Abstract:

Objective: The entire northeast region, including the Daxing’an Mountains, Xiaoxing’an Mountains, and Changbai Mountains, was taken as the research area, and the region is divided into five parts: the Daxing’an Mountains, Xiaoxing’an Mountains, Zhangguangcai Mountains, Changbai Mountains, and the Liaodong Mountainous area. The “region” was used as a dummy variable to evaluate the site quality of natural secondary forests in northeast China. The differences in the distribution patterns of site quality across different areas were compared and analyzed, in order to provide a scientific basis for forest management and planning. Method: In this study, 456 research plots with an area of 0.1 hm² in northeast China were selected as the data source. Dominant trees were screened using Assmann’s criteria, and the height of dominant trees weighted by the cross-sectional area of all tree species in the plot was used as the basic modeling data. The optimal guide curve was selected from seven commonly used theoretical growth equations. The generalized algebraic difference approach (GADA) was used to derive the site form model, and the “region” was used as a dummy variable to construct a final dummy variable model. The coefficient of determination (R²), root mean square error (RMSE), and mean absolute error (MAE) were used to assess the goodness of fit of the model. Finally, the site form values of each region were obtained according to the model and analyzed and compared to explore the distribution pattern of forest site productivity in northeast China. Result: The research results showed that R² of the site form model derived by the GADA was 0.307. However, after using “region” as a dummy variable, the R² values of each region increased significantly, and both the RMSE and MAE of the model decreased. The spatial distribution of forest site quality in northeast China showed significant heterogeneity, which is characterized as follows: highest site quality level (A) and high-intermediate site quality level (B) were mainly concentrated in the main areas of the Daxing’an Mountains and the Changbai Mountains. The proportion of the highest quality level (A) sites in the Daxing’an Mountains exceeded 90%, and the proportion of site quality levels increased from D to A. The proportion of C and D level sites in the Xiaoxing’an Mountains was over 80%, and the site quality in the south was significantly better than that in the north. A and B level sites in Zhangguangcai Mountains accounted for only 35.6%, and the site quality in the areas closer to the Changbai Mountains was better. In the Changbai Mountains, the proportion of A and B level sites exceeded 60%, and the overall site quality showed an upward trend. In the Liaodong Mountainous area, C and D level sites were dominant, and the spatial distribution showed the characteristic of improved quality in the areas adjacent to the Changbai Mountains. It can be seen that there are also significant differences in the distribution of site quality in different regions. Conclusion: This study has established a dummy variable site form model based on the GADA. The introduction of dummy variables enhances the model’s performance. Based on this, an assessment of the site quality of natural secondary forests in northeast China is conducted. The results of the site quality assessment indicate significant differences in site quality across different regions. This study provides theoretical support for achieving sustainable development of forest resources and ecological protection. In addition, this study also emphasizes that future research should comprehensively consider environmental factors in order to gain a deeper understanding of the dynamic changes in forest ecosystems and provide guidance for forest management and conservation strategies.

Key words: site form, site quality assessment, dummy variables, uneven-aged mixed forest, generalized algebraic difference approach (GADA)