Near Real-time Seasonal Drought Forecasting and Retrospective Drought Analysis using Simulated Multi- layer Soil Moisture from Hydrological Models at Sub- Watershed Scales

This study proposes a stratified approach of drought severity assessment using multi-layer simulated soil moisture. SWAT (Soil and Water Assessment Tool) models are calibrated for 50 watersheds in the South-Atlantic Gulf region of the Southeastern US and a high-resolution daily soil moisture dataset is obtained at Hydrologic Unit Code (HUC-12) resolution for a period of January 1982 through December 2013. A near real-time hydrologic simulation framework by coupling the calibrated SWAT models with the National Centers for Environmental Prediction (NCEP) coupled forecast system model version 2 (CFSv2) weather data is developed to forecast various water balance components including soil moisture (SM), actual evapotranspiration (ET), potential evapotranspiration ET (PET), and runoff (SURQ) for near-real time drought severity assessment, and drought forecasting for a lead of 9-months. A combination of the surface and total rooting depth soil moisture percentiles proves to be an effective increment over conventional drought assessment approaches in capturing both, transient and long-term drought impacts. The proposed real-time drought monitoring approach shows high accuracy in capturing drought onset and propagation and shows a high degree of similarity with the U.S. Drought Monitor (USDM), the long-term (PDSI, PHDI, SPI-9 and SPI-12), and the short-term (Palmer Z index, SPI-1 and SPI-6) drought indices.