Plausible Drying and Wetting Scenarios for Summer Rainfall in Southeastern South America: a storyline approach

Summer rainfall trends in Southeastern South America (SES) have received a lot of attention in the past decades because of the socioeconomic impacts affecting an area where around 200 million people live. Literature identifies more than one driving mechanism for them, some of which have opposing effects. However, it is still not clear how much each mechanism has contributed to the observed trends or how their combined influence will affect future trends. In this work, we study SES summer rainfall future trends that can be explained by mechanisms related to large-scale extratropical circulation responses in the Southern Hemisphere (SH) to remote drivers of regional climate change. We find that regional uncertainty in SES during summer is well represented by combining the influence of four remote drivers, two of them characterizing tropical Pacific SST warming patterns, another one related to tropical upper tropospheric amplification of surface warming, and a fourth one related to the delay of the SH stratospheric polar vortex breakdown date. To do so, we quantify the contributions of the different remote drivers to SES rainfall trends and assess the sensitivity of circulation in SH to these drivers. Then, we analyze how the circulation response can mediate the relationship between regional precipitation changes and each of the remote driver responses. We use these quantifications and a storyline framework to answer the following research questions: 

• Are all of the plausible scenarios for summer rainfall in the SES region characterized by wet conditions only? Is it possible that some plausible scenarios could also involve very weak wetting or even drying trends?
• Can the differences between these contrasting scenarios be explained by variations in atmospheric circulation patterns forced by large-scale remote drivers?

By applying a storyline approach, the representation of uncertainty in summer precipitation trends in SES is improved and plausible drying scenarios are found in addition to the wetting ones. Our storyline approach is able to identify the highest impact storylines and deal with the complexity of working with four remote drivers. In addition, we show how the definition of the SES regional box can impact the results, given that if the spatial pattern characterizing the dynamical influences are complex and the impacts can be averaged out.