The preservation of the δ18O and δD composition of rain and snow, and resulting isotope-elevation gradients, in terrestrial waters derived from precipitation, such as streams, lakes, snowmelt, and soil waters, is often an inherent assumption in paleoclimate, hydrologic, and paleoelevation studies utilizing paleo–meteoric water proxy records. Modern spring and surface water samples from the southwestern United States along a latitudinal transect at ∼36°N from the southern Sierra Nevada, California, to the central Basin and Range reveal that spring and surface waters collected from the orographic slope on the western side of the Sierra Nevada preserve isotopic compositions and δ18O-elevation gradients (–2 to –3‰/km) that are consistent with published regional precipitation records and that are similar to values expected based on simple Rayleigh distillation processes. In contrast, high-elevation (≥2000 m) spring waters from the Panamint and Spring Mountains in the orographic rain shadow on the eastern side of the Sierra Nevada exhibit δ18O values (∼–14‰ to –13‰ relative to Vienna standard mean ocean water [VSMOW]) that are significantly higher than those of the winter season precipitation (–22‰ to –15‰) from which the terrestrial waters are derived, yielding spring water δ18O-elevation gradients (∼–0.8‰/km) that are 2–3 times lower than those derived from regional precipitation values (∼–2.0‰/km) and those predicted by Rayleigh distillation models. The observed discrepancy between precipitation and terrestrial water isotopic compositions and the reduced spring water δ18O-elevation gradients in the Panamint and Spring Mountains suggest that elevation-dependent modification of the isotopic composition of precipitation following deposition plays a key role in the isotope hydrology of meteoric water systems in the continental interior southwestern United States. We demonstrate that altitude-dependent sublimation of the winter snowpack is a viable mechanism for the observed isotopic enrichment and reduced δ18O-elevation gradients of Basin and Range spring waters. This finding suggests that atmospheric processes alone may be insufficient to explain globally observed variations in δ18O-elevation gradients, and that the geologic record of the δ18O composition of paleo–meteoric waters may not accurately reflect the δ18O compositions of paleoprecipitation, particularly in arid environments dominated by winter precipitation in the form of snow. If unaccounted for, sublimation effects can lead to significant underestimates of both paleoelevation (on the order of 2–3 km) as well as the contribution of winter season precipitation to annual groundwater recharge. As a result, consideration of potential sublimation influence on paleo–meteoric water systems, particularly as can be constrained by paleoclimate reconstructions, is necessary in order to accurately derive quantitative estimates of paleoelevation and seasonal groundwater recharge.