Collaborative Research: GEO OSE Track 1: Advanced cloud-based Data- and Visualization-Integrated Simulation EnviRonment (ADVISER) to Advance Computational Glaciology

近年来，云计算的采用得到了极大的发展，尤其是在商业领域。然而，这项技术在计算地球科学中的应用，特别是在极地科学和冰川学方面，却落后了。对于在这些领域工作的科学家来说，云计算可以极大地简化用户访问和利用高性能计算资源以及开源多物理场仿真软件的方式。为了实现这些好处，有必要创建开放、包容的网络基础设施，弥合现有的知识差距，并开展外展活动，以改善云和现有地球科学软件平台的采用。本项目通过以下方式实现了这些目标：(1)构建一个可供计算地球科学家使用的开放式云计算平台——Advanced cloud-based Data and Visualization Integrated Simulation EnviRonment (ADVISER)；(2)扩展现有的地球科学软件（Icepack），使其能够在ADVISER上轻松高效地运行。将ADVISER扩展到计算冰川学研究和教育中，使云计算的力量民主化，并使那些很少或没有云专业知识的人受益。通过它的设计，使用ADVISER不需要特殊的硬件或知识，所以学生在课堂环境中可以参与到新的课程计划中，比如在课堂上模拟冰川流动。通过计划中的研讨会/黑客马拉松，该项目将涉及几所大学和学院的学生，教育他们更好地利用计算能力（通过云计算），这是一种传统上无法通过地球科学或工程课程获得的技能。该项目包括四项具体任务：(1)扩展云原生平台ADVISER，用于在云中运行和管理科学软件和数据；(2)将冰川流模拟软件Icepack应用于advisor平台；(3)创建数据同化管道，以便将各种现有的和新的数据集以适合在Icepack等软件中使用的格式导入ADVISER平台；(4)举办研讨会和黑客马拉松，促进ADVISER和Icepack在更广泛的计算地球科学界的可持续采用和扩展。研究人员的愿景是，他们的ADVISER平台将允许将新兴的遥感数据、机器学习算法和基于物理的正反问题计算模型无缝结合。研究人员将特别关注南极冰架的真实数据同化。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。

The adoption of cloud computing has increased tremendously over the years, especially in business. However, the use of this technology in computational geosciences—and particularly, in polar sciences and glaciology—has lagged behind. For scientists working in these areas, cloud computing can greatly simplify how users access and leverage high-performance computing resources, as well as open source multiphysics simulation software. To realize these benefits, it is necessary to create open, inclusive cyberinfrastructure that bridges existing knowledge gaps and to conduct outreach to improve adoption of the cloud and existing geosciences software platforms. This project achieves these goals by: (1) building an open cloud computing platform, Advanced cloud-based Data and Visualization Integrated Simulation EnviRonment (ADVISER), that can be used by computational geoscientists; and (2) by extending existing geosciences software (Icepack) to run easily and efficiently on ADVISER. Extending ADVISER for use in computational glaciology research and education democratizes the power of cloud computing and gives its benefits to those with little or no expertise in the cloud. By its design, no special hardware or knowledge is required to use ADVISER, so students in classroom environments will be able to participate in new kinds of lesson plans, such as running glacier flow simulations on ADVISER during class sessions. Through a planned workshop/hackathon, this project will involve students across several universities and colleges, educating them to better harness computational power (via the cloud), a skillset not traditionally available through course curricula in geosciences or engineering.The project comprises four specific tasks: (1) extending a cloud-native platform, ADVISER, for running and managing scientific software and data in the cloud; (2) adapting the Icepack software for modeling of glacier flow to run on ADVISER; (3) creating data assimilation pipelines that enable importing various existing and novel data sets into the ADVISER platform in formats amenable to usage in software like Icepack; and (4) hosting a workshop and a hackathon that spur sustainable adoption and extensions of ADVISER and Icepack for the broader computational geosciences community. The investigators' vision is that their ADVISER platform will allow for the seamless incorporation of emerging remote sensing data, machine learning algorithms, and physics-based computational models for forward and inverse problems. The investigators will particularly focus on real-world data assimilation from Antarctic ice shelves.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.