基于叶片空间分布的植被遥感适宜尺度方法

A method for determining suitable scales for vegetation remote sensing based on the spatial distribution of leaves

高空间分辨率遥感为植被定量遥感提供了新的数据源,同时也带来了新的挑战和机遇。传统基于辐射传输理论的叶面积指数遥感方法,主要的理论依据是比尔朗伯(Beer-Lambert)定律,其前提是叶片在像元内的分布服从泊松分布,本研究探究的是连续植被叶片在像元中的空间分布服从泊松分布的情况下的适宜尺度问题。选择封垄小麦为研究对象,以小麦冠层为例,利用植被三维真实模拟模型LESS(LargE-Scale remote sensing data and image Simulation framework,LESS)模拟不同分辨率的连续小麦冠层遥感影像;在此基础上,利用三维模拟的叶片冠层分析小麦连续冠层叶片服从泊松分布的适宜尺度,构建了连续植被叶面积指数(leaf area index,LAI)反演适宜尺度的计算方法。结果表明适宜尺度受到LAI数值和聚集效应的影响。选择河南省漯河市为主要研究区,利用无人机高光谱飞行数据和LAI反演结果验证了该方法的可行性。

High spatial resolution remote sensing data serve as a new data source for quantitative remote sensing of vegetation,bringing in both new challenges and opportunities.The traditional leaf area index(LAI)inversion method based on the radiative transfer theory takes Beer-Lambert Law as the primary theoretical basis.The prerequisite for its application is that the leaf distribution in pixels follows a Poisson distribution.This study explored the appropriate scale in the case that the spatial distribution of continuous vegetation leaves in pixels follows a Poisson distribution.Focusing on the wheat canopy,this study used the LESS(LargE-Scale remote sensing data and image Simulation framework)software to simulate the remote sensing images of continuous wheat canopy.Based on this,this study analyzed the appropriate scale on which continuous wheat canopy leaves follow a Poisson distribution through the three-dimensional simulation of leaf canopy.Moreover,this study constructed a method for calculating the appropriate scale of the LAI inversion of continuous vegetation.The results show that the appropriate scale is influenced by the LAI value and the aggregation effect.The UAV hyperspectral data and the LAI inversion results from Luohe City,Henan Province validated the feasibility of this method.