Evaluation of daily SPI and SPEI indices for near-real time drought monitoring over CONUS

Two drought indices; the Standardized Precipitation Index (SPI) and the Standardized Precipitation Evapotranspiration Index (SPEI) are computed over CONUS using daily precipitation and temperature estimates from the NOAA Daily U.S. Climate Gridded Dataset (NClimGrid-Daily). The NClimGrid-Daily dataset consists of four climate variables derived from the GHCN-D dataset: maximum, minimum, and average temperatures and precipitation from 1951 to the present with a 5-km grid resolution. While SPI only uses precipitation as an input to assess drought conditions, SPEI uses both precipitation and potential evapotranspiration (PET). The daily SPI and SPEI are computed over various time scales (30-, 90-, 180-, 270-, 365-, 730-day). The differences between the two indices are being evaluated focusing on the influence accumulation period, differing period of record, and various SPI (McKee et al 1993, Guttman 1999) and daily PET (Thornthwaite and Mather 1957, Camargo et al. 1999, Pereira and Pruitt 2004) formulations. The impact of the period of reference is analyzed to account for the impact of precipitation and temperature changes over time (i.e., 1952-present, 1960-1990, and 1990-2020 for instance). For the most recent period (2000-present), the NClimGrid-SPI and NClimGrid-SPEI are compared against existing droughts monitoring resources such as the weekly US Drought Monitor (USDM). The use of cloud-scale computing resources reduces considerably the computation time and allows for the near-real time computation of daily SPI and SPEI indices. The effort to transfer the SPI and SPEI from research to operation (R2O) and to provide near-real time drought monitoring capabilities is also presented.