Initial results of the land surface temperature (LST) validation with the Evora, Portugal ground-truth station measurements

Meteosat second generation (MSG) provides multispectral data with a high temporal resolution of 15 min. On the basis of these data, the Land Surface Analysis-Satellite Applications facility (LSA-SAF), patronized by Eumetsat, generates and distributes many MSG products. The land surface temperature (LST) is one of its main products. It is generated from the MSG spinning enhanced visible and infra red imager (SEVIRI) data with a split-window algorithm, using 10.8 and 12.0 mu m channels. Based on the accuracy of the parameters used in the LST algorithm, the SAF-LST quality is estimated to be +/- 2 degrees C. However, this estimation must be proven with ground-truth data. This article presents the concept of the first LST validation station for the MSG in Evora, Portugal. The validation results calculated for cloud-free days during 2006 reveal very good agreement of the satellite- and ground-based LST at night (average difference of 1.0 +/- 1.5 degrees C). During the day, however the average bias reaches 3.0 +/- 1.5 degrees C. The largest disagreements take place on hot summer days with very intense direct solar radiation. This systematical bias reveals that the algorithm overestimates LST at high surface temperatures.