Analysis of low frequency variability patterns and circulation regimes over southern South America and their response to global warming as depicted by IPCC AR4 AOGCMs

SILVINA SOLMAN YNATALIA PESSACG

Foz de Iguazu, Brasil

Conferencia; 8TH International Conference on Southern Hemisphere Meteorology and Oceanography.; 2006

American Meteorological Society (AMS)

The behavior of low-frequency variability patterns of the atmospheric circulation in response to increased greenhouse gas forcing is studied in an area encompassing southern South America. Last generation 20th and 21st century simulations performed with a set of 8 Atmosphere - Ocean General Circulation Models (AOGCMs) participating in the IPCC Fourth Assessment Report (AR4) were analyzed. The focuses of this study are twofold. First, to evaluate the capability of the AOGCMs in representing the observed (as depicted by NCEP reanalysis) main leading modes of low-frequency variability, defined as the leading EOFs calculated as the covariance matrix. Second, to evaluate the response of the main modes of variability to anthropogenic forcing, as depicted by the 21st century simulations under the A2 emission scenario. Moreover, climate change is being analyzed in terms of changes in circulation regimes population representing the more recurrent patterns at interannual and intradecadal time-scales. All calculations are based on monthly mean 500 hPa height. Anomalies were defined with respect to an 11-year running mean throughout the period 1950-2099. Overall, the AOGCMs analyzed are able to capture the broad spatial structure of the main sectorial modes of observed low-frequency variability, say the Antarctic Oscillation mode (AAO) ranging as the first leading mode, and the Pacific South America modes (PSA1 and PSA2) ranging second and third, respectively, though some features appear misplaced. Nevertheless, all the models systematically underestimate the explained variance associated to the PSA1 pattern, characterized by centers of anomalies extending from the tropical eastern Pacific towards southern South America, representing the atmospheric response to ENSO forcing. This common behavior may be associated with too weak ENSO like SSTs anomalies in all models. In all models, the main modes of low-frequency variability are found to remain stationary throughout the 21st century simulation, suggesting they depend more on the internal dynamics of the atmospheric flow than on its external forcing. The linear trend for the present-day climate conditions was compared against NCEP and it was found that only few models are capable to represent adequately the last 50 years observed trend. Nevertheless, all models capture the spatial structure of the trend throughout the 21st century resembling the AAO-like pattern. A cluster analysis, derived from the first few EOFs, was performed to identify large-scale patterns that appear recurrently at fixed locations. A cluster analysis provides very directly a chronology of events that can attributed to a given climate regime, and it is then of straightforward use in order to evaluate whether increasing greenhouse gas forcing induces a change in regime frequencies, a change in the spatial characteristics of the regimes, or both. This methodology allows the analysis of potential changes in the frequency distribution of the climate regimes. Inspection of the circulation regimes derived for the period 1950-2099 for all models showed no significant change in population and thus, in terms of low-frequency variability behavior, climate change manifests only as a positive trend whose spatial pattern is reminiscent of the structure of the AAO mode of variability.