It’s raining cats and dogs when a hydrologist, a climatologist, a statistician, an economist, and a geospatial scientist walk into a bar. The bar is just down the road from a big old dam and they’d each glanced toward it when they’d pulled into the parking lot.
After a few weeks of steady rain, the whole dam area has been on everyone’s mind. The hydrologist is thinking about the last time this dam overtopped. The climatologist and the statistician are debating how extreme this amount of precipitation would be considered and what may come in the future. The economist is remembering how disruptive flooding in this area is to the local businesses that export their goods to distant places, and the geospatial scientist is estimating the extensive numbers of vulnerable neighborhoods downriver.
Over a beverage, a few particular questions come to mind. How could we characterize the communities that are downstream from this dam? This dam overtopped before in their lifetimes, but wasn’t that when the other dam upriver had water released? How had the local governments and agencies responded to these crises in the past? Had it also been raining for so long then?
They come to the realization that these questions together are bigger than any one of them usually considers. They would need a coordinated mind meld. Then, they realize, as they nervously reach for more pretzels, there is the data they’ll need. It might be big data. Data that may take some time to access, to organize, and to process; precipitation data and streamflow data, both now and from the past; census data and elevation data, maybe at different spatial scales; text data from local news media and government sources. Never before had any one in the small group known how to find and clean up the data from outside of their disciplines, much less combine and analyze it in a meaningful and valid way. Would this be the beginning of a beautiful friendship? Or would they just end up just needing a bigger boat?
Sustainability and resiliency around topics such as aging dams are complex goals that require innovative and insightful approaches, and lots and lots of data. Tried-and-true, disciplinary-based methods aren’t up to the task now, much less what’s coming. I-GUIDE — the Institute for Geospatial Understanding through an Integrative Discovery Environment — is confronting this critical vulnerability by creating a computing environment that supports scientific integration and discovery. Based at the University of Illinois Urbana-Champaign and led by Dr. Shaowen Wang, professor of geography and geographic information science, I-GUIDE is a collaboration between 11 institutions and multiple scientific and engineering disciplines. Its mission is to transform convergence and geospatial sciences for solutions to complex societal problems. We bring together disciplinary experts and enable their work to become convergent, meaning innovative solutions that become possible when expertise is combined. This was the intent behind the National Science Foundation selecting I-GUIDE as one of the institutes to fund under its Harnessing the Data Revolution big idea.
Supporting and enabling convergence science outcomes is central to I-GUIDE’s vision. Two characteristics make research become convergent, according to NSF: that the inquiries are driven by a particular compelling problem, often with a heavy societal impact, such as situations with our aging infrastructure, and that coordinated collaboration across disciplines is what leads to solutions. This involves effective blending of research methods, theories, and, of course, the substantial use of data, which is where I-GUIDE’s vision, mission, and platform enter the scene.
The I-GUIDE platform is designed to harness the vast, diverse, and distributed geospatial data at different spatial and temporal scales and make them broadly accessible and usable to convergence research and education enabled by cutting-edge cyberGIS and cyberinfrastructure. I-GUIDE recognizes the enormous time and cost involved in data discovery, exploration, and integration — data wrangling — that are prerequisite to scientific analysis and modeling. Accelerating these data-harnessing processes will not only improve time-to-insight but, as importantly, will catalyze discovery by enabling many science questions that remain unpursued due to the high cost of data wrangling.
Dr. Michael Tischler, director of the U.S. Geological Survey’s National Geospatial Program and member of I-GUIDE’s External Advisory Board said, “FAIR+ and open data increases the transparency and defensibility of scientific studies, as well as the speed at which they are conducted. Not only can scientists across and between disciplines have ready access to reliable data that is interoperable in their own analytical workflows, the entire community and the public are able to investigate and understand the data upon which key conclusions are made.”
The I-GUIDE platform is where separate cyberinfrastructure components converge to form a seamless environment to support the kind of interdisciplinary research and education envisioned by the institute. Several leading CI groups have brought advanced geospatial data, science gateways, and high-performance computing capabilities, tools and standards into the I-GUIDE partnership. The I-GUIDE platform thus provides core cyberinfrastructure capabilities and services for data preparation, improvements in model and data interoperability and workflows, and support education and workforce training in data science literacies.
“By integrating these capabilities, the platform serves geospatial datasets on demand, providing computational capacity from the national CI ecosystem,” said Dr. Carol Song, the chief scientist at the Rosen Center for Advanced Computing at Purdue University. “I-GUIDE researchers and the broader community can use the platform to find and access datasets that are readily usable. The I-GUIDE platform is easy to access and it serves as a hands-on practice environment for I-GUIDE’s education and engagement activities.”
The platform clearly plays a central role in enabling convergence science to take place, and it’s I-GUIDE’s large set of graduate students, post-docs, and junior research staff interacting together who are often making it happen. The group — dubbed The Climbers for their status as rising stars — has developed a camaraderie over their Slack channel and during weekly Zoom meetings. Collectively, the group of about 30 represents more than a dozen different disciplines, and they embody a collegial and safe space to ask questions, share technology tips and tricks, and experience how professional research networks can be cultivated and supported.
“Being part of The Climbers group from a social science point of view has been rewarding,” said Bailey Holdaway, a research technician in environment and society at Utah State University. “I get to dive headfirst into other fields while also being able to share social science techniques like interviewing and conducting surveys with peers from other disciplines.”
Another way that I-GUIDE is pursuing its goals is through its development of a convergence curriculum for geospatial data science. The curriculum itself is convergent; by design it’s structured so that five topics (ethics, analytics, geospatial, computing, and visualization) are at the base as foundational knowledge threads. These are then combined in a deliberate manner that contributes towards convergence science, through geoethics, geospatial computing, geospatial analytics, and geovisualization into geospatial artificial intelligence and data science.
With the addition of separate domain knowledge, such as economics and hydrology, the right knowledge and skills are harnessed to tackle a convergence science problem, such as aging dams.
A guiding principle of the curriculum is to help each group of experts learn a little about what their counterparts know and do. Since its components are modular, they can be explored in any way a user needs. The effort put forth by a learner can be succinct dabbling (three sentences or three slides about a topic) or a more targeted commitment (a three-hour or a three-week module). These aren’t intended as substitutes for a degree program, but rather just the right amount for curious learners at any level to feel more confident and competent when interacting with colleagues or potential collaborators from a different field. That can have a significant positive impact within the realistic confines of a busy professional schedule.
I-GUIDE can achieve its ambitious impacts only when it’s successful at supporting new audiences and new applications beyond its initial test cases. The platform itself is ready for people to give it a try, and expect to see this collection of Jupyter Notebook access points to be expanded in the near future as current projects come online. The I-GUIDE team is eager for guests to register and provide feedback. The I-GUIDE Convergence Curriculum materials, which also include Jupyter Notebooks as components within the three-hour modules, are also open for business.
Several in-person and online opportunities are also upcoming. Graduate students and early-career scholars wanting to experience I-GUIDE’s approach to convergence science may apply for a five-day summer school in early August at the University Corporation for Atmospheric Research in Boulder, Colorado. During that week, small teams will engage in convergence science in action while deploying geospatial data science techniques and taking advantage of the new platform’s capabilities. I-GUIDE regularly hosts Virtual Consulting Offices, online sessions to learn about the types of technical, analytical, and computational methods that are increasingly part of modern geospatial data science.
I-GUIDE’s tagline is “Map, Connect, and Discover.” The project is doing just that: enabling easier access to data that drives our maps, inspiring our connections, and accelerating our discoveries. This is what they started figuring and celebrating in the bar that night.
For more information about I-GUIDE, including access to the platform and all open curricular materials, registration for upcoming virtual consulting offices and webinars, or signing up for future announcements via the mailing list, please visit https://i-guide.io. We welcome inquiries about partnerships and other connections!