A global historical and near-real time drought index from the Global Precipitation Climatology Centre (GPCC)

The Global Precipitation Climatology Centre (GPCC) was established in 1989 on request of WMO at Deutscher Wetterdienst (DWD). Under WMO auspice, it collects in situ precipitation data. Those data are quality controlled and stored in a relational data bank system. High-quality gridded precipitation analyses are produced based on these data and are made publicly available at DWD’s OpenData-Server. Using these data, a near-real time global drought index was developed to support drought monitoring. The access and usage of those data sets is not restricted.

GPCC’s drought index (GPCC-DI) combines the Standardized Precipitation Index (SPI) and the Standardized Precipitation Evapotranspiration Index (SPEI). The combination is needed, as none of both indices is globally useable due to technical limitations or near-real time data availability. The SPI is based on precipitation observations and not applicable in very dry regions. The reason lies in the fact that the gamma distribution fitted to those data cannot be standardized. The SPEI uses the water balance for the detection of droughts and therefore needs an estimation of the potential evapotranspiration (PET). Only simple parameterizations, like the one developed by Thornthwaite using only temperature data, can be applied in near-real time due to limited accessibility of data. A disadvantage of this parameterization, however, are the unrealistic PET estimates in cold regions inhibiting applicability in those areas. The combination of SPI and SPEI fills the gaps of each other and provides a global overview except for only a few places being both cold and dry, e.g. some mountain tops in the Andes and Himalayas.

One potential drawback in station-based analysis is that time-series suffer from gaps due to missing readings, instrument malfunctions, maintenance, and transmission problems. Some stations are retired and therefore not available anymore for real-time analyses, while new stations have too short time-series to calculate reliable statistics needed in the calculation of SPI and SPEI. To overcome these limitations, gridded input data providing reliable long time series are used from the GPCC for precipitation and NOAA’s Climate Prediction Centre (CPC) for air temperature, respectively.

Impacts of water shortages in various sectors happen on different time scales. Therefore, the GPCC-DI is calculated for several accumulation periods from one month up to four years. The resulting data set covers the period from 1952 to 2013. Since 2013, near-real time updates have been done on a monthly basis. GPCC-DI is provided via DWD’s OpenData-Server.

The paper presents the methodology and analyses of drought events.