Photosynthetic Gas Exchange and Water Relation Responses of Three Tallgrass Prairie Species

to Elevated Carbon Dioxide and Moderate Drought

Author(s): Erik P. Hamerlynck, Christine A. McAllister, Alan K. Knapp, Jay M. Ham, Clenton
E. Owensby
Source: International Journal of Plant Sciences, Vol. 158, No. 5 (Sep., 1997), pp. 608-616
Published by: The University of Chicago Press
Stable URL: http://www.jstor.org/stable/2474921 .
Accessed: 02/08/2011 00:36

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .
http://www.jstor.org/action/showPublisher?publisherCode=ucpress. .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to
International Journal of Plant Sciences.

http://www.jstor.org
Int. J. Plant Sci. 158(5):608-616. 1997.
C) 1997 by The Universityof Chicago. All rightsreserved.
1058-5893/97/5805-0015$03.00

PHOTOSYNTHETIC
GASEXCHANGEANDWATER
RELATION OFTHREE
RESPONSES PRAIRIE
TALLGRASS
TOELEVATED
SPECIES CARBON ANDMODERATE
DIOXIDE DROUGHT
ERIK P HAMERLYNCKI,* CHRISTINE A. McALLISTER,* ALAN K. KNAPP,* JAY M. HAM,t AND CLENTON E. OWENSBYt
*Division of Biology, and tDepartmentof Agronomy,Kansas State University,Manhattan,Kansas 66506

Undisturbedtallgrassprairiewas exposed to ambientand elevated (twice-ambient)levels of atmosphericCO2 and

experimentaldryperiods.Seasonal and diurnalmiddayleaf waterpotential(Tleaf), netphotosynthesis (Ane),and stomatal
conductance(g) responses of threetallgrassprairiegrowthforms-a C4 grass,Andropogongerardii; a broad-leaved
woody C3 shrub,Symphiocarposorbiculatus;and a C3 perennialforb,Salvia pitcheri-were assessed. kleafin A. gerardii
and S. orbiculatuswas higherunderelevated C02, regardlessof soil moisture,while Tleaf in S. pitcherirespondedonly
to drought.Elevated CO2 always stimulatedAne,in the C3 species, while A. gerardii Anet increased only under dry
conditions.However,Anetunder elevated CO2 in the C3 species declined with droughtbut not in the C4 grass. Under
wet conditions,g5reducedin elevated CO2 forall species. During dryperiods,g5at elevated CO2 was sometimeshigher
than in ambientCO2. Our resultssupportclaims thatelevated CO2 will stimulatetallgrassprairieproductivity during
dry periods and possibly reduce temporaland spatial variabilityin productivity
in these grasslands.

Introduction woody perennials,increased markedlyunder elevated

IncreasingatmosphericCO2 is expected to strongly
stimulatethe productivityof the humidtemperatetall-
grass prairie of centralNorth America (Parton et al.
1994). This is in part due to improvedseasonal water
status accompanying lower stomatal conductances
(Knapp et al. 1993b; Owensby et al. 1993b; Knapp et
al. 1994a; Parton et al. 1994; Ham et al. 1995), re-
duced photoinhibition, and increased CO2 fixationby
ribulose-1,5-carboxylase/oxygenase (Long 1991). The
latter could be importantfor C3 plants growing in
grasslandsdominatedby C4 grasses. C3 grasslandspe-
cies, thougha small portionof totalbiomass, comprise
most of the species diversity,spanninga wide range
of growthformspossessing physiological,morpholog-
ical, and phenological adaptationsforpersistencein a
C4 "sea of grass" (Fahnestock and Knapp 1994; Tur-
ner et al. 1995). This diversityof plantformand func-
tion could lead to shiftsin grasslandcommunitystruc-
ture, and subsequent function,in an elevated CO2
world (Owensby et al. 1993b).
An ongoing 6-yr studyon the impacts of elevated
CO2 on intacttallgrassprairiehas shown thatCO2 en-
richmentstimulatesbiomass accumulationmorein dry
years compared with wet years, a resultof improved
water statusand highersustainedphotosynthetic rates
in theC4 dominantgrass,Andropogongerardii(Knapp
et al. 1993b, 1994a, 1994b; Owensby et al. 1993b;
Ham et al. 1995). Even in a wet year,elevated CO2
reduced whole canopy evapotranspirationby 26%
(Ham et al. 1995) and stem sap flowin A. gerardiiby
18% (Bremeret al. 1996). In wet years,totalbiomass
was similar between elevated and ambient CO2, but
C3 species biomass, especially herbaceous forbs and
IAuthor for correspondenceand reprints.Currentaddress: De-
partmentof Biological Sciences, Universityof Nevada, 4504 Mary-
land Parkway,Box 454004, Las Vegas, Nevada 89154-4004; E-mail
EPH @nscee.nye.edu.
Manuscript received February 1997; revised manuscriptreceived
April 1997.
CO2 (Owensby et al. 1993b). Several greenhouseand
growth chamber studies implicate that C3 grassland
and grassland/steppeplants will replace C4 species,
thoughhighertemperatures and waterstresscan result
in strongerpositive effectsof elevated CO2 on C4 pho-
tosynthesis and productivity(Carter and Peterson
1984; Smith et al. 1987; Johnsonet al. 1993; Morgan
et al. 1994; Read and Morgan 1995; Hunt et al. 1996).
However,the interactionof droughtand elevated CO2
on photosynthesisin native C3 tallgrassprairieplants
in an intactC4 grasslandis notwell understood.Knapp
et al. (1996) foundthatstomatalconductanceand wa-
ter relationsresponses to seasonal water stressunder
elevated CO2 were species specific,even withincertain
growthforms.Early in the growingseason, all species
had lower stomatalconductancesand higherwaterpo-
tentialsunder elevated CO2. In drierlate-season con-
ditions,only a C4 grass and a C3 forbmaintainedthis
pattern(Knapp et al. 1996). However, the degree to
which this improvedcarbon uptake is not known.
Here, we presenta studyassessing the gas exchange
and water relationsresponses of threeimportanttall-
grass prairieplant growthforms,a fibrousrooted C4
grass (big bluestem,Andropogon gerardii), a deeply
rootedC3 perennialforb(pitchersage, Salvia pitcheri),
and a shallowly rooted C3 broad-leaved woody shrub
(buck brush,Symphiocarposorbiculatus) to elevated
CO2 and experimentaldroughtin intacttallgrassprai-
rie. These species have distinctabove- and below-
ground physiognomies (Weaver 1954) and differin
photosyntheticcapacity and seasonal water relations
(Fahnestock and Knapp 1994; Turner et al. 1995;
Knapp et al. 1996). We predictedthatcarbon enrich-
mentwould have greaterbenefitsforthe C3 species in
moist conditions,while, in drier conditions,we ex-
pected greater benefitsfor the C4 dominant. These
findingscould be relevantto predictingshiftsin spe-
cies productivity patternsthatcould alterspecies com-
position,potentiallyimpactingecosystemprocesses in
tallgrassprairie.

608
HAMERLYNCK ET AL.-TALLGRASS PRAIRIE PLANT
Material
andmethods
STUDY SITE AND DRY DOWN REGIME
Research was conductedin native tallgrassprairiein the
Flint Hills of northeastKansas near Manhattan,Kansas (lat.
39.12?N, long. 96.35?W, 385 m above sea level). This por-
tion of undisturbedtallgrassprairiehad been exposed to am-
bient (350-360 pL L-1) and twice-ambient(700 pL L-')
CO2 in four large (4.5-m diameter,4.0-m height) open-top
chambers (OTCs) for 4 yr at the time of this study.OTCs
were subjected to three drying/wetting cycles early (DOY
[day of year] 161-193), mid (DOY 198-221), and late
(DOY 198-221) in the 1995 growingseason. Dryingcycles
were initiatedby shuttingoffan irrigationsystemnormally
used to returnprecipitationinterceptedby the OTC roofinto
the chamber.A high-speed fan used to maintain OTC at-
mosphericpressurealso reduced precipitationinputinto the
chambersas well. Subsurfaceshielding(aluminumflashing)
buriedto 0.9 m blocked subsurfacewaterinfiltration, a prob-
lem in previous experiments(Nie et al. 1992). This protocol
reduced precipitationby 98%-99% in the OTCs (J.M. Ham,
unpublished data). After a minimumdry period of 14 d,
plots were monitoredfor visible signs of plant stress (e.g.,
leaf curling),then irrigatedfor 2 d. There were two water-
ings, one on July 13-14 (9.7 cm) and the otheron August
10-11 (7.2 cm). The chamberswere leftundisturbedfortwo
additionaldays, and anotherdryingcycle was initiated.
LEAF-LEVEL MEASUREMENTS
GAS EXCHANGE. Seasonal and diurnal leaf-level photo-
syntheticgas exchange measurementswere made on a C4
grass (big bluestem,Andropogongerardii Vit.), a perennial
C3 forb (blue pitchersage, Salvia pitcheri Torr.),and a C3
woody shrub (buckbrush, Symphiocarpos orbiculatus
Moench.). These species were selected as representative
growth forms in tallgrass prairie and occurred frequently
enough in the OTCs for replicate sampling.Photosynthesis
(Anet) and stomatal conductance (gs) were estimatedin the
fieldwith a closed-systemportableIRGA (LiCOR Li-6200)
using the equations of von Caemmererand Farquhar(1981).
Leaf temperaturewas measured with a fine-wiretype-T
(copper/constantan) thermocouplepressedto theabaxial side
of the leaf. Photosyntheticphoton flux density (PPFD in
pLmol m-2 s-') was measured with a silicon photodiode at-
tached to the outsideof a 0.25-L cuvette.Leaves were main-
tained as close to natural orientationas possible. The en-
closed leaf was held in the cuvette for a 10-s sampling
period, which usually resultedin a 2-5-puL L-' drawdown
of CO2. Only samples with<2% change in cuvetteRH were
analyzed to ensure accuracy of gs estimations.Four plants
of each species in each CO2 treatment-chamber and from
adjacent field controlplots were sampled for seasonal and
diurnalcourses of gas exchange. Seasonal gas exchange dy-
namics were compiled frommiddaymeasurementsmade ca.
every 3 d fromJune 9 to September 12, 1995, in ambient
light conditions. Diurnal gas exchange curves were made
early and late (DOY 165 and DOY 193) in the early-season
dryingcycle.
WATER RELATIONS. Seasonal midday (ca. 1130 to 1400
CentralDaylightTime [CDT]) and diurnalleaf waterpoten-
tial (Tleaf) was measuredfromA. gerardiiin the same cham-
bers and fieldplots sampled forgas exchangemeasurements.
Seven mature leaves were harvested,placed into sealable
plastic bags, and transportedto the lab for determination
of
RESPONSES TO ELEVATED CO2 AND DROUGHT 609
tIeafwith a Scholander-typepressure bomb (PMS Model
1000, Corvallis, Oreg.). This samplingprotocol allowed for
rapid processing of a large numberof samples gatheredin
the fieldacross a shortperiod (ca. 10-15 min) and avoided
diurnaleffects(Fahnestock and Knapp 1994). Tleaf was es-
timatedca. every 3 d throughthe season. Diurnal Tleaf was
trackedearly (DOY 158, 2 d beforestarting)and late (DOY
193, +33 d into) the early-seasondryingcycle. There were
not enough individuals of the othertwo species to measure
seasonal P,eaf; only measurementspooled fromthe startand
end of the early- and midseason dryingcycles were made
for S. pitcheriand S. orbiculatus. Four leaves of each two
species in each chamberthese species were collected,giving
a total sample size of eight.
STATISTICAL ANALYSIS
A repeated-measures,split-plot,two-wayANOVA (Gen-
eral ANOVA/ANCOVA, Statistixv4. 1, AnalyticalSoftware,
St. Paul, Minn.) was used to detect combined drought/CO2
treatment effectswithineach species. Data were pooled from
three sampling days at the start(wet treatment)and 3 d at
the end (dry treatment)of each of the threedryingcycles
(early-,mid-,and late-season),givinga whole-plottreatment
structure of wet/elevated,
wet/ambient, dry/elevated,dry/am-
bient,and field control,with dryingcycle and the cycle X
treatmentinteractionas subplot factorsfor ANOVA. Since
the same plots were measured for each sampling date, this
is a repeated-measuresdesign with four and seven replica-
tions forgas exchange and tleaf measurements,respectively.
Also, dryingcycles reflectedseasonal changes in plant per-
formance.The whole-ploterrortermforF-testswas therep-
lication X treatmentinteraction,withthe subplot-error term
being the date X treatmentX replicationinteraction.A prob-
abilitylevel of 0.05 was considered significant,withmeans
separationby LSD. Since no rainoutshelterswere construct-
ed to provide a balanced factorial3 X 2 treatmentstructure,
CO2 treatmentmain effectswere tested by post hoc gener-
alized contrastswith Scheffe's F-test using the whole-plot
errorterm(Statistixv4.1).
Results
SEASONAL DYNAMICS
GAS EXCHANGE. Net photosynthetic rates (Anet)un-
der elevated CO2 were similar between wet and dry
periods in Andropogongerardii,while Anetin ambient
CO2 OTC plants declined significantly by 23% (fig. 1;
table 1). In elevated CO2, Ane,in OTC Symphiocarpos
orbiculatus declined by 25% in response to experi-
mental drought,while Salvia pitcheri showed a 20%
decrease in assimilationrates(fig. 1; table 1). Drought-
induced declines in Anetwere most pronouncedin am-
bient OTC S. orbiculatus,reducingby 61%. Stimula-
tion of Anetby elevated CO2 was greatest in S.
orbiculatus in dry conditions, which increased by
192%, compared with a 50% stimulationunder moist
conditions.Droughtwidened the differencein Anet be-
tween elevated and ambient A. gerardii, increasing
from 6% to 40% higher photosyntheticrates going
fromwet to dry periods (fig. 1). Anetin elevated CO2
S. pitcheriincreased only slightlyover ratesexpressed
in ambientCO2, with a 28% in wet and a 33% stim-
ulationin dryconditions,respectively.There were un-
expected increases in A . in A. gerardii and S. pitcheri
610 INTERNATIONAL JOURNAL OF PLANT SCIENCES

30
A. gerardii W ELEVATED
Table1
25 25 - 2772 ~~~~~~~~~AMBIENT_
ANALYSIS OF SEASONAL PHOTOSYNTHETIC GAS EXCHANGE AND WATER
a
b a FIELD
ab
20 -a20~~~~~~
b b a: RELATIONS RESPONSES OF THREE TALLGRASS PRAIRIE SPECIES
~ ~
|
b~ b TO COMBINED C02 AND DROUGHT TREATMENTS

15
Species and Signi-
effect MS F df ficance
AG:
Anet
dropogon grd,
orbiculatusC
gaS.
Treat........ 23.9 36.5 4,15 *
10
E
I
aa Cycle....... 619.3 40.1 2, 150 *
T X C ...... 108.6 6.7 8, 150 *
E~~~~~~~~~~~~~~~~~~~~~~~~~~
shrub,d S[avi aa amb
Treat ........ 1.1 X 105 13.5 4, 15 *
Cycle........ 6.9 X 103 9.3 2, 150 *
WET
DY ~~~ WTDY ET aR WE R T x C ...... 3.4 X 103 4.6 8, 150 *
4tieaf
Treat ........ 7.15 94.6 4, 65 *
Cycle........ 17.0 145.8 2, 550 *
T x C ...... 0.4 3.4 8, 550 *
SO:
Anet-
shu,adSli6 icei Cb fob, InevadadamitCO Treat........ 202.1 43.0 4, 15 *
opntp(T)cabr0ubetdt rigcce.basidct
? S( 1),letrsdifr igifcnty ihi echpne a Cycle........ 123.3 21.5 2, 150 *
<0.05 (LSD).~~~~~~~~b T X C ...... 7.6 4.7 8,150 ns
9":
Treat........ 2.2 x 105 5.5 4, 15 *
Cycle........ 3.3 X 103 3.3 2, 150 *
T X C ...... 3.5 X 103 1.7 8, 150 *
F 1 LE af-level,et photosynteic gas exan (ane of A SP:
gerardii5
a(CLSgrass,DSymphiocarpo
dropogon rl Anet
Treat ....... 364.6 43.0 4, 15 *
Cycle........ 431.3 21.5 2, 150 *
T X C ...... 93.9 4.7 8, 150 *
os:
Treat....... 4.0 X 105 5.5 4, 15 *
Cycle....... 2.2 X 106 3.3 2, 150 *
T X C ...... 4.6 X 106 6.7 8, 150 *
during the dry portion at the end of the midseason
cycle, especially in ambientOTC A. gerardii (fig. 1), Note. AG = Andropogongerardii (C4 grass), SO = Symphiocar-
pos orbiculatus (C3 shrub),and SP = Salvia pitcheri(C3 forb).An
which led to significanttreatmentX season interac- asteriskindicates significanceat P < 0.05. Means comparisonsfor
tions in Anet forthese species (table 1). Only in S. or- each parametercan be foundin figures1, 2, and 3.
biculatus did C02/moisturetreatmenteffectson Anet
remain constantthroughthe growing season (fig. 1;
table 1). Pooled across treatment combinationsand the bientOTC S. orbiculatus(ca. -35%), while g, did not
season, photosyntheticrates were higher in elevated change underelevated CO2. Comparingg, betweenel-
thanambientCO2 OTC plantsforall threespecies (ta- evated and ambientCO2 S. orbiculatus,we foundthat
ble 2). Anetwas usually highestin fieldplants,withthe g, was 2% higherin ambientCO2 plants,while in wet
exception of S. pitcheri,in which Anet was similarbe- conditions,gs was 20% lower. In S. pitcheri,gs was
tween elevated CO2 OTC and fieldcontrols(table 2). significantlylower (48%) under elevated CO2 com-
There were significanttreatmentX season interac- pared with ambient CO2 during favorable moisture
tions on stomatalconductances(gs) in all threegrowth conditionsbut was not underdrought(fig.2). Overall,
forms(fig. 2; table 1). For A. gerardii and S. pitcheri, gs was significantlylower underelevated CO2 thanam-
gs remainedconstantbetween wet and dry conditions bient CO2 OTC or field controlsonly in A. gerardii.
but reduced significantly in ambientOTC plantsunder Elevated and ambientOTC S. orbiculatushad similar
droughtconditions.However, at the end of the mid- gs, which were both lower thanfieldcontrols,while in
season dryingcycle, gs in these species was markedly S. pitcheri,only elevated CO2 OTC plants had gs sig-
higherthan gs shortlyafterwatering,especially in S. nificantlylower than those in field conditions (table
pitcheri(fig. 2). Droughtreduced S. orbiculatusgs in 2).
the early-seasonand midseason dryingcycles but did
not at midseason (fig. 2). Pooled across the season, WATER RELATIONS. Leaf waterpotentials(Mleaf) re-
droughtdid not reduce gs in eitherCO2 treatmentin duced less in response to dryingin A. gerardii in el-
A. gerardii but did narrowthe reductionbetween el- evated (- 19%) than in ambient CO2 (-27%) OTC
evated and ambientCO2 treatments, from48% lower pooledacrosstheseason(fig.3; table1). Tleaf in OTCs
in elevated CO2 in wet conditionsto a -27% differ- were higherthanfieldcontrolsafterwatering,but only
ence in dry (fig. 2). Droughtreduced gs most in am- lowerthan
ambientOTC plantshad Tleaf consistently
HAMERLYNCK ET AL.-TALLGRASS PRAIRIE PLANT RESPONSES TO ELEVATED C02 AND DROUGHT 611

2
Table
350 A. gerardi a a AMBIENT
GENERAL CONTRASTS OF MAIN EFFECTS OF ELEVATED CO2OTC (E),
300 -
b
~~~~~~~ba FIE
a
AMBIENT CO20TC (A), AND FIELD CONTROL (F) PLANTS ON a
250 ab
GAS EXCHANGE OF THREE TALLGRASS PRAIRIE SPECIES
200 b b

Species and Signi- 150

contrast Ss F ficance Outcome
AG:
a
ab
Anet

E vs. A ....... 220.4 7.4 * E> A E450" ,._.Sb

S. orbiculatus
E vs. F........ 231.4 7.8 * F> E
A vs. F ....... 747.2 25.2 * F> A 350
Ds-
E vs. A ....... 2.8 X 105 8.8 * E < A
E vs. F........ 2.8 X 105 8.7 * E< F 300
A vs. F ....... 9.1 X 103 0.3 ns A = F
SO:

E
ab
Anet

E vs. A ....... 339.1 36.1 * E> A 1500

sia2500
d
0 b a a
E vs. F........ 42.6 4.5 * E < F
A vs. F ....... 465.2 49.5 * A < F 500d
9s: 0
E vs. A ....... 2.0 X 104 0.9 ns E= A
E vs. F........ 7.3 X 105 31.2 * E < F
A vs. F ....... 5.5 X 105 23.3 * A < F
pichr in elevated n min O2
picher S. pitchtedanamin Cepsdtodyn
OOTexoetodygccls yls
SP:
Anet:

E vs. A ....... 769.0 22.4 * E> A signifcantl

at P 0.05(LSD)
E vs.F ........ 8.7 0.3 ns E= F
A vs. F ....... 382.2 11.3 * A < F
9s -
E vs. A ....... 6.1 X 106 2.1 ns E= A
E vs. F........ 1.4 X 107 4.9 * E < F
A vs. F ....... 3.1 X 106 1.1 ns A= F

Note. AG = Andropogongerardii (C4 grass), Symphiocarposor-

biculatus (C3 shrub),and SP = Salvia pitcheri(C3 forb).An asterisk
indicates significanceat P < 0.05.
CO2 was similarto those in wet soils, but in ambient
CO2, g, was reduced to levels closer to those in ele-
vated CO2.
field controlsin dry periods (fig. 3), which led to a Symphiocarpos orbiculatus Anetwas consistently
significanttreatmentX season interaction(table 1). higher in elevated than in ambient CO2 during wet
Compared with ambientCO2 OTC, Tleaf in A. gerardii conditions,with the lattersimilar to Anet in elevated
were ca. 11% higher,regardlessof droughttreatment CO2 in drought(fig. 4). During dry periods,Anetwas
(fig. 3). Water potentials for S. orbiculatus and S. 300%-1200% higherthroughthe day compared with
pitcherideclined significantlyfromwet to dry condi- ambient CO2 (fig. 4), which was near zero through
tions,reduced by ca. 26% in the formerand 37% in much of the day (fig. 4). This decline was accompa-
the latter,regardless of CO2 treatment(table 3). For
bothof thesespecies,lpleaf was higherunderelevated EARLYSEASON MIDSEASON LATESEASON POOLED
CO2 conditions(ca. 17% in S. orbiculatus,6% forS. WET DRY WET DRY WET DRY WET DRY
pitcheri) than ambient CO2 OTC, thoughthis differ-
ence was significantonly forS. orbiculatus(table 3). -D.5

DIURNAL DYNAMICS
GAS EXCHANGE. Anetin A. gerardii was generally
slightlyhigher(in elevatedthanin ambientCO2 during c)
-1.0 _ b
wet conditions; fig. 4), reaching midday highs (ca.
16.0-20.0 Rmol m-2 s-1) between 1200 and 1400 bc c~~~~ c

CDT. Anetin elevated CO2 on a dry day was consid- a ELEVATED-

erablyhigher(25%-100%) throughthe day compared AMBIENT
-3.0 FIELD
with ambientCO2 (fig. 4). Anetin ambientCO2 OTC ardiiinelevate bn bmin OTsepsd bo drigcce.Bar

varied littlethroughoutthe day, rarelyreachingabove -3.5

8.0 Rmol m-2 s-1. Stomatal conductances in A. ger-
ardii (fig. 5) showed the greatestreductionsin g, be-
tween elevated and ambientCO2 treatments occurring
in wet soil conditions.In dryconditions,g, in elevated
Fig.3 Seasonal middaywaterpotential(T,,,,f)ofAndropogonger-
ardii in elevated and ambientOTCs exposed to dryingcycles. Bars
indicate + 1 SE of the mean (n = 14); letterswithineach panel
differsignificantly at P < 0.05 (LSD).
612 INTERNATIONAL JOURNAL OF PLANT SCIENCES

Table
3 nied by markedreductionsin g, (fig.5). The difference
POOLED MIDDAY WATER POTENTIAL MEASUREMENTS in MPa) OF
(4JjIeaf
in g, betweenelevated and ambientCO2 was morepro-
TWO TALLGRASS PRAIRIE SPECIES EXPOSED TO ELEVATED AND nounced in dry conditions,with gs in ambient CO2
AMBIENT CO2 CONCENTRATIONS AND MILD WATER STRESS oftenlower than in elevated CO2.
Symphiocarpos Moderate droughtreduced the marked diurnal re-
orbiculatus Salvia pitcheri sponse of A,,,in S. pitcheriin wet soil (fig.4), though
CO2
in both conditions,Anet in elevated and ambientCO2
treatment Wet Dry Wet Dry
OTC were quite similar.Anetenhancementin elevated
Elevated ... -1.69 -2.15 -1.09 -1.46 CO2 was highest (25%-50%) during dry conditions
(0.14)a (0. 11)b (0.08)a (0.06)b (fig.4). Stomatalconductanceswere lower in elevated
Ambient... -2.07 -2.56 -1.14 -1.59
and ambient CO2 during drought,with more pro-
(0.08)b (0. 16)c (0.08)a (0.08)b
nounced midday stomatal closure in elevated CO2
Note. Each value is the mean of four measurements,pooled treatments (fig.5). Unlike theotherC3 species, drought
across two repeatedmeasures.Lettersdiffersignificantlywithineach did not increase the differencebetween elevated and
species at P = 0.05; SE of the mean is in parentheses.
ambientCO2 gs (fig. 5).

WATER RELATIONS. Duringwet periods,A. gerardii

changedless on a diurnalbasis in elevatedCO2
tleaf
and remainedconsistently
above Tleaf in ambientCO2
30 , 15 l l l l 35

--- ELEVATED/ WET S. orbiculatus

-0- AMBIENT/WET 30 S. pitcheri
-
25 -4-- ELEVATED/IDRY 12 30S-ithr
-
-4-1-AMBIENT/IDRY

Cl) ~~~~~~~~~~~~~~~~~~~~25-
cN 20 9

E 20 -

E 15-

10 1011121314151617
3 8 91011121314151617
8 91011121314151617
A. gerardii s

0 a3

l
-150 3llllllllll10
-2
0
1200 -75-
100

z ~~~~~~~~~~~~1000 50-
w

50 60-

o H 4~~~~~~~~~~~~00 -25-
0 <

-100 -.-- DRY~~~400- -5

>~~~~~~~~~~
-75-

-150 -200 100L-

IO [-. I I I I I
8 9 1011121314151617 8 9 1011121314151617 8 9 1011121314151617

CDT(h)
Fig.4 Top row, Diurnal responses of net photosynthesis(Ane,) of Andropogongerardii (C4 grass), Symphiocarposorbiculatus(C3 shrub),
and Salvia pitcheri (C3 forb) in elevated (squares) and ambient(circles) CO2 OTC at the start(wet; emptysymbols) and end (dry; filled
symbols) of a dryingcycle. Each point is the mean of fourmeasurements;bars indicatemaximumSE of the mean. Bottomrow,Percentage
change of Ane, in elevated CO2 compared with ambient CO2 OTC plants in wet (empty symbols) and dry (filled symbols) soil moisture
conditions.
HAMERLYNCK ET AL.-TALLGRASS PRAIRIE PLANT RESPONSES TO ELEVATED CO2 AND DROUGHT 613

350 I I l 350 T 1200

-0- ELEVATED/WET S. orbiculatus S. pitcheri
-)- AMBIENT/WET
300 -
-- DRY
,ELEVATED/ 3
300 1ooo
-0-- AMBIENT/I
DRY (h)

250 -A. gerardii 250 - (

800-
C,)

IE 200 -200-
7F) ~~~~~~~~~~~~~~~~~~~~600
E 150 150-

400-
100 - - ~~~~~~100
200-
50 50

0 C 0
40-
80-
200-
20 ----WET 60-
F- -4-- DRY
w 150 -4
LLJ 0
im~~~~~~~~~~~~~~~~~~~~~2
Z < ~~~~~~~~~~~100
LLo -20 0
LLD 5

-40 ~~~~~~~~~0-20

-60

-80 -50
-80

-1001 -IOC -IOC

8 91011121314151617 8 91011121314151617 8 91011121314151617

CDT(h)
Fig.5 Top row,Diurnal responsesof stomatalconductance(gs) of Andropogongerardii (C4 grass), Symphiocarposorbiculatus(C3 shrub),
and Salvia pitcheri (C3 forb) in elevated (squares) and ambient(circles) CO2 OTC at the start(wet; emptysymbols) and end (dry; filled
symbols) of a dryingcycle. Each point is the mean of fourmeasurements;bars indicatemaximumSE of the mean. Bottomrow,Percentage
change of gs in elevated CO2 compared with ambient CO2 OTC plants in wet (empty symbols) and dry (filled symbols) soil moisture
conditions.

showed
chambers(fig. 6). Both chamberedtreatments specific seasonal dynamics.For the C4 grass,A. ger-
lowerTleaf laterin the afternoon
markedly compared ardii, elevated CO2 almost completely counteracted
with field water potentials.On a dry day, after1000 droughtstress,which resultedin markedlyhigherAnet
CDT, Tleaf in elevatedCO2 remainedat about -1.8 and Tleaf and lower g, in elevated CO2 compared with
MPa throughout mostoftheday,whileTleaf in ambient plantperformancein ambientCO2. This supportsfind-
CO2 OTC was reduced from -2.0 to -2.5 MPa (fig. ings showing that elevated CO2 can stimulatephoto-
6). Overall, elevated CO2 resultedin a 10%-20% in- synthesisin C4 grasses, especially in water-limited
of soil waterstatus(fig.6).
creasein Tleaf, regardless conditions(Reichers and Strain1987; Wrayand Strain
1987; Knapp et al. 1993b, 1994b; Cheng et al. 1994;
Discussion Morgan et al. 1994).
The dryingtreatments in thisstudyresultedin mod- Our findingsalso supportclaims thatelevated CO2
eratewaterstress,withmiddaywaterpotentialsin An- will stimulatetallgrassprairieproductivityin dryyears
dropogon gerardii similar to past seasonal drying (Knapp et al. 1993b; Owensby et al. 1993a) or prolong
trendsand much higherthanthose experiencedduring the growing season by alleviating late-season water
severe drought(Knapp 1985; Knapp et al. 1993a). stress(Knapp et al. 1996). Elevated CO2 could reduce
Photosyntheticgas exchange and water relationsre- intra-annualvariability of productivityin tallgrass
sponses to elevated CO2 and this moderate drought prairieby stimulatingproductivityin chronicallywa-
treatmentwere stronglyinfluencedby growthform ter-limitedareas of tallgrassprairie such as annually
614 INTERNATIONAL JOURNAL OF PLANT SCIENCES

O I I I I I i I
changeand Tleaf measurements
indicatethatthisshrub
WET was mostresponsiveto rewatering.Thus, elevatedCO2
co may most benefitwater status of shallowly rooted,
-J
broad-leavedC3 species (Knapp et al. 1996).
Elevated CO2 offsetdroughteffectsto a lesser ex-
<a--ELEVATED tent in Salvia pitcheri, a C3 species of considerably
AMBIENT
Z -2
-
_ FIELD greater photosyntheticcapacity than S. orbiculatus
LUJ
(Turneret al. 1995). The late-season increases in g,
0 later may coincide with floweringor be due to leaf
0L
i -3 a l l l l l age-specific increases in photosynthetic capacity
(Harper 1989). Our resultsalso suggestthatS. pitcheri
LUJ
has access to deeperwaterreservesthanS. orbiculatus.
Midday water potentialswere markedlyhigherin S.
-LJ
pitchericompared with the shrub,S. orbiculatus,and
did not respond to elevated CO2, while mild drought
produced a relativelysmall, albeit significant,reduc-
-2- tionin Tleal. Thus,elevatedCO2 maynotas strongly
affectplantwaterstatus,and subsequentgas exchange
performance,in this deep-rooted, drought-resistant,
-3 l l l
long-lived perennial C3 forb (Owensby et al. 1993b;
Fahnestock and Knapp 1994; Turner et al. 1995;
20 - t WET Knapp et al. 1996).
zDR
->Flo r We foundlimitedevidence thatdirectphysiological
LU
0
effectsof elevated CO2 will lead to rapid changes in
LLQ
the structureof this C3/C4 system.Our resultsindicate
wo thatC4 tallgrassprairiegrasses receive consistentben-
efits from atmospheric carbon enrichment,mainly
W-20-
throughimproved water status, and could continue
dominatingthese systems.Elevated CO2 did alleviate
-30 i l Il i droughteffectscompared with ambientCO2 in the C3
4 6 8 10 12 14 16 18 20 plants studied,especially for the woody shrub,S. or-
CDT (h) biculatus. Biomass data has repeatedlyshown increas-
es in C3 biomass in tallgrass prairie exposed to ele-
Fig.6 Top row, Diurnal responses of leaf xylem waterpotential vated CO2 (Owensby et al. 1993b; J. M. Ham,
of Andropogongerardii (a C4 grass) in elevated (squares) and am-
bient (circles) CO2 OTC at the start(wet; emptysymbols) and end
unpublisheddata), but it is not known which growth
(dry; filled symbols) of a dryingcycle. Each point is the mean of formsconstitutethis increase. Our data suggest that
14 measurements;bars indicate ? 1 SE of the mean. Bottom row, broad-leavedwoody C3 species mighthave thegreatest
Percentagechange in Tleaf in elevated CO2 compared with ambient potential for increased productivitycompared with
CO2 OTC plants in wet (emptysymbols) and dry (filled symbols) herbaceous forbs. However, both C3 species showed
soil moistureconditions.
significantdeclines in Anetat elevated CO2 duringdry
periods,while the C4 grass did not. Elevated CO2 also
burnedprairieor exposed uplands,featuresthatstrong- resultedin consistently
highermiddayTleaf in A. ger-
ly influenceproductivityresponse to seasonal precip- ardii, while resultsfortheC3 species were mixed. This
itationin these grasslands(Knapp and Seastedt 1986; stands in contrastto othermixed C3/C4 communities,
Knapp et al. 1993a; Briggs and Knapp 1995). Also, which showed marked improvementin C3 plant per-
themarkedmidseasonincreasein A. gerardiigs,which formancethatresultedin rapid increases in C3 species
could be accompanied by alteredcanopy transpiration (Curtiset al. 1989; Ziska et al. 1990; Drake and Lead-
(Ham et al. 1995), suggest that modeling effortsat- ley 1991; Arp et al. 1993). However, the balance be-
temptingto predictchanges in primaryproductivity in tween C3 and C4 vegetationin tallgrassprairieis the
grasslandsystemsshould take into accountthe season- resultof complex interactionsbetweenclimate(Borch-
al patternsof water use and transpirationof canopy ert 1950; Axelrod 1985), grazing (Vinton et al. 1993;
dominantsas theyare affectedby elevated CO2 (Parton Fahnestock and Knapp 1994), and fire (Towne and
et al. 1994). Owensby 1984; Hulbert 1988; Briggs and Knapp
CO2 enrichmentdirectlybenefitedthe C3 shrubspe- 1995), not simply differencesin photosynthetic rate.
cies, Symphiocarposorbiculatus,the most in drought Rather,it seems that longer-term, cumulativeeffects
conditions compared with ambient CO2. Drought at such as tissue nutrientratios, allocation patterns,be-
ambientCO2 oftenresultedin stomatalclosure before lowgroundprocesses, nutrientlimitations,and poten-
midday, negative Anet, and leaf temperatures4-5?C tial changes in the variabilityin futureclimatescould
above air temperature.Elevated CO2 eliminatedthese more stronglymediate changes in tallgrass prairie
limitations,even at similarleaf temperatures.
Gas ex- communitystructurein an elevated CO2 world (Ow-
HAMERLYNCK ET AL.-TALLGRASS PRAIRIE PLANT
ensby et al. 1993a, 1993b; Kemp et al. 1994; Monz
et al. 1994; Owensby et al. 1994; Parton et al. 1994;
Rice et al. 1994).
Acknowledgments
We would like to thankShawn Conard and Melissa
Fillipi for theirunflaggingand cheerfulgatheringof
water potentialsin the face of mind-bendingtemper-
Literature
cited
Arp WJ,BG Drake, WT Pockman, PS Curtis,DF Whigham 1993
Interactionsbetween C3 and C4 salt marsh plant species during
fouryears of exposure to elevated CO2. Vegetatio 104/105:133-
143.
Axelrod DI 1985 Rise of the centralgrasslandbiome, centralNorth
America. Bot Rev 51:163-202.
BorchertJR 1950 The climateof the centralNorthAmericangrass-
land. Ann Assoc Am Geogr 15:1-39.
BremerDJ, JM Ham, CE Owensby 1996 Effectof elevated atmo-
sphericCO2 and open top chamberson transpiration in a tallgrass
prairie.J EnvironQual 25:691-701.
Briggs JM,AK Knapp 1995 Interannualvariabilityin primarypro-
ductionin tallgrassprairie:climate,soil moisture,topographicpo-
sition, and fire as determinantsof abovegroundbiomass. Am J
Bot 82:1024-1030.
CarterDR, KM Peterson 1984 Effectsof CO2-enrichedatmosphere
on the growthand competitiveinteractionof a C3 and C4 grass.
Oecologia 58:188-193.
Cheng DX, MB Coughenour,AK Knapp, CE Owensby 1994 Math-
ematical simulationof C4 grass photosynthesisin ambient and
elevated CO2. Ecol Model 73:63-80.
Curtis PS, BG Drake, DF Whigham 1989 Nitrogen and carbon
dynamicsin a C3 and C4 estuarinemarshplants grownunderel-
evated CO2 in situ. Oecologia 78:297-230.
Drake BG, PW Leadley 1991 Canopy photosynthesisof crops and
native plant communitiesexposed to long-termelevated CO2.
Plant Cell Environ 14:853-860.
FahnestockJT,AK Knapp 1994 Plant responses to selective bison
grazing: interactionsbetween light,herbivoryand water stress.
Vegetatio 115:123-131.
Ham JM, CE Owensby,PI Coyne, DJ Bremer 1995 Fluxes of CO2
and watervapor froma prairieecosystemexposed to ambientand
elevated atmosphericCO2. Agric For Meteorol 77:73-93.
HarperJL 1989 Canopies as populationsPages 105-128 in G Rus-
sell, B Marshall,PG Jarvis,eds. Plantcanopies: theirgrowth,form
and function.CambridgeUniversityPress, Cambridge.
HulbertLC 1988 Fire and littereffectsin undisturbedtallgrassprai-
rie. Ecology 50:874-877.
Hunt HW, ET Elliot, JK Detling, JA Morgan, DX Chen 1996 Re-
sponses of a C3 and C4 perennialgrass to elevated CO2 and tem-
peratureunderdifferent waterregimes.Global Change Biol 2:35-
47.
JohnsonHB, HW Polley, HS Mayeux 1993 Increasing CO2 and
plant-plantinteractions:effectson natural vegetation.Vegetatio
104/105:157-170.
Kemp PR, DG Waldecker,CE Owensby,JF Reynolds,RA Virginia
1994 Effectsof elevated CO2 and nitrogenfertilization pre-treat-
mentson decompositionin tallgrassprairie.Plant Soil 165:115-
127.
Knapp AK 1985 Effectof fireand droughton the ecophysiology
of Andropogongerardii and Panicum virgatumin a tallgrassprai-
rie. Ecology 66:1309-1320.
Knapp AK, M Cocke, EP Hamerlynck,CE Owensby 1994b Effect
of elevated CO2 on stomataldensityand distribution in a C4 grass
and a C3 forbunderfieldconditions.Ann Bot 74:595-599.
Knapp AK, JT Fahnestock,SP Hamburg,LB Statland,TR Seastedt,
DS Schimel 1993a Landscape patternsin soil-plantwater rela-
tionsand primaryproductionin tallgrassprairie.Ecology 74:549-
560.
RESPONSES TO ELEVATED C02 AND DROUGHT 615
atures and Fred Caldwell for providingtechnicalex-
pertisein maintainingthe OTC facility.This research
was supportedby the U.S. Departmentof EnergyTer-
restrial Carbon Processes Program (DE-FG02-
95ER62001), the Long TermEcological Researchpro-
gram (DEB-9011662) at Konza PrairieResearch Nat-
ural Area, and the Kansas AgriculturalExperimental
Station.
Knapp AK, JTFahnestock,CE Owensby 1994a Elevated CO2 alters
stomatalresponses to variable sunlightin a C4 grass. Plant Cell
Environ 17:198-195.
Knapp AK, EP Hamerlynck,JM Ham, CE Owensby 1996 Re-
sponses in stomatalconductanceto elevated CO2 in 12 grassland
species thatdifferin growthform.Vegetatio 125:31-41.
Knapp AK, EP Hamerlynck,CE Owensby 1993b Photosynthetic
and water relations responses to elevated CO2 in the C4 grass
Andropogongerardii. Int J Plant Sci 154:459-466.
Knapp AK, TR Seastedt 1986 Detritusaccumulationlimitsproduc-
tivityof tallgrassprairie.BioScience 36:662-668.
Long SP 1991 Modificationof the response of photosynthetic pro-
ductivityto rising temperatureby atmosphericCO2 concentra-
tions: has its importancebeen underestimated? Plant Cell Environ
14:729-739.
Monz CA, HW Hunt, FB Reeves, ET Elliot 1994 The response
of mycorrhizal colonization to elevated CO2 and climate
change in Pascopyrum smithii and Bouteloua gracilis. Plant
Soil 165:75-80.
Morgan JA,HW Hunt,CA Monz, DR LeCain 1994 Consequences
of growthat two carbon dioxide concentrationsand two temper-
aturesforleaf gas exchange in Pascopyrumsmithii(C) and Bou-
teloua gracilis (C4). Plant Cell Environ 17:1023-1033.
Nie D, MB Kirkham,LK Ballou, DJ Lawlor, ET Kanemasu 1992
Changes in prairievegetationunderelevatedcarbondioxide levels
and two soil moistureregimes.J Veg Sci 3:673-678.
Owensby CE, LM Auen, PI Coyne 1994 Biomass productionin a
nitrogen-fertilized tallgrassprairieecosystemexposed to ambient
and elevated levels of CO2. Plant Soil 165:105-113.
Owensby CE, PI Coyne, LM Auen 1993a Nitrogenand phosphorus
dynamics of a tallgrasss prairie ecosystem exposed to elevated
carbon dioxide. Plant Cell Environ 16:843-850.
Owensby CE, PI Coyne, JM Ham, LA Auen, AK Knapp 1993b
Biomass productionin a tallgrass prairie ecosystem exposed to
ambientand elevated CO2. Ecol Appl 3:644-653.
PartonWJ,DS Ojima, DS Schimel 1994 Environmentalchange in
grasslands:assessmentusing models. Clim Change 28:111-141.
Read JJ,JA Morgan 1995 Growthand partitioning in Pascopyrum
smithii(C3) and Bouteloua gracilis (C) as influencedby carbon
dioxide and temperature. Ann Bot 77:487-496.
Reichers GH, BR Strain 1987 Growthof blue grama (Bouteloua
gracilis) in response to atmosphericCO2 enrichment.Can J Bot
66:1570-1573.
Rice CW, FO Garcia, CO Hampton,CE Owensby 1994 Soil mi-
crobial response in tallgrass prairie to elevated CO2. Plant Soil
165:67-74.
SmithSD, BR Strain,TD Sharkey 1987 Effectsof CO2 enrichment
on fourGreat Basin grasses. Funct Ecol 1:139-143.
Towne G, CE Owensby 1984 Long-termeffectsof annual burning
at differentdates in ungrazed Kansas tallgrass prairie. J Range
Manage 37:392-397.
TurnerCL, JR Kneisler, AK Knapp 1995 Comparative gas ex-
change and nitrogenresponses of the dominant C4 grass, An-
dropogon gerardii and five C3 forbs to fire and topographic
position in tallgrass prairie during a wet year. Int J Plant Sci
156:216-226.
Vinton MA, DC Hartnett,EJ Finck, JM Briggs 1993 Interactive
effectsof fire,bison (Bison bison) grazing and plant commu-
nitycomposition in tallgrass prairie. Am Midl Nat 129:10-18.
616 INTERNATIONAL JOURNAL OF PLANT SCIENCES

von CaemmererS, GD Farquhar 1981 Some relationshipsbetween
the biochemistryof photosynthesisand the gas exchange of
leaves. Planta 153:376-387.
Weaver JE 1954 NorthAmerican prairie.Johnsen,Lincoln, Nebr.
348 pp.
WraySM, BR Strain 1987 Interactionof age and competitionunder
CO2 enrichment.Funct Ecol 1:145-149.
Ziska LH, BG Drake, S Chamberlain1990 Long-termphotosynthetic
responsein singleleaves of a C3 and C4 salt marshspecies grownat
elevatedatmospheric CO2 in situ.Oecologia 83:469-472.