高级检索

    云南松林分平均高生长模型及模型参数环境解释

    Stand average height growth model and environmental interpretation in model parameter of Pinus yunnanensis

    • 摘要:
      目的云南松作为我国西南地区的主要建群树种,在西南地区占有重要地位,研建其林分平均高生长模型以及对模型参数进行环境解释,可为气候变暖背景下研究云南松林分的生长动态提供经验模型。
      方法基于云南省森林资源连续清查数据和气象数据,以云南松林为研究对象,结合7种理论生长模型,采用非线性回归方法构建林分平均高生长模型,并对最优模型的参数进行环境解释。
      结果对选定的7种理论生长模型进行拟合,以调整R2和均方根误差(RMSE)为模型拟合精度指标,从中选出林分平均高最优生长模型,将标准化的环境影响因子引入到最优生长模型参数中,对最优模型的参数进行环境解释。经方差分析可知,引入环境影响因子后的模型与基础模型之间有显著差异,研究取得了较好的结果,模型具有一定的适用性。
      结论(1) 从7个理论生长方程中选定林分平均高最优生长模型为逻辑斯蒂(Logistic)模型,调整R2达到0.616,均方根误差RMSE为2.328 m。(2)将环境影响因子引入到不同参数组合位置时表现最优的模型形式,作为该环境因子对林分平均高生长模型的参数环境解释,各环境因子引入后,模型拟合效果明显提高。调整R2后,提高最显著的是将湿润指数同时引入到参数abc位置上时的模型形式,其拟合效果提高了5.375%;提高最低的是将土壤厚度因子同时引入到参数abc位置上时的模型形式,其拟合效果提高了1.938%。(3)各环境影响因子对林分平均高生长模型的影响程度大小排序为:湿润指数>年均降水量>海拔>潜在蒸散量>年均温度>温暖指数>年均生物学温度>坡度>土壤厚度。(4)地形因子和气候因子与林分平均高生长之间的关系有正有负,地形因子中的海拔因子对林分平均高的影响不大,气象因子中温度对林分平均高生长的影响是通过对降水的制约来实现的。

       

      Abstract:
      ObjectivePinus yunnanensis is the dominant species and plays an important role in southwestern China. The empirical model for growth dynamics of Pinus yunnanensis under global climate warming conditions may be provided via investigating the stand growth model and analyzing the relationship between optimal model parameters and environment impacting factors.
      MethodEvaluating the growth model of stand average height by nonlinear regression model and analyzing the relationship between the optimal model parameters and the environment impacting factors, the continuous forest inventories (CFI) data of Yunnan Province and meteorological data, as well as seven kinds of fundamental theory equations were used.
      ResultIdentifying the average stand height optimal growth model based on the adjustment R2 by fitting seven basic theoretical equations and the root mean square error (RMSE), the standardized environmental impacting factors were introduced into the parameters of the optimal growth model to explain the influence of environmental factors on the average stand height growth model. The variance analysis showed that there was a significant difference between the basic model and the modified model with introduced environmental impacting factors. The results of the study have good applicability and effects.
      Conclusion(1) Logistic model was selected from the optimal model by adjusting R2 and the minimum root means square error. The adjusted R2 was 0.616 and the minimum root mean square error was 2.328 m. (2) The best model was the one with introduced environmental factors as the different parameters, and the modified model was efficient to explain the influence of environmental factors on the average stand height growth model. The most significant increase of the model fitting precision adjustment R2 was obtained with humid index combined into the model parameters a, b, c and the increase rate was 5.375%. The lowest increase was the model when the soil thickness factor was introduced into the parameters a, b, c combination and the value was 1.938%. (3) The influence of environmental impacting factors on the height growth model was HI>MAP>ALT>ET0>MAT > WI > BT > SLO > ST. (4) The topographic/climatic factors and stand average height growth was positively or negatively correlated. The altitude in topographic factors has no significant effects on stand height growth. The effect of temperature on the stand height growth was achieved by the control of precipitation.

       

    /

    返回文章
    返回