Author(s)

Huiyuan Cui, Chengde Wang, Fadian Lu, Min Liu

Corresponding Author

Chengde Wang

ABSTRACT

In the process of forest management, the crown profile model can not only directly reflect crown size and crown volume but also can be used to derive forest stand visualization. We used a total of 410 trees collected from 98 pure even-aged temporary sample plots established in Fujian Province to develop crown profile models of Chinese fir (Cunninghamia lanceolata). To describe the crown more accurately, three different curve equations were selected to describe the crown profiles of different age groups (i.e., young, middle and mature forests). In the model fitting process, a continuous autoregressive error structure (CAR) was used to account for the correlation between repeated measurement data from the same tree crown and a weighted regression function was used to eliminate the heteroscedasticity. Based on the developed crown profile models, we proposed a new approach which used the theory of Directed Tree to describe the semantic relation of morphological structure and constructed a digital model to express each tree in forest stand. The results showed that the predicted accuracy of model fit by the CAR had better fitting effects and the residual distribution of which showed the characteristics of equal variance and unbiasedness. The 3D digital tree constructed in this paper was more consistent with the actual growth of plantation and was of great significance for qualitative and quantitative analysis of tree competition level.

KEYWORDS

Crown profile model, Repeated measurement data, Continuous autoregressive error structure, 3D digital tree