Thornton, P.E., 1998. Regional ecosystem simulation: combining surface- and satellite-based observations to study linkages between terrestrial energy and mass budgets. Ph.D. Dissertation, The University of Montana. 280 pp.

White, J.D., S.W. Running, P.E. Thornton, R.E. Keane, K.C. Ryan, D.B. Fagre, and C.H. Key, 1998. Assessing simulated ecosystem processes for climate variability research at Glacier National Park, USA. Ecological Applications, 8: 805-823.

Kimball, J.S., P.E. Thornton, M.A. White, and S.W. Running, 1997. Simulating forest productivity and surface-atmosphere carbon exchange in the BOREAS study region. Tree Physiology, 17:589-599.

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.

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.

Terrestrial biogeochemistry process model

Study of global and regional interactions between climate, disturbance, and biogeochemical cycles

Represent the most important ecosystem processes in a computationally efficient framework.

Design criteria include:

Spatial - No intrinsic spatial scale. Depending on the availability of driving variables, the model can be run at any spatial scale. Fractional cover treatment helps to reduce aggregation error for larger cell sizes.

Temporal - Daily timestep.

Spatial - comparisons down to the stand level for one-dimensional simulations, and up to any gridcell size for gridded simulations.

Temporal - Daily comparisons against flux data (tower fluxes, hydrographs, etc). Annual or average annual comparisons against most typical NPP data.

Carbon: GPP, MR, daily allocation, GR, seasonal phenology of leaf and fine root display, daily litterfall, coarse woody debris physical fragmentation, litter turnover to soil organic matter, HR, fire losses, whole plant mortality.

Water: Precip to snow/rain, rainfall interception on canopy, evaporation of intercepted rainfall, transfer of intercepted rainfall to soil water pool, snowpack accumulation, snowmelt to soil water pool, evaporation from soil water pool, transpiration, outflow, slow drainage to field capacity.

Soils (simple bucket, saturated/unsaturated flow, controls on water movement through the profile, etc.): Simple bucket, one layer, defined by % sand, %silt, %clay, % rocky, and rooting zone depth.

Energy balance: (e.g. latent, sensible heat, aet, pet): Shortwave radiation balance, latent heat, sensible heat only by inference in
Penman- Monteith eq. No calculation of PET, only AET. For other details see the list under main heading.

Snow: snowmelt driven by temperature, vapor pressure deficit, and radiation. Includes a snowpack sublimation component.

'Order' of water balance: (e.g. incoming water is first evaporated from plant/soil surface, then infiltration, transpiration, runoff)

Carbon:

Nitrogen:

Soil water:

All the vegetation parameters are commonly measured ecophysiological variables, and we use a standard set of parameters for global runs that are averages from a large number of observations. This includes the allocation parameters, the photosynthesis parameters, the stomatal conductance and stomatal control parameters, the light extinction and rainfall interception parameters, the vegetation C:N values, and the proportions of each vegetation state variable that is labile, cellulose, and lignin.

Phenology model parameters were determined by comparison of satellite-derived phenological information with long-term meteorological data.

All the soil rate constants for turnover of organic matter are objectively derived from 14C labeling experiments, using data from the literature.

Each fractional vegetation type is simulated entirely independently, with no implicit or explicit spatial scaling. Fractional components can then be weighted by proportion to get flux densities for a grid cell, and further weighted by grid cell area to get total fluxes.

Fire, whole plant mortality, harvest. Whole plant mortality is intended to represent the combined effects of wind throw, death due to insect or disease, grazing or any process that leaves most of the dead plant material on the site. Harvest can be defined in different ways, to simulate taking or leaving different fraction

Takes a description of the fractional cover by fundamental vegetation type as an input.

Daily max and min temperature, daily total precipitation, elevation, slope, aspect, latitude, soil texture (%sand,%silt,%clay, %rocky), rooting zone depth, atmospheric deposition of mineral N, biological fixation of atmospheric N. Optional: fire frequency and intensity information, time since last stand-replacing disturbance, management practices (including fertilization, irrigation, thinning).