论文部分内容阅读
光学遥感是目前反演植被叶面积指数LAI(Leaf Area Index)的主要手段,但是当叶面积指数较大时存在光学遥感信息饱和、反演精度显著降低的问题。叶面积指数和平均叶倾角对光学、微波波段范围内反射和散射特性都有重要影响,主要表现在植被结构参数的变化可以引起冠层孔隙率和消光截面大小的改变。本文以典型农作物玉米为例,通过构建统一的PROSAIL和MIMICS模型输入参数,生成一套玉米全生长期光学二向反射率和全极化微波后向散射系数模拟库和冠层参数库。通过对模拟数据与LAI敏感性和相关性分析得出:(1)光学植被指数MNDVI(800 nm,2000 nm),在LAI为0—3时敏感,基于MNDVI与LAI的回归模型可以估算LAI变化0.4的情况,RMSE是0.33,R2是0.958。(2)微波植被指数SAR SRVI(1.4 GHz HH,9.6 GHz HV),在LAI为3—6时敏感,基于SAR SRVI与LAI的回归模型可以估算LAI变化1的情况,RMSE为0.22,R2是0.9839。研究表明,采用分段敏感的植被指数,协同光学和微波遥感反演玉米全生长期叶面积指数是可行的。
Optical remote sensing is the main method of inverting leaf area index (LAI). However, when the leaf area index is large, there is a problem that the optical remote sensing information is saturated and the inversion accuracy is significantly reduced. Leaf area index and average leaf inclination have important influence on the reflection and scattering characteristics in the optical and microwave bands. The changes of vegetation structure parameters mainly result in changes of canopy porosity and extinction cross section. In this paper, taking typical crop corn as an example, a set of simulation library and canopy parameter library of full-length optical birefringence and all-polarization microwave backscatter coefficient were generated by constructing uniform input parameters of PROSAIL and MIMICS models. Through the analysis of sensitivity and correlation between simulated data and LAI, we can get: (1) The optical vegetation index MNDVI (800 nm, 2000 nm) is sensitive when the LAI is 0-3. Based on the MNDVI and LAI regression model, 0.4, RMSE is 0.33 and R2 is 0.958. (2) The microwave vegetation index SAR SRVI (1.4 GHz HH, 9.6 GHz HV) is sensitive when LAI is 3-6. The regression model based on SAR SRVI and LAI can estimate LAI variation 1 with RMSE of 0.22 and R2 of 0.9839 . Studies have shown that it is feasible to use section-sensitive vegetation indices, synergistic optics and microwave remote sensing to retrieve the leaf area index of maize whole growth period.