基于林分及地形因子的落叶松人工林林分生物量模型构建

Construction of the stand⁃level biomass model of Larix olgensis plantations based on stand and topographic factors

【目的】落叶松在我国东北地区广泛分布,是重要的造林和用材树种,具有生长速度快、耐寒等优点。为了准确地估算落叶松人工林林分生物量,构建了落叶松林分可加性生物量模型。【方法】以落叶松人工林为研究对象,基于黑龙江省的304块人工落叶松固定样地数据,采用非线性似乎不相关回归的方法建立了可加性生物量模型系统,使用留一交叉验证法对建立的模型进行检验。【结果】林分断面积和林分平均高对树干、树枝、树叶和树根生物量模型有显著影响,林龄和海拔也显著影响林分树干、树叶、树根生物量;坡率和坡向对树枝生物量有显著影响。树叶生物量与林分平均高、林龄和海拔呈显著负相关,树干与树根生物量则与之呈显著正相关,树枝生物量与林分平均高呈显著正相关。在所建立的可加性生物量模型中,调整后决定系数(Radj2)均在0.94以上,均方根误差(RMSE)较小。检验指标平均误差(MPE)和平均误差百分比(MPE%)均接近0,拟合指数(IF)均大于0.93,平均绝对误差(MAE)较小,且平均绝对误差百分比(MAE%)均小于11%。【结论】建立的落叶松人工林可加性生物量模型的拟合与预测效果均较好,可以进行黑龙江省林分尺度的落叶松人工林生物量预测。

【Objective】Larix olgensis is widely distributed in northeastern China and it is considered to be an important afforestation and timber species with advantages like the rapid growth and cold tolerance.To accurately estimate the biomass of larch plantations,an additive system of biomass equations for larch plantations was established.【Method】This study investigated larch plantations using the forestry inventory data of 304 larch plantations in Heilongjiang Province.An additive system of biomass equations was established using the method of non⁃linear and seemingly uncorrelated regression and employed a leave⁃one⁃out cross method for the model validation.【Result】The stand basal area and stand mean tree height had a significant effect on the stem,branch,foliage and root biomass,the stand age and elevation also significantly affect the stem,foliage and root biomass.The branch rate and aspect had a significant effect on branch biomass.Foliage biomass was negatively correlated with stand mean tree height,stand age and elevation,whereas stem and root biomasses were positively correlated with the explanatory variables,and branch biomass was positively correlated with stand mean tree height.In terms of the additive system of the biomass equation,the adjusted coefficients of determination(Radj 2)were greater than 0􀆰94,and the root mean square error(RMSE)was small.The average bias(MPE)and average bias percentage(MPE%)were both close to 0,the fit indices were all greater than 0􀆰93,the average absolute bias(MAE)was small,and the average absolute bias percentage(MAE%)was less than 11%.【Conclusion】The additive biomass model of larch plantations established in this study is effective for fitting and prediction,and can be used to predict the stand⁃level biomass of larch plantations in Heilongjiang Province.