We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
A Spatio-Temporal Model for Mountain Pine Beetle Damage.
- Authors
Kaufeld, Kimberly; Heaton, Matthew; Sain, Stephan
- Abstract
Forest composition in the western region of the United States has seen a dramatic change over the past few years due to an increase in mountain pine beetle damage. In order to mitigate the pine beetle epidemic, statistical modeling is needed to predict both the occurrence and the extent of pine beetle damage. Using data on the front range mountains in Colorado between the years 2001-2010 from the National Forest Service, we develop a zero-augmented spatio-temporal beta regression model to predict both the occurrence of pine beetle damage (a binary outcome) and, given damage occurred, the percent of the region infected. Temporal evolution of the pine beetle damage is captured using a dynamic linear model where both the probability and extent of damage depend on the amount of damage incurred in neighboring regions in the previous time period. A sparse conditional autoregressive model is used to capture any spatial information not modeled by spatially varying covariates. We find that the occurrence and extent of pine beetle damage are positively associated with slope and damage in previous time periods.
- Subjects
SPATIO-temporal variation; CLIMATE change; BLACK pine bark beetle; FORESTS &; forestry; UNITED States. National Forest Service; REGRESSION analysis
- Publication
Journal of Agricultural, Biological & Environmental Statistics (JABES), 2014, Vol 19, Issue 4, p437
- ISSN
1085-7117
- Publication type
Article
- DOI
10.1007/s13253-014-0182-1