Drought Hazard Analysis in the Jilin Province Based on a Three-Dimensional Copula Method

Abstract

Based on the precipitation and potential evapotranspiration at 23 stations in the Jilin Province, the standardized precipitation evapotranspiration index (SPEI) was calculated. Based on run theory and combined with previous research in Jilin Province, we redefine the drought threshold, with three characteristics, that is, drought duration, intensity, and severity, were identified and extracted. Two- and three-dimensional copula functions were selected to fit and combine drought variables to calculate the joint recurrence period. In order to make the alternative copula model more complete, 25 representative copula functions are selected for the first time to construct the Joint probability distribution function. The Mann–Kendall test was used to examine abrupt changes in Meihekou County. SPEI trends and the joint cumulative probability of drought variables in the Changbai area were analyzed. The results show that: (1) the SPEI in the Meihekou area was unstable around 2000 and 2010, with abrupt change points; (2) the trend of the SPEI in the Changbai Mountains area indicates that the drought occurrence is similar at various timescales; the drought range and duration increase with increasing timescale; and (3) the correlation between the three drought variables is strong and the fit is good. The results of two- and three-dimensional joint recurrence period calculations are similar, with a high probability of recurrence within three years.

Characteristics of the Spatial and Temporal Distribution of Drought in Northeast China, 1961–2020

“…Besides, while the work presented here assessed drought duration, severity, and frequency in Northeast China, there are some limitations that still need to be addressed. There are other factors that could also be considered, including drought persistence and peak, and the mutual correlations between variables could also be explored [63][64][65]. Thus, a non-stationary approach for the estimation of the joint return period of multidimensional drought variables will be a key addition in future studies.…”