论文部分内容阅读
MODIS等全球叶面积指数(LAI)产品空间分辨率偏低(250m～7km),不能满足高空间分辨率遥感应用的需求。为获取大区域高空间分辨率LAI,有必要对物理模型用于高空间分辨率遥感影像LAI反演的可行性进行探讨。本文基于4-scale模型LAI反演算法,以甘肃省张掖为研究区,利用TM影像实现研究区人工林LAI反演。反演算法考虑了反射率入射-观测角度对LAI与植被指数关系的影响和植被冠层尺度的集聚程度。利用地面实测LAI数据对反演结果进行验证与分析,并与NDVI经验模型进行对比,同时分析LAI反演结果对波段反射率敏感性。结果表明:4-scale模型LAI反演结果与实测LAI一致性良好(R2=0.67,RMSE=0.50),且优于NDVI经验模型(R2=0.59,RMSE=0.67);当LAI大于2时,4-scale模型LAI反演算法误差小于NDVI经验模型,能有效避免植被指数的饱和现象;红光波段反射率减小时,引起4-scale模型LAI反演结果的变化幅度比其增大时更高,且影响程度大于近红外波段反射率。研究表明,4-scale模型LAI反演算法可用于TM数据反演人工林LAI,模型应用普适性较强。
The spatial resolution of MODI such as Global Leaf Area Index (LAI) products is low (250m ~ 7km), which can not meet the needs of high spatial resolution remote sensing applications. In order to obtain high spatial resolution LAI in large area, it is necessary to discuss the feasibility of physical model for LAI inversion of high spatial resolution remote sensing image. Based on 4-scale model LAI inversion algorithm and Zhangye of Gansu Province as a research area, LAI inversion of plantation in the study area was realized by using TM images. The inversion algorithm takes into account the influence of reflectivity incidence-observation angle on the relationship between LAI and vegetation index and the extent of vegetation canopy scaling. The inversion results were validated and analyzed by LAI data measured on the ground and compared with the NDVI empirical model. Meanwhile, the sensitivity of LAI inversion to the band reflectance was also analyzed. The results show that LAI inversion results of the 4-scale model are in good agreement with the measured LAI (R2 = 0.67, RMSE = 0.50) and better than the NDVI empirical model (R2 = 0.59, RMSE = 0.67) The LAI inversion error of -scale model is less than that of NDVI empirical model, which can effectively avoid the saturation of vegetation index. When the reflectivity of red band decreases, the LAI inversion result of 4-scale model is higher than that of LAI inversion. And the impact is greater than the reflectivity of the near infrared band. The results show that the LAI inversion algorithm of 4-scale model can be used to retrieve LAI of TM plantation data, and the model is more universal.