COS 150-7 - How novel are 21st century climates? A global assessment of future climates and their analogs back through the Eocene

Thursday, August 10, 2017: 3:40 PM
E141, Oregon Convention Center
Kevin D. Burke, Nelson Institute for Environmental Studies, University of Wisconsin-Madison, Madison, WI, John W. (Jack) Williams, Geography, University of Wisconsin-Madison, Madison, WI and Stephen T. Jackson, DOI Southwest Climate Science Center, U.S. Geological Survey, Tucson, AZ

Current rates of resource consumption have likely committed our planet to a future whose climate will be warmer than present. Given this knowledge and the fact that environmental change is a multivariate process that leads to a shift in the environmental space of a location (both the mean state as well as the extremes), it is important to assess how much the climate will shift. Yet it is also important to assess where analogous climates will occur in the future in order to determine the viability of a species to track and persist in a suitable environment. To quantify how different a future climate will be relative to a baseline period, one can calculate the dissimilarity (Standardized Euclidean and Mahalanobis Distances) between these time periods using a series of many-to-one comparisons. Studies have assessed climate novelty relative to the modern era, however we extend this baseline to quantify how dissimilar projected future climates will be relative to a paleo-baseline as well. Using modeled simulations of past climates (including HadCM3, CCSM, and GISS Model E-R) we apply this technique to the last interglacial, as well as past earth warm periods such as the mid-Pliocene and early Eocene.


These time periods are of interest because various geochemical proxies suggest global climate that was similar, if not warmer, than the present. Modeled climate simulations at the global scale offer the best opportunity to assess how similar these past climates may have been to our projected future. When comparing 21st century climates to individual snapshots of the past, dissimilarity is, on average, highest when compared to past cool periods like the Last Glacial Maximum and the Penultimate Glacial Maximum, and is lowest when compared to the last interglacial, the mid-Pliocene and the early Eocene. Spatial patterns differ between time periods though climate dissimilarity is frequently high in the high latitudes, as well as portions of equatorial Africa and the Tibetan plateau. Regions where climate dissimilarity remains high across all time periods represent locations whose future is more climatically unprecedented than anywhere else on the planet. An important secondary consideration is the relative contribution of different climate variables to dissimilarity. While past earth warm periods like the Eocene may be climatically similar in terms of temperature, their precipitation patterns are more uncertain and lead to greater dissimilarity.