Objective This research aims to develop an individual-tree diameter increment model of natural Larix gmelinii forests based on climatic factors in order to predict the DBH growth and provide a theoretical basis for the management of natural Larix gmelinii forests in Daxing’an Mountain Region, Inner Mongolia of northern China.
Method Based on the 187 permanent sample plots of natural Larix gmelinii forests (national forest continuous inventory in 2013 and 2018) in Daxing’an Mountains and climate data, we employed the step-by-step regression analysis to develop a traditional individual-tree diameter growth model considering climatic factors. On this basis, sample plot effects were further introduced to develop individual-tree mixed-effect model of Larix gmelinii. Finally, the basic and mixed-effect models were tested using independent test sample data.
Result Mean annual temperature (MAT) and mean precipitation of growing season (Pgm) were the main climatic factors which affected DBH growth of Larix gmelinii in the region. MAT and Pgm were positively correlated with the DBH growth. Other factors significantly affecting the DBH growth included the reciprocal of the initial DBH (1/DBH), basal area of trees larger than the subject tree (BAL) and the number of trees (NT), and all of them showed negative correlation with the DBH growth. The coefficient of determination (R2), mean absolute error (MAE) and root-mean-square error (RMSE) for the mixed-effect model were 0.760 4, 0.386 6 and 0.486 3 cm2, respectively. Compared with traditional model, the R2 of mixed-effect model increased by 0.321 7, and the MAE and RMSE reduced by 0.230 6 and 0.267 4 cm2, respectively. The mixed-effect model also showed better fitting accuracy in the model tests.
Conclusion Using individual-tree mixed-effect model based on climatic factors can well describe the DBH growth process of Larix gmelinii in Daxing’an Mountain Region, Inner Mongolia.