DO CLOUDS CREATE LAKES? - THE CASE OF PLEISTOCENE LAKE BONNEVILLE

A major paradox of Quaternary geology is the existence of lakes in the Great Basin during glacial cycles when one would expect the cooler temperatures and reduced saturation vapor pressure would lead to reduced fluxes in the hydrologic cycle. Many explanations focus on the possibility of decreased evapotranspiration, which could result from lower temperatures or from increased cloud cover. Cloud cover can influence evapotranspiration in two ways: (1) indirect radiative forcing from water vapor emissivity and transmissivity and (2) albedo changes. Here we quantify #1 in the context of the Late Pleistocene hydrologic cycle of the western U.S. with a radiation balance model (Pease, 1987) and two variables: specific humidity and surface pressure, derived from a control scenario and Late Pleistocene scenarios (21ka, 16ka, 14ka and 11ka) of a general circulation model (CCM1). Water vapor pressure is calculated from specific humidity (re-expressed in terms of density of water vapor and of dry air) and surface pressure using the equations of state for water vapor and dry air. Emissivity consists of two components: to space and to the surface, and is calculated from water vapor pressure and longwave emittance. The proportion of longwave emittance to the surface depends upon the emissivities of CO2 and vapor in each scenario. We also adjust CO2 emissivity for the changing CO2 load in the atmosphere in each scenario. Results are calculated for the CCM1 grid cell containing Lake Bonneville and for a maritime grid cell for comparison.

The Late Pleistocene reduction in specific humidity was generally greater at the coast than inland so atmospheric emissivity was generally larger inland in each scenario. We find that outgoing longwave radiation was up to 9% higher at 21ka and reached its lowest values at 11ka in both locations and this pattern may be related to the rise and collapse of the Great Basin lakes in the latest Pleistocene.