Abstract

This study was conducted to improve estimation of concomitant variables for implementation of a stand density management diagram (SDMD) for ponderosa pine (Pinus ponderosa Laws.) in northern California and Oregon. In traditional SDMD, isolines for variables such as stand volume are presented in such a way that uncertainty with estimation is not available. We developed the new top height and stand volume equations, as well as aboveground biomass and percent canopy cover, for building isolines in the SDMD using high-quality data collected from well-managed even-aged stands. The data were selected from the USDA Forest Service’s Pacific Southwest Research Station database. A total of 829 observations (from 113 plots across 15 sites in Oregon and California) were used for model construction. In addition, covariance-variance structures of all of the estimated parameters were provided so that users can evaluate the uncertainty associated with predictions. The model validation results indicated that the predictions made from fixed-effects model forms performed better than the current volume equation of SDMD, as well as those from mixed-effects model forms using the population average effect. The proposed equations provide enhanced predictions and additional useful information about managed ponderosa pine stands, including their uncertainty.

中文翻译：

---

黄松松林密度管理图等值线的改进方程

摘要

本研究的目的是改善伴随变量的估计实施黄松有一林分密度管理图（SDMD）的（黄松法律。）在加利福尼亚北部和俄勒冈州。在传统的SDMD中，变量等值线（如林分体积）的显示方式使得估计的不确定性不可用。我们开发了新的顶部高度和机架体积方程式，以及地上生物量和冠层覆盖百分比，以便使用从管理良好的均匀老化机架中收集的高质量数据在SDMD中构建等值线。数据选自美国农业部森林服务局的太平洋西南研究站数据库。总共829个观测值（来自俄勒冈州和加利福尼亚州15个站点的113个地块）用于模型构建。此外，还提供了所有估计参数的协方差-方差结构，以便用户可以评估与预测相关的不确定性。模型验证结果表明，用固定效应模型形式进行的预测要好于当前SDMD的体积方程，以及使用总体平均效应的混合效应模型形式的预测。提出的方程式提供了增强的预测和有关管理的美国黄松林分的更多有用信息，包括其不确定性。