Building improved models of sugar maple mortality
- Download PDF (166499)
- This publication is available only online.
In: Morin, Randall S.; Liknes, Greg C., comps. Moving from status to trends: Forest Inventory and Analysis (FIA) symposium 2012; 2012 December 4-6; Baltimore, MD. Gen. Tech. Rep. NRS-P-105. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. [CD-ROM]: 204-209.
The decline of sugar maple (Acer saccharum Marsh.) in the northern United States is causing concern, and several studies have identified soil properties that are linked to the observation of dead/dying trees. Unfortunately, the sample of trees supporting these studies is purposive in nature; soil properties are assessed only on those plots where dead trees are observed. In this study, we used the U.S. Forest Service's Forest Inventory and Analysis database (FIADB) to conduct an exploratory analysis of a broader population of sugar maple (live and dead) across a wide range of soil types. This population of plots has a highly skewed, zero-inflated distribution: the number of plots in the sample without dead trees is an order of magnitude greater than the number of plots with dead trees. One effective method of analysis is a hurdle--or conditional--model approach. In the first phase, the response variable is the presence or absence of dead sugar maple and the inferential space is the entire population of plots with sugar maple trees. The second phase uses the relative abundance of dead sugar maple as the response variable; in this case, inference is restricted to those plots where dead sugar maple trees are observed. In both sets of models, basal area and geology are significant predictors of dead sugar maple, but the most significant soil variables vary between these two inferential spaces. Our study highlights important analytical considerations when using FIADB for analysis of forest health conditions and presents simple methods to create a more comprehensive space for statistical inference.
Keywordsstatistics; estimation; sampling; modeling; remote sensing; forest health; data integrity; environmental monitoring; cover estimation; international forest monitoring
Note: This article is part of a larger document. View the larger document
Perry, Charles H.; Zimmerman, Patrick L. 2012. Building improved models of sugar maple mortality. In: Morin, Randall S.; Liknes, Greg C., comps. Moving from status to trends: Forest Inventory and Analysis (FIA) symposium 2012; 2012 December 4-6; Baltimore, MD. Gen. Tech. Rep. NRS-P-105. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. [CD-ROM]: 204-209.