融合Landsat ETM+和MODIS数据估算高时空分辨率地表短波反照率

Estimation and validation of high spatio-temporal resolution albedo by fusing Landsat ETM + and MODIS data

提出一种通过融合高空间低时间分辨率、低空间高时间分辨率地表短波反照率,来估算高时空分辨率地表短波反照率的方法。首先,利用Landsat ETM+数据,通过窄波段到宽波段的转换得到一景或多景空间分辨率较高的ETM+蓝天空短波反照率;然后,在MODIS短波反照率产品基础上,以天空光比例因子为权重,得到空间分辨率较低的MODIS蓝天空短波反照率;最后,利用STARFM(Spatial and Temporal Adaptive Reflectance Fusion Model)模型融合ETM+短波反照率的空间变化信息和MODIS短波反照率的时间变化信息,得到高时空分辨率的地表短波反照率。针对STARFM模型在异质性区域估算精度降低的问题,通过以MODIS反照率影像各像元的端元(各地类)反照率取代MODIS像元反照率来提取时空变化等信息参与STARFM模型的融合过程,达到提高异质性区域估算精度的目的。结果显示,直接利用STARFM模型估算得到的高空间分辨率地表短波反照率处在合理的精度范围内(RMSE<0.02),用改进后的STARFM模型估算得到的异质性区域短波反照率和真实ETM+短波反照率间的相关系数增大。

We propose an available method to generate land surface shortwave albedo at a 30 m spatial resolution in eight-day interval. The method uses at least one pair of fine-resolution remote sensing images （e.g., Landsat data） and coarse-resolution remote sensing albedo products （MODIS albedo） to estimate the 30 m albedo on the date when the MODIS albedo is given without Landsat data. When MODIS albedo time-series is given, the time-series of a 30 m resolution albedo can be estimated. The method consists of three main steps： （1） generating fine-resolution shortwave albedo images based on Landsat ETM＋ data. （2） generating coarse-resolu-tion shortwave albedo images based on MODIS albedo products. （3） using STARFM model to blend spatial information from fineresolution shortwave albedo images and temporal information fi＇om coarse resolution shortwave albedo images to estimate high spatio-temporal resolution albedo. Comparison with actual ETM＋ albedo shows that the predictions are of reasonable accuracy, and maintain a high level of spatial detail. For the problem of degrading estimation accuracy in the STARFM model for heterogeneous regions, this study presents an improvement by introducing the idea of mixed pixel decomposition. In using this method the predicted albedo shows bigger correlation coefficients and smaller deviations with actual ETM＋ albedo for heterogeneous regions.