Drought Occurrence Probability Analysis Using Multivariate Standardized Drought Index and Copula Function Under Climate Change

Abstract

This study aims to investigate the effect of climate change on the probability of drought occurrence in central Iran. To this end, a new drought index called Multivariate Standardized Drought Index (MSDI) was developed, which is composed of the Standardized Precipitation Evapotranspiration Index (SPEI) and the Standardized Soil Moisture Index (SSI). The required data included precipitation, temperature (from CRU TS), and soil moisture (from the ESA CCA SM product) on a monthly time scale for the 1980-2016 period. Moreover, future climate data were downloaded from CMIP6 models under the latest SSPs-RCPs emission scenarios (SSP1-2.6 and SSP5-8.5) for the 2020-2056 period. Based on the NRMSE, Sn, and NS evaluation criteria, the Galambos and Clayton functions were selected to derive copula-based joint distribution functions in both periods. The results showed that more severe droughts and longer will occur in the future compared to the historical period and in particular under the SSP5-8.5 scenario. From the derived joint return period, a drought event with defined severity or duration will happen in a shorter return period as compared with the historical period. In other words, joint return period indicated a higher probability of drought occurrence in the future period. Moreover, the joint return period analysis revealed that the return period of mild droughts will remain the same, while it decresed over extreme droughts in the future.

Multivariate Drought Monitoring, Propagation, and Projection Using Bias‐Corrected General Circulation Models

“…Consequently, climate change impact projections require correcting these biases (Dieng et al, 2022;Laux et al, 2021). Multivariate bias correction (BC) strategies take into account the entire multivariate dependency structure between several dependent and independent variables (Adeyeri, Laux, Lawin, & Oyekan, 2020;Dieng et al, 2022), whereas univariate BC strategies, such as quantile mapping (Adeyeri et al, 2019) or linear scaling (Laux et al, 2021;Naderi et al, 2022), map the source distribution's quantiles to the target distribution's quantiles or applies a change factor from a particular observed variable to its model simulated counterpart. For example, in the study by Naderi et al (2022), the analysis was performed at the microscale level using climate models for drought projection.…”

“…For example, in the study by Naderi et al. (2022), the analysis was performed at the microscale level using climate models for drought projection. However, a change factor BC method was used in correcting the GCM biases, and no information was provided about drought propagation.…”

“…Also, they performed local‐scale analysis. Additionally, a holistic evaluation of the bias correction method used in correcting General Circulation Models (GCMs) projections in drought studies is lacking (Naderi et al., 2022). Nevertheless, as the climate continues to change, it is essential to comprehensively assess drought's impact, propagation, and projections in the context of the changing climate.…”

“…Naderi et al. (2022) applied the Standardized Precipitation Evaporation Index (SPEI) and the SSI for the 1980–2016 and 2020–2056 periods under two climate scenarios. Xu et al.…”

Assessment of Drought Hazard Change in China under 2 ℃, 3 ℃ and 4 ℃ Temperature Rise Scenarios Based on CMIP6 Models

“…Some studies have pointed out that compared with CMIP5, CMIP6 has improved the simulation of extreme events (Chen et al, 2020;Fan et al, 2020;Luo et al, 2020). Therefore, some researchers have analyzed the characteristics and changes in extreme climate in different regions of the world in the future based on CMIP6, such as extreme precipitation (Dike et al, 2022;Tang et al, 2021;Wu et al, 2022;Xu et al, 2022), extreme temperature (Babaousmail et al, 2022;Kuang et al, 2021;Rajulapati et al, 2022;Zhao et al, 2021), heat waves (Zachariah et al, 2021) and drought (Chen et al, 2021;Cook et al, 2020;Naderi et al, 2022;.…”