|Understanding Coastal Hydrodynamic Processes and Mitigating Risk through High Fidelity Computer Simulations
The Computational Hydraulics Laboratory (CHL) at the University of Notre Dame is focused on developing cutting edge technologies to simulate wave and circulation environments in the coastal ocean and adjacent coastal floodplain. Through the development of high resolution coastal circulation and wave models, CHL has played a pivotal role in the development of hurricane and extra-tropical storm hazard modeling and has shaped today’s state-of-the-art high fidelity and high performance codes and methodologies that are applied by the U.S. Army Corps of Engineers (USACE), the Federal Emergency Management Agency (FEMA), the National Oceanic and Atmospheric Administration (NOAA), the U.S. Nuclear Regulatory Commission (NRC), the U.S. Coast Guard (USCG), and universities and private sector companies worldwide.
CHL and its partners have been key movers in the development of comprehensive multi-process, multi-scale, integrated domain, and high resolution models. The laboratory has pioneered unstructured heterogeneous mesh hydrodynamic models that account for:
- Riverine flow
- Winds and nearshore atmospheric boundary layer development
- Wind waves
- Atmospheric pressure
- Wind wave-current interaction
- Sediment erosion
Our models apply large domains that extend from the deep ocean, over the continental shelf, into coastal waters, and across the floodplain and into vulnerable coastal riverine systems to capture the wide range of scales and processes that occur. The models provide high mesh resolution locally to resolve the system’s physical features and the hydrodynamic processes. The laboratory is involved in fundamental algorithmic development to solve the underlying partial differential equations; in optimized coding targeted to available and evolving computer architectures including high performance parallel architectures; in coupling multiple physical processes; in applications to oceans, continental shelves, estuaries, rivers, and coastal floodplains across the world; and in verification, validation and uncertainty quantification.
CHL’s graduate students and postdoctoral fellows have gone on to faculty positions at leading research universities, to planning and high-tech companies, and to government agencies including NOAA, the USACE, and the U.S. Navy.
The laboratory has focused on developing tomorrow’s technology and creating innovations that will yield long-term improvements in understanding physical processes along coasts and in accessing the associated risks to populations and infrastructure. The laboratory has very successfully transitioned their latest codes, technologies and models to practitioners for a wide range of applications including the analysis and design of major flood control projects as well as hurricane related risk assessments. Application regions include coasts along the North Atlantic Ocean, the Gulf of Mexico, the Caribbean Sea, the Pacific Ocean, the Arctic Ocean, the South China Sea and the Indian Ocean.
|Founding Developers of ADCIRC, Today's Leading Modeling Technology for Evaluating Coastal Flood Risk
CHL has co-developed, with the University of North Carolina at Chapel Hill and the University of Texas at Austin, the widely used ADCIRC finite element based coastal ocean circulation code. ADCIRC has evolved into a community-based code and has a wide range of users within academia, government, and the private sector worldwide. An annual user group meeting, a listserv, and a training camp facilitate usage and enable our user base.
A partial list of applications includes:
- The USACE extensively uses the ADCIRC code to assess coastal currents and water surface elevations.
- ADCIRC was used to design the $15 billion flood risk mitigation system recently completed in Southern Louisiana by the USACE.
- The USACE is now applying ADCIRC for a comprehensive post Hurricane Sandy coastal flood assessment of the entire east coast of the U.S.
- FEMA applies ADCIRC to evaluate hurricane flood risk along the U.S. East and Gulf coasts.
- NOAA uses ADCIRC in support of operational forecasting of tides, extra-tropical storms, and tropical storms for U.S. coasts and estuaries.
- The U.S. Coast Guard uses ADCIRC forecasts in their operational missions.
- Industrial applications of ADCIRC range from U.S. nuclear power station flood risk assessment required by the NRC to tidal power station design.
- U.S. industrial users include Michael Baker Corporation, AECOM, Arcadis, URS, Dewberry, IBM, FMGlobal and Taylor Engineering.
Our work has led to significant advancements in understanding and evaluating hurricane related risk associated with waves, storm surge, and currents in the coastal floodplain, coastal rivers and on the continental shelf. Target applications include risk and vulnerability assessments of coastal structures, ports, ships, moored barges, utility and energy distribution systems, offshore and onshore pipelines and energy production rigs on the continental shelf. In addition to deterministic and stochastic assessments of water levels, evaluation of hydrodynamic forces due to waves, currents and erosion potential make it possible to design and/or strengthen infrastructure to be sufficiently robust and resilient in order to withstand storm forces. The high resolution grids applied make it possible to examine areas far into the coastal floodplain and within complex deltaic riverine systems.
Advancing the State-of-the-Art in Coastal Simulations
Coastal ocean hydrodynamic modeling technology has evolved rapidly over the past two decades with exponential growth in computational power and concurrent improvements in observational data from wind observations, land use and cover, topography, bathymetry, and water level, wave, and current observations. There have been substantial investments from federal agencies as well as the private sector.
Today, we have modeling technology available that is high fidelity and that is fully portable. Specifically, the technology can be applied to the entire range of historical storms and can be applied globally with no storm or region specific tuning being necessary. All model coefficients are data driven, making the technology applicable globally. This also means that accurate assessments of hurricane risk can now be developed that are simply not available from measured historical records and that can determine risk in a changing climate.
CHL continues to improve and refine the state-of-the-art coastal hydrodynamic codes and models. The emphasis is on further improving accuracy while dramatically reducing computational costs, and in developing model automation, thus making the technologies more widely accessible.
Recent research directions include:
- Advancing high order Discontinuous Galerkin (DG) based unstructured grid algorithms and the associated coding paradigms to improve computational efficiency more than two orders of magnitude as compared to current generation high fidelity codes.
- Developing high order adaptive meshes in space and time that evolve automatically with the flow and wave fields. This will further improve both efficiency and accuracy and will substantially reduce model development time and effort.
- Designing codes that interact with high resolution big data bases to automatically build meshes for the problems of interest. Thus the codes will query and mine topographic, bathymetric, LIDAR, land use, and GIS data bases; prior hydrodynamic studies in the region and areas of interest; and historical wind, wave, water elevation, and current data bases in order to automatically build initial meshes and evolve them with the hydrodynamics. This will result in significant reductions in manpower and lead time to develop models for sites of interest.
- Designing the code to operate efficiently on GPU based computer architectures.
- Incorporating phase resolving wave processes including run-up directly into circulation codes in order to better predict hydrodynamic forces on coastal infrastructure on deep ocean islands.
- Understanding resonant basin and shelf modes and shelf dissipation processes to better predict tides and storm surges on shelves and in seas such as the Caribbean and South China Sea.
- Better representing reef and mangrove systems in the models.
- Incorporating event based sediment erosion in storm models to understand the impact of storm induced channel deepening or dune destruction.
- Incorporating local rainfall and small scale channel routing capabilities into shallow water based codes to directly account for the large amounts of rainfall that can impact upland regions.