Wildfires are one of the greatest threats to human infrastructure and the ecosystem services humans value in the western US, but are also necessary in fire-adapted ecosystems. Wildfire activity is widely projected to increase in response to climate change in the Northwest, but we currently lack a comprehensive understanding of what this increase will look like or what its impacts will be on a variety of ecological and hydrologic systems. This project addressed one critical part of those impacts: the islands of unburned vegetation within wildfires. Unburned islands occur naturally as wildfires burn across landscapes, and are important habitat refuges for species -- places where plants and animals survive the fire and subsequently regenerate across the recently burned landscape. Since they are naturally resistant to burning, understanding how climate change impacts these islands helps us understand what species may be in even more danger from wildfires if these refuges disappear. Studying them and copying their characteristics also helps humans build our homes and communities to be more resistant to wildfire. This project aimed to understand and model unburned islands within wildfires to inform both conservation and restoration planning and community wildfire protection planning efforts.