Estimating the land surface albedo using Level1-G and CDR Landsat-7 satellite imagesandnbsp;

Land surface albedo, is defined as the ratio of the surface reflected to received solar radiation. It is one of the major meteorological variables required for calculation of the surface radiation budget, crop water requirement and boundary conditions of climate models. During the last several decades, the albedo has been estimated using remotely sensed data. In this study, land surface albedo is estimated by using Level1-G and CDR Landsat-7 satellite images and compared with in-situ data in a farmland located in Hamedan province in the proximity of the climatology station located at faculty of agriculture, Bu-Ali Sina University. For this purpose, 14 satellite images were used. In-situ surface albedo data were also obtained from the net radiometer installed at the study site. Surface albedo was calculated at various wavelengths after removing SLC-Off, cloud effect, and atmospheric and topographic errors.The results showed that the estimated green cover albedo values were within the range of 0.21-0.26and snow cover albedo was approximately 0.65. Estimates of surface albedo from CDR and Level1-G images were evaluated against the surface measurements by using root mean square errors (RMSE). The calculated RMSE was within the range of0.02 to 0.09. Furthermore, the maximum error of estimated albedo was observed in NIR wavelengths. According to the Fisher test results, there was no significant difference between the estimated surface albedo from CDR and Level1-G images. Hence, for crude estimation of surface albedo, both approaches can be recommended with no priority; especially, for coarse regional estimates.