Deriving a Sea Surface Temperature Record Suitable for Climate Change Research from the along - Track Scanning Radiometers

(SST) products currently available from the series of (Advanced) Along Track Scanning Radiometers, or (A)ATSRs, are found have small biases on regional scales (typically up to 0.3 K) and excellent error statistics in validation (~0.25 K or better). Long-term stability is at present poorly characterized. Work towards " version 3 " SSTs is under way within a project called (A)RC: (A)ATSR Re-analysis for Climate. The main objectives are to reduce regional biases in retrieved sea surface temperature (SST) to less than 0.1 K for all global oceans while creating a very homogenous record with a stability (lack of drift in the observing system and analysis) of 0.05 K decade-1 , while maximizing the independence of the record from existing analyses of SST used in climate change research. If these stringent targets are achieved, this record will enable new (and significantly improved) estimates of surface temperature trends and variability of the quality required to advance questions of climate change attribution, climate sensitivity and historical reconstruction of surface temperature changes. The approach includes development of new, consistent estimators for SST for each of the ATSRs, and detailed analysis of overlap periods. Novel aspects of the approach include generation of multiple versions of the record using alternative channel sets and cloud detection techniques, to assess for the first time the effect of such choices. There will be extensive effort in quality control, validation and analysis of the impact on climate SST data sets. The need for " climate-quality " SSTs from satellite observations is clear. Both science and society need changes in global sea surface temperature (SST) over recent decades to be more robustly quantified, at global and regional scales. This paper describes a five-year project, started in 2006, to address this need, called (A)RC – (Advanced) along-track scanning radiometer Re-analysis for Climate. This highly collaborative project will lead to the " version 3 " SST record from the (A)ATSR series covering 1991 to 2007 and beyond. Creating a homogeneous SST record of " climate quality " is challenging but feasible using (A)ATSR data, and we present in this paper evidence to support this view. (A)ATSR data have several features that make them uniquely suitable. (1) (A)ATSRs are exceptionally well calibrated. The two-point calibration system is specified to drift less than 0.03 K over each sensor lifetime (1-sigma drift over ~5 years) [1,2]. (2) The instrument response functions are sufficiently well characterized to …