Inter-comparison of multiple angle remotely sensed data across different spatial resolutions and sensors for determination of albedo

Abstract

Surface albedo is one of the critical parameters required by studies of surface energy balance and climate models. Albedo is defined as the ratio of outgoing radiances to incoming irradiances over hemispherical view-illumination geometry. Remotely sensed albedo is usually based on multiple view angle observations and a Bidirectional Reflectance Distribution Function (BRDF) model. The accuracy of remotely sensed albedo depends on a variety of factors of which the main ones are the accuracy of atmospherically corrected observations, the reliability of observations, and the validity of the applied BRDF model. Fine spatial resolution airborne and satellite data are valuable for the validation of coarse spatial resolution satellite albedos as they may be validated using field measurements with higher reliability. In this study, a variety of remote sensing data and field measurements were used to estimate, validate and analyse albedo at different spatial resolutions. The main aim was to validate the MODIS albedo product under UK conditions using the methods of direct and indirect comparisons with other available data. The source of the fine spatial resolution data used was the NCAVEO Field Campaign 2006 that took place at the Chilbolton calibration test site in southern England. The CHRIS/PROBA albedo was used as a fine spatial resolution (34 m) albedo map to investigate the spatial variation of albedo. The results of this investigation provided valuable information about the possibility of the extension of the obtained albedo map from CHRIS/PROBA data. The MODIS albedo product with a coarser spatial resolution (500 m), relative to the NCAVEO datasets, was compared with the CHRIS/PROBA albedo map to examine the effect of spatial scale on the accuracy of albedo (direct comparison). The uncertainties in the obtained albedo maps, from both MODIS and CHRIS/PROBA, were mainly examined by testing the accuracy of the input reflectance data and the applied BRDF model (indirect comparison). The results showed the accuracy of the MODIS albedo product inferior to that claimed by the MODIS team.

More information

Identifiers

ORCID for P.M. Atkinson: orcid.org/0000-0002-5489-6880

Catalogue record

Export record