摘要
【目的】1)评估模型的线性和非线性形式、模型残差假设对推断不确定性的效应;2)比较2种总体均值的方差估计方法(自助法和解析法);3)评估多种因素对推断不确定性的效应,构建基于遥感模型的统计推断经验法则用于指导实践。【方法】应用基于模型的统计推断方法,以森林蓄积量估算为例,基于非洲稀树草原的薪材材积实测样地数据和Landsat 8遥感辅助数据,使用二阶抽样从总体中选择160块样地形成样本,在不同模型假设下进行总体参数推断,量化分析参数模型假设对估计量不确定性的效应,并辅以置信椭圆等诊断方法确保分析的有效性。【结果】1)不同模型假设下的总体均值估计值μ_(mb)为7.159~7.331 m^(3)·hm^(-2),解析方差估计值Var(μ_(mb))为0.147~0.221,抽样精度为93.59%~96.64%,总体均值的经验方差估计值Var(μ_(mb))为0.143~0.237。模型假设会影响模型参数估计,进而影响推断精度Var(μ_(mb))。自助法是检验总体参数解析估计量无偏性的有效方法。2)基于设计的统计推断方法得出的总体均值估计值μ_(db)为6.774 m^(3)·hm^(-2),其方差估计值Var(μ_(db))为0.965,抽样精度为85.50%。既定条件下,相比基于设计的统计推断,基于模型的统计推断能够有效将推断精度提升77.10%~84.77%,对抽样精度的提升为9.46%~13.03%。【结论】基于模型的统计推断在小样本推断中具有更高的推断精度和抽样精度,有助于实现高精度、低样本量、短周期的森林资源调查目标,但建模过程中的不确定性会影响推断精度,其中残差变异性对推断不确定性的影响最大。忽略方差异性和空间自相关效应在同方差假设下进行总体参数推断,会低估Var(μ_(mb)),在考虑方差异性的同时应进一步检验空间自相关性并使用相应的权函数和自相关函数模拟残差变异性。
【Objective】1)Evaluate the effects of linear and nonlinear model forms of the model,as well as residual assumptions on inferential uncertainty.2)Compare two methods of estimating variance of the population mean-bootstrap and analytical method.3)Assess the effects of multiple factors on inferential uncertainty,and construct empirical rules of statistical inference based on remote sensing models to guide practice.【Method】160 sample plots were selected from the population using a two-stage sampling design.The variable of interest was denoted by forest volume as an example.Under the model-based inference,based on the measured sample plots of firewood volume in African savannahs and Landsat 8 remote sensing auxiliary data,the population parameters were estimated under different modeling assumptions,which aimed to quantitatively analyze the effects of analytical parameter model assumptions on estimating uncertainty and using diagnostic methods such as confidence ellipses to ensure the validity of the analysis.【Result】1)Under the different model assumptions,the population mean estimates Var(μ_(mb)) ranged from 7.159 to 7.331 m^(3)·hm^(-2).Analytical variance of the population mean estimates ranged from 0.147 to 0.221.The sampling precision ranged from 93.59%to 96.64%.Empirical variance of the population mean estimates Var(μ_(boot)) ranged from 0.143 to 0.237.Model assumptions will affect inferential the estimation of model parameters,which will ultimately affect inferential precision Var(μ_(mb)).The bootstrap method is an effective method for testing the unbiasedness of the analytical estimate of population parameters;2)Under design-based inference,the estimated mean was 6.774 m^(3)·hm^(-2)with a variance Var(μ_(db))of 0.965,the sampling precision is 85.50%.Under established conditions,compared with design-based inference,model-based inference effectively increased the inferential precision by 77.10%-84.77%and improved the sampling precision between 9.46%-13.03%.【Conclusion】Model-based inference has
作者
齐元浩
侯正阳
刘太训
徐晴
Qi Yuanhao;Hou Zhengyang;Liu Taixun;Xu Qing(Key Laboratory for Silviculture and Conservation of Ministry of Education,Beijing Forestry University,Beijing 100083;Ecological Observation and Research Station of Heilongjiang Sanjiang Plain Wetlands,National Forestry and Grassland Administration,Shuangyashan 518000;CCCC Tianjin Dredging Co.,Ltd.,Tianjin 300461;International Center for Bamboo and Rattan Key Laboratory of National Forestry and Grassland Administration/Beijing for Bamboo and Rattan Science and Technology,Beijing 100102)
出处
《林业科学》
EI
CAS
CSCD
北大核心
2024年第9期111-123,共13页
Scientia Silvae Sinicae
基金
雄安新区科技创新专项“白洋淀生态固碳能力评估与调控”(2022XACX1000)
国家社会科学基金项目“森林生态系统碳汇监测核算体系构建与评价研究”(22BTJ005)。
关键词
森林资源遥感调查
基于模型的统计推断
回归模型
方差估计
remote sensing-assisted forest inventory
model-based inference
regression model
variance estimation
