total above- and
below-ground biomass) of seven graminoid emergents was unaf-
fected by clipping and water regime (C. bulbosus, E. palustris, O.
rufipogon, P. punctatum, P. distichum, S. helvolus, and V. zizanioides
(p > 0.05; Tables 1 and 2). Differences in total weight (biomass)
were proportionally less in inundated than saturated treatments
for some species including C. dactylon, P. paludosum, and S. helvo-
lus with lower total weight in inundated treatments, but these
species also had lower biomass when clipped in saturated condi-
tions (Fig. 3A(iii) and B(iii); Tables 1 and 2). C. alopecuroides and D.
bipinnata had lower total biomass when clipped in the inundated
treatment (p < 0.05; Tables 1 and 2). Only P. paludosum flowered
when clipped during the study (Table 1).
Root to shoot ratio was affected by either water regime or
clipping for four of the graminoid emergents (p < 0.05; Table 2);
however, O. rufipogon and S. littoralis had higher root to shoot ratios
in the inundated treatment (120.0 ± 11.0 vs. 29.5 ± 4.9, respec-
tively; t = 9.4, p < 0.0001) while S. tuberosus had a higher root to
shoot ratio when clipped in the saturated treatment (root to shoot
ratio: 26.7 ± 4.9 vs. 1.6 ± 4.9, t = 3.6, p = 0.0003). Scirpus littoralis had
a higher root to shoot ratio if not clipped in the inundated treatment
(root to shoot ratio: 29.5 ± 4.9 vs. 1.2 ± 4.9, respectively; t = 4.1,
p < 0.0001).
3.3. Non-graminoid emergent macrophytes
During the study, percent survival time of the non−graminoid
emergent T. angustata did not differ significantly in water and/or
cutting treatments (p > 0.05; Fig. 4A(i) and B(i); Tables 1 and 2;
42 B.A. Middleton et al. / Aquatic Botany 126 (2015) 38–47

Table 1
Species response (positive = +; negative = −; and unaffected = 0) of biomass (B) and percent survival time (S) to clipping and flooding treatments and interactions.

Species Functional type Clipping B/S Flooding B/S Interaction B/S Flowering Major impact(s)

Hydrilla verticillata S 0/− 0/0 0/−* i, u Cutting

Potamogeton crispus* S −/− +/0 −/−* Cutting/drought
Potamogeton indicus S 0/0 0/+ 0/0 Drought
Vallisneria spiralis S 0/+ 0/+ 0/− i, u Drought
Ipomoea aquatica Non-GE 0/− 0/+ 0/0 Cutting/drought
Nymphoides cristatum FL 0/0 +/+ 0/0 i,c & m,u Drought
Nymphoides indica FL −/− +/+ −/− Cutting/drought
Eichhornia crassipes FF 0/0 0/0 0/0
Typha angustata Non-GE 0/0 0/0 0/0
Hemiadelphus polysperma C 0/0 0/0 0/0 m,c & m,u
Cynodon dactylon G −/0 −/0 −*/0 Cutting/flooding
Cyperus alopecuroides G −/0 0/0 −/− Cutting
Cyperus bulbosus G 0/0 0/0 0/0
Desmostachya bipinnata G −/− 0/0 −/0 Cutting
Eleocharis palustris G 0/0 0/0 0/0
Oryza sativa G 0/0 0/0 0/0
Panicum paludosum G −/− −/0 −*/−* m,c Cutting/flooding
Paspalidium punctatum G 0/0 0/0 0/0
Paspalum distichum G 0/0 0/0 0/0
Scirpus littoralis G 0/− 0/0 0/0 Cutting
Scirpus tuberosus G −/− −/0 −*/−* Cutting/flooding
Sporobolus helvolus G 0/0 0/0 0/0
Vetiveria zizaniodes G 0/0 0/0 0/0

A score of −* indicates that a negative interaction occurred in the saturated soil treatment. Flowering occurred in saturated (m), inundated (i), unclipped (u), or clipped (c)
conditions. In assessing the major stressor (cutting, flooding or drought), its impact on longevity is primarily considered. Macrophyte species are grouped by life form (S –
submersed; FL – floating leaved; FF – free floating; E – emergent; C – creeping emergent; Non-GE – non-graminoid emergent; G – graminoid emergent). These summary
responses are based on multiple comparisons based on significant main effects and interactions in the ANOVA (Table 2).

Fig. 3. Responses of graminoid emergent species to experimental treatments for: (A) water regime (saturated vs. flooded), and (B) clipping (clipped vs. not clipped), with
differences expressed as ratios of control vs. treated responses as (i) % survival, (ii) total accumulated stem length, and (iii) total mass. As an example for comparing the
control vs. treated response to water regime, differences in % survival were calculated as the ratio of a species’ mean survival in unflooded vs. flooded treatments.
 B.A. Middleton et al. / Aquatic Botany 126 (2015) 38–47
Fig. 4. Responses of non-graminoid/creeping emergent species to experimental treatments for: (A) water regime (saturated vs. flooded), and (B) clipping (clipped vs. not
clipped), with differences expressed as ratios of control vs. treated responses as (i) % survival, (ii) total accumulated stem length, and (iii) total mass.
Appendix 1B), while I. aquatica was negatively affected by satu-
rated soil and/or cutting treatments (p < 0.05; Fig. 4A(i) and B(i);
Tables 1 and 2; Appendix 1B). Hemiadelphus polysperma (creep-
ing emergent macrophyte) was unaffected by water and/or cutting
treatment including percent survival time, total stem length, stem
biomass, root biomass and total biomass (p > 0.05; Fig. 4A and B;
Table 1; Appendix 1B and D and Appendix 2 B, D and F), except
that this creeping emergent had more leaves in clipped treatments
(p < 0.05; Appendix 1F). H. polysperma flowered in all treatments
during the study but the other two non-graminoid emergents
did not (Table 1). Total biomass (total above-and below-ground
biomass) of I. aquatica was unaffected by cutting and water regime,
but for T. angustata was higher in saturated soil treatments (p > 0.05;
Fig. 4A(iii) and B(iii); Tables 1 and 2). Root to shoot ratios of these
two non-graminoid emergent were not affected by water and/or
cutting treatments (p > 0.05). Note that H. polysperma had negligible
root biomass.
3.4. Submersed macrophyte
Two of the submersed aquatic species had longer percent sur-
vival times under inundated conditions including P. indicus and V.
spiralis (p < 0.05), although H. verticillata and P. crispus did not differ
in inundated versus saturated treatments (p > 0.05; Tables 1 and 2;
Fig. 5A(i); Appendix 1B). H. verticillata and P. crispus had longer
percent survival time in unclipped treatments in moist conditions
(p < 0.05; Fig. 5B(i); Tables 1 and 2; Appendix 1B). Total biomass
(total above- and below-ground biomass) of H. verticillata, P. indi-
cus, and V. spiralis were unaffected by cutting and water regime
43
(Fig. 5A(iii) and B(iii) Tables 1 and 2; Appendix 2F), while for P.
crispus in inundated, unclipped treatments both total accumulated
stem length and total biomass were higher (Fig. 5A(ii) and B(ii),
respectively; Table 2; Appendix 1D and 2F). Both H. verticillata and
V. spiralis only flowered in inundated, unclipped treatments. Root to
shoot ratios of these submersed aquatics do not warrant discussion
because these species had negligible root biomasses.
3.5. Floating-leaved/freely-floating macrophyte
Percent survival time of the free-floating (invasive) aquatic E.
crassipes did not differ significantly in the water and/or cutting
treatments (p > 0.05; Fig. 6A(i) and B(i); Tables 1 and 2; Appendix
1B). The floating-leaved aquatic species, N. cristatum had longer
percent survival times under inundated treatments but was not
affected by clipping (p < 0.05 and p > 0.5, respectively; Fig. 6A(i)
and B(i), Tables 1 and 2; Appendix 1B), while N. indica had shorter
percent survival times when clipped under inundated conditions
(p < 0.0001; Fig. 6B(i); Tables 1 and 2; Appendix 1B). N. crista-
tum had higher accumulated stem length in inundated conditions
(p < 0.0001; Table 2; Fig. 6A(ii)). N. cristatum and N. indica had
higher total biomass in the inundated and unclipped treatments
(p < 0.003; Fig. 6A(iii) and B(iii); Tables 1 and 2). N. cristatum flow-
ered in all treatments (Table 1). N. indica had a higher root to shoot
ratio if clipped in the inundated treatment (root to shoot ratio:
36.9 ± 4.9 vs. 1.9 ± 4.9, respectively; t = 5.2, p < 0.0001) but not the
other species of this functional type (p > 0.05).
44 B.A. Middleton et al. / Aquatic Botany 126 (2015) 38–47

Fig. 5. Responses of submersed species to experimental treatments for: (A) water regime (saturated vs. flooded), and (B) clipping (clipped vs. not clipped), with differences
expressed as ratios of control vs. treated responses as (i) % survival, (ii) total accumulated stem length, and (iii) total mass.

4. Discussion species performed the same whether clipped or not in saturated

4.1. Functional group vs. individualistic responses to disturbances
Species within a functional type did not always respond in
a similar way to cutting and flooding. Thus, we must reject our
hypothesis that species with similar morphologies respond simi-
larly to disturbances. Our data indicated that even closely related
species within a functional type responded individualistically to
flooding and/or grazing. For example, we found that N. cristatum
and N. indica differed in percent survival and total biomass after
clipping underwater. These findings further suggest that N. crista-
tum is much more likely to be resilient to aquatic grazing than
N. indica. Such differences in species’ responses to the mix of dis-
turbances ultimately could affect the species composition of these
wetlands. Based on the propensity of such responses in our study,
we concluded that disturbance responses (sensu Bond et al., 2008)
should be based more on species than functional group response.
Certainly, a better understanding of how each species responds to
various kinds of disturbances is important for effective wetland
management.
The survival or biomass production of about 43% of the species
of all functional groups were unaffected by the clipping or water
regime treatments. Therefore, most of the emergent species in
our study did not fit functional group expectations because these
or flooded conditions. These unresponsive species likely have a
high flexibility to water and cutting regime (Table 1). This is an
important finding for understanding vegetation dynamics because
species that can tolerate a variety of disturbance conditions may
fare best in a future of increased drought, harvesting and grazing
(Table 1). In contrast, other emergent species such as C. dactylon,
C. alopecuroides, D. bipinnata, I. aquatica, P. paludosum, S. tuberosus
and S. littoralis had lower biomass or percent survival time when
clipped (Table 1), so that an increase in harvesting or grazing will
likely reduce their dominance.
The reasons underlying the differences of species’ responses to
cutting and flooding are undoubtedly complex. Englehardt’s (2006)
study predicted that species with high rates of root growth may be
able to recover more quickly from stresses. In contrast, our results
suggest that it may not be root size, morphology and/or leaf growth
alone that predict the survival of wetland species. For example,
despite root mass, submersed species did not survive well when
they were clipped in conditions with no standing water, which
does not follow Englehardt’s (2006) prediction. More consistent
with our results are other studies, which show widely varying root
biomasses in response to clipping and flooding, even for species
of the same functional type. In other studies, root biomasses of
the emergents Bolboschoenus caldwellii, Phragmites australis and
Schoenoplectus validus did not vary in different water depth and
 B.A. Middleton et al. / Aquatic Botany 126 (2015) 38–47
Fig. 6. Responses of floating-leaved/free-floating species to experimental treatments for: (A) water regime (saturated vs. flooded), and (B) clipping (clipped vs. not clipped),
with differences expressed as ratios of control vs. treated responses as (i) % survival, (ii) total accumulated stem length, and (iii) total mass.
clipping treatments (Hayball and Pearce, 2004), while root biomass
of Eleocharis acuta decreased at all clipping levels (Blanch and Brock,
1994). In our study, root to shoot ratio generally decreased with
increased water depth, a finding similar to that of Coops et al.
(1996); nevertheless, the root to shoot ratios of some species in
our study were not affected by either water depth or clipping (e.g.,
T. angustata and I. aquatica).
4.2. Additional work
Among the shortcomings of our work was that our study only
continued for a few months, and it is possible that the responses of
the species would have differed with more intense or longer dura-
tion treatments of cutting and flooding. Nevertheless, studies of
related species seem to show different responses depending on the
situation. For example, in a three month study of Typha in Michi-
gan, T. latifolia and T. angustifolia died after underwater cutting (Sale
and Wetzel, 1983). Our study was of a similar duration to the Michi-
gan study, but T. angustata survived when cut underwater. Clipping
intensity does not explain the differences in results between the
two studies because the individuals in the Sale and Wetzel (1983)
study were cut once per month, while in our study were cut every
two weeks. The differences in Typha responses could be attributed
to species-specific responses, although other experimental differ-
ences could also be at play. We might have found different species’
responses if our study had been conducted during another season
of the year. Clearly much more research is necessary to determine
45
how aquatic species respond to disturbances such as cutting and
flooding.
4.3. Implications for management
In monsoonal wetlands in northern India, aquatic species may
be compromised in the future because of longer dry periods with-
out standing water due to both water extraction from rivers (Fig. 2;
Chauhan and Gopal, 2005) and/or climate change (Mani et al.,
2009). While we are not aware of any comprehensive field stud-
ies of dried wetlands in northern India, our experimental results
on water depth tolerances were consistent with those of a field
study of a newly impounded wetland along the Yamuna River near
New Delhi (Chauhan and Gopal, 2005). They found that popula-
tions of submersed and rooted floating-leaved species increased in
abundance after impoundment, while less flood tolerant species
decreased in abundance or established at higher elevations. Mon-
soonal wetlands in this region may be more likely to be subjected
to drier conditions in the future, and the varying abilities of these
species to tolerate altered hydrology and cutting will likely be an
important determinant of aquatic community composition. Our
experiments provided some insights into how common wetland
species in north India might respond to long-term changes in
hydrology.
Ultimately, the maintenance of biodiversity during hydrologic
change depends on the nature of surviving seeds and/or propagules
(van der Valk and Davis, 1978; Middleton, 1999; Brock, 2011). After
46 B.A. Middleton et al. / Aquatic Botany 126 (2015) 38–47

ANOVAs of species responses to water level (saturated or flooded: not inundated or inundated to 12 cm, respectively) and clipping regime (not clipped and clipped) and the interactions of these factors as based on appropriately
transformed data. Responses include % survival time (as a percentage of total days of the experiment), total stem length accumulated (cm), total leaf number, total stem biomass (g), total root biomass (g), total biomass, and (g)
a prolonged period of little flooding, resistant aquatic species are

0.0012

<0.0303

0.0075
<0.0001

<0.0001

<0.0001

<0.0001
often clonal, with roots or rhizomes that can survive desiccation
(Mony et al., 2011). In one study of subalpine wetlands, Carex ros-

p
Root/shoot ratio
trata and Juncus balticus re-colonized from roots and rhizomes after

3.9
10.7
4.1
4.7
4.3
7.2
4.2
a drought. Scirpus acutus and Nuphar polysepalum did not survive as

F
roots or rhizomes, so that zones that were once dominated by these

10206
1283
10834
569
11326
870
11126
species did not re-establish after the drought (Rejmánková et al.,

ss
1999). For species without drought-resistant roots and rhizomes
or effective dispersal mechanisms, long-lived seed banks are espe-

0.9237
0.0404
<0.0001

<0.0001
<0.0001
<0.0001

<0.0001
ratio of shoot to root biomass. Plants were grown in an experimental tank study in the Keoladeo National Park, Bharatpur, India, October 1986–February 1987. P values indicate level of significance.

cially important in re-establishment (Middleton, 1999a,b; Brock,

2011).
p
Total biomass

Drought occurred in the Keoladeo National Park after the upriver

30.1
4.2
3.5
68.2
2.6
<0.1
3.5
diversion of water beginning in the 1990s (Fig. 1). Based on species
F

responses to flooding in this experiment as well as other stud-

14.2
<0.1
1.7
1.5
1.3
<0.1
1.7
ss

0.5084 ies (e.g., Finlayson, 2005), floating leaved and submersed species
0.0024

0.0031
<0.0001

0.0002
<0.0001

<0.0001 may be greatly reduced in abundance or even eliminated as a

consequence of drought (e.g., P. crispus, P. indicus, V. spiralis, I.
p

aquatica, N. cristatum, and N. indica). Nevertheless, these types of

Root biomass

27.3
9.3
2.5
41.8
2.1
0.4
2.6

species generally have long-lived seed banks that could support

F

their re-establishment after flooded conditions return to the wet-

9.2
0.1
0.9
0.6
0.7
<0.1
0.9

land (Middleton et al., 1991). Furthermore, our study suggests that

ss

aquatic species with robust roots and propagules in the short-term

0.1695

0.1951
<0.0001

<0.0001
<0.0001
<0.0001

<0.0001

and seed banks in the long-term might be used for the restoration
of monsoonal wetlands (Wang et al., 2013).
p
Stem biomass

While certain species are able to survive drought in seed banks,

26.3
1.9

70.4
4.1
1.7
4.8

seed banks are only present if plants can maintain episodes of

5
F

active reproduction, and reproduction can be severely affected by

5.9
<0.1
1.1
0.7
0.9
<0.1
1.1

cutting (Middleton, 1999b). In our study, only 3 of the 23 species

ss

(13%) flowered in clipping treatments, suggesting that cutting may

0.2382
<0.0001

<0.0001
<0.0001
<0.0001

<0.0001

reduce the reproductive output of most of these species (Table 1).

0.29

Similarly, a study by Crosslé and Brock (2002) showed that mechan-

Leaf number per pot

ical simulation of grazing by clipping could ultimately reduce the

369.8
49.1

47.8

33.9

29.5

germination of species from the soil seed banks of temporary

1.1

1.4
F

wetlands, even though these wetland species are otherwise well

859211

294356
593417
837840

515830

adapted to fluctuating water conditions.

1111
894

Because future water management and grazing activities will

ss

dictate the nature of biodiversity in the Keoladeo National Park, it

0.1666

0.2198
<0.0001

<0.0001
<0.0001
<0.0001

<0.0001

is important to recognize that the species of these wetlands have

differing inherent abilities to survive these disturbances. Recent
p

hydrologic restoration via the Govardhan drain project has begun to

Total stem length

5.7
1.9
6.3
747315 156.2
4.8
1.5
5.3

increase flooding in the Keoladeo National Park (Sebastian, 2012),

F

and monsoonal precipitation levels have been high in recent years

562171
599066

660173

503585

(USDA, 2013). For the immediate future, these conditions can favor
9189

7227

aquatic species in this national park. In the longer term future,

ss

drought or heavy cutting/grazing may cause the loss of species in

0.9859

0.0882
<0.0001

<0.0001
<0.0001
<0.0001

0.0003

such nature preserves. As demands for freshwater increase around

the world with predicted climate changes, studies of the life history
p
% Survival time

constraints of species in future environments are essential in devel-

9.2
<0.1
4.9
115.2
5.4
2.9
2.5

oping management strategies to avoid species loss in protected

areas.
F

22,92 16.3
<0.1
8.7
9.3
9.5
0.2
4.4
ss

22,88

22,88

22,88
1,8

1,4

1,4
df

Acknowledgments
Species × water depth × clipping type

The work was conducted under Smithsonian Grant No. 4013600

to Iowa State University as part of the project, “The ecology
Water depth × clipping type

of a semitropical monsoonal wetland in India (Bharatpur)” in

1984–1987. Our continued thanks to the State Government of
Species × clipping type
Species × water depth

Rajasthan (Jaipur) and the Bombay Natural History Society (Mum-

bai) for their assistance. We also thank our field assistants: Om
Prakash Mudgal, Laxmi Kant Mudgal, Sunil Mudgal, Raj Mudgal,
Clipping type
Water depth

Ashok Faujdar, Mahendra Singh, Gopal Singh, Nihal Singh, Bihari

Species

Singh (deceased), Prakash Singh, and Soren Singh. We thank the

Source
Table 2

associate editor, three anonymous reviewers, and Michael Osland

for comments on earlier versions of this manuscript. Any use of
 B.A. Middleton et al. / Aquatic Botany 126 (2015) 38–47
trade, product, or firm names is for descriptive purposes only and
does not imply endorsement by the U.S. Government.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.aquabot.2015.06.
004
References
Bates, B.C., Kundzewicz, Z.W., Wu, S., Palutikov, J.P., 2008. Climate Change and
Water. Technical Paper of the Intergovernmental Panel on Climate Change.
IPCC, Secretariat, Geneva, Switzerland.
Blanch, S.J., Brock, M.A., 1994. Effects of grazing and depth on two wetland plant
species. Aust. J. Mar. Freshwater Res. 45, 1387–1394.
Bond, N.R., Lake, P.S., Arthington, A.H., 2008. The impacts of drought on freshwater
ecosystems: an Australian perspective. Hydrobiologia 600, 3–16.
Boutin, C., Keddy, P.A., 1993. A functional classification of wetland plants. J. Veg.
Sci. 4, 591–600.
Brock, M.A., 2011. Persistence of seeds in Australian temporary wetlands.
Freshwater Biol. 56, 1312–1327.
Casanova, M.T., Brock, M.A., 2000. How do depth, duration and frequency of
flooding influence the establishment of wetland plant communities? Plant
Ecol. 147, 237–250.
Chauhan, M., Gopal, B., 2005. Vegetation structure and dynamics of a floodplain
wetland along a subtropical regulated river. River Res. Appl. 21, 513–534.
Coops, H., van den Brink, F.W.B., van der Velde, G., 1996. Growth and
morphological responses of four helophyte species in an experimental
water-depth gradient. Aquat. Bot. 54, 11–24.
Crosslé, K., Brock, M.A., 2002. How do water regime and clipping influence wetland
plant establishment from seed banks and subsequent reproduction. Aquat. Bot.
74, 43–56.
Del-Val, E., Crawley, M.J., 2005. Are grazing increasers better tolerators than
decreasers? An experimental assessment of defoliation tolerance in eight
British grassland species. J. Ecol. 93, 1005–1016, eFloras.org, 2014. Flora of
China http://www.efloras.org/index.aspx
Engelhardt, K.A.M., 2006. Relating effects and response traits in submersed aquatic
macrophytes. Ecol. Appl. 16, 1808–1820.
Finlayson, C.M., 2005. Plant ecology of Australia’s tropical floodplain wetlands: a
review. Ann. Bot. 96, 541–555.
Fisher, A., Hunt, L., James, C., Landsberg, J., Phelps, D., Smyth, A., Watson, I., 2004.
Review of Total Grazing Pressure Management Issues and Priorities for
Biodiversity Conservation in Rangelands: A Resource to Aid NRM Planning.
Desert Knowledge CRC Project No 3 (August 2004). Desert Knowledge CRC and
Tropical Savannas CRC, Alice Springs, Australia.
Hayball, N., Pearce, M., 2004. Influences of simulated grazing and water-depth on
the growth of juvenile Bolboschoenus caldwellii, Phragmites australis and
Schoenoplectus validus plants. Aquat. Bot. 78, 233–242.
Hejný, S., Segal, S., Raspopov, I.M., 1998. General ecology of wetlands. In: Westlake,
D.F., Květ, J., Szczepański, A. (Eds.), The Production Ecology of Wetlands.
Cambridge Univ. Press, Cambridge, U.K, pp. 1–77.
Hickman, K.R., Harnett, D.C., Cochran, R.C., Owensby, C.E., 2004. Grazing
management effects on plant species diversity in tallgrass prairie. J. Range
Manage. 57, 58–65.
Hockings, M., Mathur, V.B., 2003. UNESCO-IUCN enhancing our heritage project:
monitoring and managing for success in natural world heritage sites: Initial
management effectiveness evaluation report. In: Keoladeo National Park,
Bharatpur, India, July 2003. United Nations Foundation-IUCN, Wildlife Institute
of India, Dehradun, India.
Hutchinson, G.E., 1975. A Treatise on Limnology, Vol. 3. Limnological Botany.
Wiley, New York, NY, USA.
JMP SAS, 2012. Statistical Analysis System, v. 10.0.2. Cary, North Carolina. USA.
47
Koteswaram, P., 1978. Meteorology and climatology of the Rajasthan desert. Proc.
Indian Natl. Sci. Acad. 44, 401–410.
Květ, J., Westlake, S., Dykyjová, D., Marshall, E.J.P., Ondok, J.P., 1988. Primary
production in wetlands. In: Westlake, D.F., Květ, J., Szczepański, A. (Eds.), The
Production Ecology of Wetlands: The IBP Synthesis. Cambridge Univ. Press,
Cambridge, UK, pp. 78–169.
Mani, N.J., Suhas, E., Goswami, B.N., 2009. Can global warming make Indian
monsoon weather less predictable. Geophys. Res. Lett. 36, 1–5.
Mathis, M., Middleton, B.A., 1999. Simulated herbivory and vegetation dynamics in
coal slurry ponds reclaimed as wetlands. Restor. Ecol. 7, 392–398.
Middleton, B.A., 1999a. Succession and herbivory in monsoonal wetlands.
Wetlands Ecol. Manage. 6, 189–202.
Middleton, B.A., 1999b. Wetland Restoration, Flood Pulsing and Disturbance
Dynamics. John Wiley and Sons, New York, USA.
Middleton, B.A., 2009. Vegetation Status of the Keoladeo National Park, Bharatpur,
India. Science Investigation Report 2009–5193, U.S. Geological Survey.
Lafayette, LA, USA.
Middleton, B.A., 2013. Rediscovering traditional vegetation management in
preserves: trading experiences between cultures and continents. Biol. Conserv.
158, 271–279.
Middleton, B.A., van der Valk, A.G., Williams, R.L., Mason, D.J., Davis, C.B., 1991.
Vegetation dynamics and seed banks of a monsoonal wetland overgrown with
Paspalum distichum in northern India. Aquat. Bot. 40, 239–259.
Mony, C., Puijalon, S., Bornette, G., 2011. Resprouting response of aquatic clonal
plants to cutting may explain their resistance to spate flooding. Folia
Geobotany 46, 155–164.
Oesterheld, M., McNaughton, S.J., 1991. Interactive effect of flooding and grazing
on the growth of Serengeti grasses. Oecologia 88, 153–156.
Osland, M.J., Gonzalez, E., Richardson, C.J., 2011. Restoring diversity after cattail
expansion: disturbance, resilience, and seasonality in a tropical dry wetland.
Ecol. Appl. 21, 715–728.
Prasad, V.P., Mason, D., Marburger, J.E., Ajith Kumar, C.R., 1998. Illustrated Flora of
Keoladeo National Park. Oxford University Press, USA, Bharatpur, Rajasthan.
Rejmánková, E., Rejmánek, M., Djohan, T., Goldman, C.R., 1999. Resistance and
resilience of subalpine wetlands with respect to prolonged drought. Folia
Geobotany 34, 175–188.
Rejmánková, E., 2011. The role of macrophytes in wetland ecosystems. J. Ecol. Field
Biol. 34, 333–345.
Sale, P.J.M., Wetzel, R.G., 1983. Growth and metabolism of Typha species in relation
to cutting treatments. Aquat. Bot. 15, 321–334.
Scott, D.A., 1993. Wetland inventories and the assessment of wetland loss: a global
overview. In: Moser, M., Prentice, C., van Vessem, J. (Eds.), Waterfowl and
Wetland Conservation in the 1990: A Global Perspective. Proceedings of the
IWRB Symposium. IWRB, Slimbridge, UK, pp. 154–163.
Sebastian, S., 2012. Good Times Ahead for the Keoladeo Birds With the Completion
of the Govardhan Pipeline. The Hindu, Jaipur, India, October 4.
USDA, United States Department of Agriculture, 2013. Excess Monsoonal Rainfall
Raise Concerns for India’s 2013/2014 Soybean Productivity. September 5,
2013. USDA Foreign Agricultural Service, Washington, D.C., USA http://www.
pecad.fas.usda.gov/highlights/2013/09/India/
van der Valk, A.G., Davis, C.B., 1978. The role of seed banks in the vegetation
dynamics of prairie glacial marshes. Ecology 59, 322–335.
van der Valk, A.G., Middleton, B.A., Williams, R.L., Mason, D.H., Davis, C.B., 1993.
The biomass of an Indian monsoonal wetland before and after being
overgrown with Paspalum distichum. Vegetatio 109, 81–90.
van der Valk, A.G., Volin, J.C., Wetzel, P.R., 2015. Predicted changes in interannual
water-level fluctuations due to climate change and its implications for the
vegetation of the Florida Everglades. Environ. Manage. 55, 799–806.
Wang, G., Middleton, B.A., Jiang, M., 2013. Restoration potential of sedge meadows
in hand-cultivated soybean fields in northeastern China. Restor. Ecol. 21,
801–808.
Wesche, K., Krause, B., Culmsee, H., Leuschner, C., 2012. Fifty years of change in
Central European grassland vegetation: large losses in species richness and
animal-pollinated plants. Biol. Conserv. 150, 76–85.