RII Track-2 FEC: Natural Resource Supply Chain Optimization using Aerial Imagery Interpreted with Machine Learning Methods

RII Track-2 FEC：使用机器学习方法解释的航空图像优化自然资源供应链

• 批准号: 2119689
• 负责人: Jesse Johnson
• 金额: $ 391.41万
• 依托单位: University of Montana
• 依托单位国家: 美国
• 项目类别: Cooperative Agreement
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-02-15 至 2026-01-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2119689&HistoricalAwards=false
• 关键词: RII; Track; FEC; Natural; Resource

The University of Montana and the University of Alaska, Anchorage will conduct scientific research that is responsive to the needs of their natural resource based economies. Specifically, questions from the areas of snow water resources, wildland fire management, and abandoned oil well monitoring will drive the research agenda. The scientific work is united by a common set of techniques for answering questions. Moreover, the common approach is to: 1) Use unmanned flying vehicles called drones to collect pictures and other measurements. While the information acquired by drones is high quality, it is also a large amount of complex data. 2) To aid in the data’s interpretation and address the science questions, we use machine learning methods that train computers to identify patterns in data. Collecting data using drones and the use of machine learning are critical skills for America’s future workforce. Our activities are aligned with career training through partnerships with local companies and internships for participating students. With the emphasis on internships, the focus will be retention, rather than recruitment of students. Diversity and inclusion efforts will work in tandem with workforce development to ensure that Indigenous, low income, and rural members of our jurisdictions are integral to the efforts. It is proposed to use machine learning (ML) to process imagery and other data acquired by autonomous aerial systems (UAS). Processed data will support scientific research in natural resource management by providing a clear means of testing hypotheses. The three areas of natural resource management to apply this approach are: 1) snow water resources, because energy production, agricultural output, and economic growth require improved assessment of the natural capital banked in the mountain snowpack. 2) fire management and science, because an advanced understanding of the physical and ecological processes driving wildfire is required for management practices that better protect forests and the critical infrastructure within them. 3) abandoned oil well monitoring, because detecting and mapping uncapped or improperly sealed oil and gas wells will provide critical data for improved mitigation, site reclamation, and hazard removal. The team bring together the University of Montana and the University of Alaska Anchorage to conduct locally relevant research. The local focus strengthens the relations with nearby commercial interests to identify questions common to academic and commercial endeavors in modern natural resource based economies. Attention is called to UAS and ML as skills vital to both commercial and academic work. To advance the natural resource based advanced manufacturing industries of the jurisdictions, special consideration is given to developing a workforce that spans from two-year college graduates to junior faculty members. This award's plan centers on paid, credit-bearing internships with key partners, drawing commercial interests into the proposed work, and bringing junior faculty to the research. Diversity and inclusion efforts will work in tandem with workforce development to ensure that Indigenous, low income, and rural members of the jurisdictions are an integral part of efforts. With the emphasis on internships, the focus will be retention, rather than recruitment of students. The project engages with disruptive technologies that have long reaching consequences for workers. To address the consequences related to these technologies, the team enlisted the aid of a social scientist to conduct studies evaluating the social and economic impacts.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.

蒙大拿大学和阿拉斯加大学安克雷奇大学将进行科学研究，以响应其基于自然资源经济的需求。具体而言，来自雪水资源，野外消防管理和废弃石油井监测的问题将推动研究议程。科学工作由回答问题的一系列通用技术团结在一起。此外，常见的方法是：1）使用名为无人机的无人飞行器收集图片和其他尺寸。尽管无人机获取的信息是高质量的，但它也是大量复杂数据。 2）为了帮助数据的解释并解决科学问题，我们使用机器学习方法来训练计算机来识别数据中的模式。使用无人机收集数据并使用机器学习是美国未来劳动力的关键技能。我们的活动与与当地公司的合作伙伴关系以及为参与学生的实习而进行的职业培训保持一致。着重于实习，重点将是保留，而不是招募学生。多样性和包容性工作将与劳动力发展协同工作，以确保我们司法管辖区的土著，低收入和粗糙的成员是努力不可或缺的。建议使用机器学习（ML）处理由自动空中系统（UAS）获取的图像和其他数据。处理后的数据将通过提供清晰的测试假设的方法来支持自然资源管理中的科学研究。采用这种方法的自然资源管理的三个领域是：1）雪水资源，因为能源生产，农业产出和经济增长需要改善对山脉中养护的自然资本的评估。 2）消防管理和科学，因为对驱动野火的物理和生态过程的深入理解是更好地保护森林和其中关键基础设施的管理实践。 3）废弃的石油井监测，因为检测和映射未封闭或密封的石油和天然气井将提供关键数据，以改善缓解，现场开垦和危害的清除。该小组将蒙大拿大学和阿拉斯加大学的锚定汇集在一起​​，进行本地相关的研究。当地的重点增强了与近乎商业利益的关系，以确定现代基于自然资源经济中学术和商业努力常见的问题。注意UAS和ML是对商业和学术工作至关重要的技能。为了促进司法管辖区的基于自然资源的高级制造业，对开发从两年年大学毕业生到初级教师的劳动力进行了特殊考虑。该奖项的计划集中在与主要合作伙伴的付费信用实习中，将商业利益纳入拟议的工作，并将初级教师带入研究。多样性和包容性工作将与劳动力发展协同工作，以确保司法管辖区的土著，低收入和粗糙的成员是努力不可或缺的一部分。着重于实习，重点将是保留，而不是招募学生。该项目与颠覆性技术的互动对工人产生了长期影响。为了解决与这些技术有关的后果，该团队邀请社会科学家的帮助进行评估社会和经济影响的研究。该奖项反映了NSF的法定任务，并通过使用该基金会的知识分子的优点和更广泛的影响来审查标准，认为NSF的法定任务是宝贵的支持。

项目成果

Application of LiDAR Derived Fuel Cells to Wildfire Modeling at Laboratory Scale

LiDAR 衍生燃料电池在实验室规模野火建模中的应用

• DOI: 10.3390/fire6100394
• 发表时间: 2023
• 期刊: Fire
• 作者: Marcozzi, Anthony A.;Johnson, Jesse V.;Parsons, Russell A.;Flanary, Sarah J.;Seielstad, Carl A.;Downs, Jacob Z.
• 通讯作者: Downs, Jacob Z.

Cold Season Rain Event Has Impact on Greenland's Firn Layer Comparable to Entire Summer Melt Season

冷季降雨事件对格陵兰岛冷杉层的影响相当于整个夏季融化季节

• DOI: 10.1029/2023gl103654
• 发表时间: 2023
• 期刊: Geophysical Research Letters
• 影响因子: 5.2
• 作者: Harper, J.;Saito, J.;Humphrey, N.
• 通讯作者: Humphrey, N.