USING THE LAST GLACIAL MAXIMUM TO INFORM ECOLOGICAL NICHE MODELING

Climate change is rapidly altering the geographic ranges in which species can survive. Predictions of how these ranges might shift are crucial for conservation efforts, but the ecological niche models (ENMs) used to make these predictions are based predominately on modern distributions and may not adequately reflect a species’ fundamental niche. This could lead to inaccurate forecasts of range shifts. However, the past can illuminate the climatic extremes that a species lived in historically and can give clues to what a species can endure in the future. Species distributions from the Last Glacial Maximum (LGM, ~20 ka) have the potential to be particularly informative as they reflect the conditions in which species survived in a climate very different from today’s. If, as expected, LGM distributions more accurately indicate a species’ fundamental niche, then forecasts based on these distributions may predict less drastic range shifts than those based on modern data alone. Using online collections databases, we compiled modern species occurrence data and occurrence data from the LGM for North American artiodactyls and carnivorans. We then used these data combined with climatic reconstructions of the LGM, historical climate data from the 19^th-21^st Centuries, and predictions of climate at the end of the current century. This produced two sets of maps: one predicting where a species could live based on fossil occurrence data and one predicting where a species could live based on modern occurrence data. As predicted, range shifts based on LGM occurrences forecasted less dramatic range shifts than those based on modern distributions. ENMs based on both LGM and modern distributions have the potential to be more accurate than those based on just a single time period.