张广才岭三大硬阔树高曲线的研究

Tree Height-diameter Model for Three Hardwood Tree Species in the Zhangguangcai Mountains

树高曲线不仅在计算材积、生物量、碳储量等方面具有重要作用,还是森林调查监测的重要依据。本文以张广才岭水曲柳(Fraxinusmandschurica Rupr.)、胡桃楸(Juglans mandshurica Maxim.)和黄菠萝(Phellodendronamurense Rupr.)为研究对象,基于2 696株实测胸径和树高数据,分别选用9个树高曲线模型,对模型参数进行求解,并采用调整后确定系数(R2a)、均方根误差(RMSE)对模型拟合效果进行评价,采用平均预测误差(MPE)、平均预测误差百分比(MPE%)、平均绝对误差(MAE)和平均绝对误差百分比(MAE%)4个"刀切法"指标对模型进行独立性检验。结果表明:模型8(Gompert方程)能较好地描述水曲柳和黄菠萝树高曲线,模型9(Schumacher方程)能较好地描述胡桃楸树高曲线;三大硬阔最优树高曲线的R2a均大于0.60,平均绝对误差(MAE)和平均绝对误差百分比(MAE%)都较小。总的来说,所建立的张广才岭三大硬阔树高曲线拟合效果较好,具有一定的预估性。

The tree height curve can not only be used as a key indicator to calculate the volume, biomass, and carbon storage, etc.,but also be an important basis for forest survey and monitoring. In this paper, we used 9 tree height curve models based on the actual measurement data of DBH and height from 2 696 individual trees and took the three hardwoods （ Fraxinus mandschurica Rupr . Juglans mandshurica Maxim. Phellodendron amurense Rupr .） of Zhang guang cai M ountains as the research object to evaluate the fitness of the model by the adjusted coefficient determination（R2/a） and root mean square error（RMSE）. Four “Jackknife” indexes including mean prediction error （MPE） , mean prediction p ercent error （MPE % ） , mean absolute error （MAE） and mean absolute percent error （MAE % ） were used to take the independence test of model. The results showed that model 8（ Gompert equation） was better for height curve of Fraxinus mandschurica Rupr . and Phellodendron amurense Rupr , a nd the model 9 （ Schumacher equation ） was better for height curve of Juglans mandshurica Maxim ; the best R2a for all three hard wood was greater than 0. 6 while the mean absolute error （MAE） and mean absolute percentage error （MAE% ） were small. In summary , the established height curve model fitting results was better and would be a good precision of height for three hardwood of the Zhangguangcai Mountains.