Recent intensification of extreme precipitation events in the La Plata Basin in Southern South America (1981-2018)

Changes and trends in ten extreme daily rainfall indices in the La Plata Basin (LPB) during the 1981-2018 period were analyzed. These indices were evaluated for three regions in the LPB, namely the northern, central, and southern LPB (N-LPB, C-LPB, and S-LPB, respectively). A decrease in the maximum rainfall over five consecutive days (RX5day) and an increase in the consecutive dry days (CDD) were observed in N-LPB, mainly during the austral winter. An increase in the total wet-day precipitation was found in S-LPB, particularly during September-November (SON), and during December-February (DJF). The number of heavy precipitation days (R10mm) and CDD indices in S-LPB during SON showed a breaking point in 1999 and were followed by an increase in R10mm after 1999. Therefore, we discussed the large-scale climatic features that could be related to the breaking point in the extreme precipitation indices found in the S-LPB during SON, through correlation analysis. Our results indicate that R10mm shows significant positive correlations with the sea surface temperature (SST) in the Southwestern Subtropical Atlantic Region, mainly between 32 degrees S-22 degrees S and 50 W-34 degrees W. The SST anomaly index in this oceanic region for SON during 1981-2018 displayed a significant upward change in 1999. Furthermore, the correlation map between this index and the vertically integrated moisture flux (VIMF) and its divergence, shows negative correlations in the S-LPB, which means a decrease of the moisture divergence in this region. This implies an increase (decrease) of R10mm, and a decrease (increase) of CDD in the S-LPB after (before) 1999. This relationship results from an anticyclone in the Southwestern Atlantic extending into the LPB up to 60 degrees W, which causes an increase in the northwesterly moisture flux over central and southern parts of the basin, enhancing the moisture convergence over S-LPB. These insights have not been discussed previously, and might be relevant for future modeling studies and monitoring tasks.