Detection and Attribution of Meteorological Drought to Anthropogenic Climate Change (Case Study: Ajichay basin, Iran)

Abstract

It is not clear to what extent anthropogenic activities increase meteorological drought based on regional-scale observations. This study provides a detection and attribution (D&A) analysis of external forcing on meteorological drought using the standard precipitation index for a 12-month time scale (SPI-12) on a regional scale, particularly in the Ajichay basin, from 1972 to 2020, based on models from the Coupled Model Intercomparison Project Phase 6 (CMIP6). The Regularized Optimal Fingerprinting (ROF) method is performed for D&A analyses on two SPI-12 time series (inter-annual/decadal and long-term), which are decomposed by the Ensemble Empirical Mode Decomposition (EEMD). Observed annual precipitation, greenhouse gas (GHG) forcing, and anthropogenic-plus-natural (ALL) forcing show 62%, 33%, and 17% upward trends, respectively, based on the Mann-Kendall test. Additionally, the EEMD method reveals that the long-term trends of observed SPI-12, GHG, and ALL forcings exhibit nonlinear trends that have 7%, 3.5%, and 4.5% variance contribution rates of components, respectively. The scaling factor (β) presents the responses SPI-12 to external forcing using total least squares regression estimates in the ROF method. External forcing is detectable and attributable should β and an uncertainty range be greater than zero and spanning unity. The results show that for inter-annual/decadal SPI-12, greenhouse gas can be detected and separated from natural (NAT) and other anthropogenic forcings (\(\:\beta\:\) =0.96 with 95% confidence interval of 0.64–1.2) in single, two, and three-signal analyses. In long-term evaluations, greenhouse gas forcing (\(\:\beta\:\) =1.27 with a 95% confidence interval of 0.95–1.59) can be detected and separated from natural and other anthropogenic forcing in single, two, and three-signal analyses.