Observed and Simulated Summer Rainfall Variability in Southeastern South America

LEANDRO DÍAZ; CAROLINA VERA; RAMIRO SAURRAL

Cape Town

Conferencia; Joint IAPSO-IAMAS-IAGA Assembly; 2017

IAPSO-IAMAS-IAGA

p { margin-bottom: 0.25cm; line-height: 120%; }Theevolution of precipitation during the last 150 years exhibits inSoutheastern South America (SESA) considerable multi-decadalvariations that have been identified in previous works as forced bythe tropical ocean variability. Therefore, in order to betterunderstand the influence of the observed large-scale interannualvariability of the sea surface temperatures (SST) on austral summerrainfall in SESA in a global warming context, a singular valuedecomposition analysis was performed over the 1902-2010 period. Theleading mode (SVD1) shows a clear global warming signal, mainlyrelated to warming in the Pacific and Indian Oceans, in associationwith a rainfall increase in SESA. The temporal series of the modeexhibits significant variability ranging from the interannual scaleto long-term trends. In particular, the decadal variability signal isremarkable, with a particular phase shift at around the middle 1970s.After detrending the series, the spatial distribution of both SSTanomalies and precipitation anomalies in SESA associated with thefirst mode resembles that typically related with El Niño-SouthernOscillation. Moreover, the mode temporal evolution has a remarkablevariability on decadal scales, which shows that the relationshipbetween SST anomalies, especially in the Tropical Pacific Ocean, andSESA precipitation is non stationary. Historicaland Decadal simulations simulations included in the Fifth phase ofthe World Climate Research Program-Coupled Model IntercomparisonProject (WCRP/CMIP5) are considered in order to make a preliminaryevaluation of the representation of austral summer rainfallvariability and trends in SESA and its connection with oceanvariability. Models are able to represent a significant positivetrend over SESA in agreement with the observations, although weakerthan observed. The model representation of the SVD1 was alsoassessed. Preliminary results using short-term climate predictionsshow that some models are able to reproduce the main spatial featuresassociated with the mode, on both SST anomalies and precipitationanomalies in SESA. In particular, these models are able to reproducethe teleconnections linking the tropical Pacific-Indian Ocean sectorwith SESA. An assessment of the temporal evolution of this modeprovided by those simulations will be also presented in theConference.