Analysis of oceans' influence on spring time rainfall variability over Southeastern South America during the 20th century

Southeastern South America (SESA) rainfall is influenced by the tropical Pacific, Atlantic and Indian Oceans. At the same time, these tropical oceans interact with each other inducing sea surface temperature anomalies in remote basins through atmospheric and oceanic teleconnections. In this study, we employ a tool from complex networks to analyse the collective influence of the three tropical oceans on austral spring rainfall variability over SESA during the 20th century. To do so we construct a climate network considering as nodes the observed Nino3.4, Tropical North Atlantic (TNA), and Indian Ocean Dipole (IOD) indices, together with an observed and simulated precipitation (PCP) index over SESA. The mean network distance is considered as a measure of synchronization among all these phenomena during the 20th century. The approach allowed to uncover two main synchronization periods characterized by different interactions among the oceanic and precipitation nodes. Whereas in the 1930s El Nino and the TNA were the main tropical oceanic phenomena that influenced SESA precipitation variability, during the 1970s they were El Nino and the IOD. The influence of El Nino on SESA precipitation variability might be understood through an increase of the northerly transport of moisture in lower-levels and advection of cyclonic vorticity in upper-levels. On the other hand, the interaction between the IOD and PCP can be interpreted in two possible ways. One possibility is that both nodes (IOD and PCP) are forced by El Nino. Another possibility is that the Indian Ocean warming influences rainfall over SESA through the eastward propagation of Rossby waves as suggested previously. Finally, the influence of TNA on SESA precipitation persists even when the El Nino signal is removed, suggesting that SST anomalies in the TNA can directly influence SESA precipitation and further studies are needed to elucidate this connection.