Surface melting observations in Antarctica by microwave radiometers: Correcting 26-year time series from changes in acquisition hours

Abstract Surface melting duration and extent of the Antarctic coasts and ice-shelves is a climatic indicator related to the summer temperature and radiative budget. Surface melting is easily detectable by remote sensing using passive microwave observations. The preliminary goal of this study is to extend to 26 years an existing data set of surface melting [Torinesi, O., Fily, M., Genthon, C. (2003), Interannual variability and trend of the Antarctic summer melting period from 20 years of spaceborne microwave data, J. Climate, 16(7), pp. 1047–1060] by including the most recent years of observation. These data come from 4 microwave sensors (the Scanning Multichannel Microwave Radiometer (SMMR) and three Special Sensor Microwave Imager (SSM/I)) observing the surface at different hours of the day. Since surface melting varies throughout the day as the air temperature or the radiation, the interannual melting extent and duration time series are biased by sensor changes. Using all the sensors simultaneously available since 2002, we were able to model the diurnal variations of melting and use this hourly model to correct the long-term time series. This results in an unbiased 26-year long time series better suited for climate analysis. The cooling trend found by Torinesi et al. using uncorrected time series for the 1980–1999 period is confirmed but the decreasing rate is weaker after correction. Furthermore, extending the series up to summer 2004–2005 reveals recent changes: the last 2 summers have been particularly warmer over all the East Antarctica compared to the 10 previous years, thus ending the cold period of the 1990s. The trend over 1980–2005 is no longer toward cooling but complex climatic variations appear. The time series are available at http://www.lgge.obs.ujf-grenoble.fr/~picard/melting/ .