Abstract: To develop a reasonable and simplified TOPMODEL (Topographic Index model) that can be integrated into large-scale land surface models, the classic TOPMODEL is extended to spatially heterogeneous land surfaces, where saturated hydraulic conductivity at the ground surface k_0 , water recharge rate to groundwater R , and decline factor f are all spatially heterogeneous. The extended TOPMODEL with a spatial variation of f is coupled with the land surface model SSiB4 (SSiB4/GTOP). In addition, numerical experiments with an assumed spatial variable pattern of f are conducted with the coupling model to evaluate the hydrological effects of the spatial variation of f . The effects of spatial heterogeneity on soil moisture, evapotranspiration, surface runoff, base flow, and total runoff are analyzed. The main conclusions are as follows: (1) The spatial variation of k_0 or R does not change the original relation of the classical TOPMODEL; as long as the new topographic indexes are defined, no differences exist between the spatially heterogeneous TOPMODEL and the spatially homogeneous TOPMODEL. (2) For heterogeneous f in space, because the local groundwater table depth is also related to the local value of f , the conclusion that the locations with the same topographic index are hydrologically similar is no longer valid. (3) Compared with the simulation results of the spatial homogeneity of f , the spatial variation of f , which linearly decreases with the increase in altitude, will reduce the soil moisture, surface runoff, and evapotranspiration of the simulated basins but increase the base flow and total runoff. 4) The spatial heterogeneity of f affects hydrological simulations; however, the topographic effects are more important than the influence of the spatial variation of f .