湖南楠木次生林生长与收获预估模型

Growth and yield models of Phoebe zhennan secondary stand in Hunan province

【目的】建立湖南省楠木次生林林分断面积与蓄积相容性生长收获预估模型,为林分的生长预测和经营决策提供理论依据。【方法】以湖南省1989—2014年6期一类清查样地中的楠木次生林为研究对象,建立其林分断面积和蓄积生长的联立方程组,在此基础上,加入样地效应,构建基于混合效应的联立方程系统。【结果】基础模型包括3个含林分变量的内生变量,模型M1、G2、M2拟合的确定系数(R2)分别为0.860、0.907、0.778;相比基础模型,混合效应模型不仅在R2上有所提高(0.960,0.913,0.915),其平均误差(Bias)和均方根误差(RMSE)也有明显的降低;对比残差分布图,发现混合效应模型拥有分布范围更小、分布更均匀的残差值;通过对比混合效应模型不同模拟的AIC、BIC值,同时结合似然比(LRT)检验,确定了M1、G2和M2的随机参数分别为(β1,β3)、(β6)和(β1,β3,β6)。【结论】基于混合效应的湖南楠木次生林相容性联立方程系统能够提高模型的精度与适用性,更准确地预估该林分的生长与收获,为其经营管理提供科学依据。

【Objective】Establishing a compatible growth and yield prediction model for the basal area and volume of Phoebe zhennan secondary forest in Hunan province, to provide theoretical basis for its growth prediction and management decision-making.【Method】A simultaneous equation group of stand basal area and volume growth was established for Phoebe zhennan secondary forest by the data of national forest inventory in Hunan province from 1989 to 2014. The simultaneous equation system based on mixed effect was constructed by adding stand effect.【Result】The basic model includes three endogenous variables including stand variables. The determination coefficients(R2) of model M1, G2 and M2 are 0.860, 0.907 and 0.778. Compared with the basic model, the mixed effect model not only improves on R2(0.960, 0.913, 0.915), but also has smaller average error(Bias) and root mean square error(RMSE). Contrasting the distribution of residuals, it is found that the mixed effect model has smaller range and more uniform distribution of residuals. By comparing the AIC and BIC values simulated by different mixed effect models, and combining with likelihood ratiotest(LRT), the random parameters of M1, G2 and M2 are determined to be(β1,β3),(β6) and(β1,β3,β6).【Conclusion】The compatibility simultaneous equation system of Phoebe zhennan secondary forest in Hunan based on mixed effect can improve the accuracy and applicability of the model, more accurately to predict the growth and yield of the forest, and provide scientific basis for its management.