| Summary: | The GHub science gateway, https://theghub.org, is a new collaboration space for scientists, educators, and students who are working to understand our planet's ice sheets. Our gateway hosts ice sheet datasets, modeling and analysis workflows, and community codes, provides access to high-performance computing, and enables tool building by the cryospheric science community. GHub bridges the gap between the numerical ice sheet modeling community with observational datasets of past and present ice sheet states that will ultimately improve predictions of sea level rise. GHub also provides a common point of engagement for other scientists seeking to engage with the ice sheet community. GHub is based on the open-source HUBzero platform. When users run computational tools on GHub, the gateway spins up Docker containers for Jupyter Notebooks, Linux workspaces, or interactive development environments (IDEs) such as RStudio. Tools needing more computational power leverage remote high-performance computing resources at University at Buffalo's Center for Computational Research (CCR). GHub users may develop, share and release their own tools and documentation on the gateway, using compilers, source control, and a guided tool development workflow. Users can also submit tool reviews, add to wishlists for future enhancements, and submit questions to tool developers. During the project's pilot, we developed 8 computational tools, and hosted the Ice Science Modeling Intercomparison Projects (ISMIP6) dataset and other data, totalling 7 TB. Now in the first year of a five-year grant, our user base is 85 members strong, and growing. We are working to host important community codes such as NASA's Ice-sheet and Sea-level System Model (ISSM) and Cryosphere model Comparison tool (CmCt). Furthermore, we are integrating the gateway's tools with crucial datasets stored at locations such as National Snow and Ice Data Center (NSIDC). These tools will be used by ice sheet scientists, educators, and for workshops and outreach. GHub is supported by National Science Foundation and EarthCube grants.
|
|---|