Constraining Global Coastal Ocean Methane Emissions to the Atmosphere

该项目将确定全球沿海海洋环境的甲烷排放量，这是大气中最不确定的甲烷自然来源之一。甲烷是一种温室气体，其对未来气候变暖的影响将取决于人类来源和不断变化的自然环境的排放。因此，了解天然甲烷源向大气的基线排放率及其对变化的敏感性至关重要。虽然人们认为公海环境只向大气排放少量甲烷，但甲烷的浓度和排放率在接近海岸线时会大幅增加。沿海海洋甲烷排放量在全球范围内可能很大，但由于缺乏准确捕捉沿海分布的观测结果，因此仍然存在很大的不确定性。此外，从沿海表面沃茨排放的甲烷的来源并不为人所知，这限制了我们预测未来排放量如何变化的能力。该项目将确定甲烷从沿海海洋到大气的来源和全球排放率，并建立一个框架来预测气候变暖下的未来排放率。除了这些科学和社会的影响，该项目将有很强的教育影响，因为它将为本科生提供机会，体验整个科学过程，从想法的概念，最终结果的出版。罗切斯特大学的PI已经建立了一系列课程，从海洋科学的角度指导学生完成这一过程。这个项目将作为下一次课程迭代的重点，参与的学生将是研究的重要贡献者。在以前进行时，这种教育推广使本科生能够追求自己的科学兴趣，并导致大量学生追求海洋科学的研究生职业。该项目还将支持一个博士学位。学生在一个真正独特的经验，使她/他将有机会进行有意义的研究，在这两个航海测量以及建模实验室，从而融入两个往往不同的社区。这个项目将通过一个独特的和平等的结合观测和统计建模工作，利用PI和Co-PI实验室中已建立的方法，在项目的2.5年期间取得快速进展。总的来说，四个沿海地区的地表甲烷浓度“跨越三个不同的海洋盆地和亚热带到亚极地纬度范围”，将使用超快速真空提取方法进行测量，从而产生沿海数据覆盖范围，这在以前的数据集中是无与伦比的。此外，还将测量地表甲烷的放射性碳含量，以确定其在化石和微生物来源之间的来源，并收集包括叶绿素、营养物质和溶解氧浓度在内的地球化学数据。最初的巡航数据（第1年）将用于训练人工神经网络模型，以预测地表甲烷过饱和度随地球化学变量的变化，以后的巡航（第2年）将允许在未用于训练的区域进行独立的模型验证。在建立了该模型可以在沿海环境之间推广的保真度之后，它将被应用于外推甲烷过饱和度的地图，并估计区域和全球尺度的沿海甲烷排放量，同时量化其不确定性。总的来说，这项工作将填补我们在天然甲烷预算方面的知识空白，产生迄今为止最可靠的沿海海洋排放量估计，并对维持这些排放量的机制有了新的理解。该奖项反映了NSF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。

This project will determine global methane emissions from coastal marine environments, one of the most uncertain natural sources of methane to the atmosphere. Methane is a greenhouse gas whose impact on future climate warming will depend on emissions from both human sources and the changing natural environment. It is therefore critical to understand the baseline emission rates of natural methane sources to the atmosphere as well as their sensitivity to change. While the open ocean environment is thought to emit only minor amounts of methane to the atmosphere, concentrations and emission rates of methane increase substantially approaching coastlines. Coastal ocean methane emissions are potentially significant at the global scale but remain highly uncertain due to a lack of observations that accurately capture coastal distributions. Furthermore, the source of methane emitted from coastal surface waters is not well known, limiting our ability to predict how emissions will change in the future. This project will determine the source and global emission rate of methane from the coastal ocean to the atmosphere, and establish a framework to predict future emission rates in a warming climate. In addition to these scientific and societal impacts, this project will have strong educational impacts as it will provide undergraduate students the opportunity to experience the entire scientific process from idea conception to publication of the final results. A sequence of classes has been established by the PI at the University of Rochester to guide students through this process from an ocean science perspective. This project will serve as the focus for the next iteration of the class sequence, and participating students will be vital contributors to the research. When conducted previously, this educational outreach has empowered undergraduates to pursue their own scientific interests and has led to significant numbers of students pursuing graduate careers in the ocean sciences. This project will also support a Ph.D. student in a truly unique experience whereby she/he will have the opportunity to conduct meaningful research in both sea-going measurement as well as modeling laboratories, and thus integrate into two often disparate communities.This project will be accomplished through a unique and equal combination of observational and statistical modeling work, leveraging methodologies that are well established in the PI and co-PI's laboratories to make rapid progress over the 2.5-year duration of the project. In total, surface methane concentrations in four coastal regions "spanning three different ocean basins and subtropical to subpolar latitude ranges" will be measured using an ultra-fast vacuum extraction method, yielding coastal data coverage that is unparalleled in previous datasets. Additionally, the radiocarbon content of surface methane will be measured to fingerprint its provenance between fossil and microbial sources, and biogeochemical data including chlorophyll, nutrient, and dissolved oxygen concentrations will be collected. Initial cruise data (year 1) will be used to train Artificial Neural Network models to predict surface methane supersaturation as a function of biogeochemical variables, and later cruises (year 2) will allow for independent model validation in regions that were not used for training. Having established the fidelity with which this model can generalize between coastal environments, it will be applied to extrapolate maps of methane supersaturation and estimate regional and global scale coastal methane emissions while quantifying their uncertainty. Overall, this work will close these gaps in our knowledge of the natural methane budget, yielding the most robust estimates to date of coastal ocean emissions and a new understanding of the mechanisms that sustain them.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.