断面积
数学
胸径
站点索引
混合模型
统计
树(集合论)
森林资源清查
变量
林业
竞赛(生物学)
森林经营
生态学
地理
生物
数学分析
作者
Patricia Adame,Jari Hynynen,Isabel Cañellas,Miren del Rı́o
标识
DOI:10.1016/j.foreco.2007.10.019
摘要
In this study, a distance-independent mixed model was developed for predicting the diameter growth of individual trees in Mediterranean oak (Quercus pyrenaica Willd.) coppices located in northwest Spain. The data used to build the model came from 41 permanent plots belonging to the Spanish National Forest Inventory with the dependent variable being 10-year diameter increment over bark for trees larger than 7.5 cm at breast height. The basic field data required for predictions had been divided into four main groups: size of the tree, stand variables, competition indices and biogeoclimatic variables. The most significant independent variables were the individual-tree diameter, the basal area of trees larger than the subject tree, dominant height, site index and biogeoclimatic stratum. The model was defined as a mixed linear model with random plot effect, achieving an efficiency of 44.38%. The accuracy of the model was tested against the modelling data and against independent data from the same stands. Mixed model calibration of diameter increment was carried out with the independent data using a different sample of complementary observations of the dependent variable. The calibrated model was an improvement on the trivial model, which assumes constancy in diameter increment for a short projection period, especially the pattern of residuals with respect to predicted diameter and the independent variables.
科研通智能强力驱动
Strongly Powered by AbleSci AI