Estuarine and Coastal Modeling (2005)

The Northwest Fork of the Loxahatchee River contains one of the last examples of a pristine subtropical riverine cypress swamp in South Florida. The permanent opening of the Jupiter Inlet along with sea level rise and reduced dry season freshwater flows in the Northwest Fork has allowed salt water to migrate upriver causing the introduction of mangroves into areas once dominated by freshwater floodplain swamp vegetation, particularly bald cypress. Restoration of this natural resource requires reducing or reversing the current condition of saltwater intrusion and maintaining the freshwater habitats in the upper reaches of the Northwest Fork to protect existing populations and distribution of wildlife that require freshwater habitat. A computer program was developed to predict salinity and calculate other performance parameters under various ecosystem restoration scenarios. Field data, regression analysis and results of multi-dimensional hydrodynamic computer models were integrated into a system simulation and management tool. Salinity prediction and other computations were conducted over a 39-year period. Such long-term simulations are required to investigate ecosystem response and assess the effectiveness of the proposed restoration approach. The model also contains system management functions such as the calculation of additional freshwater demand for salinity management under various restoration alternatives.

A three-dimensional (3-D) hydrodynamic model for the Charleston Harbor System (CHS), South Carolina, has been developed for the Berkeley-Charleston-Dorchester Council of Governments (BCD-COG) to assist in developing the dissolved oxygen Total Maximum Daily Load (TMDL). The hydrodynamics of the CHS are largely influenced by incoming tides, salinity, and vegetated marshes. The CHS drains over 1,200 square miles directly, with a larger area contributing freshwater flow to Lake Moultre, the largest contributor of freshwater flow to the Cooper River. The Environmental Fluid Dynamics Code (EFDC) was used for the hydrodynamic simulation. The CHS EFDC model extends beyond the main Harbor into the Atlantic Ocean and it includes the Ashley River, Wando River, and Cooper River including both east and west Cooper River branches. Both instream flows and lateral overland flows generated by the watershed model, the Loading Simulation Program in C++ (LSPC), were input to the EFDC model. Effluent discharges received from facilities and the State were also input as lateral flows to represent point source inflows to the system. The CHS marshes were delineated from landuse maps. Wetting and drying grid cells, equal in area to the delineated marsh area, were initially simulated in the model. Using continuous monitoring data collected 1996 and 2004 and several special studies the hydrodynamic model was calibrated and validated. Measured water levels, flows, salinity, and temperature at several stations in the CHS were compared with simulated model results.

NOAA's National Ocean Service (NOS) and Lamont-Doherty Earth Observatory of Columbia University, have designed and conducted a sulfur hexafluoride (SF₆) experiment in the Houston Ship Channel. SF₆ was released at the confluence of the HSC and Patrick Bayou, an EPA superfund site, to measure the dispersion characteristics of the upper Houston Ship Channel within the Port of Houston. Approximately 2 moles of SF₆ were released on 17 May and surface and vertical profiles were measured with a gas chromatograph over the period 17–26 May 2005. Initial measurements are provided and dispersion characteristics are presented. Numerical simulation results, which were performed prior to the experiment using the NOS Galveston Bay Operational Forecast System (GBOFS), were used to estimate the extent of tracer dispersion to plan the survey. The Bay and Channel hydrodynamic models within GBOFS were used to simulate the transport and dispersion of SF₆ over the ten day period, November 2–11, 2004, by adding two concentration algorithms to each model to simulate the movement of the passive tracer with and without surface gas transfer. The simulated water levels, currents, and density were compared with PORTS observations to quantify model accuracy. Model simulated tracer concentration distributions and the total tracer mass balance were studied. Residence time, turnover times of the injection cells, and exposure level and duration were computed. Areal extents of tracer concentration above background levels of 2 fmol L^–1 were determined to assist in the planning of the tracer release experiment. Surface gravity wave algorithms have been incorporated in both hydrodynamic models to simulate short period waves with and without wave-current interaction and indicate the feasibility of including the wave algorithms within GBOFS. In conclusion, plans for additional SF₆ data analysis and recommendations for additional SF₆ numerical simulations are discussed.

The Mississippi River has major economic, environmental, ecological, and industrial values to the entire United States. At present, the Mississippi River Delta area of coastal Louisiana is deprived of practically all the sediment transported by the river to the Gulf of Mexico. Therefore, alternative solutions to recover or re-direct a portion of this massive amount of valuable sediment to benefit the restoration of Louisiana coastal lands are being investigated. These investigations consider the impact of management and restoration projects on the conditions of the river (supply side) and on the surrounding wetland and water bodies (demand side). This paper evaluates the use of a suite of numerical models to aid in the simulation of the bed-material and wash-load components in the Lower Mississippi River. This approach provides information on the river's hydrodynamics and sediment characteristics with large spectra of temporal and special scales. The models serve as viable and efficient management and analysis tools for the Lower Mississippi River. They provide detailed information on the availability of fresh water and sediment for diversion to surrounding wetlands, and determine quantitatively the impact of existing and planned diversion projects on the dynamics of the river. The models developed herein would also provide sediment and water information needed for larger scale models encompassing the Mississippi River Delta and the continental shelf.

The Peace River, located in southwest Florida, flows southwestward and drains approximately 6213 km² of watershed into Charlotte Harbor, a relatively large estuary in the state. The river is generally meandering and narrow except for its downstream area where the river width varies from several hundred meters to several kilometers. To establish a regulatory freshwater inflow rate for the Lower Peace River, a dynamically coupled 3D-2DV model was developed and used to simulate hydrodynamics in the Lower Peace River — Upper Charlotte Harbor system. The simulation domain includes a 3D sub-domain that consists of the northern portion of Charlotte Harbor and the most downstream 15.5 km of the Lower Peace River and a 2D sub-domain that extends about 43 km further upstream and covers the entire tidal reach of the river. In the coupled model, 3D Reynolds Averaged Navier-Stokes (RANS) equations are solved in the 3D sub-domain, while laterally averaged (2DV) RANS equations are solved in the 2D sub-domain. A free-surface correction (FSC) method is used to solve governing equations in both the 3D and 2D sub-domains. The FSC method is unconditionally stable with respect to gravity waves, wind and bottom shear stresses, and vertical eddy viscosity terms. The use of the FSC method allows both 3D and 2DV equations to be solved simultaneously instead of in sequence. The coupled model was run for a simulation period of about 12.6 months, from June 13, 2003 to July 7, 2004. Model results were compared with measured surface elevations, velocities, salinities, and temperatures at four stations in the simulation domain. Model runs show that the coupled model is very efficient and yields good model results for the complex Lower Peace River — Upper Charlotte Harbor system.

In this study, Environmental Fluid Dynamic Code (EFDC) is used to simulate the salinity plume distribution in the mouth of the Cape Fear River Estuary (CFRE). Effects of astronomical tide, river discharge and wind on the CFRE salinity plume were investigated. The spatial and temporal characteristics of the model simulated salinity plume are compared with observations measured by the Coastal Ocean Research and Monitoring Program (CORMP). The results indicate that model results and observations show a good agreement in water level and salinity. The simulations also indicate that strong winds tend to reduce the surface CFRE plume size and distorting the bulge region near the estuary mouth due to enhanced wind induced surface mixing. Even moderate wind speeds could fully reverse the buoyancy-driven plume structure in CFRE under normal river discharge conditions. Tide and the river. discharge also important factors ....

In this paper, various turbulence closure models are compared for use with a hydrodynamic 3D code in the Delaware Bay. More specifically, six (6) different turbulence closures, i.e. a constant eddy viscosity, an algebraic model, and 4 two-equation closure models have been used for comparison. These models have been implemented in the UnTRIM hydrodynamic code using the Generic Length Scale approach that mimics through its parameter combinations 3 of the two-equation closures (k-epsilon, k-omega, and k-kl) plus a separate Yamada Mellor 2.5 code. The UnTRIM code was used to model flows and salinity transport for a 2 month period by keeping the boundary and initial conditions unchanged for all model comparisons. The performance of each closure scheme was tested against salinity time series at a single station in the estuary. The comparisons reveal that the zero and first order approaches perform fair, but less accurate than the two-equation models that warrants the increased computational cost associated with these higher order approaches. Among the four 2-equation models, the k-epsilon model best represented the measured salinity time histories in the bay. The differences when compared to the other 2-equation models were significant enough to conclude that an arbitrary choice from the list of available or commonly used turbulence closure models is not sufficient. As a result, it is recommended that a test series be conducted to identify the best choice of turbulence closure in a specific application.

A three-dimensional shelf circulation model is used to simulate the seasonal variability of circulation and hydrography on the Scotian Shelf over the eastern Canadian seaboard. The model is a modified form of an operational forecast system known as DalCoast3. The model is forced by forecast winds and atmospheric pressure produced by GEM, a Canadian numerical weather prediction model. Climatological temperature and salinity are used to initialize the model, and as lateral open boundary inputs. The monthly mean currents from a baroclinic circulation model of the northwest Atlantic Ocean are used as inputs to represent remotely forced flows. A new data assimilation method developed by Thompson et al. (2006b) is used to constrain model temperature and salinity to their respective climatologies, while allowing short-term (O(days)) variations to evolve prognostically. The model produces southwestward flows along the Nova Scotian coast and along the shelf break, which have been observed in the past. The pattern of simulated volume transport over the Scotian Shelf is qualitatively similar to those from past model studies, and the transport across a section on the inner shelf is also comparable to values calculated from observations.

A three-dimensional nonlinear finite element model is used to simulate climatological monthly-mean circulation in the Flemish Cap region. The model is forced with monthly-mean wind stresses from the NCEP-NCAR reanalysis data prescribed at the sea surface, and large-scale remote forcing specified at the open boundary from the steric height and determined from a North Atlantic wind-driven circulation model. The model results are consistent generally with previous study results and in approximate agreement with observations for major patterns. They indicate that shelf-edge and upper-slope Labrador Current water is the important source for the Flemish Cap for all seasons. The deep Labrador Current and the North Atlantic Current may modify the water properties and circulation of the Flemish Cap. The Labrador Current bifurcates north of the Flemish Pass, with one branch flowing southward through the Flemish Pass and the other flowing eastward around the Flemish Cap. The southward Flemish Pass flow has a seasonal evolution (peak in December) that is consistent with the upstream shelf-edge Labrador Current off central Labrador that peaks in October. The monthly-mean model solutions indicate that the Flemish Cap features a weak year-round anticyclonic eddy with a mean transport of 0.2 Sv within the 700-m isobath, intensified in August and winter. The eddy seems to be generated by the baroclinicity of the Labrador Current and the sloping bottom topography of the Cap. There is also an anticyclonic partial gyre around the Cap's northern, eastern and southern deep flanks. The associated transport within the 3000-m isobath decreases significantly downstream, and is strongest in summer on the seasonal scale.

Three-dimensional simulations of circulation in the San Francisco Estuary were performed with the three-dimensional hydrodynamic model, TRIM3D, using a generic length scale turbulence closure model. The model was calibrated to reproduce observed tidal elevations and tidal currents in the San Francisco Estuary and then was applied during a period of intensive data collection. The model predicts tidal currents accurately and realistically simulates variability in salinity at both the seasonal and tidal time scale. The model results are consistent with the current conceptual understanding of stratification in the San Francisco Estuary and approximately predict the magnitude and tidal phasing of observed stratification. The simulation results were analyzed to improve understanding of periodic stratification. During spring tides, compression of salinity gradients near high water and weak tidal currents at and following high water allow the formation of weak and transient stratification. During neap tides stronger stratification is observed and predicted. This stratification is strongest during ebb when tidal straining is effective in creating stratification. Vertical turbulent mixing may be adequate to diminish vertical stratification during strong ebb tides while stratification can persist or increase through weak ebb tides.

Cobscook Bay is situated in the easternmost part of Maine, US, bordering Canada's maritime province New Brunswick. Vigorous tidal currents mix the water column rich of nutrients, which make this small bay very productive biologically. Rapid growth in salmon aquaculture has a great impact on the ambient coastal environment. Since the intertidal flats account for a sizeable percentage of the total bay surface area at low water, it is necessary to include the wetting/drying process in numerical models when simulating the circulation of Cobscook Bay. The 3D nonlinear finite element ocean model Quoddy with a drying/wetting routine is used in this study to examine the capacity of tides to flush nutrients, pollutants and waste materials from the inner reaches of Cobscook Bay. Driven by major tidal constituents and realistic winds, the modeled circulation agrees well with observed data. Comparisons of time series from the model with in situ observations show that the ellipses at the GoMOOS buoy J are almost rectilinear following the regular cycle of the semidiurnal tide. Water level at Eastport matches observation very well. Several passive drifter experiments are conducted using a Lagrangian particle-tracking program, which agree very well with observed drifter trajectories. The fundamental mechanisms for exchanges between Cobscook Bay and the adjoining Passamaquoddy Bay are strong flooding/ebbing tidal currents and to a lesser degree the flows driven by the instantaneous local winds.

Recreational shellfish beds in the Southport Beach area of Southport Harbor, Connecticut are impacted by elevated levels of fecal coliform bacteria. However, the locations and relative magnitudes of the sources of these bacteria have not yet been identified and quantified. In this study, a computer modeling approach to characterize the sources of the bacteria responsible for closure of recreational shellfish beds in Southport Harbor is demonstrated. Field studies provided data to support the modeling and included physical and chemical measurements, bacterial sampling, and dye studies for two suspected sources was conducted to better understand flow in the area and to support the modeling effort. Circulation in Southport Harbor was found to be primarily driven by the tides and secondarily by winds. The tidal response in the harbor is of a mixed standing and progressive wave. The bacteria surveys indicated highest fecal coliform (FC) concentrations in the Mill River and at the mouth of Sasco Brook, locations on either side of the beach area. An integrated hydrodynamic and pollutant transport modeling framework (WQMAP / OILMAP) was used to identify the location and evaluate the bacteria sources responsible for closure of the recreational shellfish beds. The hydrodynamic model was applied to the greater Southport Harbor estuary. The model, which was calibrated to data from the field program, provided temporally and spatially varying currents for use in the subsequent pollutant transport calculations. A Lagrangian (particle-based) approach was used to model pollutant transport both forward and backward in time. In the forward mode the Lagrangian model predicts the evolution of pollutant plumes from specified source locations, while in the backward mode it estimates the likely source locations that would affect a specified end point or resource area, in this case the Southport Beach shellfish bed. Forward transport modeling indicated that pollutants from a distant source (Mill River) were more likely to impact the Southport Beach area than pollutants from an adjacent source (Sasco Brook). Backward mode results from the Lagrangian transport model confirmed that the Mill River was the likely source based on six representative receptor sites located in the beach area. These results indicated that offshore sources might also be significant, but these can be discounted since there are no known sources there. Modeling results thus suggest that control of the more distant Mill River bacteria sources is more likely to improve water quality at Southport Beach than control of the nearby Sasco Brook sources.

A recently developed age concept is capable of quantifying transport characteristics of dissolved substances over a Eulerian coordinate framework. This allows for the estimation of transport time to be presented as a spatial variable and reveals detailed transit time information for transport both within a basin and to reach its exterior. This paper outlines the calculation of spatial distribution of age-of-water within coastal embayments by using a high-resolution numerical hydrodynamic model. The Lynnhaven River, located in Virginia Beach, Virginia, is used for the case study. The surface elevation, current velocity, and salinity were calibrated using the UnTRIM model. Implementing the age-of-water calculation into a calibrated hydrodynamic model, one can account for the localized effects of tide and river discharge. To quantify transport properties in the context of transport timescales, flushing time, residence time, turnover time, and age-of-water have often been used to determine water contamination, nutrient retention, and biochemical processes in estuaries. First-order timescale (integrated) calculations have been traditionally used for shallow water coastal embayments, assuming a completely mixed waterbody. This may not be the best approach for complex coastal embayments, even if they are "small".

The problem of water pollution has become more widespread and severe in Chinese estuarine and coastal areas, such as the Bohai Bay and the mouth of the Yangtze River. The Bohai Sea has experienced the highest increase of pollution levels caused by rapid economic development and discharge of land waste. In order to recover the ocean environment in the Bohai Sea, it is necessary to understand the hydrodynamics of the sea. The present study focuses on the evolution of thermal stratification and the dispersion processes as a result of river water from the Yellow River discharging into the Bohai Sea. A three-dimensional baroclinic flow model was applied to simulate the current system and the density field, using the NCEP/NCAR Reanalysis (available from the NOAA Climate Diagnostics Center) and JCOPE reanalysis data (Miyazawa and Yamagata, 2003) as a boundary condition. A long-term hindcast simulation for 2003 was carried out. This simulation makes the spatial variability of the density stratification clear in the Bohai Sea. Moreover, the detailed particulars of the behavior of suspended solids in the Yellow River water are investigated using a modified Lagrangian particle tracking method. The present model can predict well the distribution of high turbidity water mass which can be seen in the southwestern sea area in the Laizhou Bay in Landsat satellite images. Wind effects on the sediment distribution were evaluated.

Brunnsviken in the inner Stockholm archipelago, close to the City of Stockholm and popular for recreational activities, is a semi-enclosed bay with a very narrow passage to the adjacent archipelago and consequently has a limited water exchange. Various attempts have been made over the past three decades to improve the water quality in the bay. Since 1986, the drinking water authorities of Stockholm withdraw bottom water by pumping it from one of the deep basins of the bay through a pipe to be eventually discharged into the nearest embayment of the archipelago. There are, however, some questions regarding the cost-benefit aspect of this strategy. In particular, it is unclear if the location and the rate of pumping are well chosen in order to increase the ventilation of the bay at large. In addition, it should be possible to eventually optimize the pumping schedule, so that these energy-demanding and thus costly efforts are concentrated to times when they act in concert with natural forcing to increase water exchange. To help improve the pumping program, a modelling project was started in spring 2004. First, a one-dimensional (1-D) layer model that resolves the bay into one basin with multiple-layer stratification was attempted. Second, a three-dimensional (3-D) model with the capacity to be run under non-hydrostatic assumptions was set up at high resolution to study a number of scenarios. Both models were forced by wind, river discharge, surface temperature and the exchange driven by density fluctuations across the boundary to the adjacent archipelago. In the 3-D model, the pumping is included as a virtual divergence of the flow at the location of the pipe. The main results are that the 1-D model performs considerably better than the 3-D model in simulating the measured salinity profiles, even though it does not resolve the basin in the horizontal direction. The poor performance of the 3-D model with regard to salinity is however mainly due to boundary problems: too little inflow of saline water through the narrow and shallow entrance channel, and underestimation of the freshwater supply. The dynamics of the thermocline is better captured by the 3-D model but can certainly also be improved.

A triply nested-grid ocean circulation modeling system is used to simulate the circulation and associated seasonal variability on the Meso-American Barrier Reef System (MBRS) of the northwest Caribbean Sea. The nested-grid system consists of three subcomponents: a coarse-resolution outer model of the western Caribbean Sea (WCS); an intermediate-resolution middle model of the southern Meso-American Barrier Reef System; and a fine-resolution inner model of the Belizean shelf. The two-way nesting technique based on the smoothed semi-prognostic method developed by Sheng et al. (2005) is used to exchange information between the three subcomponents of the system. The nested-grid system is forced by 12-hourly NCEP wind and climatological monthly mean sea surface heat and freshwater fluxes and integrated for 11 years from 1990 to 2000. Model results of the last 10 years are used to calculate the monthly mean currents, temperature/salinity distributions and associated variability over the MBRS. Model results demonstrate that the circulation has significant temporal and spatial variability in the study region.

Water environment in brackish lake is very complex. To analyze the water qualities, it needs to approach from physical aspect and ecological aspect. Moreover, it needs to consider not only the influence of ecology on the water qualities but also the influence of water qualities on ecology itself. We developed ecological model to consider the biochemical processes of the bivalve, that calculates water temperature, salinity, chlorophyll-a, inorganic nitrogen (NH₄, NO₃) and phosphorous, organic nitrogen and phosphorous, COD and DO. Applying this model to Lake Ogawara, we analyzed the characteristics of the water qualities in the brackish lake. In order to clarify the influence of water qualities on ecosystem, we establish the model that predicts the variation of Biomass of Corbicula japonica, that is the dominant bivalve in Lake Ogawara. Main results in the present study are summarized as follow. It can be calculated the spatial and seasonal variation of water qualities with accuracy. It is found that spatial distribution of water qualities is dependent on the physical factor such as a flow scale and a lake size, and on the biochemical factor such as a time scale of production and decomposition. We can clarify the influence of water qualities on the bivalve's growth and population. It is found that the growth rate is faster in the southern area than northern area, and the population tends to decrease because of the lack of spawning.

The tidal Christina River is located at the lower portion of the Christina River Basin, and connects to the Delaware River. The enterococci level is one of the major concerns for watershed management. The tidal Christina River receives enterococci loads from Brandywine Creek, Red Clay Creek, White Clay Creek, non-tidal Christina River, and 38 Combined Sewer Overflows (CSO) from the city of Wilmington, DE. The Environmental Fluid Dynamics Code (EFDC) model was modified and configured to address the fate and transport of enterococci in the tidal Christina River. Enterococci loads from the creeks were provided from four watershed loading models using the Hydrological Simulation Program FORTRAN (HSPF). The loads from CSOs were calculated from the output of XP-SWMM model developed by the city of Wilmington. The hydrodynamics were calibrated with water surface elevations at two USGS tidal stations, temperature and salinity time series at five water-quality monitoring stations. The EFDC code was modified to include impacts of solar radiation, temperature, and salinity on the die-off of enterococci. The modeled enterococci were compared against field data at five monitoring stations. The model results reveal that the tidal Christina River is mainly affected by the enterococci from the upstream creeks due to wash off of manure on agricultural land during storms and animal direct deposit of waste in streams in warm weather under existing conditions. CSOs only affect limited areas around the outfalls since the dilution is strong in the tidal Christina River. However, to achieve the water quality standard for the entire tidal Christina River, significant reduction of CSO loads is still required.

Hog Island Bay, Virginia, is a shallow back barrier lagoon that is subject to seasonal inputs of inorganic nitrogen and related episodes of hypoxia. Numerical simulations were carried out to estimate the importance of physical flushing times relative to biochemical turnover times known to be a few days or less within the system. A 2D vertically averaged finite element hydrodynamic model, which was designed to accommodate regular flooding and dewatering of shallow flats and marshes, was coupled with a particle tracking model to estimate median lagoon residence time and the spatial distribution of local residence time in the lagoon. The model was forced with observed tidal elevations and winds from the end of the growing season when hypoxia tends to occur. The median residence time estimated by numerical modeling is on the order of weeks (358 hours), and variations in tidal stage, tidal range and wind produced deviations in median residence time on the order of days. Residence times near the inlets were very short, while those near the mainland were long, showing that (1) horizontal mixing in the Bay is insufficient to successfully apply integral methods to obtain residence times, and (2) residence times near the mainland are long compared to timescales of biologically driven chemical transformations.

A model to examine the impacts of long term sea level rise (SLR) is being implemented in the coastal North Carolina ecosystem. This area is particularly vulnerable to SLR, as a fragile system of barrier islands protects an extensive but sensitive estuarine system. The primary impact of SLR is to the hydrodynamic response of the system: circulation, tidal amplitude, and inundation patterns due to tides, winds, and storms can all change in response to rising sea level. Rates of SLR in the region are just under 3 mm/year and are increasing, and inundation is tied to inlet conveyance which can be modified by SLR. A Coastal Flooding Model (CFM) has been developed for the region by combining a finite element hydrodynamic model with a continuous bathymetric and topographic elevation dataset. The CFM domain extends from 90 km offshore of the Outer Banks to the 15 m topographic contour and from northern Currituck Sound south to the New River. The CFM provides high resolution of coastal features down to 20 m. High resolution topographic elevation data relative to the North American Vertical Datum of 1988 (NAVD 88) was combined with bathymetric sounding data relative to local tidal datums by transforming the tidal datums to NAVD 88 with the VDatum vertical transformation tool developed by the National Oceanic and Atmospheric Administration's (NOAA) National Ocean Service (NOS). The VDatum tool allows for transformation among nearly 30 different tidal, orthometric, and ellipsoidal vertical datums. A 6 m horizontal resolution continuous bathymetric/topographic (bathy/topo) Digital Elevation Model (DEM) was constructed for accurate modeling of inundation. The CFM is relative to the NAVD 88 vertical datum and was populated with DEM elevations where available and other topographic and bathymetric data relative to NAVD 88 elsewhere to create a continuous bathy/topo elevation field. A two-dimensional barotropic model is used to simulate the tidal response in the CFM to study changes due to SLR and will also be used to model regional synoptic wind events and hurricane storm surge propagation with SLR. Accurate simulation of inundation patterns is accomplished by high localized resolution in the coastal zone, continuous bathy/topo data, and an accurate wetting/drying algorithm. The CFM is validated against observational data before modification of initial and boundary water levels to represent eustatic SLR. The RMS error at tidal stations was calculated for the primary constituent amplitudes and phases. The average of these RMS amplitude errors is 7.8 x 10^–3 m and the average of the RMS phase errors is 6.57°. Shoreline migration can be dynamically computed from the CFM simulation output as a function of SLR. Finally, the CFM will be coupled to submodels that characterize the ecological impact of SLR.

Grid computing is emerging as an attractive approach to solve large problems of importance to the estuarine and coastal modeling community through the coordinated sharing of resources. Using the simulation of storm surge as a proxy for a typical estuarine and coastal modeling application, two examples of computation using Grid computing are undertaken. The applications used in the examples are Grid-enabled through virtualization techniques available through the In-VIGO web portal and Grid middleware. Virtualization allowed both examples to be Grid-enabled without modification of the source codes in any way, thus requiring minimal IT expertise by the modeler. In the first example, regional Grid computational resources at three institutions, UF, LSU and VIMS are used to perform a series of ensemble simulations of storm surge using the CH3D component of the CH3D Storm Surge Modeling System (CH3D-SSMS). Ensembles consist of the response of Tampa Bay, FL to various forecasted tracks of Hurricane Charley (2004). The Grid-calculated ensemble Maximum of Maximum (MOM) water level is identical to the MOM simulated on a single host or set of resources at one institution with the advantage that the computations can be performed in parallel across the regional Grid's resources. A 1% increase in wall clock time is shown when solving the simulations on different Grid resources. In the second example, the execution of the local-scale surge model (CH3D), the local-scale wave model (SWAN) and the regional-scale surge model (ADCIRC) components of the CH3D-SSMS are used to determine if coupled modeling systems can perform adequately across Grid resources. With CH3D running at VIMS, SWAN at LSU and ADCIRC at UF and coupling performed via simple file I/O through a virtual file system, the entire coupled system shows an increase in wall clock time of 13%. As with the ensemble simulations, simulated parameters are identical to those computed on one host. The relative ease of Grid-enabling both examples combined with the minimal increase in wall clock time demonstrates that typical estuarine and coastal modeling applications can successfullly use Grid technologies across a regional Grid, which, in turn, opens up estuarine and coastal modeling to the possibility of using resources from larger Grid projects, e.g. SURAgrid or Open Science Grid.

There are many water quality models (e.g., WASP, QAUL2E/QUAL2K, CE-QUAL-ICM, etc.) that have been employed by practitioners in surface water quality modeling. All of these models are similar to each other. The major differences among them are the number of water quality parameters included and the number of biogeochemical processes considered. There is a need to develop a model that would allow the inclusion of any number of water quality parameters and enable the hypothesis of any number of biogeochemical processes. This paper presents the development of a numerical water quality model using a general paradigm of diagonalizing reaction networks. In a reaction-based approach, all conceptualized biogoechemical processes are transformed into a reaction network. Through the decomposition of species governing equations via Gauss-Jordan column reduction of the reaction network, fast reactions and slow reactions are decoupled, which enables robust modeling of fast/equilibrium reactions. For example, the model alleviates the needs of using simple partitions for fast reactions. The system of species governing equations is transformed into three sets: algebraic equations (either mass action equations or users' specified) of equilibrium-variables, differential equations of kinetic-variables and differential equations of component-variables. With the diagonalization strategy, it makes the inclusion of arbitrary number of fast and kinetic reactions relatively easy, and, more importantly, it enables the formulation and parameterization of kinetic reactions one by one. To demonstrate the flexibility and generality, the eutrophication model in WASP5, QUAL2E, and CE-QUAL-ICM are recast in the mode of reaction networks. This illustrates that the model embeds the most widely used water quality models as specific examples. Based on these three examples, the deficiencies of current practices in water quality modeling are discussed and the actions that must be taken to improve these practices are addressed.

Previous studies have shown that utilizing the conservative form of the momentum (CM) equation improves the mass conserving ability of the ADCIRC (ADvanced CIRCulation) finite element model, especially in areas of steep bathymetry gradients. Local spatial accuracy also improves, due to a decrease in the truncation error, yet does so without significantly impacting temporal or global spatial accuracy. That study raised other research questions, two of which are examined herein. The first deals with the lateral stress closure scheme utilized in the model. As the ADCIRC community begins to incorporate more near shore physics and baroclinic capabilities, a more sophisticated closure scheme is needed that allows both spatial and temporal variation, such as the Smagorinsky formula. The second question explores the treatment of the advective terms in the governing equations. In this paper, we examine various algorithms using a variety of metrics in order to determine the "best" formulation for the lateral stress and advective terms in the governing equations. Initial results indicate that mass conservation errors are impacted more by the treatment of the advective terms than the lateral stress formulations, with algorithms based on the CM equation producing the lowest error. Results also indicate that the lateral stress formulation in the continuity equation only slightly impacts the elevation results; other experiments were unaffected by the different formulations. When analyzing the CM equation, we observe that the native treatment (i.e., derivatives not expanded) of the advective terms produces smaller differences in spatial accuracy for most of the elevation and velocity components. Taken together, results do not single out a "best" algorithm, so additional work is needed, particularly with the comparison of velocity-based versus flux-based lateral stress formulations.

We extend two fractional-step pressure methods (pressure projection and pressure correction) to nonhydrostatic free-surface flows by employing them to discretize the nonhydrostatic pressure on a staggered grid and applying a semi-implicit time discretization to the free-surface elevation. The time accuracy and efficiency for each pressure method is investigated in detail. Using a sloshing free-surface wave as a test case, we show analytically and numerically that the correction method is always second-order accurate in time, while the projection method is second-order accurate in time only if the nonhydrostatic pressure is included in the depth-averaged continuity equation through the use of an iterative procedure. We demonstrate that this convergence behavior holds for the linearized equations.

Efficiency and accuracy of a fully non-hydrostatic three-dimensional model are examined through simulation of linear and nonlinear dispersive waves. The model uses an implicit method to solve the unsteady, three-dimensional, Navier-Stokes equations and the fully nonlinear free-surface boundary conditions. A new top-layer pressure treatment that takes the non-hydrostatic pressure into account is implemented. The model results are verified against analytical solutions, indicating that the model using a small number of vertical layers is capable of effectively simulating the free-surface elevation. By further examining the performance of the model in resolving nonlinearity and dispersion of surface wave characteristics, we demonstrate the efficiency and accuracy of the model. To extend the model's application, a generic length scale approach is adopted in the non-hydrostatic model, yielding the k-kl (Mellor-Yamada Level 2.5), k-epsilon, k-omega and gen turbulent closure models. This newly developed non-hydrostatic model is used to simulate wind-driven circulations under different turbulent closure models.

The flux of hydrophobic organic chemicals (HOCs) between the bottom sediments and overlying water is primarily due to sediment erosion/deposition, molecular diffusion, and bioturbation. All of these processes are continuously and often significantly modified by the finite rate of adsorption and desorption of the HOC between the solid particles in the sediments and the surrounding pore water. Molecular diffusion can be a significant contributor to the overall flux, is always present, and modifies all of the other flux processes. In order to investigate the flux of HOCs due to molecular diffusion as modified by sorption, experiments have been conducted in two different sediments with eight HOCs which have a range of partition coefficients, K_p, from 5 x 10⁴ to 10 L/kg. Lengths of the experiments ranged from 32 to 256 days. For large K_p's, the results demonstrate that (1) significant changes in HOC concentrations are limited to a few millimeters near the sediment-water interface, (2) measured values of the HOC concentration are generally much less than their equilibrium value, and (3) the flux depends on K_p and is much greater than that for a non-sorbing chemical or for a sorbing chemical with equilibrium sorption assumed. As K_p decreases, the results reduce to those expected for pure molecular diffusion without sorption. A theoretical model which includes non-equilibrium sorption has been developed and is uniformly valid for all HOCs tested. Good agreement between the numerical calculations, asymptotic approximations, and experimental results is obtained.

Of the three processes (sediment erosion/deposition, molecular diffusion, and bioturbation) that primarily govern the sediment-water flux of hydrophobic organic chemicals (HOCs), bioturbation is probably the least well understood and quantified. The reason is the large variety of benthic organisms which may be present in surficial sediments and the different ways in which they affect the sediments. This is further complicated by the effects of non-equilibrium sorption which is significant for HOCs with high partition coefficients. In order to understand and quantify the flux due to bioturbation, laboratory experiments have been done with three benthic organisms (Lumbriculus variegatus, a vertical burrower and an oligochaete; Chironomus tentans, a horizontal burrower; and Hyalella azteca, a mechanical mixer of surficial sediments) and an HOC (hexachlorobenzene) in Lake Michigan sediments. Each of these organisms acts on the sediments in a different way; as a result, they are representative of a wide range of benthic organisms found in the sediments of surface waters. A convection-diffusion model including non-equilibrium sorption has been developed in order to analyze these experiments. Good agreement between the experimental and calculated results is obtained. The model is quite general and should be applicable to other organisms, HOCs, and sediments with the appropriate change of parameters.

Indian River Lagoon (IRL) is a long, narrow and shallow coastal lagoon located on the east coast of Florida. It is a nationally renowned aquatic ecosystem that supports tremendous biodiversity and provides recreational and commercial fishing resources as well. The St. Johns River Water Management District (SJRWMD) is charged with the responsibility of protecting this important coastal water body by developing seagrass-based Pollutant Load Redcution Goals (PLRGs). The development of justifiable PLRGs involves identification of the pollutants impacting the health and distribution of seagrass in the IRL and estimation of the response of the IRL to reductions of pollutants. To aid in these efforts, SJRWMD initiated and funded the University of Florida (UF) to develop a CH3D (Curvilinear Hydrodynamic Three Dimensional, based IRL specific integrated estuarine model. The integrated model, consisting of hydrodynamic, wave, suspended sediment transport, light attenuation and eutrophication model components, allows the examination of physical, chemical and biological interactions within the IRL system. UF has calibrated the model using one-year data. In this study, a two-year simulation was performed using HSPF (Hydrologic Simulation Program Fortran) predicted nutrient and sediment loads as boundary conditions. Simulated dissolved oxygen, chlorophyll_a and nutrients were compared with observed data in the IRL. Nutrient budgets computed using the model show the possible importance of nutrient recycling within sandy bottom sediments.

ADCIRC (ADvanced CIRCulation), developed over the last 20 years, is a hydrodynamic model capable of simulating water surface elevation and velocity fields in lakes, bays, estuaries, and oceans. The model is based on the full non-linear St. Venant (shallow water) equations, using the traditional hydrostatic pressure and Boussinesq approximations; the equations are discretized in space using linear finite elements, while time is discretized using an efficient split-step Crank-Nicolson algorithm. Model features include the following: both 2D depth integrated and 3D modes; linear or fully nonlinear equations; Cartesian or spherical coordinates; consistent or lumped mass matrices; Dirichlet, flux, or radiation boundary conditions; surface stress, flux, and tidal potential forcings; least squares analysis of harmonic constituents; wetting/drying of near-shore elements; fully-implicit time marching and, a preprocessor to optimize the code for various computer architectures. Over the years, the ADCIRC model has been supported by a number of federal and state agencies, with applications ranging from coastal dredging to larval transport to Naval fleet operations to hurricane storm surges. Recently, a transport algorithm has been added so that it is capable of prognostic baro-clinic simulations. Herein, we compare the performance of continuous and discontinuous Galerkin ("CG" and "DG," respectively) transport codes within the framework of the continuous hydrodynamic model; more specifically, the new codes are verified against known analytical solutions, examined for stability and convergence rates, and validated against experimental data. Results from verification studies show that both codes work well for low Peclet numbers (second order convergence), but only the DG code can capture the shock at high Peclet numbers. Results from the validation study show that both codes are able to capture the overall behavior of the "lock-exchange" experiment, although the DG is slightly more accurate. Parameter studies in the context of this lock-exchange experiment show that simulation results (both codes) are most sensitive to the value of the horizontal eddy viscosity. Global mass balance errors for both codes are much less than 1%.

Many coastal modeling applications rely on accurate bathymetry and/or topography data. The best data sets are often synthesized products with data collected from several sources which may be referenced to different vertical datums including Mean Low Water (MLW), Mean Lower Low Water (MLLW), North American Verical Datum of 1988 (NAVD 88), and others, depending on the age and source of the data. Transforming all of the data to a common vertical datum is a non-trivial problem, and so the National Ocean Service (NOS) of the National Oceanic and Atmospheric Administration (NOAA) has created a software tool called VDatum, which is designed to transform among approximately 30 vertical reference datums (including ellipsoidal, orthometric, and tidal datums). This program transforms among ellipsoidal datums using Helmert transformations, between the North American Datum of 1983 (NAD 83) (an ellipsoidal datum) and NAVD 88 (an orthometric datum) using a geoid model; between NAVD 88 and Mean Sea Level (MSL) using a topography of the sea surface model; and among tidal datums using tidal datum models. Regions where tidal datum models have been implemented include: Puget Sound and the Strait of Juan de Fuca, central California, southern coastal Louisiana, Lake Charles, Louisiana, Tampa Bay, the Outer Banks area in North Carolina, Delaware Bay, the New York Bight and Long Island Sound. A large collaboration of partners within NOS has participated to create this tool. The focus of this paper is on the work done in the Coast Survey Development Laboratory (CSDL) on the tidal datum modeling. Three types of models that have been employed to determine the tidal datum fields for the regional scale models are discussed: spatial interpolation of the datums, computation of the datums from a harmonic constant database, and computation of the datums from a regional hydrodynamic model. The results from these models are then interpolated onto the structured grid used with the VDatum software.

The majority of oil and chemical spill, search and rescue, and particle transport models employ a random walk procedure to predict the trajectory and dispersal of the relevant variables. The random walk model is the first in a hierarchy of stochastic particle models, where the object's position (random walk), velocity (random flight), and acceleration are progressively represented as Markovian processes. This hierarchy of models has been extensively developed and applied to study the movement of Lagrangian drifters in the ocean and the dispersal of air pollutants in flows over complex terrain. They have not however been incorporated in state of the art spill, research and rescue, and particle transport models. The goal of this paper is to investigate the impact of applying a random flight model to predict the trajectories in the above referenced models. Numerical implementation and testing of the first and second order models (random walk and random flight) against an analytic solution to the diffusion equation show very good agreement, provided that a sufficient number of independent simulations are ensemble averaged. The random flight model is shown to predict smaller dispersal areas than the random walk model, with a long term reduction in the area proportional to the dispersion coefficient times the velocity autocorrelation time scale. This offset ramps in immediately after the release, and is fully developed over multiples of the velocity autocorrelation time scale. In order for a random flight model to provide improved predictions over its random walk counterpart, accurate estimates of the dispersion coefficient and the autocorrelation time scale must be available. Simulation of typical coastal trajectory problems shows that the uncertainty in the value of the dispersion coefficient used as input to the model must be no greater than 1/n, where n (>3) is the number of autocorrelation time scales, if the random flight model is to provide improved predictive performance relative to a random walk model. Application of random walk and flight models to typical search and rescue are provided to illustrate the use of both models.

In support of a vertical datum transformation tool called VDatum, a tide modeling application was developed for the Long Island Sound, New York Bight, and Narragansett Bay area to compute spatially-varying tidal datum fields. The tidal datum fields, when combined with orthometric and ellipsoidal models, are input to VDatum, a software tool developed by the National Oceanic and Atmospheric Administration's (NOAA) National Ocean Service (NOS) to transform bathymetry-topography data among approximately 30 vertical datums. Tidal datum fields were derived from numerically simulated time histories of tidal water levels using the two-dimensional, depth-integrated version of the ADCIRC model. The model was applied to simulate barotropic tides and depth-integrated tidal currents. A high-resolution grid with 332,524 triangular elements and 180,042 nodes was created to represent the water area up to the Mean High Water (MHW) shoreline. The spatial resolution of the elements, ranging approximately from 25 m to 5.5 km, resolves fine coastline features and preserves local bathymetric variability. The grid bathymetry was derived from NOAA/NOS sounding data and was adjusted from Mean Lower Low Water (MLLW) to Mean Sea Level (MSL) iteratively using the results of a series of tidal simulations. The model was forced with harmonic constants for seven tidal constituents (K₁, O₁, Q₁, M₂, S₂, N₂, and K₂) extracted from a tidal database of harmonic constants from a regional tide model of the Western North Atlantic. A quadratic formulation was used to calculate the bottom friction. To optimize model and observation agreement, the open boundary conditions were adjusted by increasing the M₂ tidal amplitudes along the northeast portion of the boundary. Spatially-varying bottom friction coefficients were also used with values increasing westward through Long Island Sound in an effort to mitigate spatially variable overpredictions of water levels. Tidal datum fields of Mean Higher High Water (MHHW), MHW, Mean Low Water (MLW), and MLLW were calculated from the model results based on a 30-day period of water levels. They were compared to observations at 123 NOAA/CO-OPS water level stations across the study area. Good agreement was achieved, with the average discrepancy of the above four tidal datums less than 4 cm.

An immersed boundary method (IBM) for unstructured grids is presented for the simulation of flows over bottom topography. We implement the method in a nonhydrostatic, unstructured-grid code that solves the Navier-Stokes equations under the Boussinesq approximation. The equations are discretized on a staggered Cartesian z-level grid in the vertical plane and an unstructured triangular grid in the horizontal plane, and the solution to the hydrodynamics is based on a three dimensional semi-implicit free-surface approach together with the pressure-correction method for the non-hydrostatic pressure. In the simulation domain, the cells which contain the physical boundaries are treated as the ghost cells. Thus, each ghost cell itself has an immersed boundary, which is used to represent the presence of the physical boundaries based on the assumption that boundaries are piecewise linear within each cell. In the ghost cell, rather than directly imposing the desired boundary condition on the cell edge, which is the case for the z-level grids in the absence of an immersed boundary, the ghost-cell quantities are imposed at those edges to satisfy the desired boundary conditions at the piecewise-linear immersed boundaries. The ghost-cell quantities are obtained using linear extrapolations from the known interior points. Before solving the discrete governing equations, the spatial coefficients for the extrapolation are obtained based on geometries. Unlike previous studies on IBM, which impose the forcing term to make velocities to satisfy the desired boundary conditions, this study focuses on applying the immersed boundary method for solving the pressure field. The potential flow over a Gaussian hump is simulated using the proposed immersed boundary method and the results are compared with the simulation without IBM.

Hurricanes cause property damages and loss of lives in the U.S. every year. In 2004, four major hurricanes caused more than 5 billion dollars in damage each in Florida alone. Katrina in 2005 caused more than $200 billion dollars in damage. Major damage caused by a hurricane is usually associated with storm surge and coastal flooding. This paper presents CH3D-SSMS, a robust storm surge and coastal flooding forecasting system for tropical and extratropical storms. The basic forecasting system couples a storm surge and coastal flooding model (CH3D) with a shallow water wave model (SWAN) in a high resolution (50–500m) coastal/estuary/inland model grid. The coupled surge-wave model receives open boundary conditions of surge and wave from a global surge model (ADCIRC) and a global wave model (WAVEWATCH-III) which cover the Gulf of Mexico and Western Atlantic. Currently high resolution grids have been set up for the East Florida Coast (including St. Johns River, Indian River Lagoon, and adjacent coastal water), Tampa Bay and Charlotte Harbor, Florida Panhandle, and Chesapeake Bay and adjacent coastal waters. The forecast wind and atmospheric pressure information are provided by a number of NOAA wind models including NAM (North Atlantic Mesoscale), GFDL-Hurricane model, and an analytic model which uses GFDL hurricane parameters. Based on the forecast wind, the CH3D-SSMS produces an 84-hour forecast of water level, wave height/period/direction, flow field, and maximum of maximum (MOM) water level and inundation, every 6 hours. Prior to forecasting, the model performs a 24-hour nowcast using analysis wind. Performance of the CH3D-SSMS forecasting system has been demonstrated using results from 2003 (Isabel) and 2004 (Frances and Ivan) hurricanes. Forecasting of the impact of Hurricane Wilma on the Charlotte Harbor region is illustrated to reflect the current capabilities and weaknesses of the CH3D-SSMS as well as hurricane wind model. To improve the forecasting system, more accurate and efficient hurricane wind models and coupling of models of various processes and scales are needed. Hurricane-ground interaction should be incorporated into the hurricane wind model to produce more realistic near ground wind after hurricane landfall. Cyberinfrastructure should also be employed to facilitate regional and community collaboration.

The New York Harbor Observing and Prediction System (NYHOPS) has been quasioperational since late 2003. The system provides hindcasts/nowcasts/forecasts of water properties in an urban ocean environment. The environment includes the waters of the New York /New Jersey Harbor Estuary, from the coast eastward to the continental shelf break of the New York Bight and all of Long Island Sound in one contiguous fashion. Each day NYHOPS produces a 24-hour hindcast/nowcast and a 48-hour forecast of the physical state of water level, currents, temperature, and salinity. The model hindcast/nowcast/forecast skill assessment has been quantified for the entire year 2004 in terms of mean error (ME), standard deviation of the error (SD), RMS error and correlation coefficient. For the hindcast/nowcast, at most of the locations, errors in computed water levels are less than 8.5% of the local tidal range and correlations between the data and model exceed 0.94. The mean RMS errors and correlations degrade to 10 % and 0.91 for the 1^st day forecast and to 14 % and 0.83 for the 2^nd day forecast. The performance of the model for salinity and temperature are very encouraging. The vertical salinity stratification and its temporal variations in the different regions in the model domain are accurately reproduced in the model. For the hindcast/nowcast, the correlation coefficients range at various stations from 0.64 to 0.79 for salinity, and are typically greater everywhere than 0.96 for temperature. The 1^st day and 2^nd day forecasts of salinity and temperature have almost the same skill as the hindcast. The experience gained over more than one year of quasi-operational status has led to a NYHOPS system that is capable of describing the entire spectrum of time scales of the computed quantities. NYHOPS data and model products are all available in real-time at www.stevens.edu/maritimeforecast). Future extensions of NYHOPS include the assimilation of temperature, salinity and CODAR derived surface current measurements.

The National Ocean Service (NOS) has developed an experimental nowcast/forecast system over Galveston Bay using a three-dimensional hydrodynamic model. In addition, a one-way coupled fine resolution Houston Ship Channel model has also been incorporated into the system. Nowcast and forecast results have been assessed over the one-year period April 2000 through March 2001 based on the NOS formal acceptance statistical criteria and the system was made operational in June 2004. To move towards an all weather capability, the following algorithms were considered and initially tested: (1) rainfall/runoff inflows from four major basins within the City of Houston, (2) overland flooding, (3) tropical storm and hurricane wind and pressure fields, and (4) surface waves. Herein, we present initial test results for Hurricane Carla in September 1961, Hurricane Alicia in August 1983, and for Tropical Storm Allison in June 2001. Longwave model results for water levels are compared with tide gauge measurements and are in agreement for peak water levels within 35 cm (~1 foot), the informal NOS target level. High water mark data comparisons are more difficult to make, since they may often include, short wave information as well. Peak current, salinity and temperature response were also simulated and areas of inundation determined, but no data are available for comparison. Similarly, short wave results for significant wave height and period are discussed in terms of representativeness since no data are available for comparison. Based on these initial test results, plans for further development and testing are advanced.

Coastal ocean hydrodynamic models are used to simulate water surface elevations and circulation in oceans, lakes, estuaries, rivers and floodplains. One such model is ADCIRC (ADvanced CIRCulation), which was originally developed more than 15 years ago and which is used for a variety of purposes, including naval fleet operations, storm surge predictions, and larvae transport. ADCIRC assumed fixed land boundaries until a wetting and drying algorithm was implemented in 1995. However, this algorithm had two components that limited the performance of the model. First, nodes were required to remain "wet" or "dry" for a user-specified number of time steps before changing states. This component became restrictive in relatively flat regions, such as a flood plain, where it caused oscillations and slowed the propagation of flood waves. Second, in regions with steep bathymetry, mass balance problems and instabilities would occur when a thin film of water was allowed to flow uninterrupted. Changes based on a more physically-accurate description of the wetting and drying process were made recently to address these two problems. This paper describes the wetting and drying algorithm and those changes, and it applies the improved algorithm to an idealized domain that was designed specifically to test the two problem areas. The improved algorithm provides better stability and mass balance properties.

COASTMAP, a globally re-locatable marine environmental modeling and monitoring system, has been extensively upgraded, extended, and applied as an initial demonstration of the national backbone of the integrated ocean observing system (IOOS). Following the guidelines from the IOOS Data Management and Communications (DMAC) plan, the system provides end-to-end implementation, from data supplier to user, and includes data discovery; data and archive access; data fusion; and product generation, assessment, and product dissemination. The geographic information based system has two primary users; a professional. Windows ArcGIS- based (thick) client and a web browser-based (thin) client. The implementation features a web services approach to obtaining data from OPeNDAP and COASTMAP servers and supports a wide variety of data protocols and types. To illustrate the system's capabilities, it has been applied to demonstrate the use of disparate data sources (observations and model predictions) from a wide variety of providers to predict when and where ship traffic should be re-routed to avoid potential interactions with protected marine resources (marine mammal-marine transportation interaction) and to predict the oceanographic and climatic conditions that indicate a hindrance to naval sea basing. For the marine mammal ship interaction application the system has been applied to assess the impact of ship traffic on North Atlantic right whales in Cape Cod Bay. Biweekly data on right whale distributions are based on aerial surveys. US Coast Guard Automated Identification System (AIS) provides historical and real time information on the location of all vessels transiting the area. An embedded ship routing model provides the ability to forecast vessel paths. Supporting environmental data includes real time and historical water level and wind data from monitoring stations, nowcasts and forecasts of currents, salinity, temperature, and meteorological variables from regional and national scale models, and a whale-sighting database. The system allows the user to animate any of the environmental variables for selected areas (windows) and times. It also includes an embedded model to predict the probability that a ship transiting the area will or has struck a whale. For the sea basing application, the demonstration focused on the use of the system to assist in the pre-positioning and continued environmental assessment of the deployment of a maritime force platform in the eastern Gulf of Mexico. The system provides real time access to model predicted sea surface currents, winds, and significant wave forecasts from Navy models of the Gulf of Mexico, to climatological data bases for the same variables, and to observations from NOAA buoy observations.

This paper presents a forecast model to predict summer anoxia in the Chesapeake Bay using preceding spring information of nitrogen and phosphorus loads, water temperature, algae concentration, and freshwater flows. It's based on a regression of over 20 years of observed data. The use of anoxic intensity, a DO-normalized anoxic volume, versus the traditional anoxic volume as the forecasting estimator is described. Anoxic volume can be influenced by partial mixing or other factors that sometimes causes a higher anoxic volume but with actually less intensity in the anoxic condition. This paper recommends using quadratic terms of separate nitrogen and phosphorus loads, and other conditions such as freshwater flows from the upper Bay and water temperature, as regressors. More than 18 regression equations with different combinations of explanatory variables were compared, and the importance of their effects on the anoxia forecast is discussed.

The pilot Coastal Observatory in the eastern Irish Sea (http://coastobs.pol.ac.uk) integrates (near) real-time measurements with coupled models in a pre-operational coastal prediction system. Started in 2002 the initial setup was reported in Proctor et al. The aim is to develop the underpinning science for marine management, focusing on the impacts of storms, eutrophication and the relative importance of events viz-a-viz the mean. Measurements include: in situ vertical profiles of current, temperature, salinity, turbidity, nutrients and chlorophyll; shore-based HF radar measuring waves and surface currents out to 50 km; the Birkenhead-Belfast ferry measuring surface properties; waves; tide gauges; satellite data including infrared (for sea surface temperature) and visible (for chlorophyll and suspended sediment). In cooperation with the National Centre for Ocean Forecasting (NCOF), a suite of nested three-dimensional models (the Proudman Oceanographic Laboratory Coastal Ocean Modelling System - POLCOMS) is run daily, focusing on the Observatory area by covering the ocean/shelf of northwest Europe (at 12 km resolution), the NW European continental shelf (at 7km), the Irish Sea (at 1.8 km) and (soon) Liverpool Bay (at 200–300m resolution). Recently near-real time river discharges have become available for input to the higher resolution models. Nutrient and plankton dynamics are simulated with the ERSEM (European Regional Seas Ecosystem Model) component of POLCOMS. A critical analysis of the approach can now be made based on more than 2 years' operation. In particular the seasonal cycles of physics, nutrients and biology, and their interaction with the tides and coastal discharges is becoming apparent.

The three-dimensional, curvilinear, hydrostatic hydrodynamics model CH3D-Z was modified to include single-grain, cohesive sediment suspended transport and bed transport of fluid mud on a slope. CH3D-Z is a z-plane finite difference model that in this case has 25 layers each 60-cm thick, except the surface layer that is thicker to include all water level variations. Both settled mud and fluid mud processes are included in the new model CH3DZ-FM. A seven-layered bed structure simulates the formation of a surface layer through hindered-settling consolidation of newly-deposited material and, if necessary, tracks the descent of an erosive surface into the bed. Depending on the sediment surface density, cohesive sediment transport is treated either as settled mud (particle erosion and floc deposition) or as fluid mud (fluid entrainment and settling). Wind wave resuspension is an important process to suspended sediment, and winds are used by hydrodynamic and sediment transport sub-models. A model module computes down-slope gravity forces, and density-dependent yield stresses through the fluid mud layer and moves material accordingly. The Atchafalaya River flow averages 9,340 m³/sec and carries about 98 million tonnes of sediment annually forming two deltas in the shallow, micro-tidal Atchafalaya Bay. A 6.1-m-deep by 122-m-wide navigation channel transects the bay and extends some 28 km offshore unconfined by jetties. Fine, cohesive material forms a peak shoal height near the extent of salinity intrusion. Because of funding limitations, annual maintenance dredging of the bar channel by the Corps of Engineers has been limited to about 7.6 million m³, even though project depths are not always maintained. Other operational measures have been tried yet project depths have not been maintained for more than a couple of months each year. A model mesh extending 97 km long-shore and about 64 km cross-shore was developed. The model has about 3,000 surface and 25,000 total cells. Year-long simulations are performed for the period of 2001–2002 when extensive field data were collected. Structural alternatives tested in the model have included: jetties, artifical reefs, sediment traps, channel destratification measures, and a sloped channel reach with a sump. Deeper channel dimensions have also been tested.

A hydrodynamic and hydrologic modeling analysis was conducted to evaluate the feasibility of restoring natural estuarine functions and tidal marine wetlands habitat in the Chinook River estuary, located near the mouth of the Columbia River in Washington. The reduction in salmonid populations is attributable primarily to the construction of a Highway 101 overpass across the mouth of the Chinook River in the early 1920s with a tide gate under the overpass. This construction, which was designed to eliminate tidal action in the estuary, has impeded the upstream passage of salmonids. The goal of the Chinook River Restoration Project is to restore tidal functions through the estuary, by removing the tide gate at the mouth of the river, filling drainage ditches, restoring tidal swales, and reforesting riparian areas. The hydrologic model (HEC-HMS) was used to compute Chinook River and tributary inflows for use as input to the hydrodynamic model at the project area boundary. The hydrodynamic model (RMA-10) was used to generate information on water levels, velocities, salinity, and inundation during both normal tides and 100-year storm conditions under existing conditions and under the restoration alternatives. The RMA-10 model was extended well upstream of the normal tidal flats into the watershed domain to correctly simulate flooding and drainage with tidal effects included, using the wetting and drying schemes. The major conclusion of the hydrologic and hydrodynamic modeling study was that restoration of the tidal functions in the Chinook River estuary would be feasible through opening or removal of the tide gate. Implementation of the preferred alternative (removal of the tide gate, restoration of the channel under Hwy 101 to a 200-foot width, and construction of an internal levee inside the project area) would provide the required restorations benefits (inundation, habitat, velocities, and salinity penetration, etc.) and meet flood protection requirements. The alternative design included design of storage such that relatively little difference in the drainage or inundation upstream of Chinook River Valley Road would occur as a result of the proposed restoration activities.

The Skagit River is the largest river in the Puget Sound estuarine system. It discharges about 39% of total sediment and more than 20% of freshwater into Puget Sound. The Skagit River delta provides rich estuarine and freshwater habitats for salmon and many other wildlife species. Over the past 150 years, economic development in the Skagit River delta has resulted in significant losses of wildlife habitat, particularly due to construction of dikes. Diked portion of the Skagit River delta is known as Fir Island where irrigation practices for agriculture land over the last century has resulted in land subsidence and reduced efficiency of drainage network and impeded fish passages through the area. In this study, a three-dimensional tidal circulation model was developed for the Skagit River delta to assist estuarine restoration in the Fir Island area. The hydrodynamic model used in the study is the Finite Volume Coastal Ocean Model (FVCOM). The hydrodynamic model was calibrated using field data that was collected from the study area specifically for the model development. Wetting and drying processes in the estuarine delta are simulated in the hydrodynamic model. The calibrated model was applied to simulate different restoration alternatives. The results provide guidance for estuarine restoration and management. Specifically, the model was used to help select and design configurations that would improve the supply of sediment and freshwater to the mudflats and tidal marsh areas outside of diked regions and improve the estuarine habitats for salmon migration.

Sarasota County, Florida is undertaking the work necessary to plan, design and permit a project to reopen Midnight Pass. In support of the design studies for the project, a hydrodynamic model study was conducted that served two main purposes: (1) provide a tool to aid the engineering design by evaluation of design alternatives and optimization of the inlet dimensions, and (2) provide a framework for evaluation of potential project induced environmental changes (e.g., salinity, temperature, estuary water residence time, etc.) and identify alternatives with the least adverse impacts and greatest potential benefits. This paper presents the setup, calibration and application of a two-dimensional hydrodynamic model (EFDC) to determine the environmental and scientific merits of this project. The model study area includes an interconnected bay system that includes Little Sarasota Bay, several other bays and four existing inlets (Longboat Pass, New Pass, Big Sarasota Pass, and Venice Inlet). Model calibration and verification comparisons to observed data demonstrate that the model is capable of representing the hydrodynamics and salinity transport of the bay system. Several project alternatives were simulated to evaluate the current velocities in the newly opened inlet, the impacts to the circulation and flushing characteristics of Little Sarasota Bay, and impacts to the stability of nearby inlets also connected to the bay system. Based on an evaluation of each alternative, a preferred alternative with the least adverse impacts and greatest potential benefits was recommended.

The Willamette Falls Hydroelectric Power Dam, operated by Portland General Electric (PGE), is located on the Willamette River, Oregon. The Project site consists of T.W. Sullivan Power Plant and a 2,950-ft-long spillway located on the top of the Willamette Falls Dam. As part of the effort to improve the flow field and enhance the downstream juvenile fish passage in the region just upstream of the forebay (pre-forebay) where a flow control structure at the dam was proposed. The flow in the pre-forebay of Willamette Falls Dam is affected by the complex bathymetry, geometry of the structure, powerhouse flow, fish ladder flow and the spillway around the dam. The expectation was that the flow would be sensitive to the proposed flow control structures and could be modified to enhance downstream migration. In this study, a three-dimensional, free-surface hydrodynamic model (EFDC) was developed for the pre-forebay region of Willamette Falls to evaluate the feasibility of the proposed alternative and its effect on the flow field in two different flow regimes (low and high river flow). The model was also used to assess the hydraulic capacity of flow control structures. One of the challenges in this modeling study was the proper specification of the free open boundary conditions along the 2,950-feet-long spillway. In this study, a pressure boundary condition, calculated using hydraulic rating curves, was applied to the free spillway boundary. The numerical model was calibrated with Acoustic Doppler Profiler (ADP) velocity measurements at 17 stations for the existing condition. The measurement period was in the low flow season. Good agreements between model results and measured data were obtained, indicating the successful application of pressure boundary condition on the spillway boundary. The calibrated model was applied to simulate the flow field and free surface elevation in the high flow region near the control flow structures under different alternative conditions. The model results were used to evaluate the effectiveness of flow control structure alternatives for downstream fish passage. The model was also used to estimate the hydraulic capacity based on the head loss upstream of the structures. This model application demonstrated that a free surface coastal model can be used successfully to examine free surface hydraulic problems near high velocity regions upstream of spillways at dams.

A fully three-dimensional, nested grid, integrated circulation-wave-sediment-geomorphology numerical model, COCIRM-SED, was developed to study sediment dynamics on Roberts Bank, Fraser River foreslope, Strait of Georgia, Canada. Roberts Bank is an extensive intertidal zone, located just south of where the main (South Arm) Fraser River channel enters the Strait of Georgia. The substrate of Roberts Bank is primarily silty sand, and a total of four grain size classes were considered in the model. The study area of interest is modeled using a fine grid size of 100 m by 100 m, and is nested within the much larger domain of the Strait of Georgia with a coarser grid size of 500 m by 500 m. These two model domains are solved together at every time step using a fully dynamic and two-way connection scheme. Following initial testing with idealized analytical cases, the COCIRM-SED modeled flows and suspended sediment concentration were calibrated and verified using the data obtained with a configurable multi-sensor tripod system on the Bank and at the edge of the Bank, respectively. These field data, including month-long ADCP current profiles, OBS sediment concentrations, and directional wave measurements, are used to examine the model performance. The preliminary model results provide improved understandings of the sediment dynamics on Roberts Bank.

The UK government funded Environment Agency (EA) has commenced its Thames Flood Risk Management Strategy investigation and has initiated a series of studies to provide a sound basis to furnish its decision making process. Within the Thames Estuary, cohesive sediments play an important role and understanding their present behaviour and the effects of any flood risk management strategy on them is an important issue to be addressed. The studies have been undertaken in three phases. Firstly an Estuary-wide data collection exercise was done covering the whole 100 km length of the Estuary for a range of tides. The through-tide flow and suspended solids data obtained were processed using the SEDIVIEW technique to calibrate the attenuation of the acoustic backscatter signal from ADCP results to determine suspended solids concentrations and hence sediment fluxes along a measurement transect. These fluxes provided the key calibration data for the establishment of the model. The second phase of the study was a pilot modelling exercise designed to provide a foundation to the third phase of the study, the detailed calibration. The pilot study explored the choice of modelling methodology, key processes and parameters. In particular it was found that even a well mixed Estuary like the Thames which features a prominent turbidity maximum requires resolution in the vertical to establish the correct sediment balance in the model. This conclusion resulted in the choice of the fully coupled mud transport and flow model SEDI-3D, developed by EDF-LNHE as the modelling tool. The third phase of the study was the detailed model calibration including the use of objective statistical methods for assessment of model performance.

Following paper contains a presentation of an integrated model approach combining tidal current and wave impact on a graded sediment transport model. The tidal currents are modelled by using a two-dimensional hydrodynamic numerical model. A coupled wave climate model calculates directional frequency-spectra. In order to describe hydrodynamic and wave influenced sediment transport developments, a graded sediment model is used. The following paper will show that the graded sediment transport model allows a site-specific description of sediment distribution in a bay. The development of different tidal flats, e.g. mud flat and sand flat, can be reproduced. It will be shown that this integrated model approach is capable of simulating the accumulation of fine sediments in shallow areas and bed coarsening in the deeper channels due to tidal currents. Combined impact of tidal currents and waves leads to resuspension of Fine sediments in shallow areas. A part of the resuspended sediments are brought back into the deeper areas of the bay. Tidal current transports fine sediments from deep water regions with high bed shear stress to shallow areas in a bay. In the shallow regions the sediment accumulates based on low bottom shear stress induced by tidal current. Waves have a converse effect in shallow regions in comparison to tidal currents. The combined bottom shear stress can become higher than the current induced bottom shear stress, so that the accumulated sediment is mobilized again. This correlation points out the necessity for simultaneously consideration of both effects.