摘要
以贵州省黎平县为研究区,着重研究森林叶面积指数(LAI)的ETM遥感信息反演和向1km空间尺度转换算法。通过LAI-2000的针叶林和阔叶林等植被类型的LAI实地观测,建立实测LAI与ETM影像归一化植被指数(NDVI)的相关关系并进行LAI遥感制图,并在陆地覆盖类型遥感分类信息提取的基础上,发展了针叶林、混交林和空旷地三种地表类型LAI的向上空间尺度转换算法,以对粗分辨MODIS遥感数据的LAI产品实现LAI算法的转换与校正,并通过示例应用显示了本研究空间尺度转换算法的有效性。
In this article,we developed the algorithms for forest leaf area index(LAI) calculation and spatial scaling based on sub-pixel information of remote sensing image.The study area of interest is near Liping county,Guizhou Province.Methods for LAI spatial scaling were investigated on LAI images of 960m resolution derived in two ways: LAID is derived from the ETM+ data(30m),LAIL is derived from the coarse resolution,but resampled by the ETM+ data.It is found from this investigation that lumped(coarse resolution) calculations can be considerably biased from the distributed(ETM+) case,suggesting that global and regional LAI maps can be biased by the same amount if surface heterogeneity within the mapping resolution is ignored.Based on these results,algorithms for removing these biases in lumped LAI are developed using sub-pixel land cover type information.Then we apply the algorithm to correct one scene of coarse resolution MODIS LAI product and greatly improved its precision.
出处
《遥感信息》
CSCD
2006年第4期5-11,插页1,共8页
Remote Sensing Information
基金
加拿大CIDA项目"中国碳循环研究"资助
