GC51D-0455:
A Novel Method making direct use of AIRS and IASI Calibrated Radiances for Measuring Trends in Surface Temperatures

Friday, 19 December 2014
Hartmut H Aumann and Alexander Ruzmaikin, NASA Jet Propulsion Laboratory, Pasadena, CA, United States
Abstract:
Making unbiased measurements of trends in the surface temperatures, particularly on a gobal scale, is challenging: While the non-frozen oceans temperature measurements are plentiful and accurate, land and polar areas are much less accurately or fairly sampled. Surface temperature deduced from infrared radiometers on polar orbiting satellites (e.g. the Atmospheric Infrared Sounder (AIRS) at 1:30PM, the Interferometer Atmosphere Sounding Interferometer (IASI) at 9:30 AM and the MODerate resolution Imaging Spectro-radiometer (MODIS) at 1:30PM), can produce what appear to be well sampled data, but dealing with clouds either by cloud filtering (MODIS, IASI) or cloud-clearing (AIRS) can create sampling bias. We use a novel method: Random Nadir Sampling (RNS) combined with Probability Density Function (PDF) analysis. We analyze the trend in the PDF of st1231, the water vapor absorption corrected brightness temperatures measured in the 1231 cm-1 atmospheric window channel. The advantage of this method is that trends can be directly traced to the known, less than 3 mK/yr trend for AIRS, in st1231. For this study we created PDFs from 22,000 daily RNS from the AIRS and IASI data. We characterized the PDFs by its daily 90%tile value, st1231p90, and analysed the statistical properties of the this time series between 2002 and 2014. The method was validated using the daily NOAA SST (RTGSST) from the non-frozen oceans: The mean, seasonal variability and anomaly trend of st1231p90 agree with the corrsponding values from the RTGSST and the anomaly correlation is larger than 0.9. Preliminary results (August 2014) confirm the global hiatus in the increase of the globally averaged surface temperatures between 2002 and 2014, with a change of less than 10 mK/yr. This uncertainty is dominated by the large interannual variability related to El Niño events. Further insite is gained by analyzing land/ocean, day/night, artic and antarctic trends. We observe a massive warming trend in the Artic between 2002 and 2007, which has since level off, but no significant trend in the Antarctic. The AIRS results since 2002 are confirmed by IASI data since 2007.

This work was supported by the Jet Propulsion Laboratory of the California Institute of Technology, under a contract with the National Aeronautics and Space Administration.