Abstract:An individual-tree basal area increment model was developed for masson pine based on 26276 observations of 13,138 trees in 987 sample plots from the 7th (2004), 8th (2009), and 9th (2014) Chinese National Forest Inventory in Hunan Province, South-central China. The model was built using a linear mixed-effects approach with sample plots included as random effects since the data have a hierarchical stochastic structure and biased estimates of the standard error of parameter estimates could be a consequence of applying ordinary least square (OLS) for regression. In addition, within-plot heteroscedasticity and autocorrelation were also considered. The final mixed-effects model was determined according to the Akaike information criterion (AIC), Bayesian information criterion (BIC), log-likelihood (Loglik), and the likelihoodratio test (LRT). The results revealed that initial diameter (DBH), the sum of the basal area (m2/ha) in trees with DBHs larger than the DBH of the subject tree (BAL), number of trees per hectare (NT), and elevation (EL) had a significant impact on individual-tree basal area increment. The mixed-effects model performed much better than the basic model produced using OLS. Additionally, the variance structure of the model errors was successfully modeled using the power function. However, the autocorrelation structures were not defined because there was no autocorrelation amongst the data. It is believed that the final model will contribute to the scientific management of the masson pine.
What problem does this paper attempt to address?
This paper aims to solve the problem of developing a prediction model for the basal area increment (BAI) of individual Masson pine trees, especially in the uneven - mixed forests in Hunan province in central - southern China. Specifically, the research objectives are:
1. **Develop a linear mixed - effects model**: In order to predict the basal area increment of individual Masson pine trees more accurately, the researchers adopted a linear mixed - effects model. This method can handle the hierarchical structure and longitudinal correlation of data, thereby improving the predictive ability of the model.
2. **Improve model performance**: By introducing random effects, the research hypothesized that this would significantly improve the performance of the model, especially compared with the traditional ordinary least squares (OLS). The researchers hope to provide a more solid foundation for the formulation of forest management strategies through this method.
### Research background
- **The importance of Masson pine**: Masson pine is an evergreen pioneer tree species widely distributed in southern China. It has the ability to grow rapidly and adapt to harsh environments, so it is widely used in ecological restoration projects. At the same time, Masson pine wood has high economic value and is used for construction, furniture and packaging materials.
- **Limitations of current management methods**: At present, the management of Masson pine plantations mainly adopts a continuous rotation system, which has a negative impact on the environment. Therefore, more and more research focuses on the management methods of uneven - mixed forests.
### Model development
- **Data sources**: The study used data from the 7th (2004), 8th (2009) and 9th (2014) national forest resources inventories in China, covering 26,276 observations of 13,138 trees in 987 plots in Hunan province.
- **Variable selection**: The researchers selected four main variables: initial diameter at breast height (DBH), competition index (BAL), number of trees per hectare (NT) and elevation (EL). These variables were determined by the stepwise regression method and a multicollinearity test was carried out.
- **Model form**: The final form of the linear mixed - effects model is as follows:
\[
\log(\text{DDS}+ 1)=(\beta_1 + b_1)+(\beta_2 + b_2)\text{DBH}+(\beta_3 + b_3)\text{BAL}+(\beta_4 + b_4)\text{EL}+(\beta_5 + b_5)\text{NT}+e_{ij}
\]
where \(\beta\) represents the fixed effect, \(b\) represents the random effect, and \(e_{ij}\) represents the error term.
### Results
- **Basic model**: The preliminary basic model shows that DBH, BAL, NT and EL have a significant impact on the basal area increment of individual trees, but there is obvious heteroscedasticity.
- **Mixed - effects model**: By introducing random effects and heteroscedasticity functions, the performance of the mixed - effects model is significantly improved. The AIC, BIC and Loglik values of the final model are all better than those of the basic model, and the residual plot and QQ plot show that the heteroscedasticity and normality of the model have been significantly improved.
### Discussion
- **Variable interpretation**: DBH is positively correlated with BAI, indicating that larger trees have more advantages in resource competition. BAL has a negative impact on BAI, indicating that larger trees have a larger basal area increment due to their stronger resource competition ability.
- **Model application**: The research results show that the mixed - effects model performs excellently in predicting the basal area increment of individual Masson pine trees, providing strong support for the scientific management of uneven - mixed forests.
In conclusion, this study improves the prediction accuracy of the basal area increment of individual Masson pine trees by developing a linear mixed - effects model, providing a scientific basis for the formulation of forest management strategies.