摘要
基于PROSPECT+SAIL植被辐射传输模型,通过控制不同的植被生化变量、地表参数和土壤光谱参数建立光谱数据集,定量地分析了归一化植被指数(NDVI)、比值植被指数(SR)、土壤调节植被指数(SAVI)等10种常用的植被指数(VIs)对叶面积指数(LAI)的响应.利用敏感性函数定量地筛选出具有较强适用性的转换型土壤调节植被指数(TSAVI).在此基础上,分别建立了TSAVI及常用植被指数NDVI反演LAI的模型.以张掖市南部地区的TM影像为数据源,进行了LAI的反演,并利用黑河生态水文遥感试验获得的中游LAI数据集对模型进行精度评价.结果表明:TSAVI–LAI模型最佳拟合关系为指数形式,其反演结果与LAI实测值的偏差最小(0.200),R2最大(0.686),RMSE最小(0.397).TSAVI可以作为较强适用性植被指数来进行LAI的反演.
A simulated spectral dataset was built by using PROSPECT and SAIL vegetation radiative transfer models with various vegetation biochemical parameters, surface parameters and soil spectral parameters. Based on the dataset, the responses of ten vegetation indexes （VIs）, such as commonly used normalized difference vegetation index （NDVI）, simple ratio （SR） and soil adjusted vegetation index （SAVI）, etc, to leaf area index （LAI） were quantitatively analyzed. A sensitivity analysis indicated that the transformed soil-adjusted vegetation index （TSAVI） was more applicable for LAI estimation than other VIs. Then models of TSAVI-LAI and NDVI-LAI were constructed respectively. Using TM images of the southern region in Zhangye as a data source, the LAI was inversed and validation done by employing the data set of vegetation LAI measured in the middle reaches of the Heihe River Basin. The results indicated that the TSAVI-LAI exponential model was the best for LAI inversion with a bias error of 0.200, R2 of 0.686 and RMSE of 0.397. It is concluded that TSAVI can be used as a vegetation index with strong applicability for LAI inversion.
出处
《兰州大学学报（自然科学版）》
CAS
CSCD
北大核心
2014年第1期89-94,100,共7页
Journal of Lanzhou University(Natural Sciences)
基金
中央高校基本科研业务费专项资金项目(lzujbky-2013-m02)
国家自然科学基金项目(91025010)
