Evapotranspiration of Pineapple in Hawaii
Paul C. Ekern3
Hawaii Institute of Geophysics, University of Hawaii, Honolulu, Hawaii
The plants of the cerra(lo and caatinga ecological
provinces of Brazil, the suggested original home of
pineapple (3), have many and varied mechanisms
whereby they have adapted to the frequent develop-
ment of moisture stress. The pineapple planit presenits
a peculiarly interestinig combination of those traits
that have been l)ropose(l for xerophytic and sc!ero-
phyllous plants (10, 15). Preferential collapse of
the wvater storage tissue of the leaf in time of mois-
ture stress is further evidence of the ilntriguing adap-
tation of the pineapple p)lant to water (leficit (15. 19).
Reduction in the daytime rate of v-apor exchange
from the pineapple leaf (14) could well be ac-
complished bv the highly cutinized upper epidermis
and the deeply entrenched stomatal pores with an
overlying mat of trichomes on the undersurface.
The very lowN rate of but 0.3 to 0.5 niig of water lost
per Cm112 of leaf surface in anl houtr conitrasts sharply
with the 26 mg from a corn leaf, and the 43 mg from
a cocklebur leaf (23). Such restricted daytime gas
exchange through the leaf requires some compensa-
tory mechanism for a CO., supply for photosynthesis.
The acid metabolism of crassulacean plants has been
shown to provide such a miiechainisml, and it has been
reported specifically for l)ineapple (21. 22).
Materials and Methods
Continuous record was nude of evaporationi from
a class A pan (5)). Bermtuda grass sod, (Cynod.on
dactvlon L. Pers., and smiiootlh cayeinne pineapple,
A4nanas coiniosius L. AMerr., Avere plante(l in \Vahiawa
Loxw Huimic Latosol (26) on hydratulic lysiiietprs (6.
12) at the Pineapple Research Inistitute field station.
Wahiawa (index 820.2, 210 28' N, 217 m elev.) (24).
The lvsimeters were 1.525 mn square ail(l 0.5n ni deep.
Conitinuous record of miioistuire consuimption was miiade
with a float recorder oni the hydrauilic scale of the
lvsimeter. Sulnliglht was recorded by anl Eppley
pyrheliomiieter calibrate(d oni the Smnithsonian scale
(2). Each lvsimiieter has a self-border-planting 5 iin
in wvidth.
Correlative stti(lies of the leaf area in(lices of pine-
ap)l)le l)lants were miade by tracinig onito car(ll)oar(l or
Revised manuscript received January 26, 1965.
2 Pineapple Research Institute Technical Paper No.
308 and Hawaii Institute of Geophysics Contribution No.
82.
3 Formerly Soil Physicist, Pineapple Research Insti-
tute, Honolulu, Hawaii.
736
photographic paper the otutline of leaves strippe(1
from field plants. Leaf area was interpolated from
the accumulated weight of the cardboard. Tin scv-
eral instances the leaf area was also calculated from
the leaf dimensions based upoIn a trapezoidal shape.
Pineapple crowins newrly rooted in wvater ctultture at
the Honolulu glasshouse of the Pineapple Researclh
Institute were used for the collection of the gas
evolved froml the roots of the stunlit planlts.
Results
The fraction of the day s evaporation that oc-
ctirred at night (20: 00 to 08: 00) was nmeasturedl
for the pan, Bermutda grass sod, Latosol, and(i pine-
apple plantings in the spring of 1959. Year-old
potted pineapple plants had been transplanite(d onto a
lvsimeter in October 1958. and were in the red-buid
stage of flowering at the time of these measuiremiients.
The 0.38 fraction of the water use that occturred
at nig-ht for the pineapple was in distinct contrast
to the 0.17 fractioin from the pan, the 0.13 fromii the
sod, and the 0.19 from the Latosol. The pattern
of loss for 2-hour intervals clear!v demiionstrate(d the
suippression of daytimiie transl)iration 1w the pineapple
plants.
TIn August 1959 the diuirnal patternl of the water
uise of the several surfaces again ha(l the sam,e rela-
tive aspect. Sixteen potted pineapple plants about
a year old were set on one of the lxsimeters to ex-
aggerate the effect of the pineapp'e cover, since a
niormiial field would have but '3plants in this area.
The Bermiiudla sod had been clipped short on Auguist
7 after a 10-cm rain hadl thoroughly wNetted tlle
1vsimeter. The pattern of water loss fromii the pan
anid the sod closely followeld stunlight. Nvhereas the
pattern for pineapple inclicated (laytime suppression
of transpiration. On days when the leaf axils of the
pineapl)le were filled with walter fri-om rainfall, somlle
inicrease in the daytimie cons)iniptive use w\as fotlun(l.
More precise (leterminiatioll of the moisture uise pat-
tern of pineapl)le was obtainie(l in the fall of 195(
froml the lysimieter on which 1t pineapple planits grex-
through a complete cover ol 0.038 mmni black poly-
ethylene. Bermuda grass so(d and moist L,atosol
had nearly idenitical use rates with a str-ong- midday
miiaximultm whiclh corresponldedI closely to the p)atterni
of suinlight. h'l'e pinieapple had greatly restricted
daytimiie tranlspiration. The pineapple plants hadl
more nearly complete cover on this lysimeter in
April 1960 than in the fall of 1959, and the sharp
contrast between the use pattern of pineapple anid of
 EKERN-EVAPOTRANSPIRATION OF PINEAPPLE IN HAWAII 737
sod persisted. As soil moisture stress beneath the by the time of 60 % canopy closure, in August 1961.
sod increased, the use pattern of the wilted sod This reduction took place despite a 2-fold increase
gradually approached that of pineapple, with a much in the potential evaporation, as indexed by the rate
reduced rate of loss that was nearly constant day of pan evaporation.
and night. Comparison of the relative use rates of Bermuda
The effect of pineapple growth and the increased grass sod, pineapple plants 8 months old, and pine-
canopy coverage which accompanied that growth was apple plants 24 months old demonstrated the increase
dramantically demonstrated by a series of measure- in suppression of daytime water use as the nontrans-
ments which were begun in October 1960, when a piring pineapple canopy increased (fig 1). The
lysimeter was planted with 9 slips (table I). The moisture use of pineapple after the canopy closed
plantings were through black polyethvlene mulch in the fall of 1961 averaged 0.15 of the energy of
0.038 mm thick and 1 m wide. A 50 % reduction in the incident sunlight.
the daily rate of consumptive use of water occurred Measurement of the water use by the 1960 plant-
ing was continued into 1962 as the fruit matured,
Table I. Fraction of Sunlight Used to Evaporate Water and records for weekly periods (table II) gave what
by a Class A Pan and by Pineapple Planted in
October 1960
- . I.. . . . ,177
Sunlight, Fraction used by .8 L
Date ly/day Pan Pineapple ' SUNLIGHT
i
1961 2C A i \.
Jan 346 0.77 0.32 2
Feb 444 0.56 0.28
Mar 422 0.58 0.24 -2.01 118
Apr 440 0.65 0.20 I- B SOD
May 546 0.70 0.15 .6
June 533 0.62 0.13 6 4 E
N

604 0.67 0.16 E 8 . *-. t PINEAPPLE

July w
Aug 606 0.65 0.13 m .2 CS j °i / (8 mo.)
Sept 511 0.68 0.14 w 1.0 59
Oct 358 0.69 0.22
-

.8
*
Lu 10:o
.*
j
*.2 PINEAPPLE
(24 mo.) -i
z
Nov 306 0.62 ... : .6 Co
U)
Dec 363 0.54 0.15 z
o .4
1962 j:- -.......^-
Jan 349 0.49 ... .2(
Feb 421 0.49 0.19
(gain)
Mar 453 0.58 0.20 Ans
v.Z I 2 ]1
.81 . 6 8202 2
.

Apr 513 0.59 ... 2 4 6 8 10 c

0
14 16 IS 20 22 24 2 4 6 8 10
May 499 0.58 0.19 c
TIME

Jun 568 0.67 0.15 7 JULY, 1961 8 JULY, 1961

July 607 0.72 0.09
Aug 536 0.72 0.09 FIG. 1. Measured consumptive use of water by Ber-
Sept 500 0.77 0.08 muda grass sod and by pineapple 8 and 24 months after
Oct 441 0.67 0.10 planting compared to the pattern of sunlight at Wahiawa,
Oahu.
Table II. Fraction of Sutnliqht Used to Evaporate Watcr by) a Class A Pan and by Pineapple
18 MIonths aftcr Planting During a Period of Increasing Soil Moisture Stress

Sunlight, Fraction used for evaporation Tensiometer, Rainfall,

1962 ly/day Pan Pineapple bars mm
June
2-4 437 0.61 0.19 0.06 0
4-8 594 0.66 0.15 ... 2.38
8-12 595 0.67 0.15 0.12 0
12-18 612 0.69 0.12 0.22 1.52
18-25 560 0.73 0.14 0.50 1.78
25-2 565 0.72 0.16 0.80 5.35
July
2-9 576 0.63 0.09 0.85 12.70
9-16 610 0.69 0.11 .. * 8.13
16-23 624 0.73 0.09 ... 3.81
23-30 653 0.79 0.08 ... 0.76
* Soil moisture beneath plastic: July 13, 0.33 cc/cc = one-third bar stress; July 27, 0.27 cc/cc = 8 bar stress.
738 '11\SIOL,OGY.
PLANTA PHY
Table III. Leaf Area( In;tdice-s for Pinac(ipple
Plant age, Plant weight, Leaf area,
months g cm2
3 2618
3 3037
6 6200
6 7205
11 3182 10,150
8000
12 12,985
12 17,798
19 5000 13,650
23 3382* 9700
23 3294* 10,460
23 3200* 10,250
* Grown at density 2.5-fold conventional practice.
might be evidence of some (lecrease in consuml)iptive
use as the soil moistuire approached the 15 bar wiltng
poinlt.
Peculiar evidenice for the effective closure of the
leaf to gas exchange by day was found in the pro-
duction of gas from the root tips of pineapple plants
growing in water culture. Abel (1) had shown that
on bright days with active photosynthesis, this gas,
collected from intact root tips, contained 30 to 40 %
O., and was free from CO.,. Gas volumes of 13.7 cc
accumulated in a day. The same phenlomiienion was
noted again in 1961. and samples of gas which
evolved from broken roots of slips in water culture
accumuilated as much as cc in an hour at midday.
5 axils
Analysis by gas chromatography indicated the sam-
ples to be 78 % O.,. The evolution of this gas was
closely keyed to sunilight sinice evolution stoppe(d
withini a few minuites after a plant became shaded.
Excised pineapple leaves also produced the gas,
wvhich could be seen escaping from the air channiiels
of the leaf when sunlight struck the leaf. In these
leaves, too, evolution ceased shortly after the leaf
became shaded.
Lest it be thought that the conservative use of
moisture by pineapple planting indicates transpiration
from a limited amount of leaf area, measuremenit
was made of the leaf area inidex (27) (tab'e III).
Leaf area index does niot equal total leaf sturface
but rather only the area of 1 side of the leaf. The
leaf area of 13,000 cm112 for a pineapple plant gave a
comlparatively high leaf area ind(lex of 5.8 for con-a
entional p!anting of 4.3 p)lants p)er s(luare miieter.
Discussion
'I'he initial evalporationi rate of the lysiinieter with
the surface one-half black polyethylene and onie-half
bare Latosol, newly I)lanted to l)ineapple, was one-
third the rate of pan evaporation and bore out the
self-mulching action of the Low Humic Latosol (9).
The reduction to rates one-fifth that of pan evapora-
tion or oiic-sixth of the energy in the sunlight, as the
ILeaf area Index
at 4.3/M2't 10.8X/mn2*
1.1
1.28
2.60
3.02
4.25
3.36
5.45
7.50
6.0
10.40
11.20
11.05
pineapple plant grew to give full vegetative cover,
was a remarkable (leparture. The Bermiuda grass
sod used water that was equivalent to six-tentlhs of
the eniergy of the suniilight, a rate reporte(d to he
typical for convenitional vegetation (25).
The design of the pineapple leaf (15) is ideal for
the suppression of water vapor exchange since the
stomates are entrenched on the unidersurface andl are
covered with a mass of trichomes (direct observation
of the status of stomatal openinigs is thuts precluded).
In addition, the heart-shaped array of the erect
leaves and the 5/13 and N + 1 phyllotaxy of the
leaves concentrates dew and light showers inito the
where the water canl be efficienitly conserved
(16).
The diurnal cvcle of the leaf thickness suiggests
that the plant is milost turgid in the afternooni and
least turgid early in the morning. (Linford, A1. B.
1934. Private publications of the Pineapple Research
Institute.) SUch a patterni is in accord with the
restricted water vapor escape from the leaf by (lay.
A similar reversal of the wsater balance has beeni
reported for opuntia and other stucculent plants
where the rate of water uptake from the soil by day
has exceeded transpiration (17 20).
The design of the leaf array and the persistent,
small, zenith angles of the noonl sun in the subtropics
make the net radiationl hurden on the pineapple very
great (7, 8). The thick suiccnilenit pilleapple leaf
was reported (18) to develop tenlpertures 50 above
air temperatnire. This .5 elevationi of the leaf teni-
peralture is in close accor(l with that forecast b)y
Gates (11) for a nonitranispiring- leaf cooled by conl-
vection forced by a 65() cmn l)er seconld wind tlnder
a 1.1 ly per mlitnuite ra(liationi load when the ambient
air temiiperature is 30(). Tlshis elevate(d leaf temil)era-
ture ani(l a total leaf mass som11e 10-fold that of coni-
ventionial crops miiakes the assnlll)tioll of ne-ligible
heat storage in the planit canopy a mlatter of quiestion-
able validity when an attemnl)t is nlade to strike a
heat balance for the pineapple crop.
The restricted development of the pineapple root
system is yet another phase of the unusual water bud-

EKERN-EVAPOTRANSPIRATION OF PINEAPPLE IN HA\\WAI3
get of the plant. Though there miiay be potential
extension of the root system through a large soil
depth, the actual development of the roots in the
Latosol is confined to the tilled area (4, 13). Evi-
dence of the shallow depth of rooting was foutnid
from the measurements of water withdrawal made
with Boyoucos resistance blocks at Wahiawa in 1955.
No moisture withdrawal by the pinleapple roots was
ilndicated from depths greater than the surface 30
cm of 1.5-m deep percolate lysimeters, whereas mois-
ture withdrawal was indicated for the entire 1.5-ni
deep profile beneath Bermuda grass sod in a similar
lysimeter. This restricted root zone of the pineapple
plant makes only a small soil water reservoir avail-
able to the plant, a feature that would make most
conventional plants extremely susceptible to drought.
Summary
Full pineapple plant canopy effectively suspended
water vapor exchange by midday. This restriction
of the water loss rate was accomplished despite a leaf
area index of nearly 6. The control of transpiration
was a direct physiological function of the pineapple
leaf, which is excellently desiglne(d for the control of
gaseous exchange.
Literature Cited
1. ABEL, F. A. E. 1924. Physiological studies of the
pineapple plant. M.S. Thesis, Univ. Hawaii.
2. ANONYMOUS. 1958. Annals of the IGY 1957-1958.
Part VI. Radiation instruments and measure-
ments. London, Pergamon Press. p 367-467.
3. BAKER, K. F. AND J. L. COLLINS. 1939. Notes on
the distribution and ecology of Antanas and Pseu-
danantas in South America. Am. J. Botany 26:
697-702.
4. BLACK, R. F. 1962. Pineapple growth and nutri-
tion over a plant crop cycle in South Eastern
Queensland. Queensland J. Agri. Sci. 19: 435-51.
5. CAMPBELL, R. B., J. H. CHANG, AND D. C. Cox.
1959. Evapotranspiration of sugar cane in Hawaii
as measured by infield lysimeters in relation to
climate. Proc. 10th Cong. Intern. Soc. Sugar
Cane Technologists. p 637-44.
6. EKERN, P. C. 1959. Evapotranspiration patterns
under trade wind weather regime on Central Oahu,
Hawaii (Abstract). Agronomy Abs. p 4.
7. EKERN, P. C. 1962. Net radiation balance over
pineapple plants on Oahu, Hawaii (Abstract). J.
Geophys. Res. 67: 3532.
8. EKERN, P. C. 1965. The fraction of sunlight re-
tained as net radiation in Hawaii. J. Geophys.
Res. 70: 785-93.
739
9. EKERN, P. C. 1965. The disposition of inet radia-
tion over bare Low Humic Latosol in Hawaii. J.
Appl. Meteorol. (in press).
10. FERRI, M. G. 1961. Problems of water relations
of some Brazilian vegetation types vith special
consideration of the concepts of xeromorphy and
xerophytism. Proc. Madrid Symp. Arid Lands
Res. 16: 191-97.
11. GATES, D. M. 1964. Leaf temperatures and trans-
piration. Agron. J. 56: 273-77.
12. GLOVER, J. AND J. A. FORSGATE. 1962. Measure-
ment of evapotranspiration from large tanks of
soil. Nature 195: 1330.
13. GwYNNE, M. D. 1962. Root systems of pineapple
plants. East African Agri. Forestry J. 27: 204-
06.
14. KRAUss, B. H. 1930. The transpiration of pine-
apple. M.S. Thesis, Univ. Hawaii.
15. KRAUss, B. H. 1949. Anatomy of the vegetative
organs of the pineapple Antanas comnosus L. Merr.
II. The leaf. Botan. Gaz. 110: 333-404.
16. LEOPOLD, L. B. 1952. Dew as a source of plant
moisture. Pacific Sci. 6: 259-61.
17. MAcDOUGAL, D. T. 1936. Dendrographic meas-
urements of the tree cactus (Caritegiea gigaittea).
Carnegie Inst. Wash. Publ. 462: 174-94.
18. NOFFSJNGER, T. L. 1961. Leaf and air tempera-
ture under Hawaiian conditions. Pacific Sci. 15:
304-06.
19. SANFORD, W. G. 1962. Pinieapple crop log-conicept
and development. Better Crops with Plant Food
46(3): 32-43.
20. SCHREVE, E. B. 1916. AAn analysis of the causes of
the variation in the transpiring power of cacti.
Physiol. Researches 2: 73-127.
21. SESHAGIRI, P. V. V. AND A. SURYANARAYA-MURTHY.
1957. Studies in the organic acid metabolism of
Aiaizas sativa. J. Indian Botan. Soc. 36: 25461.
22. SIDERIS, C. P., H. Y. YOUNG, AND H. H. Q. CHUNN.
1948. Diurnal changes and growth rates as asso-
ciated with ascorbic acid, titratable acidity, carbo-
hydrate and nitrogenous fractions in the leaves of
A;zanaas comIostus L. Merr. Plant Physiol. 23: 38-
69.
23. SPECTOR, W. S. 1956. Handbook of biological data.
Wright Air Develop. Center Tech. Rept. 56-273,
Table 300, p 336.
24. TALIAFERRO, W. J. 1961. A key to climatologic ob-
servations in Hawaii. U. S. Dept. Comm. Weath-
er Bur. Documentation No. 1.11. 118 pp.
25. TANNER, C. B. AND E. R. LEMON. 1962. Radiant
energy used in evapotranspiration. Agron. J. 54:
207-11.
26. U. S. DEPT. AGRI. 1955. Soil survey of the terri-
tory of Hawaii. (Soil Survey Series 1939 No. 25.)
Washington, D. C. p 207-313.
27. WATSON, D. J. 1947. Comparative physiological
studies in the growth of field crops. Ann. Botany
N.S. 11(41): 41-76.