Copulas-Based Drought Characteristics Analysis and Risk Assessment across the Loess Plateau of China

The Loess Plateau (LP) of China is famous with soil erosion and water shortage problems. Droughts were frequently occurred in this region, which becomes a critical limiting factor to the socioeconomic development, ecology and food production. Therefore, the major motivation of the present study is to investigate the drought characteristics and assess the potential drought risk in this area, which is crucial for drought resistance, water resource management as well as agricultural production. This study analyzes the variations of meteorological drought, characterized by the Standardized Precipitation Evapotranspiration Index (SPEI), and assesses the drought hazards in the LP during 1950-2014. The results show that the northwest of LP is more likely to experience long duration and large severity droughts than the southeast of LP. From the perspective of statistical probability models, the exponential distribution and Gamma distribution can well fit the drought duration and severity, respectively. Compared to Frank and Clayton copula, the Gumbel copula can better model the dependence structure between the drought variables in our study area. Moreover, the estimation of the upper tail dependence coefficient between drought duration and severity also demonstrate that Gumbel copula can provide the best description of the upper tail. The spatial distribution of joint return period under different cases indicates that drought risk in northwestern LP is relatively higher than that in other areas of LP. The results presented in this study can provide some scientific basis for the strategic planning of drought resistance and water resource management in the LP.