Intercomparison of Landsat albedo retrieval techniques and evaluation against in situ measurements across the US SURFRAD network

Surface albedo is an essential parameter not only for developing climate models, but also for most energy balance studies. While climate models are usually applied at coarse resolution, the energy balance studies, which are mainly focused on agricultural applications, require a high spatial resolution. In this context Landsat is one of the most used remote sensing sensors. bedo, estimated through the angular integration of the Bidirectional Reflectance Distribution Function (BRDF), requires an appropriate angular sampling of the surface. However, Landsat sampling characteristics, with nearly constant observation geometry and low illumination variation, prevent from deriving a surface albedo product. s paper we present an algorithm to derive a Landsat surface albedo based on the BRDF parameters estimated from the MODerate Resolution Imaging Spectroradiometer (MODIS) Climate Modeling Grid (CMG) surface reflectance product (M{O,Y}D09) using the VJB method (Vermote, Justice, & Bréon, 2009). We base our method on Landsat unsupervised classification to disaggregate the BRDF parameters to the Landsat spatial resolution. We tested the proposed algorithm over five different sites of the US Surface Radiation (SURFRAD) network and inter-compare our results with Shuai, Masek, Gao, and Schaaf (2011) method, which also provides Landsat albedo. The results show that with the proposed method we can derive the surface albedo with a Root Mean Square Error (RMSE) of 0.015 (7%). This result supposes an improvement of 5% in the RMSE compared to Shuai et al.ʹs (2011) method (with a RMSE of 0.024, 12%) that is mainly determined by the correction of the negative bias (lower retrieved albedo than in situ data).