Estuarine and Coastal Modeling (2009)

Over the last century, considerable near-shore tidal marshland as well as upriver freshwater habitat has been lost due to diking for agriculture, flood protection, and other development. While efforts to restore degraded coastal / estuarine habitats and re-establish migratory pathways for salmon, receive considerable attention and are supported through complex 3-D coastal circulation modeling, upstream restoration actions are sometimes neglected. The traditional approach of modeling upstream reaches has been to utilize one-dimensional hydraulic models developed for FEMA flooding analyses. While this approach may be sufficient for engineering design for peak flood protection, it is inadequate for most high flow events where flood waters often leave the channel and flow across the floodplain in a complex manner. This paper presents a hydrodynamic modeling analysis of the Skagit River and its floodplain to evaluate the effect of a proposed restoration project on flooding. It also shows the feasibility and advantages of using an unstructured multidimensional model for conducting such analyses. The site was located at approximately RM 28.5, within the lower Skagit River floodplain but upstream of tidal influence. The restoration proposal involves removal of a section of the existing dike as it is thought to be degrading habitat of the floodplain downstream by preventing flow into historic channels that could potentially provide refugia for salmon during migration. Previously a hydrodynamic model of the Skagit Basin including the Skagit Bay, estuary, and the flood plain was developed using the unstructured, three-dimensional model FVCOM. The approach used in the current study was to include the river channel and floodplain in the model domain, with inflow specified at the upstream end covering approximately 37 miles of river channel and floodplain. Because the study area is above tidal influence, measure water surface elevation data were used as the downstream boundary conditions. The results show the change in hydrodynamic conditions from the proposed project. A brief discussion is given on the potential of the modeling approach to provide the connectivity needed to properly analyze the Skagit River system from the North Cascades Mountains to the Skagit Bay for water quality, food web dynamics, and climate change.

The New York Harbor Observation and Prediction System, now in its 3^rd generation (NYHOPS v3), combines a network of real time sensors and a hydrodynamic forecasting computer model to assess prevailing ocean, environmental and meteorological conditions and to provide long and short term forecasts of the mentioned conditions. The older NYHOPS v2 model used spatially uniform surface heat flux forcing. Barometric pressure gradient forcing has also been neglected. The scope of this work was to assess sensitivity of the NYHOPS Sea Surface Temperature (SST) predictions to the spatial variability of the surface boundary condition. We compared two runs using different meteorological forcing: 1) Spatially varying wind stress and air pressure forcing, but spatially uniform heat flux forcing for the entire NYHOPS region (NYHOPS v2 surface boundary conditions with air pressure). 2) Spatially varying wind stress, air pressure, and heat flux forcing (NYHOPS v3 surface boundary conditions with air pressure). The SST modeled with NYHOPS was then compared against the validated GOES 12 satellite SST mapped on the NYHOPS grid nodes. A remarkable improvement (error reduction to the extent of 60%) in the prediction of SST by NYHOPS v3 was observed compared to NYHOPS v2. Similar error ranges were observed on comparison of NYHOPS v3 modeled SST and the satellite SST against in-situ observations, indicating that NYHOPS provides an effective SST prediction tool.

The 2D hydrodynamic numerical (HN) model Marina is introduced to simulate physical processes in the Delaware Bay. It was chosen because it is embedded in a complete development environment from pre- to post processing, and it includes a wave and sediment module, which will be necessary for future research. NOAA's Model Evaluation Environment (MEE) for the Delaware River and Bay was used to assess its performance and to compare it to other models. The evaluation is based on two scenarios, an astronomical tide only simulation to show the model's ability to predict tidal dynamics correctly, and a hindcast run during the 1984–1985 National Ocean Service (NOS) circulation survey (Klavans et al., 1986) to show the model's capability of reproducing realistic physical processes throughout the Delaware Bay. The comparison parameters include lime series plots, statistical metrics, and harmonic constituents for water levels and currents. Special attention was given to a comparison of the 2D Marina vs. the 3D FVCOM results with respect to salinity because our ultimate goal are long term simulations with respect to sediment exchange between the bay and the adjacent marshes. Hence, two-dimensional simulations would significantly reduce the computational effort. Overall the results showed good agreement of water levels and currents for both scenarios. In the upper estuary further calibration is still necessary to improve the performance. The quality of salinity results is insufficient in the turbidity zone, as expected due to vertical stratification, but in the well mixed upper estuary and weakly stratified Delaware Bay reasonably good performance was reached.

The Hurricane Protection Office (HPO) has requested that the Engineer Research and Development Center (ERDC) investigate the effects of proposed hurricane protection structures on the transport of larval fish recruitment into Lake Pontchartrain. Although the protection of human life is the priority, environmental agencies are concerned that these structures may adversely impact the fish population in Lake Pontchartrain. ERDC has modified passive particles within the Particle Tracking Model (PTM) with characteristic larval fish transport behaviors to study the effects of the hurricane protection structures on the transport of the representative particles into Lake Pontchartrain. PTM is a Lagrangian particle tracker designed to determine the fate of discrete constituents within hydrodynamic flowfields. Particles are initiated within specified areas of the Mississippi River Gulf Outlet (MRGO), the Gulf Intercoastal Waterway (GIWW), and Lake Borgne. Hydrodynamic solutions used to drive particle transport are developed by the Adaptive Hydraulics model (ADH). Analysis is performed comparing recruitment of representative particles into Lake Pontchartrain based on construction phases, initial locations, final recruitment paths, and characteristic larval fish transport behaviors. Results show that reduction of flow within the MRGO generates alternate preferred pathways for larval fish recruitment. In addition, characteristic larval fish behaviors affect transport into recruitment zones. This study is able to give insight into the effects generated by these proposed changes to the system and how these effects may impact the fish population.

To support the development of the Delaware River and Bay Operational Nowcast/Forecast System (DBOFS), high resolution hindcasts have been made using the Regional Ocean Modeling System (ROMS) community model from the head of tide at Trenton, NJ out to the continental shelf break. The construction of the high resolution grid (HRG) using the DELFT3D RGFGRID software package is first discussed and contrasted with the previous medium resolution grid (MRG). Results for two 15-day hindcasts in 1984 (one 27 March – 10 April during high flow and the other 10–24 September 1984 during low How) on both grids are compared with the National Ocean Service (NOS) Delaware River and Bay 1984 Circulation Survey measurements of water surface elevation, currents, salinity, and temperature. Total water level RMS errors range from 15 cm at Cape May, NJ to 37 cm at Trenton, NJ on both grids under high flow. RMS errors on the MRG (11–22 cm) are less than on the HRG (16–37 cm) under low flow, due to differences in the specification of the ocean boundary subtidal water level. Estuarine station current speed and direction RMS errors are generally reduced on the HRG by 5 cm/s and 10 degrees, respectively. Salinity (0.8–3.5 PSU) and temperature (0.4–3.1 °C) RMS errors are at most stations comparable on both grids. Total water level results for the HRG are evaluated for an extended seven-month, March – September 1984, hindcast and are comparable to the two 15-day hindcast results. The tidal water level error component is determined to range from 15–24 cm. To seek further reduction, several HRG tide simulations using different open boundary conditions are performed. Plans for further testing and development are then presented.

Hurricane storm surges cause coastal inundations and saltwater intrusions into land and wetland areas where freshwater dominates at normal times. The extent of the salty ocean water intrusion in time and space, over the normally low-salinity regimes, determines how much damage the surge does to the coastal ecosystem. With few exceptions, little effort has been made to understand the dynamics of the surge flushing out of hurricane-flooded coastal bays, particularly those in the northern Gulf of Mexico, known for its broad continental shelf and mild-slope near-shore region. This study focuses on the process during which the flood water is flushed out through both inundated land and tidal passages. For this purpose, an investigation of storm surge impact to the coast of Louisiana is conducted using the state-of-the-art finite-volume coastal ocean model FVCOM (Chen et al., 2003). We apply FVCOM to storm surges caused by Hurricane Rita of September 2005 and Hurricane Ike of September 2008 and analyze their inundation of coastal areas. Data from existing platforms along the coast are used in the validation of the model. These model results of storm surges allow an analysis of the effects of (1) the geometry of the water body affected, (2) the specific hurricane characteristics, and (3) the trajectory of the hurricane relative to the bay. Our case-studies in Sabine Lake. Calcasieu Lake and Vermilion Bay in southwest Louisiana demonstrate how the relative geometry of the bay/track is crucial in determining a storm's impact, and how the different dynamics of surge build-up versus receding stages produce distinct asymmetric in-and-out flushing patterns. Flushing rates are calculated and compared with the normal tidal conditions.

A three-level nested-grid coastal ocean modeling system was developed recently by Tang et al. (2009) for the Pearl River Estuary (PRE) in South China's Guangdong Province. The modeling system has three downscaling subcomponents: (a) a coarse-resolution outer model for China Seas from Bohai Sea to the northern South China Sea; (b) an intermediate-resolution middle model for coastal waters over the northern shelf of South China Sea; and (c) a fine-resolution inner model for the PRE and adjacent waters. The modeling system is forced by tides, meteorological forcing and buoyancy forcing associated with freshwater runoff from the Pearl River system. Multi-year model results in 1993–95 are used in this study in examining circulation and salinity distributions during the dry (December–March) and wet (May–August) seasons in the PRE. The seasonal mean circulation and salinity distributions in the dry season are affected significantly by freshwater runoff, wind and tides over the northern and western PRE and mainly by wind and tidal forcing over the outer PRE and adjacent inner shelf waters. In the wet season, the estuarine plume extends significantly offshore with river discharge and tides to be the main driving for circulation and salinity distributions inside the PRE, and tides and wind forcing to be the main driving forcing over the inner shelf waters off the PRE.

A recent model application has been developed by the Washington State Dept. of Ecology that simulates water circulation in South Puget Sound (SPS) using the Generalized Environmental Modeling System for Surface waters (GEMSS) with an open northern boundary near Edmonds, WA. Post-calibration root mean square errors (RMSEs) are <5% of Ehe tidal range in all but Oakland Bay where the error was 10% of the tidal range. The model simulates salinity with a RMSE of 0.6 psu near the surface and 0.5 psu near the seafloor compared with field observations at 22 key stations.

A three-dimensional, data-assimilative, regional ocean circulation model is used in simulating circulation, hydrography and associated variability in the Intra-Americas Sea (IAS). The model domain covers the region between 8°N and 32°N and 99°W and 54°W, with a horizontal resolution of 1/6°. The ocean circulation model is driven by 6 hourly wind fields produced by the National Centers for Environmental Prediction and boundary forcing extracted from 5-day reanalysis data produced by the British Atmospheric Data Centre. The model is integrated for 3 years from January 1999 to December 2001. The model performance is assessed by comparing model results with oceanographic observations made in the IAS during this period. Model results are used in the study of the "compensation effect" in which transport variations through the Yucatan Channel are partially compensated by flow through the Old Bahama and Northwest Providence Channels. The compensation effect is found to be associated with baroclinic (2-layer) flow through the Yucatan Channel at timescales longer than 20 days, while at shorter timescales (less than 20 days) the vertical structure of the flow is barotropic.

This study employs a model to examine, through numerical experiments, the potential for material transport across the channel of the upper, non-saline tidal Delaware River in the vicinity of the City of Philadelphia. The goal is to explore possible dynamics of tidally-driven transverse transport, and the potential implications for two drinking water intakes of pollutant sources such as the discharge of a local stream. The model formulation includes astronomical tides and river inflows driving the depth-integrated shallow water equations that include energy losses to bottom friction, and Reynolds stresses and fluxes represented with a Smagorinsky turbulence model. Cross-river transport is explored by performing a tracer simulation. Model simulation results support the potential for solute transport across the channel of the upper tidal-Delaware River solely as a result of tidal interaction with local bathymetry.

A three-dimensional hydrodynamic model using unstructured Cartesian grids has been developed for an estuarine system called Crystal River/Kings Bay in southwest Florida. The 3D model uses a cut-cell approach to fit the bottom topography and shorelines and solves flux-based finite difference equations using an efficient semi-implicit method. At the same lime, the model has the ability of discretizing complex geometries with Cartesian grids that can be liberally downsized in the two horizontal directions. With the help of a geophysical information system, the grid generation process is quite straightforward and does not suffer the grid generation headache often seen in other unstructured models, because the unstructured Cartesian model does not require grids to be orthogonal. This allows the arrangement of model grids to be very flexible, as any of the Cartesian grids can be split into two without affecting its surrounding grids. The newly developed model was validated with a couple of test cases in rectangular basins, before it was qualitatively tested for the Crystal River/Kings Bay system. Model tests show that the unstructured Cartesian grid model works well in these shallow water cases.

Habitat in open water is not simply a 2-D to 2.5-D surface such as the ocean bottom or the air-water interface. Rather, pelagic habitat is a 3-D volume of water that can change over time, leading us to the term habitat space. Visualization and analysis in 2-D is well supported with GIS tools, but a new tool was needed for visualization and analysis in four dimensions. Observational data (cruise profiles (x_o, y_o, z, t_o)), numerical circulation model fields (x,y,z,t), and trajectories (larval fish, 4-D line) need to be merged together in a meaningful way for visualization and analysis. As a first step toward this new framework, UNIDATA's Integrated Data Viewer (IDV) has been used to create a set of tools for habitat analysis in 4-D. IDV was designed for 3-D+time geospatial data in the meteorological community. NetCDF Java^TM libraries allow the tool to read many file formats including remotely located data (e.g. data available via OPeNDAP ). With this project, IDV has been adapted for use in delineating habitat space for multiple fish species in the ocean. The ability to define and visualize boundaries of a water mass, which meets specific biologically relevant criteria (e.g., volume, connectedness, and inter-annual variability) based on model results and observational data, will allow managers to investigate the survival of individual year classes of commercially important fisheries. Better understanding of the survival of these year classes will lead to improved forecasting of fisheries recruitment.

To meet the challenge of developing a more accessible educational content delivery system for the Southeastern Universities Research Association's (SURA) Coastal Ocean Observing and Prediction Program (SCOOP), a SCOOP Educational Virtual Appliance (EVA) is being developed. The EVA is based on the Grid Appliance (http://www.grid-appliance.org) which is a self-configuring Virtual Machine (VM) appliance that contains all necessary cyberinfrastructure, applications and tools to demonstrate a variety of SCOOP services in a single easy-to-use container. Applications and tools contained within the EVA include examples of how to find, download, and plot SCOOP data as well as the ability to perform simulations of storm surge and inundation. The storm surge application includes the ability to setup, perform and visualize the simulation of storm surge and inundation using CH3D-SSMS in several hypothetical domains as well in Charlotte Harbor, FL during the landfall of Hurricane Charley (2004). In addition to the EVA, a companion web-based tutorial is also under development. Lessons plans cover many aspects of the SCOOP program including: file naming conventions; how to find, download and plot archived data; how to use SCOOP web services; how to use the EVA; and an introduction to hurricane storm surge. The EVA is currently being used in several educational and research settings and is publically available on the Internet (http://www.grid-appliance.org/scoopwiki/index.php/Main_Page).

Model data interoperability for the United States Integrated Ocean Observing System (IOOS®) was initiated with a focused one year project. The problem was that there were many regional and national providers of oceanographic model data; each had unique file conventions, distribution techniques and analysis tools that made it difficult to compare model results and observational data. To solve this problem, a distributed system was built utilizing a customized middleware layer and a common data model. This allowed each model data provider to keep their existing model and data files unchanged, yet deliver model data via web services in a common form. With standards-based applications that used these web services, end users then had a common way to access data from any of the models. These applications included: (1) a 2D mapping and animation using a web browser application, (2) an advanced 3D visualization and animation using a desktop application, and (3) a toolkit for a common scientific analysis environment (Matlab®). Due to the flexibility and low impact of the approach on providers, rapid progress was made. The system was implemented in all eleven US IOOS regions and at the NOAA National Coastal Data Development Center, allowing common delivery of regional and national oceanographic model forecast and archived results that cover all US waters. The system, based heavily on software technology from the NSF-sponsored Unidata Program Center, is applicable to any structured gridded data, not just oceanographic model data. There is a clear pathway to expand the system to include unstructured grid (e.g. triangular grid) data.

It is well know that standard Galerkin finite element methods are prone to numerical instabilities for advection-dominated flows and a variety of methods have been proposed to stabilize the advection terms. One large class is upstream weighting; herein, three variants of that method are examined. The first, which is unique to the ADvanced CIRCulation (ADCIRC) model, looks at the possibility of introducing an upstream weighting coefficient to the elemental fluid velocities in the non-conservative form of the advection terms in the generalized wave continuity, momentum and transport equations. Results from truncation error analysis show consistent convergence, as well as the presence of a numerical diffusion term. However, this term is not sufficient to increase the stability of the model. The second method is flux corrected transport (FCT). Herein, we examine this method qualitatively by comparing results of the FCT implementation within the Quoddy model to spatially and temporally refined ADCIRC model runs using the same parameter values. Results indicate that with enough resolution ADCIRC can match these FCT results; however this is not practical for real world applications. Finally, we examine the Streamline upwind Petrov Galerkin (SuPG) method applied to the transport equation.

To accurately capture the fluid dynamics present in shallow straits and near-coastal zones, development efforts have centered around enhancing the 3D baroclinic capabilities of the coastal model, ADCIRC. The scale of resolution necessary to resolve shallow straits and coastal zones is not practical with structured global/regional models, such as HYCOM. However, ADCIRC utilizes an unstructured grid, which can capture complicated topography and coastal fluid dynamics. Typically, the barotropic ADCIRC domain encompasses large portions of the ocean; however, in baroclinic simulations, we restrict the domain to the region of interest due to computational demands. Consequentially, the open boundary is placed in an area dominated by complex, nonlinear processes, so proper specification of the open boundary conditions is difficult. Hence, one solution is coupling the offshore conditions provided from HYCOM to the near-coastal model, ADCIRC. This paper examines the coupling protocol needed to interface ADCIRC and HYCOM and provides preliminary results.

The US Army Engineer Alaska District is currently assessing hydrodynamic and port sedimentation issues associated with proposed expansion and deepening of Port of Anchorage, Alaska. The Port of Anchorage (POA) is located in lower Knik Arm, a macrotidal subestuary of Cook Inlet, characterized by large, seasonally variant suspended sediment loads; strong tidal currents; dynamic eddy structure associated with headland points and shoals, and expansive intertidal mudflats. Given the energetic conditions, large sediment load, and proximity to eddies, POA has recently experienced high sedimentation rates. Sedimentation rates at POA vary on seasonal and interannual to decadal scales. The interannual variability and remote proximity of POA to dredging resources make planning and scheduling of dredging resources complex and crucial towards successful execution of the USACE navigation mission. To anticipate sedimentation rates under proposed expansion and deepening of the port and to better accomplish the USACE navigation mission, numerical and physical modeling studies were performed.

The US Army Engineer Alaska District is currently assessing hydrodynamic and port sedimentation issues associated with proposed expansion and deepening of Port of Anchorage. Port of Anchorage is located in lower Knik Arm, a macrotidal subestuary of Cook Inlet, Alaska characterized by large, seasonally variant suspended sediment loads; strong tidal currents; and dynamic eddy structure associated with headland points and shoals. Two field experiments were conducted to support numerical hydrodynamics and sediment transport modeling, quantifying physical processes from subgrid to estuary scales. Measurements include bottom- and vessel-mounted Acoustic Doppler Current Profilers (ADCPs), suspended sediment settling velocities, and cohesive sediment erosion. Analysis of tidal volume exchange indicates a strong influence of upper Knik Arm intertidal mudflats on currents at the study site. Prior to the field experiments, eddy shedding by headland points within lower Knik Arm was known, but poorly quantified. The field experiments and data analysis quantified time-dependent development and growth of these eddies. Analysis of sediment data indicates weak cohesiveness in the physically weathered sediments of Knik Arm and pronounced tidal, seasonal, and decadal variation in suspended sediment concentration. The findings from field experiments and data analysis were found to be crucial in representing the dynamic physical processes at the study site with numerical models.

The US Army Engineer Alaska District is currently assessing hydrodynamic and port sedimentation issues associated with proposed expansion and deepening of Port of Anchorage. Port of Anchorage is located in lower Knik Arm of Cook inlet, Alaska. A series of numerical hydrodynamic simulations were performed to assess the impact of the proposed expansion configuration on circulation within the port. Critical to the understanding of the hydrodynamic transport processes within and around the port, is the formation, strength and evolutions of eddies formed during all phases of the tide. Initial comparisons of model predictions with field measures suggested the need to improve upon an existing grid and the lateral eddy viscosity formulations. Subsequent to these improvements, good agreement between the formation of the eddies and field measurements was achieved.

The US Army Engineer Alaska District is currently assessing hydrodynamic and port sedimentation issues associated with proposed phased expansion and deepening of the Port of Anchorage. A series of numerical model simulations were performed to assess the impact of the proposed expansion configurations on sedimentation rates within the port. The EFDC surface water modeling system was used to perform these simulations. The hydrodynamic model in EFDC was driven by ADCIRC simulated tides at the forelands near Nikiski, AK in Upper Cook Inlet. The sediment transport model was calibrated using a two-month average sedimentation rate in the port calculated from dredging records. EFDC was applied to represent six different expansion configurations of the port in order to assess the impact of the phased port expansion and deepening on the sedimentation rate inside the port. The ratios of dredging volumes, calculated as the ratio of the predicted sedimentation volume for each of the expansion configurations to the base configuration sedimentation volume, for the six expansion configurations vary from 3.2 to 5.2. These increases in dredging volumes are mainly due to the larger area to be dredged for the expansion configurations than for the base port configuration. Variations in dredging volume ratios among the six expansion configurations are attributed to altered gyre circulation and therefore current speeds along the dock face in the port due to 1) the phased closure of the North and South replacement areas, and 2) the deepening of the bathymetry immediately offshore of the port in the 2009 and 2013 port configurations.

An experimental study and modeling analysis was performed to determine the relative importance of physical transport versus swimming behavior in determining the distribution of zooplankton and their retention in patches of high phytoplankton density. The experiment involved establishing, maintaining and sampling a natural plankton patch in a ∼1-km long tank (3.3 m deep by 7.3 m wide). Natural plankton containing a mixture of Chesapeake Bay algal species dominated by the calanoid copepod Acartia hudsonica were inoculated into the lighted center of the tank within a plastic bag such that the plankton patch was held in place. The patch was grown (via lighting and nutrient additions) to a relatively high density as compared to the surrounding water. After balancing the water density throughout the tank, the patch was released by removing the plastic bag. The physical regime was such that the patch was sheared. Copepod distributions were sampled and modeled, including physical transport and various swimming behaviors, to determine if the observed movements could be accounted for purely by physical transport, or whether the animals needed to swim either vertically or horizontally to account for the changing densilies in space and lime. Modeling results indicated that younger copepod stages moved purely by physical transport and randomly-directed swimming, but that there was evidence of directed movement by older copepods. Vertical migration behavior, coupled with physical transport and randomized foraging behavior, was sufficient to explain the distribution of older stage copepods; oriented horizontal swimming behaviors were not needed to account for distributions. Such directed movements, coupled with physical transport, are important determinants of plankton distributions, both for holo-zooplankton and larval fishes. Additional species and conditions need to be evaluated to fully understand the roles of directed swimming behaviors and physical transport in determining plankton distributions.

To support marine ecological resource management and emergency response and to enhance scientific understanding of physical and biogeochemical processes in Puget Sound, a real-time Puget Sound Operational Forecast System (PS-OFS) was developed by the Coastal Ocean Dynamics & Ecosystem Modeling group (CODEM) of Pacific Northwest National Laboratory (PNNL). PS-OFS employs the state-of-the-art three-dimensional coastal ocean model and closely follows the standards and procedures established by National Oceanic and Atmospheric Administration (NOAA) National Ocean Service (NOS). PS-OFS consists of four key components supporting the Puget Sound Circulation and Transport Model (PS-CTM): data acquisition, model execution and product archive, model skill assessment, and model results dissemination. This paper provides an overview of PS-OFS and its ability to provide vital real-time oceanographic information to the Puget Sound community. PS-OFS supports pacific northwest region's growing need for a predictive tool to assist water quality management, fish stock recovery efforts, maritime emergency response, nearshore land-use planning, and the challenge of climate change and sea level rise impacts. The structure of PS-OFS and examples of the system inputs and outputs, forecast results are presented in details.

Recent and frequent fish-kills in waters otherwise known for their pristine high quality, created increased awareness and urgent concern regarding potential for degradation of water quality in Puget Sound through coastal eutrophication caused by increased nutrient loading. Following a detailed review of leading models and tools available in public domain, FVCOM and CE-QUAL-ICM models were selected to conduct hydrodynamic and water quality simulations for the fjordal waters of Puget Sound. A high resolution hydrodynamic model of Puget Sound has already been developed and tested. The objective of this effort was to develop an associated sophisticated water quality model of Puget Sound to help nearshore restoration managers and planners in the management of nutrient loading to Puget Sound. In this study we present a one-way coupling of FVCOM hydrodynamic solution to CE-QUAL-ICM water quality kinetics. The coupled model uses FVCOM model grid and transport solution scheme to conduct water quality kinetic calculations in CE-QUAL-ICM. Through this approach, it is possible to construct the linked model with sufficient resolution in the nearshore region to address dynamics of water quality state variables in the complex coastline, multiple tidal channels, tidal flats, and a deep fjord such as Puget Sound. In this paper, we demonstrate the scalar transport and water quality module tests (a propagated pulse, diffused plumes, a BOD/oxygen sag, and phytoplankton/nutrients reactions) and compare model results with their analytical solutions as part of model validation.

The Craney Island Dredged Material Management Area (CIDMMA) is a 2400- acre, federally-owned and operated facility located in Hampton Roads, Virginia adjacent to the city of Portsmouth. A 520-acre designed expansion of the CIDMMA to the east was shown in a previous study to have the least impact to the hydrodynamic circulation of all evaluated land expansions. The 3D baroclinic hydrodynamic model HEM-3D coupled with the water quality model was applied over a domain spanning the James and Elizabeth Rivers. The model was calibrated, validated, and used to assess the long-term environmental impact of the terminal expansion, as well as the impact resulting from an intermediate construction phase. The study showed that a slight reduction in tidal prism was compensated for by an increase in the non-tidal residual flow during flood and ebb tide phases, thereby enhancing flushing. Dissolved oxygen (DO) levels due to the eastward expansion were mitigated by the enhanced exchange of Elizabeth River water with that from the oxygen-rich Lower James River. As a result, the increase of advective DO flux from the James River and local vertical mixing overcome the increase of low DO volumes. Overall, the impacts to DO levels due to both the south cell and the full expansion are minimal and are well within the range of detection limits.

Water hyacinth (Eichhornia crassipes) poses numerous water quality problems in many surface water bodies in tropical and sub-tropical regions. In this paper, a three-dimensional coupled hydrodynamic model in the Environmental Fluid Dynamics Code (EFDC) modeling framework and a water quality model using the Water Quality Analysis Simulation Program (WASP7.2) were developed for Lake Nokoué, Benin for the period of September 01, 2006 to December 31, 2007. A third predictive model that simulates the suitability of salinity, nutrient and temperature conditions in the lake for water hyacinth growth was developed. Results from this study showed that during the dry season no water hyacinth infestation was expected in the lake since the salinity was above the lethal level whereas in the rainy season, the environmental conditions in many parts of the lake were suitable for the weed's growth except in the Cotonou area where salinity remains limiting almost throughout the year. Investigation of the lake's response to removal of nutrients from the different sources showed that a maximum reduction of 17% in themulti-parameter suitability index was achieved when 66% of the nutrient loads from the Ouemé River and flooded marshes were removed.

Cone Bay is located in the Kimberly region of Western Australia. It is an area of fast flushing rates and dynamic circulatory patterns due to a 9 to 11 m tide, the second largest tide in the world after the Bay of Fundy in Canada. Marine Produce Australia (MPA) currently operates at a maximum production of 150 tonnes of arramundi per year within Cone Bay and was seeking a licence to increase the production into a commercialised venture of 1,000 tonnes per year. As part of the environmental assessment proposal, a staged desktop modelling study was carried out to examine the potential changes in circulation following the presence of sea cages and the potential for deposition and accumulation of fish waste and fish food. A side-by-side comparison of the circulation patterns indicated that for the existing environment currents within the operational area would be reduced to by the inclusion of the sea cages, but no changes in current directions were indicated. A numerical dye-study showed that Hushing times for the modified setting took approximately one day longer, due to retardation of the tidal flows. The fate of fish waste and food pellets was simulated at various sites of the existing operational area, and there was a tendency for the waste material to settle close to the release sites and along the east-west tidal axis. Based on these results, MPA had decided to position the sea cages a set distance from the boundaries to ensure that the potential effects are contained and that the environmental integrity of the area is maintained.

Watershed and estuarine modeling were conducted in the Myakka River basin to assess the effects of watershed management plans on the salinity characteristics of the estuarine portion of the lower Myakka River. Analysis of long-term data indicate that an increase in agricultural land and water use has contributed to increasing streamflow in the upper river sub-basin which has resulted in extensive tree mortality in freshwater riparian wetlands. In order to address these adverse ecological impacts, an integrated surface water/ground water model (MIKE SHE/MIKE 11) was developed to simulate continuous streamflow in the upper river sub-basin as a function of climate and watershed characteristics. Simulations of streamflow under historic (circa 1950) and existing watershed conditions allowed for the estimation of excess flows the river currently receives due to an increase in agricultural activities and other human alterations of the river's watershed. A dynamically coupled 3D-2DV hydrodynamic model of the estuarine reach of the river was used to simulate the effects of reducing these excess flows on the salinity characteristics of the lower river. Using the existing inflow regime of the river as the baseline, reductions in the area and volume of biologically important low salinity zones were calculated to assess potential ecological changes that could result from reducing the excess flows.

Naturally occurring sites with sufficient kinetic energy suitable for tidal power generation with sustained currents > 1 to 2 m/s are relatively rare. Yet sites with greater than 3 to 4 m of tidal range are relatively common around the U.S. coastline. Tidal potential does exist along the shoreline but is mostly distributed, and requires an approach which allows trapping and collection to also be conducted in a distributed manner. In this paper we examine the feasibility of generating sustainable tidal power using multiple nearshore tidal energy collection units and present the Modular Tidal Prism (MTP) basin concept. The proposed approach utilizes available tidal potential by conversion into tidal kinetic energy through cyclic expansion and drainage from shallow modular manufactured overland tidal prisms. A preliminary design and configuration of the modular tidal prism basin including inlet channel configuration and basin dimensions was developed. The unique design was shown to sustain momentum in the penstocks during flooding as well as ebbing tidal cycles. The unstructured-grid finite volume coastal ocean model (FVCOM) was used to subject the proposed design to a number of sensitivity tests and to optimize the size, shape and configuration of MTP basin for peak power generation capacity. The results show that an artificial modular basin with a reasonable footprint (approx. 300 acres) has the potential to generate 10 to 20 kw average energy through the operation of a small turbine located near the basin outlet. The potential of generating a total of 500 kw to 1 MW of power through a 20 to 40 MTP basin tidal power farms distributed along the coastline of Puget Sound, Washington, is explored.

The three-dimensional, coastal circulation numerical model COCIRM-SED was recently adapted and optimized to simulate and predict the flows in southern Discovery Passage and Canoe Pass, BC, Canada. These model results provided reliable, detailed flow information for use in assessing potential locations of installing and operating underwater tidal current turbines. In the application of modeling flows in southern Discovery Passage, the model involved tidal forcing with 69 constituents at open boundaries as well as the effects of the Campbell River freshwater input and the Coriolis. Detailed hourly model flows for typical neap and spring tides were extracted for use in the site assessment of potential locations for tidal current turbines. In the other application of modeling flows in Canoe Pass, the model was used to predict the water flows and water levels through Canoe Pass if the dam in Canoe Pass, which has been in place since the 1940's, were completely removed and replaced by a passage of 40 m wide between Quadra and Maude Islands. In both studies, the model went through extensive calibration and verification processes using available measurements of water levels and ocean currents at various sites in the modeling areas. It was demonstrated that the 3D model has very good capabilities for simulating water level and currents in both model areas under different conditions.

A tide model is being developed for the Northwestern Gulf of Mexico in support of VDatum, the National Oceanic and Atmospheric Administration's (NOAA) vertical datum transformation tool. The model domain extends from south of the US/Mexico border in the west to Atchafalaya Bay, Louisiana in the east. A high-resolution unstructured grid was created to represent the complex shoreline and fine bathymetric features. Efforts have been made to include most of the Intracoastal Waterway and to resolve the major navigational channels and barrier islands. The grid sizes vary from about 40 m along the channels/islands to 300 m along the coasts to 30 km at the open ocean boundary. The two-dimensional version of the ADvanced CIRCulation model is employed to simulate barotropic tides and depth-integrated tidal currents. Tidal datums, including mean lower low water, mean low water, mean high water and mean higher high water, as well as tidal harmonics, are derived from the simulated tidal water levels and validated against available observations at various tidal gauges obtained from NOAA's Center for Operational Oceanographic Products and Services and the Texas Coastal Ocean Observation Network. Tides in the area are mostly diurnal. K₁ and O₁ are of similar magnitudes and are the two dominant tidal constituents in the open coast. Both K₁ and O₁ amplitudes change only slightly in the coastal waters. However, M₂ amplitudes have a great spatial variability with magnitude comparable to the diurnal components over the northern Louisiana-Texas shelf. The results of sensitivity studies on the boundary conditions, bottom friction formulation, and other parameters are also described.

NOAA's National Ocean Service (NOS) has developed the VDatum software to allow users the ability to easily transform vertical elevation data among tidal, orthometric, and ellipsoid-based 3D reference systems, VDatum applications are developed on a regional basis with the goal of attaining seamless coverage for all U.S. near-coastal waters. Development of VDatum is coordinated among three NOS offices: the Coast Survey Development Laboratory (CSDL), the National Geodetic Survey (NOS), and the Center for Operational Oceanographic Products and Services (CO-OPS). As part of this collaborai ion, CSDL develops the tidal datum fields used as input to the VDatum software. This paper describes the standard modeling procedures CSDL uses to develop the tidal datum fields, how these are built into the VDatum framework of vertical datum transformations, and the community modeling framework that is envisioned for VDatum-derived models and products.

The present study simulates barotropic tides and currents in the Florida (FL) coastal waters and the Straits of Florida (SFL) using a high-resolution, two-dimensional version of the Advanced Circulation model (ADCIRC -2DDI). The model domain spans from eastern Gulf of Mexico (GOM) across the SFL to the entire Southern Atlantic Bight (SAB). The finite-element model grid consists of 353,718 nodes and 622,367 triangular elements, with a resolution ranging from 16 m to 41 km. A finely resolved spatial model resolution is an important factor to ensure realistic reproductions of tidal and current fields. The fields of tidal harmonics, current ellipses, tidal datums, energy fluxes, and the bottom friction dissipation rates are derived from the modeled water level and current time series. The results indicate favorable model-data agreement. In addition to reproducing tidal and current fields, this study focuses on examining the tidal dynamics and energetics over the SFL and its adjacent waters. This study reveals an amphidromic point of the principle solar constituent (K₁) centered around the Bimini Island. It is located offshore the southeast Florid Shelf; its formation is attributed to superposition of tides propagating in opposite directions. This study identifies that energy in the SFL originates from the SAB. The energy flux is attributed predominantly to the principle lunar semidiurnal constituent (M₂). It propagates into the Straits through three pathways: along the east FL shelf (EFS); a deep channel between islands in the northern Bahamas; and the northeast coast of Cuba. Across the Straits the energy flux progresses from east to west. Once into the eastern GOM, it turns rapidly to a northward direction, then finally turning eastward onto the west FL shelf (WFS). The shallow, nearshore regions display the strongest energy dissipation rate of the entire domain. The areas of less than 10 m water depth dissipates about 5 GW energy, accounting for over 75 percent of the total tidal energy loss over the entire model domain. Of the rest 25 percent energy loss, waters between 10–50 m depths make 24 percent and even deeper waters cause the remaining 1 percent.

In order to meet the Australian State and Federal Government Environmental regulations, any consequences of industrial activities disturbing marine sediments (such as dredging, drilling, etc.) have to be thoroughly evaluated and presented to the satisfaction of the regulators. An evaluation normally consists of two interrelated components, field measurements and numerical model investigations. Asia Pacific Applied Science Associates (APASA) developed an integrated approach to such evaluations, which satisfy the regulating entities' requirements. The approach was successfully validated against field measurements and implemented in Australia. The proposed modelling approach with validation results is the subject of this contribution.

The Magdalena River is the second longest river in South America and is home to the Port of Barranquilla, one of the largest Colombian Ports. A numerical model including hydrodynamics, sediment transport, and morphological updating was constructed using Delft3D. This was implemented in order to design river training structures that would deepen the navigation channel to the port, allowing it to accommodate larger ships and become more competitive in the region. The existing model was enhanced and re-validated to conditions during the project construction. It was then employed to re-evaluate and modify the structural design during construction of the system. It was found that the response of the system to the construction was slightly different than was originally expected and the design was modified to incorporate this response. This adaptive modeling approach can be useful in determining the most effective design for projects with a phased construction.

Phase-resolving wave models are best suited for the class of coastal applications with strong wave diffraction and reflection caused by coastal structures. Often, a deepwater spectral wave transformation model has to be coupled to a phase-resolving wave model to ensure accurate modeling of the local wave field near structures. This paper presents the development and evaluation of simplified methods for coupling a phase-resolving model CGWAVE with a phase-averaged model CMS-Wave. Four methods, which are, single point input, geometric mean, spatially varying wave height input, and quasi-spectral wave input, were tested at the Grays Harbor/Half Moon Bay area, a very dynamic site with complex geometry and bathymetry. Comparison of results with field data indicate that monochromatic CGWAVE solutions with a single-point wave input may not be reliable in certain applications since this method ignores the spatial variation and spectral properties of the incident waves at the open boundary. Results also indicate that directional wave spreading in the coupling of two wave models is important even for swell-like wave conditions. By combining a finite number of monochromatic CGWAVE solutions using the proposed quasi-spectral methods, the wave height estimates were improved in the Half Moon Bay as compared to data.

Hydrodynamics and water renewal of intermittent coastal streams are highly variable, at various time scales, due to the very active morphodynamic behavior of their inlets. Due to this strong dynamics, the pathways of water-borne materials — and the consequences of contaminated discharges — can depend strongly on the morphology and environmental conditions. Predicting the fate of contaminants in these systems requires coupled numerical models accounting for the major physical and water quality processes. We aim at improving the understanding of the impact of inlet morphology and wave action on the pollutant and sediment pathways of these small coastal systems, based on a suite of calibrated and validated coupled models. Two analyses, based on particle simulations, are presented to assess sediment dynamics and pollutant pathways for several conditions. Results show that waves have a major effect on the fate of water-borne materials in the estuary. Wave-induced currents sweep away materials coming out of the estuary, while wave-induced setup has a profound effect on tidal propagation, water levels and velocities in the estuary, promoting the upstream transport of pollutants.

The National Ocean Service presently has an Operational Forecast System (CBOFS) for the Chesapeake Bay which generates only water levels and depth-integrated currents. As a next generation system, a fully three-dimensional, baroclinic Forecast System (CBOFS2) was developed, calibrated and validated; this system will produce water levels, currents, temperature and salinity. First, a two-month tides only simulation was conducted to validate the water levels and currents and thereafter, a synoptic hindcast simulation from June 01, 2003–September 01, 2005 was conducted to validate water levels, currents, temperature and salinity. Upon comparison with observations, CBOFS2 for the most part met the target NOS water level error criteria and for current error, the criteria were met exceptionally well; the temperature and salinity errors were frequently less than 1 C and 3 PSU respectively. Hence, the predictive accuracy of CBOFS2 warranted it being accepted as a suitable three-dimensional upgrade to CBOFS.

The NOAA National Ocean Service Coast Survey Development Laboratory is partnering with NOAA's National Weather Service National Centers for Environmental Prediction (NCEP) to investigate an integrated prediction system of weather, storm surge, and inundation. These initial investigations are aimed for routine operations at NOAA. For that purpose, NCEP operational weather-ocean forecast systems are interfaced to a high resolution Coastal Flood Model (CFM). The model used to develop a high resolution CFM is the ADvanced CIRCulation (ADCIRC) finite element model. The ADCIRC hydrodynamic model has several features beneficial for this application and has been demonstrated to be effective at predicting water levels in complex coastal systems. Its unstructured grid methodology can not only propagate the hurricane-generated storm surge wave from offshore, across the shelf, and inland, but also can readily and accurately represent irregular shoreline and coastal topographic features, including barrier islands, rivers and waterways. In the approach illustrated here, a high resolution CFM of the North Carolina sound system is used to examine the impact of meteorological forcing from a suite of NOAA models. The meteorological forcings (e.g. the hurricane wind and pressure fields) being tested include the H*Wind hurricane wind analysis system, the Geophysical Fluid Dynamics Laboratory (GFDL) Hurricane Model, and the Hurricane Weather Research and Forecasting (HWRF) Model. The test cases are focused on simulating Hurricane Isabel (2003).

We briefly describe the new NYHOPS v3 OFS (Operational hydrodynamic Forecast System) and quantify its performance against National Ocean Service (NOS) standard OFS evaluation metrics. Given the relatively large area of the NYHOPS v3 OFS (including the NY/NJ Harbor estuary, Long Island Sound, and their coastal ocean), and the proliferation of sensor networks, the presented skill assessment is one of the most extensive performed to date: model results are compared to in situ observations of water level, currents, temperature, salinity, and waves from over 100 locations, collected in a 2 year period. The model's ability io describe the hydrodynamic conditions in the extensive area it is employed is remarkable. The average index of agreement for water level is 0.98, for currents is 0.87, for water temperature is 0.98, for salinity is 0.77, and for significant wave heights is 0.88. Respective, average root-mean-square errors are: 10cm for water level, 13cm/s and 9° for currents, 1.4°C for water temperatures, 2.8psu for salinities, and 32cm for significant wave heights.

A Tampa Bay operational forecast system is under development using the Regional Ocean Modeling System community model to provide short-term forecast guidance of water levels, current velocities, salinity and temperatures. The model domain includes Tampa Bay, the intra-coastal waterway, inner portions of the west Florida continental shelf, and Sarasota Bay. The model system was calibrated with two scenario model simulations: a tidal simulation and a synoptic hindcast running from June 1990 to September 1991. The tidal simulation was forced with tide constituents from ADCIRC EastCoast 2001 (Advanced Circulation Model) and did not include river inputs and atmospheric forcing, and water temperature and salinity stayed constant. The synoptic hindcast was initialized with temperature and salinity conditions generated from monthly measurements from the Environmental Protection Commission of Hillsborough County, and forced by surface atmospheric forcing generated from the North America Regional Reanalysis, daily river discharges from USGS measurements, and the same lateral open boundary conditions as the tidal simulation and observed non-tidal water levels. The simulated model results of water levels, current velocities, salinity and temperatures were compared with a comprehensive observed dataset collected during the National Ocean Service's Tampa Bay Oceanographic Project (TOP) conducted between June 1990 and September 1991. This comparison was made using the NOS standard skill assessment software for operational forecast models. The TOP data set includes ADCP current meter data at 40 fixed stations, water levels at 16 stations, time series of salinity and temperature data at 36 fixed stations mostly co-located with current meters, and salinity and temperature profiles (CTD) along six transects.

Forecast and hindcast simulations were carried out using the ADCIRC storm surge prediction model with two adjacent high-resolution grid domains to assess the impact of Hurricane Ike in Louisiana and Texas. The model was forced by winds generated with an asymmetric parametric wind model that is coupled to the ocean model at every node and every time step. The wind model uses the storm parameters from the National Hurricane Center's best track and forecast advisories to generate the wind and atmospheric pressure fields that drive the storm surge model. A set of 20 forecast advisories combined with best track information, starting 5 days before the predicted landfall, were used to examine the variability in the predicted storm surge resulting from changes in the storm parameters over the course of the advisories. The simulations were grouped into 5-, 4-, 3-, 2-, 1-day ensembles to assess the predictive skill of the forecast and examine the evolution of the forecast convergence. The closer the advisories are to the time of landfall, the lower the range of variability in the surge predictions. The forecasts began to converge with the 2-day ensemble forecast. The RMS error in maximum elevation between each advisory and the best track at each node for each simulation shows that the error decreases substantially over Louisiana and Texas in the last 8 forecast advisories encompassing the last 2 days of the storm. Validations of the hindcast results against NOAA and temporary USGS stations were performed in both Louisiana and Texas to assess absolute model performance.

The hydrodynamic model ADCIRC has been used extensively to model hurricane storm surge inundation. ADCIRC utilizes an unstructured triangular mesh that allows sufficient resolution of important features and has been coupled to STWAVE or SWAN to account for wind waves. However, river inflows applied to ADCIRC are usually constant values representative of flow rates during landfall. This may be adequate for systems producing flooding dominated by storm surge, but some systems cause significant precipitation-induced flooding as well as surge, e.g.. Hurricane Floyd produced only 3 meters of storm surge, but precipitation resulted in over 500-year flood levels on the lower Tar River. Current NOAA and DBS projects seek to produce holistic coastal flood inundation modeling by coupling hydrologic models, HL-RDHM and Vflo, to ADCIRC. Results from an idealized study show routing using a dynamic wave approximation for the momentum equation is necessary to accurately handle backwater effects; kinematic wave approximation routing should only be used upstream of backwater effects. The initial target area is the Tar and Neuse River region of coastal North Carolina. Preliminary application of ADCIRC, to a domain that includes major rivers in this area subject to backwater effects, forced with a discharge hydrograph at the upstream boundary produces promising results.

An integrated storm surge modeling system, CH3D-SSMS, based on the CH3D (Curvilinear-grid Hydrodynamics in 3D), has been further enhanced to enable simulation of storm surge, wave, inundation, and 3D circulation along Florida coasts during recent tropical cyclones. Until recently, CH3D was coupled to a barotropic basin-scale model to simulate 2D/3D barotropic response of coastal and estuarine waters to tropical cyclones. In this study, CH3D is coupled to a 3D baroclinic circulation model HYCOM which covers the North Atlantic and Gulf of Mexico to enable simulation of salinity, temperature, and currents in addition to water level in coastal waters. Tidal open boundary condition is obtained from ADCIRC. The CH3D is also coupled to a coastal wave model SWAN which receives open boundary condition from the basin-scale WAVEWATCH-III model. The 3D baroclinic CH3D-SSMS is used to simulate the surge, inundation, and salinity in the NE Florida coast, including St. Johns River and Guana-Tolomato-Matanzas National Estuarine Research Reserve (GTM-NERR) near St. Augustine, during Tropical Storm Fay in 2008. The slow moving storm with heavy precipitation caused significant inundation in St. Johns River while lowering salinity inside the GTM-NERR. While the water level during Fay was accurately simulated, accurate salinity simulation required the use of precipitation forecasts from HYCOM and river flow forecasts from NWS' AHPS. Inadequate river flow forecast results in salinity simulation errors.

Differences in ADCIRC 2D depth-integrated and ADCIRC 3D results are explored with particular attention to strong, wind forced applications. The 3D model results are sensitive to the turbulent vertical eddy viscosity (EV) formulation and bottom stress formulation. At high values of EV, the 3D velocity profile becomes uniform with depth and the simulation results match the 2D results. EV values typically range from 0.05 to 1.0 m²/s, with higher values in regions of wave breaking. Sensitivity to EV is tested using steady state, spatially uniform forcing and a domain of uniform along shelf depth. Maximum water elevations from the 3D model can differ from those predicted by the 2D model in a range of 0 percent for high values of constant EV to 9 percent for EV computed using Mellor-Yamada 2.5. We expect that the Mellor-Yamada parameterization may underestimate the vertical mixing during strong wind events such as tropical cyclones due to the effects of breaking wave induced turbulence, and therefore, that the actual differences will be substantially less than 9% during these events. The bottom roughness parameter, z₀, governs the 3D bottom drag coefficient which determines the bottom stress response, and can range from 0.10 m to 0.0001 m. We present results from a series of 3D simulations of Hurricane Isabel using the Mellor-Yamada 2.5 turbulence closure model and constant bottom roughness for a range of roughness values. These results are compared against 2D simulations of Hurricane Isabel using a Manning's n type bottom friction. Our results show that the differences in maximum elevations between the 3D and 2D results range from ±5% over most of the domain to ±20% in localized regions of inundation. There is no indication that, with the current state of the science as applied to storm surge modeling, the 2D and 3D results differ beyond the levels of intrinsic error in specifying the vertical mixing or bottom friction parameterization that currently exists for each method. We show that the differences in maximum elevation of storm surge between a 2D and 3D model is probably on the order of ±5%.