CLIMATOLOGICAL DRIVERS OF PLIO-PLEISTOCENE PLUVIAL LAKE DISTRIBUTIONS IN WESTERN NORTH AMERICA

The persistence of large inland lakes in western North America during the Pliocene and Pleistocene provides first-order constraints on changes in the regional water balance. Furthermore, the spatial distribution of terminal lakes records changes in the regional moisture delivery dynamics. We investigate the climatological conditions driving terminal basin lake levels, and the spatial pattern of pluvial lakes in western North America during the mid-Pliocene warm period and the latest Pleistocene glacial maximum. Geologic proxy evidence suggests wet conditions persisted in regions of the southwest during both periods despite dramatically different global climate and pCO2 levels, including great differences in continental ice sheet configuration and sea surface temperature gradients. Lacustrine deposits suggest that lakes persisted in many terminal basins during both warmer and colder periods relative to today. We investigate the hydroclimate drivers of lake distributions using two methods: (1) a quantitative proxy-model comparison using compilations of geologic proxies and an ensemble of climate models, and (2) forward modeling of lake area distributions for terminally draining watersheds. We utilize archived climate model simulations of the Last Glacial Maximum (21 ka, LGM) and mid-Pliocene (3.3 Ma) produced by the Paleoclimate Modelling Intercomparison Project (PMIP and PlioMIP). Our proxy network is made up of stable isotope records from caves, soils and paleosols, lake shoreline dating, outcrop extent, and sediment cores, glacier extent, and packrat middens. Steady-state modeling of terminal basin lake levels and distributions suggest that reduced evaporation and moderate increases in precipitation, relative to modern, drove moderate to large pluvial lakes during the LGM in the Great Basin. In contrast, larger precipitation increases appear to be the primary driver of lake levels during the Pliocene in the southwest, with this spatial difference suggesting a role for El Niño teleconnections during the mid-Pliocene. The implications of these results for the terrestrial hydrologic cycle is that during past periods of global change patterns of ‘dry-gets-drier, wet-gets-wetter’ do not hold true for some regions of the mid-latitudes such as western North America.