Welcome to visit Scientia Silvae Sinicae,Today is

Scientia Silvae Sinicae ›› 2013, Vol. 49 ›› Issue (7): 75-85.doi: 10.11707/j.1001-7488.20130711

Previous Articles     Next Articles

Compatible Tree Biomass Models for Natural White Birch (Betula platyphylla) in Northeast China Forest Area

Dong Lihu, Li Fengri, Jia Weiwei   

  1. College of Forestry, Northeast Forestry University Harbin 150040
  • Received:2012-10-09 Revised:2013-04-26 Online:2013-07-25 Published:2013-07-23

Abstract:

Based on data of tree biomass for natural white birch (Betula platyphylla) in northeast China forest area, two methods (controlling jointly from level to level and controlling directly under total biomass) were used to establish the mono-element and dual-element compatible tree biomass model of natural white birch. And using nonlinear measurement error simultaneous equation estimated the parameters in the model, at the same time, the weighted regression was used to eliminate the heteroscedasticity. The results showed that R2 of mono-element and dual-element compatible model in this paper was 0.800-0.988, and the fit efficiency(EF)was 0.80-0.97. Besides, the precision of these models reached beyond 80%. On the other hand, the precision of foliage and branch model was relatively lower, and it was more than 69%. Among the compatible models established, the effect of total tree and stem was better than root, foliage and branch. On the whole, the compatible models established by two methods had the precision that we could receive and they could be used to predict the biomass of natural white birch with good precision. But the mono-element and dual-element compatible tree biomass model based on controlling directly under total biomass was better than the way that using controlling jointly from level to level under total biomass. To be concluded, it was advised to use the compatible model that based on controlling directly under total biomass.

Key words: natural white birch (Betula platyphylla), tree biomass, compatible model, error-in-variable model

CLC Number: