Co-variability between summer southeastern South America rainfall anomalies and tropical sea surface temperatures anomalies in CMIP5 decadal predictions

DÍAZ, LEANDRO B.; VERA, CAROLINA S.; SAURRAL, RAMIRO I.

Buenos aires

Conferencia; Humboldt Colloquium; 2018

Alexander von Humboldt Foundation

The year-to-year variability of summer rainfall anomalies in Southeastern South America (SESA) along the last century exhibits the combination of variability in different time scales (Interannual, Decadal, Multidecadal) and trends. Thus, how the rainfall in the region will evolve in the next years-decades largely depends on the combined influence of the internal natural variability, mostly associated with the tropical ocean evolution, and the external climate forcing, associated with both natural and anthropogenic sources. Therefore, the objective of this work is to describe the influence of the observed large-scale interannual variability of the sea surface temperatures (SST) on austral summer rainfall in SESA, and to assess the ability of decadal hindcast simulations from the Coupled Model Intercomparison Project Phase 5 (CMIP5) of the World Climate Research Programme in reproducing it.In order to better understand the influence of the observed large-scale interannual variability of the SST on austral summer rainfall in SESA in a global warming context, a singular value decomposition analysis was performed over the 1962-2013 period. The leading co-variability mode (SVD1) shows a clear global warming signal, mainly related to warming in the Pacific and Indian Oceans, in association with a rainfall increase in SESA. Also, the mode exhibits significant variability ranging from the interannual scale to long-term trends. In particular, the decadal variability signal is remarkable, with a particular phase shift at around middle 1970s. After detrending the series, the spatial distribution of both SST anomalies and precipitation anomalies in SESA associated with the first mode resembles that typically related with El Niño-Southern Oscillation (ENSO). Moreover, the mode temporal evolution has a remarkable variability on decadal scales, which shows that the relationship between SST anomalies, especially in the Tropical Pacific Ocean, and SESA precipitation is not stationary.The ability of a set of 5 CMIP5 models decadal hindcasts ?initialized? every year from 1960-onward to reproduce SVD1 activity was assessed. To estimate the importance of the initialization, results were compared with those obtained by historical or ?unitialized? simulations. It was found that CMIP5 decadal hindcasts are able to represent SVD1 spatial structures with and without considering trends, improving in both cases results from uninitialized simulations. Also, detrended SVD1 activity shows skill in the first two prediction years. Good skill is a result from the combination of a good representation of the low and high frequency oscillations (associated with periods longer and shorter than 5 years, respectively) in SVD1 activity. When trends are also considered, skill increase in the successive prediction years, indicating additional value from global warming effect over climate variability. SVD1 activity prediction was assessed during extreme positive ENSO events, obtaining good results for the first two prediction years. These facts represent a promising result for the predictions of rainfall in the SESA region on interannual and longer time scales.