Detecting legacies of millennial scale climate oscillations on modern biodiversity: lessons from a proxy-model comparison

The ecological legacies of past climate changes can confound efforts to forecast biodiversity responses to contemporary climate trends. Other studies have shown imprints of glacial climates on contemporary patterns of biodiversity, but the legacy effects of millennial-scale climate variability, one of the dominant modes of variability over the last 250,000 years remain under-evaluated. We assess the legacies of millennial-scale temperature and precipitation change in the contemporary distribution of species richness for amphibians, birds, mammals, reptiles, and trees in eastern North America (ENA). We use a suite of 13 paleoclimate estimates from 11 different climate simulations with transient or meltwater-forcing experimental designs, and we compare results from climate simulations with proxy-based reconstructions. These alternative paleoclimate estimates are used in spatial error models (SEM) to determine the relationship between millennial-scale paleoclimate change and contemporary biodiversity. Our SEMs demonstrate detectible but weak legacies of millennial-scale climate change to predict the contemporary distribution of all classes of organisms. Millennial-scale precipitation variations carry the strongest effects, in which the increased precipitation in portions of ENA following a weakening of the Atlantic meridional overturning circulation is predictive of greater contemporary richness, particularly for avian, mammalian, and tree richness. Millennial-scale temperature changes are less predictive of contemporary biodiversity than precipitation, with relationships that change across alternative paleoclimate estimates. Regions of ENA where temperatures increased after the onset of meltwater forcing harbor greater reptilian, arboreal, and mammalian richness. These findings suggest that the warmer and wetter climates of the southeastern United States may have been a refuge from millennial-scale climate variability, and that contemporary patterns of biodiversity may be partially shaped by legacies of past millennial-scale climate change. Hence this work points to the importance of past and present climate variations in regulating species distributions and suggests that forecasting species distributions will need to account for legacies of past climatic change.