STOMATAL SENSITIVITY TO PHOTOSYNTHETIC AND ENVIRONMENTAL SIGNALS

IN GLYCINE MAX GROWN AT ELEVATED ATMOSPHERIC CONCENTRATIONS OF CO2 AND O3

BY

KATHERINE THERESE RICHTER

THESIS

Submitted in partial fulfillment of the requirements for the degree of Master of Science in Plant Biology

in the Graduate College of the University of Illinois at Urbana-Champaign, 2011

Urbana, Illinois

Master’s Committee:

Assistant Professor Andrew Leakey, Chair Assistant Professor Carl Bernacchi Assistant Professor Micheal Dietze

ii    

ABSTRACT

Rising atmospheric concentrations of carbon dioxide (CO2) and ozone (O3) will stimulate

and impair, respectively, yield and productivity of both managed and natural ecosystems in the

21st century. The ease of diffusion of these gases, as well as water vapor, between the

atmosphere and interior of leaves is actively controlled by stomatal pores and termed stomatal

conductance (gs). gs is highly regulated in response to environmental factors and a key

determinant of plant carbon gain, water use and ozone uptake. The Ball, Woodrow and Berry

(1987, Progress in Photosynthetic Research, vol IV, 221-224) model of stomatal conductance is a

central component of ecosystem and global models of carbon and water cycling used to predict

the impacts and feedbacks of vegetation and climate. It is essential that the m parameter of the

model, which characterizes the sensitivity of gs to net photosynthetic CO2 assimilation rate (A),

fractional relative humidity at the leaf surface (h) and atmospheric CO2 concentration ([CO2]), be

appropriate for the conditions in consideration to avoid errors in large-scale predictions of

ecosystem function and global biogeochemical cycling. Characterizing any changes in m with

elevated [CO2] and [O3] will be necessary for realistic simulations of the future. Variation in

maximum carboxylation capacity (Vcmax) as a result of plant growth in elevated [CO2] or

elevated [O3] will change the A that a given ci, permitted by a given gs, will support. Such a

Vcmax-induced change in the sensitivity of gs to A may contribute to a change in m. Soybean was

exposed to a range of [O3] in the field, and to combinations of different [O3] and [CO2] in

environmental growth chambers, to examine the effect of these conditions on m. A positive

correlation between m and [O3] was observed in the field. This was reproduced in the controlled

environment experiment at the lowest growth [CO2] treatment, but the response was ameliorated

by elevated [CO2]. This suggests gs becomes more sensitive to A, h and/or [CO2] when plants are

iii    

grown in higher [O3] conditions and ambient [CO2]. The estimate of gs was 46% lower for the

unparameterized relative to the parameterized Ball et al. (1987) model for the highest growth

[O3] of 120ppb and ambient [CO2]. Reparameterization of large-scale models will be necessary

to account for this change in stomatal function under these conditions. A strong, negative

correlation between Vcmax and m was present whenever m was significantly altered. Estimation of

m values based on this correlation may present an easy means to improve simulations of large-

scale models for different species and growth conditions involving elevated [O3] without

additional parameterization measurements.

iv    

TABLE OF CONTENTS

CHAPTER 1: INTRODUCTION…….………………………………………………………….1 CHAPTER 2: GROWTH AT ELEVATED [O3] LEADS TO ACCLIMATION IN SOYBEAN THAT INCREASES THE SENSITIVITY OF STOMATAL CONDUCTANCE TO PHOTOSYNTHETIC RATE, HUMIDITY AND [CO2] …………..………………...…………..4 CHAPTER 3: GROWTH AT ELEVATED [CO2] AMELIORATES THE CHANGES IN SENSITIVITY OF STOMATAL CONDUCTANCE TO PHOTOSYNTHETIC AND ENVIRONMENTAL SIGNALS AT ELEVATED [O3] ………………………….…………….24 CHAPTER 4: CONCLUSION…………………...……..………………………………….........47 REFERENCES…...……………………………...………………….………………………...…50      

1

CHAPTER 1: INTRODUCTION

Humans are highly dependent on ecosystem services, including for provision of

food, water and raw materials. Every year humans harvest approximately 1/8th of global

terrestrial, potential net primary productivity for consumption as food, fiber and fuel

(Haberl et al. 2007) and remove freshwater resources from rivers and aquifers at a rate of

4000 km2 per year (Arnell 2004). These harvesting activities, plus anthropogenic

emissions of carbon dioxide (CO2) from fossil fuel burning, are driving changes in the

carbon and water cycles that influence the stability and availability of indispensable

ecosystem services. The continually increasing global population, with an annual growth

rate of 1.2% (World Bank 2009; http://search.worldbank.org), is exacerbating the

challenge to sustainable development. Understanding human-driven changes to the

biosphere is, therefore, essential if we are to mitigate and adapt to environmental changes

in the coming century.

Anthropogenic emissions currently represent a flux of approximately 9 GtC yr-1

into the atmosphere (Canadell et al. 2009). While increased uptake from terrestrial plants

and oceans aids in offsetting this source, a net increase of 5 GtC yr-1 remains, thus

disrupting a system that had previously functioned near equilibrium for thousands of

years (Canadell et al. 2009; Le Quere et al. 2009). The Intergovernmental Panel on

Climate Change (IPCC) was formed to determine the causes, mechanisms and

consequences of human-driven changes to climate in order to guide efforts for adaptation

to, and mitigation of, this problem. A key aspect of the research synthesized by the IPCC

is the use of coupled models to simulate future climate (global circulation models,

2

GCMs) and the influence upon it of biogeochemical cycling by the land surface, which is

dominated by plants (Dynamic global vegetation models, DGVMs).

DGVMs are tools for integrating the effects of many different factors of climate

change on fundamental plant mechanisms to predict leaf and canopy level responses

across time and space that can then be used to characterize the overall effects on a region.

With the incorporation of more biological processes, simulations of the future are

becoming realistic. In addition, levels of certainty in IPCC assessment reports evaluating

the nature of climate change and its attribution to human activity are increasing (IPCC

2001, 2007). Two key sub-models of numerous DGVMs are the Ball et al. (1987) model

of stomatal conductance (gs) (and its derivatives including that of Leuning 1995) and the

Farquhar et al. (1980) model of photosynthesis (A). DGVMs that employ such enzyme

kinetic models tend to perform better than those using light use efficiency models for

estimating GPP (Schwalm et al. 2010). Together the Ball et al. (1987) and Farquhar et

al. (1980) models can be used to predict leaf-level gas exchange, the most basic point of

carbon and water exchange between the terrestrial and atmospheric pools. Incorporation

of mathematical descriptions of leaf-level processes represented by these equations into

DGVMs enables inferences about carbon and water cycling at the global level in the

context of various expected facets of climate change (Foley et al.1996; Sellers et al.1996;

Bounoua et al.1999; Moorcroft et al. 2001).

The integrity of DGCM models is significantly dependent on the accuracy of the

leaf-level models. Small errors have the potential to be amplified to substantial

magnitudes, so any effects of changing concentrations of atmospheric components such

as CO2 (slated to rise to 730-1020ppm by 2100; IPCC 2007) or O3 (40-60% increases

3

through 2100; IPCC 2007) on leaf-level model functioning must be accounted for. For

example, the acclimation response of components of the Farquhar et al. (1980)

photosynthesis model to conditions of elevated CO2 has been widely studied and

incorporated into modeling efforts (Wullschleger 1993; Drake et al.1997; Ainsworth &

Long 2005, Cramer et al.2001). However, less is known about acclimation of parameters

of the Ball et al.(1987) model of stomatal function. The effects of elevated [O3] have also

been less characterized than other aspects of environmental change, despite recognition

that O3 is the most damaging air pollutant to plants (Ashmore 2005; Karnosky et al.2007;

Paoletti et al. 2007). This study aims to address the effects of elevated O3 in isolation,

and in combination with elevated CO2, on the sensitivity of gs to A, fractional

atmospheric relative humidity (h) and [CO2], as described by the Ball et al. (1987) model.

While numerous models of gs exist, the Ball et al. (1987) model was used for this study as

it is one of the simplest but also provides accurate predictions under variable

environments (Damour et al. 2010; Mott & Peak 2010). Evidence for acclimation of this

nature will require parameterization of the Ball et al. (1987) model and its derivatives for

growth [O3] and [CO2] of interest to achieve precise predictions of climate and large-

scale carbon and water fluxes affecting ecosystem goods and services upon which

humans inextricably depend.

4

CHAPTER 2: GROWTH AT ELEVATED [O3] LEADS TO ACCLIMATION IN SOYBEAN THAT INCREASES THE SENSITIVITY OF STOMATAL

CONDUCTANCE TO PHOTOSYNTHETIC RATE, HUMIDITY AND [CO2] Introduction

Tropospheric O3 is one of the most damaging pollutants to plants, and background

[O3] have doubled over the last 50-100 years (Royal Society 2008). Current tropospheric

[O3] are sufficient to reduce crop yield, with annual crop losses for soybean alone

estimated between 1.8 and 3.9 billion dollars in the USA, and between 3 and 5.5 billion

dollars in China (Van Dingenen et al. 2009). Since O3-induced damage is known to

increase linearly with [O3] (Krupa et al. 1994) crop losses can be expected to rise further

as the Midwest USA and eastern China experience increases in mean July [O3] of 30 and

50 ppb, respectively, by the end of this century (Prather et al. 2003). O3 significantly

reduces net primary productivity (NPP) in forests and grasslands (King et al. 2005;

Wittig et al. 2007; Gonzalez-Fernandez et al. 2008) with effects being of sufficient

magnitude to play an important role in the future carbon cycle and climate change (Sitch

et al. 2007). In both natural and managed ecosystems O3 reduces plant water use, with

potential subsequent effects on atmospheric and soil water cycling (Ollinger et al. 1997;

Bernacchi et al. 2011).

The exchanges of gases between a leaf and the atmosphere, including CO2, water

and O3, are mediated by stomata. While the mechanisms regulating gs are not completely

understood, Ball et al.(1987) discovered an empirical, linear equation which relates gs to

A, h and atmospheric [CO2] at the leaf surface:

Eqn. 1 gs = g0 + m(Ah/[CO2])

where g0 is the y-axis intercept and m is the slope of the line. In combination with the

Farquhar et al. (1980) model of leaf photosynthesis, this model has been widely used to

5

predict leaf level gas exchange, which can be scaled up to the canopy and has been

incorporated into many ecosystem and global models of carbon and water cycling (e.g.,

Foley et al. 1996; Sellers et al.1996; Bounoua et al.1999; Moorcroft et al. 2001). In the

event that either the slope or intercept parameters of the Ball et al (1987) model change

with growth [O3], the model will require reparameterization at any given [O3] of interest

in order to ensure accurate predictions. Without this correction, any errors could scale-up

to result in substantial discrepancies when predicting large-scale carbon and water

cycling. However, while this possibility has been recognized for some time, lack of

mechanistic understanding and parameterization information has prevented such a

mechanism from being incorporated in modeling exercises (e.g. Ollinger et al.1997).

Reparameterization of the Ball et al.(1987) model would be necessary if the

sensitivity of gs to either A, h or [CO2] were to vary at different growth [O3]. It is

commonly observed that growth of diverse plant species, including soybean, at elevated

[O3] impairs A primarily by reducing photosynthetic capacity, with reduced gs being a

consequence rather than a driver of lower A (Reich 1987; Farage et al.1991; Farage &

Long 1995; Morgan et al. 2004; Paoletti & Grulke 2005). A reduction in photosynthetic

capacity will theoretically increase the gs required to support a given A. Consistent with

this prediction, there is a long record of experiments indicating that the ratio (i.e.

sensitivity) of gs to A increases with rising [O3], leading to reduced water use efficiency

(WUE; e.g. Reich & Lassoie 1984; Matyssek et al.1991; Tjoelker et al.1995; Maurer et

al.1997). In the context of the Ball et al.(1987) model, such responses would lead to

greater m. However, it is possible that other features of photosynthetic and stomatal

function are also altered at elevated [O3] leading to alternative outcomes. For example

6

growth at elevated [O3] reduced the response of steady-state gs to declining soil moisture

(Pearson & Mansfield 1993); slowed stomatal opening and closure in response to a range

of environmental factors including light (Paoletti & Grulke 2010); and reduced sensitivity

of gs to ABA (Wilkinson & Davies 2009). In addition, a network of cellular signaling that

drives stomatal closure in leaves and epidermal peels in response to acute O3 doses is also

being elucidated (Vahisalu et al. 2010). This presents the possibility that O3 may also be

working directly on the stomatal apparatus, rather than only indirectly via changes in

carboxylation rate. A concentration driven change in this relationship could also drive

changes in m for different growth [O3].

This study tested the prediction that growth of soybean at elevated [O3] results in

acclimation of gs to A, h, and/or [CO2], which manifests as greater values of the m

parameter from the Ball et al. (1987) model. To test this, soybean was grown at eight

[O3] (39-121 ppb) over its entire lifecycle at the soyFACE (soybean Free Air

Concentration Enrichment) facility in central Illinois. Laboratory measurements of leaf

gas exchange were used to parameterize the Ball et al. (1987) model, which was then

validated against independent measurements of leaf gas exchange in the field. The use of

FACE technology allowed O3 fumigation with minimal interference to the natural soil-

plant-atmosphere continuum experienced by crop plants. The homogeneous nature of the

field site in combination with the low genetic variation of the inbred soybean results in

low variability between plants that will aid in detection of treatment differences where

they occur (Leakey et al. 2006a). Soybean has often been used as a model crop for

examination of the effects of [O3] on an annual crop species (Morgan et al. 2003) and

represents an important food crop, ranking in the top four most important agricultural

7

crops. Additionally, the soybean-maize agroecosystem covered more than 80 million

hectares in 2008, equal to approximately 10% of the contiguous states of the U.S.

(USDA; http://www.nass.usda.gov).

Materials and Methods Field site and cultivation

This study was performed at the soyFACE facility near Champaign, IL (40°02'N,

88°14'W, 228 m above sea level), which covers 32ha of land in the heart of the Midwest

U.S. Corn Belt. SoyFACE is located on a tile-drained field with Drummer-Flanagan

series soil formed from loess and silt parent material deposited on the till and outwash

plains. Half of this area was planted with soybean, while the other 16ha were planted

with maize. Detailed descriptions of the SoyFACE facility and cultivation practices have

been previously reported (Rogers et al. 2004; Leakey et al. 2004). In 2009 soybeans were

planted on June 9.

The experiment was conducted as a gradient design, with eight 20m-diameter

octagonal plots each fumigating soybean with a different [O3] over the growing season. A

5.7 x 2.7 m subplot within each FACE plot was planted with the cultivar used in this

study (Glycine max cv 93B15; Pioneer Hi-Bred). Target [O3] ranged from 40 to 200ppb.

This corresponded to the range of [O3] from current, ambient [O3] in the Midwest U.S. up

to elevated [O3] experienced currently in areas of the world with severe air pollution and

predicted to become more common over a wider geographic area over the course of this

century (Prather et al.2003). The plots were positioned at least 100m away from one

another to prevent cross-contamination. Details of the FACE technology used at

SoyFACE have been described previously (Miglietta et al. 2001; Morgan et al. 2004).

The O3 fumigation system was turned off at night and when leaves were wet. The

8

growing season average of the maximum daily 8-h [O3] for each of the eight plots in the

experiment was 39, 47, 56, 59, 75, 94, 96 and 121 ppb O3.

Model parameterization Photosynthetic gas exchange of mature, sun leaves was measured under

laboratory conditions in order to parameterize the Ball et al. (1987) and Farquhar et al.

(1980) models at two crop developmental stages during the growing season: during

vegetative development and during seed fill. Over four dates during each developmental

stage (July 21-25 and Aug 21-24) parameterization data was collected from two leaves

per FACE plot. On a given measurement day, mature, sun-exposed leaves were collected

pre-dawn from four of the eight O3 rings. The four plots from which leaves were assessed

on each day were randomly selected in order to prevent confounding growth [O3] with

measurement date. Leaves were brought back to the lab where petioles were submerged

in water, re-cut and kept in water for the duration of measurements. Predictions of gs

based on parameterization data from leaves handled in this manner have been

successfully validated against leaf gas exchange of soybean in the field (Leakey et al.

2006b).

Rates of leaf photosynthetic gas exchange were measured using four open gas

exchange systems with 2 cm2 leaf chambers housing infrared gas analyzers to measure

fluxes of both CO2 and water (LI-6400; LI-COR Inc., Lincoln, NE, USA). These systems

were calibrated prior to measurements by passing gas of known CO2 and H2O

concentrations through the system. A, gs and intercellular [CO2] (ci) were calculated using

equations from von Caemmerer & Farqhuar (1981). Data were collected following a

simplified version of the protocol described by Leakey et al. (2006b) in which

photosynthetic photon flux density was maintained at 1500 µmol m-2 s-1, leaf temperature

9

was 25 ±.3 °C and the vapor pressure deficit from leaf to air was < 2 kPa while [CO2]

entering the chamber was varied stepwise (400, 150, 250, 250, 450, 650, 850, 1200,

1500ppm). A minimum of 20 minutes was allowed for A and gs to completely stabilize

before data were collected at each [CO2].

Linear least squares regression was used to estimate the slope (m) and intercept

(g0) parameters for each individual leaf. A second linear least squares regression was

used to quantify the relationship between m and growth [O3]. There was no significant

effect of developmental stage on m (p=0.3713), so all data were pooled for subsequent

analyses. There is significant variation in [O3] over time and distinct acute and chronic

effects on plant physiology have been characterized (Chen et al. 2009). Therefore, the

regression analysis of m and [O3] was repeated using averages of maximum daily 8-h

[O3] over periods varying in duration from 1 day to 59 days (i.e. average since fumigation

began) prior to collection of the parameterization data. The maximum carboxylation

capacity of Rubisco (Vcmax) and the maximum capacity for RubP regeneration by electron

transport (Jmax) were parameterized by fitting the responses of A to ci (A/ci curves) as

described by Ainsworth et al. (2007).

Model validation with in situ gas exchange Mid-day gs and A of field grown soybean were measured in situ on two dates

(July 27 and September 1) during the 2009 growing season with the same gas exchange

systems used and described for collection of the parameterization data. Immediately prior

to collecting gas exchange measurements, PPFD and air temperature were measured at

the top of the canopy and the leaf chambers of all gas exchange systems were configured

to duplicate these ambient conditions. The [CO2] entering the leaf chambers was set to

400 ppm and humidity was maintained between 60-70% for the duration of the sampling

10

period. Four gas exchange systems were used, each one measuring soybean growing in

two FACE plots. The plots measured by a particular gas exchange system were

randomized between measurement dates to avoid confounding equipment with [O3]. The

youngest fully expanded leaf was measured on three plants in each plot.

For each leaf, gs was predicted using estimates of m for the corresponding plot

(from best-fit line of m versus average 8-h [O3] to date) in conjunction with A, h and

[CO2] measured in the chamber for each leaf. The accuracy of these predictions was

evaluated by regression against measured gs.

Results During laboratory parameterization measurements, systematic variation in [CO2]

at the leaf surface resulted in a wide range of values for gs and A (Fig 2.1). This allowed

m and g0 to be estimated from a regression of gs against Ah/[CO2] (Fig. 2.2). g0 was not

significantly different from zero, which allowed for the relationship between gs and the

index comprising A, h and [CO2] to be treated as a straight line passing through the

origin.

There was a significant, positive correlation between m and the average 8-h [O3],

which varied in strength depending on the number of days prior to measurements over

which [O3] was averaged (Figs. 2.3,2.4). Long-term [O3] (>45 days) generally explained

a greater fraction of variation in m than short-term [O3]. However, the strength of the

correlation was moderate (r2 > 0.4) when using [O3] on the day prior to measurements or

average [O3] over the preceding 6 days. Since m was most closely related to long-term

average [O3], all subsequent analyses were based on the relationship between m and the

average 8-h [O3] since fumigation began after crop emergence. In this case, m increased

by approximately 1 for every 10 ppb rise in average 8-h [O3] (Fig. 2.4; p < 0.0001, r2 =

11

0.49). Consequently, while m of soybean grown at current, ambient [O3] of 39 ppb was

8.5, this rose by 92% to 16.3 at 121 ppb. In summary, the effect of rising [O3] on gs can

be visualized (Fig. 2.5) as the combined effect of greater m (steeper slope) as well as a

lower operating point within the range of [O3] tested in this study as described by the Ball

et al. (1987) model.

To assess the applicability of the laboratory parameterization to simulation of

crop physiological status in the field, the model predictions of gs were compared against

measurements of in situ gs (Fig. 2.6). Model predictions corresponded well with

measured values of gs, with the slope of the regression line between measured and

modeled gs being close to one (gmeasured=.94*gmodeled + .12; slope: p<0.0001; intercept

p=0.0287; r2=0.57).

To determine whether the effect of varying growth [O3] on m might be related to

changes in photosynthetic capacity, m was regressed against Vcmax and Jmax. There were

significant, negative relationships between m and Vcmax (Fig. 2.7; p<0.0001, r2=0.48), as

well as between m and Jmax (Fig. 2.8; p=0.001, r2=0.29).

Discussion As predicted, this study found that variation in growth [O3] caused acclimation in

soybean that altered the sensitivity of gs to A, h and [CO2]. This was determined by

parameterizing the Ball et al. (1987) model of gs for soybean across a range of [O3] (39 -

121 ppb) at a FACE facility in central Illinois, and finding a significant, linear increase in

a key model parameter (m) with greater growth [O3]. This provides unique evidence from

field experimentation for the need to reparameterize a key component of models that

simulate leaf, ecosystem and global fluxes of carbon, water and trace gases in order to

predict the impacts of tropospheric [O3] as a major present day pollutant and component

12

of future global environmental change. It also provides the necessary data to support

reparameterization of models over a wide range of [O3] for soybean in, at least its

primary area of production, the U.S. Midwest. Use of the Ball et al. (1987) model without

parameterization for [O3] produces estimates of gs almost 50% lower than predictions

based on the parameterized run, suggesting the potential for substantial discrepancies

between models considering growth [O3] or not.

The increase in m with rising [O3] is consistent with previous studies’ findings of

an increase in the ratio of gs to A with rising [O3] (e.g. Reich & Lassoie 1984; Matyssek

et al. 1991; Tjoelker et al. 1995; Maurer et al. 1997), when measured under consistent

conditions of h and [CO2]. While this study does not explicitly test the mechanism

altering m in response to [O3], the increase in m is also consistent with the reduced Vcmax

often observed with increasing O3 exposure (Reich 1987; Farage et al. 1991; Farage &

Long 1995; Morgan et al. 2004; Paoletti & Grulke 2005). This may be resulting from

lower Vcmax decreasing the A for a given amount of CO2 entering the leaf, as controlled

by gs. Consistent with this expectation, there were significant negative, correlations

between m and in vivo measures of photosynthetic capacity, Vcmax and Jmax. While

reduced photosynthetic capacity would theoretically drive increases in m, it is possible

that other stomatal and/or mesophyll responses to [O3] either enhanced or counteracted

the response of m. For example, acute doses of elevated [O3] have been shown to drive a

cellular signal transduction pathway in guard cells that leads to stomatal closure

(Vahisalu et al. 2010). In addition, growth at elevated [O3] has also been observed to

change stomatal anatomy in some cases (Woodward & Kelly 1995). Further investigation

is needed to address these open questions.

13

A relationship between m and Vcmax could be extremely useful, where it exists, as

it could allow estimation of the m based on measured Vcmax rather than a separate set of

parameterization measurements. Few measurements of m exist across species and

conditions relative to Vcmax often resulting in use of default values that only consider

photosynthesis type (C3 or C4). The increase in m observed for soybean in elevated [O3]

demonstrates that m can vary within a species under different atmospheric conditions. It

follows that variation in m across species is likely to exist and could also be quite

variable, especially considering the range of photosynthetic capacity, even within

functional groups, as well as different inherent sensitivity to O3 and typical growing

conditions (Ashmore 2002). The relationship between m and Vcmax may arise from [O3]

damage affecting Ball et al. (1987) parameters differently, therefore conclusions

regarding this relationship in general should not be made from this data alone. At least for

conditions of elevated [O3] however; a relationship between m and Vcmax could be used to

generating more appropriate m values than a stock constant. Thus errors in estimates

employing the Ball et al. (1987) model could theoretically be reduced for

species/conditions combinations with known Vcmax values without explicit tests of m.

Based on parameterization measurements, an equation was developed to

determine the appropriate m value for use in predictions requiring the Ball et al. (1987)

function to account for acclimation to growth [O3]. This equation determines m based on

the average growth [O3] of interest (m = .0944*(average [O3]) + 4.815). The positive

slope of this equation quantifies the observed tendency for the necessary m to increase

with average growth [O3]. Incorporation of this equation into the Ball et al. (1987) model

14

will enable predictions that compensate for acclimation to [O3] observed here in

soybeans.

Based on both the parameterization equation developed in this study and the

projection of a continued rise in [O3] (IPCC 2007), predictions of future ecosystem

functioning using the current, unparameterized Ball et al. (1987) model will result in

lower gs than with the model corrected for growth [O3]. Even a small misestimation of

such an important leaf-level process can lead to increasingly substantial errors when

scaled to the canopy, regional and global level for predictions regarding climate change

and carbon and hydrologic cycles. O3-induced reductions in gs, for instance, reduce

transpiration, and thus evapotranspiration (ET) (Bernacchi et al. 2011), which can

contribute directly to lower precipitation or indirectly to higher radiation budget via

reduced cloud cover (Sellers et al. 1997). Lower gs also results in less CO2 uptake and

NPP in forests, grasslands, and agricultural crops (King et al. 2005; Morgan et al. 2006;

Gonzalez-Fernandez et al. 2008; Wittig et al. 2007), thereby reducing sink potential for

vegetation.

Accumulation of CO2 in the atmospheric pool resulting from O3-induced declines

in NPP represents indirect radiative forcing that will further drive climate change and all

associated consequences (Sitch et al. 2007). Simultaneous effects on the water cycle from

altered ET can also impact plant drought stress (Bernacchi et al. 2011) that could

influence productivity, and hence NPP and climate change. The presence of gs at the

initial steps in calculation of these estimates means that accurate parameterization of the

Ball et al. (1987) model for growth [O3] is necessary to prevent amplification of small

15

discrepancies into serious misestimates of future climate and carbon and hydrologic

cycles.

It is also important to consider inference space in the use of the Ball et al. (1987)

model with incorporated average growth [O3] parameterization. For instance, it may be

desirable to generate predictions of gs for values of atmospheric [CO2] expected in the

future. No stomatal acclimation was observed to occur between field-grown soybean

exposed to conditions of either ambient or elevated [CO2], therefore no additional

parameterization is necessary to account for conditions of elevated growth [CO2] alone

(Leakey et al. 2006b). However, the effects of elevated concentrations of O3 and CO2 on

A and seed yield are known to interact with one another. Elevated [CO2] reduces the

negative effects of elevated [O3] on A in soybean grown in heightened concentrations of

both gases (Morgan et al. 2003). Further investigation is needed to determine whether

there is an interaction between these gases in terms of their effect on sensitivity of gs to

A, h and [CO2] that will require further parameterization of the Ball et al. (1987) model.

16

Figures

PAR

(µm

ol m

-2 s

-1)

0

200

400

600

800

1000

1200

1400

1600

[CO

2] (µ

mol

CO

2 mol

-1)

Vpd

L (k

Pa)

0

1

2

3

4

Time (min)0 100 200 300 400

A (µ

mol

CO

2 m-2

s-1

)

0

5

10

15

20

25

30

35

g s (m

ol H

2O m

-2 s

-1)

0.0

0.2

0.4

0.6

0.8

1.0

Figure 2.1. Representative course of (a) incident photosynthetic photon flux density (PPFD), [CO2] and vapor pressure deficit between the leaf and atmosphere (VpdL) during gas exchange measurements of (b) steady-state net photosynthetic CO2 assimilation (A) and stomatal conductance (gs) during data collection from mature sun leaves of field-grown soybean for model parameterization.

a)

b)

17

Ah/[CO2]0.00 0.01 0.02 0.03 0.04 0.05 0.06

g s (m

ol H

2O m

-2 s

-1)

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

111 ppb O3111 ppb O3 56 ppb O3 56 ppb O3

Figure 2.2. Representative relationship of stomatal conductance (gs) with the product of net photosynthetic CO2 assimilation (A) and fractional relative humidity (h) divided by the [CO2] of mature sun leaves of field-grown soybean grown at different [O3]. Points represent an individual gas exchange measurement of a single leaf. Only data for which [CO2] > 150 µmol mol-1 were used. Representative data from two plants grown at high (111 ppb; gs = 17.4109*Ah/[CO2] - .0107; p<0.0001) and low [O3] (55 ppb; gs =8.7880*Ah/[CO2] + .0365; p<0.0001) [O3] are depicted along with the regression fits of the Ball et al. (1987) model to the measured data from each leaf.

18

Average [O3] ppb (fumigation to measurement)

0 20 40 60 80 100 120 140

m

0

5

10

15

20

25

Figure 2.3. Scatterplot of maximum daily 8-h [O3] averaged from beginning of fumigation to measurement and slope parameter from the Ball et al. (1987) model (m). Each point represents data from a single leaf.

m = 0.0944*[O3] + 4.8154 p < 0.0001 r2=0.49

19

period prior to estimation of m used for estimation of O3 exposure (days)0 10 20 30 40 50 60

R2 (f

rom

m v

s. [O

3] gr

aph)

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

Figure 2.4. The fraction of variability (r2) in the slope parameter from the Ball et al. model (1987; m) explained by regression of m with [O3] depending on the period prior to estimation of m over which O3 exposure was calculated (1 – 56 days).

20

Figure 2.5. Summary of relationship between gs and Ah/[CO2] for soybean grown at 40, 60, 80, 100 and 120 ppb O3 drawn from lines of best fit for regression relationships of m versus [O3] (Fig 1.3) and light-saturated gs versus [O3]. g0 was fixed at 0.. Lines represent the increasing slope parameter in the Ball et al. (1987) model (m) resulting from acclimation to greater growth [O3]. Black points represent the rates of photosynthetic gas exchange observed under light-saturated conditions in the field (gs was observed to decline significantly with greater growth [O3]). Grey points represent the decline in gs that would be predicted to result from reduced A at greater [O3] assuming no change in m. The insert shows the underestimation of gs resulting from assuming no change in m relative to the observed response where m increased with greater [O3].

120ppb [O3]

60ppb [O3]

40ppb [O3]

80ppb [O3]

100ppb [O3]

21

gmodeled (mol H2O m-2 s-1)

0.0 0.2 0.4 0.6 0.8 1.0

g mea

sure

d (m

ol H

2O m

-2 s

-1)

0.0

0.2

0.4

0.6

0.8

1.0

Figure 2.6. Relationship between stomatal conductance (gs) of the youngest fully expanded leaf of soybeans grown in the field under various growth [O3] (gmeasured) and gs predicted by the model parameterized from laboratory gas exchange measurements (gmodeled). Field measurements of gs were taken at mid-day on two dates during the 2009 growing season. Red line shows 1 to 1 line.

gmeasured = .9286*gmodeled + 0.1215 p < 0.0001 r2 = 0.57

22

Vc,max (µmol m-2 s-1)

0 20 40 60 80 100 120 140 160 180

m

0

5

10

15

20

25

Figure 2.7. Scatterplot of slope parameter from the Ball et al. (1987) model (m) and the maximum carboxylation capacity of Rubisco, estimated from A/ci curves using the Farquhar et al. (1980) (Vcmax). Each point represents data from a single leaf. The line of best fit and statistical results from regression analysis are shown.

m = 19.85 - 0.08*Vcmax p < 0.0001 r2 = 0.48

23

Jmax (µmol m-2 s-1)0 50 100 150 200 250 300

m

0

5

10

15

20

25

Figure 2.8. Scatterplot of slope parameter from the Ball et al. (1987) model (m) and the maximum rate of electron transport, estimated from A/ci curves using the Farquhar et al. (1980) (Jmax). Each point represents data from a single leaf. The line of best fit and statistical results from regression analysis are shown.

m = 19.9 - 0.05*Jmax p = 0.0014 r2 = 0.29

24

CHAPTER 3: GROWTH AT ELEVATED [CO2] AMELIORATES THE CHANGES IN SENSITIVITY OF STOMATAL CONDUCTANCE TO

PHOTOSYNTHETIC AND ENVIRONMENTAL SIGNALS AT ELEVATED [O3] Introduction

Concentrations of both O3 and CO2 are rising in the atmosphere. These gases are

known to exhibit opposing effects on plant productivity and ecosystem function through

different mechanisms of action on carbon metabolism and water use. The effects of [CO2]

on A and water use are well understood and have been incorporated into models of

ecosystem and global biogeochemistry. Plant responses to elevated [O3] have also been

extensively studied and modeled, but not to the same degree. Significant uncertainty still

remains regarding how plants and ecosystems will respond as they experience

simultaneous increases in both CO2 and O3 this century. Recently, acclimation to elevated

[O3] that altered the sensitivity of gs to A, h and [CO2] has been reported in soybean

(Chapter 2). Previously it has been demonstrated that there is no such change in soybean

grown at elevated [CO2] (Leakey et al 2006b). This study investigates how model

representations of stomatal function will need to be parameterized in order to account for

the interaction between elevated [CO2] and elevated [O3] in the future.

O3 levels have already increased approximately four-fold from 10ppb before the

Industrial Revolution (Volz & Kley 1988) to current background concentrations of 35-

40ppb (Royal Society 2008). O3 acts to reduce the activity and amount of Rubisco,

leading to reduced assimilation capacity (Reid et al.1998). Reductions in A with elevated

[O3] have been observed in both crop and tree species for decades (Reich & Amundson

1985). Damages to rice, soybean, maize and wheat were already estimated between $14-

26 billion using data from 2000 (Royal Society 2008). [O3] can be expected to rise a

further 40-60% through the rest of this century (IPCC, 10.4.3, 2007) and global crop

25

losses are expected to continue rising even under somewhat optimistic scenarios of future

[O3] (Van Dingen et al. 2009). Significant O3-induced reductions in NPP in forests and

grasslands demonstrate that the damaging effects of O3 also exist in natural ecosystems

(King et al. 2005; Gonzalez-Fernandez et al. 2008; Wittig 2007). The diminished carbon

sink capacity of all vegetation from such [O3]-induced reductions in NPP are enough to

drive climate change to the same degree as the direct radiative forcing of O3 (Sitch et al.

2007). Elevated [O3] also lowers gs, causing reduced plant water use in both natural and

managed ecosystems (Ollinger et al. 1997, Bernacchi et al. 2011). Meanwhile reductions

in gs from O3 are of a magnitude sufficient to alter regional and continental hydrology

(Sellers et al. 1996).

[CO2] is expected to reach 730-1020 ppm by the end of 2100 (IPCC 2007, Meehl

et al. 2007). Increased levels of CO2 contribute to a rise in A as [CO2] around Rubisco

becomes more saturated while simultaneously outcompeting oxidative photorespiration

(Drake et al.1997). Acclimation of A to [CO2] results in reductions in Vcmax and Jmax, with

a shift in control of A from Vcmax towards Jmax (Ainsworth & Long 2005; Bernacchi et al.

2005). Unlike [O3], elevated [CO2] increases A for many vegetation types including trees

and crop species, which represents the most important factor increasing yields in elevated

[CO2] conditions (Will & Teskey 1997; Norby et al.1999; Ainsworth et al. 2002; Rogers

et al. 2004; Bernacchi et al. 2006). Another common response to elevated [CO2] is a

reduction in gs (Ainsworth et al. 2002; Morgan et al. 2003; Wittig et al. 2007). Similarly

to [O3], the decrease in gs resulting from elevated [CO2] also has the potential to alter

different aspects of climate including temperature and precipitation (Field et al. 1995).

The 6% rise in global mean river-runoff since the start of the industrial revolution, for

26

example, can be attributed to [CO2]-induced reductions of gs (Betts et al. 2007).

Additional runoff may contribute to an increase in precipitation, which is expected to rise

between 3 and 11% with a doubling of [CO2] (Wigley & Jones 1985). CO2 will be one of

the greatest determinants of twenty first century climate (Houghton et al. 2001) and is a

component in many models of future climate change. A study of six global dynamic

vegetation models found [CO2] to be the most important determinant of NPP, which

increased with elevated [CO2] in all models (Cramer et al. 2001).

There are many examples in the literature of harmful effects of [O3] on plants

being reduced or fully ameliorated when exposed to concomitant elevated [CO2]. In two

tree species of differing O3 sensitivity, the positive effects of [CO2] were completely

ameliorated by simultaneous treatment with elevated [O3] (Karnosky et al. 2003). A

study of six plants representing a variety of functional groups and photosynthetic

pathways found that O3-induced reductions in A and relative growth rate were not

observed in plants that were grown in conditions of elevated [CO2] in addition to elevated

[O3] (Volin et al. 1998). Species with the highest gs were most susceptible to O3 damage

and it was postulated that the amelioration of O3 effects could result from reduced gs,

with elevated [CO2] decreasing amount of O3 able to enter the leaf and producing damage

(Volin et a. 1998). A meta-analysis of [O3] effects on soybean also found significant

decreases in [O3]-induced damaged with elevated [CO2] and cite restricted uptake of [O3]

arising from reduced gs as a possible mechanism of this result (Morgan et al. 2003).

The exchanges between the leaf and atmosphere of CO2 and O3, as well as water,

volatile organic carbons and many pollutants are controlled by stomata. gs is a measure of

the ease with which guard cells allow diffusion of these gases between leaf and

27

atmosphere in mol m-2s-1. Predictions of gs can be made using an empirical model

developed by Ball et al.(1987) based on a linear relationship between gs and A, h, and

[CO2]:

Eqn. 1 gs = go + m(Ah/[CO2])

where g0 is the y-intercept and m is the slope of the line. This model, or its derivatives, is

used in combination with the Farquhar et al. (1980) equations to predict leaf level gas

exchange. They are also incorporated into many large-scale models of ecosystem

functioning and carbon and water cycling (e.g., Foley et al. 1996; Sellers et al. 1996;

Bounoua et al. 1999; Moorcroft et al. 2001). Changes in the sensitivity of gs to A, h

and/or [CO2], represented by m, resulting from the combined effects of elevated CO2 and

O3 must be incorporated into models employing the Ball et al. (1987) function to achieve

accurate estimates of future conditions. The importance of exploring the potential

acclimation of m in varying environmental conditions has been recognized for some time

(Ollinger et al. 1997) and explicit analyses of m in response to elevated [CO2] and [O3]

has been conducted for each gas separately (Leakey et al. 2006b, Chapter 2). However, m

has yet to be examined for elevated levels of both gases in combination, as they are

ultimately predicted to exist in real-life.

The effects elevated [CO2] and [O3] on m have been observed individually with

different results. m increases linearly with [O3] in field-grown soybean exposed to a range

of growth [O3] (Chapter 2) but there was no significant change in m between ambient and

elevated [CO2] treatments (Leakey et al. 2006b). Vcmax experienced substantial decreases

with elevated [O3] (Chapter 2), functionally reducing the amount of A that could be

supported by a given gs. Such an increase in the ratio of gs to A is a common response to

28

elevated [O3] (Reich & Lassoie 1984; Matyssek et al. 1991; Tjoelker et al. 1995; Maurer

et al. 1997). m must increase to maintain the balance of the Ball et al. (1987) model with

a rise in gs /A, which is consistent with the increase in m observed in the field [O3]

experiment (Chapter 2). If O3-induced reductions in Vcmax contribute to the increase in m

observed in the field, m will similarly be expected to increase with O3 at 400ppm CO2. As

[CO2] rises however, additional stomatal closure will progressively limit [O3] uptake,

dampening the effect of [O3] on Vcmax and subsequent increases in m. It follows that

parameterization of the Ball et al. (1987) model under combinations of a range of

increases in both [O3] and [CO2] would be expected to produce m values for different

treatment combinations representative of this pattern. The small decrease in Vcmax in

elevated [CO2] (Leakey et al. 2006b) may not have been of sufficient magnitude to

produce a visible effect on m. Alternatively, other features of stomatal and/or

photosynthetic functioning altered with elevated [O3] could drive opposing changes in m.

For example, [O3] can also slow stomatal functioning and reduce the sensitivity of gs to

light and ABA (Pearson & Mansfield 1993; Paoletti & Grlke 2010; Wilkinson & Davies

2009).

The steady, homogenous rise projected for atmospheric [CO2] (IPCC 2007) in

combination with the heterogeneous and dynamic nature of atmospheric [O3] (Royal

Society 2008) suggests plants will be exposed to a multitude of combinations of these

gases in the future. The difficulty of modeling interactions combined with the

consequences of misestimations of gs in models necessitates characterization of the

general response surface of the variation in m with [CO2] and [O3]. The objective of this

study was to test the effects of co-occurring elevated [CO2] and [O3] on m on a model

29

crop for [O3], soybean, due to its prevalence as a food crop and the substantial land area

of the soybean-maize agroecosystem (10% of the contiguous states of the USA; USDA

www.nass.usda.gov). Soybean was grown in environmental growth chambers and

exposed to a combination of one of four [CO2] set points (400, 600, 800, 1200 ppm) and

one of four [O3] set points (0, 50, 100, 150 ppb) in a fully factorial design (Table 3.1).

The large number of treatment combinations employed in this experiment rendered field-

grown assessment using FACE technology impractical despite limitations of artificial

growth environments. Laboratory measurements for parameterization of the Ball et al.

(1987) model were made on half of the plants in the style of Leakey et al. (2006b).

Parameterizations were validated using light saturated measurements of A for the

remaining plants. Growth at elevated levels of [O3] is hypothesized to increase the

sensitivity of gs to Ball et al. (1987) parameters A, h, and/or [CO2], represented by m,

with elevated [O3]. Addition of elevated [CO2] is expected to increasingly diminish, and

eventually mask this response.

Materials and Methods Plant Material

Soybean (Glycine max cv 93B15; Pioneer Hi-Breed) was grown in 20L pots of

sterilized soil treated with 50% Long Ashton solution. Seeds were treated with inoculant

(Rhizo-Stick, Becker Underwood, Inc., http://oda.state.or.us/fertilizer) according to

product instructions. Three seeds were planted in each pot and later thinned to one plant

per pot. Due to limitations in available growth chambers, the experiment was conducted

in two runs with half of all treatment combinations represented in each run.

The experiment was conducted as a 2-way complete factorial design with plants

receiving one of four set point [CO2] (400, 600, 800, 1200 ppm) as well as one of four set

30

point [O3] (0, 50, 100 or 150 ppb). Target [O3] concentrations correspond to the range of

[O3] between current Midwest U.S. conditions and conditions of highly polluted areas,

the latter of which being likely to grow in geographic extent through the end of this

century (Prather et al. 2003). Target [CO2] encompass levels near current conditions

through levels expected for 2100 (IPCC 2007). Twelve plants were placed in each of

eight environmental growth chambers. Each set of twelve plants received a different

combination of [CO2] and [O3] treatments, controlled by the environmental growth

chamber fumigation system. Experiment-long averages and standard errors for [CO2] and

[O3] levels are given in Table 1. Light, humidity and temperature were the same for all

treatments.

Each chamber contained high pressure sodium bulbs and metal halide bulbs and

was set to replicate a 14hr day. Lights in the chamber were lowered to 40cm above the

chamber bottom and raised in 10cm increments as plants grew. This allowed for light

levels between 1000 and 1200 µmol/ms PAR depending on position within the chamber.

Plants were rotated within chambers every two days and between chambers every week

to remove the effect of chamber as a confounding variable and allow for consideration of

individual plant as replicates. Humidity was set to 70% and day/night temperatures were

set to 25/22 ºC for all chambers. Plants were watered every other day with 50% Long

Ashton solution (12.5 mm NH4NO3) throughout the experiment. After four weeks plants

were also watered with tap water on days when they did not receive nutrient solution.

Model Parameterization Photosynthetic gas exchange of mature, upper-canopy leaves was measured under

laboratory conditions on intact plants to allow for parameterization of the Ball et al.

(1987) and Farquhar et al. (1980) models, as described previously (Chapter 2). Over

31

twelve days for each growth period, parameterization data was collected from six leaves

per [CO2] and [O3] treatment combination. On each day, four to six plants from different

treatments were removed from the chambers before the lights went on in the morning and

a mature, upper leaf from each was measured while still affixed to the plant. The

treatment combinations measured on a given day were selected at random to prevent

confounding [CO2] and [O3] treatments with measurement date.

Rates of leaf photosynthetic gas exchange were measured using four to six open

gas exchange systems with 2 cm2 leaf chambers housing infrared gas analyzers to

measure fluxes of both CO2 and water (LI-6400; LI-COR Inc., Lincoln, NE, USA). These

systems were calibrated prior to measurements by passing gas of known CO2 and H2O

concentrations through the system. A, gs and ci were calculated using equations from von

Caemmerer & Farquhar (1981). Data were collected following a simplified version of the

protocol described by Leakey et al. (2006b) in which photosynthetic photon flux density

was maintained at 1500 µmol m-2 s-1, leaf temperature was 25±.3 ºC and the vapor

pressure deficit from leaf to air was <2 kPa. [CO2] entering the chamber was varied step-

wise (400, 50, 150, 250, 350, 450, 650, 850, 1200, 1500 ppm). A and gs were allowed to

stabilize for at least 20min at each [CO2] before data were collected. Parameterization

values obtained in this fashion validated well with in situ gas exchange measurements

(Leakey et al. 2006b).

The slope (m) and y-intercept (g0) parameters were estimated via linear least

squares regression for each individual leaf. The effect of growth [CO2] and [O3] on m was

assessed using a two by two analysis of variance in the MIXED procedure of SAS (SAS

Institute, Vary, NC, USA) where growth [CO2] and [O3] were both treated as fixed

32

effects. The threshold of significance was defined as a probability level of p≤0.1 in order

to avoid type two errors given the low replicate size. The maximum carboxylation

capacity of Rubisco (Vcmax) as well as the maximum capacity for RubP regeneration by

electron transport (Jmax) were parameteriezed as described by Ainsworth et al. (2007) by

fitting the responses of A to ci (A/ci curves).

Model Validation Light saturated photosynthetic gas exchange was measured with the leaf chamber

of the gas exchange systems configured to duplicate daytime conditions in the

environmental growth chamber (with the exception of [O3]): [CO2] entering the leaf

chamber at 400ppm, PPFD at 1500 µmol/ms PAR, temperature at 25 ºC and VpdL below

1.5. The center trifoliate of a mature, upper leaf was measured for five to six plants in

each [CO2] and [O3] treatment combination. These six plants were separate from those

used for parameterization of the model.

gs was predicted for each leaf using the earlier parameterization as well as the A, h

and [CO2] measured in the chamber. The accuracy of these predictions was evaluated by

regression of each predicted gs value against its corresponding measured gs.

Results Systematic variation in [CO2] entering the leaf chamber during laboratory

parameterization measurements resulted in a wide range of values for gs and A (Fig. 3.1),

which allowed m and g0 to be estimated from a regression of gs against Ah/[CO2] (Fig.

3.2). Average g0 was not significantly different from zero, allowing the relationship

between gs and the index comprising A, h and [CO2] to be treated as a straight line

passing through the origin.

33

There was a significant interaction between [CO2] and [O3] on m (Fig. 3.3;

p=0.089). There were significant increase in m with [O3] at 400 ppm CO2, but not at 600,

1000, 1200 ppm CO2 (Fig. 3.3; p=0.0073). There was also an interaction of [CO2] and

[O3] on Vcmax (Fig. 3.4, p=0.0005) and Jmax (Fig. 3.5; p=0.0002). Again [O3] effects on

Vcmax and Jmax occurred only at 300 ppb [CO2].

To ensure the relevance of the parameterization, model predictions of gs were

compared against measurements of light saturated gas exchange taken from the chamber

plants not measured for parameterization (Fig. 3.6). Measured and modeled values of gs

were well correlated (r2=0.71) and exhibited reasonable agreement (gmeasured=.77*gmodeled

+ .02; slope p<0.0001; intercept p=0.1591). Bias that did exist may be attributable to

leaves measured for validation receiving a smaller O3 dose than leaves measured for

parameterization since validation measurements were taken the day before

parameterization measurements began. Resulting overestimations of gs would manifest as

a slope value below 1, as seen here. m was regressed against Vcmax and Jmax to assess

whether the effect of varying growth [CO2] and [O3] combinations on m might be related

to photosynthetic capacity. Significant, negative relationships existed between Vcmax and

m (Fig 3.7; p<0.0001) and Jmax and m (Fig. 3.8; p=0.0017), but only for treatments

including 400 ppm [CO2].

Discussion Consistent with predictions, there was a significant interaction effect of [CO2] and

[O3] on the m parameter of the Ball et al. (1987) model, which represents sensitivity of gs

to A, h and [CO2]. Elevated [CO2] ameliorated the effect of [O3] on m. m progressively

increased with greater growth [O3] under 400 ppm [CO2] (p=0.0005) but there was no

effect under conditions of 600, 800 or 1200 ppm CO2 (p=0.7299). Results at 400 ppm

34

CO2 were consistent with parameterization results of field-grown soybean exposed to

varying [O3] from (39-121 ppb) which also found an increase in m with [O3]. The lack of

variation in m with greater [CO2] on a background of low [O3] is also consistent with that

observed in field-grown soybean at elevated [CO2] (Leakey et al. 2006b). The

disappearance of O3 effects on m with elevated [CO2] also parallels the interaction of O3

and A found in previous research focusing on A and seed yield (Volin et al. 1998,

Eichelmann et al. 2004). This is consistent with a scenario where greater [CO2]

ameliorates reductions in A as well as the acclimation in sensitivity of gs to A, h, [CO2].

These results demonstrate that elevated [O3] increases m but this effect is completely

removed at some level of [CO2]. Thus, models incorporating the Ball et al. (1987)

function must be parameterized for growth [O3] and [CO2] for accurate predictions of

these conditions. Predictions of gs will be lower when models are not parameterized to

account for greater m at elevated [O3]. This could lead to overestimation of WUE and

underestimation of transpiration that can scale up to large errors in predictions of climate

and hydrologic cycles.

Regardless of treatment combination, m was negatively correlated with Vcmax.

This was also seen when the effect of [O3] on m was examined in field-grown soybean

(Chapter 2). Although m did not change along with Vcmax decreases in field grown

soybean exposed to elevated [CO2] the reduction in Vcmax reported was small (4-6%;

Leakey et al. 2006b) and may not have been sufficient to discern a relationship even if

one was present. The persistence of the negative correlation between m and Vcmax across

chamber and field studies warrants further examination of the relationship between m and

Vcmax aimed at determining whether this relationship due to a biological link or merely a

35

statistical artifact. The correlation was present within [CO2] treatments and equations

describing the relationship between Vcmax and m were statistically indistinguishable

between [CO2]. Increases in m with reductions in Vcmax are consistent with an increase in

gs/A due to Vcmax contributing to raise m. A relationship between Vcmax and m, even if

only extant in elevated [O3], could be helpful in improving models with the Ball et al.

(1987) function for predictions of future conditions since typically only standard m values

are used for C3 (~10) and C4 plants (~4). With the observed variation here and the breadth

of photosynthetic capacities among plants within a functional group, it is likely that m

values for different plants exist and are quite variable. Scaling of leaf level models

employing the Ball et al. (1987) function to the canopy, regional or global level may

amplify small errors in gs, making minimization of errors in m critical. Further support

and definition of conditions displaying a link between Vcmax and m, may enable

estimations of m values that are more appropriate than a constant that takes nothing into

account beside photosynthetic functional. Furthermore, accurate m values are time

consuming to produce and are considerably less common than Vcmax measurements.

Under conditions demonstrated to exhibit this relationship, the negative correlation

between m and Vcmax could allow estimation of m for many species using existing data.

Without considering an increase in m with [O3], predictions for a given Ah/[CO2]

will result in lower predictions of gs with elevated [O3] and 400 ppm [CO2] conditions. gs

is tightly linked to transpiration (Bernacchi et al. 2007). Water will be transported

through the plant faster than anticipated if stomata are more open than expected, leading

to overestimations of WUE and underestimations of soil drying with subsequent

implications for onset of drought. A decrease in transpiration from CO2-driven stomatal

36

closure was found to be an important contributor to increased river runoff through the

twentieth century (Gedney et al. 2006, Betts et al. 2007). Assuming the inverse is true,

depleted soil water resulting from higher than expected gs will lead to less runoff than

expected. ET, of which transpiration is a fundamental component, also has important

effects on climate. More ET than expected means more clouds reducing insolation

reaching the earth’s surface and subsequent surface temperature rise, as well as more

precipitation than expected (Sellers et al. 1997). Since predictions of gs obtained with the

unparameterized model will be lower than those generated from the corrected model,

lower predicted transpiration rates could subsequently produce underestimations of future

precipitation and temperature increaseses. In general, it has been shown that the effects of

stomatal closure on vegetation can double the radiative forcing from direct effects of

[CO2] and [O3] from being greenhouse gas, and thus contribute to all processes associated

with global climate change (Sellers et al. 1997, Sitch et al. 2007).

While the use of controlled environment chambers for this study was necessary to

examine the effect of such a large number of treatment combinations, it is important to

keep in mind the limitations of chamber studies, especially for the purposes of

extrapolating results. Long et al. (2005) reviewed many concerns that must be

acknowledged in artificial growing conditions including feedback from restricted rooting

volume that interfere with normal root growth as well as a decoupling of atmospheric gas

conditions and leaf level within the canopy. Lower absolute values of m, and higher

values of Vcmax and Jmax observed in this study relative to field-grown soybeans indicates

that these factors likely influenced the results seen here (Leakey et al. 2006b; Chapter 2).

For instance, soybeans in this study were grown on an artificial soil mixture and were

37

treated with frequent nutrient and nitrogen applications. This creates the potential for

higher Vcmax and Jmax values, and therefore would lead to lower m values than observed in

field conditions. This discrepancy may manifest as a higher threshold [CO2] required to

fully ameliorate [O3]-driven rises in m, but, future research is necessary before specific

claims such as the specific threshold [CO2] for which [O3] stops influencing changes in m

can be made with confidence.

This study examined the effects of a variety of [CO2] and [O3] combinations on

the m parameter of the Ball et al. (1987) model in soybean grown in environmental

growth chambers, as interactions of these gases represents a key uncertainty in modeling

future food supply and ecosystem functioning. m increased with [O3] unless there was

concurrent elevated [CO2], in which case there was no change. Depressions in Vcmax that

are negatively correlated with m may contribute to the effect of [O3] at ambient [CO2],

with increased gs dampening this effect via restriction of O3 uptake at high [CO2]. Other

stomatal or photosynthetic processes influencing m at higher [CO2] could also be

contributing to the overall response. These results suggest parameterization will be

necessary for large-scale models of climate and carbon and water cycling to avoid

substantial errors in predictions for future conditions of 400 ppm [CO2], but potentially

slightly higher, in combination with elevated [O3]. Additional research is necessary to

draw specific conclusions about the pattern and thresholds of this response.

38

Figures Table 3.1. Average [O3] and [CO2] values ± standard error for each of 16 treatments. Row and column headings report target concentrations for O3 and CO2, respectively. Values in bold text are the average measured [O3] over the experiment ± standard error and the values in italics are the average measured [CO2] over the experiment ± standard error.

39

[CO

2] (µ

mol

CO

2 mol

-1)

0

200

400

600

800

1000

1200

1400

1600

PAR

(µm

ol m

-2 s

-1)

VpdL

(kPa

)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

Time (min)

0 100 200 300 400 500

A (µ

mol

CO

2 m-2

s-1

)

0

10

20

30

40

g s (m

ol H

2O m

-2 s-1

)

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

Figure 3.1. Representative course of (a) incident photosynthetic photon flux density (PPFD), [CO2] and vapor pressure deficit between the leaf and atmosphere (VpdL) during gas exchange measurements of (b) steady-state net photosynthetic CO2 assimilation (A) and stomatal conductance (gs) during data collection from mature sun leaves of chamber-grown soybean for model parameterization.

a)

b)

40

Ah/[CO2]0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07

g s (m

ol H

2O m

-2 s

-1)

0.0

0.1

0.2

0.3

0.4

0.5

0 ppb O3

150 ppb O3

Figure 3.2. Representative relationship of stomatal conductance (gs) with the product of net photosynthetic CO2 assimilation (A) and fractional relative humidity (h) divided by the [CO2] of mature sun leaves grown at different [O3]. Points represent an individual gas exchange measurement of a single leaf. Only data for which [CO2] > 150 µmol mol-1 were used. Representative data from two plants grown at high (150 ppb; gs= -0.05201+ 10.05697*Ah/[CO2]; p<0.0001) and low [O3] (0 ppb; gs= -0.04225 + 5.71463*Ah/[CO2]; p<0.0001) [O3] are depicted along with the regression fits of the Ball et al. (1987) model to the measured data from each leaf.

41

Figure 3.3. Estimates of the slope parameter (m) from the Ball et al. (1987) model from mature, upper leaves of soybean grown under one of sixteen combinations of [CO2] and [O3] (5, 47, 97, 145 ppb [O3] along with 453, 584, 764, 1175 ppm [CO2]). Values are means (±SE), n = 6. The statistical significance of main treatment effects and the interaction effect from ANOVA are shown on the figure.

42

Figure 3.4. Estimates of the maximum apparent carboxylation rate (Vcmax) based on analysis of A/ci curves from mature, upper leaves of soybean grown under one of sixteen combinations of [CO2] and [O3] (5, 47, 97, 145 ppb [O3] along with 453, 584, 764, 1175 ppm [CO2]). Values are means (±SE), n = 6. The statistical significance of main treatment effects and the interaction effect from ANOVA are shown on the figure.

43

Figure 3.5. Estimates of the maximum apparent electron transport rate (Jmax) based on analysis of A/ci curves from mature, upper leaves of soybean grown under one of sixteen combinations of [CO2] and [O3] (5, 47, 97, 145 ppb [O3] along with 453, 584, 764, 1175 ppm [CO2]). Values are means (±SE), n = 6. The statistical significance of main treatment effects and the interaction effect from ANOVA are shown on the figure.

44

gmodeled (mol H2O m-2 s-1)

0.0 0.1 0.2 0.3 0.4

g mea

sure

d (m

ol H

2O m

-2 s

-1)

0.0

0.1

0.2

0.3

0.4

Figure 3.6. Relationship between stomatal conductance (gs) of mature, upper leaf of soybeans grown in environmental growth chambers under various growth [O3] and [CO2] combinations measured during separate, light-saturated gas exchange (gmeasured) and gs predicted by the model parameterized from laboratory gas exchange measurements (gmodeled). Light-saturated gas exchange measurements were taken at the beginning of each set of parameterization measurements. Red line shows 1 to 1 line.

gmeasured = 0.0163*gmodeled + 0.7671 p<0.0001 r2=0.71

45

Vc,max (µmol m-2 s-1)0 50 100 150 200 250 300

m

0

2

4

6

8

10

12

14

Figure 3.7. Scatterplot of slope parameter from the Ball et al. (1987) model (m) and the maximum carboxylation capacity of Rubisco, estimated from A/ci curves using the Farquhar et al. (1980) (Vcmax). Each point represents data from a single leaf. The line of best fit and statistical results from regression analysis are shown.

m = 10.8 - 0.025*Vcmax r2 = 0.35 p < 0.0001

46

Jmax (µmol m-2 s-1)0 100 200 300 400

m

0

2

4

6

8

10

12

14

Figure 3.8. Scatterplot of slope parameter from the Ball et al. (1987) model (m) and the maximum rate of electron transport, estimated from A/ci curves using the Farquhar et al. (1980) (Jmax). Each point represents data from a single leaf. The line of best fit and statistical results from regression analysis are shown for treatments including 400 ppm [CO2] (black circles). No significant relationship exists for all other treatment combinations (grey circles; p=0.4223).

m = 11.6 - 0.03*Jmax r2 = 0.63 p < 0.0001

600, 800, 1200 ppm CO2 400 ppm CO2

47

CHAPTER 4: CONCLUSION

Large scale models of global and ecosystem carbon and water cycling are critical

for making predictions of future ecosystem services such as food and water supply that

support human well-being. The ability of these models to simulate ecosystem function is

highly dependent on sound representations of leaf-level processes. The research

conducted in these experiments explored the relationship between gs and A, h and [CO2]

in the context of rising atmospheric [O3], and in conditions of simultaneous rises in [O3]

and [CO2]. It was discovered that the sensitivity of gs to A, h and [CO2] increased with

elevated [O3], but only in conditions of 400 ppm [CO2]. As a result, the Ball et al. (1987)

model and its derivatives will require parameterization for growth [O3] in conditions of

400 ppm [CO2].

Without correcting for increasing sensitivity with [O3] for the conditions

described, predictions of the Ball et al. (1987) model will yield lower gs than if

parameterized for growth [O3]. Due to the close relationship between gs and transpiration

(Bernacchi et al. 2007), higher estimates of gs than expected may result in

underestimation of transpiration and all of the associated consequences. For example,

more ET than expected will increase precipitation and decrease runoff estimates (Zhang

et al. 2001). Per capita availability of freshwater ecosystem services are already

anticipated to decline with increases in river runoff that are not projected to keep pace

with rising global populations (Jackson et al. 2001). Even less runoff and WUE predicted

by the parameterized relative to the unparameterized model form greater than expected gs

will only increase the shortages expected of this critical resource (Jackson et al. 2001).

Actual ET also accounts for about half the variability of litter decomposition rates,

48

representing the potential for effects on the hydrological cycle to extend into the carbon

cycle (Meentemeyer 1978).

More research is necessary to elucidate the threshold [CO2] at which [O3] effects

on m are completely ameliorated. This is because while field and chamber experiments

were qualitatively consistent, the magnitude of environmental influences on m varied as a

result of differences in growth conditions and plant physiological status. This is a

recognized issue and has been the driver for FACE experimentation (Long et al. 2005).

Furthermore, other elements of environmental change that were not studied here such as

drought, may influence m as well especially considering its known influences on gs and A

(Saliendra et al. 1995; Tezara et al. 1999; Rizhsky et al. 2002). The finding that elevated

[CO2] and elevated [O3] interact to alter m suggests testing the influence of other

environmental perturbation is warranted.

While the Ball et al. (1987) model was successfully parameterized for growth in

elevated [O3] in soybean, questions remain concerning the pattern of response for plants

of other species and functional groups. The persistence of a negative correlation between

m and Vcmax values across experimental conditions of elevated [CO2] and [O3] implicates

Vcmax as a potential driver of this change. More research is necessary to test the resilience

of this correlation across species and functional groups, to determine if a more broadly

operational relationship between Vcmax and m is supported. Estimates of m based on its

relationship to existing Vcmax values, even if only applicable to conditions of elevated

[O3], would likely improve model predictions relative to model runs using a constant m

value taking nothing into account besides the general functional group of the species in

question. This improvement could come at a fraction of the effort required for explicit

49

parameterizations as Vcmax values already exist for many species and conditions. Ignoring

the need for corrections in the m parameter of the Ball et al. (1987) model of gs used in

ecosystem or global models of carbon and water cycling could potentially scale to

substantial errors in predictions of the future. Accurate predictions of such models will be

important for informed development of adaptation and mitigation strategies to protect

ecosystem goods and services in the face of environmental change.

50

REFERENCES Ainsworth E.A., Davey P.A., Bernacchi C.J., et al. (2002) A meta-analysis of elevated [CO2] effects on soybean (Glycine max) physiology, growth and yield. Global Change Biology. 8, 695-709 Ainsworth E.A. & Long S.P. (2005) What have we learned from 15 years of free-air CO2 enrichment (FACE)? A meta-analytic review of the responses of photosynthesis, canopy properties and plant production to rising CO2. New Phytologist 165, 351–371. Ainsworth E.A., Alistair R., Leakey A.D.B., et al. (2007) Does elevated atmospheric [CO2] alter diurnal C uptake and the balance of C and N metabolites in growing and fully expanded soybean leaves? Journal of Experimental Botany 58, 579-591. Arnell, N. (2004). Global Water Cycle Encyclopedia of Life Sciences (pp. 1-10): John Wiley & Sons, Ltd. Ashmore M.R. (2002) Effects of oxidants at the whole plant and community level. In JNB Bell, M Treshow, eds, Air Pollution and Plants, Ed 2. J. Wiley, London. Ashmore M.R. (2005) Assessing the future global impacts of ozone on vegetation. Plant, Cell & Environment 28, 949–964. Ball J.T., Woodrow I.E. & Berry J.A. (1987) A model predicting stomatal conductance and its contribution to the control of photosynthesis under different environmental conditions. In Progress in Photosynthesis Research (ed. J.Biggens), pp. 221–224. Martinus-Nijhoff Publishers, Dordrecht, the Netherlands. Bernacchi C.J., Morgan P.B., Ort D.R., et al. (2005) The growth of soybean under free air [CO2] enrichment (FACE) stimulates photosynthesis while decreasing in vivo Rubisco capacity. Planta 220, 434-446. Bernacchi C.J., Leakey A.D.B., Heady L.E., et al. (2006) Hourly and seasonal variation in photosynthesis and stomatal conductance of soybean grown at future CO2 and ozone concentrations for 3 years under fully open-air field conditions. Plant, Cell & Environment 29, 2077–2090. Bernacchi C.J., Kimball B.A., Quarles D.R., et al. (2007) Decreases in stomatal conductance of soybean under open-air elevation of [CO2] are closely coupled with decreases in ecosystem evapotranspiration. Plant Physiology 143, 134-144. Bernacchi C.J., Leakey A.D.B., Kimball B.A., et al. (2011) Growth of soybean at future tropospheric ozone concentrations decreases canopy evapotranspiration and soil water depletion. Environmental Pollution 159, 1464-1472.

51

Betts R.A., Boucher O., Collins M., et al. (2007) Projected increase in continental runoff due to plant responses to increasing carbon dioxide. Nature 448, 1037-U5. Bounoua L., Collatz G.J., Sellers P.J., et al. (1999) Interactions between vegetation and climate: radiative and physiological effects of doubled atmospheric CO2. Journal of Climate 12, 309–324. von Caemmerer S. & Farquhar G.D (1981) Some relationships between the biochemistry of photosynthesis and the gas exchange of leaves. Planta 153, 376–387. Canadell J.G., Raupach M.R. & Houghton R.A. (2009) Anthropogenic CO2 emissions in Africa. Biogeosciences 6, 463-468. Chen C.P., Frank T.D. & Long S.P. (2009) Is a short, sharp shock equivalent to long-term punishment? Contrasting the spatial pattern of acute and chronic ozone damage to soybean leaves via chlorophyll fluorescence imaging. Plant, Cell & Environment 32, 327-335. Cramer W., Bondeau A., Woodward F.I., et al. (2001) Global response of terrestrial ecosystem structure and function to CO2 and climate change: results from six dynamic global vegetation models. Global Change Biology 7, 357-373. Damour G., Simonneau T., Cochard H., et al. (2010) An overview of models of stomatal conductance at the leaf level. Plant, Cell & Environment 33, 1419-1438. Drake B.G., Gonzalez-Meler M.A. & Long S.P. (1997) More efficient plants: a consequence of rising atmospheric CO2? Annual Review of Plant Physiology and Plant Molecular Biology 48, 609–639. Eichelmann H., Oja V., Rasulov B., et al. (2004) Photosynthetic parameters of birch (Betula pendula Roth) leaves growing in normal and in CO2- and O-3-enriched atmospheres. Plant, Cell & Environment 27, 479-495. Farage P.K., Long S.P., Lechner E. & Baker N.R. (1991) The sequence of change within the photosynthetic apparatus of wheat following short-term exposure to ozone. Plant Physiology 95, 529-535. Farage P.K. & Long S.P. (1995) An in-vivo analysis of photosynthesis during short-term O-3 exposure in 3 contrasting species. Photosynthesis Research 43, 11-18. Farquhar G.D., von Caemmerer S. & Berry J.A. (1980) A biochemical model of photosynthetic CO2 assimilation in leaves of C3 species. Planta 149, 78-90. Field C.B., Jackson R.B. & Mooney H.A. (1995) Stomatal responses to increased CO2 – implications from the plant to the global-scale. Plant, Cell & Environment 18, 1214–1225.

52

Foley J.A., Prentice I.C., Ramankutty N., Levis S., Pollard D., Sitch S. & Haxeltine A. (1996) An integrated biosphere model of land surface processes, terrestrial carbon balance, and vegetation dynamics. Global Biogeochemical Cycles 10, 603–628. Gedney N., Cox PM., Betts R.A., et al. (2006) Detection of a direct carbon dioxide effect in continental river runoff records. Nature 439, 835-838. Gonzalez-Fernandez I., Bass D., Muntifering R., et al (2008) Impacts of ozone pollution on productivity and forage quality of grass/clover swards. Atmospheric Environment 42, 8755-8769. Haberl H., Erb K.H., Krausmann F., et al. (2007) Quantifying and mapping the human appropriation of net primary production in earth's terrestrial ecosystems. Proceedings of the national academy of sciences of the United States of America 104, 12942-12945. Houghton R.A., Lawrence K.T., Hackler J.L., et al. (2001) The spatial distribution of forest biomass in the Brazilian Amazon: a comparison of estimates. Global Change Biology 7, 731-746. IPCC (2001) Climate Change 2001: The Scientific Basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK. IPCC (2007) IPCC Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change In: S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller, Editors, Climate Change 2007: The Physical Science Basis, Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA (2007) Jackson R.B., Carpenter S.R., Dahm C.N., et al. (2001) Water in a changing world. Ecological Applications 11, 1027-1045. Karnosky D.F., Zak D.R., Pregitzer K.S., et al. (2003) Tropospheric O-3 moderates responses of temperate hardwood forests to elevated CO2: a synthesis of molecular to ecosystem results from the Aspen FACE project. Functional Ecology 17, 289-304. Karnosky D.F., Skelly J.M., Percy K.E., et al. (2007) Perspectives regarding 50 years of research on effects of tropospheric ozone air pollution on US forests. Environmental Pollution 147, 489-506. King J.S., Kubiske M.E., Pregitzer K.S. et al. (2005) Tropospheric O3 compromises net primary production in young stands of trembling aspen, paper birch and sugar maple in response to elevated atmospheric CO2. New Phytologist, 168, 623–635.

53

Krupa S.V., Nosal M. & Legge A.H. (1994) Ambient ozone and crop loss – establishing a casue-effect relationship. Environmental Pollution 83, 269-276. Le Quere C., Raupach M.R., Canadell J.G., et al. (2009) Trends in the sources and sinks of carbon dioxide. Nature Geoscience 2, 831-836. Leakey A.D.B., Bernacchi C.J., Dohleman F.G., Ort D.R. & Long S.P. (2004) Will photosynthesis of maize (Zea mays) in the US Corn Belt increase in future CO2 rich atmospheres? An analysis of diurnal courses of CO2 uptake under free-air concentration enrichment (FACE). Global Change Biology 10, 951–962. Leakey A.D.B., Uribelarrea M., Ainsworth E.A., Naidu S.L., Rogers A., Ort D.R. & Long S.P. (2006a) Photosynthesis, productivity and yield of maize are not affected by open-air elevation of CO2 concentration in the absence of drought. Plant Physiology 140, 779–790. Leakey A.D.B., Bernacchi C.J., Ort D.R., et al. (2006b) Long-term growth of soybean at elevated [CO2] does not cause acclimation of stomatal conductance under fully open-air conditions. Plant, Cell & Environment 29, 1794-1800. Long S.P., Ainsworth E.A., Leakey A.D.B. & Morgan P.B. (2005) Global food insecurity. Treatment of major food crops with elevated carbon dioxide or ozone under large-scale fully open-air conditions suggests recent models may have overestimated future yields. Philos The Royal Society B 360, 2011–2020. Matyssek R., Gunthardtgoerg M.S., Keller T., et al. (1991) Impairment of gas-exchange and structure in birch leaves (Betula-pendula) caused by low ozone concentrations. Trees-Structure and Function 5, 5-13. Maurer S., Matyssek R., GunthardtGoerg M.S., et al. (1997) Nutrition and the ozone sensitivity of birch (Betula pendula) .1. Responses at the leaf level. Trees-Structure and Function 12, 1-10. Meehl G.A., Stocker T.F., Collins W.D. et al. (2007) Global climate projections. In: Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Inter-governmental Panel on Climate Change (eds Solomon S., Qin D., Manning M., et al.), pp. 747–845. Cambridge University Press, Cambridge, UK/New York, NY, USA. Meentemeyer V. (1978) Macroclimate and lignin control of litter decomposition rates. Ecology 59, 465-472. Miglietta F., Peressotti A., Vaccari F.P., Zaldei A., DeAngelis P. & Scarascia-Mugnozza G. (2001) Free-air CO2 enrichment (FACE) of a poplar plantation: the POPFACE fumigation system. New Phytologist 150, 465–476.

54

Moorcroft P.R., Hurtt G.C. & Pacala S.W. (2001) A method for scaling vegetation dynamics: The ecosystem demography model (ED). Ecological Monographs 71, 557-585. Morgan P.B., Ainsworth E.A. & Long S.P. (2003) Elevated O3 impact on soybeans, a meta-analysis of photosynthetic, biomass, and yield responses. Plant, Cell & Environment 26, 1317–1328. Morgan P.B., Bernacchi C.J., Ort D.R. & Long S.P. (2004) An in vivo analysis of the effect of season-long open-air elevation of ozone to anticipated 2050 levels on photosynthesis in soybean. Plant Physiology 135, 2348–2357. Morgan P.B., Mies T.A., Bollero G.A., et al. (2006) Season-long elevation of ozone concentration to projected 2050 levels under fully open-air conditions substantially decreases the growth and production of soybean. New Phytologist 170, 333-343. Mott K.A. & Peak D. (2010). Stomatal responses to humidity and temperature in darkness. Plant, Cell & Environment 33, 1084-1090. Norby R.J, Wullschleger S.D., Gunderson C.A., et al (1999) Tree responses to rising CO2 in field experiments: implications for the future forest. Plant, Cell & Environment 22, 683-714. Ollinger S.V., Aber J.D. & Reich P.B. (1997) Simulating ozone effects on forest productivity: interactions among leaf-, canopy-, and stand-level processes. Ecological Applications 7, 1237–1251. Paoletti E., Bytnerowicz A., Andersen C. et al. (2007) Impacts of air pollution and climate change on forest ecosystems – Emerging research needs. The Scientific World Journal 7, 1–8. Paoletti E. & Grulke N.E. (2005) Does living in elevated CO2 ameliorate tree response to ozone? A review on stomatal responses. Environmental Pollution 137, 483-493. Paoletti E. & Grulke N.E. (2010) Ozone exposure and stomatal sluggishness in different plant physiognomic classes. Environmental Pollution 158, 2664-2671. Pearson M. & Mansfield T.A. (1993) Interacting effects of ozone and water-stress on the stomatal –resistance of beech (Fagus-sylavatical). New Phytologist 123, 351-358. Prather M., Gauss M., Berntsen T., et al. (2003) Fresh air in the 21st century? Geophysical Research Letters 30, 1100. Reich P.B. (1987) Quantifying plant response to ozone: a unifying theory. Tree Physiology 3, 63-91.

55

Reich P.B. & Amundson R.G. (1985) Ambient levels of ozone reduce net photosynthesis in tree and crop species. Science 230, 566-570. Reich P.B. & Lassoie J.P. (1984) Effects of low-level O-3 exposure on leaf diffusive conductance and water-use-efficiency in hybrid poplar. Plant, Cell & Environment 7, 661-668. Reid S.J., Rex M., Von der Gathen P., et al. (1998) A study of ozone laminae using diabatic trajectories, contour advection and photochemical trajectory model simulations. Journal of Atmospheric Chemistry 30, 187-207. Rizhsky L., Liang H.J. & Mittler R. (2002) The combined effect of drought stress and heat shock on gene expression in tobacco. Plant Physiology 130, 1143-1151. Rogers A., Allen D.J., Davey P.A., et al. (2004) Leaf photosynthesis and carbohydrate dynamics of soybeans grown throughout their life-cycle under free-air carbon dioxide Enrichment. Plant, Cell & Environment 27, 449–458. The Royal Society. (2008) Ground-level ozone in the 21st century: future trends, impacts and policy implications. Science Policy Report 15/08. London. Saliendra N.Z., Sperry J.S., Comstock J.P. (1995) Influence of leaf water status on stomatal response to humidity, hydraulic conductance, and soil drought in Betula-occidentalis. Planta 196, 357-366. Schwalm C.R., Williams C.A., Schaefer K., et al. (2010) Assimilation exceeds respiration sensitivity to drought: A FLUXNET synthesis. Global Change Biology 16, 657-670. Sellers P.J., Bounoua L., Collatz G.J., et al. (1996) Comparison of radiative and physiological effects of doubled atmospheric CO2 on climate. Science 271, 1402-1406. Sellers P.J., Dickinson R.E., Randall D.A., et al. (1997) Modeling the exchanges of energy, water, and carbon between continents and the atmosphere. Science 275, 502-509. Sitch S., Cox P.M., Collins W.J. & Huntingford C. (2007) Indirect radiative forcing of climate change through ozone effects on the land-carbon sink. Nature 448, 791–794. Tezara W, Mitchell V.J., Driscoll S.D., et al. (1999) Water stress inhibits plant photosynthesis by decreasing coupling factor and ATP. Nature 401, 914-917. Tjoelker M.G., Volin J.C., Oleksyn J., et al. (1995) Interaction of ozone pollution and light effects on photosynthesis in a forest canopy experiment. Plant, Cell & Environment 18, 895-905.

56

USDA (2010) National Agricultural Statistic Service, United States Department of Agriculture, Washington, D.C., USA USDA; http://www.nass.usda.gov Vahisalu T., Puzorjova I., Brosche M., et al. (2010) Ozone-triggered rapid stomatal response involves the production of reactive oxygen species, and is controlled by SLAC1 and OST1. Plant Journal 62, 442-453. Van Dingenen R., Dentener F.J., Raes F., et al. (2009) The global impact of ozone on agricultural crop yields under current and future air quality legislation. Atmospheric Environment 43, 604-618. Volin J.C., Reich P.B. & Givnish T.J. (1998) Elevated carbon dioxide ameliorates the effects of ozone on photosynthesis and growth: species respond similarly regardless of photosynthetic pathway or plant functional group. New Phytologist 138, 315-325. Volz A. & Kley D. (1988) Evaluation of the montsouris series of ozone measurements made in the 19th-century. Nature 332, 240–242. Wigley T.M.L. & Jones P.D. (1985) Influences of precipitation changes and direct CO2 effects on streamflow. Nature 314, 149-152. Wilkinson S. & Davies W.J. (2009) Ozone suppresses soil drying- and abscisic acid (ABA)-induced stomatal closure via an ethylene-dependent mechanism. Plant, Cell & Environment 32, 949-959. Will R.E. & Teskey R.O. (1997) Effect of elevated carbon dioxide concentration and root restriction on net photosynthesis, water relations and foliar carbohydrate status of loblolly pine seedlings. Tree Physiology 17, 655-661. Wittig V.E., Ainsworth E.A. & Long S.P. (2007) To what extent do current and projected increases in surface ozone affect photosynthesis and stomatal conductance of trees? A meta-analytic review of the last 3 decades of experiments. Plant, Cell & Environment 30, 1150-1162. Woodward F.I. & Kelly C.K. (1995) The influence of CO2 concentration on stomatal density. New Phytologist 131, 311-327. Wullschleger S.D. (1993) Biochemical limitations to carbon assimilation in C3 plants – a retrospective analysis of the A/ci curves from 109 species. Journal of Experimental Botany 44, 907–920. Zhang H., Henderson-Sellers A. & McGuffie K. (2001) The compounding effects of tropical deforestation and greenhouse warming on climate. Climatic Change 49, 309-338.