| This thesis documents the integration of a quasi-distributed, variable source area hydrologic model, TOPMODEL, with an ecosystem process simulation model, FOREST-BGC. The sensitivity of the integrated FOREST-BGC to two parameters, leaf area index (LAI), and a hydrologic similarity index, is evaluated.; Current ecosystem modelling lacked, but needs, the ability to simulate ecosystem behaviour while accounting for the spatial variability of topography, soils and vegetation. The integration of TOPMODEL with FOREST-BGC fills this methodological void. The spatially distributed FOREST-BGC is superior to the original FOREST-BGC in simulating annual evapotranspiration (ET), photosynthesis (PSN) and water use efficiency (WUE) in the Soup Creek basin, northwestern Montana.; Annual ET, PSN and WUE are quite sensitive to perturbations of LAI. The response of ET and PSN to variations in LAI is a function of the impact of LAI on the slope water balance. Hillslope total ET, PSN and WUE are only moderately sensitive to differences in the form of the hydrologic index frequency distribution. However, the index class, or interval totals are highly sensitive to distribution form, due to the variations in distribution of hillslope soil water among the intervals. Optimum states of LAI and soil water availability are identified. (Abstract shortened by UMI.)... |
