The need for accurate representation of spatial variability in precipitation, soil moisture and runoff generation processes on multiple spatial scales is prevalent in efforts at the numerical simulation of terrestrial hydrologic processes. Surface models are often oriented at hydrologic representations on specific scales, e.g. hillslopes, catchments (local), watersheds (regional), or basins (continental). The Land Information System (LIS), developed in the NASA Hydrological Sciences Branch at Goddard Space Flight Center (GSFC), provides an ideal numerical framework for the simulation of hydrologic states and fluxes across spatial scales. Current projects with LIS include the examination of precipitation downscaling methods and their impact on the accurate representation of surface processes at scales from O(10 m) to O(100 km). New downscaling methods for precipitation are constrained by the assimilation of satellite and in-situ data on land cover, soil types, vegetation, snow and ice cover, and topography. The effects of several different downscaling algorithms on the representation of hydrologic states (e.g. soil moisture content) and fluxes (e.g. runoff generation, evapotranspiration) in LIS are examined in this work.