Cross-Comparison of Snow Albedo Products Derived from Satellite (Sentinel-2 and Landsat 8) Optical Data

Snow albedo is an important factor affecting global climate change. Due to the limitation of atmospheric conditions and cloud cover, remote sensing products of snow albedo have data missing and error uncertainty. In this study, based on the method of directly inverting the reflectance of the top of the atmosphere, the difference and accuracy of the snow albedo derived from the Sentinel-2 and Landsat 8 data were compared. we used two narrowband-to-broadband conversion algorithms to obtain a total of four snow albedo products. Then we use scatterplots and three indicators (the mean absolute value error, root mean square error, and pearson correlation coefficient) to compare the correlation and accuracy of the four products. The research results show that four snow albedo products derived from satellite (Sentinel-2 and Landsat 8) optical data have a strong correlation. Finally, we use IDL to check each field for regression analysis to obtain a linear regression equation. The fitting coefficient of the equation is between 0.85-0.90. The snow albedo is inverted by a variety of remote sensing data, which provides a solution to the problem of missing data.