基于度量误差方法的杉木储备林间伐与未间伐林分生长模型研究

利用湖南永州金洞林场储备林数据,以Richards式和Schumacher式为基础模型,引入经营措施的哑变量,建立基于传统方法和哑变量方法的林分断面积和蓄积生长模型方程组,采用非线性度量误差方法拟合模型参数。经检验,哑变量方法建立的林分断面积和蓄积生长模型的决定系数和预估精度比传统方法建立的生长模型有所提高,而各项误差相应减少,且模型回归效果显著,残差分布更集中于横轴两侧。基于哑变量方法拟合出的林分断面积和蓄积生长模型不仅考虑了抚育间伐对林分生长过程的影响,而且采用非线性度量误差方法联立方程组求解模型参数,也使得林分断面积和蓄积生长模型更具有相容性和一致性。

Growth Model for Thinned and Un-thinned Stand of Chinese Fir Reserve Forest Based on Error Measurement Method

By using the data of reserve forest of Jindong Forest Farm,taking the Richards and the Schumacher models as the prototypes,we separately fit the basal area model and the volume growth model,and introduced the dummy variable of the management measures.The nonlinear measurement error method was used to respectively combine the equations constructed based on the traditional method and the equations constructed based on the dummy variable method.The results showed that compared with the traditional method,the determination coefficient and prediction accuracy of the basal area and the volume growth model based on the dummy variable method were improved,the errors were correspondingly reduced,the model regression effect was significant and the residual distribution was more concentrated on both sides of the horizontal axis.The basal area and the volume growth model constructed based on the dummy variable method was better than the traditional one.Not only considered the influence of tending and thinning on the growth process of the stand,but also used the nonlinear measurement error equations to solve the model.The parameters also made the basal area model and the volume growth model more compatible and consistent.