卫星遥感雪盖制图方法对比与分析

Comparison and Analysis on Methods of Snow Cover Mapping by Using Satellite Remote Sensing Data

利用ＬａｎｄｓａｔＴＭ、ＮＯＡＡ／ＡＶＨＲＲ和中分辨率成像光谱仪（ＭＯＤＩＳ）三个平台传感器的遥感数据，分别使用训练样本监督分类、阈值数字信号统计、雪盖指数方法制作雪盖图和提取积雪面积。结果表明：不同传感器遥感图像因时相和时空分辨率的差异，提取积雪信息的有效方法有所不同。但基于反射特性的雪盖指数计算法具有普遍的实际操作性意义，即雪盖制图精度高，分类合理，是提取积雪信息的最佳技术手段；当使用监督积雪分类时，只有取得精确的信号文件，分类结果才是可信的；而阈值数字信号统计的雪的阈值确定具有很大的经验性和随机性，但对数据不完整或只有单波段时也不失为有效和简便的途径；山影补偿处理法基本可以消除地形阴影的影响；而去云后其覆盖下的积雪恢复技术值得进一步讨论。

The three kinds of methods, such as training sites Supervised Classification (SC), Digital Numbers (DNs) statistics and Normalized Difference Snow Index ( NDSI) have been used to map snow cover distribution and then to calculate snow cover area. In this study, the remote sensing data sets from Landsat TM, Moderate Resolution Imaging Spectroradiometer(MODIS) and NOAA/AVHRR were selected to utilize for mapping snow cover since these images are most current remote sensing data to be applied and represent three platforms in levels. The results show that the best method obtaining snow index is different for different sensor images as reason of spatial and temporal resolution. DNs statistics to use white data series is incompletely and SC needs a quite accurate signature file for classifying. A valid and rational method has been certified that is to choose NDSI for extracting snow pixels. In detail, the impact of hillshading and cloud on snow distinguishing have also been discussed.