Frameworks: Collaborative Research: Integrative Cyberinfrastructure for Next-Generation Modeling Science

点击翻译按钮获取中文摘要

This project is designed to support and advance next generation, interdisciplinary science of the complexly interacting societal and natural processes that are critical to human life and well-being. Computational models are powerful scientific tools for understanding these coupled social-natural systems and forecasting their future conditions for evidenced-based planning and policy-making. This project is led by the Network for Computational Modeling in Social and Ecological Sciences (CoMSES.Net). CoMSES.Net's science gateway promotes knowledge sharing among scientists and with the general public, and enables open, online access to sophisticated computational models of social and ecological systems. CoMSES.Net's partners in this project (the Community Surface Dynamics Modeling System and Consortium of Universities for the Advancement of Hydrologic Science) also enable knowledge sharing and provide open, online repositories of models in the earth sciences. This project will enhance these science gateways and create online educational materials to make these critical technologies easier to find, understand, and use for scientists and non-scientists alike. By integrating innovative technology with training and incentives to engage in best practice standards, this project will stimulate innovation and diversity in modeling science. It will enable researchers to build on each other's work and combine it in new ways to address societal and environmental challenges. The cybertools and educational programs developed in the project will be openly accessible not just to research institutions but also to smaller colleges, state and local governments, and a broader audience beyond the science community. The project will give decision-makers and the data scientists who support them access to a larger and more varied toolkit with which to explore potential solutions to societal and environmental policy issues. A long-term aim of the project is to support an evolving ecosystem of diverse, reusable, and combinable models that are transparently accessible to anyone in the world. Sustainable planetary care and management is a challenge that confronts all of humanity, and requires knowledge, histories, methods, perspectives, and engagement of researchers, decision-makers, and private citizens across the country and throughout the world.The project will develop an Integrative Cyberinfrastructure Framework (ICF) to enable innovative next-generation modeling of human and natural systems, and build capacity in modeling science. It will support a set of activities that integrate the human and technological components of cyberinfrastructure. 1) Software tools will be developed that augment model codebases with modern software development scaffolding to facilitate reuse, integration, and validation of model code. 2) The project will provide high-throughput computing (HTC) resources for simultaneously running numerous iterations of models needed to capture stochastic variability, explore a parameter space, and generate alternative scenarios; 3) Online training activities will build expertise and capacity to make effective use of the cybertools and the HTC resources; 4) The ICF will engage a global modeling science community to provide professional incentives that encourage researchers to adopt best practices and catalyze innovative science. Leveraging existing NSF investments, the ICF will be developed and deployed by the Network for Computational Modeling in Social and Ecological Sciences (CoMSES.Net), in partnership with the Community Surface Dynamics Modeling System (CSDMS), Consortium of Universities for the Advancement of Hydrologic Science (CUAHSI), Open Science Grid, Big Data Hub/Spoke network, and Science Gateways Community Institute. Computational models have emerged as powerful scientific tools for understanding coupled social-biogeophysical systems and generating forecasts about future conditions under a range of climate, biogeophysical, and socioeconomic conditions. CoMSES.Net, CSDMS, and CUASI are scientific networks, with online science gateways and code archives that enable open access to computational models for an international community of social, ecological, environmental, and geophysical scientists. However, the full value of accessible, well-documented models only can be realized if their code is also widely reproducible and reusable, with a potential for integration with other models. In order to confront critical challenges for understanding the coupled human and natural systems of today's world, modeling scientists also need HTC environments for upscaling models and exploring high-dimensional parameter spaces inherent in representing these systems. The ICF is designed to meet these challenges. By integrating technology with intellectual capacity-building, the ICF will stimulate innovation and diversity in modeling science by letting creative researchers build on each other's work more readily and combine it in new ways to address societal-environmental challenges we have not yet perceived. The tools and training resources will be openly accessible not just to leading research institutions but also to the many smaller colleges, state and local governments, and a broader audience beyond science. They will provide decision-makers and the data scientists who support them access to a much larger and more varied toolkit with which to explore potential solution spaces to social and environmental policy issues. The proposed ICF is also designed to help transform scientific modeling practice, including incentives that can help early career researchers shift from creating models to solve problems specific to a particular project to models that are also useful for others. The project will help support a future evolving ecosystem of diverse, reusable, and integrable models that are transparently accessible to the broader community.This project is funded by the Office of Advanced Cyberinfrastructure in the Directorate for Computer and Information Science and Engineering, with the Division of Social and Economic Sciences in the Directorate for Social, Behavioral & Economic Sciences also contributing funds.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.