Determination of Solar Reflection Coefficients (Albedo) From Satellite Images Using Google Earth Engine Platform

Abstract

In many models calculating solar radiation, a combination of physical measurements and mathematical models is used to achieve results close to reality. In these calculations, the slope values and shading effects in the region being analyzed are often disregarded. Mathematical models such as ArcGIS's Area Solar Radiation (ASR) can calculate shading effects on three-dimensional surfaces. When solar radiation models are computed in three dimensions, accounting for solar rays reflected from the ground, in addition to atmospheric reflections, will increase accuracy. This study aimed to determine the surface reflectance coefficients that should be added in threedimensional radiation models. In literature, general assumptions exist for surface reflectance coefficients, which represent very broad average values. However, this study aimed to establish precise albedo values for all land classes and surfaces. An area of approximately 1600 km2 located in the mountainous region south of Karaman was chosen as the test area. This area was chosen in Karaman province because, as is known, this region has high solar energy potential. Sentinel 2A satellite images with a spatial resolution of 10 meters were used for both summer and winter seasons through the Google Earth Engine (GEE) platform. For the summer and winter applications, the albedo value for snowy surfaces was calculated as 0.86, while for light-colored buildings, it was 0.36 for summer and 0.28 for winter. Although examples were provided for some land classes, the study ultimately determined albedo values for all land surfaces without differentiation between classes.