The ADnet Bayesian belief network for alder decline: Integrating empirical data and expert knowledge

Area of Science:

• Ecology
• Forest Pathology
• Invasive Species Management

Background:

• Global trade facilitates invasive species spread, exemplified by Phytophthora ×alni impacting European riparian forests.
• Alder decline poses ecological risks to riverine ecosystems, necessitating effective disease management tools.

Purpose of the Study:

• To develop and present the Alder Decline Network (ADnet), a Bayesian belief network for predicting Phytophthora ×alni spread.
• To integrate diverse data sources, including expert knowledge and empirical data, for disease management.

Main Methods:

• Utilized Bayesian belief network methodology to model pathogen survival and infection.
• Gathered expert knowledge from 19 experts across 8 countries.
• Developed an original dataset of 1189 European locations, modeling pathogen occurrence using bioclimatic variables.

Main Results:

• ADnet identified mild temperatures and high precipitation as crucial for pathogen survival.
• Flood timing, water velocity, and soil type significantly influence disease incidence.
• The tool demonstrated sensitivity to state changes and scenario analyses.

Conclusions:

• ADnet provides a valuable tool for ecosystem management and knowledge transfer regarding alder decline.
• Management strategies include avoiding infected tree planting and modifying structures affecting flood periods.
• Future improvements involve expanding data coverage and researching host-pathogen genetic interactions.