R. F. Grant

Oak forest carbon and water simulations: model intercomparisons and evaluations against independent data

Models represent our primary method for integration of small-scale, process- level phenomena into a comprehensive description of forest-stand or ecosystem function. They also represent a key method for testing hypotheses about the response of forest ecosystems to multiple changing environmental conditions. This paper describes the eval- uation of 13 stand-level models varying in their spatial, mechanistic, and temporal com- plexity for their ability to capture intra- and interannual components of the water and carbon cycle for an upland, oak-dominated forest of eastern Tennessee. Comparisons between model simulations and observations were conducted for hourly, daily, and annual time steps. Data for the comparisons were obtained from a wide range of methods including: eddy covariance, sapflow, chamber-based soil respiration, biometric estimates of stand-level net primary production and growth, and soil water content by time or frequency domain reflectometry. Response surfaces of carbon and water flux as a function of environmental drivers, and a variety of goodness-of-fit statistics (bias, absolute bias, and model efficiency) were used to judge model performance. A single model did not consistently perform the best at all time steps or for all variables considered. Intermodel comparisons showed good agreement for water cycle fluxes, but considerable disagreement among models for predicted carbon fluxes. The mean of all model outputs, however, was nearly always the best fit to the observations. Not surprisingly, models missing key forest components or processes, such as roots or modeled soil water content, were unable to provide accurate predictions of ecosystem responses to short-term drought phenomenon. Nevertheless, an inability to correctly capture short-term physiolog- ical processes under drought was not necessarily an indicator of poor annual water and carbon budget simulations. This is possible because droughts in the subject ecosystem were of short duration and therefore had a small cumulative impact. Models using hourly time steps and detailed mechanistic processes, and having a realistic spatial representation of the forest ecosystem provided the best predictions of observed data. Predictive ability of all models deteriorated under drought conditions, suggesting that further work is needed to evaluate and improve ecosystem model performance under unusual conditions, such as drought, that are a common focus of environmental change discussions.

Multimodel ensembles of wheat growth: many models are better than one.

Crop models of crop growth are increasingly used to quantify the impact of global changes due to climate or crop management. Therefore, accuracy of simulation results is a major concern. Studies with ensembles of crop models can give valuable information about model accuracy and uncertainty, but such studies are difficult to organize and have only recently begun. We report on the largest ensemble study to date, of 27 wheat models tested in four contrasting locations for their accuracy in simulating multiple crop growth and yield variables. The relative error averaged over models was 24-38% for the different end-of-season variables including grain yield (GY) and grain protein concentration (GPC). There was little relation between error of a model for GY or GPC and error for in-season variables. Thus, most models did not arrive at accurate simulations of GY and GPC by accurately simulating preceding growth dynamics. Ensemble simulations, taking either the mean (e-mean) or median (e-median) of simulated values, gave better estimates than any individual model when all variables were considered. Compared to individual models, e-median ranked first in simulating measured GY and third in GPC. The error of e-mean and e-median declined with an increasing number of ensemble members, with little decrease beyond 10 models. We conclude that multimodel ensembles can be used to create new estimators with improved accuracy and consistency in simulating growth dynamics. We argue that these results are applicable to other crop species, and hypothesize that they apply more generally to ecological system models.

Boreal forest CO2 exchange and evapotranspiration predicted by nine ecosystem process models: Intermodel comparisons and relationships to field measurements

Nine ecosystem process models were used to predict CO2 and water vapor exchanges by a 150-year-old black spruce forest in central Canada during 1994–1996 to evaluate and improve the models. Three models had hourly time steps, five had daily time steps, and one had monthly time steps. Model input included site ecosystem characteristics and meteorology. Model predictions were compared to eddy covariance (EC) measurements of whole-ecosystem CO2 exchange and evapotranspiration, to chamber measurements of nighttime moss-surface CO2 release, and to ground-based estimates of annual gross primary production, net primary production, net ecosystem production (NEP), plant respiration, and decomposition. Model-model differences were apparent for all variables. Model-measurement agreement was good in some cases but poor in others. Modeled annual NEP ranged from −11 g C m−2 (weak CO2 source) to 85 g C m−2 (moderate CO2 sink). The models generally predicted greater annual CO2 sink activity than measured by EC, a discrepancy consistent with the fact that model parameterizations represented the more productive fraction of the EC tower “footprint.” At hourly to monthly timescales, predictions bracketed EC measurements so median predictions were similar to measurements, but there were quantitatively important model-measurement discrepancies found for all models at subannual timescales. For these models and input data, hourly time steps (and greater complexity) compared to daily time steps tended to improve model-measurement agreement for daily scale CO2 exchange and evapotranspiration (as judged by root-mean-squared error). Model time step and complexity played only small roles in monthly to annual predictions.

Simulation of carbon and nitrogen transformations in soil: Mineralization

Abstract If mathematical models of decomposition and transformation processes are to be rigorously validated, they should be tested against the microbial dynamics from which these processes arise. A mathematical model was constructed from kinetic equations for microbial activity reported in the literature and in earlier models, and was tested against C and N mineralization during incubation of labelled glucose, cellulose and crop residue on several soils. The model was able to reproduce temporal trends in the mineralization, immobilization and retention of labelled C and N to within 10% of recorded values over time scales of hours, days and years following soil amendments. By treating the humification and adsorption of microbial products as functions of soil clay content, the simulated mineralization of C amendments was reduced, and retention increased, to extents consistent with recorded data in soils with clay contents which varied from 4 to 34%. By allowing the decomposition of all soil organic matter to be determined by the total concentration of microbial biomass, the model also reproduced most of the increased mineralization of non-labelled C recorded from 14C-amended soils.

Modelling plant carbon and nitrogen dynamics of a boreal aspen forest in CLASS — the Canadian Land Surface Scheme

The formulation of the physiological processes of plant carbon and nitrogen that have recently been implemented in CLASS — the Canadian Land Surface Scheme is presented, including plant photosynthesis of sunlit and shaded leaves, root nitrogen uptake, tissue respiration, growth, and senescence. Results from this formulation provide the vegetation parameters (e.g. stomatal resistance, leaf area index, and root length and distribution) for simulating energy and water exchange, and contribute to the carbon and nitrogen biogeochemical calculations and ecosystem CO2 flux estimations in CLASS. The model was parameterized for deciduous trees and run at a time step of 30 min. Modelled results were tested with measurements of leaf CO2 exchange, half-hourly and daily plant CO2 exchange, and annual plant carbon budgets from the Old Aspen (Populus tremuloides) site in the Southern Study Area (SSA-OA) of the Boreal Ecosystem-Atmosphere Study (BOREAS). Model tests at leaf level and half-hourly timescale improve confidence in the predictive capabilities of the model at canopy level and longer timescale. At half-hourly timescale, simulated plant CO2 exchange explained 81% of the variance in the measured CO2 flux derived from eddy correlation measurements and soil chamber measurements in the three comparing weeks in 1994. At a coarser timescale, simulated daily plant CO2 exchange explained 86% of the variance in the measured values during 1994 and 1996. Annual gross primary production (GPP) accumulated from half-hourly values in the model in 1994 and 1996 averaged about 1053 g C m−2 for this aspen site, close to the estimations from field eddy correlation measurements. Of this carbon fixation, 59% was consumed by autotrophic respiration (Ra) for plant growth and maintenance in the model, similar to estimations from chamber measurements. Sensitivity analyses show that the modelled plant GPP, Ra and net primary production NPP (=GPP−Ra) were very sensitive to the changes in temperature, implying that this boreal aspen ecosystem is strongly constrained by temperature conditions.

Simulation of carbon and nitrogen transformations in soil: Microbial biomass and metabolic products

The mineralization of C and N in soils involves the dynamic behaviour of the microbial biomass. This behaviour was reproduced in a simulation model which was used to study microbial growth and mineralization following soil amendments with labelled glucose and crop residues. Temporal trends of simulated microbial C and N associated with these amendments were consistent with those of labelled C and N estimated from chloroform fumigation (SD 1.0 to <0.01 gg−1 day−1 that were caused by changes in substrate availability.

Simulation of methanogenesis in the mathematical model ecosys

Abstract A biologically-based approach to the modelling of CH 4 transformations in soil would likely be of general use in the estimation of soil–atmosphere CH 4 exchange as part of global climate studies. Such an approach was adopted as part of the ecosys modelling project to simulate the interrelated activities of anaerobic fermenters and H 2 -producing acetogens, of acetotrophic and hydrogenotrophic methanogens and of autotrophic methanotrophs. The simulation of these activities was based on the stoichiometries and energetics of the transformations mediated by each microbial community. Model results were tested with published data for CH 4 emission during anaerobic incubation of soils amended with different plant residues. The model reproduced within experimental error observed changes with temperature in CH 4 emission rates and in CH 4 -to-CO 2 emission ratios caused by different residue amendments. These changes were driven in the model by the dynamics of growth and respiration by fermentative and methanogenic communities under different temperatures and residue amendments. Growth yields of methanogenic communities in the model were consistent with growth yields observed in other studies, but specific growth rates in the model were somewhat lower. Because the model represents basic microbial behaviour during CH 4 transformation, its use will likely improve the confidence with which changes in soil–atmosphere CH 4 exchange can be estimated under hypothesized changes in soils, climates and management. However testing of the model should be extended to more complex conditions of transient anaerobiosis under natural temperatures to establish confidence in its predictive capability for soil–atmosphere CH 4 exchange.

Simulation in ecosys of root growth response to contrasting soil water and nitrogen

If mathematical models of plant growth are to perform reliably under diverse conditions of soil and climate, then the effects of these conditions on root growth must be represented. A mathematical model of root and mycorrhizal growth is proposed to represent the effects of soil and climate on growth using the hypothesis that a functional equilibrium exists among root axes and shoot branches. In this model access to growth resources (C, N, P, water) by different axes or branches depends upon (1) proximity of the axis or branch to the point of resource acquisition, and (2) the rate at which resources are consumed by the axis or branch in relation to that by other axes or branches. This model was coupled to a plant growth model as part of the ecosystem simulation model ecosys, and its sensitivity to changes in model parameters and soil boundary conditions was tested. Simulated root growth was less sensitive to changes in soil water and nitrogen than was simulated shoot growth. This lower sensitivity allowed the model to simulate changes in root:shoot ratios with changes in soil water and nitrogen that were consistent with those commonly reported in the literature. Changes in soil water also caused changes in vertical distributions of root length density to be simulated that were also consistent with those reported. Changes in root:shoot partitioning and in root density distributions allowed improved access by plants in the model to limiting growth resources. The root model was parameterized from basic root growth studies conducted independently of the model, and without reference to site-specific patterns of seasonal root growth. Consequently the model is likely to be of general value in the simulation of root growth under diverse soil conditions, although such generality needs to be established through further testing under different soils, climates and crops. The precision of some of the model parameters is uncertain and the sensitivity of the model to this uncertainty is discussed.

Simulation of ecological controls on nitrification

Abstract The accurate representation of nitrification in mathematical models is necessary to account for N distribution and transport through simulated ecosystems. Because of the complexity with which environmental factors interact to control nitrification, these representations should be based upon the kinetics of the limiting reactions mediated by microbial nitrifiers, rather than upon those of NO−3 formation as is currently done. In the nitrification model proposed here, algorithms for autotrophic nitrifier activity and growth are coupled to ones for substrate production and transport under dynamic conditions of water, temperature and NH+4 concentration. This model is used to reproduce the time course of nitrifier biomass and activity following different NH+4amendments under different CO2 concentrations, temperatures and water contents in a manner consistent with literature reports. Rates of NO−3 formation arising from simulated nitrifier activity are comparable to those measured during incubations of NH+4-amended soil under common conditions of temperature and water content. However, the rigour of comparison is limited by the detail with which soil conditions are reported, and by the temporal resolution with which NO−3 formation is measured, from most nitrification experiments.

Mathematical modelling of nitrous oxide evolution during nitrification

There is a need for a process-based model of N2O evolution during nitrification as part of larger models used to study trace gas exchange between terrestrial ecosystems and the atmosphere. The model proposed here for N2O evolution is based on the hypothesis that NO2− is used as an alternative acceptor for electrons not accepted by O2 during C oxidation for growth by NH3 oxidizers. Rates of N2O evolution simulated using this hypothesis are thereby sensitive to any physical or biological attribute of the soil that controls the demand for, or the supply of, O2 during nitrification, such as substrate concentration, temperature (T) or water content (θ). These rates were compared under a common range of T (10, 20 and 30°C) and θ (0.10, 0.20 and 0.30 m3m−3) to ones reported in the literature that were measured during incubation of an NH4+ amended soil. Simulated rates of N2O evolution reproduced a sensitivity to T and θ that increased with both T and θ, although these rates were overestimated at θ = 0.20 m3m−3. This overestimation is probably caused by uncertainty in parameterizing the model equation in which rates of gas transfer between gaseous and aqueous phases are calculated. Ratios of simulated N2O evolution to NO2− + NO3− production increased with both T and θ through a range of 1–5 × 10−3 μg N2ON μg−1 NO2− + NO3−N in a way that was consistent with ratios of measured evolution to production reported from the NH4+ amended soil as well as with those reported from other soils and pure cultures. As part of the larger ecosystem model ecosys, this model hypothesis will make a useful contribution towards the estimation of N2O evolution from terrestrial ecosystems under different climates and fertilizer managements.