Coccidioidomycosis is an endemic infectious disease in western North American deserts caused by the dimorphic ascomycete Coccidioides spp. Even though there has been an increase in the number of reported cases in the last years, few positive isolations have been obtained from soil samples in endemic areas for the disease. This low correlation between epidemiological and environmental data prompted us to better characterize the fundamental ecological niche of this important fungal pathogen. By using a combination of environmental variables and geospatially referenced points, where positive isolations had been obtained in southern California and Arizona (USA) and Sonora (Mexico), we have applied Genetic Algorithm for Rule Set Production (GARP) and Geographical Information Systems (GIS) to characterize the most likely ecological conditions favorable for the presence of the fungus. This model, based on environmental variables, allowed us to identify hotspots for the presence of the fungus in areas of southern California, Arizona, Texas, Baja California, and northern Mexico, whereas an alternative model based on bioclimatic variables gave us much broader probable distribution areas. We have overlapped the hotspots obtained with the environmental model with the available epidemiological information and have found a high match. Our model suggests that the most probable fundamental ecological niche for Coccidioides spp. is found in the arid lands of the North American deserts and provides the methodological basis to further characterize the realized ecological niche of Coccidioides spp., which would ultimately contribute to design smart field-sampling strategies.