Monitoring Many Faces of Drought over the United States based on the NLDAS systems [INVITED]

Drought indices derived from the North American Land data Assimilation Systems (NLDAS) from 1979 to the present are used to monitor drought over the United States. For meteorological drought, the Standardized Precipitation Index (SPI) is used to measure precipitation deficits. The Standardized Runoff Index (SRI) similar to the SPI is used to classify hydrological drought. Agricultural drought is measured by monthly mean soil moisture (SM) anomaly percentiles.

One important aspect of drought monitoring is to communicate uncertainties in the system. The uncertainties of indices are estimated by intercomparing two sets of the NLDAS ensemble products. One set is produced by the EMC/NCEP and another set is produced by the University of Washington. Two sets of the NLDAS have different models and forcing. While the indices derived from different models within the same system are similar, the differences between the ensemble mean indices from two systems are large. The largest differences occurred after 2002 when both systems went to real time operation. Over the eastern United States, indices derived from two systems can detect the same drought events. The monthly mean SM percentiles and runoff indices between two systems have larger differences over the western interior United States.

Verification of two NLDAS products against the Ameriflux in situ data will be presented in the meeting