A total of 13 models were used in this study covering a range of temporal scales, spatial complexity, and mechanistic detail (Figure 1). Eight of the models (8) used an hourly time step, four models used a daily time step, and only one model used a monthly time step (i.e., PnET-II). Most models provided estimates for both stand carbon and water budgets, but other were limited to either the carbon (MAESTRA) or water (LINKAGES and NuCM) budgets. The most mechanistically complex models (ecosys, CLASS, LaRS, and CANOAK) also used a complete energy balance. A brief description of each of the models is provided below.
The BGC++ model (Hunt et al. 1999) simulates carbon, nitrogen, and hydrologic cycles for different terrestrial ecosystems. It was derived from BIOME-BGC (for BioGeochemical Cycles; Running and Hunt 1993, Hunt et al. 1996), for the purpose of simulating allocation and growth over long time periods under different climatic conditions. Similar to BIOME-BGC and its predecessor the FOREST-BGC model (Running and Coughlan 1988, Running and Gower 1991), BGC++ uses a dual daily and annual time step, where the hydrologic, photosynthesis and respiration processes are simulated daily and carbon and nitrogen allocation are simulated annually. BGC++ differs from BIOME-BGC in three ways: (1) BGC++ simulates the fluxes of dominate and subdominate lifeforms separately, then combines the fluxes assuming a simple mixture of leaf area index (LAI), (2) allocation of carbon and nitrogen to the leaves, stems, coarse roots, and fine roots is based on an annual time step, and (3) BGC++ uses a new algorithm for estimating incident solar radiation (Winslow et al. 2001).
The key variable for the daily time step is LAI (Hunt et al. 1996). Transpiration is based on the Penman-Monteith equation using absorbed solar radiation. Soil evaporation is based on the transmitted solar radiation. Intercepted water evaporated from the canopy is linearly related to LAI. Using the model of Zheng et al. (1993), LAI strongly controls soil temperature over the year, which in turn controls the rates of root maintenance respiration, litter decomposition, and turnover of soil organic matter. During the annual time step, plant-available nitrogen is calculated from litter decomposition, turnover of soil organic matter, internal nitrogen from retranslocation, and nitrogen losses with water runoff (Hunt et al. 1999). Plant available carbon is the net primary production, and is allocated to new leaves and fine roots based on a soil water index and a soil nitrogen index, such that with high levels of soil moisture and nitrogen, there is preferential allocation to the leaves. Excess plant carbon and nitrogen are allocated to stems and coarse roots based on lifeform, 85% and 15% are allocated to tree stems and grass life forms, respectively. Nitrogen is allocated to leaves independently of carbon, so the leaf nitrogen to carbon ratio is a model prediction, and is used to determine the Vmax of Rubisco and Jmax of the RuBP regeneration for photosynthesis. For predictions regarding the Throughfall Displacement Experiment (TDE), phenology of the leaves and fine roots was handled with constants derived from site observations.
Biome-BGC (v 4.1.1 used in this study) is a general terrestrial ecosystem model designed to simulate the coupling of carbon, nitrogen, and water cycles in the plant-litter-soil system for both woody and herbaceous vegetation (Thornton et al., 2002). All model processes are calculated on a daily timestep. The model is driven by daily surface weather (temperature, precipitation, radiation, and humidity), and requires a set (34) of constants describing each plant functional type (White et al., 2000). All plant, litter, and soil carbon, nitrogen, and water pools and fluxes are entirely prognostic. This is in contrast to models such as Forest-BGC (Running and Coughlan 1988) and CANOAK (Baldocchi and Harley 1995) with prescribed canopy and other biomass pools. In simulations for this study the surface was represented as a single homogeneous deciduous broadleaf canopy, but mixed type simulations are possible (e.g. Law et al. 2001). Canopy carbon and water dynamics are treated separately for sunlit and shaded fractions, with the two fractions having the same mass-based leaf nitrogen concentration and different specific leaf area, a realistic assumption for these forests (Wilson et al. 2000b). Allocation to new growth depends on the availability of both assimilated carbon and soil mineral nitrogen (augmented by an internal pool of retranslocated foliar nitrogen). Plants compete with soil heterotrophs for a single pool of soil mineral nitrogen, with down-regulation of both carbon assimilation and nitrogen immobilizing steps in the trophic processing of litter and soil organic matter when nitrogen is limiting. Individual plant tissues have static C:N ratios, but whole-plant C:N changes as the relative amounts of different tissues change over time (e.g. as wood accumulates relative to foliage during forest stand development). Leaf area phenology for temperate deciduous systems is determined by the model of White et al. (1997).
The model is designed with special attention to the long-term controls on net ecosystem carbon exchange (NEE), and an important aspect of this design is the use of spin-up and perturbation simulations, the purpose of which is to bring model state variables to a configuration that is both internally consistent and consistent with known patterns of land use and disturbance history (Thornton et al. 2002, Law et al. 2001). The spin-up simulation begins with minimal soil organic matter (SOM) content and a nascent canopy, and proceeds until the SOM and plant pools have reached a steady state with respect to a repeated sequence of surface weather drivers. A second step moves this steady state from preindustrial to current conditions for atmospheric CO2 concentration and mineral nitrogen deposition. Overlaid on these simulations are instantaneous modifications to the prognostic model state variables to represent the known historical sequence of land use and/or disturbance history (e.g. removal of mass from plant carbon and nitrogen pools to simulate different levels of harvest or fire). This protocol avoids the inevitable transient responses in NEE that result from forcing a simulation to start with observed plant and soil state variables, since these observed states will never be entirely consistent with the internal dynamics of any particular model.
CANOAK is a one-dimensional, mutli-layer biosphere-atmosphere model that computes water vapor, CO2 and sensible heat flux densities. The model has been described and tested for growing season conditions (Baldocchi and Harley 1995; Baldocchi 1997, and applied to a 20 year climate record (Baldocchi and Wilson 2001). The model consists of coupled micrometeorological and eco-physiological modules. The micrometeorological modules compute leaf (sunlit and shaded) and soil energy exchange, turbulent (Lagrangian) diffusion, scalar concentration profiles and radiative transfer through the canopy using observed meteorological conditions above the canopy. The physiological modules are driven by physiological parameters that are obtained directly from extensive chamber measurements performed in the field. The predicted micrometeorology drives leaf photosynthesis and respiration, stomatal conductance and transpiration at 40 canopy layers. Canopy leaf area profiles were assumed to follow a beta distribution, with a heavier concentration of leaves in the upper canopy (Hutchinson and Baldocchi 1989). NEE in CANOAK is obtained by summing each component of the carbon flux: bole and soil/root respiration, and leaf photosynthesis and respiration.
CANOAK is the only model that does not explicitly simulate soil water content dynamics. As a result drought impacts on modeled physiological processes are only effected by changing atmospheric vapor pressure deficit. This limits the drought sensitivity of CANOAK and is discussed further later in the paper.
The EALCO (Ecological Assimilation of Land and Climate Observation) model was developed to simulate the energy, water, and carbon dynamics of global terrestrial ecosystems through assimilating land and climate observation data (remote sensing and ground based). The model addresses the fundamental physical, physiological, and biogeochemical processes of various land cover types recognized by remote sensing. EALCO is driven by meteorological variables of shortwave and longwave radiation, air temperature and humidity, precipitation, wind speed, and atmospheric pressure. It runs at an hourly time step and the spatial resolution can be determined from remote sensing information. For this study, it is parameterized and driven by in situ measurements.
The canopy energy balance equation and water balance equation are nested and solved simultaneously (Wang et al., 200?a). The water balance equation includes root water uptake, plant water storage change, and canopy transpiration. Stomatal conductance, which is a key parameter in the water transfer scheme, is determined by plant CO2 fixation rates according to Ball et al. (1987). This allows the plant carbon process to be coupled with the energy and water dynamics in the model. The two important prognostic variables, canopy temperature and water potential, are obtained in solving the energy and water balance equations and they are also used in determining other processes in the model, such as the plant CO2 fixation and respiration.
Ecosystem carbon and nitrogen cycles are described by Wang et al., (2001, 200?b). The physiologically active part of plant is divided into foliage, sapwood and fine roots. Calculations of plant include photosynthesis, root nitrogen uptake, substrate carbon/nitrogen transport and their bioconversion into structural materials, respiration, litterfall and root carbon exudation. Transformations of both litterfall and soil organic matter in/on the soil are formulated by first-order kinetics and by separating them into different pools based on the biochemical resistance to microbial decomposition. A separate microbial biomass pool and a mineral N pool are also represented in each of the three soil layers. While litterfall obtained in the plant simulations provides the carbon and nitrogen source to the soil, the organic matter turnover in/on the soil controls the mineral nitrogen release rates and its availability for plant uptake, which determines the plant N conditions and affects the plant CO2 fixation and thereafter the land surface energy and water balances.
A detailed description of the algorithms on which ecosys is based, and of the testing to which they have been subjected, is given in Grant (2001). A general description of those parts of the model most relevant to the study reported here is given below.
Net Primary Productivity
CO2 Fixation. CO2 fixation is calculated in ecosys from coupled algorithms for carboxylation and diffusion. Carboxylation rates are calculated for each leaf surface of multispecific plant canopies according to Farquhar et al. (1980) driven by irradiance (rectangular hyperbola), temperature (Arrhenius kinetics) and CO2 concentration (Michaelis-Menten function). Maximum reaction rates are calculated from specific activities and surficial concentrations of rubisco or chlorophyll determined by CO2, water, N and P uptake during leaf growth.
Diffusion rates are calculated from atmosphere - mesophyll CO2 concentration differences multiplied by leaf stomatal conductance (Grant et al. 1999b). Conductance is determined by leaf carboxylation rate and canopy turgor.
Transpiration. Canopy turgor is generated from a convergence solution for canopy water potential at which the difference between transpiration and root water uptake equals the difference between canopy water contents at the water potentials of the previous and current time steps (Grant et al. 1999b). Canopy transpiration is solved from a first order solution to the canopy energy balance (Grant et al. 1999b). Root water uptake is solved from soil-root-canopy water potential gradients and root hydraulic conductances calculated from a C-driven root growth submodel (Grant 1998a).
Autotrophic Respiration and Senescence. The product of CO2 fixation is added to a pool of stored C for each branch of each plant species from which C is oxidized to meet maintenance respiration requirements (Grant et al. 1999b). If the C oxidation rate is less than the maintenance respiration requirement, the difference is made up by respiration of remobilizable C in leaves and twigs. Upon exhaustion of remobilizable C, the remaining structural C is dropped from the branch as litterfall and added to residue at the soil surface where it undergoes heterotrophic decomposition. When the C oxidation rate exceeds maintenance respiration, the excess is used for growth respiration that drives the formation of new biomass (Grant et al. 1999b).
Nutrient Uptake. Nutrient (N and P) uptake is calculated for each plant species by solving for aqueous concentrations of NH4+, NO3- and H2PO4- at root and mycorrhizal surfaces at which radial transport by mass flow and diffusion from the soil solution to the surfaces equals active uptake by the surfaces (Grant and Robertson 1997, Grant 1998a). Soil nutrient transformations control the aqueous concentrations of NH4+, NO3- and H2PO4- in each soil layer through thermodynamically driven precipitation, adsorption and ion pairing reactions (Grant and Heaney 1997), convective-dispersive solute transport (Grant and Heaney 1997), and microbial mineralization-immobilization (Grant et al. 1993a). Active uptake is calculated from length densities and surface areas (Itoh and Barber 1983) given by a root and mycorrhizal growth submodel (Grant 1993a, 1998a; Grant and Robertson 1997). The products of N and P uptake are stored in root and mycorrhizal pools from which they are combined with stored C when driven by growth respiration to form new plant biomass. Plant species designated as legumes in the model also reduce aqueous N2 to stored N through oxidation of stored C according to the energetics of Schubert (1982).
Plant Growth. Growth respiration drives expansive growth of vegetative and reproductive organs through mobilization of stored C, N and P according to phenology-dependent partitioning coefficients and biochemically based growth yields. This growth is used to simulate the lengths, areas and volumes of individual internodes, twigs and leaves (Grant 1994a) from which heights and areas of leaf and stem surfaces are calculated for irradiance interception and aerodynamic conductance algorithms used in energy balance calculations. Growth respiration also drives extension of primary and secondary root axes and of mycorrhizal axes through mobilization of stored C, N and P (Grant 1993, 1998a). This growth is used to calculate lengths and areas of root and mycorrhizal axes from which root uptake of water (Grant et al. 1999b) and nutrients (Grant and Robertson 1997) is calculated.
The growth of different branch organs and root axes in the model depends upon transfers of stored C, N and P among branches, roots and mycorrhizae. These transfers are driven by concentration gradients within the plant which develop from different rates of C, N or P acquisition and consumption by its branches, roots and mycorrhizae (Grant 1998a). The model thus implements the functional equilibrium between roots and shoots proposed by Thornley (1995).
Decomposition. Soil organic matter in ecosys is resolved into four substrate-microbe complexes (plant residue, animal manure, particulate organic matter and non-particulate organic matter) within each of which C, N and P may move among five organic states: solid substrate, sorbed substrate, soluble hydrolysis products, microbial communities, and microbial residues (Grant 1999, Table 1). Each organic state in each complex is resolved into structural components of differing vulnerability to hydrolysis and further resolved into elemental fractions C, N and P within each structural component. Microbial communities are also resolved into functional type including obligate aerobes, facultative anaerobes (denitrifiers), obligate anaerobes (fermenters), methanogens and diazotrophs.
Litterfall is added to the plant residue complex and partitioned into carbohydrate, protein, cellulose and lignin structural components according to Trofymow et al. (1995). Rates of substrate hydrolysis are the product of the active biomass and specific activity of each microbial functional type within each complex (Grant et al. 1993a; Grant and Rochette 1994). Specific activity is constrained by substrate-microbe density relationships (Grant et al., 1993a; Grant and Rochette, 1994), and by the temperatures and O2 concentrations of surface residue and a vertically resolved soil profile (Grant 1997; Grant and Rochette 1994; Grant et al. 1998). A fraction of the hydrolysis products of lignin is coupled with those of protein and carbohydrate according to the stoichiometry proposed by Shulten and Schnitzer (1997) and the resulting compound is transferred to the particulate organic matter complex.
Microbial Growth. The concentration of the soluble hydrolysis products determines rates of C oxidation by each heterotrophic population, the total of which drives CO2 emission from the soil surface. This oxidation is coupled to the reduction of O2 by all aerobic populations (Grant et al. 1993a, Grant and Rochette 1994), to the sequential reduction of NO3-, NO2- and N2O by heterotrophic denitrifiers (Grant et al. 1993b; Grant and Pattey 1999) and to the reduction of organic C by fermenters and of acetate by heterotrophic methanogens (Grant 1998b). The energetics of these oxidation-reduction reactions determine growth yields and hence the active biomass of each heterotrophic functional type from which its decomposer activity is calculated. In addition, autotrophic nitrifiers conduct NH4+ and NO2- oxidation (Grant 1994b) and N2O evolution (Grant 1995), and autotrophic methanotrophs conduct CH4 oxidation (Grant 1999), the energetics of which determine autotrophic growth yields and hence biomass and activity. Each microbial population in the model seeks to maintain steady-state ratios of biomass C:N:P by mineralizing or immobilizing NH4+, NO3- and H2PO4-, thereby regulating solution concentrations that drive N and P uptake by roots and mycorrhizae. Microbial populations undergo first order decomposition, products of which are partitioned between microbial residues and the nonparticulate organic matter complex according to soil clay content (Grant et al. 1993a). All microbial reactants and products undergo convective-dispersive transport through, and volatilization-dissolution transfer between, aqueous and gaseous soil phases. Parameterization of soil gas transport algorithms in progress the time of the model intercomparison affected short-term soil CO2 emission rates in the model.
INTRASTAND is an hourly time step model designed for use in the interpolation of measured physiological data over time for the calculation of daily and intra-annual forest stand carbon and water budgets. The model structure contains three canopy foliage layers, branch and bole stem components, 4 soil layers and stem capacitance. The structure is designed so that it doesn’t exceed the availability of measured input data. Carbon uptake is based on the coupled Farquhar/Ball-Berry photosynthetic and stomatal conductance model as described by Harley et al. (1992) , a modified version of the water budget model PROSPER (Huff et al. 1977) with the inclusion of a stem capacitance, a model of stem respiration (Edwards and Hanson 1996), and a model of forest floor CO2 efflux (Hanson et al. 1993, 2003a). The canopy is divided into three layers of equal LAI and diffuse and direct light penetration is calculated as suggested by Norman (1982). The modeled canopy net foliar assimilation rates were calibrated to yield maximum assimilation in early June following full leaf expansion in accordance with reported values for the upland oak forest on Walker Branch watershed (Baldocchi and Vogel 1996; Baldocchi 1997; Harley and Baldocchi 1995; Verma et al. 1986). Isoprene emission was also estimated for this stand (a small component of carbon flux) based on the observations and models of Harley et al. (1997). The model is coded using "Stella” modeling software (High Performance Systems, Hanover, NH). The combined mechanistic model was developed for application to intra-annual carbon and water budgeting but has been used here over a multi-year period by transferring soil water, litter mass and stored carbohydrate data to the initial conditions of model runs for subsequent years. Growth of stems, roots, and leaves, the timing of leaf out and leaf senescence, climate variables, and required physiological variables are inputs from direct measurements on the study site. Respiratory costs of growing tissues were calculated by a modified Penning de Vries approach as outlined by Amthor (1996). Published results for the water budget and carbon flux components can be found in Edwards and Hanson (1996), Hanson et al. (1998, 2001, 2003a, 2003b) and Johnson et al. (2002). Results presented in this manuscript represent the first comprehensive test of its utility against a multiyear data set and other models.
LaRS is an hourly time step model designed to simulate tree physiology and growth within the context of a forest ecosystem. Solar radiation-forest interactions are simulated with a ray tracing procedure that accounts for diffuse solar (sky) radiation arising from 10 elevation bands and for direct beam radiation (based on solar elevation). Solar radiation is divided into PAR and NIR wave bands, with differing optical properties of canopy elements specified for each wave band. The canopy is divided into multiple horizontal layers, and radiation from each source (moving toward the ground or, after reflection, toward the sky) is either absorbed by, reflected by, or transmitted through leaves, branches, and boles in each canopy layer. The vertical distribution of leaf and branch/bole ‘clumping’ is accounted for. The forest floor also absorbs and reflects solar radiation. During rain, water is stored on leaf and branch/bole surfaces based on rain amount and the surface areas of leaves and branches/boles. That water is then evaporated during energy balance calculations in subsequent hours.
Leaf growth is initiated in the spring based on ‘temperature sums’ and daylength. Leaf growth rate during each hour (for the period from leaf bud burst to leaf physiological maturity) is based on temperature and availability of substrate (i.e., stored and current photosynthate). Maximum leaf area index and structural leaf mass per unit leaf area for a site are inputs to the model. Once leaves reach maturity (based on a spring/summer temperature sum), they mobilize nitrogen continuously, with the rate of mobilization based on temperature and soil water content. Autumnal leaf senescence is based on temperature and day length.
Leaf temperature (from an energy balance), photosynthesis, maintenance respiration, and phloem translocation are simulated for sunlit and shaded leaves in each canopy layer. Transpiration is an output of the energy balance calculations. Leaf photosynthetic rate is calculated from a light-response curve defined by the quantum efficiency and photosynthetic potential (i.e., maximum, light-saturated photosynthesis). Quantum efficiency is defined by intercellular CO2 level, leaf temperature, and specificity of ribulose 1,5-bisphosphate carboxylase/oxygenase (rubisco) for CO2 (Farquhar and von Caemmerer, 1982). Photosynthetic potential is defined by leaf nitrogen content (which is distributed among leaves according to vertical position in the canopy), CO2 level, and a species-dependent empirical parameter. Photorespiration is calculated from photosynthetic rate and rubisco specificity for CO2. Leaf isoprene emissions are also calculated, based on absorbed PAR and leaf temperature. Most canopy physiological processes are estimated iteratively (see Amthor et al. 1994 for the same approach applied to a big-leaf canopy).
Bole and branch growth are based on time of the year (linked to timing of leaf growth), temperature, and substrate availability. Respiratory costs of leaf growth, leaf maintenance, and phloem loading are estimated according to standard principles (Amthor, 2000). Wood maintenance respiration is a function of temperature. Herbivores consume leaf tissue and convert it to CO2 based on specified base rates and temperature.
Physically based equations are used to transfer heat and water (liquid and vapor) vertically between adjacent layers in the soil, and between the air above the forest floor and the top soil layer. The physical properties of the soil (clay, silt, sand, and rock fractions) in each layer are used to define thermal and hydraulic characteristics in that layer. Fine roots absorb or release (through ‘hydraulic lift’) water in soil layers containing roots (layers containing roots are specified as input). Net water uptake by the root system is dictated by canopy transpiration during each hour (i.e., no water storage in the plant is accounted for). Water moving horizontally through the soil is lost to streams, and water moving down out of the bottom soil layer is considered deep drainage. Root growth in each soil layer containing roots is calculated from soil temperature, fine root density in the layer (amount of root per volume of soil), water content of the layer, and substrate availability. Respiration supporting root growth, maintenance (a function of soil temperature), and ion uptake from the soil solution are calculated each hour. Fine root death (turnover) rate in each layer is based on temperature and soil water content.
Death of bole, stump, and branches contribute to a coarse woody debris pool. That pool decomposes each hour based on temperature and moisture in the top soil layer. Decomposition of soil organic matter and the forest floor is based on an empirical rate specified for the site. That rate is in turn modulated by temperature and moisture of the top two soil layers.
LINKAGES v2.1 (Wullschleger et al. 2003) is derived from LINKAGES (Pastor and Post, 1985) to study the effects of climate change (i.e., temperature and precipitation) and inter- and intra-annual variations in climate on long-term forest dynamics. LINKAGES v2.1 was modified to incorporate a more physiology-based representation of plant and soil controls on potential and actual evapotranspiration over that found in original LINKAGES. Modifications include replacing the Thornth-waite and Mather (1957) monthly calculation of potential evapotranspiration with a daily scheme in which evaporation from the soil surface and canopy transpiration are treated separately (Shuttleworth and Wallace, 1985). A maximum leaf conductance to water vapor is specified for the stand and modified according to daily radiation, temperature, vapor pressure deficit and extractable soil water. Interception losses are determined for the canopy based on leaf area and stem area index (Federer, 1995). Multiple soil layers (12) have been added to the model with water required for transpiration and evaporation extracted sequentially from each layer. Seedlings, saplings and mature trees occupy specific soil layers and size classes thus experience differential stress during the season. Reductions in diameter increment due to drought are accomplished via the concept of “stress days” whereby days on which soil water content falls below the permanent wilting point are accumulated. Stress days are weighted according to the time of year the drought occurs. LINKAGES v2.1 retains all other components of the original LINKAGES model, which was based on the individual tree model FORET (Shugart and West, 1977). Particularly, LINK-AGES v2.1 retains litter production, decomposition and associated nitrogen dynamics similar to those in the FORTNITE model (Aber and Melillo, 1982). Model predictions of species composition, basal area and stems ha-1 generated by LINKAGES v2.1 have recently been validated for the Walker Branch Watershed study site (Bugmann et al., 2001).
Although LINKAGES provides estimates of forest wood increment by species, components of the carbon budget are not provided. Outputs for LINKAGES v2.1 in the current paper are limited to the water budget comparsions.
LoTEC (Local Terrestrial Ecosystem Carbon) is the ecosystem C cycle model implemented in each grid cell of the global model GTEC 2.0 (Global Terrestrial Ecosystem Carbon). It describes C and water dynamics of local, homogeneous vegetation stands at scales of several square meters to perhaps a hectare. It is a generic ecosystem simulator, with no features specific to boreal forests. LoTEC litter and soil C dynamics are a modification of the Rothamsted model (Jenkinson, 1990) used in GTEC 1.0 (King et al., 1997; Post et al., 1997). The statistical NPP model in GTEC 1.0 (Lieth, 1975) was replaced with a process-based model including big-leaf canopy physiology and plant growth/senescence. Hourly simulations of big-leaf CO2 and water vapor fluxes are used, but plant growth and soil C dynamics are modeled with daily time steps. LoTEC predicts PG, plant growth, plant growth and maintenance respiration, litter production, decomposition, transpiration and precipitation interception losses, and soil water balance.
MAESTRA is a three dimensional model of forest canopy radiation absorption, photosynthesis, and transpiration. The canopy is represented by an array of tree crowns whose positions and dimensions are specified. Each crown is divided into 6 horizontal layers with each layer divided into 12 grid points of equal volume. Each layer is specified by a number of physical and physiological properties, including radiation, temperature, leaf area index, and leaf nitrogen content. Radiation absorption is calculated for a ‘target crown’ in the canopy. Positions and dimensions of trees surrounding the target crown are used to determine the amount of radiation incident on the target crown after passing through the neighboring crowns. Radiation penetration to each grid point is calculated for three wavebands (i.e. PAR, near infrared and thermal radiation), including consideration of direct, diffuse and scattered radiation. Photosynthesis and transpiration at each grid point are calculated from the absorbed radiation. Leaf photosynthesis is estimated by the Farquhar photosynthesis model (Farquhar et al. 1980) coupled to the Ball-Berry stomatal conductance model (Ball et al. 1987). Transpiration is calculated by applying the Penman-Monteith formula (Jarvis and McNaughton 1986) to each grid point. Environmental variables driving model simulations are radiation, air temperature, air humidity, wind speed and atmospheric CO2 concentration above the canopy. The model assumes that air humidity, temperature, and CO2 concentration are uniformly distributed within the canopy. The time step in the model simulation is one hour.
The model has previously been applied to study canopy carbon and water fluxes of Picea sitchensis (Wang and Jarvis 1990), Pinus radiata (McMurtrie and Wang 1993), Betula pendula (Wang et al. 1998), and Pinus taeda (Luo et al. 2001). More details of the model description can be found in Wang and Jarvis (1990) and Luo et al. (2001).
MAESTRA does not attempt to simulate belowground root or decomposition processes or soil water depletion, and like CANOAK has a limited ability to capture the response of ecosystems to drought.
The NuCM model was designed by a team of investigators in the Integrated Forest Study (see Johnson and Lindberg, 1992) and the code was written by Tetra-Tech, Inc (Liu et al., 1992). NuCM depicts the cycling of N, P, K, Ca, and Mg at a stand level but also includes the fluxes of major cations (H+, NH4+, Ca2+, Mg2+, K+, Na+) anions (NO3-, SO42-, ortho-phosphate, Cl-, HCO3-, organic anion), and Si in precipitation, throughfall, and soil solution. Because NuCM was designed primarily for simulating the effects of atmospheric deposition on nutrient cycling processes, its construction emphasizes soil and soil solution chemistry (Liu et al., 1992). The ecosystem is represented as a series of vegetation and soil components. The overstory consists of one generic conifer and one generic deciduous species of specified biomass and nutrient concentration (foliage, branch, bole, roots). For mixed-species stands, average values for biomass and nutrient concentration by component must be used. NuCM also allows an understory that can be divided into canopy, bole, and roots. Maximum potential growth in the model is defined by the user and is constrained in the model by the availability of nutrients and moisture. The forest floor is simulated from litterfall inputs and litter decay. Litterfall mass inputs are defined by the user, and litter decay is represented as a four-stage process where (1) litter decays to fine litter, (2) fine litter decays to humus and cations, (3) humus decays to organic acids, NH4+, SO42-, H+, and CO2, and 4) organic acids decay to NH4+, SO42-, H+, and CO2. The user defines bulk density, cation exchange capacity, exchangeable cations, adsorbed phosphate and sulfate, and four soil minerals and their composition. These inputs define the initial soil exchangeable/adsorbed pools and total pools. Initial total soil N pools are simulated from litterfall and decay, as described above, and user-defined C/N ratios. Vegetation, litter, and soil pools change over a simulation in response to growth, litterfall and decomposition, and nutrient fluxes via deposition, leaching and weathering, as described below.
The processes which govern interactions among these pools include translocation, uptake, foliar exudation and leaching, organic matter decay, nitrification, anion adsorption, cation exchange and mineral weathering. Translocation, defined as the removal of nutrients from foliage prior to litterfall, is user-specified (as a percentage of foliage nutrient content). Maximum uptake is calculated from biomass and nutrient concentrations; actual uptake is equal to this maximum value when sufficient nutrients are available and reduced when nutrients become limiting. Reduced uptake first allows reduced nutrient concentrations in plant tissues (by a user-specified percentage) then causes a reduction in growth. Foliar exudation and leaching rates are simulated in the model as proportional to foliar concentrations and user-defined coefficients.
Precipitation in the NuCM model is defined by input meteorological files (typically 1 to 5 years long) which are repeated in order to generate long-term simulations. The meteorological file contains daily values for precipitation quantity (cm); maximum air temperature (oC); minimum air temperature (oC); cloud cover (decimal fraction); dewpoint (oC); atmospheric pressure (mbars); and wind speed (m sec-1). Monthly soil temperatures are set in a separate file and can also be entered on a menu. Precipitation is routed through the canopy and soil layers and evapotranspiration, deep seepage, and lateral flow are simulated. Potential evapotranspiration (ETp) is calculated as:
ETp = (Fet/n)(TmCeHc)
where Fet = the evapotranspiration factor, which is a function of latitude (r) (Hargreaves, 1974), Tm = mean ambient daily temperature (oF); Hc = a humidity correction factor; and Ce = a calibration factor. The movement of water through the system is simulated using the continuity equation, Darcy's equation for permeable media flow, and Manning's equation for free surface flow. Percolation occurs between layers as a function of layer permeabilities and differences in moisture content. Nutrient pools associated with soil solution, the ion exchange complex, minerals, and soil organic matter are all tracked explicitly by NuCM. Wet deposition is calculated from precipitation amounts and user-input air quality files which define precipitation concentrations on a monthly basis. Dry deposition is calculated from air concentrations in the air quality files combined with user-defined deposition velocities and simulated leaf areas. Leaching is calculated from soil water flux and simulated soil solution concentrations using the soil chemical and biological algorithms defined above for each soil horizon.
The only processes in the NuCM model that are explicitly temperature-dependent are ET, occurrence of precipitation as rainfall versus snowfall, snowpack melting, litter decay, and nitrification. Temperature affects processes such as cation exchange, mineral weathering, uptake, etc., only indirectly. Precipitation effects are manifested strictly through the hydrologic simulations; none of the nutrient processes are explicitly dependent upon soil water content . Although NuCM executes calculations daily, only annual water cycle outputs are available for comparison with other models. Predictions of annual nutrient availability and flux available from the NuCM model (Johnson et al. 2002, 2003) are not presented here because they have no analogue from the other 12 models.
The PnET-II model was originally developed for studying forest ecosystem processes in northern forests (Aber and Federer, 1992). It is a lumped-parameter, monthly-time-step, and stand-level model that describes carbon and water dynamics in mature forests. It simulates both carbon and water cycles in a forest ecosystem using simplified algorithms that describe key biological and hydrologic processes. This model has been validated with field data from northern deciduous upland hardwood forests (Aber et al. 1995; Aber et al. 1996) and southern pine forests (McNulty et al. 1996; Sun et al. 2000), and it has been applied at a regional scale to study the potential effects of climatic change on U.S. forests (U.S. Global Change Research Program 2000).
Input parameters for vegetation, soil and site locations, and climate may be derived from the literature or measured from a local study site. Stand level vegetation parameters include those regulating the physiological and physical processes such as photosynthesis, light attenuation, foliar nitrogen concentration, plant and soil respiration, and rainfall interception. Only one soil parameter, soil water holding capacity (field capacity in percentage ¥ rooting depth), is required. Climate input variables include minimum and maximum monthly air temperature, total monthly photosynthetic active radiation (PAR), and total monthly precipitation.
The model simulates the carbon cycle by tracking absorbed carbon during photosynthesis, allocation to foliage, wood, and root, and respiration from leaf, stem and roots. PnET calculates the maximum amount of leaf-area which can be supported on a site based on the soil, the climate and parameters specified for the vegetative type. The model assumes that leaf area is equal to the maximum amount of foliage that could be supported due to soil water-holding capacity, species, and climate limitations. Predicted NPP equals total gross photosynthesis minus growth and maintenance respiration for leaf, wood, and root compartments. PnET calculates respiration as a function of the current month's minimum and maximum air temperature. Changes in water availability and plant water demand also place limitations on leaf area produced, so total leaf area decreased as vapor pressure deficit and air temperature increased above optimal levels. Reduced leaf area decreased total carbon fixation and altered ecosystem hydrology. The hydrologic cycle is simulated by the water balance equation. The input component of soil water storage is represented by net precipitation (i.e., precipitation - canopy interception), and outputs consist of canopy interception, plant transpiration, fast or macro-pore flow representing water not available for extraction by plant roots, and lateral and deep drainage. Soil evaporation is neglected in fully stocked forest ecosystems. Evapotranspiration is defined as the sum of plant transpiration and canopy interception. The model assumes that water that is not subjected to evapotranspiration eventually flows to streams as runoff. Transpiration is directly linked to forest photosynthesis and forest carbon gain processes by modeling transpiration as a function of water use efficiency and vapor pressure deficit. Therefore, PnET-II closely integrates forest hydrology with the biological processes.
The Soil-Plant-Atmosphere model (SPA, Williams et. al 1996) is a process-based model that simulates ecosystem photosynthesis and water balance at fine temporal and spatial scales (30 minute time-step, ten canopy and soil layers). The scale of parametrization (leaf-level) and prediction (canopy level) was designed to allow the model to diagnose eddy flux data, and to provide a tool for scaling up leaf level processes to canopy and landscape scales (Williams et al. 2001).
The model employs a detailed radiative transfer scheme that determines the time-varying transmittance, reflectance and absorption of longwave, near infra-red and direct and diffuse photosynthetically active radiation (PAR) by canopy layers and the soil surface. Absorption of PAR in each canopy layer is partitioned between sunlit and shaded foliage fractions. The SPA model employs some well tested theoretical representations of eco-physiological processes, such as the Farquhar model of leaf-level photosynthesis (Farquhar & von Caemmerer 1982), and the Penman-Monteith equation to determine leaf-level transpiration. These two processes are linked by a novel model of stomatal conductance that optimizes daily carbon (C) gain per unit leaf nitrogen (N), within the limitations of canopy water storage and soil-to-canopy water transport. The maximum flux rate of water through vegetation is determined by the difference between soil water potential and the minimum sustainable leaf water potential, and by the hydraulic resistance of the soil-root-leaf pathway. Stomata adjust to equalize evaporative losses with the maximum hydraulic supply, minimizing the risk of cavitation.
The model contains a detailed representation of soil hydrology and thermal dynamics. From the estimated transmission of radiation through the canopy, we determine the down-welling radiation at the soil surface. We then solve the surface energy balance by estimating the soil surface temperature, and partition net radiation into sensible, latent and ground heat fluxes. The soil is divided into 10 layers of varying thickness, each with a specified organic matter and mineral content. The flux of heat through the soil profile is determined on the basis of the ground heat flux, the thermal gradient between soil layers, the soil thermal conductivity and thermal heat content of each layer. The thermal parameters are dependent on soil organic matter and mineral fractions and soil water content, and phase transitions between liquid water and ice. The field capacity of each layer is determined according to soil texture and soil water retention curves. Heat is redistributed through the soil profile according to water movement, and from patterns of freezing and thawing we determine the ice content of each soil layer daily. Root water uptake is explicitly linked with soil water potential and soil hydraulic conductivity, as determined from soil water retention curves, through the plant hydraulic model outlined above. Roots are distributed through the upper soil layers and water is withdrawn from the layers with greatest moisture content. Precipitation inputs to soils are calculated after canopy interception, drainage and evaporation from the canopy water store, and infiltration through the soil surface. A snow sub-model tracks the dynamics of the snow-pack and its effects on soil temperature.
The model is readily applied to different ecosystems as there are relatively few parameters to be changed. The most critical are LAI and foliar N (accounting for phenological changes), plant hydraulic conductance, minimum leaf water potential, rooting depth, and soil texture. The SPA model has been applied in ecosystems ranging from 70oN to 2oS. The SPA model has been extensively tested against independent eddy covariance data for the temperate oak-maple forest at the Harvard Forest (Wofsy et al. 1993; Williams et al. 1996). The model was able to explain more than 90% of the variability in measured daily gross primary productivity at Harvard Forest (Williams et al. 1997). The version of SPA used in this intercomparison provides complete water cycle data and data for annual gross primary production, but does not estimate ecosystem respiration components.
Aber J.D., and J.M. Melillo. 1982. FORTNITE: a computer model of organic matter and nitrogen dynamics in forest ecosystems. Univ. Wisconsin Res. Bull. R3130.
Aber, J.D., and C.A. Federer. 1992. A generalized, lumped-parameter model of photosynthesis, evapotranspiration, and net primary production in temperate and boreal forest ecosystems. Oecologia 92:463-474.
Aber, J.D., S.V. Ollinger, C.A. Feder, P.B. Reich, M.L. Goulden, D.W. Kicklighter, J.M. Mello, and R.G. Lathrop, Jr. 1995. Predicting the effects of climate change on water yield and forest production in northeastern U.S. Climate Research 5:207-222.
Aber, J.D., P.B. Reich, and M.L. Goulden. 1996. Extrapolating leaf CO2 exchange to the canopy: a generalized model of forest photosynthesis validated by Eddy correlation. Oecologia 106:257-265.
Amthor, J.S. 1996. Models of plant growth costs and their application to trees in Walker Branch Watershed. UCRL-AR-125547. Lawrence Livermore National Laboratory, Livermore, California.
Amthor, J.S. 2000. The McCree—de Wit—Penning de Vries—Thornley respiration paradigms: 30 years later. Annals of Botany 86:1-20.
Amthor, J.S., M.L. Goulden, J.W. Munger, and S.C. Wofsy. 1994. Testing a mechanistic model of forest-canopy mass and energy exchange using eddy correlation: carbon dioxide and ozone uptake by a mixed oak—maple stand. Australian Journal of Plant Physiology 21:623-651.
Baldocchi, D.D. 1997. Measuring and modeling carbon dioxide and water vapour exchange over a temperate broad-leafed forest during the 1995 summer drought. Plant, Cell and Environment 20:1108-1122.
Baldocchi, D.D., and P.C. Harley. 1995. Scaling carbon dioxide and water vapour exchange from leaf to canopy in a deciduous forest II. Model testing and applications. Plant, Cell and Environment 18:1331-1340.
Baldochhi, D.D., and C.A. Vogel. 1996. Energy and CO2 flux densities above and below a temperate broad-leaved forest and a boreal pine forest. Tree Physiology 16:5-16.
Baldocchi D.D., and K.B. Wilson. 2001. Modelling CO2 and water vapour exchange of a temperate broadleaved forest on daily to decadal time scales. Ecological Modelling 142:155-184.
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, pp. 221-224 In: Biggins, I. (ed.), Progress in Photosynthesis Research. Vol 1V.5, Proceedings of the VII International Photosynthesis Congress. Martinus Nijhoff, Dordrecht, Netherlands.
Bugmann, H.K.M., S.D. Wullschleger, D.T. Price, K. Ogle, D.F. Clark, and A.M. Solomon. 2001. Comparing the performance of forest gap models in North America. Climatic Change (in press).
Edwards N.T., and P.J. Hanson. 1996. Stem respiration in a closed-canopy upland oak stand. Tree Physiology 16:433 –439.
Farquhar, G.D., and S. von Caemmerer. 1982. Modelling of photosynthetic response to environmental conditions, pp. 549-587 In: Lange, O.L., P.S. Nobel, C.B. Osmond, and H. Ziegler (eds.), Encyclopedia of Plant Physiology, NS vol. 12B: Physiological Plant Ecology II. Springer-Verlag, New York.
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.
Federer, C.A. 1995. BROOK90: a simulation model for evaporation, soil water, and streamflow, Version 3.1. Computer freeware and documentation. USDA Forest Service, PO Box 640, Durham, NH, 03824.
Grant, R.F. 1993. Simulation model of soil compaction and root growth. I. Model development. Plant and Soil 150:1-14.
Grant, R.F. 1994a. Simulation of competition between barley (Hordeum vulgare L.) and wild oat (Avena fatua L.) under different managements and climates, Ecological Modelling 71:269-287.
Grant, R.F. 1994b. Simulation of ecological controls on nitrification. Soil Biology & Biochemistry 26:305-315.
Grant, R.F. 1995. Mathematical modelling of nitrous oxide evolution during nitrification. Soil Biology & Biochemistry 27:1117-1125.
Grant, R.F. 1997. Changes in soil organic matter under different tillage and rotation: mathematical modelling in ecosys. Soil Science Society of America Journal 61:1159-1174.
Grant, R.F. 1998a. Simulation in ecosys of root growth response to contrasting soil water and nitrogen. Ecological Modelling 107:237-264.
Grant, R.F. 1998b. Simulation of methanogenesis in the mathematical model ecosys. Soil Biology & Biochemistry30:883-896.
Grant, R.F. 1999. Simulation of methanotrophy in the mathematical model ecosys. Soil Biology & Biochemistry 31:287-297.
Grant, R.F. 2001. A review of the Canadian ecosystem model ecosys, pp. 175–264 In: Shaffer, M. (ed.), Modeling Carbon and Nitrogen Dynamics for Soil Management. CRC Press, Boca Raton, Florida.
Grant, R.F., N.G. Juma, and W.B. McGill. 1993a. Simulation of carbon and nitrogen transformations in soils. I. Mineralization. Soil Biology & Biochemistry 27:1317-1329.
Grant, R.F., M. Nyborg, and J. Laidlaw. 1993b. Evolution of nitrous oxide from soil: I. Model development. Soil Science 156:259-265.
Grant, R.F., and P. Rochette. 1994. Soil microbial respiration at different temperatures and water potentials: Theory and mathematical modelling. Soil Science Society of America Journal 58:1681-1690.
Grant, R.F., D.J. Heaney. 1997. Inorganic phosphorus transformation and transport in soils: mathematical modelling in ecosys. Soil Science Society of America Journal 61:752-764.
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., R.C. Izaurralde, M. Nyborg, S.S. Malhi, E.D. Solberg, and D. Jans-Hammermeister. 1998. Modelling tillage and surface detritus effects on soil C storage under current vs. elevated CO2 and temperature in ecosys, pp. 527-547 In: Lal R., J.M. Kimble, R.F. Follet, and B.A. Stewart (eds.). Soil Processes and the Carbon Cycle. CRC Press. Boca Raton, Florida.
Grant, R.F., and E. Pattey. 1999. Mathematical modelling of nitrous oxide emissions from an agricultural field during spring thaw. Global Biogeochemical Cycles 13:679-694.
Grant, R.F., G.W. Wall, K.F.A. Frumau, P.J. Pinter, Jr., D. Hunsaker, B.A. Kimball, and R.L. LaMorte. 1999b. Crop Water Relations under Different CO2 and Irrigation: Testing of ecosys with the Free Air CO2 Enrichment (FACE) Experiment. Agricultural & Forest Meteorology 95:27-51.
Hanson, P.J., S.D. Wullschleger, S.A. Bohlman, and D.E. Todd. 1993. Seasonal and topographic patterns of forest floor CO2 efflux from an upland oak forest. Tree Physiology 13:1-15.
Hanson, P.J., D.E. Todd, M.A. Huston, J.D. Joslin, J. Croker, and R.M. Augé. 1998. Description and field performance of the Walker Branch Throughfall Displacement Experiment: 1993-1996. ORNL/TM-13586, Oak Ridge National Laboratory, Oak Ridge, Tennessee.
Hanson, P.J., D.E. Todd, and J.S. Amthor. 2001. A six year study of sapling and large-tree growth and mortality responses to natural and induced variability in precipitation and throughfall. Tree Physiology 21:345-358.
Hanson, P.J., E.G. O’Neill, M.L.S. Chambers, J.S. Riggs, J.D. Joslin, and M.E. Wolfe. 2003a. Soil respiration and litter decomposition. Chapter 10 in Hanson, P.J., and S.D. Wullschleger (eds.), North American Temperate Deciduous Forest Responses to Changing Precipitation Regimes, Springer, New York (in press).
Hanson, P.J., N.T. Edwards, T.J. Tschaplinski, S.D. Wullschleger, and J.D. Joslin. 2003b. Estimating the net primary and net ecosystem production of a southeastern upland Quercus forest from an 8-year biometric record. Chapter 22 in in Hanson, P.J., and S.D. Wullschleger (eds.), North American Temperate Deciduous Forest Responses to Changing Precipitation Regimes, Springer, New York (in press).
Harley, P.C. and D.D. Baldocchi. 1995. Scaling carbon dioxide and water vapor exchange from leaf to canopy in a deciduous forest: leaf level parameterization. Plant, Cell and Environment 18:1146-1156.
Harley, P.C., R.B. Thomas, J.F. Reynolds, and B.R. Strain. 1992. Modelling photosynthesis of cotton grown in elevated CO2. Plant, Cell and Environment 15:271-282.
Harley, P., A. Guenther, and P. Zimmerman. 1997. Environmental controls over isoprene emission in deciduous oak canopies. Tree Physiology 17:705–714.
Huff, D.D., R.J. Luxmoore, J.B. Mankin, and C.L. Begovich. 1977. TEHM: a terrestrial ecosystem hydrology model. International Biological Program Technical Publication, EDFB/IBP-76/8, ORNL/NSF/EATC-27, Oak Ridge National Laboratory, Oak Ridge, Tennessee.
Hunt, E.R., Jr., M.B. Lavigne, and S.E. Franklin. 1999. Factors controlling the decline of net primary production with stand age for balsam fir in Newfoundland assessed using an ecosystem simulation model. Ecological Modelling 122:151-164.
Hunt, E.R., Jr., S.C. Piper, R. Nemani, C.D. Keeling, R.D. Otto, and S.W. Running. 1996. Global net carbon exchange and intra-annual atmospheric CO2 concentrations predicted by an ecosystem process model and three-dimensional atmospheric transport model. Global Biogeochemical Cycles 10:431-456.
Hutchinson, B.A., and D.D. Baldocchi. 1989. Forest meteorology, pp. 21-95 In Johnson D.W., and R.I. VanHook (eds.), Analysis of Biogeochemical Cycling Processes in Walker Branch Watershed, Springer-Verlag, New York.
Itoh, S., and S.A. Barber. 1983. Phosphorus uptake by six plant species as related to root hairs. Agronomy Journal 75:457-461.
Jarvis, P.G., and K.G. McNaughton. 1986. Stomatal control of transpiration: scaling up from leaf to region. Advances in Ecological Research 15:1-49.
Jenkinson, D.S. 1990. The turnover of organic carbon and nitrogen in soil. Philosophical Transactions of the Royal Society of London, Series B 329:361-368.
Johnson, D.W., and S.E. Lindberg SE (eds.). 1992. Atmospheric Deposition and Forest Nutrient Cycling. Springer-Verlag, New York.
Johnson, D.W., P.J. Hanson, and D.E. Todd. 2002. The effects of throughfall manipulation on soil leaching in a deciduous forest. Journal of Environmental Quality 31:204-216.
Johnson, D.W., P.J. Hanson, and D.E. Todd. 2003. Nutrient availability and cycling. Chapter 23 in Hanson P.J., and S.D. Wullschleger (eds.), North American Temperate Deciduous Forest Responses to Changing Precipitation Regimes, Springer, New York (in press).
King, A.W., W.M. Post, and S.D. Wullschleger. 1997. The potential response of terrestrial carbon storage to changes in climate and atmospheric CO2. Climatic Change 35:199-227.
Law, B.E., P.E. Thornton, J. Irvine, S. Van Tuyl, and P.M. Anthoni. 2001. Carbon storage and fluxes in ponderosa pine forests at different developmental stages. Global Change Biology 7:755-777.
Lieth, H. 1975. Modeling the primary productivity of the world, pp. 237-263 In Lieth, H., and R.H. Whittaker (eds.), Primary Productivity of the Biosphere, Springer-Verlag, New York.
Liu, S., R. Munson, D.W. Johnson, S. Gherini, K. Summers, R. Hudson, K. Wilkinson, and L.F. Pitelka. 1992. The nutrient cycling model (NuCM): overview and application, pp. 583-609 In Johnson D.W., and R.I. VanHook (eds.), Analysis of Biogeochemical Cycling Processes in Walker Branch Watershed, Springer-Verlag, New York.
Luo, Y., B. Medlyn, D. Hui, D. Ellsworth, J.F. Reynolds, and G. Katul. 2001. Gross primary productivity in the Duke Forest: modeling synthesis of the free-air CO2 enrichment experiment and eddy-covariance measurements. Ecological Applications 11:239-252.
McMurtrie, R.E., and Y.P. Wang. 1993. Mathematical models of the photosynthetic response of tree stands to rising CO2 concentrations and temperature. Plant, Cell and Environment 16:1-13.
McNulty, S. G, J.M. Vose, and W.T. Swank. 1996. Loblolly pine hydrology and productivity across the southern United States. Forest Ecology and Management 86:241-251.
Norman, J.M. 1982. Simulation of microclimates, pp. 65-99 In Hatfield, J.L., and I.J. Thomason (eds.), Biometeorology and Integrated Pest Management, Academic Press, New York.
Pastor, J., and W.M. Post. 1985. Development of a Linked Forest Productivity-Soil Process Model. ORNL/TM-9519. Oak Ridge National Laboratory, Oak Ridge, Tennessee.
Post, W.M., A.W. King, S.D. Wullschleger. 1997. Historical variations in terrestrial biospheric carbon storage. Global Biogeochem Cycles 11:99-109.
Running, S.W., and J.C. Coughlan. 1988. A general model of forest ecosystem processes for regional application. Ecological Modelling 42:125-154.
Running, S.W., and S.T. Gower. 1991. FOREST-BGC: A general model of forest ecosystem processes for regional applications. II. Dynamic carbon allocation and nitrogen budgets. Tree Physiology 9:147-160.
Running, S.W., and E.R. Hunt, Jr. 1993. Generalization of a forest ecosystem process model for other biomes, BIOME-BGC, and an application for global-scale models, pp. 141-158 In: Ehleringer and C. Field (eds.), Scaling Physiological Processes: Leaf to Globe, Academic Press, Orlando.
Schubert, K.R. (ed.). 1982. The energetics of biological nitrogen fixation. American Society of Plant Physiologists, Rockville, Maryland, 30 pp.
Shugart, H.H., and D.C. West. 1977. Development of an Appalachian deciduous forest simulation model and its application to assessment of the impact of the Chestnut Blight. Journal of Environmental Management 5:161-179.
Shulten, H.-R., and M. Schnitzer. 1997. Chemical model structures for soil organic matter and soils. Soil Science 162: 115-130.
Shuttleworth, W.J., and J.S. Wallace. 1985. Evaporation from sparse crops – an energy combination theory. Quarterly Journal of the Royal Meteorological Society 111:839-855.
Sun, G., D.M. Amatya, S.G. McNulty, R.W. Skaggs, and J.H. Hughes. 2000. Climate change impacts on the hydrology and productivity of a pine plantation. Journal of the American Water Resources Association 36(2):367-374.
Thornley, J.H. 1995. Shoot:root allocation with respect to C, N and P: an investigation and comparison of resistance and teleonomic models. Annals of Botany 75:391-405.
Thornthwaite, C.W., and J.R. Mather. 1957. Instructions and tables for computing potential evapotranspiration and the water balance. Publications in Climatology 10:183-311.
Thornton, P.E., B.E. Law, et al. 2002. Modeling and measuring the effects of disturbance history and climate on carbon and water budgets in evergreen needleleaf forests. Agricultural and Forest Meteorology 113:185-222.
Trofymo, J.A., C.M. Preston, and C.E. Prescott. 1995. Litter quality and its potential effect on decay rates of materials from Canadian forests. Water, Air, and Soil Pollution 82:215-226.
U.S. Global Change Research Program. 2000. Climate Change Impacts on the United States: The Potential Consequences of Climate Variability and Change. Overview. A Report to the National Assessment Synthesis Team. Cambridge University Press. 154 pages.
Verma, S.B., D.D. Baldocchi, D.E. Anderson, D.R. Matt, and R.J. Clement. 1986. Eddy fluxes of CO2, water vapor, and sensible heat over a deciduous forest. Boundary-Layer Meteorology 36:71-91.
Wang, S., R.F. Grant, D.L. Verseghy, and T.A. Black. 2001. Modelling plant carbon and nitrogen dynamics of a boreal aspen forest in CLASS - the Canadian Land Surface Scheme. Ecological Modelling 142:135-154.
Wang, S., R.F. Grant, D.L. Verseghy, and T.A. Black. 200?a. Modelling carbon-coupled energy and water dynamics of a boreal aspen forest in a GCM land surface scheme (in review).
Wang, S., R.F. Grant, D.L. Verseghy, and T.A. Black. 200?b. Modelling carbon dynamics of boreal forest ecosystems using the Canadian Land Surface Scheme (in review).
Wang, Y.P., and P.G. Jarvis. 1990. Description and validation of an array model - MAESTRO. Agricultural and Forest Meteorology 51:257-280.
Wang, Y.P., A. Rey, and P.G. Jarvis. 1998. Carbon balance of young birch tree grown in ambient and elevated atmospheric CO2 concentration. Global Change Biology 4:797-807.
White, M.A., P.E. Thornton, and S.W. Running. 1997. A continental phenology model for monitoring vegetation responses to interannual climatic variability. Global Biogeochemical Cycles 11:217-234.
White, M.A., P.E. Thornton, S.W. Running, and R.R. Nemani. 2000. Parameterization and sensitivity analysis of the Biome-BGC terrestrial ecosystem model: Net primary production controls. Earth Interactions 4:1-85.
Williams, M., E.B. Rastetter, D.N. Fernandes, M.L. Goulden, G.R. Shaver, and L.C. Johnson. 1997. Predicting gross primary productivity in terrestrial ecosystems. Ecological Applications 7:882-894.
Williams, M., E.B. Rastetter, D.N. Fernandes, M.L. Goulden, S.C. Wofsy, G.R. Shaver, J.M. Melillo, J.W. Munger, S.-M.Fan, and K.J. Nadelhoffer. 1996. Modelling the soil-plant-atmosphere continuum in a Quercus-Acer stand at Harvard Forest: the regulation of stomatal conductance by light, nitrogen and soil/plant hydraulic properties. Plant, Cell and Environment 19:911-927.
Williams, M., E.B. Rastetter, G.R. Shaver, J.E. Hobbie, E. Carpino, and B.L. Kwiatkowski. 2001. Primary production in an arctic watershed: an uncertainty analysis. Ecological Applications 11:1800-1816.
Winslow, J.C., E.R. Hunt, Jr., and S.C. Piper. 2001. A globally applicable model of daily solar irradiance estimated from air temperature and precipitation data. Ecological Modelling 143:227-243.
Wofsy, S.C., M.L. Goulden, J.W. Munger, S.-M. Fan, P.S. Bakwin, B.C. Daube, S.L. Bassow, and F.A. Bazzaz. 1993. Net exchange of CO2 in a mid-latitude forest. Science 260:1314-1317.
Wullschleger SD, C.A. Gunderson, L.M. Tharp, and D.C. West. 2003. Simulated Patterns of Forest Succession and Productivity as a Consequence of Altered Precipitation. Chapter 25 in Hanson, P.J., and S.D. Wullschleger (eds.), North American Temperate Deciduous Forest Responses to Changing Precipitation Regimes, Springer, New York (in press).
Zheng, D., E.R. Hunt, Jr., and S.W. Running. 1993. A daily soil temperature model derived from air temperature and precipitation for continental applications. Climate Research 2:183-191.
CDIAC Home Page