Coastal flooding associated with hurricanes and other major storm events along the U.S. Coast results from complex interactions between freshwater flows, tides, storm surge, and wave effects. We have developed a two-way coupled model consisting of the National Water Model (NWM), the Advanced Circulation Ocean Model (ADCIRC), and WAVEWATCH III (WWIII) to quantify these interactions and compute total water levels in the coastal zone after significant riverine and coastal flooding events. This coupled continental coastal model covers the US Gulf and Atlantic Coasts, extending from the US-Canada border to the US-Mexico border. The Delft3D FM, D-Flow Flexible Mesh (D-Flow FM) model simulates coastal flooding on a 2D unstructured mesh within the National Water Model (NWM)/ADCIRC/WWIII coupled system. We developed a high quality 2D unstructured mesh using a sizing function that assigns element sizes based on proximities of coastal features at given spatial locations. Data sources used to identify relevant coastal features included NWM streamlines, the National Hydrography Dataset (NHD), and United States Army Corps of Engineers (USACE) data, allowing integration of D-Flow FM with the NWM and optimization of the number of computational points. The system obtains freshwater inflow boundary conditions to D-Flow FM from the NWM channel network. Offshore water levels boundary conditions for D-Flow FM come from the coupled ADCIRC-WWIII model. Domain sub-setting keeps runtimes within reasonable limits, as it does execution of the detailed hydrodynamic model within a user-defined area enclosing the storm landfall site. The advantage of this approach comes from the fact that the same coupled model setup allows simulation of coastal flooding for different storm events; only the sub-setting enclosure and the atmospheric forcing require updating from case to case. Model validation, consisting of water level comparisons against observations from simulations using the coupled system for historical storm events. The model simulations satisfactorily reproduced observed spatial and temporal variations of total water levels. In conclusion, this study presents performance of the sub-setting approach in reducing runtime considerably without compromising the accuracy of the coupled modeling system solution.

Fluctuations of the total water level in the U.S. East Coast depends on the complex interactions of freshwater flow, tide, storm surge and wave actions. In order to include all major forcings of water movement in this area, a coupled modeling system consisting of the National Water Model (NWM), the Advanced Circulation Ocean Model (ADCIRC), and the WAVEWATCH III model has been developed. In this system, a coupled inland hydrologic model is linked to an ocean hydrodynamic and wave model to compute total water levels in the coastal zones. In the freshwater component of the hydrodynamic model, 1D river components were included in the model to capture an accurate representation of tributaries to the 2D model of the estuary and oceans. The model domain included several states of the US East Coast starting from New Jersey to the St. Croix River at the US-Canada border. Model simulations were compared with 2012 superstorm Sandy measured tidal water levels and hurricane surge. Initial simulations reproduced satisfactory spatial and temporal variations of water levels due to riverine discharge and storm surge. The model predictions showed that using 1D component allowed better representations of the inland rivers and produced accurate river water levels. Simulations indicated that water levels in the inland areas depends on both river discharges and backwater effects of the ocean. These results showed the strengths of the coupled modeling system used in this research to compute total water levels during river flooding that coincides with extreme hurricane surge. Initial results showed that the coupled modeling framework used in this study is capable of total water estimation in the coastal zones and the accuracy of the water levels highly depends on the availability of reliable topographic, bathymetric, and bottom roughness data.

The current technology used by the Extratropical Surge and Tide Operational Forecast System (ESTOFS) on the East of the US and Gulf of Mexico coasts uses a sub-optimal unstructured grid, that over-resolves some straight portions of the coastline, under-resolves complex estuaries and coastal features, and employs roughly uniform resolution depending on the different water depths. The ESTOFS model is very efficient in terms of computational run time because it was designed for operational use, but accuracy is sub-optimal as the details of the complex inland water bodies is not captured with the 200 m minimum mesh resolution. ADCIRC is a robust high-fidelity depth-integrated model, widely used for the coastal community, including ESTOFS, for tides, storm surge, and wave-induced coastal setup. ADCIRC is a continuous-Galerkin based finite element unstructured grid framework that allows using meshes with a heterogeneous resolution to better represent the complexity of the ocean, shelf and nearshore regions. Recent advances on mesh generation tools now allow generating replicable high-resolution grids in times much shorter than the hand-edited processes used to develop the current version of ESTOFS. This opens the opportunity to study the effect of the different resolutions to represent topo-bathymetric and far inland water body features, in order to reduce the computational cost and improve the accuracy of the models. Thus, the objective of this research is to develop an ADCIRC-based model to accurately and efficiently simulate the dynamics of the ocean and riverine system in the Atlantic coast of the US and Gulf of Mexico for tide/storm predictions. The new ADCIRC-based model will incorporate a representation of the riverine system far up to the point where the ocean has no effect on water levels, efficiently use the resolution to reduce the minimum grid-size from 250 m to 50 m, with no significant increase in the number of nodes, and will combine pseudo-quadrilateral elements to efficiently represent narrow channels. This new generation of ESTOFS will represent a significant enhancement of the current technology for tides and storm surge prediction, but also will set up the required conditions for future approaches focused on coupling inland hydrology to the coastal modeling.

The mechanisms and geographic locations of tidal dissipation in barotropic tidal models is examined using a global, unstructured, finite element model. From simulated velocities and depths, the total dissipation within the global model is estimated. This study examines the effect that altering bathymetry can have on global tides. The Ronne ice shelf and Hudson Bay are identified as a highly sensitive region to bathymetric specification. We examine where dissipation occur and find that high boundary layer dissipation regions are very limited in geographic extent while internal tide dissipation regions are more distributed. By varying coefficients used in the parameterizations of both boundary layer and internal tide dissipation, regions that are highly sensitive to perturbations are identified. Particularly sensitive regions are used in a simple optimization technique to improve both global and local tidal results. Bottom friction coefficients are high in energetic flow regions, across the arctic ocean, and across deep ocean island chains such as the Aleutian and Ryuku Islands. Global errors of the best solution in the $M_2$ are 3.10 \si{cm} overall, 1.94 \si{cm} in areas deeper than 1000 \si{m}, and 7.74 \si{cm} in areas shallower than 1000 \si{m}. In addition to improvements in tidal amplitude, the total dissipation is estimated and compared to astronomical estimates. Greater understanding of the geographical distribution of regions which are sensitive to friction allows for a more efficient approach to optimizing tidal models.

We present a high-resolution continental-scale compound flood modeling system. It aims to quantify inland flooding resulting from the composite effects of riverine discharge and surface runoff and storm surge, in the inland-coastal zone during significant riverine and coastal storm events. This is achieved by coupling three continental models: the National Water Model (NWM) for the hydrology component, the Advanced Circulation Ocean Model for the coastal storm surge component, and the WAVEWATCH III model for the surface wave component with a detailed inland-coastal inundation model as the mediator between coastal and inland hydrology module. The inundation model, Delft3D FM, D-Flow Flexible Mesh (D-Flow FM), uses a high quality 2D unstructured grid with high-resolution (~100 m) near coastal features and lower-resolution in other areas to resolve the geometry of the study area. The coastal features are collected from NWM streamlines, National Hydrography Dataset, US medium shorelines and bathymetric features from the United States Army Corps of Engineers . The D-Flow FM model is forced by time-varying water levels and riverine discharges applied at its offshore and inland boundaries, respectively, by spatially- and time-varying wind and pressure fields and incorporates the contributions of surface and subsurface runoff to the total discharge in rivers, channels and streams. We conducted model validations for the following four major flooding events across the US coast: Hurricanes Ike (2008), Sandy (2012), Irma (2017), and Florence (2018). The results highlight the importance of including composite effects of compound flooding to accurately predict water levels during combined river flooding and extreme storm surge events.