Assessing the use of small mammal isotopes as paleoclimatic proxies: case studies from the Pacific Northwest and Southern California

Small mammals record many aspects of their diet and thus local environment in their tissues via stable isotopes. The isotopic composition of extant small mammals has been shown to track climate and environmental gradients; however, these relationships have rarely been applied to the past. Small mammals are ubiquitous in the fossil record and represent a potentially useful and currently under-utilized proxy for climate and environmental change. Key questions remain about the spatial scale represented by small-mammal isotopes and the aspects of the environment that they record. Here, we aim to characterize the climatic and environmental variables recorded in small mammal δ¹⁸O and δ²H. We analyzed deer mice (Peromyscus maniculatus) hair samples from specimens collected over the past 100 years from natural history museums across the country. We targeted specimens from two regions of the western United States, the Pacific Northwest (PNW) and southern California (CA), that capture a wide range of temperature and precipitation regimes. We focused on δ¹⁸O and δ²H of hair, which reflect aspects of the hydroclimate. In the PNW, we found strong relationships between the δ²H of mice hair and growing season temperature and longitude. These relationships indicate that P. maniculatus accurately record growing season conditions and Rayleigh isotopic fractionation. Surprisingly, we did not find strong correlations between δ¹⁸O_hair and climatic variables in the PNW. Preliminary data from CA mice show strong relationships between δ¹⁸O_hair and vapor pressure deficit, but not with δ²H_hair and growing season precipitation, as seen in the PNW. Our results suggest that the δ¹⁸O and δ²H of a single widespread, generalist species may be controlled by different climatic variables depending on the region. As O is a common isotope system used for proxy reconstructions from the fossil record in deep time, our work suggests that a priori knowledge of the general climate and environmental state (e.g., arid vs. humid) is necessary for the application of small mammal δ¹⁸O as a paleoclimatic proxy. The isotopic composition of small mammals, be it from hair, bone or enamel in the fossil record, can then provide an additional layer of hydroclimate information, potentially more finely tuned to the local ecosystem conditions than other proxies.