Shrub encroachment into grasslands of the northern Chihuahuan Desert has resulted in often dramatic and geographically extensive changes in vegetation composition and soil structure. The nature of these transformations and their dependencies on soil characteristics have been largely conceptualized in state and transition models. Reduced perennial plant cover and heightened erosion in shrub interspaces suggests a reduced capacity for shrub-dominated sites to convert rainfall, typically the limiting resource in arid ecosystems, into plant biomass. We examined this hypothesis in the Jornada Basin, New Mexico using the MODIS Enhanced Vegetation Index (EVI) as a surrogate for primary productivity. The time frame of the study (2002-2008) coincided with a regional transition from below to above normal rainfall, providing an opportunity to assess vegetation response to increased moisture availability. A relatively dense network of rain gauges on the Jornada Experimental Range and Chihuahuan Desert Rangeland Research Center allowed for satisfactory rainfall interpolation at the scale of MODIS 250 meter imagery. Problems arising from spatial heterogeneity in soil brightness and vegetation structure were reduced by computing linear relationships between annual integrated EVI (I-EVI) and rainfall on a per-pixel basis across the study area. The coefficient of determination (r2), slope and intercept of these regressions were then mapped as a means of assessing spatial patterns in rainfall use efficiency (RUE). Here, we illustrate the results of this approach. We evaluate RUE metrics against ground-based ecological state classifications and satellite-derived shrub cover estimates, and we show that temporal, pixel-based assessments of RUE offer valuable insight into the functioning of Chihuahuan Desert vegetation states, despite some common problems associated with interpreting vegetation indices in ecosystems with low and diverse plant cover.