|Pimstein, Agustin -|
Submitted to: BARC Poster Day
Publication Type: Abstract Only
Publication Acceptance Date: March 10, 2010
Publication Date: April 21, 2010
Citation: Pimstein, A., Anderson, M.C. 2010. Spatial and temporal assessment of a new spaceborne drought index in comparison with standard precipitation-based indices[abstract]. Abs. 32. BARC Poster Day. Technical Abstract: Near real-time and high-resolution spatial information about soil moisture status is becoming an urgent need for water resource managers for accurate and timely detection of impending drought conditions. Methods for identifying agricultural droughts at early stages are of primary importance, so as to facilitate quick and effective response to localized reductions in food production. The United States Drought Monitor (USDM) reports the current hydrological conditions over the U.S. based on a combination of drought indicators. The drought indices used in the USDM combine different components of the hydrologic budget (e.g. soil moisture and evapotranspiration (ET) modeling), and ground-based meteorological observations (e.g. precipitation, groundwater storage and streamflow). The accuracy of these indicators is limited by errors in assumed model forcings, deficient spatial resolution of some of the observations, and biases/noise in current techniques for mapping precipitation. In this ongoing research, temporal ET anomalies retrieved from a satellite-based inverse model of Atmosphere-Land Exchange (ALEXI) are being evaluated for utility in the USDM suite of drought indices. Using principles of surface energy balance, and measurements of land-surface temperature derived from thermal band satellite imagery, the model determines the evaporation rate required to keep the soil and vegetation scene components at the observed temperatures. Due to its independence from precipitation data, the Evaporative Stress Index (ESI) provides a more robust monitoring in regions with minimal ground-based meteorological infrastructure. The ESI was compared to the USDM drought classifications, and to precipitation-based drought indicators considered in the production of the USDM (i.e., the suite of Palmer drought indices and the Standardized Precipitation Indices). This comparison examined spatial similarity drought index maps at different points in time, and temporal correlations between drought index pairs as a function of location across the U.S. Intercomparison of monthly drought maps during the growing seasons from 2000 to 2009 shows that spatiotemporal patterns in the ESI are similar to those in the USDM reports, capturing strong drought events in the western and southeastern U.S. Seasonal discrepancies between the ESI, USDM and precipitation indices were attributed in part to the partial decoupling of ET from monthly rainfall rates in areas where ET is enhanced by non-precipitation-based moisture inputs, such as in regions of shallow water table or a high density of irrigated land area. These results confirmed the high degree of confidence on the remotely sensed ESI for characterizing drought conditions, opening the possibility of applying it in other regions of the world in which the meteorological infrastructure is less developed. Future evaluation of ESI will include further comparison to other satellite drought indicators (e.g., VHI, VCI, VegDRI, etc), as well as assessment of alternative comparative techniques such as the Vector Change Analysis approach for temporal analysis and Kriging techniques for spatial analysis. Additionally, at the operational level, ESI production is being transitioned to NOAA facilities for being included in the North American Drought Briefing, and for having data available through the National Integrated Drought Information System (NIDIS) portal.