Although Mexico's central arid and semiarid lands face numerous conservation problems, they have largely been neglected from conservation, despite their apparent biological importance. We studied the semiarid Llanos de Ojuelos, where the original plant communities of grasslands, shrublands, and patches of oaks have been modified strongly by cattle grazing and for cultivation, affecting animal populations and communities, as well as the landscape-level ecological dynamics. Very little about the populations of wild animals in the area and the processes underlying their distribution and dynamics is known. We focused on whether satellite information was useful to explain rodent distributions and abundances in this area with a complex landscape of diffuse plant communities. In the spring of 2008 we inventoried the rodents at 74 locations in the area, through the use of Sherman live-traps (40 traps during two nights, per site; total=5920 night-traps). We covered the major perennial vegetation types. The existence of a complex landscape of intergrading grasslands, Opuntia comunities and shrublands (dominated by Acacia, Mimosa, Dodonaea, and Quercus) of different types and composition creates rich habitat mosaic that promotes a rich rodent fauna. We captured 20 species of rodents (458 individuals), while richness was estimated at 21-25 species. Reithrodontomys fulvescens, Chaetodipus nelsoni, Dipodomys phillipsii, Peromyscus melanophrys, Peromyscus gratus, and Peromyscus difficilis were very common at our sites, while Chaetodipus pennicillatus, Liomys irroratus, Perognathus flavus, Dipodomys merriami, Reithrodontomys megalotis, Peromyscus boylii, Neotoma leucodon, Peromyscus eremicus, Dipodomys ordii, and Peromyscus manicultatus were common, and Neotoma goldmani, Chaetodipus hispidus, Onychomys arenarius, and Baiomys taylori were rare. To explore the value of satellite information to understand and predict the rodents' distributions, we derived reflectance data from all bands of a Landsat5 TM Image (5 March 2008) for each trapline's midpoint. We will compare each rodent species' abundance with this data as well as with the normalized difference vegetation index (NDVI) and tasseled cap transformations (TCs: brightness, greenness, and wetness). We will link rodent presence/absences to reflectance values through linear and non-linear discriminant analysis, as well as through general linear models and multiple adaptive regression splines. Satellite images classified following the best models will then be interpreted according to field expertise. As end results we expect to have a better understanding on the rodent species' ecology, an adequate image classification procedure, and potential distribution maps for the rodent species in the study area.