Journal of Experimental Botany, Vol. 48, No. 317, pp. 2095-2102, December 1997
Journal of
Experimental
Botany
Effect of irradiance and vapour pressure deficit on
stomatal response to CO2 enrichment of four tree species
Rodney E. Will1 and Robert O. Teskey
School of Forest Resources, University of Georgia, Athens, GA 30602, USA
Received 18 March 1997; Accepted 26 August 1997
Abstract
The stomatal response of seedlings grown in 360 or
720 fitno\ m o r 1 to irradiance and leaf-to-air vapour
pressure deficit (VPD) at both 360 and 720 //mol mol
1
CO2 was measured to determine how environmental
factors interact with CO2 enrichment to affect stomatal
conductance. Seedlings of four species with different
conductances and life histories, Cercis canadensis (L.),
Quercus rvbra (L.), Populus deltoides (Bartr. ex
Marsh.) x P. nigra (L.), and Pinus taeda (L.), were measured in hopes of identifying general responses.
Conductance of seedlings grown at 360 and 720 //mol
m o l 1 CO2 were similar and responded in the same
manner to measurement CO2 concentration, irradiance
and VPD. Conductance was lower for all species when
measured at 720 than when measured at 360 //mol
m o l 1 CO2 at both VPDs ( - 1 . 5 and - 2 . 5 kPa) and all
measured irradiances greater than zero (100, 300,600,
>1600 //mol m~2 s~1). The average decrease in conductance due to measurement in elevated CO2 concentration was 32% for Cercis, 29% for Quercus, 26% for
Populus, and 11 % for Pinus. For all species, the absolute decrease in conductance due to measurement in
CO2 enrichment decreased as irradiance decreased or
VPD increased. The proportional decrease due to
measurement in CO2 enrichment decreased in three
of eight cases: from 0.46 to 0.10 in Populus and from
0.18 to 0.07 in Pinus as irradiance decreased from
>1600 to 100 //mol m" 2 s " 1 and from 0.35 to 0.24 in
Cercis as VPD increased from 1.3 to 2.6 kPa.
Key words: Stomatal conductance, C0 2 enrichment, irradiance, vapour pressure deficit.
Introduction
Stomatal conductance generally decreases when the atmospheric CO 2 concentration is elevated, but the magnitude
1
To whom correspondence should be addressed. Fax: +1 706 542 8356.
© Oxford University Press 1997
of stomatal response to CO 2 enrichment is quite variable
between and within species (Curtis, 1996; Field et al.,
1995; Morison, 1987; Raschke, 1986). Between-species
differences in physiological responses to the environment
are to be expected, but the large amount of variation
reported in stomatal sensitivity to CO 2 concentration
within species is somewhat surprising and particularly
troublesome for trying to predict the effect of atmospheric
CO 2 enrichment on plant water use. Loblolly pine (Pinus
taeda L.) is an excellent example of a species for which
large differences in stomatal sensitivity to CO 2 concentration have been reported. Responses of loblolly pine to
CO 2 enrichment include a decrease in stomatal conductance of as much as 50% depending on irradiance and
water status (Tolley and Strain, 1985), a decrease in
conductance up to 40% depending on growth CO 2 concentration (Fetcher et al., 1988), a decrease in conductance
between 10% and 21% depending on nutrient status
(Thomas et al., 1994), a 7% decrease in sap flow
(Ellsworth et al., 1995) and no significant effect on
stomatal conductance (Ellsworth et al., 1995; Liu and
Teskey, 1995; Teskey and Shrestha, 1985).
These large intraspecific differences in stomatal sensitivity to CO 2 concentration suggest factors such as genotype,
foliage age, nutrition, measurement environment or
growth environment interact with the effect of CO 2 concentration. To predict accurately the effect of increasing
atmospheric CO 2 concentration on plant water use, these
interactions must be better understood. An excellent way
to determine how the stomatal response to CO 2 enrichment will be affected by environmental fluctuations is to
grow a set of plants in ambient and a set of plants in
enriched CO 2 concentrations and then to measure all
plants in both ambient and enriched CO 2 concentrations
while manipulating environmental conditions.
A few studies have taken this approach. For two of
three species, Hollinger (1987) found stomatal conduct-
2096
Will and Teskey
ance was lower when measured in CO2 enrichment, and
that seedlings grown and measured in CO2 enrichment
exhibited a smaller per cent change in stomatal conductance in response to higher leaf-to-air vapour pressure
deficit (VPD) than seedlings grown and measured in
ambient CO2 concentration. In three out of four species
tested, Bunce (1993) found that the absolute and relative
effect of increasing VPD on stomatal conductance
decreased when plants were grown and measured in CO2
enrichment. In two of the species, Bunce found that
growth in CO2 enrichment may have altered the response
of stomatal conductance to CO2 concentration. Berryman
et al. (1994) found that in Eucalyptus tetrodonta (F.
Muell) seedlings, the absolute difference in stomatal
conductance due to measurement in ambient and elevated CO2 concentrations decreased when leaf-to-air
VPD increased or irradiance decreased. Needless to say,
there is great uncertainty as to the importance of
CO2 x environment interactions and their effect on
stomatal conductance.
In these experiments, seedlings of four species were
grown in 360/^molmol"1 (ambient) or 720 /imol mol ~'
(2 x ambient) CO2 concentrations under controlled conditions in growth chambers. Species were chosen which
represent a range of conductance, life histories and taxonomic groups in the hope of identifying general relationships. Stomatal response to irradiance and, in a separate
experiment, leaf-to-air VPD, of all seedlings was measured
at both 360 and 720 ^,mol mol" 1 CO2 concentration thus
testing the effect of growth CO2 concentration, measurement CO2 concentration, either irradiance or VPD, and
all interactions that may contribute to the conflicting
results concerning the effect of CO2 enrichment on
stomatal conductance.
It was hypothesized that measurement CO2
concentration x irradiance and measurement CO2
concentration x VPD interactions occur that would cause
the absolute decrease in stomatal conductance due to
measurement in CO2 enrichment to diminish as irradiance
decreased or VPD increased. It was also hypothesized
that as irradiance decreased or VPD increased, the proportional decrease due to measurement in CO2 enrichment
would also decrease. If both hypotheses are correct and
the absolute and proportional effect of CO2 enrichment
on stomatal conductance both decrease as environmental
conditions become less optimal, the effect of CO2 enrichment on stomatal conductance is magnified under optimal
conditions and approaches zero under suboptimal
conditions.
Materials and methods
Between 29 April and 2 May 1996, one-year-old bare root
seedlings of Cercis canadensis (L.), Quercus rubra (L.) Pinus
taeda (L.), and the Populus hybrid, Robusta (Populus deltoides
Bartr. ex Marsh, x P. nigra L.), obtained from commercial
sources, were planted in 5 1 pots filled with Fafard 3B potting
mix (Conrad Fafard Inc., Agawam, MA, USA). Seedlings were
maintained in a greenhouse under ambient conditions for
several days after planting and then moved into growth
chambers at or before bud burst. The four growth chambers
used in the experiment were 2.5 m wide x 1.3 m deep x 2 m tall
and controlled temperature at 23 °C, relative humidity at 60%,
and carbon dioxide at a concentration of either 360 or
720mmolmol~' (Environmental Growth Chambers, Chagrin
Falls, OH, USA). Irradiance at the top of the seedlings was
approximately 700 ^.mol m~ 2 s" 1 and was supplied for a 13 h
photoperiod by a combination of incandescent and high output
florescent bulbs so that all wavelengths within the visible
spectrum were present. Seedlings were placed in two chambers,
one of which was maintained at 360 and the other at
720 /jjnol mol" 1 CO 2 . Seedlings were rotated between chambers
once per week to balance any chamber effects. Chamber CO 2
concentrations were adjusted when seedlings were rotated to
maintain the appropriate treatments. Seedlings were watered as
needed and fertilized every 5 d with a solution containing 0.06
g l " 1 N, 0.12 g l " 1 P, 0.08 g l " 1 K., 0.0002 g l " 1 B, 0.0004 g l " 1
Cu, 0.0008 g l " 1 Fe, 0.0004 g l " 1 Mn, 0.0004 g l " 1 Zn, and
0.000004 g l " 1 Mo supplied as Stern's Miracle-Gro (Stern's
Miracle-Gro Products Inc., Port Washington, NY, USA). In
addition, supplemental iron was supplied to the Pinus at a rate
of 0.004 g l " 1 chelated Fe (Miller Iron Chelate DP, Miller
Chemical, Hanover, PA, USA).
The stomatal conductance-irradiance relationship was measured at both CO 2 concentrations (360 and 720 /xmol mol" 1 ) for
all seedlings. Seedlings were continuously exposed to their
growth CO 2 concentrations (360 or 720 /xmol mol" 1 ) from bud
burst to when they were measured: 6-10 weeks for Populus,
10-14 weeks for Cercis, 16—19 weeks for Quercus, and 19-22
weeks for Pinus. Stomatal conductance was measured in the
chambers using a LiCor Ll-6400 portable photosynthesis system
(Li-Cor Inc., Lincoln NE, USA) at the target irradiances of
> 1600, 600, 300, 100, and 0 /anol m" 2 s" 1 PPFD. Supplemental
light was provided by a heat shielded 50 W quartz halogen bulb
(EXN/CG, General Electric Co., Nela Park, OH, USA) 20 cm
above the cuvette. Irradiance in the cuvette was manipulated
by placing screens between the light source and the cuvette.
Actual measured mean irradiances were 1597, 601, 302, 99,
and O f t m o l m ^ s " 1 for Populus, 1612, 609, 303, 98, and
O ^ m o l m ^ s " 1 for Cercis, 1609, 601, 300, 101, and
O / t m o l m ^ s " 1 for Quercus, and 2129, 645, 313, 102, and
0 ^ m o l m " 2 s " 1 for Pinus. Leaf temperature during measurements was maintained at 26 °C and cuvette CO 2 concentration
was maintained at either 360 or 720 ftmol mol" 1 by the LI-6400
system. A similar VPD was maintained at all irradiances by
adjusting the flow rate and the proportion of dry air entering
the cuvette. Mean VPD at the different irradiances ranged from
1.10-1.35 kPa for Populus, 1.39-1.58 kPa for Cercis,
1.49-1.58 kPa for Quercus, and 1.41-1.50 kPa for Pinus. A
curve at each CO 2 concentration was measured on consecutive
days on the same leaf or needles. The order of measurement
CO 2 concentration was randomized. Measurements were not
taken until the leaves had fully equilibrated to the new
irradiance. Equilibration time ranged between 45 min for
Populus and 10 min for Quercus. Ten Populus seedlings (n = 5)
and twelve Quercus, Cercis and Pinus seedlings (n = 6) were
measured.
In a separate experiment, stomatal conductances of the same
seedlings continuously grown at either 360 or 720 jtmol m " 2 s" 1
CO 2 were measured at a low and high leaf-to-air VPD at both
Light, VPD, CO2, and stomata
CO 2 concentrations. For each species, the VPD response was
measured after the light response: approximately 9-10 weeks
after bud burst for Populus, 12-14 weeks after bud burst for
Cercis, 18-19 weeks after bud burst for Quercus, and 21-22
weeks after bud burst for Pinus. As in the light experiment, the
reciprocal CO 2 measurements were conducted on the same
foliage over two consecutive days with the order of measurement
CO 2 randomized. Mean measurement VPDs were restricted by
the conductance of the species and were 1.1 (0.075 s.d.) and
2.5 kPa (0.072 s.d.) for Populus, 1.3 (0.045 s.d.) and 2.6 kPa
(0.033 s.d.) for Cercis, 1.5 (0.063 s.d.) and 2.6 kPa (0.094 s.d.)
for Quercus, and 1.4 (0.039 s.d.) and 2.5 kPa (0.026 s.d.) for
Pinus. Leaf temperature was controlled at 26 °C and mean
irradiance was 2053 /imol m " 2 s " ' for Pinus, 1506 ^molm~ 2
s~' for Populus, 1459 ^mol m~ 2 s" 1 for Quercus, and
1590 jxsao\ m~ 2 s" 1 for Cercis. For all species, 12 plants, six
grown in 360 and six grown in 720/^mol mol ~' CO 2 , were
measured.
To determine how differences between stomatal conductance
measured at 360 and 720^molmol~ 1 CO 2 translated to
differences in water use, whole plant water use was gravimetrically measured at both CO 2 concentrations on consecutive
days for all plants. Pots were watered beyond saturation at
night and then weighed at 08.00 and 17.00 h the following day.
Immediately following the weighing at 17.00 h, CO 2 concentration was adjusted to the reciprocal and the protocol was
repeated with the seedlings in the same chamber and chamber
position. The difference in weight minus a constant value
representing evaporation (averaged from several pots without
seedlings) was considered to be the amount of water transpired
by the plant. The sample size was 17 for Populus, 10 for Cercis,
12 for Quercus, and 13 for Pinus.
Stomatal density was measured by counting the number of
stomata using a light microscope at 400 x magnification for
Populus, Cercis and Quercus or using a dissecting microscope
at 40 x magnification for Pinus. Ten fields per leaf were
averaged. A fan was placed near the microscope to prevent
condensation of water on the lens.
For all experiments, species were analysed separately. The
light experiment was analysed with a multivanate repeatedmeasures analysis of variance (SAS Institute Inc., 1987) with
two repeated measures factors, light and measurement CO 2
concentration, and the between subject factor, growth CO 2
concentration. To examine the linear, quadratic and higher
order components of the effect of light and light x measurement
CO 2 interaction, orthogonal polynomial contrasts were constructed using the 'Polynomial' transformation (SAS Institute
Inc., 1987). A similar repeated-measures analysis of variance
was performed on the data from the VPD experiment with the
exception that contrasts were not necessary because each factor
in the analysis had only two levels. Because of significant
light x measurement CO 2 and VPD x measurement CO 2 interactions, data at each irradiance and VPD level also were
analysed separately with a repeated-measures analysis with the
single repeated factor, measurement CO 2 concentration, and
the between subject factor, growth CO 2 concentration. To
examine the how the relative effect of measurement CO 2
concentration changed as irradiance changed, the proportional
decrease in stomatal conductance due to measurement in CO 2
enrichment was calculated at each irradiance and VPD by
subtracting from one the ratio of stomatal conductance
measured in 720 ^mol mol" 1 CO 2 divided by the stomatal
conductance measured in 360 /imol mol"' CO 2 .
The data from the whole plant water use experiments were
tested with a repeated-measures analysis with the repeated
factor, measurement CO 2 concentration and the between subject
2097
factor, growth CO 2 concentration. Stomatal density data were
analysed with an ANOVA to determine whether the effect of
growth CO 2 concentration was significant.
Results
Light experiment
For all species, there were no differences in stomatal
conductance between seedlings grown in 360 and seedlings
grown in 720 fimol mol ~ 1 CO 2 for the effect of CO 2
concentration during measurement, the effect of light
during measurement or their interactions (no growth CO 2
concentration effect or interactions involving growth CO 2
concentration). Therefore, discussion of results concentrates on the effect of measurement CO 2 concentration.
Data from seedlings of both growth concentrations were
pooled for presentation in figures and tables.
For all four species, stomatal conductance increased
(P<0.05) with increasing irradiance (light effect) and was
greater (P<0.05) when measured at 360 than when
measured at 720/xmolmoF 1 CO 2 (measurement CO 2
effect) (Fig. 1). In addition, the absolute decrease in
stomatal conductance due to measurement in CO 2 enrichment increased (/ > <0.05) with increasing irradiance
(light x measurement CO 2 interaction) (Fig. 1). For all
species, the linear and quadratic components of the light
effect and the light x measurement CO 2 interaction were
significant (/><0.05). When the effect of measurement
CO 2 concentration was examined separately at each
irradiance, conductance of Cercis, Quercus and Pinus was
lower (P<0.05) when measured at 720 than when measured at 360/^mol mol" 1 CO 2 at all measured irradiances
above 0 /^mol m " 2 s ~ ' PPFD and conductance of Populus
was lower (/ > <0.05) at all measured light intensities
except 0 and lOO^mol m " 2 s~l PPFD (Fig. 1).
Given the significant linear and quadratic components
of the measurement CO 2 x light interactions, the proportional decrease in stomatal conductance due to measurement in CO 2 enrichment was calculated to determine how
irradiance was influencing the relative effect of measurement CO 2 concentration. For all species, the proportional
decrease at 0 ^ m o l m " 2 s ~ ' PPFD was approximately 0
and less than the proportional decreases calculated at the
other irradiances (Table 1). For Cercis and Quercus, the
proportional decrease due to measurement in CO 2 enrichment was similar at 100, 300, 600, and 1600 ^mol m" 2
(Table 1). For Populus and Pinus however, the
s - i PPFD
proportional decrease due to measurement in CO 2 enrichment increased as irradiance increased (Table 1).
VPD experiment
As in the light experiment, there were no differences in
the stomatal response between seedlings grown in 360
2098
Will and Teskey
0.08
~
0.10 -
Cercis
Quercus
*
0.06 -
0.04
G
c
360 umol mol"1 CO2
o
o
0.02
360 umol mol'1 CO2
720 umol mol1 CO2
720 umol mol'1 CO2
0.00
000
400
800
1200
1600
0
400
800
1200
1600
0.12
360 umol mol CO2
720 umol mol"1 CO2
0.00
0.0 J
400
800
1200
2
400
1600
800
1200
2
1600
1
PPFD (umolm sec' )
1
PPFD (umolm" sec" )
Fig. 1. Mean light response curves for stomatal conductance of Cercis canadensis, Quercus rubra, Populus deltoides x P. nigra, and Pinus taeda
measured at 360 and 720 /imol mol"' CO2 concentration. Growth CO2 concentration did not influence the response to light or measurement CO2
so measurements from seedlings grown in 360 and 720^molmoP' CO2 were pooled. An asterisk indicates a significant difference due to
measurement CO2 concentration at a particular irradiance. Vertical bars represent standard errors.
Table 1. Mean proportional decrease in stomatal conductance due to measurement in CO2 enrichment [I — (measurement at 720
fjMiol mol~x CO Jmeasurement in 360 fjmolmor1 COJJ for seedlings of Cercis canadensis, Quercus rubra, Populus deltoides x P.
nigra and Pinus taeda at five target irradiances (actual irradiances are listed in Materials and methods)
Growth CO2 concentration did not influence the stomatal response to CO2 concentration or light so measurements of seedlings grown in 360 and
720 ^mol mol"1 CO2 were pooled. Standard errors are in parentheses.
Irradiance
"2s"')
0
100
300
600
1600
Cercis
Quercus
Populus
Pinus
0.03 (0.047)
0.38 (0.044)
0 42(0.059)
0.37(0.051)
0.38 (0.047)
0.13(0.080)
0.27 (0.045)
0.31 (0 035)
0.32 (0.036)
0.29 (0.042)
-0.11 (0.133)
0.10(0 140)
0.36(0.057)
0.37(0 045)
0.46 (0.039)
-0.02(0.038)
0.07(0.028)
0.11 (0.030)
0.18(0 026)
0.18 (0.025)
and those grown in 720 /xmol mol ' CO2 when measured
at same CO2 concentrations and VPDs (no growth CO2
concentration effect or interactions involving growth CO2
concentration) so results of seedlings from both growth
concentrations were pooled for presentation in figures
and tables. For all four species, stomatal conductance
was lower at the higher VPD regardless of measurement
CO2 concentration (VPD effect), stomatal conductance
of seedlings measured at 720 ^mol mol ~' CO2 was lower
than that measured at 360 junol mol ' CO2 regardless of
VPD (measurement CO2 effect) and the absolute difference in stomatal conductance between plants measured
at 720 and 360^molmol~1 was smaller at the higher
VPD than at the lower VPD (VPD x measurement CO2
interaction) (Fig. 2).
When data from the low and high VPDs were analysed
separately, stomatal conductance at both VPDs was lower
(P<0.05) when measured at 720 ^mol mol ~' CO2 than
Light, VPD, C02, and stomata
2099
0.10
Cercis
Quercus
0.08
•y 008
"5
0.06 -
f 0.06
360 umol mol"1 CO2
720 umol mol"1 CO2
o
2.0
1.5
360 umol mol CO2
720 umol mol"1 CO2
0.04
2.5
2.0
1.5
Populus
2.5
Pinus
0.10 -
0.08 -
10
o 0.2
O
360 umol mol'1 C
720 umol mol"1 CO2
1.0
1.5
2.0
360 umol mol"1 CO2
720 umol mol"1 CO2
0.06 -
2.5
20
1.5
VPD (kPa)
2.5
VPD (kPa)
Fig. 2. Mean vapour pressure deficit response curves for stomatal conductance of Cercis canadensis, Quercus rubra, Populus deltoides x P. nigra, and
Pinus taeda measured at 360 and 7 2 0 ^ m o l m o r ' CO 2 concentration. Growth CO 2 concentration did not influence the response to VPD or
measurement CO 2 so measurements from seedlings grown in 360 and 720 ^mol mol" 1 CO 2 were pooled. An asterisk indicates a significant difference
due to measurement CO 2 concentration at a particular vapour pressure deficit. Vertical bars represent standard errors.
when measured at 360 /xmol mol ' CO 2 concentration
(Fig. 2). Given the significant VPD x measurement CO 2
interactions, the proportional decrease in stomatal conductance due to measurement in CO 2 enrichment was
calculated at each VPD to determine how the relative
effect of measurement CO 2 concentration changed as
VPD changed. Only Cercis showed a large reduction in
the proportional decrease due to CO 2 enrichment as VPD
increased (Table 2).
Whole plant water use and stomatal density
In all four species, whole plant water use was lower
(P<0.05) during exposure to 720 ^mol mol" 1 CO 2 than
during exposure to 360^molmol"' CO 2 (Table 3). For
Cercis, Quercus, and Populus, neither the effect of growth
CO 2 concentration or the growth CO 2 x measurement
CO 2 interaction influenced this response. For Pinus, however, the growth CO 2 x measurement CO 2 interaction was
significant because whole plant water use of seedlings
grown in 720^molmol" 1 CO 2 was more sensitive to
measurement CO 2 concentration than seedlings grown in
360 ^.mol mol" 1 CO 2 (P<0.05) (Table 3). Stomatal den-
Table 2. Mean proportional decrease in stomatal conductance
due to measurement in CO2 enrichment [ I — (measurement at
720 fji/nol mol~l CO Jmeasurement in 360 fj/nol mol~x CO2J]
for seedlings of Cercis canadensis, Quercus rubra, Populus
deltoides x P. nigra and Pinus taeda at two vapour pressure
deficits (actual VPDs are listed in Materials and methods)
Growth CO 2 concentration did not influence the stomatal response to
CO 2 concentration or VPD so measurements of seedlings grown in 360
and 720fimol mol" 1 CO 2 were pooled. Standard errors are in
parentheses.
VPD
(kPa)
-1.5
-2.5
Cercis
Quercus
Populus
Pinus
0.35(0 039)
0.24(0.035)
0.35 (0.019)
0.34 (0.020)
0.33 (0.048)
0.30(0.091)
0.16 (0.027)
0 10(0.034)
sity was not statistically different between seedlings grown
in 360 and 720 ^mol raol"1 CO 2 concentration for any of
the species. Stomatal densities for seedlings grown in 360
and 720 /xmol mol" 1 CO 2 concentration for Cercis were
225 and 229 mm" 2 , for Quercus were 436 and 417 mm" 2 ,
for Populus were 140 and 136mm" 2 (abaxial), and 90
2100
Will and Teskey
Table 3. Mean proportional decrease in whole plant water use
due to measurement in CO2 enrichment [ 1 — (measurement at
720fjjnolmol~x
COiJmeasurement in 360 p/nolmol~' CO2JJ
for seedlings of Cercis canadensis, Quercus rubra, Populus
deltoides x P. nigra and Pinus taeda grown at two CO2
concentrations: standard errors are in parentheses
Growth CO2
(/xmol mol"1)
Cercis
Quercus
Populus
Pmus
360
720
0.41(0.023)
0.37(0.018)
0.30(0.041)
0.36(0.023)
0.19(0.016)
0.19(0.026)
0.10(0.016)
0.16(0.013)
and 90 mm
91 mm" 2 .
2
(adaxial), and for Pinus were 91 and
Discussion
For all species in our study, the absolute difference
between stomatal conductance measured in ambient and
twice ambient CO2 concentration was smaller at lower
irradiance or higher VPD. In addition, for three of the
eight experiments (Pinus-hght, Populus-light, CercisVPD), the proportional decrease due to measurement
in CO2 enrichment decreased as irradiance decreased or
VPD increased. Thus, the absolute response, and in some
cases, the relative response of stomatal conductance to
CO2 enrichment was not constant. This changing response
makes experimental outcomes dependent on environmental conditions during measurement, and predicting
the effects of CO2 enrichment on stomatal conductance
and water use difficult.
Although the proportional decrease of stomatal conductance due to measurement in CO2 enrichment
decreased as environmental conditions became less
optimal in three out of eight experiments, the proportional
effect of CO2 enrichment probably always becomes negligible as stomata close in response to stress. Interestingly,
the effect of CO2 concentration on the stomatal conductance of Cercis, the species most commonly found in the
high humidity understory environment, became proportionately smaller at the higher VPD, indicating that stress
may have been beginning to override the CO2 effect.
Similarly, the effect of CO2 concentration on stomatal
conductance of Pinus and Populus, the two shade- intolerant species, became proportionately smaller as irradiance
decreased.
In Commelina communis L. (Morison and Jarvis, 1983)
and Eucalyptus tetrodonta (Berryman et al., 1994) the
absolute difference in stomatal conductance due to measurement CO2 concentration was greater under high than
under low irradiance, but the proportional effect due to
CO2 concentration did not change as irradiance changed.
In contrast, for Eucalyptus pauciflora Sieb. ex Spreng
(Wong et al., 1978) and five herbaceous species (Sharkey
and Raschke, 1981), little change in the absolute differ-
ence in stomatal conductance due to measurement CO2
concentration occurred as irradiance increased. However,
the internal CO2 concentrations used by Wong et al.
(1978) and Sharkey and Raschke (1981) corresponded
to current ambient CO2 concentration and below. Quite
possibly, the difference in stomatal conductance due to
CO2 concentration could have been larger under high
irradiance if conductance had been measured in ambient
and enriched CO2 concentrations.
Morison and Gifford (1983), Hollinger (1987), Bunce
(1993) and Berryman et al. (1994) all found that the
absolute difference in stomatal conductance between
plants measured under ambient and elevated CO2 concentrations decreased with increasing VPD. Although
Berryman et al. (1994) and Morison and Gifford (1983)
did not find any large changes in the proportional effect
of CO2 concentration at different VPDs, Bunce (1993)
found a decrease in the proportional effect of CO2 enrichment as VPD increased for three of four species measured.
Environment x CO2 concentration interactions do not
appear to be the sole cause of the large range in reported
stomatal response of some species to CO2 enrichment.
When the proportional effect due to measurement CO2
concentration changed with changing irradiance or VPD,
three times out of eight, the changes were not very large.
For instance, the proportional decrease due to measurement in CO2 enrichment changed from 0.46 to 0.34 for
Populus and 0.18 to 0.11 for Pinus when irradiance
decreased from > 1600 to 300^molm~2 s"1 PPFD and
changed from 0.35 to 0.24 for Cercis when VPD increased
from 1.3 to 2.6 kPa.
Even though some of the differences in the reported
effects of CO2 concentration on stomatal conductance
may be due to differences in irradiance and VPD during
measurement, other factors must also contribute to the
variation. For instance, other studies have found that
foliage age (Curtis and Teeri, 1992), growth environment
(Talbott et al., 1996), CO2 concentration during growth
(Berryman et al., 1994; Conroy et al., 1988), genotype,
nutrition (Conroy et al., 1988; Thomas et al., 1994), and
time of day (Surano et al., 1986) may all affect the
magnitude of the stomatal response to CO2 concentration.
Another factor that may affect the reported stomatal
response to CO2 concentration is equilibration time
because measurements made without allowing adequate
equilibration will include potentially large experimental
errors. In the present study, 10-45 min was necessary for
complete equilibration of stomatal conductance to
changes in irradiance or VPD whereas photosynthesis
reached steady state in less than five minutes. Others have
also reported large differences between the time required
for photosynthesis and stomatal conductance to equilibrate to changing environmental conditions (Whitehead
and Teskey, 1995; Barradas and Jones, 1996; Ogren and
Sundin, 1996).
Light, VPD, CO2, and stomata
A doubling of CO 2 concentration during measurement
decreased the stomatal conductance of the four species
that were measured. Mean decreases were 32% for Cercls,
29% for Quercus, 26% for Populus, and 11% for Pinus.
In 23 out of 25 studies conducted primarily on herbaceous
species, stomatal conductance decreased approximately
40% when measured at twice ambient CO 2 concentration
(Morison, 1987). The reported effect of CO 2 enrichment
on tree species tends to be smaller. For instance, stomatal
conductance decreased by 23% on average when data
from 29 studies of 23 tree species were compiled (Field
et al., 1995). In the present study, the results of the whole
plant water use experiment verified not only the largemeasured decrease in stomatal conductance due to CO 2
enrichment, but that decreases in stomatal conductance
translate to decreases in whole plant water use.
Stomatal conductance of seedlings grown in ambient
and elevated CO 2 concentrations responded the same to
measurement CO 2 concentration, VPD, and irradiance
and had similar stomatal densities. Therefore, the
stomatal conductance and stomatal density of leaves that
developed under CO 2 enrichment did not acclimate to
growth in twice ambient CO 2 concentration even though
leaves were continuously exposed to CO 2 enrichment
from bud burst. Rather, stomata of all seedlings reduced
water loss at higher CO 2 concentrations via short-term
and reversible changes in aperture. The only exception
was whole plant water use of Pinus taeda seedlings. In
this case, seedlings grown in 720 /xmol mol ~' CO 2 displayed a greater decrease in water use in response to CO 2
enrichment than those grown in SoD^molmol" 1 CO 2 .
The Pinus seedlings grown in 720/imol mol" 1 CO 2 were
larger and supported new foliage growth while those
grown in 360/nmol mol" 1 were smaller and at an earlier
phenological stage. If sensitivity of stomatal conductance
to CO 2 concentration varies with plant size or foliage age
than these size and phenological differences could have
caused this difference in whole plant water use.
Alternatively, the size differences between the seedlings
grown in ambient and twice ambient CO 2 may have led
to differences in canopy or boundary layer conductance
which could have affected transpiration and water use
even though stomatal conductance was not affected.
Nevertheless, stomatal conductance measurements on
fully expanded needles of similar phenological age did
not show a difference due to growth CO 2 concentration.
In a review of the literature, Woodward and Kelly
(1995) found that 60% of controlled experiments measuring stomatal density have shown a decrease in response
to growth in CO 2 enrichment. Previous reports on the
effect of growth in elevated CO 2 concentration on
stomatal sensitivity to CO 2 concentration during measurement vary, with growth in CO 2 enrichment sometimes
increasing, sometimes decreasing or sometimes not
affecting the stomatal response (Eamus, 1991). No differ-
2101
ence was found in stomatal response or stomatal density
between seedlings grown in 360 and 720 ^mol mol" l CO 2
in the four different species which were tested.
Although growth in CO 2 enrichment does not necessarily affect the stomatal response to measurement CO 2
concentration (no effect in this study), CO 2 enrichment
usually decreases the stomatal conductance of trees. Even
though an interactive combination of several internal and
external factors appears to control the magnitude of
stomatal response to CO 2 concentration, irradiance and
VPD do affect the response. Accordingly, the effect of
irradiance and VPD should be considered when examining intraspeciiic differences in stomatal sensitivity to CO 2
concentration, when performing future experiments and
when modelling the effect of CO 2 enrichment on plant
water use.
Acknowledgement
We would like to thank William Jennings (Jay) Brown for his
help planting the seedlings and his assistance keeping them
healthy and vigorous.
References
Barradas VL, Jones HG. 1996. Responses of CO 2 assimilation
to changes in irradiance: laboratory and field data and a
model for beans (Phaseolus vulgaris L.). Journal of
Experimental Botany 47, 639—45.
Berryman CA, Eamus D, Duff GA. 1994. Stomatal responses to
a range of variables in two tropical tree species grown with
CO 2 enrichment. Journal of Experimental Botany 45, 539—46.
Bunce JA. 1993. Effects of doubled atmospheric carbon dioxide
concentration on the responses of assimilation and conductance to humidity. Plant, Cell and Environment 16, 189-97.
Conroy JP, KOppers M, KOppers B, Virgona J, Barlow EWR.
1988. The influence of CO 2 enrichment, phosphorous deficiency and water stress on the growth, conductance and
water use of Pinus radiata D. Don. Plant, Cell and
Environment 11, 91-8.
Curtis PS. 1996. A meta-analysis of leaf gas exchange and
nitrogen in trees grown under elevated carbon dioxide. Plant,
Cell and Environment 19, 127-37.
Curtis PS, Teeri JA. 1992. Seasonal responses of leaf gas
exchange to elevated carbon dioxide in Populus grandidentata.
Canadian Journal of Forest Research 22, 1320—5.
Eamus D. 1991. The interaction of rising CO 2 and temperatures
with water use efficiency. Plant, Cell and Environment
14, 843-52.
Ellsworth DS, Oren R, Huang C, Phillips N, Hendrey GR. 1995.
Leaf and canopy responses to elevated CO 2 in a pine forest
under free-air CO 2 enrichment. Oecologia 104, 139-46.
Field CB, Jackson RB, Mooney HA. 1995. Stomatal responses
to increased CO 2 : implications from the plant to the global
scale. Plant, Cell and Environment 18, 1214-25.
Fetcher N, Jaeger CH, Strain BR, Sionit N. 1988. Long-term
elevation of atmospheric CO 2 concentration and the carbon
exchange rates of saplings of Pinus taeda L. and Liquidambar
styraciflua L. Tree Physiology 4, 255-62.
2102
Will and Teskey
Hollinger DY. 1987. Gas exchange and dry matter allocation
responses to elevation of atmospheric CO 2 concentration in
seedlings of three tree species. Tree Physiology 3, 193-202.
Liu S, Teskey RO. 1995. Responses of foliar gas exchange to
long-term elevated CO 2 concentrations in mature loblolly
pine trees. Tree Physiology 15, 351-9.
Morison JIL. 1987. Intercellular CO 2 concentration and
stomatal response to CO 2 . In: Zeiger E, Farquhar GD,
Cowan IR, eds. Stomatal function. Stanford University
Press, 229-52.
Morison JIL, Gifford RM. 1983. Stomatal sensitivity to carbon
dioxide and humidity: A comparison of two C 3 and two C 4
grass species. Plant Physiology 71, 789-96.
Morison JIL, Jarvis PJ. 1983. Direct and indirect effects of
light on stomata. II. In Commelina communis L. Plant, Cell
and Environment 6, 103-9.
dgren E, Sundin U. 1996. Photosynthetic responses of variable
light: a comparison of species from contrasting habitats.
Oecologia 106, 18-27.
Raschke K. 1986. The influence of CO 2 content of the ambient
air on stomatal conductance and the CO 2 concentration in
leaves. In: Enoch HZ, Kimball BA, eds. Carbon dioxide
enrichment
of greenhouse
crops, Vol II. Boca R a t o n : C R C
Press Inc., 87-102.
SAS Institute Inc. 1987. SAS user's guide: version 6. SAS
Institute Inc., Box 8000, Cary, NC.
Sharkey TD, Raschke K. 1981. Separation and measurement of
direct and indirect effects of light on stomata. Plant Physiology
68, 33-40.
Surano KA, Daley PF, Houpis LJ, Shinn JH, Helms JA,
Palassou RJ, Costella MP. 1986. Growth and physiological
responses of Pinus ponderosa Dougl ex P. Laws, to long-term
elevated CO 2 concentrations. Tree Physiology 2, 243-59.
Talbott LD, Srivastava A, Zeiger E. 1996. Stomata from growthchamber-grown Vicia faba have an enhanced sensitivity to
CO 2 . Plant, Cell and Environment 19, 1188-94.
Teskey RO, Shrestha RB. 1985. A relationship between carbon
dioxide, photosynthetic efficiency and shade tolerance.
Physiologia Plantarum 63, 126-32.
Thomas RB, Lewis JD, Strain BR. 1994. Effects of leaf nutrient
status on photosynthetic capacity in loblolly pine (Pinus taeda
L.) seedlings grown in elevated atmospheric CO 2 . Tree
Physiology 14, 947-60.
Tolley LC, Strain BR. 1985. Effects of CO 2 enrichment and
water stress on gas exchange of Liquidambar styracifiua and
Pinus taeda seedlings grown under different irradiance levels.
Oecologia 65, 166-72.
Whitebead D, Teskey RO. 1995. Dynamic response of stomata
to changing irradiance in loblolly pine (Pinus taeda L.). Tree
Physiology 15, 245-51.
Wong SC, Cowan IR, Farquhar GD. 1978. Leaf conductance in
relation to assimilation in Eucalyptus panciflora Seib. ex
Spreng. Plant Physiology 62, 670-4.
Woodward FI, Kelly CK. 1995. The influence of CO 2 concentration on stomatal density. New Phytologist 131, 311-27.