Graduation Date

Spring 2017

Document Type

Dissertation/Thesis

Program

Master of Science degree with a major in Biology

Committee Chair Name

Erik Jules

Committee Chair Email

ejs4@humboldt.edu

Committee Chair Affiliation

HSU Faculty or Staff

Second Committee Member Name

Christopher Dugaw

Second Committee Member Email

dugaw@humboldt.edu

Second Committee Member Affiliation

HSU Faculty or Staff

Third Committee Member Name

Leonel Arguello

Third Committee Member Email

leonel_arguello@nps.gov

Third Committee Member Affiliation

Community Member or Outside Professional

Fourth Committee Member Name

Richard Cobb

Fourth Committee Member Email

rccobb@ucdavis.edu

Fourth Committee Member Affiliation

Community Member or Outside Professional

Fifth Committee Member Name

Jeffrey Kane

Fifth Committee Member Email

Jeffrey.Kane@humboldt.edu

Fifth Committee Member Affiliation

HSU Faculty or Staff

Subject Categories

Botany

Abstract

The pathogen Phytophthora ramorum, the causal agent of Sudden Oak Death (SOD), is responsible for the deaths of millions of oak (Quercus spp.) and tanoak (Notholithocarpus densiflorus) trees in California and Oregon (USA). A recent infection in Redwood National Park (RNP) in California (USA) provided an opportunity to adapt an existing SOD model to assess the efficacy of current and proposed management strategies. A common method of SOD treatment includes killing both infected and uninfected hosts in the area of infection, as well as the area surrounding the infection to create buffers to account for undetected or cryptic infections. I used the existing SOD model for a larger spatial area (380 ha) and included host density data. Using this model, I show that buffers of plausible width are not effective methods for managing SOD infections in RNP because they do not control spread of the pathogen. Additionally, I ran each model with two dispersal kernels (exponential and power-law) with equal mean spread distances and showed that the shape of the distribution kernel used can significantly alter the outcome of the model. For example, models using 300 m and 400 m buffers with an exponential dispersal kernel predicted containment of P. ramorum, but spread beyond these buffers was predicted with a power-law distribution kernel. Lastly, my work provides the first evidence of significant stream-to-land spread of P. ramorum. I show laboratory-confirmed P. ramorum infections along a creek in RNP, which included low-hanging branches with cankers on host trees concealed by debris. I also used the adapted SOD model to compare two scenarios, one including and one excluding stream-to-land transmission, and found that the model that included stream transmission predicted future spread significantly better than the model that did not include stream transmission. This work not only highlights the problems associated with treating SOD infections by removing hosts in buffer zones surrounding infections, but also demonstrates how precise knowledge about the dispersal distance and dispersal frequency is required to derive accurate model predictions. Additionally, my work points to a novel transmission pathway for an important forest pathogen and highlights the need to determine the prevalence of this dispersal mechanism across the range of the pathogen.

Citation Style

APA

Share

Thesis/Project Location

 
COinS