A Multiple Logistic Regression Model for Predicting the Development of Phytophthora ramorum symptoms in Tanoak (Lithocarpus densiflorus)

Abstract

Phytophthora ramorum attacks tanoak (Lithocarpus densiflorus) in California and Oregon. We present a stand-level study examining the presence of disease symptoms in individual stems. Working with data from four plots in redwood (Sequoia sempervirens)/tanoak forests in Marin County, and three plots in Mendocino County, California, we evaluated a number of stem variables such as dbh, height, crown class, foliar condition and the presence of insect, fungal infection or disease symptoms for significance in multiple logistic regression analysis. We built a maximal model of variables and simplified this down following Akaike’s Information Criterion. Our final model includes only variables significant at p = 0.01. We compare disease presence and mortality results in Marin stands with results collected in Mendocino County.

Parent Publication

• Proceedings of the Redwood Region Forest Science Symposium: What does the future hold?

Citation

Spencer, Mark; O''Hara, Kevin. 2007. A Multiple Logistic Regression Model for Predicting the Development of Phytophthora ramorum symptoms in Tanoak (Lithocarpus densiflorus). In: Standiford, Richard B.; Giusti, Gregory A.; Valachovic, Yana; Zielinski, William J.; Furniss, Michael J., technical editors. 2007. Proceedings of the redwood region forest science symposium: What does the future hold? Gen. Tech. Rep. PSW-GTR-194. Albany, CA: Pacific Southwest Research Station, Forest Service, U.S. Department of Agriculture; p. 469-474