基于随机森林算法的FY-4A地表入射太阳辐射空间订正

SPATIAL CORRECTION OF FY-4A SURFACE SOLAR RADIATION BASED ON RANDOM FOREST ALGORITHM

摘要: 筛选时次、天顶角、FY-4A AGRI全圆盘地表入射太阳辐射（SSI）、云覆盖率（CFR）、云检测（CLM）以及云类型（CLT）产品构建特征向量，采用随机森林方法开展FY-4A SSI逐时产品的空间订正研究。结果表明、随机森林算法对于提升FY-4A SSI产品的空间分布精度具有明显作用，订正后相关系数、平均偏差、平均绝对误差均有不同程度的改善，且能很好地解决FY-4A SSI产品的过高估计及FY-4A SSI产品在太阳天顶角大于70°时无观测问题，有效提高FY-4A SSI产品在高纬度地区的可用性。

Abstract: FY-4A AGRI full disk surface solar radiation（SSI）,cloud fraction ratio（CFR）,cloud mask（CLM）and cloud type（CLT）products are used to construct feature vectors and the random forest algorithm is used to carry out the research on hourly spatial correction of FY-4A SSI products. The results show that the random forest algorithm has a significant role in improving the spatial distribution accuracy of FY-4A SSI products,the correlation coefficient,average deviation and average absolute error are improved to varying degrees. It can also solve the overestimation of FY-4A SSI products and the no observation problem when the solar zenith angle is greater than 70° of FY-4A SSI products,the availability of FY-4A SSI products in high latitudes is improved effectively.