利用度量误差模型和分段建模方法建立云南云杉相容性立木材积和地上生物量模型

以云南云杉实测立木材积和地上生物量数据为例，利用度量误差模型方法和分段建模方法建立相容性的一元和二元立木材积和地上生物量模型。结果表明，无论常规模型还是分段模型二元立木材积模型的相关统计指标得到了大幅度的改进，而二元地上生物量模型的相关指标与一元模型相比差异不大；常规二元立木材积模型在小径阶下存在明显的偏差，分段模型从整体上能够有效地解决系统偏差问题；所建的分段一元立木材积模型和地上生物量模型的平均预估精度分别到了90％和95％，同时分段二元立木材积模型和地上生物量模型的平均预估精度均到了97％。

Using Measurement Error Modeling Method and Segmented Modeling Method to Establish Compatible Single-Tree Biomass Model System for Spruce in Yunnan

Based on the tree volume and above-ground biomass data of spruce in Yunnan , compatible single- tree biomass model systems of one-variable and two-variables were constructed respectively by using the error- in-variable simultaneous equations and segmented modeling method in this paper. The results showed that ： 1 ） The regressions of volume equations improved significantly when tree height was used together with diameter at breast height （DBH） for both conventional model and segmented model, while the regressions of biomass equa- tions improved slightly; 2） There were obvious system bias at small diameter classes in conventional two-varia- ble model, the segmented model could effectively resolve this problem; 3 ） The prediction precision of volume and above-ground biomass segmented one-variable model were more than 90% and 95% respectively, the pre- diction precisions of volume and above-ground biomass segmented two-variable model were more than 97%.