Evaluation of Sea Surface Temperature from FY-3C VIRR Data in the Arctic

Sea Surface temperature (SST) is an essential indicator for climate change. High accuracy and stability of the satellite SST products are required for long-term climate data record of global SST. The Arctic warming is faster than the global average and has significant impact on the global climate. Satellite SST retrievals in the Arctic are more complicated than low-mid latitude due to persistent cloud, sea ice contamination and dry atmosphere. The daily 5km SST products from the Visible and Infrared Scanning Radiometer (VIRR) onboard Fengyun -3C (FY-3C) satellite are evaluated against the buoy data and the daily 4km SST products from the Moderate Resolution Imaging Spectroradiometer (MODIS) onboard the Terra satellite. The five-month data acquired during August to December, 2014 are used for the evaluation. The buoy data are averaged and resampled in the spatial resolution of 4km and 5km similar to the MODIS and VIRR SST data. The bias and standard deviation of the SST difference between VIRR and buoy data are -0.39°C and 1.25°C respectively. Averaged to 0.25° grid, daily SSTs of VIRR and MODIS are compared. The statistic shows a bias of 0.43°C and standard deviation of 0.67°C. The results indicate the accuracy of FY3C/VIRR SST products is relatively lower in the Arctic. The error sources are discussed and the SST and cloud detection algorithms need to be improved.