基于地理环境要素的叶面积指数遥感定量反演

Inversion of Leaf Area Index Based on Geographical Environment Factors

叶面积指数(LAI)是分析冠层结构最常用的参数之一,它控制着植被的生物、物理过程,如光合、呼吸、蒸腾、碳循环和降水截获。但是通过地面直接测量来获取大面积的LAI十分困难,而传统的基于单植被指数的LAI反演方法也具有一定的缺陷。以福州市辖区与闽侯县的阔叶林和处于生殖生长阶段的水稻为研究对象,在传统单植被指数的LAI反演方法的基础上引进植被含水量、植被覆盖度和地形3个核心环境因子来建立LAI估算模型。结果表明:基于最佳植被指数与环境因子的LAI估算模型与未考虑环境因子的单植被指数LAI估算模型相比,其验证精度有所提高。其中,就阔叶林的LAI定量反演模型而言,R2由0.706~0.717提升至0.755,RMSD由0.292~0.297降低至0.271;就生殖生长阶段的水稻LAI定量反演模型而言,R2由0.724~0.879提升至0.952,RMSD由0.696~1.054降低至0.441,实现了较高精度的LAI定量反演模型,为福州市辖区及其周边闽侯县区域的LAI快速定量监测奠定基础。

Leaf area index (LAI) is one of the most commonly used parameters in the analysis of canopy structure.Based on the traditional inversion method of leaf area index of single vegetation index, three core environmental factors, vegetation water content, vegetation coverage and landform, are introduced to establish the estimation model of LAI in Fuzhou municipal districts and Minhou County. The results show that the estimation model of leaf area index based on optimal vegetation index and environmental factor is better than that of single vegetation index estimation model without environmental factor. In terms of quantitative inversion model of leaf area index of broad-leaved forest, R 2 increased from 0.706~0.717 to 0.755, RMSD reduced from 0.292~0.297 to 0.271. As far as the quantitative inversion model of leaf area index for rice at reproductive growth stage is concerned, R 2 raised from 0.724~0.879 to 0.952, RMSD reduced from 0.696~1.054 to 0.441. The relatively high precision quantitative inversion model of leaf area index is realized, which lays a foundation for rapid quantitative monitoring of leaf area index in Fuzhou municipal district and its surrounding Minhou county.