A Non-stationary Standardized Streamflow Index for hydrological drought using climate and human-induced indices as covariates

The aim of this paper is to propose a new non-stationarity hydrological drought index, which incorporates the climate-driven and human-induced non-stationarities in streamflow. For this purpose, significant teleconnection indices have been selected by correlation analysis to represent large-scale climate variability, and human-induced indices have been calculated using the Soil and Water Assessment Tool (SWAT) to indicate varying anthropogenic forcing. Whereafter, a non-stationary probability model fitted to streamflow series has been developed using the climate-driven and human-induced indices as covariates. Base on the non-stationary model, we present a variation of the classical Standardized Streamflow Index (SSI), named Non-stationary Standardized Streamflow Index (NSSI). Focusing on the streamflow records of Luanhe River basin from 1958 to 2011, a comparison of performance between NSSI and SSI has been conducted to demonstrate the capability of the NSSI. Finally temporal-spatial patterns of drought during the last 40 years over the basin have been estimated by using the NSSI. The results show that the non-stationary model describes the variability of streamflow better than a stationary one, and the covariates selected with Akaike information criterion (AIC) provide insights into non-stationary behaviors. Since the NSSI effectively accounts for the non-stationarities of streamflow associated with climate changes and human activities, it provides more reasonable and satisfactory results than the SSI. Additionally, it is indicated that serious long-term droughts generally appeared more frequently in the southeast of Luanhe River basin, and an obvious aggravating tendency of drought was observed in this area during 1971-2011. The presented NSSI enables hydrological droughts to be better characterized in a non-stationary context, thus providing valuable references for the improvement of drought index and the drought related policy-making.