【摘要】 作为研究地表能量平衡的重要参数，地表温度是研究地表物理过程、监测全球资源环境以及气候变化的重要指标之一。通过降尺度处理提高MODIS LST数据空间分辨率，对LST和气候的研究都具有重要的意义。选取黑龙江省作为研究区，选择MOD11A2数据，通过随机森林算法对MODIS LST(1 000 m空间分辨率)进行降尺度处理，并通过气象数据作为验证。试验结果表明：利用随机森林算法对黑龙江省LST数据进行降尺度，降尺度结果良好，降尺度LST的均方根误差（Root Mean Square,RMSE）为2.13 K。更多还原

【Abstract】 As an important parameter to study the surface energy balance, the surface temperature is one of the important indicators to study the surface physical process, monitor the global resource environment and climate change. The spatial resolution of MODIS LST data can be improved by downscaling processing, and the study of LST and climate is of great significance. Heilongjiang Province was selected as the study area, MOD11A2data was selected, and MODIS LST(1 000 m spatial resolution) was downscaled through random forest algorithm, and verified by meteorological data. The experimental results showed that the LST data of Heilongjiang Province were downscaled using the random forest algorithm, and the downscaling results are good. The root mean square error(RMSE)of the downscaled LST is 2.13 K.更多还原