Search

Abstract Compound hot and dry events (CHDEs) pose an increasing threat to global warming and have received growing attention in recent decades. We conducted a detection and attribution analysis (D&A) using the optimal fingerprint technique to analyze the changes in observed CHDEs at continental and sub‐continental scales from 1961 to 2014. The responses of CHDEs changes to anthropogenic (ANT), greenhouse gases (GHG), aerosols (AER), and natural (NAT) forcings were obtained from the Coupled Model Intercomparison Project phase 6 forcing experiment. In a two‐signal D&A analysis that considered ANT and NAT simultaneously, ANT signals were robustly detected and separated from NAT in all continental domains and more than 65% of subcontinental domains (29 out of 44 domains). The trend attributed to the ANT forcing is comparable to the observed trend over domains where the ANT signal is detectable. We also conducted a three‐signal D&A analysis that regressed observation onto GHG, AER, and NAT simultaneously. The GHG signals were detectable in all continents except Oceania, and it contributed dominantly to the observed trend over 22 out of 29 regions where ANT signals had emerged. Multivariate biases correction can largely reduce the models' biases in simulating the marginal distribution and the dependence structure, which can provide a more accurate future risk of CDHEs. We found that limiting global warming to 1.5℃ rather than 2.0℃ can significantly mitigate the impact of compound events in most areas of global land, particularly in South America, Africa, and Oceania.