Chapter 1

1.1 THE EFFECTS OF GLOBAL ATMOSPHERIC CHANGE ON VEGETATION

Since the industrial revolution the chemical composition of the Earth's atmosphere has been altered by emissions of anthropogenic gases, which has lead to dramatic changes in the atmospheric concentrations of several greenhouse gases within the last few decades (Watson et al., 1990; Schimel et al., 1994). Changing atmospheric concentrations of greenhouse gases cause both direct effects on vegetation and indirect effects via changes to the climate, particularly surface temperatures (Melillo et al., 1990). Plant responses to these changes, for example, a modification in net carbon exchange and water balance of vegetation, have, in turn, the potential to cause feedback effects on the atmosphere and climate. Therefore, the ability to predict the effect of global warming on crop production will depend on the ability to predict the interactive effects of global atmospheric change, climate, net carbon exchange, transpiration and the production of vegetation, in addition to the feedback effects associated with the responses.

The direct effects of changing atmospheric concentrations of some greenhouse gases on vegetation have been well researched, particularly carbon dioxide (CO2) and ozone (O3) (Campbell et al., 1988; Drake and Leadley, 1991; Lawlor and Mitchell, 1991; Long, 1991; Harley et al., 1992; Heath, 1994). Carbon dioxide concentrations have been increasing at an annual rate of 0.5%, and one direct effect of elevated [CO2] is an enhancement of vegetative growth rates in C3 plants, due to a preferentially increased supply of the substrate for photosynthesis (CO2) relative to the supply of substrate for the CO2-releasing process of photorespiration (O2) (Melillo et al., 1990; Watson et al., 1990; Lawlor and Mitchell, 1991; Long, 1991). Therefore, taking just this one effect in isolation from others, the increased growth rates from elevated [CO2] might be expected to cause an increase in the rates at which CO2 is sequestered by the biosphere, thereby causing a negative feedback effect on global warming.

Concentrations of tropospheric ozone have also been rising, and as ozone is harmful to vegetation, one direct effect of increasing ozone is expected to be a reduction in vegetative growth (Krupa and Manning, 1988; PORG, 1993).

1.2 TROPOSPHERIC OZONE

Of all the tropospheric pollutants, ozone causes the greatest damage to plants in industrial regions (Krupa and Manning, 1988). The phytotoxic effects of ozone are of particular concern as background tropospheric concentrations are already close to potentially damaging levels, and concentrations in the northern hemisphere have been rising at an annual rate of 1 - 2% for the last three decades (Watson et al., 1990; Manning and Krupa, 1992). Recent estimates suggest ozone exposure is already causing a loss of $3 billion per annum to crop production in the USA (PORG, 1993).

The toxicity of ozone exposure to vegetation is evident by the visible symptoms of leaf mottling caused by chlorosis (Heath, 1987; 1994) and invisible damage to the photosynthetic capacity (Farage et al., 1991). Ozone enters the leaf via the stomata, where it is rapidly broken down to form damaging oxidative radicals within the leaf, which induce the production of protective scavenging mechanisms in the plant (Heath, 1980; Tingey and Taylor, 1982; Guderian et al., 1985; Scandalios, 1993). Ozone affects cell membrane permeability, so disrupting metabolism (Heath, 1987). Exposure to ozone causes reduced rates of photosynthesis, via a reduced in vivo maximum capacity for carboxylation (Vcmax), as well as a decline in stomatal conductance (gs) (Hill and Littlefield, 1968; Reich et al., 1985; Olszyk and Tingey, 1986; Amundson et al., 1987; Miller, 1987; Farage et al., 1991; Farage and Long, 1995). A reduction in quantum use efficiency may eventually result (Nie et al., 1993). The decrease in rates of photosynthesis and changes in photosynthate allocation from harmful levels of ozone exposure suppresses growth and yield (Miller, 1987). The biochemical effects of ozone within the leaf have been recently reviewed by Heath (1994) and a summary of effects at the molecular level can be found in Pell et al. (1994).

Considering the direct effect of increasing concentrations of phytotoxic ozone in isolation from other effects, the reduction in vegetative growth might be expected to cause a positive feedback effect, via a reduction in the biomass available for sequestering CO2. The control of harmful ozone concentrations to reduce the loss of crop production requires a knowledge of the processes which generate ozone in the troposphere.

Ozone is a secondary pollutant produced by the photochemical oxidation of carbon monoxide and hydrocarbons in the presence of nitrogen oxides (NOX) (Chameides and Lodge, 1992). The chemicals required to produce O3 occur naturally but are emitted at enhanced rates by vehicle exhaust fumes (Heath, 1994). The generation of ozone also requires the presence of the hydroxyl radical, which is produced by the photolysis of O3, via Equations 1.1 and 1.2 (Table 1.1). In relatively pristine environments, where there are few non-methane hydrocarbons, ozone formation is initiated by the oxidation of carbon monoxide, Equations 1.3 to 1.8 (Table 1.1) (Chameides and Lodge, 1992). (M denotes a molecule in the atmosphere, such as N2 or O2, that takes part in a reaction and acts to stabilise the molecule produced in the reaction.) However, in more polluted areas, or as polluted urban air is transported over rural areas, ozone can be produced by the oxidation of non-methane hydrocarbons, such as isoprene, and methane oxidation by hydroxyl radicals, as listed in Tables 1.2 and 1.3, respectively. Several of the reactions are shared by all three oxidation processes (Equations 1.5, 1.6 and 1.8) and rates of ozone generation are complex, and are non-linearly dependent upon the concentrations of the various nitrogen oxides and hydrocarbons present (Chameides and Lodge, 1992).

Although rates of ozone production are generally higher in regions of polluted air, the slow transportation of air masses during warm sunny periods of anticyclonic weather can cause a build up of ozone to relatively high concentrations in rural areas, giving rise to phytotoxic responses and a considerable loss of crop production (PORG, 1993; Chameides et al., 1994). The frequency of these high ozone episodes is increasing, and any attempt to control tropospheric ozone concentrations is further complicated by the significant role that natural hydrocarbons, emitted by trees, play in the generation of high

Table 1.1: Table of the photochemical reactions that produce the hydroxyl radical (equations 1.1 and 1.2) and ozone by the oxidation of carbon monoxide (Equations 1.3 to 1.8) (Chameides and Lodge, 1992).

Table 1.2: Table of the photochemical reactions that produce ozone by the oxidation of non-methane hydrocarbons, such as isoprene (Chameides and Lodge, 1992). (RH, R and R' denote a hydrocarbon, a hydrocarbon chain, and a chain with one fewer carbon atoms, respectively.)

Table 1.3: Table of the photochemical reactions that produce ozone by the oxidation of methane (Chameides and Lodge, 1992).

ozone concentrations in both urban and rural areas (Trainer et al., 1987; Chameides et al., 1988). An effective precursor of ozone, and the most prolific and reactive of the non-methane hydrocarbons emitted by trees, is isoprene (2-methyl 1, 3-butadiene) (Trainer et al., 1987; Chameides et al., 1988).

1.3 INCREASED RATES OF ISOPRENE EMISSION

Isoprene is emitted mainly by deciduous tree species and emission rates are dependent on light and are highly temperature sensitive (Monson and Fall, 1989; Loreto and Sharkey, 1990; Monson et al., 1991a; Sanadze, 1991). Increasing temperatures from putative global warming would therefore be expected to cause an increase in the flux of isoprene to the atmosphere, thereby increasing levels of tropospheric ozone (Monson et al., 1991a). Why plants emit isoprene is uncertain, although recent research suggests isoprene may play a role in protecting photosynthetic mechanisms during stress, for example, by providing thermal tolerance under conditions of water stress, where stomatal closure causes an increase in leaf temperature (Sharkey and Singsaas, 1995; Fang et al., 1996; Litvak et al., 1996; Sharkey, 1996).

Although isoprene is only emitted in trace amounts by individual leaves, the cumulative emission from a forested area under warm, summer conditions is considerable, with estimates of annual global isoprene production comparable to estimates of the natural global emission rates of methane, approximately 400 TgC yr-1 (Zimmerman et al., 1978; Rasmussen and Khalil, 1988; Monson et al., 1991a). Isoprene interferes with the chemical re-cycling of ozone and nitrogen oxides within the troposphere, indirectly causing increased production of PAN and organic peroxides, in addition to ozone (Lloyd et al., 1983, Trainer et al., 1987, Chameides et al., 1988).

Isoprene molecules react with the OH. radical, the principal oxidising agent of air in daylight, and therefore not only affect air quality directly, but also compete preferentially with methane for the principal sink of methane, OH. (Crutzen and Fishman, 1977; Rasmussen and Khalil, 1988; Zimmerman et al., 1988). Thus global warming might be expected to increase the residence time and effective concentration of methane in the atmosphere (Greenberg et al., 1985; Zimmerman et al., 1988; Monson et al., 1991a). As methane has a greater global warming potential than carbon dioxide an increase in methane concentration has the potential to cause a significant additional positive feedback effect on climate (Shine et al., 1990; Monson et al., 1991a). For clarity, the interactions between climate change, vegetation and increasing concentrations of carbon dioxide, ozone and isoprene are summarised in Figure 1.1. The other indirect effects of increasing isoprene emission rates are the increased ozone concentrations which have the potential to interfere with the sequestering of carbon dioxide in vegetation biomass (Figure 1.1).

Despite an increased awareness of the key role that isoprene plays in atmospheric chemistry, and a recent increase in the volume of research into isoprene emission and synthesis, to date there is a lack of a rigorous method to predict the effects of changing environmental conditions on rates of biogenic isoprene emission.  

1.4 INTERACTIVE EFFECTS

Due to the complex nature of the Earth's system and the interrelationships between the atmosphere, climate and biosphere, future global change will consist of concurrent alterations of many interrelated environmental factors and their associated feedback effects. In order to predict accurately the effects of global change on crop production and the dynamics of natural vegetation, it will be eventually necessary to predict all the interactive effects of changes in atmospheric composition and climate on vegetation. The effects of vegetation response on climate and atmosphere will also need to be taken into account, thus requiring a "holistic" approach. However, this can only be achieved by first understanding the effects of individual factors on vegetation, and then using this knowledge to progressively build up an understanding of the interactive effects of factors on canopy carbon gain.

The prediction of the effects of an individual factor on vegetation requires an in depth knowledge of the mechanisms underlying the processes within vegetative growth. The most fundamental process of plant growth, and the one that is most well understood,

Figure 1.1.

is photosynthesis. Using the interactive effects of elevated [CO2] and temperature on photosynthesis as an example, it can be shown how the prediction of interactive effects of two factors can be very different to the effects of the individual factors on vegetation. Elevated [CO2] might be expected to cause a net negative feedback on global warming, due to an increase in the rate of sequestering of CO2 in the biosphere, on account of the reduction in photorespiratory loss of carbon from C3 plants. However, an indirect effect of rising [CO2] on vegetation is an anticipated increase in surface temperatures. An increase in temperature alters the solubility of both oxygen and carbon dioxide, as well as the specificity of Rubisco for O2 and CO2 (Jordan and Ogren, 1984). Therefore, any prediction of future climate change effects needs to take account of the interactive effects of elevated CO2 and increasing temperature, as rising Ca has the potential to alter the magnitude of the response of photosynthesis to increasing surface temperatures, or even to change the direction of that response (Long, 1991).

It has similarly been recognised that the prediction of whole-plant carbon dynamics with global change needs to take account of the effects of interacting factors other than CO2 and temperature, in particular the increasing atmospheric concentrations of phytotoxic ozone in Western Europe and North America (Melillo et al., 1990; McKee et al., 1995). There are relatively few published results concerning the interactive effects of CO2 and ozone on crop species (Barnes and Pfirrmann, 1992; Polle et al., 1993; Heagle et al., 1993; Balaguer et al., 1995; McKee et al., 1995). Elevated carbon dioxide concentrations may alleviate the harmful effects of tropospheric ozone, as increased [CO2] causes a decrease in stomatal conductance, thereby reducing the uptake of ozone into the leaf (Morison, 1985; McKee et al., 1995). Other investigations have found that phytotoxic levels of ozone reduce the stimulation in growth resulting from high CO2 concentrations (Barnes and Pfirrmann, 1992; Balaguer et al., 1995). However, a lack of mechanism for leaf ozone damage has prevented a method of predicting the interactive effects of CO2 and O3 on photosynthesis.

An additional unknown is the interactive effect of increasing concentrations of ozone and isoprene on vegetation. The highly reactive isoprene molecules react with ozone, and it has been suggested that the reaction between ozone and isoprene, or other non-methane hydrocarbons, may produce harmful free radicals and hydroperoxides within the leaf, thus contributing to the plant damage seen in areas of forest die-back (Hewitt et al., 1990). However, the experimental evidence for the ozone-hydrocarbon reactions within the plant are not supported by model calculations. Further research will be required, to establish the concentrations and reaction rates between chemicals that occur under the conditions found within plant tissue, before this issue can be resolved (Salter and Hewitt, 1992).

The response of vegetation to atmospheric and climate change, in turn, affects the local environment, as vegetation has a decisive influence on both climate and soil (Larcher, 1995). For example, vegetation cover increases the stability and water holding capacity of the soil via the physical presence of the rooting system, the shading of the soil from direct sunlight by the vegetation canopy, and the addition of dead organic matter. An extreme example of the influence of vegetation on soil and climate is the extension of desert conditions induced by the removal of the natural vegetation. It has been suggested that progressive desertification may result in reduced rates of precipitation. The increase in albedo of the earth's surface, caused by removal of vegetation, increases the reflectance of incoming radiation, and the reduced absorption of radiation by the earth's surface reduces the warming of the Earth's surface, thus reducing the likelihood of convective cloud formation (Money, 1988). Local climate is also influenced by vegetation via gas exchange, for example, increasing temperatures may cause increased emissions from forests of a suite of hydrocarbons, including monoterpenes and isoprene, which can increase haze formation and so reduce incoming radiation (Went, 1960).

Although it is not possible to predict the interactive effects of all changes in environmental conditions immediately, it is possible to address some of the issues by the application of the findings of the most recent research into individual effects to the construction of process-based models. A lack of knowledge has prevented the prediction of the interactive effects of ozone damage, temperature rise and elevated CO2 on vegetation and whether or not these effects might be exacerbated by changing rates of biogenic isoprene emission. The unknowns that have prevented these predictions include a lack of knowledge of the processes underlying ozone damage and isoprene synthesis, thus preventing the quantitative assessment of both the effects of ozone uptake by vegetation, and the interactive effects of changing environmental conditions on leaf isoprene emission rates (Monson et al., 1991a; Runeckles, 1992). However, the recently acquired knowledge of the processes of ozone damage and isoprene synthesis can be used to formulate mechanistic models to predict the changes in leaf ozone damage and isoprene emission rates with varying environmental conditions (Loreto and Sharkey, 1993; Sharkey and Loreto, 1993; Kuzma and Fall, 1993; Heath, 1994; Monson et al., 1994; Silver and Fall, 1991; 1995). Predictions of the effects of atmospheric and climate change on vegetation have to rely on model simulations, due to the complexity of the atmosphere-climate-vegetation system.

Models are simple representations of real systems based upon knowledge of the system, requiring a comprehensive understanding of the processes underlying the system, and how the processes respond to changing external variables. Modelling is the only satisfactory method of predicting the effects of future long term interacting factors which are outside our experience and cannot be determined experimentally.

1.5 SCALES AND TYPES OF MODEL

Mathematical modelling reduces a system to representative equations which describe the principle processes of the system. The use of mathematics to describe a system is an essential tool in predictive methods. Mathematical equations allow relationships to be described definitively, and not only aids the analysis of a problem into its main parts, but also helps to synthesise an understanding of a problem from a knowledge of those parts (Charles-Edwards et al., 1986). After validation, the model may be used to predict responses of the system under varying conditions. The size, scale and scope of models vary enormously. Modelling large scale complex systems requires relatively elaborate models, for example, the General Circulation Models used to predict the effects of anthropogenic greenhouse gases on climate (Manabe and Stouffer, 1980; Cess and Potter, 1988). At the opposite end of the scale are small, relatively simple mathematical models which may consist of only a few equations, for example, the equations used to simulate the change in time of amounts of reactants and products in a chemical reaction involving one enzyme.

There are several types of mathematical model. The type of model applicable to a study will depend on the hypothesis being tested and the objectives that are to be achieved. The models used within this study are dynamic deterministic models. Dynamic models predict the outcome of events with the passage of time, employing numerical integration methods. Deterministic models predict definite outcomes, with no associated probability distribution (Thornley and Johnson, 1990). Two extremes of dynamic deterministic model, used in biology, can be identified: empirical and mechanistic models. Empirical models are useful where there is insufficient data or knowledge to construct a model based on the mechanisms of a process, but cannot be realistically used to predict beyond experience. Such models are associated with the process at only one hierarchical level of organisation, and often involve curve-fitting and regression analysis to fit mathematical equations to observed data, with few constraints set by any knowledge of the underlying processes of the system. Mechanistic models, on the other hand, are reductionist, based on knowledge of the mechanisms of the processes underlying the system being modelled. Mechanistic models are always more complex than empirical models and will generally fit the data well under a variety of conditions, as it has many constraints built into its structure by means of the assumptions of the model. Because it is so constrained, a mechanistic model can often be applied to a greater range of phenomena, relating them to each other. Thus, a mechanistic model not only offers more possibilities for manipulating and improving the understanding of the system being modelled, but also allows the predictions of interactive effects to be made with some confidence. In reality, few biological processes can be modelled as pure mechanistic models, most require empirically determined parameters and assumptions. Prediction may therefore be best served by including as much known mechanism as relevant, in what might be described as mechanistically-rich or process-based models. This study aims to develop such models for the prediction of the effects of climate change on rates of isoprene emission and the effects of acute and chronic ozone exposure on photosynthesis.

1.6 AIMS AND OBJECTIVES

The overall aim of the thesis is to construct new process-based models of the effects of environmental change on rates of isoprene emission, and the effects of ozone on photosynthesis, and then to use the models to predict the interactive effects of concurrently changing variables on canopy photosynthesis. Due to both the long-term nature of atmospheric change and the complexity of the Earth's system, predictions of the effects of climate change on atmosphere, vegetation and the associated feedback effects on climate must rely on the simulations of mechanistically-rich models. However, the use of mechanistically-rich models for predicting the effects of each variable acting independently of other variables will fail to provide the accurate predictions necessary for policy-making decisions. Accurate predictions will require simulation of the interactive effects of severall concurrently changing variables, thus providing a more holistic approach. Therefore, the objectives of the present research are to construct and use mechanistically-rich models of isoprene emission and ozone effects on photosynthesis to enable the prediction of the interactive effects of rising concentrations of ozone, carbon dioxide and temperature on canopy CO2 exchange and isoprene emission. In so doing, it is aimed to show how mechanistically-rich models of a few interacting variables present predictions that are very different from those predicted by a change in just one variable, and how the interactive approach to modelling is a step towards achieving the ultimate goal of a holistic approach.

For example, the two major effects of current climate change are an increase in the concentration of carbon dioxide, and, due to the thermal absorptive properties of CO2, global warming. To assess the effects of climate change on vegetation, the interactive effects of changing variables, in addition to any feedback effects between vegetation and the atmosphere/climate system, must be appraised. Although elevated [CO2] might be expected, if acting in isolation, to cause an increase in biomass, thus sequestering more CO2 and causing a negative feedback effect on global warming, concurrent increasing temperatures may reduce, or even reverse, this effect (Long, 1991). An effect of increasing temperature, if acting in isolation, would be an increase in emission rates of isoprene, an important precursor of ozone, and increasing concentrations of phytotoxic ozone, if acting alone, would be expected to cause a decrease in vegetative biomass, which would reduce the negative feedback caused by elevated [CO2]. Also, at the leaf level scale, elevated [CO2] may cause a decrease in ozone damage by reducing stomatal conductance. Therefore, as none of these factors act in isolation, it is only by analysing the combined effects of global warming and atmospheric change on vegetation, that a more complete picture of atmosphere-biosphere interactions can be attained, and greater accuracy and reliability in predicting feedback effects between atmosphere, vegetation and climate can be realised. To this end, this study aims to develop and use mechanistically-rich models to predict vegetation responses to interacting changing environmental conditions, in terms of rates of isoprene emission and CO2 assimilation rates.

Thus, a process based model of isoprene emission rates is presented and then used to predict the interactive effects of increasing temperature and elevated [CO2] on isoprene emission rates, both at the leaf level, and at the canopy level, by scaling-up the leaf-level model within WIMOVAC (Humphries and Long, 1995). Isoprene is an important pre-cursor of ozone, and future air quality policy will depend upon assessing the likely change in emission rates under changing climate conditions. At present there is no process-based model of isoprene emissions to predict future levels of emission rates, thus, this model is a significant step towards providing air quality standards for the future.

Similarly, there is, at present, no mechanistically-rich model available to predict the effects of ozone exposure on photosynthesis, which is needed, not only for air quality reasons, but also to predict the future effects of increasing tropospheric ozone concentrations on crop productivity. Thus, a mechanistically-rich model of acute ozone effects is constructed, and used to predict the interactive effects of increasing [CO2] and acute ozone exposure on wheat leaf photosynthesis. The theory behind the acute ozone model is used to construct a mechanistically-rich model of the effects of chronic ozone exposure on wheat leaf photosynthesis, and this model is used to predict the interactive effects of elevated [CO2] and chronic ozone on the productivity of a wheat canopy.

The overall aim of Chapter Two is to establish a process-based model of isoprene emission rates, based on published data, and to use the model to compare rates of emission under present day conditions to those predicted under conditions of elevated [CO2] and warmer temperatures. The objectives of are: First, to construct and present a process-based model of leaf isoprene emission rates. Second, to test the model against published data. Third, to conduct a sensitivity analysis of the model coefficients, and Fourth, to use the model to predict both the individual and interactive effects of a temperature rise of 2oC and increasing [CO2], from 350 to 650 µmol mol-1, on leaf isoprene emission rates. Although increasing temperature is expected to cause an increase in isoprene emission rates, isoprene may be emitted at lower rates under elevated [CO2]. Interactive effects between elevated [CO2] and temperatures are also expected at the leaf level, due to increased leaf temperatures caused by elevated [CO2]-induced stomatal closure. Process-based modelling is the only way to assess the percentage changes from interactive effects.

The overall aim of Chapter Three is to predict the change in isoprene emitted by different forest ecosystems under conditions of climate change. The objectives of Chapter Three are: First, to present a canopy model of isoprene flux, by scaling up the mechanistically-rich model of leaf isoprene emission to the canopy level. Second, to validate the model by comparison with measured data published in the literature for comparable canopies of the same region. Third, to predict both the individual and interacting effects of increasing temperature by 2oC and an increased [CO2] from 350 to 650 µmol mol-1 on rates of isoprene emission from three types of forest ecosystem, temperate deciduous forest in Britain, temperate deciduous forest of Pennsylvania, and a tropical forest of the Amazon.

The overall aim of Chapter Four is to construct a model of the effects of acute ozone exposure on wheat leaves, using the data of Farage et al. (1991), where wheat was exposed to very high concentrations of ozone for a short duration of time. The exposure of wheat to ozone under this regime causes a decline in both Vcmax and stomatal conductance. The objectives of this chapter are first, to examine the relationship between the uptake of ozone and into the leaf and ozone damage. Second, to determine the relationship between the reduction in Vcmax and stomatal conductance, and so enable the Third objective to be achieved, the construction of a mechanistic model of acute ozone effects on wheat. Fourth, to predict the interactions of elevated [CO2] and acute ozone exposure on wheat leaf photosynthesis.

The aim of Chapter Five is to predict the effects of chronic ozone exposure on wheat, and to use this model to predict the interactive effects of elevated background ozone and carbon dioxide concentrations on wheat. There are six objectives to this chapter. First, to test whether the damage caused to wheat photosynthesis by chronic ozone exposure can be related to an effective ozone dose, above a threshold flux. Second, to construct a model to simulate the effects of chronic ozone exposure on wheat photosynthesis. Third, to use the model to simulate the interactive effects of ozone and carbon dioxide on leaf photosynthesis in wheat. Fourth, to test the model against data at [CO2] of 700 µmol mol-1, measured by I.F. McKee, to determine whether elevated [CO2] affords additional protection against ozone damage, above that provided by stomatal closure. Fifth, to test the model against one data point from experimental work conducted by P.K. Farage on a different cultivar of wheat. Sixth, to predict the effects of chronic ozone exposure on wheat productivity, in terms of biomass, using a simple canopy model, scaled-up from the leaf level model. As chronic ozone dose has an effect on the whole plant throughout development, the ozone effect canopy model required the modelling of wheat growth to a mature canopy.

Chapter Six is a discussion of the findings of this study, and, by analysing the limitations and uncertainties, indicates where further research is needed to improve the accuracy and reliability of model predictions.

By attaining the objectives of each chapter it is hoped to attain the overall aim of the thesis, to predict some of the interactive effects of atmospheric and climate change on vegetation response, and so identify potential feedback effects.

[ Home ] [ Up ]

Send mail to humph@essex.ac.uk with questions or comments about this web site.
Copyright © 1998 WIMOVAC Ltd.
Last modified: August 19, 1997