Evaluation of evapotranspiration deficit index for agricultural drought monitoring in North China

Drought is a major natural hazard that causes significant economic loss across the world. Drought indexes derived from remote sensing data can provide dynamic information in real time, facilitating drought assessments. The evapotranspiration (ET) deficit index (ETDI) is derived from observations of ET and potential ET. ETDI is an indicator of agricultural drought. In this study, actual ET was modeled using the Boreal Ecosystem Productivity Simulator with remote sensing data and ground-station observation data as input. We investigated the efficacy of ETDI to indicate regional agricultural drought in a major region of winter wheat production in Northern China. Spatiotemporal changes in ETDI agreed well with the standardized precipitation index (SPI) on a 5 month scale and with the Palmer drought severity index (PDSI). ETDI also agreed well with the composite index of yield reduction (CI), which represents the provincial percentage area of cropland subject to drought stress. CI performed better in agricultural regions where water was supplied by precipitation rather than irrigation. Changes in winter wheat yield due to meteorological events were highly correlated with ETDI, rather than SPI or PDSI, averaged across specific growth stages. Weather conditions had significant impacts in April (jointing and boot stages of winter wheat development) in Hebei province, in May (heading and grain-filling stages) in Shanxi province, and from January through April in other provinces. These results demonstrate that ETDI is a suitable index for indicating regional agricultural drought in order to inform drought mitigation activities.