Cooperative Ecosystem Studies Unit, Rocky Mountain CESU

archived

The purpose of the Northern Rocky Mountain Science Center (NOROCK) funding opportunity is develop quantitative models that best explain the distribution and abundance of blister rust infection and mountain pine beetle mortality in whitebark pine within the Greater Yellowstone Ecosystem. In addition, to develop quantitative models that best explain recruitment rates of juvenile whitebark pine within existing stands of mature whitebark pine and the implications for forest regeneration. The intent is to explore the rate of blister rust infection and mountain pine beetle mortality in the Greater Yellowstone Ecosystem using spatial regression models and a suite of spatially explicit covariates. To use model comparison tools based on information criteria (e.g., Akaike Information Criterion – AIC) to select the most parsimonious models. If necessary, model averaging methodology, also based on AIC, may need to be used. Similar models and covariates will also be used to best explain stem densities of juvenile trees within existing stands. The Center will be consulted to develop a set of a priori models to be tested regarding blister rust infection rates, mountain pine beetle mortality, and whitebark pine recruitment. As a necessity, analyses will be conducted at both the stand and tree level. Stand level covariates will include: Slope, elevation, aspect, latitude, longitude, a measure of solar radiation (Keating et al, 2006), habitat type (Steele et al 1983), stem density, basal area, and the presence (absence) of blister rust or mountain pine beetle in the stand. Individual tree covariates will include: Diameter at breast height, level of rust infection recorded, number of stems in a tree cluster (multiple stem growth is a common form of whitebark pine), height class, live canopy volume, and general health (categorical). Use the information-theoretic approach (Burnham and Anderson 1998, 2002) instead of classical null hypothesis testing. This new paradigm of data analysis is based on the Kullback-Leibler information that avoids many fundamental limitations of null hypothesis testing.

High