Our research is in the interdisciplinary field of hydroinformatics that combines water resources engineering, hydrology, computer science, and data science. The goal of our research is to leverage advances in computing and data science to address problems and challenges associated with water resources management such as improving community and infrastructure resilience to climate change.
Climate change is resulting in more intense storms and rising sea levels that are causing significant water-related impacts for many communities across the world. We believe that innovations in hydroinformatics, such as real-time flood warnings or real-time control of stormwater infrastructure, can help communities adapt to these new environmental conditions.
UVA Hydroinformatics Group members at the American Geophysical Union Fall Meeting, 2018
Stormwater infrastructure is designed to mitigate the impacts of urbanization caused by increased runoff. This increased runoff, if not properly handled by stormwater infrastructure, can not only increase flooding risks, but also pollute downstream waterbodies. Many other civil infrastructure systems have benefited from real-time sensing and control capabilities to improve their utility. These capabilities are just now coming to stormwater systems. Taking general approaches and ideas from the broader field of cyber-physical systems (CPS), we are helping to advance smart stormwater systems where real-time sensing and control through actuated gates, valves, and pumps can help to improve the utility of existing systems. We see climate change as a major driver for this needed change where existing infrastructure could be enhanced at a lower cost to better address changing climate conditions.
While our ability to forecast severe weather has significantly improved over past decades, our ability to forecast on-the-ground flooding impacts at a high spatial resolution are lacking. This problem is challenging due to the complexity of modeling floods over landscapes, especially landscapes with low topographic relief or with significant urbanization. We are advancing automated data processing workflows and big data management schemes to quickly create detailed hydrologic and hydraulic models for flood forecasting, cloud-computing architectures for dynamic resource allocation in support of real-time, high-resolution flood modeling, and decision support systems to engaging end users in flood warning. Much of this work is focused on impacts to transportation infrastructure before, during, and following flooding events.
Underlying these advances is the need for novel cyberinfrastructure to support hydrologic modeling and data analysis at scale. This cyberinfrastructure must enable access to high performance computing (HPC), cloud computing, automated workflows, containerization of legacy modeling codes, and other advances needed for the general reproducibility of computational hydrologic modeling. We have been long-term members of national teams building advanced cyberinfrastructure for the hydrologic science and engineering communities through our partnership with the Consortium for the Advancement of Hydrologic Science, Inc. (CUAHSI). This research has resulted in the CUAHSI Hydrologic Information System (his.cuahsi.org) and the CUAHSI HydroShare system (www.hydroshare.org). It has also resulted in partnerships with computer scientists funded through the NSF EarthCube and related cyberinfrastructure programs to advance the state of the art in scientific cyberinfrastructure.