【摘要】 近年来,遥感技术在中小区域的地表能量、热环境、气候研究等领域应用广泛,对高分辨率、高精度的反照率产品需求日益增多。鉴于此,提出了一种基于物理意义的降尺度方法,用于高效、准确地生成高分辨率的反照率结果。首先,假设地表朗伯反射,利用30 m的Landsat 8反射率估算得到Landsat 8初级反照率;然后,在500 m尺度上,发现分类后Landsat 8初级反照率与MCD43A3反照率相关性更好。因此,基于地表分类建立二者的线性回归函数,在MCD43A3基础上,融合30 m高分辨率Landsat 8初级反照率,得到降尺度反照率结果。与MCD43A3反照率相比,该结果提供了丰富的细节信息;利用SURFRAD观测数据的验证实验表明,降尺度反照率结果偏差为-0.01,标准差为0.012,对于不同地表类型具有良好的适用性,这说明该算法对于生产高分辨率反照率产品具有一定的应用价值。更多还原

【Abstract】 In recent years,remote sensing technology has been widely used in the field of surface energy,thermal environment,and climate research in small and medium-sized regions.The demand for high resolution,high precision albedo products has increased.In view of this,a physical-based downscaling method is proposed for efficiently and accurately generating high-resolution albedo results.First,under the Lambert hypothesis,the primary albedo of Landsat 8 could be obtained based on Landsat 8 reflectance at 30 m resolution.On the 500 m scale,it is found that after classification,the primary albedo of Landsat 8 has a better correlation with the broad-band albedo of MODIS MCD43A3 product.Therefore,a linear regression function based on surface classification is established to integrate the high-resolution primary albedo of Landsat 8 with high-precision MCD43A3 albedo to obtain downscaled albedo.Compared to MODIS albedo,the downscaled albedo provided more rich details.Verification experience based on SURFRAD observation data shows that Bias of the albedo downscaling is 0.01,and standard deviation is 0.012,which has good adaptability for different surface categories.It shows that the algorithm has great application value for producing high-resolution albedo products.更多还原