Climate change poses significant challenges to biodiversity in the Pacific Northwest, particularly for species of conservation concern. Species distribution models (SDMs) offer a powerful tool to predict habitat suitability under changing environmental conditions, helping managers make informed decisions. This research addresses critical knowledge gaps identified by the Oregon Department of Fish & Wildlife (ODFW) in their State Wildlife Action Plan (SWAP), especially regarding how species may respond to increasing temperatures and shifting precipitation patterns. 

In this project, Rebecca Hutchinson will collaborate with ODFW to enhance the utility of SDMs for climate adaptation planning. This project applies recent machine learning advances to two key areas — leveraging community science data and evaluating models under changing conditions. In their work applying machine learning to community science data, the project team will use eBird data and spatial clustering algorithms to improve occupancy models for bird species, addressing the challenge of imperfect detection in opportunistic datasets. These methods will be adapted for use by ODFW managers and evaluated with structured survey data. In their second focus area around evaluating models under changing conditions, the team will develop evaluation strategies that reflect real-world use cases — such as predicting species distributions in novel climates or regions — ensuring SDMs are robust and applicable to management decisions. 

By co-producing actionable science and creating accessible workflows, this project supports climate-informed conservation strategies and strengthens the partnership between academic research and resource management.