High-fidelity three-dimensional coastal compound flooding prediction system in support of disaster mitigation and safe navigation

PI: Pe’eri, Shachak (NOAA Office of Coast Survey)
Co-PI(s): Myers, Ed (NOAA Office of Coast Survey) : Moghimi, Saeed (NOAA Office of Coast Survey) : Zhang, Joseph (Virginia Institute of Marine Science) : Mani, Soroosh (collaborator, Spatial Front Inc) : Cui, Linlin (collaborator, Virginia Institute of Marine Science) : Wang, Zhengui (collaborator, Virginia Institute of Marine Science) : Lopez, Jesse (collaborator, Axiom) : Cockerill, Tim (collaborator, NSF Texas Advanced Computing Center)
Start Year: 2020 | Duration: 2 years
Partners: NOAA, Office of Naval Research, U.S. Geological Survey, Axiom, Virginia Institute of Marine Science

Project Abstract:

This multi-agency (NOAA, U.S. Geological Survey, and Office of Naval Research) project in collaboration with academia (Virginia Institute of Marine Science) and industry (Axiom) proposes to improve storm surge models using a fast response coastal ocean model engine. This work is a part of the NOAA effort to develop disaster risk assessments tools and practical technical applications to reduce and mitigate coastal vulnerability to natural disasters. These new products will improve communication and collaboration with other international agencies by understanding charting, bathymetry, water level, currents, and other products that NOAA produces. Such efforts will also support products and services that are vital for: 1) safe maritime navigation, 2) world class geodetic infrastructure, and 3) sustainable use of ocean resources for economic health and growth.

In particular within the context of this project the following goals are achieved:

  • Development of a system for coastal flooding forecast and hindcast with the following capabilities:
  • Fully automated: All the steps of preparing the model inputs and processing the outputs are done automatically in sequence without the need for human intervention.
  • End-to-end: The system starts from the basic inputs such as topography/bathymetry data as well as other NOAA operational products and includes components to process these inputs all the way to modeling and then generating visualization from the results of the model prediction.
  • Cloud based: Utilizes cloud technology to reduce the cost of running and maintaining computation infrastructure (only pay for when you use).
  • Hybrid: Has the ability to use multiple cloud service providers or a combination of cloud and existing on-premise computation infrastructure.
  • The ability to concurrently execute multiple simultaneous on-going storms or model parameterizations
  • The use of a graphical user interface for interaction and feedback on the state of the system and simulations
  • Improvements and integration of the components used within the system above:
  • Automatically setting up the coastal model to account for the effects of:
  • Inland hydrology: Ingestion of National Water Model discharges (all tributaries and rivers) to account for riverine flooding
  • Atmospheric forces (wind and precipitation):
  • Using combination of NOAA’s Global Forecast System and High-Resolution Rapid Refresh products
  • Hurricane wind modeling from basic inputs by using National Hurricane Center Tropical Cyclone data (parametric wind)
  • Automated modeling domain discretization (mesh generation)
  • Discretize the modeling domain such that regions impacted by the tropical cyclone have higher resolution, resulting in more accurate model predictions only for projected impacted region (optimize computational cost)
  • Utilizing the latest topography/bathymetry data updates from various sources
  • Automated processing of the model results and generation of visualization such as maps and water elevation graphs