We explore in this paper how animals can be affected by variation in climate, topography, vegetation characteristics, and body size. We utilize new spatially explicit state-of-the-art models that incorporate principles from heat and mass transfer engineering, physiology, morphology, and behavior that have been modified to provide spatially explicit hypotheses using GIS. We demonstrate how temporal and spatial changes in microclimate resulting from differences in topography and vegetation cover alter animal energetics, and behavior. We explore the impacts of these energetic predictions on elk energetics in burned and unburned stands of conifer in winter in Yellowstone National Park, chuckwalla lizard distribution limits in North America, California Beechey Ground squirrel and Dusky Footed woodrat mass and energy requirements and activity patterns on the landscape, their predator prey interactions with a rattlesnake, Crotalus viridis, and shifts in that food web structure due to topographic and vegetative variation. We illustrate how different scales of data/observation provide different pieces of information that may collectively define the real distributions of a species. We then use sensitivity analyses of energetic models to evaluate hypotheses about the effects of changes in core temperature (fever) global climate (increased air temperature under a global warming scenario) and vegetation cover (deforestation) on winter survival of elk, the geographic distribution of chuckwallas and the activity overlap of predator and prey species within a subset of commonly observed species in a terrestrial food web. Variation in slope and aspect affect the spatial variance in solar radiation incident on the ground, hence ground surface temperature, at the same elevation, same hourly 2 m air temperatures, and wind speeds. We illustrate visually how spatial effects and landscape heterogeneity make statistical descriptions of animal responses problematic, since multiple distributions of their responses to climate, topography, and vegetation on the landscape can yield the same descriptive statistics, especially at high (30 m) resolution. This preliminary analysis suggests that the model has far-reaching implications for hypothesis testing in ecology at a variety of spatial and temporal scales.