Global spaceborne assessment of the relationship between terrestrial water storage and evaporative demand

Tuesday, 16 December 2014: 4:15 PM
Paul Alexander Levine1, Caroline De Linage1, Isabella Velicogna1,2 and James Tremper Randerson1, (1)University of California Irvine, Department of Earth System Science, Irvine, CA, United States, (2)Jet Propulsion Laboratory, Pasadena, CA, United States
Feedbacks between terrestrial water storage (TWS) and the atmosphere can impact climatic extremes such as droughts and heat waves. High evaporative demand (high temperature and/or low humidity) removes more water by evapotranspiration, leading to lower TWS. At the same time, low TWS availability limits evapotranspiration, which increases evaporative demand by reducing the ratio of latent to sensible heat fluxes from the land surface. Studies using coupled land-atmosphere models have explored the strength of these relationships, and have been validated using local observational data, but global-scale observational studies are limited by data availability. We quantified these relationships globally using TWS data from the Gravity Recovery and Climate Experiment (GRACE), surface air temperature and relative humidity from the Atmospheric Infrared Sounder (AIRS), and precipitation from the Global Precipitation Climatology Project (GPCP). Our approach demonstrated a strong relationship between antecedent TWS and subsequent evaporative demand in certain key regions of the globe. These regions are consistent with the “hot spots” of land-atmosphere coupling found in modeling studies, and represent semi-arid transitional areas where TWS strongly limits evapotranspiration. We also found a strong relationship between antecedent evaporative demand and subsequent TWS throughout most of the global land surface. These observational relationships provide a benchmark for comparison with climate model simulations. We found good agreement between the satellite-derived relationships and the equivalent relationships calculated with output from the Community Land Model (CLM) version 4.5.