Authors: Melissa Gearman*, Saint Cloud State University, MN, Dr. Mikhail S. Blinnikov, Saint Coud State University, MN
Topics: Natural Resources, Geographic Information Science and Systems, Spatial Analysis & Modeling
Keywords: Maxent, oak wilt, Minnesota, GIS, species distribution model
Session Type: Poster
Start / End Time: 1:20 PM / 3:00 PM
Room: Napoleon Foyer/Common St. Corridor, Sheraton, 3rd Floor
Presentation File: No File Uploaded
Forest diseases and pathogens can cause significant damage to an ecosystem. Understanding where they are going to occur and what variables are important in their distribution can stave off the detrimental effects they have on established and at risk ecosystems. With the advancement of spatial analysis and remote sensing technology, these diseases can now be managed through modeling. Modeling allows researchers to determine the extent of the disease, which variables lead to the increase in infection centers, and predict the distribution of the disease. This study used Maxent, a presence-only species distribution model (SDM), to map the potential probability distribution of the invasive forest pathogen oak wilt (Ceratocystis fagacearum) in eastern and southeastern Minnesota. The model related oak wilt occurrence data with environmental variables including climate, topography, land cover, soil, and population density. Results showed areas with the highest probability of oak wilt occur within and surrounding the Minneapolis/St. Paul metropolitan area. The jackknife test of variable importance indicated land cover and soil type as the most important variables contributing to the prediction of the distribution. Multiple methods of analysis showed the model performed better than random at predicting the occurrence of oak wilt. This study shows Maxent has the potential to be an accurate tool in the early detection and management of forest diseases.