Validation and consistency assessment of land surface temperature from geostationary and polar orbit platforms: SEVIRI/MSG and AVHRR/Metop

Abstract

Land Surface Temperature (LST) is a fundamental state variable for land surface processes, long available from satellite observations in the thermal infrared. Here we make an assessment of LST products derived from the Spinning Enhanced Visible and Infrared Imager (SEVIRI) on-board the geostationary satellite Meteosat Second Generation (MSG), and from the Advanced Very High Resolution Radiometer (AVHRR) on-board Metop polar-orbiting satellites, as provided by the Satellite Application Facility on Land Surface Analysis (LSA-SAF). The validation of LSA-SAF LST products uses ground observations taking into account the representativeness of each station at pixel scale. The in situ - satellite comparisons reveal overall accuracies of 0.13 °C for SEVIRI LST and −0.32 °C for AVHRR-based LST. Better matches are usually found for night-time observations, highlighting the influence of LST spatial and temporal variability as well as of sensor view angle on satellite daytime estimates. The averaged difference between AVHRR and SEVIRI LST products remains mostly within ±1 °C. The impact of different viewing and illumination angles is particularly pronounced for daytime matchups, leading to higher standard deviation of differences than in the case of night-time. Over heterogeneous landscapes, this primarily results from illumination-shadowing effects in daytime observations present in all LST estimates derived from infrared observations. These differences are smaller for night-time observations, or over homogeneous surfaces in general: in these cases, the dependency of LST differences on viewing geometry is smoother and likely explained by the impact of view angle on surface emissivity. Overall the LSA-SAF SEVIRI and AVHRR LST datasets are consistent and meet high accuracy standards. Their production is ensured by the LSA-SAF throughout the life-cycle of the sensors' satellite platforms, making them good candidates for long term applications and studies.