GCOM-W1 AMSR2 and MetOp-A ASCAT wind speeds for the extratropical cyclones over the North Atlantic

Abstract

This study presents the results of the comparison between sea surface wind speed (SWS) estimates made by MetOp-A scatterometer ASCAT and GCOM-W1 Advanced Microwave Sounding Radiometer 2 (AMSR2) relating to winds developing in the extratropical cyclones (ECs) over the North Atlantic. One season of winter ECs of 2012–2013 is considered in which 33 most intensive ECs are selected for the study. ASCAT Level 2 operational coastal wind vector product is used for the comparison, whereas AMSR2 SWS values are calculated with newly developed algorithms. Two algorithms for SWS retrieval from AMSR2 are based on numerical simulation of AMSR2 brightness temperatures (TB) over the oceans and their following inversion with Neural Networks. The first algorithm uses TB measurements at higher frequency AMSR2 channels (HF algorithm), the second one uses TB measurements at lower frequency channels (LF algorithm). It is demonstrated that both AMSR2 SWS estimates are highly correlated with ASCAT SWS for the range of low and moderate wind speeds which opens new potential for possible merging of active and passive microwave SWS products. It is shown also that LF AMSR2 algorithm overestimates ASCAT winds greater than 15m/s. To support the comparison results SWS estimates by AMSR2 are validated against in-situ wind speed measurements from platform weather stations in the North Sea and Norwegian Sea, including high wind events.