Integrated drought evaluation index: considering the ecological feedback of the soil moisture and vegetation on wheat

With the acceleration of climate instability, drought is causing increasing losses that seriously threaten food security in China. In consideration of the feedback of the ecological environment vulnerability on drought, this study selects the temperature vegetation dryness index to evaluate the boundaries of the regional ecological drought index and integrates many factors, such as precipitation, temperature and human activities, from the four aspects of natural disaster risk management—hazard, vulnerability, exposure and resistance—to establish an integrated drought evaluation index for wheat. The results showed that drought was the main reason for the reduction of wheat yield. Standardized precipitation index (SPI), crop water deficit anomaly index and temperature vegetation dryness index (TVDI) were the most important factors affecting drought. On the monthly scale, precipitation showed seasonal characteristics, and the most severe water shortage period occurred from March to May, which was also the physiological water demand period of wheat; on the annual scale, the precipitation fluctuates more frequently, and the alternate from dry to wet occurs average every 3–4 years, which is basically consistent with the changing trend of yield reduction rate. Spatially, precipitation scarcities were concentrated throughout the north of the study region, where drought was most frequent and severe. There were highly positive spatial correlations between the integrated drought evaluation index and the annual yield reduction rate of wheat in dry years, whose bivariate Moran's I values reached 0.39, 0.42, 0.31 and 0.38 in the 2002, 2005, 2011 and 2016, respectively; further, the yield reduction rate increased with drought aggregation. This study clearly demonstrates that the evaluation accuracy of Integrated drought evaluation index of wheat (IDEIW) with TVDI is 11.8% higher than that of IDEIW without TVDI due to the strong effects of vegetation and soil moisture on drought. Moreover, population density has a strong impact on SPI and TVDI in the long-term time series. Therefore, the IDEIW in the study area is stronger and more stable than TVDI and SPI in terms of availability, precision and sensitivity, which can be used as an important tool to assess and monitor dynamic variations in agricultural drought and provide a new means for the early warning and forecast management of agricultural drought.