Recent climate variability in Antarctica from satellite-derived temperature data

Recent Antarctic climate variability on month-to-month to interannual time scales is assessed through joint analysis of surface temperatures from satellite thermal infrared observations (T-IR) and passive microwave brightness temperatures (T-B). Although T-IR data are limited to clear-sky conditions and T-B data are a product of the temperature and emissivity of the upper; 1 m of snow, the two datasets share significant covariance. This covariance is largely explained by three empirical modes, which illustrate the spatial and temporal variability of Antarctic surface temperatures. The T-B variations are damped compared to T-IR variations, as determined by the period of the temperature forcing and the microwave emission depth; however, microwave emissivity does not vary significantly in time. Comparison of the temperature modes with Southern Hemisphere (SH) 500-hPa geopotential height anomalies demonstrates that Antarctic temperature anomalies are predominantly controlled by the principal patterns of SH atmospheric circulation. The leading surface temperature mode strongly correlates with the Southern Annular Mode (SAM) in geopotential height. The second temperature mode reflects the combined influences of the zonal wavenumber-3 and Pacific - South American (PSA) patterns in 500-hPa height on month-to-month time scales. ENSO variability projects onto this mode on interannual time scales, but is not by itself a good predictor of Antarctic temperature anomalies. The third temperature mode explains winter warming trends, which may be caused by blocking events, over a large region of the East Antarctic plateau. These results help to place recent climate changes in the context of Antarctica's background climate variability and will aid in the interpretation of ice core paleoclimate records.