Validation of S-NPP VIIRS Provisional Land Surface Temperature Product

The Visible Infrared Imaging Radiometer Suite (VIIRS), aboard the Suomi National Polar-orbiting Partnership (S-NPP) satellite, is the sensor to provide measurements of the atmospheric, land and oceanic parameters which are referred to as Environmental Data Records (EDRs). Land Surface Temperature (LST) EDR is the measurement of the skin temperature over global land coverage including coastal and inland-water. The VIIRS LST EDR is produced from a baseline split-window regression algorithm. Coefficients of the LST algorithm are surface type dependent, referring 17 International Geosphere-Biosphere Programme (IGBP) types. This study presents validation works of the provisional VIIRS LST generation, which was approved by the JPSS Algorithm Evaluation Review Board (AERB) in June 2013. The provisional LST data will be available to public in November, 2013, through NOAA's Comprehensive Large Array-data Stewardship System (CLASS). The validation is mainly carried out using two approaches including the conventional temperature-based approach by comparisons between the VIIRS LSTs and in-situ LSTs, and the radiance based simulation approach.

The evaluation results suggest that the VIIRS LST EDR meets the provisional maturity criteria and it is ready for operational evaluation. And, the result shows an overall improved performance over beta maturity. However, provisional VIIRS LST presents an uneven performance over surface types. For example, the VIIRS LST EDR does not meet the requirement over cropland surface type according to the ground evaluation. The further efforts are being made toward the water vapor correction and the improvement of the LST inconsistency by adding emissivity information into the coefficient.