Drought assessment of China in 2002-2017 based on a comprehensive drought index

Drought, as an extreme natural disaster event, can cause or exacerbate water, food, and national security hazards. However, because of differences in the regional characteristics and applicability of drought indices, a more comprehensive drought index is necessary for drought assessment over large areas. Here, we propose a new detrended combined climatologic deviation index (Detrended-CCDI), and monitor drought over China from 2002 to 2017 on the basis of pixels on pixels by Detrended-CCDI. Results show that Detrended-CCDI has recorded monthly drought distribution results for China from April 2002 to June 2017, which are similar to those recorded by the Gravity Recovery and Climate Experiment - Drought Severity Index (GRACE-DSI), the Palmer Drought Severity Index (PDSI), the Standardized Precipitation-Evapotranspiration Index (SPEI), and the China Flood and Drought Disaster Bulletin. Detrended-CCDI has a good drought monitoring capability and captures drought recovery more quickly. Three main drought characteristics identified by Detrended-CCDI is similar to those demonstrated by other indices. The drought trend obtained by Detrended-CCDI is similar to that obtained by GRACE-DSI, but is more accurate and can detect more trends associated with becoming wetter or drier than PDSI and SPEI. The drought frequency obtained using Detrended-CCDI was lower than those identified by other indices, due to Detrended-CCDI combines all water layer and precipitation data. The regional drought events obtained by Detrended-CCDI and GRACE-DSI were almost identical (kappa average > 0.8), higher than those obtained by other indices. Our findings provide scientific guidance for drought monitoring and assessment over China.

Automated summary

This study proposes a new drought index (Detrended-CCDI) and uses it to monitor drought in China. The results show that Detrended-CCDI has a good drought monitoring capability and captures drought recovery more quickly than other indices. It is more accurate in detecting drought trends and can identify more trends associated with becoming wetter or drier. Additionally, Detrended-CCDI has a lower drought frequency and higher similarity in regional drought events compared to other indices.