MODELING Modeling Interactions among Carbon Dioxide, Nitrogen, and Climate on Energy Exchange of Wheat in a Free Air Carbon Dioxide Experiment Robert F. Grant,* Bruce A. Kimball, Talbot J. Brooks, Gary W. Wall, Paul J. Pinter, Jr., Doug J. Hunsaker, Floyd J. Adamsen, Robert L. Lamorte, Steven W. Leavitt, Thomas L. Thompson, and Allan D. Matthias ABSTRACT
upon site conditions. The direction and extent of these changes are important in estimating whether ecosystems will become more or less water-limited when both Ca and air temperatures are rising as hypothesized in current climate change scenarios. Mathematical models are being used to simulate changes in mass and energy exchange under changing climate for site-specific conditions. Earlier models relied on simple multiplication factors for radiation, water and N use efficiencies to simulate the effects of Ca on plant CO2 fixation, and transpiration (e.g., Parton et al., 1995; Rosenzweig and Parry, 1994; Stockle et al., 1992). The use of these factors requires the assumption that each is unique and independent of site conditions, whereas the diversity of results from field studies (Lawlor and Mitchell, 1991) suggests that these factors are not likely to be unique. In fact, these multiplication factors depend in complex ways on climate and soil conditions. Future simulations of changes in CO2 fixation and transpiration under rising Ca could more reliably account for sitespecific conditions if mathematical models were based on a more fundamental treatment of the physical and biological processes known to be affected by Ca. A Ca ⫻ N interaction on mass and energy exchange is important because CO2 fixation and transpiration are limited by N in most terrestrial ecosystems under ambient Ca, and are likely to be more limited under elevated Ca. These greater limitations may be caused by:
Changes in mass and energy exchange by crops under rising atmospheric CO2 concentration (Ca ) may be affected by N and weather; Ca interacts with weather on mass and energy exchange through limitations on latent heat flux imposed by stomatal conductance, which is affected by Ca, and aerodynamic conductance, which is affected by weather. We examined the bases for these interactions with the ecosystem model ecosys. Simulation results were tested with energy flux data from a Free Air CO2 Enrichment (FACE) experiment in which wheat (Triticum aestivum L.) was grown under 548 vs. 363 mol mol⫺1 Ca and fertilized with 7 vs. 35 g N m⫺2. Both model and experimental results indicated that raising Ca from 363 to 548 mol mol⫺1 reduced midday latent heat fluxes by ca. 50 W m⫺2 for wheat fertilized with 35 g N m⫺2, and by ca. 100 W m⫺2 for wheat fertilized with only 7 g N m⫺2 when N deficits developed later in the growing season. These reductions were smaller under low wind speeds (⬍5 km h⫺1 ) and stable boundary conditions when aerodynamic conductance became the dominant constraint to transpiration. At a seasonal time scale, raising Ca from 363 to 548 mol mol⫺1 reduced simulated (measured) evapotranspiration of wheat by 9% (7%) when fertilized with 35 g N m⫺2, and by 16% (19%) with 7 g N m⫺2. Changes with Ca in mass and energy exchange used in climate change studies should therefore reflect the site-specific availability of N, as well as climate attributes such as wind speed.
he exchange of mass and energy between terrestrial ecosystems and the atmosphere strongly affects atmospheric thermodynamics and hence climate (Chahine, 1992), as well as terrestrial hydrology. Rising atmospheric CO2 concentrations (Ca ) increase CO2 fluxes into, and reduce latent heat fluxes out of, terrestrial vegetation. The extent to which these fluxes change with Ca depends on environmental conditions such as irradiance and air temperature (Cure and Acock, 1986; Idso et al., 1987), and on soil conditions such as water availability (Gifford, 1979; Kimball et al., 1995; Rogers et al., 1986), and nutrients (Hocking and Meyer, 1991; McKee and Woodward, 1994; Rogers et al., 1996; Wolf, 1996; Wong, 1979). Therefore changes in mass and energy exchange under rising Ca are highly dependent
1. A greater increase in CO2 fixation vs. C consumption under rising Ca that causes primary products of C fixation to accumulate in leaves (Wong, 1990). The consequent sequestration of inorganic phosphate in these products may reduce CO2 fixation (Azcon-Bieto, 1983; Herold, 1980) and hence stomatal conductance and transpiration, although this reduction may be a short-term phenomenon pending longer-term adaptations (Stitt, 1991). 2. A relatively smaller increase in N uptake than in CO2 fixation under elevated Ca that reduces distribution of N to leaves compared with other plant organs such as roots (Makino et al., 1997; Rogers et al., 1993), thereby rebalancing CO2 fixation and N uptake. Leaf N concentrations are thereby reduced, constraining the increase in CO2 fixation with Ca (Baxter et al., 1997; Wong, 1979).
R.F. Grant, Dep. of Renewable Resources, Univ. of Alberta, Edmonton, AB, Canada T6G 2H1; B.A. Kimball, T.J. Brooks, G.W. Wall, P.J. Pinter, Jr., D.J. Hunsaker, F.J. Adamsen, and R.L. Lamorte, USDA-ARS, U.S. Water Conserv. Lab., 4331 E. Broadway, Phoenix, AZ 85040; S.W. Leavitt, Laboratory of Tree Ring Res., Univ. of Arizona, Tucson, AZ; and T.L. Thompson and A.D. Matthias, Dep. of Soil, Water, and Environ. Sci., Univ. of Arizona, Tucson, AZ. Received 13 June 2000. *Corresponding author ([email protected]
These processes may explain why rates of CO2 fixation and plant growth increase less with Ca under low vs. high N availability (Owensby et al., 1994; Rogers et al., 1993; Wolf, 1996). If CO2 concentration ratios across
Published in Agron. J. 93:638–649 (2001).
GRANT ET AL.: MODELING CO2, N, AND CLIMATE EFFECTS ON ENERGY EXCHANGE
plant stomates are conserved at different Ca and N (Wong et al., 1979), then stomatal conductance should be reduced more by higher Ca under low vs. high N levels. Greater reductions in stomatal conductance should cause greater reductions in transpiration, as has been observed in glasshouse experiments (Wong, 1979). These greater reductions in transpiration should be manifested in field experiments as greater reductions in latent heat fluxes and greater increases in sensible heat fluxes and canopy temperatures when Ca is raised under low vs. high N. These reductions in latent heat fluxes under higher Ca should also be greater when stomatal conductance ⬎⬎ aerodynamic conductance, because the former is directly affected by Ca, whereas the latter is not. Such effects have been postulated by Jarvis and McNaughton (1986) and modeled by Carlson and Bunce (1996). Modeling Ca ⫻ N ⫻ weather interactions on mass and energy exchange should be based on modeling the processes described above by which plants adapt their CO2 fixation and N partitioning under rising Ca. It is important that results of this modeling be subjected to well-constrained tests before the use of models in predictive studies of climate change effects. Generally speaking, model tests become better constrained as their spatial and temporal resolutions increase. The use of mass and energy fluxes in model testing is well constrained in that test data are highly resolved temporally and flux theory is comparatively well defined. Measured fluxes could therefore provide a well-constrained test for hypotheses of Ca ⫻ N ⫻ climate interactions on mass and energy exchange. The simulation of the processes by which N limits increases in CO2 fixation under elevated Ca would therefore be corroborated by the accurate simulation of a Ca ⫻ N interaction on mass and energy exchange. Simulating the processes by which wind speed, temperature, and vegetation affect aerodynamic conductance would be corroborated by the accurate simulation of a Ca ⫻ weather interaction on mass and energy exchange. In earlier work (Grant et al., 1995a, 1995b, 1999b) we simulated a Ca ⫻ water interaction on mass and energy exchange of wheat with the detailed ecosystem model ecosys (Grant, 1996), supported by findings from a Free Air CO2 Enrichment (FACE) experiment. Those simulations were conducted with high levels of N fertilization. In this study we extend our earlier work to include a simulation with the same model of a Ca ⫻ N interaction on mass and energy exchange of wheat under high irrigation. We describe hypotheses for simulating Ca effects on N partitioning and hence on CO2 fixation, and for simulating CO2 fixation effects on stomatal conductance and energy exchange. We also describe hypotheses for simulating wind speed effects on aerodynamic conductance and hence on energy exchange. We then test these simulations with data for mass and energy exchange, crop growth, and water use reported from the 1995–1996 FACE experiment at Phoenix, AZ (Kimball et al., 1999). Our objective is to develop a predictive capability for Ca effects on mass and energy exchange over crops as affected by N and weather.
MODEL DEVELOPMENT Energy Exchange Energy exchanges between the atmosphere and terrestrial surfaces are resolved in ecosys into that between the atmosphere and the leaf and stem surfaces of each population (i.e., species or cohort) within the plant community, and that between the atmosphere and each of the surfaces (soil, plant residue, snow) of the ground beneath (Grant et al., 1999b). Total energy exchange between the atmosphere and terrestrial surfaces is calculated as the sum of exchanges with all plant and ground surfaces. Canopy energy exchange in ecosys is calculated from an hourly two-stage convergence solution for the transfer of water and heat through a multilayered multipopulation soil– root–canopy system. The first stage of this solution requires convergence to a value of canopy temperature Ti for each plant population i at which the first-order closure of the canopy energy balance (net radiation, sensible heat flux, latent heat flux, and change in heat storage) is achieved. Net radiation is calculated from the absorption, reflection, and transmission of shortwave and longwave radiation by each leaf and stem surface defined by height, azimuth, and inclination through a multilayered canopy. Nonuniformity in horizontal distribution of leaf surfaces is modeled with a clumping factor for leaf surface exposure to incident radiation in each canopy layer (given a value of close to one for cereals, i.e., mininal nonuniformity). Sensible heat fluxes are calculated from air temperature differences between the atmosphere and the aerodynamic height of the plant canopy. Latent heat fluxes are calculated from vapor pressure differences between the atmosphere and either free water on plant surfaces if present, or water within plant leaves if not. Fluxes between the atmosphere and each plant population are controlled by aerodynamic (gAi ) and stomatal (gCi ) conductances. Aerodynamic conductance is calculated from zero plane displacement and surface roughness heights derived from canopy height and leaf area (Perrier, 1982) calculated in the model, and corrected for nonisothermal effects from the bulk Richardson number according to Van Bavel and Hillel (1976). This correction reproduces the sensitivity of aerodynamic conductance to the bulk Richardson number calculated by Ottoni et al. (1992) from the aerodynamic conductance formulation of Choudhury et al. (1986). Two controlling mechanisms are postulated for gCi: 1. At the leaf level, maximum leaf conductance gLi is that which allows an assumed constant Ci/Ca ratio of 0.725 for C3 plants to be maintained at carboxylation rates calculated under ambient irradiance, temperature, Ca, and nonlimiting water potential. The assumption of constant Ci/Ca is implicit in other stomatal models in which gL varies directly with CO2 assimilation and inversely with Ca (e.g., Ball, 1988). Carboxylation rates are calculated as the lesser of dark and light reaction rates according to Farquhar et al. (1980). Maximum dark or light reaction rates are assumed to be driven by the specific activities and areal concentrations of rubisco or chlorophyll, respectively, set maximum values of which are given in Table 1 of Grant and Nalder (2000). These activities and concentrations may be reduced from maximum values by environmental conditions (radiation, temperature, Ca, water, N, P) as described under “Coupling of C and Nutrient Transfers” below. Maximum gLi is then aggregated by surface area to the canopy level for use in the energy balance convergence (Grant et al., 1999b). 2. At the canopy level, gCi is then reduced from that at
AGRONOMY JOURNAL, VOL. 93, MAY–JUNE 2001
nonlimiting water potential through an exponential function of canopy turgor potential Ti (Grant et al., 1999b) determined from total Ci and osmotic i water potentials (Grant et al., 1999b). The calculation of canopy water potentials is described under “Water Relations” below. The exponential function of Ti used here is based on that proposed by Zur and Jones (1981) to account for the effects of osmotic adjustment on stomatal conductance.
Water Relations After convergence for Ti is achieved, the difference between canopy transpiration Ei from the energy balance and total water uptake Ui from all rooted layers in the soil is tested against the difference between canopy water content from the previous hour and that from the current hour. This difference is minimized by adjusting Ci, which determines each term from which this difference is calculated. The value of Ci determines that of Ti, and hence gCi from which Ei is calculated, through its effect on i. The difference between Ci and soil water potential Sl determines Ui by establishing potential differences across soil–root–canopy hydraulic resistances in each rooted soil layer l (Grant et al., 1999b). Hydraulic resistances require values for root length and surface area provided by a root system submodel driven by shoot–root C transfers and root C oxidation (Grant, 1998). The value of Ci determines that of canopy water content according to plant water potential–water content relationships (e.g., Saliendra and Meizner, 1991). Because Ti drives Ei, the canopy energy balance described under “Energy Exchange” above is recalculated for each adjusted value of Ci during convergence.
Carbon Dioxide Fixation After successful convergence solutions for Ti and Ci, leaf carboxylation rates are adjusted from those calculated under nonlimiting Ci to those under ambient Ci. This adjustment is required by the decrease in gCi from its maximum value (calculated under “Energy Exchange” above) to that at ambient Ti (calculated under “Water Relations” above). The adjustment is achieved through a convergence solution for Ci and its aqueous equivalent at which the diffusion rate of gaseous CO2 between Ca and Ci through gLi equals the carboxylation rate of aqueous CO2 at the equivalent of Ci (Grant et al., 1999b). The CO2 fixation rate of each leaf surface at convergence is added to arrive at a value for gross canopy CO2 fixation by each tiller (or branch) of each plant population (i.e., species or cohort) in the model.
Carbon Respiration The product of CO2 fixation in each branch (dicots) or tiller (monocots) of each plant population described under “Carbon Dioxide Fixation” above is added to a C storage (nonstructural) pool from which C is oxidized to meet requirements for maintenance and growth respiration (Grant et al., 1999b). Low C storage may cause oxidation of stored C to be less than maintenance respiration requirements, in which case the shortfall is made up through respiration of remobilizable protein C (set in the model to 0.5 of total protein C) withdrawn from lamina and petioles (dicots) or sheath (monocots) C in each tiller. Withdrawal starts at the lowest node at which laminae and sheaths are present, and proceeds upward. Protein N and P withdrawn with protein C are added to storage (nonstructural) N and P pools. Upon exhaustion of the remobilizable protein C in each lamina or sheath, the remaining protein and nonprotein C is dropped from the tiller and added to the soil surface as litter. Environmental constraints such as nutrient,
heat or water stress that reduce the fixation and hence oxidation of stored C with respect to maintenance requirements will therefore hasten the loss of lamina and sheath C from the plant. Net canopy CO2 fixation is calculated as the difference between aggregated leaf carboxylation rates and the sum of canopy maintenance and growth respiration.
Nutrient Uptake and Translocation Between germination and emergence, all C, N, and P storage pools in the roots and tillers are supplied heterotrophically in the model from seed reserves. After emergence all storage C is supplied autotrophically from CO2 fixation by branches as described above, and all storage N and P is supplied from active uptake by roots and mycorrhizae in each rooted soil layer (Grant, 1998; Grant and Robertson, 1997). Uptake of N and P is calculated by iteratively converging toward solution [NH4⫹], [NO3⫺], and [H2PO4⫺] at root and mycorrhizal surfaces at which radial transport by mass flow and diffusion from the soil solution to these surfaces equals active uptake by the surfaces (Grant, 1998). Solution N and P concentrations are controlled by precipitation, adsorption and ion pairing reactions (Grant and Heaney, 1997), vertical and horizontal solute transport (Grant, 1991; Grant and Heaney, 1997) and microbial activity including mineralization (Grant et al., 1993a, 1993b), nitrification (Grant, 1994b, 1995), denitrification (Grant et al., 1993c, 1993d), volatilization and N2 fixation. Radial transport and active uptake require values for root length and surface area provided by a root growth model (Grant, 1998). Shoot, root, and mycorrhizal growth in ecosys depends on the transfer of storage C, N, and P among them (Eq. – in Grant, 1998). These transfers are driven by concentration differences in storage C, N, and P (Brugge and Thornley, 1985) that depend on: 1. The proximity of root or mycorrhizal axes and tillers to the sites of C, N, and P acquisition. Thus, storage C concentration gradients develop such that shoot ⬎ root ⬎ mycorrhizae, and storage N and P concentration gradients develop such that mycorrhizae ⬎ root ⬎ shoot. 2. The rates at which C, N, and P are consumed by root or mycorrhizal axes and tillers in relation to that by other axes and tillers. In this way storage C, N, and P migrate to those parts of the plant in which they are most rapidly used.
Coupling of Carbon and Nutrient Transfers A mechanism is required in ecosystem models to constrain CO2 fixation and hence stomatal conductance when N or P uptake is limiting. In ecosys the storage product of CO2 fixation is coupled with the storage products of N and P uptake to form storage pools in mycorrhizae and root axes and in tillers from which new organs are formed according to organ-specific C/N/P ratios. Ratios of storage C to storage N or P in branch or tiller pools greater than those at which organ growth occurs indicate excess CO2 fixation with respect to N or P uptake for shoot growth. Such ratios in the model have two effects: 1. In the short term, they reduce specific activities of rubisco and chlorophyll (i.e., deactivation), thereby simulating the suppression of CO2 fixation by leaf carbohydrate accumulation widely reported in the literature (reviewed by Stitt, 1991 and Bowes, 1991). The constraint to CO2 fixation imposed by low storage N/C and P/C ratios also reduces leaf and canopy stomatal conductances to conserve the Ci/Ca ratio. 2. Because tiller C, N, and P storage pools are the substrates for leaf growth, the N/C and P/C ratios in the tiller storage pools determine the N/C and P/C ratios at which
GRANT ET AL.: MODELING CO2, N, AND CLIMATE EFFECTS ON ENERGY EXCHANGE
leaves are formed in ecosys. Low storage ratios cause N/C and P/C ratios of current leaf growth to be reduced from set maximum values of 0.15 g N g⫺1 C and 0.015 g P g⫺1 C (Penning de Vries, 1982). This reduction simulates the modulation by nitrate of amino acid vs. carbohydrate synthesis in leaves proposed by Champigny and Foyer (1992). The reduction of leaf N/C and P/C ratios in ecosys causes a proportional reduction in areal concentrations of rubisco and chlorophyll because rubisco and chlorophyll N are assumed to be constant fractions of total leaf N. Reduced areal concentrations of rubisco and chlorophyll cause reduced leaf CO2 fixation and hence reduced leaf and canopy stomatal conductances, thereby conserving the Ci/Ca ratio. The reduction in leaf N/C and P/C ratios in ecosys also causes less storage N and P to be used for leaf growth, which reduces root– shoot concentration gradients and allows more storage N and P to remain available for growth of other organs (e.g., roots). Such changes in N distribution have been observed experimentally under N-limited conditions (Makino et al., 1997; Rogers et al., 1993). These changes allow root/shoot phytomass ratios to increase under limiting N or P, which, by improving N and P uptake, redress to some extent the storage C–N–P imbalance. The model thus implements the functional equilibrium between roots and shoots proposed by Thornley (1995).
FIELD EXPERIMENT Crop Management To reduce initial concentrations of mineral N at the experimental site, oat (Avena sativa L) plants were planted in 1994 and grown without fertilizer in a 4-ha field of Trix clay loam (fine-loamy mixed [calcareous] hyperthermic Typic Torrifluvent) at the University of Arizona’s Maricopa Research Center 30 km south of Phoenix, AZ. The oat plants were harvested several times as green forage during 1994–1995, after which the field was tilled and remaining oat plants were killed with herbicide. Spring wheat (Triticum aestivum L. ‘Yecora rojo’) was planted on 15 Dec. 1995 in 0.25-m rows at a density of 190 plants m⫺2. The field was irrigated with a subsurface drip system at a depth of about 0.20 m in 0.50-m rows parallel to those of the wheat. Irrigation fully replaced evapotranspirational demand whenever ca. 30% of plant-available water in the rooted soil zone was depleted (Table 1). The field was located within an irrigated area that extended for more than 1 km in all directions, so that most of the area surrounding the site was irrigated during the entire experiment. Hourly averages of solar radiation, air temperature, wind speed and humidity, and hourly totals of precipitation were recorded at a height of 2 m on the field site. Dry masses of different components (leaves, crowns, stems, chaff, grain) were measured every 10 d from approximately 24 plants in each replicate of each treatment. Leaf area indices were calculated from measurements of lamina mass and specific leaf area. Soil water contents were measured before and after each irrigation in all subplots using time domain reflectometry from the surface to 0.3 m and a neutron scattering device from 0.4 to 2.0 m at 0.2-m increments. Total evapotranspiration was calculated for each Ca ⫻ N treatment as Irrigation ⫹ Rainfall ⫺ Change in soil water content (0–2 m) be-
Table 1. Irrigation and fertilization schedules at the FACE site during 1995–1996. Irrigation Date
21 Dec. 1995 30 Jan. 1996 31 Jan. 1996 22 Feb. 1996 5 Mar. 1996 11 Mar. 1996 22 Mar. 1996 30 Mar. 1996 31 Mar. 1996 5 Apr. 1996 10 Apr. 1996 15 Apr. 1996 18 Apr. 1996 22 Apr. 1996 26 Apr. 1996 1 May 1996 6 May 1996
30 30 25 40 33 45 45 54 2 45 45 30 43 43 42 54 47
Low N g m⫺2
mm 30 30 25 40 33 45 45 54 2 45 45 30 43 43 42 40
tween 18 Dec. 1995 and 16 May 1996. Further details of this experiment were reported in Kimball et al. (1999).
Carbon Dioxide and Nitrogen Treatments Four replicates of control and FACE treatments (Hendrey, 1993), consisting of toroidal plenum rings constructed from 30-cm diameter pipe with 2.5 m vertical pipes located every 2.4 m around the periphery, were established in 25-m diameter circular plots shortly after seeding. From 1 Jan. 1996 (emergence) to 16 May 1996 (maturity) a computer control system used wind speed, wind direction, and CO2 concentration measured at the center of each control vs. FACE ring to direct CO2 emission from vertical pipes upwind of the plots. During emission ambient air (control) or CO2-enriched air (FACE) was blown over the plots through holes in the vertical pipes at elevations from 0.5 to 1.5 m, depending on crop height, such that average CO2 concentration maintained over the FACE plots (548 mol mol⫺1 ) was 185 mol mol⫺1 above that over the control plots (363 mol mol⫺1 ). One half of each of the FACE and control plots received 35.0 g N m⫺2 (high N) and the other half 7.0 g N m⫺2 (low N) as ammonium nitrate (Table 1) added to irrigation water. The plots received an additional 3.3 (high N) and 3.0 (low N) g N m⫺2 in the irrigation water itself.
Mass and Energy Exchange Measurements The plots were semicircular with a useable radius of only about 10 m, providing limited fetch for calculating CO2 ⫻ N effects on evapotranspiration from profiles of wind speed, temperature, and water vapor measured above the crop. Therefore, a residual energy balance method was used to calculate latent heat fluxes from the individual plots (e.g., Huband and Monteith, 1986). This technique was reasoned to be less sensitive to fetch constraints because: 1. All plots were in a field of wheat in which all structural elements were close to the same size
AGRONOMY JOURNAL, VOL. 93, MAY–JUNE 2001
and geometry. Therefore, aerodynamic resistance would not be expected to vary much among plots. 2. Turbulent transfer processes are a logarithmic function of height above the crop surface so that gradients close to the crop are largest and most important in determining rates of heat transfer. Crop surface temperatures were measured with infrared thermometers, which were not affected by wind speed, thereby minimizing fetch requirements. The parameters for canopy emittance and reflected sky radiation used in the calculation of surface temperatures for this study were found by Huband and Monteith (1986) to give unbiased differences between radiative and aerodynamic temperatures of ⱕ1.5 K and usually ⬍1 K. Under the conditions of the FACE experiment, a difference of 1 K between radiative and aerodynamic temperatures would cause a difference of as much as 60 W m⫺2 in the calculation of sensible heat flux. However, this difference would not be systematic. During most of the experiment net radiation was measured every 15 min using net radiometers (Model Q6, Radiation and Energy Balance Systems, Seattle, WA) calibrated before and after the experiment against a pyranometer (Model 15, Eppley Laboratory, Newport, RI) and another net radiometer (Model Q7, Radiation Energy Balance Systems, Seattle, WA) (Kimball et al., 1999). The net radiometers were mounted 1.0 m above the crop in two replicates of each Ca and N treatment, where they were raised, cleaned and leveled weekly. Soil heat flux was measured from four soil heat flux plates (Model HFT-3, Radiation Energy Balance Systems, Seattle, WA) placed at a depth of 10 mm in each Ca ⫻ N plot of one replicate. Sensible heat flux was calculated from canopy temperatures measured with stationary infrared thermometers (IRTs) (Model 4000a, 15⬚ field-of-view, Everest Interscience, Tustin, CA) calibrated against an extended-area black-body source (Model EABB-250, Advanced Kinetics, Huntingdon Beach, CA), and from dry and wet bulb temperatures measured with a pair of aspirated psychrometers. The IRTs were positioned 1 m above each Ca ⫻ N plot in two replicates to view the crop canopy northward at a zenith angle of 45⬚. The IRT temperatures were corrected for canopy emittance and reflected sky radiation to account for differences between radiative and aerodynamic surface temperatures (Huband and Monteith, 1986). The aspirated psychrometers were mounted at a height of 2 m in the same plots as the net radiometers. Aerodynamic resistances used in sensible heat flux calculations were computed from wind speed measured at 2 m with a cup anemometer and photochopper (Model 12102D, R.M. Young Co., Traverse City, MI), and from zero plane displacement and roughness length calculated from canopy height, using a nonisothermal stability correction. On selected days during the experiment, canopy CO2 fixation was measured on individual replicates of each Ca ⫻ N treatment using four 0.75 m by 0.75 m by 1.3 m steady-state canopy gas exchange enclosures (Brooks, 1998; Garcia et al., 1990). The enclosures were covered
with propafilm C (ICI Americas, Wilmington, DE), a material with high thermal transmittance. Temperatures of the air entering and leaving the enclosures were monitored with thermocouples and were generally within 1⬚C of ambient. Air pressure inside the enclosures was maintained slightly above atmospheric levels to minimize influx of soil CO2. Differences in the concentration of CO2 entering and leaving the enclosures were measured with infrared gas analyzers in absolute and differential modes (LI-COR Model LI-6262, Lincoln, NE). During measurement CO2 concentrations in the chambers were maintained within 10 mol mol⫺1 of FACE treatment levels.
SIMULATION EXPERIMENT The simulation model ecosys was initialized with the physical and chemical properties of Trix clay loam (Table 1 in Grant et al., 1999b) including NH4⫹ and NO3⫺ concentrations measured on 15 Dec. 1995 in each soil horizon of each treatment. The model was also provided with the biological properties of spring wheat (Grant et al., 1995a, 1999b). The model was run at Ca ⫽ 363 or 548 mol mol⫺1 under the irrigation and fertilization schedules for the high (35 g N m⫺2 ) and low (7 g N m⫺2 ) N treatments (Table 1) using management practices and hourly meteorological data reported between 15 Dec. 1995 and 20 May 1996 from the field site. All site-specific inputs required by ecosys were confined to site, soil, and plant properties that could be measured independently of the model. All model parameters for CO2 fixation, respiration, and partitioning ⫻ plant and microbial populations were the same as those used in earlier studies of C and energy exchange over agricultural crops (Grant, 1994a; Grant and Baldocchi, 1992; Grant et al., 1993e, 1995a, 1995b, 1999b) and forests (Grant et al., 1999a). Hourly totals of Canopy ⫹ Ground net radiation, Rn (Eq.  and  in Grant et al., 1999b), latent heat, LE (Eq.  and  in Grant et al., 1999b), and sensible heat, S (Eq.  and  in Grant et al., 1999b), simulated in ecosys were compared with hourly averaged fluxes measured under 363 vs. 548 mol mol⫺1 Ca in the high and low N treatments during two 5 d periods, 6–10 Mar. 1996 (DOY 66–70) and 24–28 Apr. 1996 (DOY 115–119). These periods were selected because they included changes in weather during the early preanthesis and late postanthesis stages of the experiment. Hourly net canopy CO2 fixation (Eq.  in Grant et al., 1999b) simulated in ecosys was compared with hourly averaged fluxes for each Ca ⫻ N treatment measured during 21 Feb. 1996. Phytomass and LAI simulated for each Ca ⫻ N treatment was compared with weekly measurements. Agreement between simulated and measured values was tested by comparing standard differences between the two with standard errors of the measured values.
RESULTS Mass and Energy Exchange During the first comparison period from 6 to 10 Mar. 1996 (DOY 66–70) the weather changed from cool (max.
GRANT ET AL.: MODELING CO2, N, AND CLIMATE EFFECTS ON ENERGY EXCHANGE
Fig. 1. Hourly averages of air temperature and wind speed from 6 to 10 Mar. 1996 (DOY 66–70) at the FACE site near Phoenix, AZ.
air temperatures ca. 20⬚C) and windy (wind speeds ⬎10 km h⫺1 ) to warm (max. air temperatures ca. 25⬚C) and still (wind speeds ⬍10 km h⫺1 ) (Fig. 1). Canopy phytomass and LAI at this time were ca. 500 g m⫺2 and 4, with only a small increase from 548 vs. 363 mol mol⫺1 Ca and none from 35 vs. 7 g N m⫺2 (Fig. 8 below). Under 363 mol mol⫺1 Ca and both 35 g N m⫺2 (Fig. 2a) and 7 g N m⫺2 (Fig. 2b) upward LE offset downward Rn so that both modeled and measured S remained close to zero, except when warmer temperatures combined with higher wind speeds to raise LE and cause downward S (e.g., advection during afternoons of DOY 67 and 69). Under 363 mol mol⫺1 Ca and 7 g N m⫺2 both modeled and measured Rn were reduced by ⬍2 W m⫺2 while upward LE and downward S were reduced by at most
30 W m⫺2 from those under the same Ca and 35 g N m⫺2 (Fig. 2b vs. 2a). These small reductions in energy exchange under 7 vs. 35 g N m⫺2 were attributed in the model to limited constraints of N on CO2 fixation and hence on gCi, because soil N was not yet depleted at this preanthesis stage of the experiment. Under 548 mol mol⫺1 Ca, upward LE and downward S were reduced from those under 363 mol mol⫺1 Ca by 50 to 100 W m⫺2 at midday at both 35 (Fig. 3a) and 7 (Fig. 3b) g N m⫺2. These reductions were larger when wind speeds were greater (DOY 66–67) and smaller when wind speeds were less (DOY 68–70). Reductions in LE under 548 vs. 363 mol mol⫺1 Ca occurred in the model because increases in CO2 fixation were proportionately smaller than those in Ca–Ci, forcing reductions
Fig. 2. Hourly averages of net radiation (Rn ), latent heat (LE), and sensible heat (S ) fluxes measured (symbols ⫾ SE) and simulated (lines ) from 6 to 10 Mar. 1996 (DOY 66–70) under 363 mol mol⫺1 CO2 and (a ) 35 g N m⫺2 and (b ) 7 g N m⫺2 fertilization. Negative values represent upward fluxes.
AGRONOMY JOURNAL, VOL. 93, MAY–JUNE 2001
Fig. 3. Differences in latent and sensible heat fluxes under 548 vs. 363 mol mol⫺1 CO2 with (a ) 35 g N m⫺2 and (b ) 7 g N m⫺2 fertilization measured (symbols ) and simulated (lines ) from 6 to 10 Mar. 1996 (DOY 66–70). Positive values represent reductions in upward fluxes under 548 vs. 363 mol mol⫺1 CO2.
in gCi. These reductions were partially offset in the model by increases in gAi caused by higher canopy temperatures under 548 vs. 363 mol mol⫺1 Ca (manifested as more negative S in Fig. 3a and 3b). The reductions in LE under 548 vs. 363 mol mol⫺1 Ca were greater under higher vs. lower wind speeds in the model because gAi was smaller and hence comparatively less important in determining LE. The failure of the model accurately to simulate reductions of up to 100 W m⫺2 measured during the windiest days could be attributed to an underestimate of gAi under high wind speeds and stable conditions, although the model used standard formulations for nonisothermal effects on gAi.
During the second comparison period from 24 to 28 Apr. 1996 (DOY 115–119) the weather was hot (max. air temperatures ⬎30⬚C) while wind speeds rose from ⬍10 km h⫺1 (DOY 115–117) to ⬎10 km h⫺1 (DOY 118–119) (Fig. 4). Canopy phytomass and LAI at this time were ca. 1500 g m⫺2 and 4 (35 g N m⫺2 ) or 2 (7 g N m⫺2 ), with an increase of 10 to 20% in phytomass from 548 vs. 363 mol mol⫺1 Ca and a doubling of LAI from 35 vs. 7 g N m⫺2 (Fig. 8, below). Higher irradiance and air temperatures caused measured and modeled LE to rise from ca. 450 W m⫺2 during early March (Fig. 2) to ca. 750 W m⫺2 during late April (Fig. 5). Under 363 mol mol⫺1 Ca and 35 g N m⫺2, upward LE often
Fig. 4. Hourly averages of air temperature and wind speed from 24 to 28 Apr. 1996 (DOY 115–119) at the FACE site near Phoenix, AZ.
GRANT ET AL.: MODELING CO2, N, AND CLIMATE EFFECTS ON ENERGY EXCHANGE
Fig. 5. Hourly averages of net radiation (Rn ), latent heat (LE), and sensible heat (S ) fluxes measured (symbols ⫾ SE) and simulated (lines ) from 24 to 28 Apr. 1996 (DOY 115–119) under 363 mol mol⫺1 CO2 and (a ) 35 g N m⫺2 and (b ) 7 g N m⫺2 fertilization. Negative values represent upward fluxes.
exceeded downward Rn, causing downward S of as much as 250 W m⫺2 during later afternoons on windier days (advection on DOY 118–119 in Fig. 5a). Under 363 mol mol⫺1 Ca and 7 g N m⫺2, midday Rn was reduced by ⬍10 W m⫺2 while upward LE and downward S were reduced by ca. 100 W m⫺2 from those under the same Ca and 35 g N m⫺2 (Fig. 5b vs. 5a). These reductions were consistent with rises in canopy temperature observed by Seligman et al. (1983) in N limited vs. N fertilized wheat. In the model these reductions were greater than those during the first comparison period (Fig. 2b vs. 2a) because N deficits became more limiting to CO2 fixation as soil mineral N was depleted after anthesis in the low N treatment. Lower CO2 fixation rates forced lower stomatal conductances to conserve Ci/Ca in the model. Under 548 mol mol⫺1 Ca upward LE and downward S were reduced from those under 363 mol mol⫺1 Ca by ⬍50 W m⫺2 at 35 g N m⫺2 (Fig. 6a) and by ⬎100 W m⫺2 at 7 g N m⫺2 (Fig. 6b). Reductions in LE under higher Ca were smaller when wind speeds were ⬍10 km h⫺1 (DOY 115–117) although those at 7 g N m⫺2 were less affected by lower wind speeds. Earlier senescence and leaf loss under 548 vs. 363 mol mol⫺1 Ca and 7 g N m⫺2 (Fig. 8, below) caused larger reductions in LE after DOY 117 (Fig. 6b). Lower fertilizer application caused greater reductions in LE under 548 vs. 363 mol mol⫺1 Ca in the model (Fig. 6b vs. 6a) because N became more limiting to CO2 fixation and hence to stomatal conductance under higher Ca. Stomatal conductance was therefore comparatively more limiting to LE than was
aerodynamic conductance so that reductions in LE were less affected by low wind speeds under 548 vs. 363 mol mol⫺1 Ca. These larger reductions in stomatal conductance caused average midday canopy temperatures simulated under 548 vs. 363 mol mol⫺1 Ca to rise during this period by 0.5⬚C at 35 g N m⫺2 and by 1.2⬚C at 7 g N m⫺2. Corresponding increases of 0.6 and 1.1⬚C were reported by Kimball et al. (1999). Comparisons of modeled vs. measured LE during both comparison periods indicated that the model accounted for ⬎95% of measured variation with no apparent bias for any Ca ⫻ N treatment (Table 2). Standard differences between modeled and measured fluxes approached the standard errors of the measured fluxes. These differences were largely affected by uncertainty Table 2. Results from regressions of simulated latent heat fluxes on measured latent heat fluxes from 1 to 10 Mar. 1996 and from 19 to 29 Apr. 1996. Ca mol
SE (measured) W m⫺ 2
1–10 Mar. 1996 363 548 363 548
35 35 7 7
363 548 363 548
35 35 7 7
0.95 0.97 0.98 0.97 1.00 0.96 1.00 0.97 18–28 Apr. 1996 1.00 1.00 1.04 1.05
0.98 0.98 0.95 0.96
31 27 32 28
13 20 12 24
28 40 57 42
22 33 42 44
AGRONOMY JOURNAL, VOL. 93, MAY–JUNE 2001
Fig. 6. Differences in latent and sensible heat fluxes under 548 vs. 363 mol mol⫺1 CO2 with (a ) 35 g N m⫺2 and (b ) 7 g N m⫺2 fertilization measured (symbols ) and simulated (lines ) from 24 to 28 Apr. 1996 (DOY 115–119). Positive values represent reductions in upward fluxes under 548 vs. 363 mol mol⫺1 CO2.
in the timing of hourly averaged calculations when fluxes were changing by as much as 200 W m⫺2 h⫺1 during early mornings and late afternoons.
changes in stomatal conductance, and hence in mass and energy exchange (Fig. 3a, 3b, 6a, 6b).
Carbon Dioxide Fixation
The 7 g N m⫺2 treatment caused only a limited reduction in phytomass growth (Fig. 8a) and LAI (Fig. 8b) from that in the 35 g N m⫺2 treatment before anthesis (DOY 85), and hence only a limited reduction in energy exchange measured and modeled during the early March comparison period (Fig. 2b vs. 2a). This limited reduction was consistent with the limited constraint of N on canopy CO2 fixation before anthesis (Fig. 7). The 7 g N m⫺2 treatment caused a greater reduction in phytomass growth from that in the 35 g N m⫺2 treatment after anthesis as soil mineral N became depleted. This depletion caused a greater reduction in mass and energy exchange to be measured and modeled with 7 vs. 35 g N m⫺2 during the late April comparison period (Fig. 5b vs. 5a) vs. the early March comparison period (Fig. 2b vs. 2a). Late season depletion of soil N in the modeled 7 g N m⫺2 treatment was more rapid under 548 vs. 363 mol mol⫺1 Ca, causing a larger reduction in late season phytomass growth (Fig. 8a) and also in LAI (Fig. 8b) through more rapid remobilization and senescence and through higher canopy temperatures (cf. Seligman et al., 1983). This more rapid depletion also caused the larger reductions in mass and energy exchange measured and modeled in the 7 vs. 35 g N m⫺2 treatment under 548 vs. 363 mol mol⫺1 Ca, especially after DOY 117 (Fig. 6b vs. 6a).
Lower storage N/C ratios developed in the model under both 548 vs. 363 mol mol⫺1 Ca (due to more rapid CO2 fixation) and 7 vs. 35 g N m⫺2 (due to less rapid N uptake). These lower ratios caused reductions to be modeled in the specific activities of rubisco and chlorophyll and in their areal concentrations by lowering leaf N/C ratios during leaf growth. On 10 Mar. 1996 (DOY 70 during the preanthesis comparison period), leaf N concentrations simulated (measured) under 363 vs. 548 mol mol⫺1 Ca were 89 (92 ⫾ 3) vs. 87 (88 ⫾ 7) mg N g⫺1 C at 35 g N m⫺2 and 79 (76 ⫾ 7) vs. 72 (70 ⫾ 6) mg N g⫺1 C at 7 g N m⫺2 (measured data are reported in Sinclair et al., 2000). Lower activities and areal concentrations of rubisco and chlorophyll caused lower leaf CO2 fixation capacities under higher Ca and lower N. These lower capacities interacted with ambient Ca, irradiance, temperature, and water status to determine leaf and hence canopy CO2 fixation rates for the Ca ⫻ N treatments in the model. During 21 Feb. 1996 leaf CO2 fixation capacities allowed a measured increase in canopy CO2 fixation of 3 to 6 mol m⫺2 s⫺1 to be simulated under 548 vs. 363 mol mol⫺1 Ca and a measured increase of 1 to 3 mol m⫺2 s⫺1 to be simulated under 35 vs. 7 g N m⫺2 (Fig. 7). These modeled increases in CO2 fixation under higher Ca and N determined
Seasonal Growth and Water Use
GRANT ET AL.: MODELING CO2, N, AND CLIMATE EFFECTS ON ENERGY EXCHANGE
Fig. 7. Canopy CO2 fluxes measured (symbols ) and simulated (lines ) on 21 Feb. 1996 under 363 vs. 548 mol mol⫺1 CO2 and 35 vs. 7 g N m⫺2 fertilization.
Reductions in mass and energy exchange and LAI under 548 vs. 363 mol mol⫺1 Ca and 7 vs. 35 g N m⫺2 caused corresponding reductions in seasonal water use. Total ET modeled between mid-February and the end of April 1996 under 548 vs. 363 mol mol⫺1 Ca was reduced by 9 and 16% at 35 and 7 g N m⫺2, respectively. These reductions corresponded to ones of 7 and 19% in seasonally averaged daily ET calculated from energy balance measurements during the same period by Kimball et al. (1999).
DISCUSSION Low N imposed a greater constraint upon seasonal phytomass growth under 548 vs. 363 mol mol⫺1 Ca (Fig. 8a) as has been found in several other studies (Hocking and Meyer, 1991; McKee and Woodward, 1994; Owensby et al., 1994; Rogers et al., 1993, 1996; Wolf, 1996; Wong, 1979). In the model this greater constraint was imposed by reduced rubisco activity and concentration caused by lower N/C ratios in leaves and storage pools that developed under higher Ca and lower N. These reductions in rubisco activity and concentration represent acclimation to higher Ca as described by Bowes (1991). Because the Ci/Ca ratio is conserved by the stomatal model in ecosys (Grant et al., 1999b), lower rubisco activity and concentration forced greater reductions in stomatal conductances and hence in hourly LE and seasonal ET under 548 vs. 363 mol mol⫺1 Ca when N was limiting to CO2 fixation (Fig. 3 and 6), as has also been reported from other experiments (Wong, 1979). In other research at the FACE site, Kimball et al. (1999) reported from an energy balance study that the decrease in canopy conductance at low vs. high N was larger at elevated Ca. Wall et al. (1997) demonstrated from a porometer study that leaf conductance was lowest under low N and elevated Ca. The direct effects of elevated Ca and low N on LE through stomatal conductance and LAI in the model were partially offset by indirect effects of elevated Ca and low N on LE through: 1. Higher canopy temperatures caused by reduced LE that raised vapor pressure gradients between the canopy and the atmosphere. These higher can-
Fig. 8. Seasonal time course of (a ) aboveground phytomass and (b ) LAI measured (symbols ) and simulated (lines ) under 35 vs. 7 g N m⫺2 fertilization and 363 vs. 548 mol mol⫺1 CO2 concentration.
opy temperatures were manifested as greater upward S (Fig. 2 and 5). 2. Higher aerodynamic conductance due to greater buoyancy over warmer canopies, especially under low wind speeds and stable conditions. These increases in aerodynamic conductance could almost entirely offset reductions in stomatal conductance due to higher Ca and lower N (Fig. 3 and 6) when atmospheric conditions caused aerodynamic conductance to be less than stomatal conductance. The effect of wind speed on changes in LE with Ca was less apparent under low N because stomatal conductance was more limiting to LE (Fig. 6b vs. 6a). 3. Wetter soils under reduced LE that allowed more rapid soil evaporation, especially following irrigations. The accurate coupling of changes with Ca and N of both CO2 fixation (Fig. 7) and LE (Fig. 2, 3, 5, and 6) in ecosys supports the use of stomatal models in which the Ci/Ca ratio is conserved so that stomatal conductance varies directly with CO2 fixation rate and inversely with Ca or Ca – Ci (e.g., Ball, 1988; Grant et al., 1999b).
AGRONOMY JOURNAL, VOL. 93, MAY–JUNE 2001
However, the simulation of stomatal effects alone on LE may cause reductions in LE under elevated Ca and low N to be overestimated, especially under low wind speeds and stable conditions when aerodynamic conductance is limiting (Fig. 3 and 6). Stomatal models therefore need to function as part of a comprehensive energy balance model to ensure that indirect effects of Ca, N, and climate on LE are fully represented. Such comprehensive models are a necessary component of soil– plant–atmosphere models used in site-specific studies of climate change effects on terrestrial ecosystems. Recent FACE research indicates that an elevation of 200 mol mol⫺1 Ca would raise wheat yields by 10 to 15% and reduce ET by 7% under well irrigated and fertilized conditions (Kimball et al., 1995, 1999). However, these changes would not likely be realized in most terrestrial ecosystems, which are N-limited. These ecosystems typically have lower LE and higher S than reported here, indicating a greater stomatal limitation to mass and energy exchange even when water is not limiting, as found for example in boreal coniferous forests (e.g., Jarvis et al., 1997). Our experimental and modeling work indicate that an elevation of 200 mol mol⫺1 Ca would likely raise the biological yields of an N-limited ecosystem less, and therefore reduce its ET more, than it would the yield and ET of a well-fertilized ecosystem. The extent to which ET is reduced under elevated Ca may be also be greater in windier climates with taller vegetation where stomatal conductance is the dominant control on transpiration. The greater partitioning of energy exchange away from LE and toward S in N-limited ecosystems under rising Ca may cause more sensible heat to be returned to the atmosphere and more water to be conserved within terrestrial ecosystems than currently foreseen in hypothesized climate change scenarios. ACKNOWLEDGMENTS This research was supported by Grant no. DE-FG03-95ER62072 from the Department of Energy Terrestrial Carbon Processes Research Program to the University of Arizona, Tucson, AZ, and Maricopa, AZ (S. Leavitt, T. Thompson, A. Matthias, R. Rauschkolb, and H.Y. Cho are principal investigators) and by Interagency Agreement no. IBN-9652614 between the National Science Foundation and the USDA-ARS Water Conservation Laboratory as part of the NSF/DOE/ NASA/USDA Joint Program on Terrestrial Ecology and Global Change (TECO II) (G.W. Wall, F.J. Adamsen, B.A. Kimball, and A.N. Webber principal investigators). Operational support was also provided by the USDA-ARS U.S. Water Conservation Laboratory, Phoenix, AZ, and by the Potsdam Institute for Climate Impact Research, Potsdam, Germany. We also acknowledge the helpful cooperation of Dr. Roy Rauschkolb and his staff at the Maricopa Agricultural Center. The FACE apparatus was furnished by Brookhaven National Laboratory, and we are grateful to Mr. Keith Lewin, Dr. John Nagy, and Dr. George Hendrey for assisting in its installation and consulting about its use. Computational facilities were provided by the Multimedia Advanced Computational Initiative (MACI) at the Universities of Alberta and Calgary. This work contributes to the Global Change Terrestrial Ecosystem (GCTE) Core Research Programme, which is part of the International Geosphere-Biosphere Programme (IGBP).
REFERENCES Azcon-Bieto, J. 1983. Inhibition of photosynthesis by carbohydrates in wheat leaves. Plant Physiol. 73:681–686. Ball, J.T. 1988. An analysis of stomatal conductance. Ph.D. diss. Stanford Univ., Stanford CA. Baxter, R., T.W. Ashenden, and J.F. Farrar. 1997. Effect of elevated CO2 and water status on growth, dry matter partitioning and nutrient content of Poa alpina var. vivipara L. J. Exp. Bot. 48:1477–1486. Bowes, G. 1991. Growth at elevated CO2: Photosynthetic responses mediated through Rubisco. Plant Cell Environ. 14:795–806. Brooks, T.J. 1998. Free-air carbon dioxide enrichment and nitrogen stress effects on wheat canopy architecture and gas exchange. M.S. thesis. Arizona State Univ., Tempe AZ. Brugge, R., and J.H.M. Thornley. 1985. A growth model of root mass and vertical distribution dependent on carbon substrate from photosynthesis and with non-limiting soil conditions. Ann. Bot. 55:563–577. Carlson, T.N., and J.A. Bunce. 1996. Will a doubling of atmospheric carbon dioxide concentration lead to an increase or decrease in water consumption by crops? Ecol. Model. 88:241–246. Chahine, M.T. 1992. The hydrological cycle and its influence on climate. Nature (London) 359:373–380. Champigny, M., and C. Foyer. 1992. Nitrate activation of cytosolic protein kinase diverts photosynthetic carbon from sucrose to amino acid biosynthesis. Plant Physiol. 100:7–12. Choudhury, B.J., R.J. Reginato, and S.B. Idso. 1986. An analysis of infrared temperature observations over wheat and calculation of latent heat flux. Agric. For. Meteorol. 37:75–88. Cure, J.D., and B. Acock. 1986. Crop responses to carbon dioxide doubling: A literature survey. Agric. For. Meteorol. 38:127–145. Farquhar, G.D., S. von Caemmerer, and J.A. Berry. 1980. A biochemical model of photosynthetic CO2 assimilation in leaves of C3 species. Planta 149:78–90. Garcia, R.L., J.M. Norman, and D.K. McDermitt. 1990. Measurements of canopy gas exchange using an open chamber system. Remote Sens. Rev. 5:141–162. Gifford, R.M. 1979. Growth and yield of CO2-enriched wheat under water-limited conditions. Aust. J. Plant Physiol. 6:367–378. Grant, R.F. 1991. The distribution of water and nitrogen in the soil– crop system: A simulation study with validation from a winter wheat field trial. Fert. Res. 27:199–214. Grant, R.F. 1994a. Simulation of competition between barley (Hordeum vulgare L.) and wild oat (Avena fatua L.) under different managements and climates. Ecol. Model. 71:269–287. Grant, R.F. 1994b. Simulation of ecological controls on nitrification. Soil Biol. Biochem. 26:305–315. Grant, R.F. 1995. Mathematical modelling of nitrous oxide evolution during nitrification. Soil Biol. Biochem. 27:1117–1125. Grant, R.F. 1996. ecosys. p. 65–74 In Global Change and Terrestrial Ecosystems Focus 3 Wheat Network. Model and Experimental Meta. Data. 2nd ed. GCTE Focus 3 Office, NERC Centre for Ecology and Hydrology, Wallingford, Oxon, UK. Grant, R.F. 1998. Simulation in ecosys of root growth response to contrasting soil water and nitrogen. Ecol. Model. 107:237–264. Grant, R.F., and D.D. Baldocchi. 1992. Energy transfer over crop canopies: Simulation and experimental verification. Agric. For. Meteorol. 61:129–149. Grant, R.F., T.A. Black, G. den Hartog, J.A. Berry, H.H. Neumann, P.D. Blanken, P.C.Yang, C. Russell, and I.A. Nalder. 1999a. Diurnal and annual exchanges of mass and energy between an aspen– hazelnut forest and the atmosphere: Testing the mathematical model ecosys with data from the BOREAS experiment. J. Geophys. Res. 104:27,699–27,717. Grant, R.F., R.L. Garcia, P.J. Pinter, Jr., D. Hunsaker, G.W. Wall, B.A. Kimball, and R.L. LaMorte. 1995a. Interaction between atmospheric CO2 concentration and water deficit on gas exchange and crop growth: Testing of ecosys with data from the Free Air CO2 Enrichment (FACE) experiment. Global Change Biol. 1:443–454. Grant, R.F., and D.J. Heaney. 1997. Inorganic phosphorus transformation and transport in soils: Mathematical modeling in ecosys. Soil Sci. Soc. Am. J. 61:752–764. Grant, R.F., N.G. Juma, and W.B. McGill. 1993a. Simulation of carbon and nitrogen transformations in soils: I. Mineralization. Soil Biol. Biochem. 27:1317–1329.
GRANT ET AL.: MODELING CO2, N, AND CLIMATE EFFECTS ON ENERGY EXCHANGE
Grant, R.F., N.G. Juma, and W.B. McGill. 1993b. Simulation of carbon and nitrogen transformations in soils: II. Microbial biomass and metabolic products. Soil Biol. Biochem. 27:1331–1338. Grant, R.F., B.A. Kimball, P.J. Pinter, Jr., G.W. Wall, R.L. Garcia, and R.L. LaMorte. 1995b. Carbon dioxide effects on crop energy balance: Testing ecosys with a Free-Air CO2 Enrichment (FACE) Experiment. Agron. J. 87:446–457. Grant, R.F., and I.A. Nalder. 2000. Climate change effects on net carbon exchange of a boreal aspen–hazelnut forest: Estimates from the ecosystem model ecosys. Global Change Biol. 6:183–200. Grant, R.F., M. Nyborg, and J. Laidlaw 1993c. Evolution of nitrous oxide from soil: I. Model development. Soil Sci. 156:259–265. Grant, R.F., M. Nyborg, and J. Laidlaw 1993d. Evolution of nitrous oxide from soil: II. Experimental results and model testing. Soil Sci. 156:266–277. Grant, R.F., and J.A. Robertson. 1997. Phosphorus uptake by root systems: mathematical modelling in ecosys. Plant Soil 188:279–297. Grant, R.F., P. Rochette, and R.L. Desjardins. 1993e. Energy exchange and water use efficiency of crops in the field: Validation of a simulation model. Agron. J. 85:916–928. Grant, R.F., G.W. Wall, B.A. Kimball, K.F.A. Frumau, P.J. Pinter, Jr., D.J. Hunsaker, and R.L. Lamorte. 1999b. Crop water relations under different CO2 and irrigation: Testing of ecosys with the free air CO2 enrichment (FACE) experiment. Agric. For. Meteorol. 95:27–51. Hendrey, G.R. (ed.). 1993. FACE: Free-Air CO2 Enrichment for plant research in the field. CRC Press, Boca Raton, FL. Herold, A. 1980. Regulation of photosynthesis by sink activity: the missing link. New Phytol. 86:131–144. Hocking, P.J., and C.P. Meyer. 1991. Effects of CO2 enrichment and nitrogen stress on growth and partitioning of dry matter and nitrogen in wheat and maize. Aust. J. Plant Physiol. 18:339–356. Huband, N.D.S., and J.L. Monteith. 1986. Radiative surface temperature and energy balance of a wheat canopy. Bound. Layer Meteorol. 36:1–17. Idso, S.B., B.A. Kimball, M.G. Anderson, and J.R. Mauney. 1987. Effects of atmospheric CO2 enrichment on plant growth: the interactive role of air temperature. Agric. Ecosyst. Environ. 20:1–10. Jarvis, P.G., J.M. Massheder, S.E. Hale, J.B. Moncrieff, M. Rayment, and S.L. Scott. 1997. Seasonal variation in carbon dioxide, water vapor, and energy exchanges of a boreal black spruce forest. J. Geophys. Res. 102:28,953–28,966. Jarvis, P.G., and K.G. McNaughton. 1986. Stomatal control of transpiration: scaling up from leaf to region. Adv. Ecol. Res. 15:1–49. Kimball, B.A., R.L. LaMorte, P.J. Pinter, Jr., G.W. Wall, D.J. Hunsaker, F.J. Adamsen, S.W. Leavitt, T.L. Thompson, A.D. Matthias, and T.J. Brooks. 1999. Free-air CO2 enrichment (FACE) and soil nitrogen effects on energy balance and evapotranspiration of wheat. Water Resour. Res. 35:1179–1190. Kimball, B.A., P.J. Pinter, Jr., R.L. Garcia, R.L. LaMorte, G.W. Wall, D.J. Hunsaker, G. Wechsung, F. Wechsung, and T. Kartschall. 1995. Productivity and water use of wheat under free-air CO2 enrichment. Global Change Biol. 1:429–442. Lawlor, D.W., and R.A.C. Mitchell. 1991. The effects of increasing CO2 on crop photosynthesis and productivity: A review of field studies. Plant Cell Environ. 14:807–818. Makino, A., M. Harada, T. Sato, H. Nakano, and T. Mae. 1997. Growth and N allocation in rice plants under CO2 enrichment. Plant Physiol. 115:199–203. McKee, I.F., and F.I. Woodward. 1994. CO2 enrichment responses of wheat: interactions with temperature, nitrate and phosphate. New Phytol. 127:447–453. Ottoni, T.B., A.D. Matthias, A.F. Guerra, and D.C. Slack. 1992. Comparison of three resistance methods for estimating heat flux under stable conditions. Agric. For. Meteorol. 58:1–18. Owensby, C.E., L. Auen, and P. Coyne. 1994.. Biomass production in a nitrogen-fertilized tallgrass prairie ecosystem exposed to ambient and elevated levels of CO2. Plant Soil 165:105–113.
Parton, W.J., J.M.O. Scurlock, D.S.Ojima, D.S. Schimel, D.O. Hall, and SCOPEGRAM group members. 1995. Impact of climate change on grassland production and soil carbon worldwide. Global Change Biol. 1:13–22. Penning de Vries, F.W.T. 1982. Crop production in relation to availability of nitrogen. p. 213–221. In F.W.T. Penning de Vries and H.H. van Laar (ed.) Simulation of plant growth and crop production. Centre for Agricultural Publ. and Documentation, Wageningen, the Netherlands. Perrier, A. 1982. Land surface processes: Vegetation. p. 395–448. In P.S. Eagleson (ed.) Atmospheric general circulation models. Cambridge Univ. Press, Cambridge, UK. Rogers, H.H., J.D. Cure, and J.M. Smith. 1986. Soybean growth and yield response to elevated carbon dioxide. Agric. Ecosyst. Environ. 16:113–128. Rogers, G.S., P.J. Milham, M. Gillings, and J.P. Conroy. 1996. Sink strength may be the key to growth and nitrogen responses in Ndeficient wheat at elevated CO2. Aust. J. Plant Physiol. 23:253–264. Rogers, G.S., L. Payne, P. Milham, and J. Conroy. 1993. Nitrogen and phosphorus requirements of cotton and wheat under changing CO2 concentrations. p. 257–260 In J. Barrow (ed.) Plant nutrition: From genetic engineering to field practice. Kluwer Academic Publ., Dordrecht, the Netherlands. Rosenzweig, C., and M.L. Parry. 1994. Potential impact of climate change on world food supply. Nature (London) 367:133–138. Saliendra, N.Z., and F.C. Meinzer. 1991. Symplast volume, turgor, stomatal conductance, and growth in relation to osmotic and elastic adjustment in droughted sugarcane. J. Exp. Bot. 42:1251–1259. Seligman, N.G., R.S. Loomis, J. Burke, and A. Abshahi. 1983. Nitrogen nutrition and canopy temperature in field-grown spring wheat. J. Agric. Sci. 101:691–697. Sinclair, T.R., P.J. Pinter, Jr., B.A. Kimball, F.J. Adamsen, R.L. LaMorte, G.W. Wall, D.J. Hunsaker, N. Adam, T.J. Brooks, R.L. Garcia, T. Thompson, S. Leavitt, and A. Matthias. 2000. Leaf nitrogen concentration of wheat subjected to elevated CO2 and either water or N deficits. Agric. Ecosyst. Environ. 79:53–60. Stitt, M. 1991. Rising CO2 levels and their potential significance for carbon flow in photosynthetic cells. Plant Cell Environ. 14:741–762. Stockle, C.O., J.R. Williams, N.J. Rosenberg, and C.A. Jones. 1992. A method for estimating the direct and climatic effects of rising atmospheric carbon dioxide on growth and yield of crops: I. Modification of the EPIC model for climate change analysis. Agric. Syst. 38:225–238. Thornley, J.H. 1995. Shoot:root allocation with respect to C, N and P: An investigation and comparison of resistance and teleonomic models. Ann. Bot. 75:391–405. Van Bavel, C.H.M., and D.I. Hillel. 1976. Calculating potential and actual evaporation from a bare soil surface by simulation of concurrent flow of water and heat. Agric. Meteorol. 17:453–476. Wall, G.W., N.R. Adam, T.J. Brooks, A.N. Webber, B.A. Kimball, P.J. Pinter, Jr., R.L. LaMorte, and S.W. Leavitt. 1997. Acclimation response of the photosynthetic apparatus of spring wheat to fullseason CO2 enrichment and nitrogen stress. p. 21. In Agronomy abstracts. ASA, Madison, WI. Wolf, J. 1996. Effects of nutrient supply (NPK) on spring wheat response to elevated atmospheric CO2. Plant Soil 185:113–123. Wong, S.C. 1979. Elevated atmospheric partial pressure of CO2 and plant growth: I. Interactions of nitrogen nutrition and photosynthetic capacity in C3 and C4 plants. Oecologia 44:68–74. Wong, S.C. 1990. Elevated atmospheric partial pressure of CO2 and plant growth: II. Non-structural carbohydrate content in cotton plants and its effect on growth parameters. Photosynth. Res. 23:171–180. Wong, S.C., I.R. Cowan, and G.D. Farquhar. 1979. Stomatal conductance correlates with photosynthetic capacity. Nature (London) 282:424–426. Zur, B., and J.W. Jones. 1981. A model for the water relations, photosynthesis and expansive growth of crops. Water Resour. Res. 17: 311–320.