Terrestrial biogeochemical cycling and vegetation response to climate in an earth system model

摘要

We are developing (1) a seasonal model of terrestrial productivity and biogeochemical cycling (TERRA) and (2) a model of vegetation response to climate (HABITAT). Both of these models are designed to be components of an Earth System Model being developed at Lawrence Livermore National Laboratory (LLNL). We have implemented the grid-cell model of TERRA and have calibrated it for the 17 vegetation types that it simulates. In a sensitivity analysis, we have found that total system response is most sensitive to parameters affecting soil moisture content which, in turn, affects soil respiration. Albedo was found to be an important factor in tundra systems. Carbon sequestration was strongly influenced by only a handful of parameters for each of the five types discussed in this paper. Parameters affecting soil respiration, soil moisture, littefall rate, CO(sub 2) assimilation, C:N ratio in litterfall, and nitrogen uptake were, in most circumstances, found to be the most important parameters. As an initial phase in constructing the vegetation response model, we have investigated an existing method of classifying vegetation life zones based on total annual precipitation and average monthly temperature to determine if this method discriminates between vegetation types. We used global vegetation and climatic data sets for this analysis. We found that these variables lack power to resolve vegetation types.