基于随机森林方法的中国地表短波辐射估算  

ESTIMATION OF SURFACE INCIDENT SHORTWAVE RADIATION OVER CHINA BASED ON RANDOM FOREST REGRESSION METHOD

作　　者：侯宁 张晓通[1] 魏瑜 张为宇 姚云军 贾坤 Hou Ning;Zhang Xiaotong;Wei Yu;Zhang Weiyu;Yao Yunjun;Jia Kun(State Key Laboratory of Remote Sensing Science,Institute of Remote Sensing Science and Engineering,Faculty of Geographical Science,Beijing Normal University,Beijing 100875,China)

机构地区：[1]北京师范大学遥感科学国家重点实验室,遥感科学与工程研究院,地理科学学部,北京100875

出　　处：《太阳能学报》2021年第2期31-36,共6页Acta Energiae Solaris Sinica

基　　金：国家重点研发计划(2017YFA0603002);国家自然科学基金面上项目(41571340)。

摘　　要：利用中国气象局的辐射观测资料和气象观测资料,基于随机森林方法构建地表短波辐射估算模型,估算模型的相关系数为0.99,绝对偏差为0.01 W/m^2,均方根误差为8.88 W/m^2。利用估算模型构建中国1958~2015年长时间序列地表短波辐射数据集,通过与已有数据集的精度对比,发现所构建的地表短波辐射数据集精度更高。In this study,the Random Forest(RF)algorithm is used to reconstruct the surface incident shortwave radiation(SSR)using the ground measured SSR data and the routine meteorological station data from China Meteorological Administration(CMA).The estimated SSR is validated using the independent ground measurements with an overall correlation coefficient(R)value of 0.99,a BIAS value of 0.01 W/m^2,and a root mean square error(RMSE)value of 8.88 W/m^2.The RF based model is then used to generate a 58-years(1958-2015)daily SSR dataset at the routine meteorological CMA stations.

关 键 词：短波辐射 随机森林 时空分布 全球变亮 全球变暗 

分 类 号：P422.1[天文地球—大气科学及气象学]