RCN: Community-based educational infrastructure for numerical simulation in the Earth Sciences: a reactive transport use case

在整个物理科学中，许多系统表现为耦合和相互依赖的化学，机械和生物过程的结果，这些过程过于复杂，如果没有数值模拟技术的帮助就无法解决。现在比以往任何时候都更重要的是，准确预测物理化学过程，如那些决定土壤健康、作物生长和水质的过程，对于维持我们不断增长的全球人口至关重要。幸运的是，现代计算能力的进步是非凡的，最近向更多开源计算平台的过渡意味着下一个十年的研究将比以往更快地开发、使用和评估模型。因此，提高所有职业阶段的模范素养至关重要。这种培训确实可能是我们利用最新技术进步来解决从资源可持续性到新材料发现等社会相关问题的能力的关键限制步骤。目前的RCN将把地球科学作为开发开环学习框架的试验台：即教育可以而且应该是一个持续的过程，而不是上过大学或上过专题短期课程的结果。为了支持模型使用、开发和管理的持续教育平台，支持者将创建一个由建模能力和规则（MAR）组成的新颖的、开放的框架。这些能力和相关规则将使个人能够匹配和利用其本国机构提供的资源、通过在线学习获得的资源和社区努力提供的资源，以加强和扩大其特定技能和知识水平。除了加速地球科学模型的采用和使用外，平台支持者将创建的平台可以被物理科学中的其他社区广泛推广，这些社区依赖于数值模型和相关的科学工作流程来培养模型素养。数值反应输运模型（RTMs）已经成为推动地球科学广泛领域知识的关键工具，从早期成岩作用和大陆风化作用到污染物的命运和运输以及核废料的储存。然而，与广泛的应用程序和感兴趣的用户相比，此类软件的早期开发人员很少，因此，不幸的是，学生的教育机会很少。这种差异也意味着现有的用户社区很难跟上计算方法和工作流程的快速发展。为了满足这些需求，该项目的目标是开发一种基于社区的教育模式，使地球科学社区能够将反应传输模型（rtm）嵌入其科学和教育工作流程中。这项工作跨越了大学和国家实验室，他们都被激励去开发一个集成的培训平台，以解决模型能力和能够应用它们的用户群之间日益增长的差距。为了实现这一点，支持者将作为培训平台，并作为模型框架新进展的存储库和社区资源。这将通过一系列活动来实现：(1)开发一个RTM-Hub，收集和有意义地组织材料和机会，并提出一个基于能力的教学框架，以支持开环教育模式；(2)开发一个概述建模能力和规则（MAR）的新框架，作为个性化教育的路线图；(3)利用现有和新开发的在线材料增强教育基础设施；以及大多数大学通用的课程，以及(4)组织和培养一个自给自足的夏季RTM学院。该RCN提案旨在通过创建基于社区和灵活的教育框架来改变用户软件环境，该框架旨在使教育平台的发展超越个人或甚至作为一个社区所能实现的目标。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。

Across the physical sciences, many systems manifest as the result of a coupled and codependent set of chemical, mechanical and biological processes which are too complex to be solved without the assistance of numerical modeling techniques. Now more than ever, the accurate prediction of physiochemical processes such as those dictating the health of soil, the growth of crops and the quality of water is paramount to sustaining our increasing global population. Fortunately, modern advancements in computational capability are extraordinary, and recent transitions to more open-source computing platforms mean that the next decade of research will develop, employ and evaluate models faster than ever before. Thus, enhancing model literacy across all career stages is critical. Such training may indeed stand as a critical limiting step in our ability to harness the latest technological advances to address societally relevant problems, ranging from resource sustainability to the discovery of new materials. The present RCN will use the geosciences as a test bed for developing a framework for open-loop learning: the idea that education can and should be a continuous process rather than the result of having attended a University or a topical short course. In order to support a continuous educational platform for the use, development and stewardship of models, proponents will create a novel, open framework consisting of Modeling Abilities and Rubrics (MAR). The abilities and the associated rubrics will allow individuals to match and leverage resources afforded by their home institutions, those available through on-line learning and those provided by community efforts, to strengthen and expand their particular skill set and knowledge level. In addition to accelerating the adoption and use of geoscience models, the platform proponents will create can be widely generalized by other communities in the physical sciences that rely on numerical models and the associated scientific workflows to foster model literacy. Numerical Reactive Transport Models (RTMs) have become a key tool for advancing knowledge across a broad swath of the geosciences, from early diagenesis and continental weathering to contaminant fate and transport and nuclear waste storage. However, the early developers of such software have been few in comparison to the broad variety of applications and interested users, such that educational opportunities for students are unfortunately rare. This disparity also means that it is difficult for the existing community of users to stay abreast of the rapid advances in both computational approaches and workflows. To address these needs, the objective of this project is to develop a community-based educational model to enable the geoscience community to embed Reactive Transport Models (RTMs) in their scientific and educational workflows. This effort spans universities and national laboratories, all of whom are motivated to develop an integrated training platform to address a growing gap between model capabilities and the user base who can apply them. To accomplish this, proponents will serve as a platform for training and as a repository and community resource for new advancements in model frameworks. This will be accomplished through a series of activities: (1) development of an RTM-Hub that will collect and meaningfully organize materials and opportunities and present an abilities-based pedagogical framework to support an open-loop educational model, (2) development of a novel framework outlining the Modeling Abilities and Rubrics (MAR) to serve as a roadmap for personalized education, (3) enhance the educational infrastructure by leveraging existing and newly developed on-line materials, as well as a curriculum that is common across most universities, and (4) organize and foster a self-sustaining summer RTM Institute. This RCN proposal is specifically intended to change the user-software landscape by creating a community-based and flexible education framework that is designed to enable to grow the educational platform beyond what one could otherwise achieve individually, or even as a community.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.