Abstract
Compound hot and dry events can cause greater impacts than those generated by individual extreme events. Understanding the physical mechanisms that lead to their development is particularly important for an early warning. The aim of this study is to assess the ability of global climate models (GCMs) to simulate hot/dry compound events in South America (SA) during the historical period 1979–2014, in comparison with observational and reanalysis datasets. Additionally, this work seeks to investigate the potential changes in these events under two future climate scenarios for the period 2065–2100. Furthermore, we analyze the spatial patterns of sea surface temperature anomalies (SSTA) in the Pacific Ocean associated with these events in tropical and extratropical SA. In the historical period, reanalysis tends to overestimate the number of hot/dry events, while the ensemble median of GCMs performs better than the individual ones. The future projections under the high emissions scenario show longer heat waves, but a low model agreement about the number of compound events in tropical SA. For southern SA, an increase in the annual frequency of compound events is projected, and more than two hot/dry events per year are expected to occur relative to the 1979–2014 baseline. Finally, we find that compound events in tropical SA are favored during the El Niño phase, even though two other SSTA patterns have gained prominence in recent years. In southern SA, hot/dry events are associated with the negative phase of the Pacific Decadal Oscillation and the La Niña phase.