Validation of non-linear split window algorithm for land surface temperature estimation using Sentinel-3 satellite imagery: Case study; Tehran Province, Iran

In recent years, land surface temperature (LST) has become critical in environmental studies and earth science. Remote sensing technology enables spatiotemporal monitoring of this parameter on large scales. This can be estimated by satellite images with at least one thermal band. Sentinel-3 SLSTR data provide LST products a spatial resolution 1 km. research, direct indirect validation procedures were employed to evaluate the over study area different seasons from 2018 2019. The method was based absolute (direct) evaluation product field comparison (indirect) MODIS using non-linear split-window (NSW) algorithm. Also, two emissivity estimation methods, (1) NDVI thresholding (NDVI-THM) (2) classification-based (CBEM), used estimate NSW according bands images. Then, accuracy these methods estimating evaluated temporal changes vegetation, which NDVI-THM generated better results. For between product, NSW, four filters separates pairs pixels pixel quality ensure consistency compared pixel. general, results Sentinel-3, showed similar trend for during seasons. With respect comparative validations products, summer highest values bias (?1.24 K), standard deviation (StDv = 2.66 RMSE (2.43 winter lowest ones (bias 0.14 K, StDv 1.13 1.12 K) provided worst best period 2018–2019, respectively. According both results, reliable all scale our studied area.