Modeling the risk for Phytophthora ramorum with an anthropogenic focus in northern California

Phytophthora ramorum is an oomycete plant pathogen found in both North America and Europe that is commonly referred to as Sudden Oak Death (SOD). Since its initial detection in California in 1995, SOD has killed over a million trees, primarily tanoaks (Lithocarpus densiflorus), coast live oak (Quercus agrifolia), California black oak (Quercus kellogii), and Shreve’s oak (Quercus parvula var. shrevei). Northeastern California has thus far remained uninfected with the exception of one Rhododendron tree in Placer County. In order to contribute to the prevention of SOD spread to northeastern California, I created several models that depict where the risk for SOD development is highest based on four variables: host vegetation, climate suitability, proximity to roadways, and proximity to nurseries and lumber mills. I used a weighted overlay analysis to combine these variables at differing weights under five conditions. I applied the same analyses in Mendocino County, where sampling for SOD has resulted in 73 positive cases and 326 negative cases, to assess the accuracy of each condition. I calculated Cohen’s Kappa coefficient to measure the agreement between each condition and the recorded points. Though the Kappa values do not show strong agreement in the conditions, they do illustrate a trend of increase in agreement with the addition of proximity to roadways and nurseries. Suitable climate and host vegetation for SOD also exist along the coastal Pacific Northwest, and most of that area lies too far away from SOD incidence for natural spread to be the primary concern. Similar risk models incorporating human introduction variables could be applied to all of this area.