关键词: 冠幅, 树高, 胸径, 回归模型

Abstract: As a key factor in forestry research, the research on crown width is of great significance for the development of precision forestry. In order to better use crown width to predict other factors of trees, 14 common tree species such as Pinus tabulaeformis Carr., Tilia tuan Szyszyl., Quercus mongolica Fischer ex Ledebour in Songshan Nature Reserve of Beijing were selected to establish the DBHcrown model and tree hightcrown model. Based on the six regression equations and the measured data of plots, the filtered optimal equations for each species were tested for accuracy. The results showed that the model of 14 tree species displayed good fitting accuracy, most of decision coefficients(R2) were greater than 0.5. Among them, the highest R2 belonged to Juglans regia Linn. DBHcrown model which was 0.951, and the R2 of Juglans regia Linn. height model was 0.887. The lowest R2 belonged to Quercus mongolica Fischer ex Ledebour tree height model which was 0.506 and also more than 0.5. The accuracy evaluation index of these models in this study provided esstatistical significance and important reference value for the prediction of largearea forest parameters in Songshan area of Beijing.

Key words: crown width, tree height, DBH, regression models