SEARCH

SEARCH BY CITATION

Keywords:

  • net primary production;
  • transpiration;
  • NDVI;
  • radiation use efficiency;
  • scaling up;
  • grasslands

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Model Description
  5. 3. Study Area and Data Sets
  6. 4. Results
  7. 5. Discussion and Conclusions
  8. Acknowledgments
  9. References

[1] Biophysical and physiological processes in plants and ecosystems occur over a wide range of spatial and temporal scales. Our knowledge (or models) of these processes is largely at small scales. It is, however, difficult to directly apply mechanistic process-oriented models over large scales due to heterogeneities in the distributions of processes, and nonlinearities in the functional responses of processes to environmental variables. On the other hand, simple parametric/empirical models in which system complexity is lumped into a small number of parameters have been widely employed to describe processes at larger scales. The variation of these parameters in these simple parametric/empirical models depends on the underlying biophysical processes. In this work, we showed that detailed process models and simple parametric models for primary production and transpiration could be effectively combined to scale leaf photosynthesis and transpiration up to large spatial scales. The integrated process model, General Energy Mass Transfer Model (GEMTM), was used to identify major factors contributing to the variability of the parameters in the parametric models for regional transpiration and primary production and quantify their responses to these factors. Simulations with the GEMTM showed that net carbon assimilation was proportional to intercepted photosynthetically active radiation (IPAR), but the radiation use efficiency (RUE) changed with leaf N concentration, temperature, and atmospheric CO2 concentration; transpiration was linearly correlated with the product of net primary production (NPP) and atmospheric water vapor pressure deficit (VPD), and the slope varied with leaf N concentration. RUE increased with leaf N content asymptotically, and responded to temperature in an asymmetric bell shape pattern with a 22°C and 26°C optimal temperature under current ambient and doubled CO2 concentration, respectively. A simple parametric NPP model and a regional transpiration model (Tr model) were developed from the relationships and parameter values obtained using the GEMTM. The NPP model reasonably simulated the seasonal and interannual variations of accumulated NPP estimated from field data. Simulated regional distribution of NPP over the Central Grassland Region of the United States was consistent with estimates obtained using other models. NPP increased from 120 gC/m2/year in the northwest to 956 gC/m2/year in the southeast. Simulated regional transpiration had a similar spatial distribution pattern as NPP, ranging from about 16 cmH2O/year to 136 cmH2O/year. The transpiration model introduced in this study provides a mechanism to explicitly couple transpiration and NPP in large-scale analyses, although more complete analysis and validation are required.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Model Description
  5. 3. Study Area and Data Sets
  6. 4. Results
  7. 5. Discussion and Conclusions
  8. Acknowledgments
  9. References

[2] Net primary production (NPP) forms an important link between plant ecosystems and the atmosphere through the exchange of CO2 and has important influences on soils, water fluxes, nutrient cycles, and climate. It describes both the removal of CO2 from the atmosphere and the potential supply of carbon to herbivores, and decomposers. It is an indicator of overall ecosystem functioning [McNaughton et al., 1989]. Understanding NPP, its controls, and its interactions with the physical environment is essential to understanding the biogeochemistry of Earth [Raich et al., 1991].

[3] Processes in plants and ecosystems occur over a wide range of spatial and temporal scales [Ojima et al., 1991]. The “scaling up” process involves taking information at smaller spatial and shorter temporal scales and using that information to derive estimates of processes at larger spatial and longer temporal scales [Jarvis, 1995]. It is nonlinearities in the relationships between processes and driving variables that makes the scaling difficult and challenging. Consider, for example, the photosynthetic response to photosynthetic active radiation (PAR). Because of the typical nonlinear asymptotic form of the function, the average photosynthetic rate computed by average PAR received by a leaf over a day differs significantly (up to 40%) from the rate obtained by averaging instantaneous photosynthetic rates [Hari et al., 1984]. Similarly, nonlinearity in the functional responses between a process and a state variable, or parameter, leads to similar difficulties (e.g., transpiration response to stomatal conductance [Jarvis and McNaughton, 1986; McNaughton, 1994]).

[4] During the past 3 decades, enormous advances have been made in our understanding of the biochemistry of photosynthesis [Farquhar et al., 1980; Harley et al., 1992; Chen et al., 1994], stomatal behavior, and biophysics of transpiration [Monteith, 1965, 1995; Cowan, 1977, 1982; Jarvis, 1976; Ball et al., 1987]. Several models which summarized this knowledge at the individual leaf level have been constructed [Collatz et al., 1991; Leuning, 1995; Nikolov et al., 1995]. Leaf level processes have been scaled up to the canopy level in many different canopy models [Goudriaan, 1977, 1989; Norman, 1979; Coughenour, 1984; Chen and Coughenour, 1994; Sellers et al., 1986]. A major challenge is to extend this understanding at leaf and canopy scales to regions and the globe [Mooney, 1991; Schimel et al., 1993].

[5] Individual leaf level processes have been integrated to the canopy level using multilayer or “big leaf” canopy models [Goudriaan, 1989; Raupach and Finnigan, 1988; Norman, 1980]. Two-layer (sunlit and shaded leaves) models have been recommended [Norman, 1980; Jagtap and Jones, 1989; Goudriaan, 1989; Chen and Coughenour, 1994]. However, it is difficult to apply mechanistic biophysical process canopy models in regional and global scale studies of biogeochemical cycles, because these process models require a large number of parameters and initial variables that are not universal constants, and are unknown except for a few point measurement locations.

[6] Although hierarchical modeling approaches which link different models addressing leaf, canopy, ecosystem, and regional scale processes have been suggested in the past [Kittel and Coughenour, 1988; Strain and Bazzaz, 1983; Reynolds and Acock, 1985], few model systems have been implemented in this way, because models at different scales usually have been developed for different purposes, and are therefore based upon different, sometimes conflicting, assumptions. On the other hand, simple parametric/empirical models in which system complexity is lumped into a small number of parameters have been employed to describe processes at larger scales [e.g., Monteith, 1977; Woodward and Smith, 1994; Schultze et al., 1994].

[7] The Monteith model of net primary production (NPP) estimates the NPP of crops as a function of intercepted radiation and radiation use efficiency (RUE) [Monteith, 1977; Kumar and Monteith, 1981]. This model or its variation has been widely used in monitoring NPP with remotely sensed data at regional and global scales under the assumption that radiation use efficiency can be approximated or assumed nearly constant [Ruimy et al., 1994; Prince, 1991; Potter et al., 1993]. Unfortunately, the radiation use efficiency is not constant, but depends on factors such as leaf nitrogen, plant type (e.g., C3 and C4), temperature, and water availability [Sinclair and Horie, 1989; Potter et al., 1993; Field et al., 1995; Prince and Goward, 1995; Landsberg and Hingston, 1996]. Detailed process models may bridge the gap between the complexity of small-scale process-oriented models and the simplicity of parametric/empirical models by identifying major factors contributing to the variability of parameters in simple parametric/empirical models.

[8] Evapotranspiration is one of the main processes of soil water depletion, and so is one of the important processes of the Earth's hydrology and biogeochemistry. It links land surfaces and the atmosphere, and interacts with carbon and nitrogen cycling. The predictive capability of any water budget model depends critically on the accuracy of measurement or estimation of antecedent evapotranspiration. However, as compared to rainfall, there is little systematic measurement of evapotranspiration. Remotely sensed data have been used to estimate evapotranspiration as the residual component of the energy balance [Jackson et al., 1977, 1983, 1987] [also see Choudhury, 1994]. Transpiration, the major plant controlled component of total evapotranspiration, can be regarded as a cost that plants pay to assimilate carbon [Cowan, 1982]. There is convincing evidence that carbon assimilation exercises a significant control on stomatal conductance, and that stomata function in such a way that carbon gain is optimized in relation to transpiration water loss [Cowan, 1977, 1982; Monteith, 1988], i.e.,

  • equation image

where TR is transpiration, A is carbon assimilation, and C is constant.

[9] The objectives of this study are to (1) quantify the relationships between the radiation use efficiency (RUE), leaf N content, and environmental factors using our existing integrated process model (GEMTM) [Chen et al., 1994, 1996; Chen and Coughenour, 1994]; (2) apply the Monteith NPP model [Monteith, 1977; Kumar and Monteith, 1981] with the derived RUE functions and remotely sensed data to explore the temporal and spatial distribution of NPP over the central grassland region of the USA; and (3) explore the use of remotely sensed data and the relationship between transpiration and carbon assimilation [Tanner and Sinclair, 1983; Monteith, 1988, 1990] in estimating regional transpiration over the central grasslands.

[10] With the above objectives, we first describe the integrated process model GEMTM, the simple parametric/empirical NPP model, and the regional transpiration (Tr) model. Then we compare the output of the GEMTM model with selected field measurements, and use the GEMTM to quantify the responses of the parameters of the NPP model and the Tr model to leaf N concentration and environmental factors. Finally, we apply the NPP model and the Tr model with remotely sensed data to estimate regional NPP and transpiration over the central grasslands.

2. Model Description

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Model Description
  5. 3. Study Area and Data Sets
  6. 4. Results
  7. 5. Discussion and Conclusions
  8. Acknowledgments
  9. References

2.1. GEMTM Model

[11] The general energy and mass transfer model (GEMTM) links leaf level processes, canopy microclimate, soil abiotic processes, plant growth, biomass production, and dynamics [Chen and Coughenour, 1994]. The model comprises a canopy microclimate submodel, a soil thermal dynamics submodel, a soil water dynamics submodel, a plant growth submodel that includes leaf photosynthesis and stomatal conductance, biomass production, root spatial distribution dynamics, and a soil respiration submodel. The model provides a fully coupled solution for carbon assimilation, stomatal conductance, evapotranspiration, sensible heat transfer, and canopy and soil temperatures. The key features of the model are (1) the separation of the canopy into sunlit and shaded leaves; (2) mechanistic accounting for differences between C3 and C4 photosynthesis; (3) inclusion of multiple-reflection of radiation within the canopy and soil surface system in calculations of surface energy balance; (4) coupling the energy balance of the canopy and soil surface with soil thermal dynamics to account for the interactions between soil canopy and the atmosphere in energy transfer; and (5) interactive determination of plant water potential based upon plant water balance which enables the model to explicitly represent the interactions between root water uptake, transpiration, and photosynthesis. The framework of the model is described here, and the full details of the model and its validation are given by Chen and Coughenour [1994], and Chen et al. [1994].

[12] Microclimate processes considered in the GEMTM include radiation transfer in plant canopies, canopy and soil surface energy balance, turbulent transfer, boundary layer, and surface resistances. Goudriaan's model was adopted to model radiation transfer within canopies [Goudriaan, 1977]. The radiation at any level in the canopy is divided into nine intensities in nine different zones in a hemisphere. Incident radiation is divided into visible, near infrared, and thermal radiation. The sunlit and shaded leaf area indices are calculated using the derived light extinction coefficient after Norman [1979]. Soil, canopy, and atmosphere are aggregated for calculating energy and mass exchanges. Latent (transpiration and evaporation) and sensible heat exchanges between soil surface, canopy, and the atmosphere are simulated using an electrical resistance analogy. Canopy and soil surface energy and water transfer processes are tightly linked through solutions for leaf and soil surface energy balance equations. The energy balance of the canopy and soil surface is coupled with soil thermal dynamics. The model predicts canopy temperature and soil surface temperature. Soil thermal dynamics are modeled by solving the heat conduction equation at given boundary conditions. Soil water transport is modeled based on Darcy's law and mass conservation. Root water uptake is a function of the difference between plant water potential and soil water potential and resistances along the water transport pathway.

[13] The plant growth and biomass production submodel includes leaf photosynthesis, stomatal conductance, plant respiration, photosynthate partitioning, and root-zone distribution dynamics. Carbon enters the system through photosynthesis and is distributed among leaves, stems, and roots. Photosynthesis and stomatal conductance are calculated for the sunlit and shaded leaves. Canopy level CO2 assimilation rate is the leaf area weighted average of sunlit and shaded leaf photosynthesis rates. Respiration of each of these biomass components consists of maintenance and growth respiration [Amthor, 1989].

[14] The single leaf photosynthesis model for C3 and C4 species is described by Chen et al. [1994, 1996]. The submodel for C3 photosynthesis is based on the model of Farquhar et al. [1980], which considers the relative limitations of rates of RubP carboxylase, RubP regeneration, and RubP oxidation. RubP fixation is limited by mesophyll CO2 concentration in C3 species and by bundle sheath CO2 concentration in C4 species. The maximum rate of RubP regeneration is asymptotically light limited. The C4 cycle of C4 species is limited by PEP carboxylase fixation, which responds to mesophyll CO2 concentration. Assimilation rate is negatively affected by low plant water potential, and low leaf nitrogen concentration. The response of photosynthesis to temperature is incorporated via temperature dependencies of model parameters. Stomatal conductance is simulated using the empirical model of Ball et al. [1987]. In this model, stomatal conductance responds to leaf surface CO2, relative humidity, and net assimilation rate.

[15] The root:shoot allocation is empirically simulated such that the fraction of photosynthate used for shoot growth declines with increasing water stress [Hunt et al., 1991; Chen et al., 1996]. Nitrogen is distributed in the canopy according to the light intensity profile (sunlit and shade) [Schimel et al., 1991; Hirose and Werger, 1987; Field, 1983]. Soil respiration is empirically modeled after Norman et al. [1992] as a function of soil temperature and soil water content.

[16] The photosynthetic submodels of the GEMTM have been validated using several experimental data sets under both current and doubled atmospheric CO2 levels [Chen et al., 1993, 1996; Coughenour and Chen, 1997]. The GEMTM model was validated using field data measured in the NASA FIFE experiment in Kansas tallgrass prairie, and was shown to accurately simulate water and energy fluxes [Chen and Coughenour, 1994].

2.2. Net Primary Production Model

[17] Monteith [1972, 1977] suggested that annual crop production is largely determined by variability in intercepted PAR (IPAR). Kumar and Monteith [1981] introduced a model that estimated plant production as a product of IPAR and radiation use efficiency or radiation conversion efficiency (RUE). Accordingly, we propose to use the following equations to estimate NPP:

  • equation image
  • equation image

where f is the fraction of incident PAR intercepted by green leaves, SR is incident solar radiation, cpar is the ratio of PAR to SR, a constant (0.5) here according to McCree [1981].

[18] According to previous field and model experimental results [Asrar et al., 1984; Daughtry et al., 1992; Sellers, 1985, 1987; Ruimy et al., 1994], a linear relationship between f and the normalized difference vegetation index (NDVI) is used,

  • equation image

where a and b are constants. The linear relationship between the NPP and the absorbed radiation has been used in remotely sensed estimates of NPP among a range of ecosystems and scales [Asrar et al., 1985; Daughtry et al., 1992; Tucker et al., 1983; Prince and Tucker, 1986; Goward et al., 1985; Goetz and Prince, 1996; Prince, 1991], with field estimated or theoretical values of RUE [Prince and Goward, 1995].

[19] According to Monteith [1977] [also see Field, 1991], RUE is a relatively conservative value among plant types of the same metabolic type. However, RUE can vary with phenological stage, temperature, and water stress [Jarvis and Leverenz, 1983; Potter et al., 1993; Landsberg and Hingston, 1996], and leaf nitrogen concentration [Sinclair and Horie, 1989]. To account for temperature (Ta) and leaf N concentration (Nl) effects on RUE, two scalar (zero-to-one) functions, EFTc and EFNc, are introduced,

  • equation image
  • equation image
  • equation image

where E, H, and S are parameters, Topt, in Kelvin, is the optimal temperature for EFTc, and b and Nb are parameters for EFNc. EFTc reflects plant CO2 assimilation rate responses to temperature in an asymmetric bell shape pattern [Farquhar et al., 1980; Long, 1991; Knapp et al., 1993; Chen et al., 1994]. EFNc reflects the response of plant CO2 assimilation to leaf N concentration in an asymptotic pattern [Sinclair and Horie, 1989].

[20] Here we ignored the short-term effect of soil water stress on the radiation use efficiency, since a monthly time step was used in this work and the changes in remotely sensed vegetation indices integrate well the effects of changes in rainfall [Paruelo and Lauenroth, 1995]. Indeed, the use of NDVI to estimate the fraction of the absorbed incident PAR allows one to take into account relatively long-term (weekly or monthly) soil moisture effects. For instance, the long-term water stress (drought) reduces the LAI and therefore the fraction of absorbed radiation (equation (4)).

2.3. Regional Transpiration Model

[21] de Wit [1958] first suggested that the ratio of carbon assimilation (crop yield) to transpiration is conservative after he reanalyzed existing measurements of transpiration and crop production. Tanner and Sinclair [1983] reviewed similar evidence for a number of crop species. Monteith [1988] and Tanner and Sinclair [1983] have shown from basic gas exchange theory for single leaves and canopies that the ratio of carbon assimilation to transpiration for a given plant type (C3, C4, or CAM) is largely determined by water vapor pressure deficit (VPD). It is worth noting that this theory was mainly tested on arable crops, where N is usually not a limiting factor. N availability is often limiting in natural ecosystems. N availability affects C assimilation, and then the ratio of the NPP to the transpiration Tr, which might explain the variation of the ratio among sites. Accordingly, we propose that long-term transpiration, particularly at large scales, can be estimated using a linear relationship between Tr and the product of NPP and VPD:

  • equation image
  • equation image

where Tr0 (cm H2O), sa (cm H2O m2/gC/kPa), sb, sc (cmH2O m2/gC/kPa), and Ns (%) are parameters. Tr is in cmH2O, NPP is in gC/m2, and VPD is in kPa.

3. Study Area and Data Sets

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Model Description
  5. 3. Study Area and Data Sets
  6. 4. Results
  7. 5. Discussion and Conclusions
  8. Acknowledgments
  9. References

[22] The study area encompassed a 698,100 km2 portion of the central grassland region of the United States (latitude: 37°N∼44°N; longitude: 94.5°W∼105°W). Annual precipitation increases from ∼40 cm/yr in the west to 100 cm/yr in the east of this region. Annual mean temperature increases from northwest to southeast (8°∼20°). The potential natural vegetation of the region is largely tallgrass prairie and mixed prairie, and partly shortgrass steppe. Regional dominant species are C4 grasses [Kim and Verma, 1990]. Estimates of NPP for this region were made previously using simple regression models driven by annual precipitation [Lauenroth, 1979; Sala et al., 1988].

[23] To support the use of the GEMTM in derivation of simple parameters in the NPP and Tr models, we present model simulations and their comparisons with single leaf gas exchange data and eddy correlation measurements of CO2 and energy fluxes in the next section. Single leaf photosynthesis and stomatal conductance responses to PAR and temperature under two levels of CO2 (350 ppm and 700 ppm) were measured on the C4 grass Andropogon gerardii, the dominant species of the tallgrass prairie [Knapp et al., 1993; Chen et al., 1994]. The eddy correlation measurements of CO2 flux and energy fluxes on 4 typical days in different parts of the growing season of 1987 in the tallgrass prairie were obtained from the FIFE database, and the relevant plant and soil moisture data were from Kim and Verma [1990]. Leaf nitrogen concentration and its distribution in the canopy were obtained from Schimel et al. [1991]. Simulation runs with the GEMTM model were conducted under different growth seasons and weather conditions. The outputs were used in parameterizing the NPP and Tr models using nonlinear regression of SAS [SAS Institute Inc., 1987]. Values for VPD in the regressions were derived from mean monthly temperature and humidity data (see below). The estimates of the parameters E, H, S, and Topt in equation (6) were 91720, 150,238, 529, and 295 at 1 × C, and 102,931, 137,602, 489, and 299 at 2 × C, respectively. The parameters b and Nb of equation (7) were estimated to be 5 and 0.55%, respectively. The parameters Tr0, sa, sb, sc, and Ns of equation (8) were estimated to be 0.34 cmH2O, 0.65 cmH2O m2/gC/kPa, 4, 0.06 cm H2O m2/gC/kPa, and 0.76%, respectively.

[24] The biweekly biomass clipping data of the Konza Long-Term Ecological Research project (LTER) of 1990 to 1993 were used to estimate plant production dynamics, which were used to test the NPP model. The NPP were estimated from aboveground biomass clipping data assuming ∼55% of total production to the belowground according to Marshall's [1977] summary and our data [Chen et al., 1996].

[25] Biweekly maximum NDVI composites of 1 km resolution of 1990 to 1994 were obtained from the EROS Data Center in Sioux Falls, South Dakota, for the conterminous United States. The procedures used to create the composites attempted to minimize cloud contamination and atmospheric attenuation [Ailts et al., 1990; Holben, 1986]. Images were geometrically registered to the Lambert Azimuthal Equal Area map projection. A 5-km grid spacing, contracted from the original 1-km-resolution data, was used in this study. The monthly average NDVI were calculated based on the biweekly data for the central grassland region. The NDVI data from the grid cell corresponding to the Konza LTER site were extracted for testing model predictions against the time series of biomass production data.

[26] Monthly average solar radiation, temperature, and relative humidity data were obtained from the Vegetation/Ecosystem Modeling Analysis Project (VEMAP) database [Kittel et al., 1995]. The VEMAP data were provided in a 0.5° × 0.5° grid spacing across the United States. We reprojected the VEMAP data in the Lambert Azimuthal Equal Area map projection, and resampled the data at 5 km × 5 km resolution for the central grassland region, consistent with the NDVI data.

4. Results

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Model Description
  5. 3. Study Area and Data Sets
  6. 4. Results
  7. 5. Discussion and Conclusions
  8. Acknowledgments
  9. References

4.1. Leaf Photosynthesis and Stomatal Conductance Responses to Environment

[27] A correct representation of leaf photosynthesis and stomatal conductance is central to simulate canopy carbon assimilation and transpiration. It has been demonstrated that the photosynthesis submodel of the GEMTM accurately simulated the responses of photosynthesis and stomatal conductance to temperature, PAR, and CO2 in our previous work [Chen et al., 1994, 1996; Coughenour and Chen, 1997]. An illustration of the modeled results of leaf photosynthesis and stomatal conductance response to different environment factors are shown (Figure 1). The quantum efficiency (initial slopes of the PAR response curves) and light compensation points were correctly simulated, as were saturation light intensities (Figures 1a and 1b). The response curves of photosynthesis to PAR at 350 ppm (1 × C) and 700 ppm (2 × C) CO2 concentration were quite close to each other (Figures 1a and 1b), which indicated that 1 × C CO2 concentration was already saturating photosynthesis as expected because of the CO2 concentrating mechanism of C4 photosynthesis [Pearcy and Ehleringer, 1984; Chen et al., 1994]. Simulated photosynthetic responses to temperature agreed well with the data at both CO2 levels (Figures 1c and 1d). The C4 grass exhibited an optimal temperature at ∼38°C at 1 × C and 39°C at 2 × C. At lower temperature (<10°C) photosynthesis was insignificant, while it was still active at a temperature as high as 50°C (Figures 1c and 1d). Optimal temperature of photosynthesis was shifted upward under doubled CO2 concentration which might be due to differential temperature sensitivities of solubilities and RuBP affinities for CO2 and O2 according to Long [1991]. Predicted responses of stomatal conductance to PAR precisely matched the data at both 1 × C and 2 × C levels (Figures 1e and 1f). Stomatal conductance was reduced significantly by doubling CO2 concentration. The light saturated stomatal conductances were about 130 mmol m−2 s−1 and 92 mmol m−2 s−1 at 1 × C and 2 × C, respectively.

image

Figure 1. Comparisons between model simulation (solid curves) and experimental data (circles) for C4 grass Andropogon gerardii: (a, b) photosynthesis responses to PAR; (c, d) photosynthesis responses to temperature; and (e, f) stomatal conductance responses to PAR.

Download figure to PowerPoint

4.2. Diurnal Variation of CO2 and Energy Fluxes

[28] The GEMTM simulated fluxes of CO2, latent heat, sensible heat, and soil surface heat were compared against the micrometeorological measurements on four selected days [Kim and Verma, 1990] (Figures 2 and 3). These days were distributed throughout the growing season of the tallgrass land. The model reproduced the features of the data well, despite a wide range of moisture and growing conditions. On 5 June, the early growing season and well-watered condition, simulated CO2 flux accurately matched the data (r2 = 0.97). The peak CO2 flux of about 20 μmol m−2 (ground area) s−1 occurred at noon (∼1800 GMT). Nocturnal CO2 flux (soil plus plant respiration) was ∼8 μmol m−2 (ground area) s−1. Maximum latent heat flux was ∼400 J m−2 (ground area) s−1 at early afternoon. On 2 July, when the prairie was in the peak growth stage and soil water was not limiting, the model precisely simulated the diurnal courses of CO2, latent, sensible, and soil surface heat fluxes (0.95). Peak CO2 flux was ∼30 μmol m−2 s−1, and peak latent heat flux was ∼500 J m−2 s−1. On 30 July, plants were in early senescence, soil water was limiting, and atmospheric evaporative demand was very high [Kim and Verma, 1990]. CO2 assimilation was suppressed dramatically, and the peak flux was ∼2 μmol m−2 s−1 and occurred at early morning. The midday depression of CO2 assimilation in the model was driven by low plant water potential, and low humidity resulted in stomata closing. On 15 August, soil moisture condition was improved due to frequent rainfall in early August [Kim and Verma, 1990]. CO2 assimilation was recovered from the previous dry period suppression, as was the evapotranspiration (latent heat flux). The peak CO2 flux was ∼10 μmol m−2 s−1 at local time 1100 (∼1700 GMT).

image

Figure 2. Diurnal courses of measured (symbols) and simulated (curves) CO2 flux, latent heat flux, sensible heat flux, and soil surface heat flux in a tallgrass prairie on 5 June and 2 July 1987.

Download figure to PowerPoint

image

Figure 3. Diurnal courses of measured (symbols) and simulated (curves) CO2 flux, latent heat flux, sensible heat flux, and soil surface heat flux in a tallgrass prairie on 30 July and 15 August 1987.

Download figure to PowerPoint

4.3. RUE and its Responses to Leaf N and Environment

[29] The NPP model assumes a linear correlation between net CO2 assimilation and the intercepted PAR by the whole canopy (equation (2)). It suggests that nonlinearity in the photosynthetic response to PAR at the leaf level (e.g., Figures 1a and 1b) is largely eliminated when daily totals are derived for the whole canopy. The GEMTM model was used to test the hypothesis and explore the causes of the variation of RUE. The linear relationship between net CO2 assimilation and IPAR was supported by the GEMTM model simulations (Figure 4). The slopes (i.e., RUE) between daily net C assimilation and IPAR changed with both leaf N and CO2 concentration. Leaf area index of the canopy had only a small effect on RUE, and then only at low LAI (Figure 5). At low LAI, RUE was decreased because of the high fraction of the canopy leaf areas subjected to radiation that approached photosynthetic light saturation, and then low RUE. However, once LAI reached 1, RUE was within 10% of the value at LAI 5. This is consistent with the reports of Horie and Sakuratani [1985] and Sinclair and Horie [1989].

image

Figure 4. Daily net carbon assimilation against intercepted PAR at two levels of leaf N concentration and atmospheric CO2 concentration. Symbols are the GEMTM simulations, and lines are linear regression curves. The correlation coefficients are all >0.97.

Download figure to PowerPoint

image

Figure 5. Radiation use efficiency (RUE) at different LAIs.

Download figure to PowerPoint

[30] The temperature response of RUE is presented in Figure 6. It shows that the C4 grass performs better in the conversion of solar radiation into biomass at warmer conditions than at cooler conditions. The RUE was close to zero at 0°C and reached the maximum at about 22°C and 26°C at 1 × CO2 and 2 × CO2 concentration, respectively (Figure 6, top). This response is the result of differential responses of leaf photosynthesis and plant respiration to temperature. The fitted normalized response function to this output (EFTc, equation (6)) is plotted in Figure 6 (bottom).

image

Figure 6. GEMTM (top) modeled and (bottom) normalized RUE responses to temperature at ambient and doubled CO2 concentrations.

Download figure to PowerPoint

[31] The effect of leaf N concentration on RUE is presented in Figure 7. The response of RUE to leaf N is a straightforward result of the effect of leaf N on leaf photosynthesis in the GEMTM. Figure 7 (top) shows that RUE increases near linearly with leaf N at low range of leaf N, and saturates at high leaf N. The fitted normalized response function to this output (EFNc, equation (7)) is shown in Figure 7 (bottom). The fitted values of EFNc correspond to the effect of nitrogen on photosynthesis used for the species being modeled here. The effect of nitrogen on photosynthesis may be a generally applicable one, but less precise for a given site or species, or it may be more precisely parameterized for the species at a given site, as was done here.

image

Figure 7. GEMTM (top) modeled and (bottom) normalized RUE responses to leaf N concentrations.

Download figure to PowerPoint

4.4. Temporal and Spatial Variation of NPP

[32] Simulated time evolution of cumulative NPP from 1990–1993 using the NPP model compared favorably with the field data collected under NSF LTER research program at the Konza Prairie Research Natural Area located near Manhattan, Kansas (Figure 8). The NPP model reproduced interannual and seasonal variation of NPP data well. The annual NPP was low in 1991. The model underestimated the peak cumulative NPP of 1993. This might be partly due to the same leaf N concentration used for all years of simulation, and the lack of consideration of short-term water stress effects and changes in allocation patterns.

image

Figure 8. Comparison between simulated NPP (curve) using the NPP model and field data (symbols) at Konza LTER site from 1990 to 1993. Vertical bars represent standard error intervals of the data points.

Download figure to PowerPoint

[33] Simulated spatial distribution of annual NPP of 1991 over the central grassland region of the USA using the NPP model is presented in Figure 9. It shows that NPP increases from northwest to southeast. NPP ranges from 120 to 956 gC/m2/yr. This is consistent with estimates obtained using other NPP models [e.g., Sala et al., 1988; Burke et al., 1991].

image

Figure 9. Simulated spatial distribution of annual NPP of 1991 over the central grasslands region of the United States using the NPP model.

Download figure to PowerPoint

4.5. Regional Transpiration

[34] Our GEMTM simulations showed that transpiration increased almost linearly with the product of NPP and VPD, which supported the hypothesis that the ratio of transpiration to NPP and VPD is conservative [Tanner and Sinclair, 1983; Monteith, 1988]. Figure 10 shows that the daily transpiration was linearly correlated with the product of NPP and VPD at two levels of leaf N concentration. Linear regressions resulted in correlation coefficients >0.97. The slopes were influenced by leaf N. The slope decreased exponentially with increasing leaf N; at higher range of leaf N, the leaf N had very little effect on the slope (Figure 11). The slope changed little with LAI (Figure 12); the value at LAI = 1 was within 20% of the value at LAI = 7. The regional pattern of estimated annual transpiration for 1991 using the Tr model was parallel to that of the NPP (Figure 13), increasing from ∼16 cm in the north west boundary to 136 cm H2O in the south east. This is comparable to the range and distribution of annual rainfall over this region [Burke et al., 1991].

image

Figure 10. Daily total transpiration against the product of daily carbon assimilation and VPD at two levels of leaf N. Symbols are the GEMTM simulations, and lines are linear regression curves.

Download figure to PowerPoint

image

Figure 11. Effects of leaf N concentration on the ratio (s) of transpiration to the product of NPP and VPD. Symbols are the GEMTM simulations, and the curve is the fitting of equation (9).

Download figure to PowerPoint

image

Figure 12. Effect of LAI on the ratio (s) of transpiration to the product of NPP and VPD.

Download figure to PowerPoint

image

Figure 13. Simulated spatial distribution of annual transpiration of 1991 over the central grasslands region of the United States using the Tr model.

Download figure to PowerPoint

5. Discussion and Conclusions

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Model Description
  5. 3. Study Area and Data Sets
  6. 4. Results
  7. 5. Discussion and Conclusions
  8. Acknowledgments
  9. References

[35] We demonstrated that the integrated process model GEMTM successfully simulated leaf photosynthesis and stomatal behavior, diurnal variation of CO2, latent heat (transpiration plus evaporation), sensible heat, and soil surface heat fluxes. The model simulations supported the notion that net carbon assimilation was proportional to intercepted PAR, but the radiation use efficiency changed with leaf N, temperature, and atmospheric CO2 concentration. The GEMTM model was also used to show that the transpiration was linearly correlated with the product of NPP and VPD [Monteith, 1988; Tanner and Sinclair, 1983], and the slope varied with leaf N.

[36] We identified and quantified the changes of parameters (RUE and s) of the simple parametric/empirical models for NPP and transpiration using the GEMTM. RUE increased with leaf N asymptotically, and responded to temperature in an asymmetric bell shape pattern with a 22°C and 26°C optimal temperature under current ambient and doubled CO2 concentration, respectively. The slope (s) of transpiration against the product of NPP and VPD decreased exponentially with leaf N.

[37] The NPP model reasonably simulated the seasonal and interannual variation of cumulative NPP data. Simulated regional distribution of NPP over the central grasslands was consistent with estimates obtained using other models. NPP increased from 120 gC/m2/year at the northwest to 956 gC/m2/year at the southeast part. The simulated regional transpiration had a similar spatial distribution pattern as the NPP, ranging from about 16 cmH2O/year to 136 cmH2O/year.

[38] The NPP and transpiration models indicated that the estimates of NPP and transpiration are sensitive to leaf N content and its spatial and temporal distribution. While information about leaf N content is not available or not easily obtained over large scales, leaf N content information might be able to be derived from remotely sensed data in the future [Wessman et al., 1990; Waring et al., 1987]. Leaf N might also be a predicted outcome of coupling a soil organic matter simulation model with NPP and transpiration models [Potter et al., 1993]. The CASA model [Potter et al., 1993; Field et al., 1995] links the Monteith NPP model with a soil organic model which simulates soil C and N cycling and soil water budget; hence the CASA model explicitly takes into account the interactions between NPP and soil resource availability, but there is no feedback of soil N on light use efficiency. Evapotranspiration in the CASA model is totally decoupled from biomass production. The transpiration model introduced in this study provides an alternative which couples transpiration and NPP explicitly. More complete analysis and validation are required, however. The NPP model here does not explicitly consider short-term water stress effects; hence the model might overestimate NPP, especially in dry years. Recently, Prince and Goward [1995] proposed to use remotely sensed data to directly derive soil moisture information to account for water stress effects. These results suggested that detailed process models can be used to identify and quantify parameter values in simple parametric models, such as the NPP and Tr models developed here, and that leaf level photosynthesis and transpiration processes can be scaled up to regional areas using simple parametric models and remotely sensed data.

[39] The primary advantage of IPAR based NPP models is the ability to use satellite data to capture spatial and temporal details at large scales. Hence it is a very powerful tool for monitoring ecosystem function over large areas. The predictive capabilities of these models are limited in that responses to climate change and rising CO2 concentration cannot be predicted.

Acknowledgments

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Model Description
  5. 3. Study Area and Data Sets
  6. 4. Results
  7. 5. Discussion and Conclusions
  8. Acknowledgments
  9. References

[40] This work was partly supported by the Joint Program for Terrestrial Ecology and Global Change NSF DEB 9524129.

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Model Description
  5. 3. Study Area and Data Sets
  6. 4. Results
  7. 5. Discussion and Conclusions
  8. Acknowledgments
  9. References
  • Ailts, B., D. Akkerman, B. Quirk, and D. Steinwand (1990), LAS 5.0—An image processing system for research and production environments, paper presented at American Society of Photogrammetry and Remote Sensing–American Congress of Surveying and Mapping Annual Convention, Am. Soc. of Photogramm., Denver, Colo.
  • Amthor, J. S. (1989), Respiration and Crop Productivity, pp. 69104, Springer-Verlag, New York.
  • Asrar, G., M. Fuchs, E. T. Kanemasu, and J. L. Hatfield (1984), Estimating absorbed photosynthetically active radiation and leaf area index from spectral reflectance in wheat, Agron. J., 76, 300306.
  • Asrar, G., E. T. Kanemasu, R. D. Jackson, and P. J. Pinter (1985), Estimation of total above ground phytomass production using remotely sensed data, Remote Sens. Environ., 17, 211220.
  • Ball, J. T., I. E. Woodrow, and J. A. Berry (1987), A model predicting stomatal conductance and its contribution to the control of photosynthesis under different environmental conditions, in Progress in Photosynthesis Research, edited by J. Biggens, pp. 221224, Martinus Nijhoff, Zoetermeer, Netherlands.
  • Burke, I. C., T. G. F. Kittel, W. K. Lauenroth, P. Snook, C. M. Yonker, and W. J. Parton (1991), Regional analysis of the central great plains: Sensitivity to climate variability, BioScience, 41, 685692.
  • Chen, D.-X., and M. B. Coughenour (1994), GEMTM: A general model for energy and mass transfer of land surfaces and its application at the FIFE sites, Agric. For. Meteorol., 68, 145171.
  • Chen, D.-X., M. B. Coughenour, A. K. Knapp, and C. E. Owensby (1994), Mathematical simulation of C4 grass photosynthesis in ambient and elevated CO2, Ecol. Modell., 73, 6380.
  • Chen, D.-X., H. W. Hunt, and J. A. Morgan (1996), Responses of a C3 and C4 perennial grass to CO2 enrichment and climate change: Comparison between model predictions and experimental data, Ecol. Modell., 87, 1127.
  • Choudhury, B. J. (1994), Synergism of multispectral satellite observations for estimating regional land surface evaporation, Remote Sens. Environ., 49, 264274.
  • Collatz, G. J., J. T. Ball, C. Grivet, and J. A. Berry (1991), Physiological and environmental regulation of stomatal conductance, photosynthesis and transpiration: A model that includes a laminar boundary layer, Agric. For. Meteorol., 54, 107136.
  • Coughenour, M. B. (1984), A mechanistic simulation analysis of water use, leaf angles, and grazing in east African graminoids, Ecol. Modell., 26, 203230.
  • Coughenour, M. B., and D.-X. Chen (1997), An assessment of grassland ecosystem responses to atmospheric change using linked ecophysiological and soil process process models, Ecol. Appl., 7, 802827.
  • Cowan, I. R. (1977), Stomatal behaviour and environment, Adv. Bot. Res., 4, 117228.
  • Cowan, I. R. (1982), Regulations and water use in relation to carbon gain in higher plants, in Encyclopedia of Plant Ecophysiology, New Ser., vol. 12, B, edited by L. Lange, P. S. Nobel, and C. B. Osmond, pp. 589613, Springer-Verlag, New York.
  • Daughtry, C. S. T., K. P. Gallo, S. N. Goward, S. D. Prince, and W. P. Kustas (1992), Spectral estimates of absorbed radiation and phytomass production in corn and soybean canopies, Remote Sens. Environ., 39, 141152.
  • de Wit, C. T. (1958), Transpiration and Crop Yields, 88 pp., Pudoc, Wageningen, Netherlands.
  • Farquhar, G. D., S. von Caemmerer, and J. A. Berry (1980), A biochemical model of photosynthetic CO2 assimilation in leaves of C3 species, Planta, 149, 7890.
  • Field, C. B. (1983), Allocating leaf nitrogen for the maximization of carbon gain: Leaf age as a control on the allocation program, Oecologia, 56, 341347.
  • Field, C. B. (1991), Ecological scaling of carbon gain to stress and resource availability, in Response of Plants to Multiple Stresses, edited by H. Mooney, pp. 3565, Academic, San Diego, Calif.
  • Field, C. S., J. T. Randerson, and C. M. Malmstrom (1995), Global net primary production: Combining ecology and remote sensing, Remote Sens. Environ., 51, 7488.
  • Goetz, S. J., and S. D. Prince (1996), Remote sensing of net primary production in boreal forest stands, Agric. For. Meteorol., 78, 149179.
  • Goudriaan, J. (1977), Crop Micrometeorology: A Simulation Study, Pudoc, Wageningen, Netherlands.
  • Goudriaan, J. (1989), Simulation of micrometeorology of crops, some methods and their problems, and a few results, Agric. For. Meteorol., 47, 239258.
  • Goward, S. N., C. J. Tucker, and D. G. Dye (1985), North American vegetation patterns observed with the NOAA-7 advanced very high resolution radiometer, Vegetatio, 64, 314.
  • Hari, P., T. Nilson, R. Salminen, L. Kaipiainen, E. Korpilahti, and J. Ross (1984), Nonlinear dependence of photosynthetic rate on irradiance and its consequences for estimates of the amount of saccharide formed, Photosynthetica, 18, 2833.
  • Harley, P. C., R. B. Thomas, J. F. Reynolds, and B. R. Strain (1992), Modelling photosynthesis of cotton grown in elevated CO2, Plant Cell Environ., 15, 271282.
  • Hirose, T., and M. T. A. Werger (1987), Maximizing daily canopy photosynthesis with respect to the leaf nitrogen allocation pattern in the canopy, Oecologia, 72, 520526.
  • Holben, B. N. (1986), Characteristics of maximum-value composite images from temporal AVHRR data, Int. J. Remote Sens., 7(11), 1417.
  • Horie, T., and T. Sakuratani (1985), Studies on crop-weather relationship model in rice: 1. Relation between absorbed solar radiation by the crop and the dry matter production, Jpn. J. Agric. Meteorol., 40, 331342.
  • Hunt, H. W., M. T. Trlica, E. F. Redente, J. C. Moore, J. K. Delting, T. G. F. Kittel, D. E. Walter, M. C. Fowler, D. A. Klein, and E. T. Elliott (1991), Simulation model for the effects of climate change on temperate grassland ecosystems, Ecol. Modell., 53, 205246.
  • Jackson, R. D., R. J. Reginato, and S. B. Idso (1977), Wheat canopy temperatures: A practical tool for evaluating water requirements, Water Resour. Res., 13, 651656.
  • Jackson, R. D., J. L. Hatfield, R. J. Reginato, S. B. Idso, and P. J. Pinter Jr. (1983), Estimation of daily evapotranspiration from one time-of-day measurements, Agric. Water Manage., 7, 351362.
  • Jackson, R. D., M. S. Moran, L. W. Gay, and L. H. Raymond (1987), Evaluating evaporation from field crops using airborne radiometry and ground-based meteorological data, Irrig. Sci., 8, 8190.
  • Jagtap, S. S., and J. W. Jones (1989), Evapotranspiration model for developing crops, Trans. ASAE, 32, 13421350.
  • Jarvis, P. G. (1976), The interpretation of the variations in leaf water potential and stomatal conductance found in canopies in the field, Philos. Trans. R. Soc. London, Ser. B, 273, 593610.
  • Jarvis, P. G. (1995), Scaling processes and problems, Plant Cell Environ., 18, 10791089.
  • Jarvis, P. G., and J. W. Leverenz (1983), Productivity of temperate, deciduous and evergreen forests, in Encyclopedia of Plant Physiology, New Ser., vol. 12, D, edited by O. L. Lange et al., pp. 233280, Springer-Verlag, New York.
  • Jarvis, P. G., and K. G. McNaughton (1986), Stomatal control of transpiration: Scaling up from leaf to region, Adv. Ecol. Res., 15, 149.
  • Kim, J., and S. B. Verma (1990), Carbon dioxide exchange in a temperate grassland ecosystem, Boundary Layer Meteorol., 52, 135149.
  • Kittel, T. G. F., and M. B. Coughenour (1988), Prediction of regional and local ecological change from global climate model results: A hierarchical modeling approach, in Monitoring Climate for the Effects of Increasing Greenhouse Gas Concentrations, edited by R. A. Pielke, and T. G. F. Kittel, pp. 173193, Colo. State Univ., Fort Collins.
  • Kittel, T. G. F., N. A. Rosenbloom, T. H. Painter, D. S. Schimel, and VEMAP Modeling Participants (1995), The VEMAP integrated database for modeling United States ecosystem/vegetation sensitivity to climate change, J. Biogeogr., 22, 857862.
  • Knapp, A. K., E. P. Hamerlynk, and C. E. Owensby (1993), Photosynthesis and water relations responses to elevated CO2 in the C4 grass Andropogoz gerardii, Int. J. Plant Sci., 154, 459466.
  • Kumar, M., and J. L. Monteith (1981), Remote sensing of crop growth, in Plants and the Daylight Spectrum, edited by H. Smith, pp. 133144, Academic, San Diego, Calif.
  • Landsberg, J. J., and F. J. Hingston (1996), Evaluating a simple radiation/dry matter conversion model using data from Eucalyptus globulus plantations in Western Australia, Tree Physiol., 16, 801808.
  • Lauenroth, W. K. (1979), Grassland primary production: North American grasslands in perspective, in Perspectives in Grassland Ecology, Ecol. Stud., vol. 32, edited by N. R. French, pp. 324, Springer-Verlag, New York.
  • Leuning, R. (1995), A critical appraisal of a combined stomatal-photosynthesis model for C3 plants, Plant Cell Environ., 6, 181194.
  • Long, S. P. (1991), Modification of the response of photosynthetic productivity to rising temperature by atmospheric CO2 concentrations: Has its importance been underestimated? Plant Cell Environ., 14, 729739.
  • Marshall, J. K. (1977), Biomass and production partitioning in response to environment in some America grasslands, in The Belowground Ecosystem: A Synthesis of Plant-Associated Processes, Range Sci. Dep. Sci. Ser., vol. 26, edited by J. K. Marshall, pp. 123132, Colo. State Univ., Fort Collins.
  • McCree, K. J. (1981), Photosynthetically active radiation, in Physiological Plant Ecology, I, Responses to the Physical Environment, vol. 12, A, edited by L. Lange et al., p. 4155, Springer-Verlag, New York.
  • McNaughton, K. G. (1994), Effective stomatal and boundary-layer resistances of heterogeneous surfaces, Plant Cell Environ., 17, 10611068.
  • McNaughton, S. J., M. Oesterheld, D. A. Frank, and K. J. Williams (1989), Ecosystem-level patterns of primary productivity and herbivory in terrestrial habitats, Nature, 341, 142144.
  • Monteith, J. L. (1965), Evaporation and environment, in The State and Movement of Water in Living Organisms, Symp. Soc. Exper. Biol., vol. 19, pp. 205234, Cambridge Univ. Press, New York.
  • Monteith, J. L. (1972), Solar radiation and productivity in a tropical ecosystem, J. Appl. Ecol., 9, 747766.
  • Monteith, J. L. (1977), Climate and the efficiency of crop production in Britain, Philos. Trans. R. Soc. London, Ser. B, 281, 277294.
  • Monteith, J. L. (1988), Does transpiration limit the growth of vegetation or vice-versa? J. Hydrol., 100, 5768.
  • Monteith, J. L. (1990), Conservative behaviour in the response of crops to water and light, in Theoretical Production Ecology: Reflections and Prospects, edited by R. Rabbinge et al., pp. 116, Pudoc, Wageningen, Netherlands.
  • Monteith, J. L. (1995), A reinterpretation of stomatal responses to humidity, Plant Cell Environ., 18, 365372.
  • Mooney, H. A. (1991), Biological response to climate change: An agenda for research, Ecol. Appl., 1, 112117.
  • Nikolov, N. T., W. J. Massman, and A. W. Schoettle (1995), Coupling biochemical and biophysical processes at the leaf level: An equilibrium photosynthesis model for leaves of C3 plants, Ecol. Modell., 80, 205235.
  • Norman, J. M. (1979), Modeling the complete crop canopy, in Modification of the Aerial Environment of Crops, edited by B. J. Barfield, and J. F. Gerber, pp. 249277, Am. Soc. of Agric. Eng., St. Joseph, Mich.
  • Norman, J. M. (1980), Interfacing leaf and canopy light interception models, in Predicting Photosynthesis for Ecosystem Models, edited by J. D. Hesketh, and J. W. Jones, pp. 4968, CRC, Boca Raton, Fla.
  • Norman, J. M., R. Garcia, and S. B. Verma (1992), Soil surface CO2 fluxes and carbon budget of a grassland, J. Geophys. Res., 97(D17), 18,84518,853.
  • Ojima, D. S., T. G. F. Kittel, T. Rosswall, and B. H. Walker (1991), Critical issues for understanding global change effects on terrestrial ecosystems, Ecol. Appl., 1, 316325.
  • Paruelo, J. M., and W. K. Lauenroth (1995), Regional patterns of normalized difference vegetation index in North American shrublands and grasslands, Ecology, 76, 18881898.
  • Pearcy, R. W., and J. Ehleringer (1984), Comparative ecophysiology of C3 and C4 plants, Plant Cell Environ., 7, 113.
  • Potter, C. S., J. T. Randerson, C. B. Field, P. A. Matson, P. M. Vitousek, H. A. Mooney, and S. A. Kleoster (1993), Terrestrial ecosystem production: A process model based on global satellite and surface data, Global Biogeochem. Cycles, 7, 811841.
  • Prince, S. D. (1991), A model of regional primary production for use with coarse resolution satellite data, Int. J. Remote Sens., 12, 13131330.
  • Prince, S. D., and S. N. Goward (1995), Global primary production: A remote sensing approach, J. Biogeogr., 22, 815835.
  • Prince, S. D., and C. J. Tucker (1986), Satellite remote sensing of rangelands in Botswana: II. NOAA AVHRR and herbaceous vegetation, Int. J. Remote Sens., 7, 15551570.
  • Raich, J. W., E. B. Rastetter, J. M. Melillo, D. W. Kicklighter, P. A. Steudler, and B. J. Peterson (1991), Potential net primary productivity in South America: Application of a global model, Ecol. Appl., 1, 399429.
  • Raupach, M. R., and J. J. Finnigan (1988), Single-layer models of evaporation from plant canopies are incorrect but useful, whereas, multilayer models are correct but useless, Discuss. Aust. J. Plant Physiol., 15, 705716.
  • Reynolds, J. F., and B. Acock (1985), Predicting the response of plants to increasing carbon dioxide: A critique of plant growth models, Ecol. Modell., 29, 107129.
  • Ruimy, A., B. Saugier, and G. Dedieu (1994), Methodology for the estimation of terrestrial net primary production from remotely sensed data, J. Geophys. Res., 99(D3), 52635283.
  • Sala, O. E., W. J. Parton, L. A. Joyce, and W. K. Lauenroth (1988), Primary production of the central grassland region of the United States, Ecology, 69, 4045.
  • SAS Institute Inc. (1987), SAS/STAT™ Guide for Personal Computers, version 6 ed., Cary, N. C.
  • Schimel, D. S., T. G. F. Kittel, A. K. Knapp, T. R. Seastedt, W. J. Parton, and V. B. Brown (1991), Physiological interactions along resource gradients in a tallgrass prairie, Ecology, 72, 672684.
  • Schimel, D. S., W. A. Weis, and T. G. F. Kittel (1993), Spatial information for extrapolation of canopy processes: Examples from FIFE, in Scaling Physiological Processes: Leaf to Globe, edited by J. R. Ehleringer, and C. B. Field, pp. 2136, Academic, San Diego, Calif.
  • Schultze, E.-D., F. M. Kelliher, C. Korner, J. Lloyd, and R. Leuning (1994), Relationships among maximum stomatal conductance, ecosystem surface conductance, carbon assimilation rate, and plant nitrogen: A global ecology scaling exercise, Annu. Rev. Ecol. Syst., 25, 629660.
  • Sellers, P. J. (1985), Canopy reflectance, photosynthesis and transpiration, Int. J. Remote Sens., 6, 13351372.
  • Sellers, P. J. (1987), Canopy reflectance, photosynthesis and transpiration: II. The role of biophysics in the linearity of their interdependence, Remote Sens. Environ., 21, 143183.
  • Sellers, P. J., Y. Mintz, Y. C. Sud, and A. Dalcher (1986), A simple biosphere model (SiB) for use within general circulation models, J. Atmos. Sci., 43, 505531.
  • Sinclair, T. R., and T. Horie (1989), Leaf nitrogen, photosynthesis, and crop radiation use efficiency: A review, Crop Sci., 29, 9098.
  • Strain, B. R., and F. A. Bazzaz (1983), Terrestrial plant communities, in CO2and Plants: The Response of Plants to Rising Levels of Atmospheric Carbon Dioxide, AAAS Selected Symp., vol. 84, edited by E. R. Lemon, pp. 177222, Westview, Boulder, Colo.
  • Tanner, C. B., and T. R. Sinclair (1983), Efficient water use in crop production: Research or re-search, in Limitations to Efficient Water Use in Crop Production, edited by J. H. M. Taylor, W. R. Jordan, and T. R. Sinclair, pp. 127, Am. Soc. of Agron., Madison, Wis.
  • Tucker, C. J., C. L. Vanpraet, E. Boerwinkel, and A. Gaston (1983), Satellite remote sensing of total dry accumulation on the Senegalese Sahel, Remote Sens. Environ., 13, 461474.
  • Waring, R. H., J. D. Aber, J. M. Melillo, and B. Moore (1987), Precursors of change in terrestrial ecosystems, BioScience, 36, 433438.
  • Wessman, C. A., B. Curtiss, and S. L. Ustin (1990), Large scale ecosystem modeling using parameters derived from imaging spectrometer data, in SPIE Technical Symposium on Optics, Electro-Optics, and Sensors, vol. 1298, pp. 164170, SPIE—Int. Soc. for Opt. Eng., Bellingham, Wash.
  • Woodward, F. I., and T. M. Smith (1994), Global photosynthesis and stomatal conductance: Modelling the controls by soil and climate, Adv. Bot. Res., 20, 241.