中国主要树种通用二元材积模型与推导形数模型研究
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

基金项目:

国家自然科学基金项目(41371001)、北京市自然科学基金项目(6161001)和北京林业大学青年教师科学研究中长期项目(2015ZCQ-LX-01)


Development of Generic Standard Volume Model and Derived Form Factor Model for Major Tree Species in China
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    以全国主要树种的二元材积模型、各省市一元材积表为基础材料,以取样径阶为1cm间隔所生成的2082组胸径(D)、树高(H)、材积(V)数据为基础数据,构建通用二元材积模型与推导形数模型。其中,通用二元材积模型利用SPSS软件进行回归建模,构建形式为山本式的1个全国通用二元材积模型、2个全国通用针/阔叶二元材积模型及6个全国分地区通用二元材积模型。结果表明,各模型的拟合决定系数R2均在0.984以上,选取6种回归模型评价指标进行模型验证,验证结果表明各模型的总相对误差和平均系统误差基本都不超过3%,在特定情况下,可以取代现有规模庞大的分地域树种一/二元立木材积模型库进行材积估算。推导形数模型采用二元材积式之一的斯泊尔式,该公式利用基础数据对形数f进行推导,得到全国六大区域总体/针叶/阔叶16个通用推导形数。结果表明,各模型的决定系数R2在0.983以上,验证结果表明各模型的总相对误差基本都控制在±3%内,总体精度较高。

    Abstract:

    In forest surveys, volume tables are usually used for standing wood volume estimation. From the national to the provincial scales and even forest farm all have their own single and binary tree volume tables which have characteristics of a large number of samples and high accuracy. However, due to the great number of models that the form is not unified, it is difficult to identify the large species and other reasons, resulting in poor efficiency in table look-up. It is inconvenient to estimate standing wood volume by using volume tables for some applications with no high precision requirement, such as estimation of wood value by lumberman, growing stock estimation by UAV and remote sensing retrieval of standing parameters in wide area. Therefore, a simple and quick standing tree volume estimation model is required. Based on standard volume model of national major tree species and single entry volume table of all provinces and municipalities, the regression model was made by SPSS to process all the data of 2082 groups of diameter at breast height (D), height of tree (H) and volume (V) generated by 1cm sampling diameter class, and then the generic standard volume models and derived form factor (f) models were built. Among them, generic standard volume models included a nationwide standard volume model, two nationwide needle-leaved and broad-leaved standard volume models and six standard volume models that made by using SPSS software with form of Yamamoto type. The result of model fitting showed that the fitting determination coefficients (R2) were all above 0.984 and the effects of model fitting were good. Evaluation indicators of six regression models were selected to conduct model verification, and the verification results showed that the indicators of total relative error (TRE) and average system error (MSE) of all models were almost within the range of ±3%. Using basic data to derive the value of f to get 16 overall, needle-leaved and broad-leaved derived form factor models. The validation result of generic standard volume models and derived form factor models showed that in specific situation, it can replace existing regional tree species’ single entry and standard tree volume model to estimate volume, which avoided the process of in-place wood recognition and table look-up, and it is of practical significance to guarantee precision and simplify evaluation process of standing tree volume.

    参考文献
    相似文献
    引证文献
引用本文

程文生,冯仲科,于景鑫.中国主要树种通用二元材积模型与推导形数模型研究[J].农业机械学报,2017,48(3):245-252.

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2016-07-08
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2017-03-10
  • 出版日期: