Spatio-temporal drought patterns of multiple drought indices based on precipitation and soil moisture: A case study in South Korea

This study aims to explore the spatio-temporal characteristics of meteorological and agricultural droughts using the Standardized Precipitation Index (SPI) and Standardized Soil Moisture Index (SSI), respectively, as well as their relationships over the past three decades (1986-2016) in South Korea. The SSI shows less frequent droughts and longer drought duration compared to the SPI, due to the gradual decrease in the autocorrelation functions of the SSI. The strongest cross-correlations are observed at a 1-month lag between the SPI and SSI for most stations. Thus, the SPI could be more appropriate for defining the onset of a drought, whereas the SSI appears to be more effective for describing drought persistence. Moreover, the transition from meteorological to agricultural droughts is significantly dependent on the season, indicating that the transition between them is highly correlated with antecedent moisture conditions. The copula-based Multivariate Standardized Drought Index (MSDI) is introduced to explicitly postulate interdependence between the SPI and SSI in the context of a multivariate probability distribution. We employ a hierarchical agglomerative clustering approach along with a quantile regression model to better understand the spatio-temporal pattern of the MSDI. More drought episodes under moderate to severe drought conditions are observed along the southern coast of South Korea. Additionally, persistent droughts with higher severity are observed in the northern part of South Korea, which may be attributed to a significant decreasing trend (or increasing drought risk).