摘要
极化干涉合成孔径雷达PolInSAR(Polarimetric Interferometry Synthetic Aperture Radar)已被广泛用于森林高度的反演,正确评估模型输入参数、模型假设、林分结构、立地条件等引起的不确定性是提高基于PolInSAR技术森林高度反演精度及准确性的关键之一。本文以贝叶斯模型为基础,以模拟的L波段PolInSAR数据为数据源,首先基于贝叶斯模型确定了随机体散射RVoG(Random Volume over Ground)模型输入参数引起的不确定性,在此基础上使用先验知识(成像中森林高度的值)对RVoG模型的消光系数进行“固定”,并反演了森林高度;然后基于RVoG模型反演结果及贝叶斯后验采样分析,讨论了树种、森林密度、地面粗糙度及土壤含水量四个因子变化引起的森林高度反演结果的不确定性。研究结果表明:对于L波段的PolInSAR模拟数据,采用RVoG模型进行森林高度反演时,使用先验知识对消光值进行固定可大大降低森林高度反演的不确定性;四个因子中,树种和森林密度引起的不确定性较显著,然后为地面粗糙度,最后为土壤含水量。阔叶林反演结果的不确定性明显高于针叶林;森林密度从150株/hm2增至1200株/hm2时,其标准误最高可下降67.5%;在针叶林纯林和阔叶林纯林中,地面粗糙度与反演结果的标准误呈现明显的正相关关系;土壤含水量引起的不确定性最小,几乎可以忽略不计。
Polarimetric Interferometry Synthetic Aperture Radar(PolInSAR)has been widely used in forest height inversion.Accurate evaluation of the uncertainty caused by model input parameters,model assumptions,stand structure,and site conditions can improve the accuracy of forest height inversion with PolInSAR technology.In practical application,the study on uncertainty of forest height inversion is as important as of forest height estimation methods.Quantification of global carbon stocks based on forest biomass calculations usually requires reducing the error in biomass estimates through forest height.The uncertainty of forest height may be attributed to model input parameters,model assumptions,observed data,and forest scene factors.However,comprehensive collaborative impact analyses on the uncertainty of forest height inversion results are few.On this basis,the uncertainty of forest height inversion should be studied using PolInSAR technique.We initially analyze the uncertainty caused by the input parameters of the RVoG(Random Volume over Ground)model based on the Bayesian model using the simulated L-band full PolInSAR data,and then prior knowledge(value of the forest height in the imaging)is applied to fix the extinction of the RVoG model.Subsequently,we inversed the forest height.The results show that a priori knowledge can greatly reduce canopy height uncertainties in some cases.On this basis,we combine the RVoG model and Bayesian framework,use L-band simulated PolInSAR data,and comprehensively explore the uncertainties that result from the input parameters of the RVoG model,model hypothesis,observation value,changes in forest tree species,forest density,surface properties,ground moisture content,and other factors in the process of forest height inversion.The research results indicated that:(1)prior knowledge can reduce the uncertainty of the forest height inversion(by fix the extinction value)with RVoG model and L-band PolInSAR data.(2)The forest height inversion results are greatly affected by forest tree species,an
作者
张庭苇
张王菲
张永鑫
黄国然
ZHANG Tingwei;ZHANG Wangfei;ZHANG Yongxin;HUANG Guoran(Forest College,Southwest Forestry University,Kunming 650224,China)
出处
《遥感学报》
EI
CSCD
北大核心
2023年第10期2431-2444,共14页
NATIONAL REMOTE SENSING BULLETIN
基金
国家自然科学基金(编号:32160365,32371869,42161059,31860240)
云南省农业基础研究联合专项面上项目(编号:202301BD070001-058)
云南省教育厅科学研究基金(编号:2019J0182,2020Y0393)。
