Drought indices have been commonly used in drought monitoring and assessments, but selecting the most appropriate one under different geographical and climatic conditions is difficult. We constructed a multi-timescale integrated drought index, the Ensemble Drought Index (EDI), for more robust and reliable drought assessments with the widely used Standardized Precipitation Index (SPI), the Palmer Drought Severity Index (PDSI), the selfcalibrated Palmer Drought Severity Index (scPDSI), and the Standardized Precipitation Evapotranspiration Index (SPEI) using the Thornthwaite and the Penman-Monteith methods. The EDI can more accurately describe the major historical drought events, and has high and significant correlation with historical monthly soil moisture and annual runoff data all over China, regardless of regional differences. in region among those indices participating in integration. We also showed that the EDI greatly improved monitoring accuracy in the arid and semi-arid regions of China, where the assessments are always thorny. In the meantime, we revealed how the reference period of fitting statistical parameters affects the accuracy of drought assessments, and concluded that the long-term stationary climate variables series (i.e., without trend) can bring more accurate conclusions.

Drought indices have been commonly used in drought monitoring and assessments, but selecting the most appropriate one under different geographical and climatic conditions is difficult. We constructed a multi-timescale integrated drought index, the Ensemble Drought Index (EDI), for more robust and reliable drought assessments with the widely used Standardized Precipitation Index (SPI), the Palmer Drought Severity Index (PDSI), the self-calibrated Palmer Drought Severity Index (scPDSI), and the Standardized Precipitation Evapotranspiration Index (SPEI) using the Thornthwaite and the Penman-Monteith methods. The EDI can more accurately describe the major historical drought events, and has high and significant correlation with historical monthly soil moisture and annual runoff data all over China, regardless of regional differences. in region among those indices participating in integration. We also showed that the EDI greatly improved monitoring accuracy in the arid and semi-arid regions of China, where the assessments are always thorny. In the meantime, we revealed how the reference period of fitting statistical parameters affects the accuracy of drought assessments, and concluded that the long-term stationary climate variables series (i.e., without trend) can bring more accurate conclusions.

Drought indices have been commonly used in drought monitoring and assessments, but selecting the most appropriate one under different geographical and climatic conditions is difficult. With the widely used Standardized Precipitation Index (SPI), and Palmer Drought Severity Index (PDSI), self-calibrated Palmer Drought Severity Index (scPDSI), Standardized Precipitation Evapotranspiration Index (SPEI) using the Thornthwaite and the Penman-Monteith methods, we constructed a multi-timescale integrated drought index, the Ensemble Drought Index (EDI), for more robust and reliable drought assessments. The EDI can more accurately describe the major historical drought events, and has high, as well as significant, correlation with historical monthly soil moisture and annual runoff data all over China regardless of difference in region among those indices participating in integration. We also showed that the EDI greatly improved monitoring accuracy in the arid and semi-arid regions of China, where the assessments are always thorny. At the meantime, we revealed how the reference period of fitting statistical parameters affects the accuracy of drought assessments, and concluded that the long-term stationary climate variables series (i.e., without trend) can bring more accurate conclusions.