An integrated system for modeling tree and stand survival

作者：Cao Quang V^*

来源：Canadian Journal of Forest Research, 2017, 47(10): 1405-1409.

DOI：10.1139/cjfr-2017-0229

• individual-tree model
• least squares
• loblolly pine
• logistic regression
• maximum likelihood

Traditionally, separate models have been used to predict the number of trees per unit area (stand-level survival) and the survival probability of an individual tree (tree-level survival) at a certain age. This study investigated the development of integrated systems in which survival models at different levels of resolution are related in a mathematical structure. Two approaches for modeling tree and stand survival were considered: deriving a stand-level survival model from a tree-level survival model (approach 1) and deriving a tree survival model from a stand survival model (approach 2). Both approaches rely on finding a tree diameter that yields a tree survival probability equal to the stand-level survival probability. The tree and stand survival models from either approach are conceptually compatible with each other but not numerically compatible. Parameters of these models can be estimated either sequentially or simultaneously. Results indicated that approach 2, with parameters estimated sequentially (first from the stand survival model and then from the derived tree survival model), performed best in predicting both tree-and stand-level survival. Although disaggregation did not help improve prediction of tree-level survival, this method can be used when numerical consistency between stand and tree survival is desired.

• 出版日期2017-10