单木生物量平差随机模型的比较与分析

Comparison and analysis of random model of single tree biomass adjustment.

为解决生物量随机模型的确定、形式、意义和权函数的统计性能等问题,该文从数据采集出发重点分析归纳了生物量模型拟合误差的来源,证明了单木生物量模型拟合误差的异方差性的特点,并揭示了其方差与单木相应的鲜重成正比的本质.该文还详细论述了几种常用随机模型的特点及其在生物量模型研究中的适用性,并以函数模型为示例,着重分析比较了采用不同随机模型对回归参数及生物量估值的影响及所确定的随机模型的物理意义,并引入稳健估计抗差理论削弱粗差的影响.

Several aspects regarding the biomass random model,like its establishment,forms,significance and statistical properties of its weight functions,were discussed.From the aspect of data collection,we analyzed the sources of fitting errors in biomass model,testified the heteroscedasticity of the fitting errors in the single tree biomass model and revealed that a positive relationship existed in the variance and corresponding fresh weight of the single tree.The characteristics of some common used random models were also discussed in detail as well as their application in the research of biomass.Taking the function model W=a(HD^2)~b as an example,we particularly compared the effects of different random models on the regressive parameters and the estimates of biomass,stressed the significance of the determined random model,and introduced the method of robust estimation to diminish the gross errors.