All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 42 18 1
PDF Downloads 11 7 0

Accuracy of Satellite Land Surface Reflectance Determination

O. ArinoLERTS, Laboratoire d'Etudes et de Recherches en Télédétection Spatiale, Toulouse, France

Search for other papers by O. Arino in
Current site
Google Scholar
PubMed
Close
,
G. DedieuLERTS, Laboratoire d'Etudes et de Recherches en Télédétection Spatiale, Toulouse, France

Search for other papers by G. Dedieu in
Current site
Google Scholar
PubMed
Close
, and
P. Y. DeschampsLERTS, Laboratoire d'Etudes et de Recherches en Télédétection Spatiale, Toulouse, France

Search for other papers by P. Y. Deschamps in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

An accuracy budget of the surface reflectance determination from Meteosat geostationary satellite data is performed. Error analysis allows identification of three main problems: calibration uncertainty of the Meteosat instrument, atmospheric corrections, and surface effects (spectral and directional). Calibration accuracy is 10%, leading to a 10% relative uncertainty on reflectance. Spectral effects of the surface lead to a maximum bias of 0.01 for a vegetated surface as sensed by Meteosat, while directional effects can lead to a bias of 0.035 between two measurements taken at two different sun zenith and azimuth angles at the same view angle over savannas. The maximum error due to the atmosphere is estimated to be of the order of 0.03 in reflectance for a surface reflectance of 0.40 and 0.01 for, a surface reflectance of 0.10. Validation with in situ measurement is within the expected error over savanna. But the difference is still high over the southwest France site of HAPEX-MOBILHY, certainly due to the joint spectral and directional errors. Comparisons with surface albedo maps from literature show the same spatial and spatial evolutions with a better spatial and temporal determination in our results.

Abstract

An accuracy budget of the surface reflectance determination from Meteosat geostationary satellite data is performed. Error analysis allows identification of three main problems: calibration uncertainty of the Meteosat instrument, atmospheric corrections, and surface effects (spectral and directional). Calibration accuracy is 10%, leading to a 10% relative uncertainty on reflectance. Spectral effects of the surface lead to a maximum bias of 0.01 for a vegetated surface as sensed by Meteosat, while directional effects can lead to a bias of 0.035 between two measurements taken at two different sun zenith and azimuth angles at the same view angle over savannas. The maximum error due to the atmosphere is estimated to be of the order of 0.03 in reflectance for a surface reflectance of 0.40 and 0.01 for, a surface reflectance of 0.10. Validation with in situ measurement is within the expected error over savanna. But the difference is still high over the southwest France site of HAPEX-MOBILHY, certainly due to the joint spectral and directional errors. Comparisons with surface albedo maps from literature show the same spatial and spatial evolutions with a better spatial and temporal determination in our results.

Save