Testing climate tracking of montane rodent distributions over the past century within the Great Basin ecoregion

Geographic ranges of many taxa have shifted poleward or up-slope under warming. Little is known, however, about the consistency of these dynamics among multiple mountains within a species’ geographic range. We assess climate tracking over the past century for four montane rodents, across the Ruby Mountains and Toiyabe Range within the Great Basin ecoregion of western North America. To do this, we generated species distribution models (SDMs, Maxent) in each mountain range during historical (1923–1940) and modern (2005–2015) time periods. Nineteen bioclimatic variables were derived from 800 m resolution monthly PRISM climate group data. SDMs trained on historic occurrences and climate were projected onto modern climates to estimate expected species distributions along each elevational gradient. Occurrence-based Observed Modern SDMs were compared to these Expected Modern SDMs, with congruence between the two representing the degree of climate tracking. Climate tracking predicted up-slope range expansions and contractions in the Ruby Mountains, while stasis or down-slope contractions were predicted in the Toiyabe Range. Observed Modern SDMs rarely matched climate tracking expectations, and the same species showed variable responses among mountains. Deviations are partially explained by changes in the contribution of temperature vs. precipitation over time. Additionally, optimal niche space in the Ruby Mountains shifted in elevation, but shifted in both elevation and latitude in the Toiyabe Range. Our results indicate that data from one elevational gradient may not be predictive of range dynamics elsewhere, even across similar environmental conditions and habitats, even for the same species. Conservation efforts will be better served by strategies that consider both inter- and intraspecific variation in responses to climate change over space and time.

Abstract licence: CC BY-NC

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.