Recently, a conservation strategy developed to restore populations of black-tailed prairie dog (Cynomys ludovicianus) suggested reintroducing animals into the Chihuahuan Desert grasslands of the southwestern United States. Rainfall in desert habitats is lower and more variable compared to rainfall near the center of the prairie dog's range. Additionally, peak rainfall comes months after prairie dogs reproduce in these desert systems. Thus, southwestern populations may be less prolific and fluctuate more than those found in northerly climes. Using mark–recapture and mark–resight techniques, we estimated reproduction and monthly survival from 577 individuals inhabiting 6 reintroduced colonies from 2003 to 2005 in the northern Chihuahuan Desert. During 2003 precipitation was 64% of the long-term average, whereas both 2004 and 2005 had near-average precipitation. Probability that a female became pregnant, number of juvenile prairie dogs emerging from maternity burrows, and date of emergence were all correlated to adult female body mass. Adult monthly survival decreased from >0.95 during spring to 0.70 in summer 2003, following a rapid loss in adult body mass that coincided with low precipitation. In 2003 monthly juvenile survival was near zero on 2 of the 3 largest colonies and growth rates of juveniles were half that of subsequent years. Estimated population size declined by 68% (range  =  18–91%) from 2003 to 2004, and 5 of 6 populations declined an average of 75% from their original introduction size. Prairie dog populations in desert environs may have a high risk of extirpation caused by weather patterns indicative of desert climates. Our results are important for those managers involved in the conservation of prairie dogs and we suggest that regional differences should be carefully considered prior to any reintroduction effort.