SCC-PG: Co-designing AI-Driven Fire Preparedness Assessment Tools to Meet Community Needs in the Wildland-Urban Interface.

由于更热、更干燥的气候和更高的燃料负荷，美国西部野火的数量和面积都在增加。除了导致野火风险增加的环境因素外，美国荒地-城市界面（WUI）的快速增长也增加了生命和房屋损失的风险。促进WUI社会生态恢复力的一个关键方面是激励限制最终火灾破坏性的行动。通过减少植被（燃料负荷）来创造可防御的空间，以防止或减缓火灾向建筑物的蔓延，这是一种广泛支持的减少野火损失风险的策略；然而，需要新的分析方法来量化与减灾规划相关的空间尺度上的燃料负荷，从而量化野火风险。美国西部的WUI社区非常需要减轻野火危害，但减轻的资源有限。荒地火灾风险与社会脆弱性相交，社会脆弱性较高的居民不太可能负担得起昂贵的燃料减少治疗或参与野火缓解计划。至关重要的是，不仅要确保易发生火灾的社区拥有减轻火灾威胁所需的科学信息，而且要根据每个社区的独特社会需求量身定制这些信息的开发、交付和应用。受社会经济地位、地点依赖、火灾经验和其他因素的影响，WUI社区内部和社区之间对火灾风险的认知、对火灾准备的态度和对野火影响的脆弱性存在差异。该项目提出了在精细空间尺度（30米）上绘制森林结构地图的技术进步，采用人工智能（AI）方法可用于估算与减灾规划相关的空间尺度上的火灾风险。燃料信息将整合到工具和应用程序中，这些工具和应用程序的设计将满足社区的信息需求，与社区共同开发，在社区进行试点，并由社区进行评估。该项目建议在加利福尼亚州内华达县社区开展基于人工智能的火灾准备评估工具的范围和试点共同开发。最终，这些工具的共同开发将为社区规划和政策中火灾准备评估工具的部署、评估和应用提供信息，并扩大到美国西部不同地区的更广泛社区。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Wildland fires have increased in number and area across the western US as a result of hotter, drier climates and higher fuel loads. In addition to the environmental factors contributing to increased wildfire risk, rapid growth in the US wildland-urban interface (WUI) has raised the risk of loss of lives and homes. A key aspect of promoting socio-ecological resilience in the WUI is incentivizing actions that limit the destructiveness of eventual fires. Creating defensible space by reducing vegetation (fuel load) to prevent or slow fire spread to and from structures is a widely supported strategy for reducing the risk of loss to wildfire; however, novel analytical methods are required to quantify fuel loads, and thus wildfire risk, at the spatial scales relevant to mitigation planning. Wildfire hazard mitigation is greatly needed in communities throughout the WUI in the American West, but resources for mitigation are limited. Wildland fire risk intersects with social vulnerability, where residents with higher social vulnerability are less likely to be able to afford expensive fuels reduction treatments or engage with wildfire mitigation programs. It is critical not only to ensure that fire-prone communities have the scientific information needed to mitigate fire threat, but also that development, delivery, and application of this information is tailored to the unique social needs of each community. Perceptions of fire risk, attitudes toward fire preparedness, and vulnerability to wildfire impacts vary within and among WUI communities, driven by socioeconomic status, place dependency, experience with fire, and other factors. This project proposes technological advances in mapping forest structure at fine spatial scales (30 m) employing artificial intelligence (AI) approaches can be used to estimate fire risk at the spatial scales relevant for mitigation planning. The fuels information will be integrated into tools and applications which will be designed to address the information needs of communities, co-developed with communities, piloted in communities, and evaluated by communities. The project proposes to scope and pilot co-development of AI-based fire preparedness assessment tools in communities in Nevada County, CA. Ultimately, the co-development of these tools will inform the deployment, evaluation, and application of fire preparedness assessment tools to community planning and policy and scaled up to a broader range of communities in diverse western U.S. geographies.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.