Validation and Comparison of ANEM-SBAC Method for Temperature and Emissivity Separation Using MODIS and VIIRS Data in the València Test Site

Abstract

Land surface temperature (LST) is critical for understanding Earths climate. Remote sensing techniques are useful for retrieving LST using thermal radiance alongside assumptions about emissivity. To address the temperature-emissivity separation (TES) problem, this study employs the adjusted normalized emissivity method with single-band atmospheric correction (ANEM-SBAC). ANEM modifies the TES NEM module, using the vegetation cover method (VCM) for initial emissivity estimation from fractional vegetation cover (FVC). The SBAC method atmospherically corrects thermal radiance and solves the radiative transfer equation (RTE) to retrieve LST. This methodology assigns appropriate initial emissivity values pixel-by-pixel with robust SBAC atmospheric correction. It avoids emissivity underestimation typically observed in TES over lowcontrast surfaces caused by residual atmospheric correction errors and band noise, improving TES accuracy and applicability. ANEM-SBAC is validated via the T-based methodology using ground measurements over several surface types (vegetation, water, soil) at the Valencia Test Site and compared to MODIS and VIIRS standard LST/LSE products based on Split-Window (SW) (MXD11, VLSTO products) and TES (MXD21, VNP21 products) algorithms. ANEM-SBAC achieves RMSE values of 1.1 K (MODIS-Terra), 1.3 K (MODIS-Aqua), and 1.5 K (VIIRS). TES reaches RMSE values of 1.4 K (MODIS-Terra), 1.7 K (MODIS-Aqua), and 1.7 K (VIIRS), while SW yields RMSE values of 1.2 K (MODIS-Terra), 0.9 K (MODIS-Aqua), and 1.9 K (VIIRS). A robust estimation method and sensitivity analysis to qualify and quantify uncertainties are also offered, serving as a reliable operational alternative and supporting adaptable solutions to improve the knowledge of the temperature-emissivity separation problem.