Notice: Wiley Online Library will be unavailable on Saturday 30th July 2016 from 08:00-11:00 BST / 03:00-06:00 EST / 15:00-18:00 SGT for essential maintenance. Apologies for the inconvenience.
Correspondence: Andrew G.Peterson Earth and Environmental Sciences, Lamont-Doherty Earth Observatory, Columbia University, 61 Route 9 W, Palisades, NY 10964, USA. Fax: USA + (914) 365–8150; e-mail: peterson@LDEO.Columbia.edu
Previous modelling exercises and conceptual arguments have predicted that a reduction in biochemical capacity for photosynthesis (Aarea) at elevated CO2 may be compensated by an increase in mesophyll tissue growth if the total amount of photosynthetic machinery per unit leaf area is maintained (i.e. morphological upregulation). The model prediction was based on modelling photosynthesis as a function of leaf N per unit leaf area (Narea), where Narea = Nmass×LMA. Here, Nmass is percentage leaf N and is used to estimate biochemical capacity and LMA is leaf mass per unit leaf area and is an index of leaf morphology. To assess the relative importance of changes in biochemical capacity versus leaf morphology we need to control for multiple correlations that are known, or that are likely to exist between CO2 concentration, Narea, Nmass, LMA and Aarea. Although this is impractical experimentally, we can control for these correlations statistically using systems of linear multiple-regression equations. We developed a linear model to partition the response of Aarea to elevated CO2 into components representing the independent and interactive effects of changes in indexes of biochemical capacity, leaf morphology and CO2 limitation of photosynthesis. The model was fitted to data from three pine and seven deciduous tree species grown in separate chamber-based field experiments. Photosynthetic enhancement at elevated CO2 due to morphological upregulation was negligible for most species. The response of Aarea in these species was dominated by the reduction in CO2 limitation occurring at higher CO2 concentration. However, some species displayed a significant reduction in potential photosynthesis at elevated CO2 due to an increase in LMA that was independent of any changes in Narea. This morphologically based inhibition of Aarea combined additively with a reduction in biochemical capacity to significantly offset the direct enhancement of Aarea caused by reduced CO2 limitation in two species. This offset was 100% for Acer rubrum, resulting in no net effect of elevated CO2 on Aarea for this species, and 44% for Betula pendula. This analysis shows that interactions between biochemical and morphological responses to elevated CO2 can have important effects on photosynthesis.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
While biochemical aspects of photosynthetic enhancement at elevated CO2 have been studied extensively, less emphasis has been placed on understanding how those enhancements interact with changes in leaf morphology and anatomy. In an important advancement, Pachepsky & Acock (1996) developed a two-dimensional model of leaf-level photosynthesis that demonstrates how leaf anatomy can have an important effect on photosynthesis. However, the general utility of this model is limited because it requires detailed maps of mesophyll cell location and information on stomatal area and density to parameterize it. Luo et al. (1994) developed the Photosynthetic Acclimation to CO2 (PAC) model which is much less data-intensive and which suggests that interactions between biochemical and morphological responses to elevated CO2 can have important effects on leaf-level photosynthesis. The PAC model uses the biochemical model of Farquhar et al. (1980) to model photosynthesis as a function of nitrogen per unit leaf area (Narea) (e.g. Harley et al. 1992 ). The model assumes that (i) nitrogen per unit leaf mass (Nmass) reflects the biochemical capacity for photosynthesis (e.g. Körner & Miglietta 1994) since the majority of leaf N is found in the proteins of the Calvin cycle (including Rubisco) and the thylakoid membranes ( Evans 1989) and (ii) that LMA summarizes aspects of leaf morphology and anatomy such as photosynthetic biomass and non-structural carbohydrate storage. The PAC model links changes in biochemical capacity and leaf morphology through the relationship Narea = Nmass×LMA. The model predicts that a reduction in biochemical capacity at elevated CO2 (estimated from a reduction in Nmass) may be counterbalanced by morphological upregulation if an increase in LMA is relatively greater than the reduction in Nmass. Leaf mass per unit area has been shown to be tightly correlated with leaf thickness in soybeans grown at ambient and elevated CO2 ( Sims et al. 1998 ), but there is also substantial evidence to show that increases in LMA at elevated CO2 can be caused by increased storage of non-structural carbohydrates (e.g. Roumet et al. 1996 ). However, if both Narea and LMA increase while Nmass decreases, then the increase in LMA must involve additional N-bearing tissue even though N concentration decreased.
The PAC model provides valuable insights into relationships between biochemical and morphological responses to elevated CO2 by showing how net changes in Nmass and LMA can interact to affect photosynthesis. However, because the PAC model focuses on net changes, it does not allow us to accurately assess the relative importance of biochemical versus morphological responses. This could be achieved by experimentally controlling Nmass or Narea in addition to LMA, but this is clearly impractical. An alternative approach would be to use a simple multiple regression of Aarea on CO2 concentration, Narea and LMA. Since the coefficients in a multiple regression are partial derivatives, they quantify the relationship between the dependent variable (Aarea) and each independent variable when the other independent variables in the equation are held constant. This allows us to statistically control both Narea and LMA, and to partition the response of Aarea among each independent variable. Using this approach, the partial regression coefficient for CO2 represents an index of CO2 limitation of photosynthesis because it quantifies the change in Aarea per unit change in CO2 when both Narea and LMA are held constant. Similarly, the partial regression coefficient for Narea represents an index of biochemical capacity per unit leaf area, and the coefficient for LMA represents an index of morphological regulation.
The multiple regression discussed above would provide useful information on the relative importance of the effects of CO2 concentration, Narea and LMA on photosynthesis. This model is, however, too simplistic to capture the range of interactions that are known, or that are likely to occur between these variables. To start with, both Narea and Nmass are correlated with LMA at ambient CO2 ( Reich et al. 1994 ; Reich & Walters 1994; Peterson et al. 1999 ). This lack of independence may confound parameter estimation and interpretation in the simple regression model. In addition, the response of Narea, Nmass and LMA to elevated CO2 are also likely to be correlated with each other. This complex set of correlations make it difficult to tease apart the individual responses. These correlations can, however, be dealt with using statistical control by framing the multiple regression model in the context of a structural equation model ( Hoyle 1995). A structural equation model (often referred to as path analysis or covariance structure analysis) can be specified as a system of multiple regression equations that partitions interactions among multiple dependent and independent variables (see the Methods section for details). In this paper we present a model that partitions the response of Aarea to elevated CO2 into independent and interactive components representing changes in an index of CO2 limitation, an index of biochemical capacity and an index of leaf morphology. Although the model is not mechanistic, it partitions known and hypothesized relationships in a conceptually and mechanistically plausible framework based on the general concepts of the PAC model. We used this linear model to re-analyse existing data from 10 C3 species (three pine and seven deciduous tree species) grown at ambient and elevated CO2 concentrations. One aim of this study was to develop a computationally and data efficient tool for assessing the relative importance of changes in biochemical capacity and leaf morphology that could be applied to biogeochemical or biogeographic models of global change. We also compared responses across species to identify whether generalizations could be made from these data based on functional or taxonomic relatedness.
MATERIALS AND METHODS
Data used in this analysis were obtained from 10 separate chamber-based elevated CO2 experiments, conducted in the field, and included three pine and seven deciduous tree species (see caption to Fig. 2 for details). Data consisted of rates of net photosynthetic carbon assimilation [Aarea, μmol (CO2) m−2 (leaf) s−1] measured at light saturation under growth conditions; leaf N per unit leaf area [Narea, g (N) m−2 (leaf)]; and leaf mass per unit area [LMA, g (leaf) m−2 (leaf)]. In most data sets leaf N was determined using the same leaves that photosynthesis was measured on, although in some cases adjacent leaves were collected for N analysis. Variation in leaf N resulted from either fertilization treatments, sun versus shade leaves, leaf developmental stage, natural variation within leaf classes, or variation due to CO2 treatment (see citations for details). Photosynthesis measurements were made at ecologically relevant temperatures for each species (20 to 30 °C depending on species) and measurements for single species were usually controlled to within ± 2 °C. Ambient CO2 concentration was either 350 or 360 μmol mol−1 and the elevated concentration was either 650 or 700 μmol mol−1 (see citations for details).
Structural equation model
Structural equation models are a class of general linear models that include analysis of variance ( ANOVA), multiple regression, path analysis and factor analysis ( Hoyle 1995). In this analysis we used a system of multiple regression equations (presented below) to approximate known and hypothesized relationships among CO2, Aarea, Narea and LMA. Path analysis (also known as covariance structure analysis) could be used in this situation but we chose the multiple regression approach for the following reasons. First, path analysis is designed to analyse hypothesized relationships among measured and latent (unmeasured) variables that are inferred from a covariance matrix ( Hoyle 1995). Because there are no latent variables in our model (other than the error terms), path analysis provides no benefit over multiple regression analysis in this case. Second, even though path analysis and regression analysis give identical slope coefficients for the relationships modelled in this study, path analysis does not provide information on the intercepts, which are important for comparing responses across species.
The system of regression equations used in this analysis was
The β coefficients for each equation are partial regression coefficients and give the slope of the relationship between the appropriate dependent and independent variable when the other independent variables in that equation are held constant. Two important coefficients in these equations are β3 (the direct effect of CO2 on Narea) and β6 (the direct effect of CO2 on Aarea). Even though β3 has units of g (N) m−2 (leaf), it is an index of change in N concentration (Nmass) because it quantifies change in the mass of N per unit leaf area when leaf mass per unit area is held constant. β6 is an index of CO2 limitation of photosynthesis because it quantifies the effect of CO2 on Aarea when Narea and LMA are both held constant. A graphical representation of this model is shown in Fig. 1.
The model was fitted to data from each species using least squares, and 95% confidence intervals for coefficients were estimated from 2000 bootstrap replicates using the bias corrected and accelerated method of Efron & Tibshirani (1993). Bootstrapping involves resampling with replacement from the original data to simulate multiple samples from a population. It is useful when the statistical properties of the comparison being made are not well understood, as is the case for calculating confidence intervals for the indirect effects of CO2 which are discussed below. Confidence intervals were used to assess the statistical significance of all coefficients.
Model 1 allows us to partition the effect of elevated CO2 on Aarea into components representing a direct effect of CO2 (β6) and indirect effects consisting of interactions among CO2, Narea and LMA. For example, the indirect effect of CO2 on Aarea due to CO2 induced changes in LMA is represented by the compound pathway CO2→LMA→Aarea ( Fig. 1). In this pathway the arrows represent specific coefficients from model 1 with the first arrow being β1 and the second arrow β8. The magnitude of this indirect effect is the product of these coefficients. Model 1 has three different pathways representing indirect effects of CO2 on Aarea: (1) CO2→LMA→Aarea; (2) CO2→LMA→Narea→Aarea; and (3) CO2→Narea→Aarea ( Fig. 1). Each indirect effect has units of μmol (CO2) m−2 (leaf) s−1 and the total indirect effect of CO2 on Aarea is the sum of these three independent effects ( Sokal & Rohlf 1981). Because these indirect effects have the same units and are additive they provide comparative information on the relative importance of each set of interactions on Aarea.
In this analysis all independent variables except CO2 were random. Least squares regression is appropriate when independent variables are random as long as their frequency distributions are not functions of the regression coefficients ( Neter, Wasserman & Kutner 1990; p 86). We assumed that this was the case for all data sets. Normality and homogeneity of residuals were checked using the original non-bootstrapped data. Because we expected correlations among the independent variables we paid particular attention to possible effects of multicolinearity on parameter estimates. Multicolinearity was checked using the original non-bootstrapped data but there was no evidence to suggest that it was a problem. We concluded that all original data sets showed adequate compliance with the assumptions of least squares regression.
Comparing responses across species
If the coefficients from model 1 were more similar within than between vegetation types (pines versus deciduous trees), then the accuracy of ecosystem and global models may be improved by incorporating the specific details of each group. We assessed the similarity of coefficients across species in two ways. First, we compared the estimated population-level frequency distributions obtained from the bootstrap replicates. If the means and 95% confidence limits for different species were similar then those species could be grouped on the premise that they had similar responses to elevated CO2. Otherwise those species were assumed to have different frequency distributions and therefore different responses.
Second, we used non-parametric and parametric cluster-analysis ( Everitt 1980; Sokal & Rohlf 1981; Digby & Kempton 1987) on the coefficients to determine if species formed discrete groups. This would imply that the complete set of coefficients were more similar within than between groups. We used hierarchical cluster-analysis to identify potential groupings and then k-means cluster analysis to test whether those groups were robust. The hierarchical procedure compares the Euclidean distance between points (species) in n-dimensional space (where n is the number of coefficients in model 1) and links those points that are closest together into clusters. The k-means procedure is an optimization procedure that assigns observations to a predefined set of k groups so that the within group variance is minimized and the between group variance is maximized. We based k on the number of major groups identified by the hierarchical procedure. All coefficients were standardized to have a mean of zero and standard deviation of one, and the hierarchical clustering criterion was complete linkage using Euclidean distances.
RESULTS AND DISCUSSION
Growth at elevated CO2 increased net photosynthesis for nine of the 10 species in this study, although this increase was marginal for both Acer saccharum and Pinus radiata ( Fig. 2a). Elevated CO2 did not have a significant effect on Narea for the majority of species even though there was a generally consistent trend for Narea to be slightly lower under elevated versus ambient CO2 ( Fig. 2b). Alnus glutinosa was an exception because it displayed a significant increase in Narea under elevated CO2, whereas P. radiata displayed a significant reduction in Narea ( Fig. 2b). In order to simplify the discussion of how these patterns in Aarea and Narea are explained by interactions among CO2 concentration, Nmass and LMA, we focus on general responses across species. Individual species that differed from these general patterns are discussed separately. First we present results for the effects of elevated CO2 on LMA and Narea, then we link those results to the direct and indirect effects of CO2 on Aarea.
Effects of elevated CO2 on LMA and Narea
All species displayed a significantly positive intercept for the regression of LMA on CO2 concentration ( Fig. 3a) and increased CO2 resulted in an increase in LMA for the majority of species ( Fig. 3b). This is a common response for C3 plants ( Curtis 1996). The Narea intercept was also positive for the majority of species ( Fig. 4a), but elevated CO2 generally had a negative direct effect on Narea when LMA was held constant ( Fig. 4b). The fact that LMA was held constant means that N concentration (Nmass) decreased and suggests that structural or non-structural carbohydrates increased with CO2 (e.g. Radoglou & Jarvis 1992; Ryle, Powell & Davidson 1992; Körner & Miglietta 1994; Schechter, Proctor & Elfving 1994; Thomas & Griffin 1994). LMA had a positive direct effect on Narea when CO2 was held constant, with N accounting for approximately 1 to 2% of the increase in leaf mass ( Fig. 4c). This positive effect of LMA on Narea suggests that leaves with higher LMA had more mesophyll tissue at a fixed CO2 concentration.
The indirect pathway CO2→LMA→Narea [units, g (N) m−2] quantifies how CO2 induced changes in LMA affects Narea. This indirect effect was not significant for Betula pendula, Fagus sylvatica, Liriodendron tulipifera, Populus euramericana and Pinus ponderosa ( Fig. 4d). Of these five species, F. sylvatica, L. tulipifera and P. ponderosa did not display a significant effect of CO2 on LMA ( Fig. 3b), which explains why the indirect effect of CO2 on Narea was not significant for these species. In contrast, B. pendula and P. euramericana both displayed higher LMA under elevated versus ambient CO2 ( Fig. 3b), but the non-significant indirect effect of CO2 on Narea suggests that this increase in LMA was due largely to the accumulation of carbohydrates. Acer rubrum, A. saccharum, Alnus glutinosa and Pinus taeda all displayed positive effects of CO2→LMA→Narea ( Fig. 4d). For these four species, the CO2-induced increase in LMA appeared to be due at least in part to increased mesophyll tissue growth. However, for both A. rubrum and A. saccharum, the net effect of joint variation in Nmass and LMA due to elevated CO2 resulted in no overall change in Narea ( Fig. 2b). Alnus glutinosa and P. taeda both tended to have higher Narea under elevated CO2 ( Fig. 2b). This was because both species displayed non-significant direct effects of CO2 on Narea (i.e. no reduction in Nmass) ( Fig. 4b) and large positive effects of CO2 on LMA ( Fig. 3b) and of LMA on Narea ( Fig. 4c). These results suggest that both of these species displayed an increase in mesophyll tissue growth and little or no accumulation of carbohydrates with increasing CO2 concentration.
Pinus radiata also differed from the majority of species because it displayed a significantly negative effect of CO2 on LMA ( Fig. 3b) (see also Hocking & Meyer (1991), Coleman & Bazzaz (1992), Ryle et al. (1992) and Knapp et al. (1994) for other examples of negative effects of CO2 on LMA. This negative effect of CO2 on LMA resulted in a significantly negative indirect effect of CO2 on Narea ( Fig. 4d). This negative indirect effect combined with a marginal reduction in Nmass to significantly reduce Narea at elevated CO2 ( Fig. 2b). These results for P. radiata suggest that this species experienced both a reduction in Nmass and a reduction in mesophyll tissue under elevated CO2.
Effects of elevated CO2 on Aarea
There was substantial variation across species for the Aarea intercept, with some species displaying significantly positive values while others displayed significantly negative values ( Fig. 5a). All species did, however, display a positive and significant direct effect of CO2 on Aarea ( Fig. 5b). This direct effect suggests a significant reduction in the CO2 limitation of photosynthesis because both Narea and LMA were held constant statistically. Thus leaves grown at elevated CO2 had higher photosynthesis for a given Narea and a given LMA compared to leaves grown at ambient CO2. The direct effect of Narea on Aarea was also positive for most species when CO2 and LMA were held constant ( Fig. 5c). This positive association between Narea and Aarea is consistent with many previous studies (e.g. Field & Mooney 1986; Walters & Field 1987; Evans 1989; Reich et al. 1994 ). In contrast, LMA tended to have either no effect or a significant negative effect on Aarea when CO2 and Narea were held constant ( Fig. 5d).
The indirect effect of CO2→LMA→Aarea ( Fig. 6a), which compares leaves with the same Narea but with LMA free to vary with CO2 concentration, was either zero or significantly negative and corresponded with the direct effect of LMA on Aarea discussed above. That is, species which displayed a negative and significant effect of LMA on Aarea also displayed a negative and significant effect of CO2→LMA→Aarea. This negative effect of CO2 on Aarea via LMA suggests the presence of a morphological mechanism of photosynthetic inhibition that is independent of any effect of CO2 on Narea.
The indirect effect of CO2→LMA→Narea→Aarea was not significant for most species ( Fig. 6b), indicating that increased Narea due to greater mesophyll tissue growth under elevated CO2 was either non-existent or too small to affect Aarea. Species that did display a positive effect of CO2→LMA→Narea→Aarea (i.e. morphological enhancement of photosynthesis) were those that also displayed a strong direct effect of Narea on Aarea. In contrast, the indirect effect of CO2→Narea→Aarea, which compares leaves with the same LMA but with Narea free to vary with CO2 concentration (i.e. variation in Nmass), was significantly or marginally negative for most species ( Fig. 6c). This negative indirect effect suggests a reduction in the biochemical capacity for photosynthesis at a given LMA, possibly due to the dilution of N caused by increased carbohydrate storage. The total indirect effect of CO2 on Aarea ( Fig. 6d), which is the sum of the indirect effects discussed above, was not significant for all but two species (see below). This analysis suggests that in general, the enhancement of Aarea under elevated CO2 was dominated by the reduction in CO2 limitation of photosynthesis. Morphological upregulation at elevated CO2 did not have an appreciable effect on Aarea for the majority of species in this study.
The total indirect effect of CO2 on Aarea discussed above was significantly negative for both A. rubrum and B. pendula ( Fig. 6d). The negative response for these two species was due to a combination of morphological reduction of photosynthesis (CO2→LMA→Aarea, Fig. 6a) and a reduction in biochemical capacity (CO2→Narea→Aarea, Fig. 6c). For A. rubrum, this combination cancelled the positive effect of reduced CO2 limitation ( Fig. 5b) and explains why there was no net effect of CO2 on photosynthesis for this species ( Fig. 2a). For B. pendula, the negative effects of morphology and lower biochemical capacity nearly halved (44% reduction) the enhancement caused by reduced CO2 limitation, resulting in an enhancement of Aarea under elevated CO2 that was still significant ( Fig. 2a), although substantially constrained.
Comparisons across species
An important result of this analysis was the variation across species for most of the coefficients in model 1. Some of this variation may represent species differences in response to elevated CO2, although some of it may reflect differences in experimental design such as nutritional status, tree age, tree density, the seasonal timing of data collection, the temperature at which measurements were made, or to differences among experiments in the CO2 concentrations used for the ambient and elevated treatments (see relevant citations for details). All experiments were, however, conducted under conditions that were considered ecologically relevant but necessarily artificial. Even though the variation between species may be reduced by using standardized experimental conditions, it may still reflect important natural variation that needs to be considered in large-scale models of plant responses to global change. Nevertheless, there seems to be little evidence to suggest that the taxonomically or functionally similar species presented in this study have comparable frequency distributions for many of these coefficients. This high degree of variation across species resulted in the averages for most coefficients being non-significant for each vegetation type (averages are presented in each figure). The only coefficients for which these averages were significant were the LMA intercept ( Fig. 4a), the Narea intercept ( Fig. 3a) and the direct effect of CO2 on Aarea ( Fig. 5b).
Despite the variation discussed above, the hierarchical cluster analysis identified four groups that were also confirmed by the k-means cluster analysis. Two of these contained only one species each, namely P. euramericana and P. radiata. The response of P. euramericana tended to be opposite that of the other species for several coefficients. This was most notable for the Narea intercept, the direct effect of CO2 on Narea and the direct effect of LMA on Narea ( Fig. 7). Pinus radiata appeared to differ from the majority of species with respect to the LMA intercept, the effect of CO2 on LMA and the direct effect of Narea on Aarea ( Fig. 7). The third group consisted of A. glutinosa and B. pendula. These two species differed from the remaining six species by having relatively higher intercepts for Narea and Aarea, a relatively larger direct effect of CO2 on Aarea and a more strongly negative effect of LMA on Aarea ( Fig. 7). The last group included A. rubrum, A. saccharum, F. sylvatica, L. tulipifera, P. ponderosa and P. taeda ( Fig. 7).
Although this analysis did identify apparently robust groupings of species, there were no clear associations based on taxonomic or functional relatedness. In fact, the analysis showed that some taxonomically divergent species were comparable for the full set of coefficients presented in model 1. This may simply represent an artifact of the conditions used in each experiment, but we may be able to confirm or refute these patterns as additional data from newer experimental technologies such as Free Air CO2 Enrichment become available. Nevertheless, it may be worthwhile conducting sensitivity analyses on models of global change by incorporating the range of variability observed here to determine if it has important effects on model predictions. Other sources of natural variation, such as temporal variation during the growing season and lag-effects from previous growing seasons, may also affect the relationships modelled in this study. Understanding the temporal dynamics of these relationships may be crucial for developing accurate and robust models of plant growth in changing environments.
In summary, this analysis extended the PAC model of Luo et al. (1994) and suggested that the general enhancement of Aarea under elevated CO2 was dominated by a reduction in CO2 limitation of photosynthesis as indicated by the direct effect of CO2 on Aarea. Contrary to the prediction of the PAC model we failed to support the hypothesis that reduction in biochemical capacity is frequently offset by increased LMA. Instead, we found that increased LMA due to elevated CO2 can significantly reduce photosynthesis in some species through a mechanism that appeared to be morphologically based but independent of Narea. The nature of this mechanism is not clear but may include a combination of factors mentioned previously, such as a reduction in N allocation to photosynthetic versus non-photosynthetic compounds ( Evans 1989), to greater allocation of biomass to structural versus photosynthetic components ( Vitousek et al. 1990 ; Lloyd et al. 1992 ), to greater internal shading ( Terashima & Hirosaka 1995), or to greater limitations to internal diffusion ( Parkhurst 1994; Pachepsky et al. 1997 ). Nevertheless, this morphological reduction in photosynthesis combined additively with a reduction in biochemical capacity to significantly reduce potential photosynthetic enhancement at elevated CO2 in two species.
This synthesis was supported by the US Department of Energy, grant number DE-FG02–95ER62084 to the CMEAL programme. The Electric Power Research Institute (EPRI) provided support for CMEAL meetings. Belinda Medlyn played an invaluable role in negotiating collaboration with members of the European Collaboration On CO2 Responses Applied to Forests and Trees programme (ECOCRAFT). Support for the initial organization of CMEAL was provided by the Desert Research Institute. Comments from three anonymous reviewers substantially improved an earlier version of this manuscript. CMEAL is a core project of the GCTE (Global Change and Terrestrial Ecosystems) programme.
CO2 Models/Experiments Activity for improved Links (CMEAL)
CMEAL is a collaborative project aimed at improving the representation of the CO2 responses in ecosystem and global models. Participants are: C. B. Field (Carnegie Institution of Washington, Co-PI); J. T. Ball (Desert Research Institute, Co-PI); J. S. Amthor (Lawrence Livermore National Laboratory); B. Drake (Smithsonian Environmental Research Center); W. R. Emanuel (University of Virginia); D. W. Johnson (Desert Research Institute); P. J. Hanson (Oak Ridge National Laboratory); Y. Luo (Desert Research Institute); R. E. McMurtrie (University of New South Wales, Australia); R. J. Norby (Oak Ridge National Laboratory); W. C. Oechel (California State University, San Diego); C. E. Owensby (Kansas State University); W. J. Parton (Colorado State University); A. G. Peterson (Desert Research Institute); L. L. Pierce (California State University, Monterey Bay); E. B. Rastetter (Marine Biological Laboratory); A. Ruimy (Universite Paris-Sud); S. W. Running (University of Montana); and D. R. Zak (University of Michigan).