Reconstructing daily snow and ice albedo series for Greenland by coupling spatiotemporal and physics-informed models

• DOI: 10.1016/j.jag.2023.103519
• Publication Date: 2023-10-11T23:19:07Z
• Authors: Fan Ye; Qing Cheng; Weifeng Hao; Dayu Yu; Chao Ma; Dong Liang; Huanfeng Shen
• Keywords: Albedo (alchemy); Greenland ice sheet; Ice-albedo feedback

Snow and ice albedo is a critical geographical indicator that reflects climate change on Earth. Quantifying the in Greenland sheet, which extensively covered with snow ice, key to studying changes energy budget Northern Hemisphere. Earth observation satellites have been regularly providing surface products. However, optical satellite-derived products many voids due persistent cloud cover over sheet. Consequently, seamless reconstruction of spatial temporal scales essential. Surface albedo, as element, spatially temporally correlated. In addition, broadband significantly modified by spectral distribution solar irradiance caused clouds. On basis such facts, this study proposes method for combines spatiotemporal information physics-informed model. This uses nonlocal filtering generate initial reference missing pixels. Then, hypothetical clear-sky reconstructed using Whittaker iterator. Finally, cloudy obtained empirical relationship between cloudy-sky albedos. We reconstruct based MOD10A1 whole region from 2001 2020. A comparison results ground measurements exhibits satisfactory accuracy an R-value 0.8162, root-mean-square error 0.0669, mean absolute 0.0486, bias 0.00001. Moreover, proposed demonstrates advantages being more accurate robust than other classical methods. Therefore, will be valuable generating 500 m daily remotely sensed large regions.