IntroductionTidal inlets are connections between bodies of water governed by hydraulic efficacies resulting in dynamic changes. The economic, ecologic, and social importance of the inlets demand anthropogenic activities to maintain the location and navigability of existing inlets. These activities often fall into two strategies: maintenance dredging, structural modifications (e.g., jetties, breakwaters), or some combination of both. Historically, the most common solution to inlet stability has been the construction of jetties or other flow-training structures.Tidal inlets are naturally maintained through the scouring action of tidal currents opposing infilling driven by longshore sediment transport. The concept of a minimum velocity required to maintain scouring efficiency was first proposed by Escoffier (1940, 1977). Subsequent research used the concept of the tidal prism as a measure of an inlet’s flushing capacity and corresponding stability in empirical relationships (Bruun and Gerritsen 1960; O’Brien 1969; Jarrett 1976). It is worth noting that this research focused on relatively large tidal prisms, with most of the data clustered around 1.5 × 108 m3. For comparison, the study inlet is two orders of magnitude lower. Alternatively, analytic solutions were developed by Keulegan (1967). However, application of analytic solutions is only appropriate to the immediate vicinity of the inlet throat, a notable shortcoming. Operational requirements necessitate maintenance of the navigational channel both in the inlet throat and the channel extending offshore. Consequently, with advances in computational efficiencies, process-based models are now routinely used to characterize existing inlet morphodynamics (e.g., Elias and Hansen 2013; Reyes-Merlo et al. 2017; Petti et al. 2020) and evaluate potential inlet management solutions (e.g., Shaeri et al. 2017; Fernández-Fernández et al. 2019; Zarzuelo et al. 2019) for both the inlet throat and offshore channel. However, studies are most often associated with either larger commercial channels, open coast, or highly energetic inlets. Small inlets (defined herein as tidal prisms on the order of 106 m3 or smaller) and those serving small-craft harbors are notably underrepresented in existing literature. This work aims to be an incremental step in quantitative comparisons of sedimentation management solutions for small-craft inlets.Study AreaSite DescriptionFortescue Inlet is a secondary inlet connecting Fortescue Creek to the Delaware Bay in southwestern New Jersey (Fig. 1) and is plagued by frequent shoaling. The inlet is used for commercial and recreation vessel traffic through a maintained channel. The channel serves two private marinas as well as the US Coast Guard Search and Rescue Station. The channel has a charted depth of 2.83 m NAVD (−6 ft MLLW) and is maintained by the New Jersey Department of Transportation (NJDOT). Fortescue Inlet is bound by a vegetated marsh (Fortescue State Wildlife Management Area) to the north and a composite rock/timber jetty to the south. The timber structure extends up (east) the creek and transitions into a timber bulkhead, terminating near the boat ramp operated by New Jersey Department of Environmental Protection. This configuration is consistent for the extent of the project area extending east to Downe Ave (Route 637 Bridge).Meteorologic and Hydraulic CharacteristicsThe inlet exhibits a Great Diurnal Range of 2.0 m and mean range of 1.8 m. Consequently, the site can be classified as microtidal (Davies and Moses 1964). Tidally driven currents for microtidal shorelines are not expected to govern shoreline morphology. Tides are semidiurnal with moderate amounts of diurnal inequality (Reeve et al. 2012). The spring and neap tidal prisms are estimated to be 1.0 × 106 and 9.1 × 105 m3, respectively.Waves present at the site include both locally generated wind waves and swell propagating from the Atlantic Ocean. A standalone SWAN model was used to assess the relative magnitude of each, concluding that offshore swell is largely dissipated before reaching the site (Table 1). Offshore swell wave heights are less than 18 cm at the site approximately 99% of the time. Winds are, therefore, the primary driver of waves at the project site. While there are some contributions to wave energy by offshore forcing in the lower Delaware Bay, it is typically dissipated by the project site (Chen et al. 2018).Table 1. Comparison of influence of swell from offshore boundary to local conditions at the siteTable 1. Comparison of influence of swell from offshore boundary to local conditions at the siteHs (m) @ boundaryFrequency of occurrence (%)Hsavg (m) @ siteReduction in wave height (%)0.2548.70.03890.7543.60.09881.256.40.14891.751.10.18902.250.20.21912.750.00.2292The closest NOAA gage with at least 15 years of wind records is located northwest of the project within Delaware Bay at Ship John Shoal (NOAA Station 8537121). The wind climate exhibits prominent seasonal trends (Fig. 2), with winds out of the northwest during the winter (October–March) and out of the south to southeast during the summer (April–September). Average and maximum hourly wind speeds are around 5 m/s (10 knots) and 23 m/s (45 knots), respectively. In general, the strongest winds of the year tend to come out of the northwest. Wind roses for each season are provided next, based on winds reported every 6 min between 2003 and 2018. It should be noted that the speeds provided have been adjusted assuming a logarithmic profile from the instrument height of 20.7 m to a standard 10 m height (CERC 1984). A more detailed frequency and direction analysis of these data is presented in Miller et al. (2020).Historical Infill VolumesThe estimated annual infill volumes are presented herein based on historical survey records. It should be noted that the surveys were performed by others for the period of 2014–2017. Annual infill volumes are estimated to be 5,300 m3 [4,400–6,300 m3] per year. A review of the surveys indicates that approximately 25,000 m3 (net) were removed from the inlet by two separate dredging projects over a 3-year period. As the channel was already partially infilled, the removed volume exceeds the annual infill volumes. Therefore, this provides an upper bound of annual infill rate on the order of 8,000 m3/year. Fig. 3 provides the elevation difference plots for these surveys ranging between 2014 and 2017 with tabulated results in Table 2.Table 2. Calculated volume changes between historical surveysTable 2. Calculated volume changes between historical surveysSurvey descriptionDateCutFillNetCondition12-05-2014———Before dredge12-17-2015−4,4274,422−5After dredge04-15-2016−11,1422,307−8,835Before dredge12-05-2016−1,4055,6694,264After dredge04-17-2017−17,054644−16,410The data reveal two distinct periods to estimate more representative annual sedimentation rates over 12- and 8-month periods. The period from April 2014 through December 2014 indicates net accretion of 4,200 m3; extrapolating would suggest an annual volume of around 6,300 m3. Curiously, net volume changes from December 2014 to 2015 are near zero net change. However, a closer inspection of the difference plots indicates significant accretion (areas of red) in the inlet mouth. This volume is offset by a small (0.1 m) cut volume on the offshore template (light blue). This relatively uniform volume loss in the offshore portion is not seen in any other difference plot. It is hypothesized that this volume may be an artifact of a survey offset error between surveys. These surveys were performed by others; detailed information including survey QA/QC was not available to the authors, making this hypothesis impossible to validate.For the purpose of this study, the net zero volume change for this period (Dec. 2014–2015) is not considered reliable. Instead, the calculated 4,400 m3 fill volume for the inlet throat during this time period is assumed to be more representative. This value is used as a lower bound on the annual infill rates. Consequently, net annual infill volumes within the channel are estimated to be on the order of 5,300 m3/year with infilling primarily occurring within 300 m of the inlet entrance.Historical Dredging CostsAs noted earlier, two separate dredging projects occurred for the period of 2014–2017. Project costs and unit costs ($/m3) are presented in Table 3. It should be noted that NJDOT has made extensive and beneficial use of the dredge spoils in the area. This includes thin layer placement (marsh enhancement), dune reconstruction, and beach nourishments. Unit costs vary based on spoil placement and project volumes. Prices varied between $180/m3 and $77/m3; no adjustment for inflation has been made. It should be noted the thin-layer placement unit price does not include further expenses of $583,106 for the placement of containment booms and preparation of the marsh. Thus, the unit prices are that exclusively of dredging and pumping activities. The large difference in unit price is associated with two factors: amortization of mobilization costs over differing project volumes, and spoil material placement. Spoil placement requiring frequent movement/extension of the discharge pipe will result in lower efficiencies. Comparing similar spoil placement (confirmed through comparable volume/day efficiencies), projects with shorter durations and lower volumes amortize the dredge mobilization costs over fewer cubic meters of material (e.g., March 2016 versus February 2017).Table 3. Dredging cost and project particulars for the period of 2014 to 2016Table 3. Dredging cost and project particulars for the period of 2014 to 2016DateDurationPay vol m3Cost ($)$/m3$/dayVolume (m3/day)Notes3/6/2016104,962893,10018089,310496Marsh enhancement (thin layer placement)3/26/201685,111665,20913083,151639Horseshoe Crab Beach enhancement2/15/20172114,0171,085,5657751,694667Dune reconstruction4/11/2017243656,80413028,402218Bathing beach replenishmentMethodsNumerical modeling was used to characterize sediment pathways and provide quantitative comparisons of the potential sediment management alternatives. A 3-week field deployment in Fall 2018 collected sufficient data to enable the calibration and evaluation of a coupled model. The hydrodynamic and morphology models were independently calibrated. The hydrodynamic (i.e., flow and wave) components were calibrated first, followed by the morphology model. The fully calibrated model was then adopted to further investigate the processes governing morphology change and evaluate the impacts of the selected management alternatives. The followed methodology is similar to that presented in Lemke et al. (2020).Model ConfigurationThe modeling package, Delft3D-FM (Flexible Mesh) (Lesser et al. 2004), was utilized for its ability to simulate the hydrodynamics and morphodyanmics of complex inlet systems (Elias and Hansen 2013; Reyes-Merlo et al. 2017; Shaeri et al. 2017; Fernández-Fernández et al. 2019). A two-way, depth-averaged (2DH) hydrodynamic module was coupled with a wave module, based on SWAN (Booij et al. 1999; Ris et al. 1999), and a sediment transport module. The model domain encompasses Delaware Bay from its entrance at the Atlantic Ocean 90 km northwest to Delaware City, Delaware. Elevations near the project site were interpolated from data collected by the authors at the beginning of the Fall 2018 field deployment using methods similar to those described in previous studies (Miller et al. 2009; Zimmerman et al. 2020). Elevations outside the field deployment’s coverage were interpolated from public sources (CIRES 2014).The composite sheet-pile and rock jetty located on the southern bank of the inlet was modeled as an impermeable structure, blocking the movement of water, waves, and sediment through it. To simplify the model domain, tidal basins were added to the terminal ends of the main channel and a smaller tributary channel to the north. This parameterized the large network of interconnected channels for which the flow particulars were outside the scope of this study. The two basins were sized to properly preserve the inlet’s tidal prism and subsequent hydrodynamics.The model has two open boundaries with one located to the north near Delaware City, Delaware, and the other to the south near the entrance to the Atlantic Ocean. The model was forced with hydrodynamic conditions at both boundaries, and waves were forced only at the bay entrance. Time-varying and spatially constant winds were applied across the entire model domain. Specifics regarding boundary conditions for calibration and applied model runs are described in subsequent sections.Calibration and EvaluationHydrodynamics and WavesCalibration of the coupled hydrodynamic-wave model was completed by comparing the modeled results with data collected during a defined period of the Fall 2018 deployment. Once calibrated, the final configuration was tested for a second defined period. These two periods are depicted in Fig. 5 and represent wind events with elevated wave heights (greater than 0.5 m).For these two periods, the hydrodynamic model was forced at the northern open boundary with a time series of 6-min water levels obtained from the Delaware City NOAA tide station (ID 8551762). The southern open boundary was forced with a spatially varying time series of water levels interpolated between those obtained from the Lewes, Delaware, NOAA tide station (ID 8557380) at the southern point and the Cape May, New Jersey, NOAA tide station (ID 8536110) at the northern point. Wind speeds and directions measured at the Ship John Shoal NOAA station (ID 8537121) were applied across the entire domain.Waves extracted from NOAA’s Wavewatch III Hindcast (NOAA/NCEP/EMC 2019) were applied at the southern open boundary and allowed to propagate into the bay. The boundary condition was applied in the form of a time series of significant wave heights, peak wave periods, and mean wave directions. Sensitivity tests showed that the offshore energy was largely dissipated before reaching the project site. Therefore, the wave forcing was ultimately dropped from the model in subsequent applied runs.Parameters tested during the hydrodynamic model calibration included the water-level boundary conditions, grid configuration, and the Manning friction coefficient. This coefficient was decreased from the default value of 0.023–0.015. The default value resulted in the modeled tides at the project site lagging the observed and in smaller predicted current magnitudes. Parameters tested during the wave model calibration included the bed friction and white-capping formulations. The JONSWAP bed friction dissipation formulation was used with the recommended coefficient for wind-seas (0.067 m2/s3) (Deltares 2020a). The Westhuysen white-capping formulation (Van der Westhuysen 2007) was used as it tends to lead to better performance over the Komen (default) formulation (Komen et al. 1984) in wind-dominated seas where there is some presence of swell (Mulligan et al. 2008; Moeini and Etemad-Shahidi 2009). All other parameters were set to their default values.Modeled hydrodynamic (water levels, current velocities) and wave (significant wave height, mean wave direction) results were compared with data collected at different locations during the field deployment (Fig. 4). A representative subset of these evaluations, including goodness-of-fit statistics (Table 4) and visual comparisons (Fig. 5), are presented. Water levels are predicted well, with errors generally 10 cm or less (5% of the spring tidal range) and lag times 15 min or less. At the midchannel location, both current magnitude and direction are well represented. Peak speeds during flood and ebb tide are similar (0.2–0.3 m/s), while headings shift from 330° to 150° (Fig. 5, fourth panel). At the throat, current speeds are predicted well during flood tide (0.2–0.3 m/s) but are underpredicted by up to 75% during ebb tide (Fig. 5, bottom panel). Directions are well reproduced during both phases. The underprediction in speed at this location during ebb tide is assumed to be the result of a combination of factors, including omission of creek discharge, and inherent simplifications of the 2DH model. Complex flow patterns (i.e., eddies) that are not likely to be fully resolved using a depth-averaged model have been observed at the location where the ADCP was deployed (just to the north of the throat; Fig. 4). This location is further complicated by rapidly changing bathymetry and nearby irregular marsh edge. It is worth noting that other deployment locations that may have better represented conditions near the throat were considered. Unfortunately, site conditions limited the placement of the ADCP to that chosen. Areas further west and south could not be utilized due to insufficient depths at low tide and those within the channel limits could not be used for risk of damage to the instrument by vessel traffic.Table 4. Goodness-of-fit statistics of current velocities for model calibration and validation periodsTable 4. Goodness-of-fit statistics of current velocities for model calibration and validation periodsParameterPeriodBiasMAERLag (min)WLBOAT SLIP (m)Calibration−−2Validation−−16Ux,MID (m/s)Calibration0.000.030.92—Validation−—Uy,MID (m/s)Calibration0.010.050.86—Validation0.010.050.93—Ux,THROAT (m/s)Calibration——Validation−—Uy,THROAT (m/s)Calibration−—Validation−—Hs (m)Calibration−—Validation0.060.100.81—MWD (deg)Calibration−8.326.90.90—Validation−—The model performance at the two locations (midchannel and throat) indicates that the overall behavior of the inlet (ebb/flood) is well represented, while more nuanced flow patterns appearing in specific areas such as that discussed earlier may be unresolved. Any implications of the underpredicted speeds during ebb tide on morphological changes are expected to affect all applied simulations similarly. Thus, they are not expected to influence conclusions regarding the relative effectiveness of each inlet management alternative.Waves are modeled well both in height and in direction. Mean errors for significant wave height are less than 10 cm. Most directions are modeled within 20° of that observed. While measured wave direction is reported as a single mean direction, it truly represents a spectrum with some directional spreading. A review of the collected spectral data indicates that the measured energy is contained within ±30° of the mean wave direction.MorphologyThe final model configuration described earlier was adopted and coupled with a sediment transport module to simulate bed level changes over the 3-week deployment period. For calibration, these modeled bed level changes were compared with those measured at the end of the field deployment. The Bijker (1971) formulation was utilized for its robust reputation and common application in coastal areas (Deltares 2020b). Bijker (1971) models bedload and suspended load separately and includes the influence of waves. Parameters with physical meanings that could be measured directly or estimated from calculations were adjusted first, followed by unitless calibration parameters. The physical parameters included grain size, density, and setting velocity. Unitless parameters scale current induced suspended and bedload transport, wave transport, and parameterize shoreline erosion. Table 5 presents a summary of the selected parameters, tested range, and default values.Table 5. Delft3D-SED model parametersTable 5. Delft3D-SED model parametersParameterAbbrev.Transport modelSelected value(s)Tested rangeDefault valueRangeCurrent-related reference concentration factorSusFlow21–810–InfCurrent-related transport vector factorBedFlow21–810–InfFactor for erosion of adjacent dry cellThetSDFlow0.20–0.200–1Median sediment diameter (mm)D50Flow + Wave0.21 & 0.550.2–0.60.2>0Sediment specific density (kg/m3)ρFlow + Wave2,650 & 2,5002,500–2,6502,650>0Settling velocity (m/s)ωWave0.02 & 0.050.01–0.060.010–InfShallow water coefficientb (shallow)Bijker (1971)65–65>0Deep water coefficientb (deep)Bijker (1971)222>0Two sediment fractions, with median sediment diameters of 0.21 and 0.5 mm, were used, supported by analysis of grab samples and review of literature (Flynn 1999; Jackson 1999). Sediment specific density was varied as a calibration parameter arriving at densities of 2,650 and 2,500 kg/m3, respectively. These values are within the expected range for quartz and feldspar sediments. Settling velocity was estimated using both the Hallermeier (1981) and Gibbs et al. (1971) formulations.Due to the relatively small bed level changes that occurred during the field deployment and limitations of the surveying techniques in discerning those changes, quantitative assessment of the model performance by a skill score or RMSE is inappropriate. Assessment of the model was therefore made by qualitatively comparing the sediment transport patterns observed and those modeled. An assessment of annualized volume errors is discussed in a later section (Assessment of Model and Climate Schematization Errors). Fig. 6 illustrates the modeled and observed elevation changes during a 3-week period. This highlights the model’s ability to accurately recreate sediment transport patterns observed during the Fall 2018 deployment. These patterns are highlighted by the white circles in the figure and include: (1) erosion of the south beach, (2) infilling along the northern side of the channel, (3) movement of the southern ebb shoal bar, and (4) erosion along the north side of the inlet throat paired with accretion along the southern limit of the inlet throat. The maximum elevation changes of 0.3 m throughout the project site are reasonable for the modeled timeframes and match the field observations, particularly with respect to the movement of the ebb shoals.Applied Model/Modeled ScenariosThe calibrated model was modified to assess the prospective inlet management techniques, each under a set of representative conditions. These conditions, defined by a set of wind- and water-level forcing, were selected to account for observed seasonal patterns. They included Calm (i.e., tidal only), and Typical and Storm conditions associated with wind directions dominant during the winter and summer months. These five conditions enable quantification of the dominant sediment pathways, evaluation of existing flushing capacity, and determination of any seasonal variations that may influence management strategy.Each condition was modeled for 7 days with a morphology change multiplication factor (MORFAC) of 4. Results, therefore, represent expected morphology change under the modeled conditions for 28 days. The MORFAC value was selected to take advantage of the model’s ability to improve computationally efficiencies and reduce required model run times. The value is low enough so ensure that (1) the model is computationally stable (Deltares 2020b) and (2) the scale of morphologic changes do not substantially alter the hydrodynamics. In addition, since each modeled scenario utilizes a constant wind velocity, the results are not sensitive to timing with tidal signal. Sensitivity runs with MORFAC values up to 7 were tested and indicate consistent results between MORFAC and real-time simulations.The selected modeling approach characterizes the effectiveness of each sediment management alternative by comparing the amount of material deposited within the channel limits during each condition to a baseline scenario. A yearly sedimentation rate is provided by estimating how often each condition is likely to occur each year and extrapolating the modeled volumes. An advantage of this approach is in the ability to test sensitivity of the yearly sedimentation volumes to the “storminess” of any given year.Wind speeds were selected to represent the typical and storm conditions based on a detailed analysis of wind speeds and directions at Ship John Shoal. A summary of the selected wind speeds for each condition is presented in Table 6. The Storm condition represents the average of the top 15% of wind speeds from a given direction. The Calm condition represents both winds that are: (1) less than 4 m/s from any direction and (2) greater than 4 m/s but from a direction without fetch. The Typical condition represents the average of those not considered in the other two scenarios. A combination of empirical formulas and a standalone wave model was used to confirm the use of a 4 m/s threshold and relevant directions for Calm conditions. The use of “Storm” is used for convenience and is not used in a strict definition (e.g., Beaufort Scale). Rather, it is used to indicate periods of elevated wind speed and, therefore, elevated wave-induced sediment transport. The storm wind speeds are determined statistically, corresponding to conditions occurring less than 5% of the time, irrespective of other meteorological conditions.Table 6. Applied model wind velocity forcingTable 6. Applied model wind velocity forcing
ConditionPercent occurrence (%)Days/
yearWind directionWind speed (m/s)Calm35125n/a<4Calm (no waves)1966SE – WSW>4Winter – typical1762292.5 (WNW)6.5Winter – storm310292.5 (WNW)11.6Summer – typical1346157.5 (SSE)5.9Summer – storm28157.5 (SSE)9.7Tidal forcing of the applied model at the open boundary was handled by creating a representative tide signal fluctuating between mean lower low water (MLLW) and mean higher high water (MHHW), and mean low water (MLW) and mean high water (MHW). This signal was generated for each of the model forcing locations, with appropriate offsets for lag based on the harmonic constituents of each station. Use of a representative tide was found to have negligible impacts on the average peak flood and ebb tide current velocities.Modeled Inlet Management ScenariosSix scenarios were assessed under applied conditions (Table 7; Fig. 7). These included a baseline (dredging as designed), two modifications to the existing jetty (extension and relocation), two modifications to the dredge template, and partially infilled channel. For the first five scenarios, the initial bathymetry reflected a post-dredge surface where the channel was fully dredged to the design template. For the deposition basin scenarios, additional areas were dredged in accordance with the description in Table 7. Each inlet management alternative was selected for its potential in interfering with the existing sediment pathways and/or modifying the channel configuration to constrain flow and improved flushing capacity. The last scenario (partial infill) was chosen to illustrate the potential consequences of leaving the channel partially filled by either not dredging or incomplete dredging (e.g., not dredging to the full design template). Historical records indicate that the channel has been partially dredged on several occasions. The partial infill scenario utilized bathymetry consistent with that observed at the site in November 2018.Table 7. Selected alternatives for modelingTable 7. Selected alternatives for modelingNameApproachBrief descriptionObjectiveBase—Channel fully dredged to existing Dredge template (min. depth −2.8 m NAVD)As designed approachPartial infill—Partially infilled channel; utilizes 2018 bathymetryEvaluate (potential) consequences of not dredging or not dredging to full templateJetty ExtensionStructuralDouble the length of the existing jetty along current heading, for a total length of 180 mAssess viability of extending the length of existing jettyJetty RelocationStructuralRealign jetty with length of 90 m to match that of existingAssess viability of narrowing the inlet throatUpland Deposition BasinDredging Management (soft solution)Dredge area of 12,000 m2 on dry beach to −2.8 m NAVD resulting in a volume of 24,000 m3 removedAssess ability of an onshore/upland deposition basin to interfere with longshore transportOffshore Deposition BasinDredging Management (soft solution)Dredge areas of 8,000 m2 (south) and 16,000 m2 (north) to −2.8 m NAVD resulting in volumes removed of 8,000 and 16,000 m3, respectively.Assess ability of offshore deposition basins to interfere with sediment bypassingResultsResults of the applied model runs are presented in both visual and tabulated volume changes. First, the graphic model output for a select number of conditions are presented. Figs. 8 and 9 present elevation changes under SSE and WNW storm conditions occurring over 28 days. Changes during typical conditions (Supplemental Materials—Figs. S1 and S2) present similar patterns as their storm counterparts. The magnitude of the changes during typical conditions however are reduced by 80%–90% from storm conditions.In addition to these visual comparisons, volume changes within the channel limits are calculated to provide a quantitative comparison. Tables 8–10 present annual estimations of fill, cut, and net volume changes within this area, respectively. These estimations were made by scaling the volumes calculated by the raw model results, which represent changes over 28 days. This scaling was performed by first converting these volumes to a volume rate per day and then multiplying by the corresponding number of days for which each environmental condition occurs during a typical year (Table 6). Fill, cut, and net volumes are presented not only to illustrate the net volume changes but also to capture the magnitude of total sediment transport. While a brief description of the results is provided here, a more detailed analysis is provided in the subsequent section.Table 8. Annualized fill volume estimate within channel limits (m3)Table 8. Annualized fill volume estimate within channel limits (m3)Design alternativeTidalSSE stormSSE typicalWNW stormWNW typicalTotalBase4901,3808602,7501,0906,570Partial infill3001,4001,2603,0701,6007,630Jetty extension7401,2208602,7601,2606,840Jetty relocation4401,3709402,7801,1806,710Upland deposition basin4801,2208802,7401,0806,400Offshore deposition basin4803904209306302,850Table 9. Annualized cut volume estimate within channel limits (m3)Table 9. Annualized cut volume estimate within channel limits (m3)Design alternativeTidalSSE stormSSE typicalWNW stormWNW typicalTotalBase−920−460−550−430−570−2,930Partial infill−420−750−960−1,110−1,090−4,330Jetty extension−870−430−510−450−580−2,840Jetty relocation−710−430−520−500−520−2,680Upland deposition basin−920−470−560−440−560−2,950Offshore deposition basin−910−620−550−500−550−3,130Table 10. Annualized net volume change within channel limits (m3)Table 10. Annualized net volume change within channel limits (m3)Design alternativeTidalSSE stormSSE typicalWNW stormWNW typicalTotalBase−4309303102,3105203,640Partial infill−1306503001,9605103,290Jetty extension−1307803502,3206804,000Jetty relocation−2709404202,2806504,020Upland deposition basin−4407503202,3005203,450Offshore deposition basin−420−230−13044080−260Calm Conditions (Tidal Only)Changes during calm (tidal only) conditions (Supplemental Materials—Fig. S3) are minimal. The net negative values for the base scenario indicate that under ideal conditions, minor flushing does occur within the inlet (Table 10). However, this flushing represents a relatively small percentage (less than 10%) of the total annual infill volumes. Flushing efficiency drops of notably when considering the partially infilled bathymetry.South–Southeast (Summer) ConditionsSSE (typical of summer) conditions result in movement of the shoal to the south of the inlet northward into the channel limits (Fig. 8). The existing jetty (in the baseline scenario) blocks some of this sediment moving parallel to the shoreline. However, as most of the transport occurs offshore of the existing structure (up to 300 m offshore), there is significant bypassing. All four management alternatives interfere with sediment movement during SSE conditions. Both structural options and the upland deposition basin provide little improvement over the existing structure. The jetty relocation and upland deposition basin exhibit similar behavior to the existing jetty in that they are largely bypassed. The jetty extension is also largely bypassed. Areas of accretion are pushed 60 m further offshore, but due to increased scour at the jetty’s tip volume, accumulation is like the baseline scenario. The offshore deposition basin effectively captures sediment before it reaches the channel limits and reduces (by 65%) sedimentation within the channel limits.West–Northwest (Winter) ConditionsMovement of the southern shoal during WNW (typical of winter) conditions is depicted opposite to that which was observed during SSE conditions (Fig. 9). Sedimentation within the channel limits is the result of southward movement of the northern shoal. Sediment already present within the channel under a partial infill scenario moves further southward, resulting in shoaling in the middle of the channel. The only alternative that interferes with this condition is offshore deposition basins. The northern basin captures most of the sediment before it moves into the channel limits, reducing fill volumes by 60% from the baseline scenario.Estimated Elevations after 1 YearFinal bathymetries after 1 year are generated for four select scenarios and presented in Fig. 10 (baseline dredging, partial infill, offshore deposition basins, and upland deposition basin). This is performed by scaling the five sedimentation patterns (i.e., Figs. 8 and 9 and those not shown) in a way that is conceptually similar to the methodology utilized to create the annualized volume tables described earlier. The daily rate for each condition was first multiplied by the annual frequency to determine annual elevation changes associated with each condition. Then the elevation changes associated with each of the five conditions were together superimposed on the initial bathymetry.Regarding Model Assumptions and Implications to ResultsOne notable simplification of the presented methodology is its lack of consideration for the sequencing of events. All model runs for a single alternative start from the same initial bathymetry, meaning that changes in bathymetry due to one condition are not reflected in the hydrodynamics of subsequent conditions. The most notable consequence is that the areas of accretion and erosion are likely to be distributed over an area larger than that depicted. This simplification is expected to be valid where the ratio of elevation changes to the water depth is small. In areas that become increasingly shallow, where elevation change approaches that of the water depth, the sequencing of events and feedback between morphologic and hydrodynamic changes are expected to be nontrivial. Therefore, while results are expected to be useful to assess offshore bypassing, they should be used with caution nearshore (shoreline and upstream in the vicinity of the state boat ramp).The method assumes that the sedimentation rate is constant. Morphology and hydrodynamics of inlets are nonlinear. However, comparison of the superposition method and best estimates of the observed annual infill rates show reasonably good agreement. This is expanded in the following section, “Discussion.” For the purposes of this study, it was therefore assumed that this simplification is appropriate. This simplification is limited to the project site and only at the temporal scales and typical (i.e., nonextreme) environmental conditions considered in this paper. It is likely as the time period of interest increases, or modeled conditions become more extreme (longer return period wind speeds), the methodology would need to be reassessed to consider nonlinear sedimentation rates.DiscussionAssessment of Model and Climate Schematization ErrorsAssessment of the model and climate schematization is completed using two comparisons. The first is the ability to capture sediment transport patterns (i.e., ability to recreate the sediment pathway vectors); the second is the ability to predict annual quantities that are consistent with historical records. To the first point, the model is shown to be capable of reproducing sedimentation patterns observed during the Fall 2018 deployment, including shoaling of the northern side of the channel, which is typical of NW wind conditions. Applied baseline model runs utilizing the developed climate schematization for NW conditions illustrate results (i.e., locations of sediment transport) that are consistent with both the calibration runs (Fig. 6) and historical observations (Fig. 3). Namely, there is agreement in the location of bar bypassing, resulting in shoaling of the channel occurring up to 300 m offshore, erosion of the southeastern beach, and sediment bypassing the existing jetty between the calibration, applied model runs, and historical dredge records.To the second point, the model estimates the annualized deposition volumes within the channel limits to be 3,600 m3, with an expected range between 3,000 and 4,800 m3/year. This range was generated through application of the methodology to 16 years of wind velocity data (2003–2018). It is notable that the mean value is roughly 30% lower than the estimated historical observations of 5,300 m3 [4,400–6,300 m3] for the period of 2014–2016 cited earlier.This difference is at least partly explained by differences between sedimentation in any specific timeframe to average conditions. The historical rate (5,300 m3/year) is based on records for a relatively narrow 3-year period (2014–2016). The modeled rate of 3,600 m3/year, on the other hand, is based on median frequencies (Table 6) developed from the analysis of 16 years of wind data. Volume changes during individual years are larger or smaller, depending upon a given year’s storm totals. The yearly variance in sedimentation is associated with changes to intensity of the WNW winds associated with the winter months (Fig. 11). It should be noted that historical wind records at nearby NOAA stations have significant periods of unavailable data. The continuity of data from an annual and seasonal perspective is presented in the bottom pane of Fig. 11. The authors attempted to find nearby stations with complete records; however, all NOAA stations had similar anomalies in the wind record.Assessment of the model and climate schematization error was made by comparing the modeled and observed volume changes for the period of April 2016 to December 2016. Wind records at Ship John Shoal (NOAA 8537121) were not continuous for this period. Thus, wind records from a nearby weather station operated by Rutgers University (2020) were utilized. Corrections were made for instrument height, and a correlation coefficient was made to relate measurements at the Rutgers station to those at Ship John Shoal. The model estimates a volume of 3,100 m3 for the period with measured volumes of 4,200 m3. This suggests that model and climate schematization have errors underestimating the sediment volumes of about 25%. Review of the historical record for the same period (April–December) suggests that typical sedimentation values are between 1,800 and 2,500 m3 confirming that the period of 2016 exhibited higher-than-average wind speeds and the ability for the model to capture increases in sediment transport rates.As the observed and annual values are on the same order of magnitude, the model and climate schematization method has been deemed acceptable by the authors for the purpose of this study. While it is conceivable the model may underestimate the annual sediment transport, any error is expected to be consistent across conditions and scenarios. Therefore, conclusions drawn from the results about the relative effectiveness of each management method will not be impacted.The volumes presented are calculated within the existing channel limits. Visual inspection of the cumulative sedimentation and erosion results (e.g., Figs. 8 and 9) indicate significant volume of sediment residing immediately outside the channel limits and thus outside the tabulated volumes presented. For the relatively short time scales modeled, this volume is outside the channel limits. However, for timescales greater than a year, it is expected to eventually reside in the channel limits. Consequently, the analysis may benefit from calculating volumes over an area larger (in the longshore direction) than that depicted. Analyzing volume changes over a larger area accounts for sediment deposited immediately adjacent to the channel. In addition, this larger area helps eliminate bias by including additional model cells and reducing the sensitivity to individual cells.To quantify this impact, volume calculations were performed to a larger area of interest. Fill, cut, and net volumes were calculated on an area with twice the width of the dredge template, as depicted in Fig. 7 (total width 200 m). Fill and cut volumes were found to be three to four times that within the dredge template. Net volumes were found to be roughly the same magnitude as that within the dredge template for most scenarios. The notable exceptions are the offshore deposition basins, the net volume change of which approached that of the baseline conditions, indicating the basins were capturing sediment bound for the channel. The sensitivity analysis indicates that the results and conclusions based on the presented volumes (Table 10) are not the result of model anomalies.Sediment PathwaysInfilling is driven primarily by offshore-bar bypassing with pronounced seasonal patterns in sedimentation based on wind direction. As depicted in the individual erosion and sedimentation maps for WNW and SSE storm conditions, these wind/wave directions tend to work against each other, with areas of fill in one being areas of erosion in the other and vice versa. It is the balancing of the two over the course of the year that determines where the offshore shoals migrate and how much sediment is deposited within the channel limits.The final bathymetry based on annualized sedimentation patterns after 1 year (Fig. 10) helps visualize the combined impacts of the five conditions. Areas of shoaling within the channel limits is present to the northern side of the channel offshore and on the channel’s southern side closer to the existing jetty. The partial infill scenario illustrates the potential consequences if the channel were allowed to continue to infill without regular dredging or with incomplete dredging. Fig. 10 shows that after 1 year, the deeper section of channel near the inlet entrance is shoaled in. Differences between the partial infill scenario and the baseline are approximately 0.5 m. This shoaling is a result of: (1) the southerly movement of sediment located in the center of the channel, and (2) the northerly movement of sediment bypassing the existing jetty. Historically, dredging projects have focused on specific sections of channel rather than the entire template. Based on this result, it is recommended that, whenever possible, complete dredging to the design template should be completed. Partial dredging only the most problematic areas has been shown to result in decreased flushing efficiencies and increased rates of sedimentation.Comparison of Inlet Management SolutionsThe applied model runs were utilized to assess the effectiveness of both structural and nonstructural solutions in mitigating sedimentation within the channel. Performance was assessed by comparing the results of the modified scenarios with the baseline. Comparisons are made with the aid of the annualized volume changes, individual sedimentation maps, and estimated final bathymetries after 1 year.Total net volume changes indicate that the offshore deposition basins provide the greatest improvement over the other three alternatives. This conclusion is explained through two primary factors. First, the set of offshore deposition basins (one on each side of the channel) is the only solution to interfere with both sediment pathways. The other three solutions only address SSE conditions. Second, for SSE conditions, the southern offshore deposition basin, whose influence extends further offshore, is more effective than the other three alternatives, whose influence is restricted to the nearshore. For a more complete discussion of these two ideas, performance of the alternatives under individual conditions is explored later.Under the existing conditions, flushing ability is limited. Cumulative net volume changes during tidal conditions accounts for less than 10% of the total magnitude infill volumes. Neither the proposed deposition basins (upland/offshore) nor the jetty extension modify the cross-sectional area of the inlet. Therefore, they are not expected to exhibit improved flushing ability over the existing baseline scenario. This is confirmed through the tidal-only condition where volume changes for the four scenarios are similar in magnitude (Table 10). The jetty relocation, which does modify the cross section just offshore of the inlet’s entrance, is not effective in increasing flushing ability.SSE conditions results in offshore-bar bypassing of the shoal located south of the channel. The tested structural solutions (jetty relocation and jetty extension) are largely ineffective in preventing this bypassing. The jetty relocation behaves similarly to the existing structure, with channel encroachment occurring within the vicinity of the inlet throat. While the jetty extension pushes this encroachment further offshore, volume changes remain similar (10% reduction in fill volumes). While longer jetties could be tested, the jetties would likely need to be prohibitively long (both in terms of economics and permitting) to block all bypassing. Structural solutions such as these are likely to provide only a temporary solution. Bypassing can be expected to continue once sediment builds up behind the new or extended structure.Upland deposition basins are predicted to be effective in capturing longshore movement that would bypass the existing jetty. Initially, a small upland deposition basin (4,000 m3) was tested but was found to be ineffective. A larger and deeper one (24,000 m3), presented here, was then modeled. The location was chosen to permit the use of land-based equipment for the removal of the material, providing potential cost savings over their offshore-based counterparts. The results indicate that the upland deposition basin is only moderately effective at interfering with sediment bypassing (10% reduction) due to SSE conditions. While it captures some longshore movement near the shoreline, it is unable to interfere with that occurring offshore.It is worth noting that the effectiveness of the upland deposition basin may be underestimated due to model limitations. The governing equations of Delft3D do not support reliable shoreline accretion modeling. Specifically, parameterization of morphology in adjacent wet/dry cells is distributed at a user-defined ratio and can only erode dry cells but not accrete. The strength of the model is robust hydrodynamics, which is more appropriately applied in primarily wet areas (i.e., offshore basins). This assessment was confirmed with model experts (C. M. Nederhoff, personal communication, May 4, 2020). It is suggested that further analysis be completed to provide a better assessment at an upland deposition basin’s performance. However, as most of the bypassing occurs further offshore, it is predicted that the upland deposition basin alone is unlikely to provide substantial improvements.No alternatives other than the offshore deposition basins were tested that address sedimentation due to WNW conditions. While it is conceivable that a jetty constructed on the northern edge of the inlet may capture some of this material, it is likely to be problematic from an environmental-permitting perspective, as the northern edge of the inlet encompasses tidal wetlands. Bypassing occurs well offshore of the inlet entrance and the structure would need to be prohibitively long (order of magnitude 500 m) to interfere with sediment movement. Installation is also unlikely to increase velocities enough to improve tidal flushing in this wave-dominated inlet. Therefore, it was not considered for the modeled scenarios.Both the southern and northern offshore deposition basins are shown to be effective in trapping offshore-bar bypassing and limiting sedimentation within the channels. Fill volumes due to SSE conditions are reduced by 65% from the baseline scenario, while those due to WNW conditions are reduced by 60%. Fig. 10 depicts the impact of this capture on the annualized bathymetry. Shoaling is restricted to the limits of the basin and does not reach the channel limits. In the northern basin, elevations reach −2.3 m NAVD, while those in the southern basin reach −1 m NAVD. It is acknowledged that the final elevations depicted here, particularly in the southern basin, are likely unrealistically high due to simplifications within the methodology as noted in an earlier section, “Regarding Model Assumptions and Implications to Results.” The accumulated volumes are likely to be dispersed over a wider area. As the area shoals, increased shear stresses would mobilize this sediment. This difference may adversely affect any estimations in the longevity of the basins. While estimations may be provided, ultimately further study including a pilot program is recommended to monitor and optimize these basins.Preliminary Basin Design ConsiderationsAlthough optimization has not been performed, initial results suggest that the effectiveness of the two offshore deposition basins is dependent on their size. In this study, the two basins were placed in the observed infilling locations under the baseline scenario. This length corresponds to the cross-shore length of the existing ebb shoals. Two basins are simulated, a northern one with a cross-shore length of 380 m (width 42 m) and a southern one with a length of 260 m (width 30 m). Depths were set to be identical to the template dredge depth (−2.8 m NAVD). Volumes removed to create the two were 16,000 m3 (north) and 8,000 m3 (south). These volumes are nontrivial, being approximately four times the annual fill volumes. In total, the volume is 1.5–2 times that dredged during previous projects (see Historical Infill Volumes). Note that these cited projects resulted in partial dredging of the channel and not to the full template.Potential Cost SavingsAs the offshore deposition basins do not influence the inlet flushing capacity, the absolute deposition rate is unchanged. Only the location is changed, potentially increasing the interval between dredging operations but not changing the total volume. It should be noted that the total volume of dredge material will likely increase, as effectiveness of the deposition basins will not be 1 for 1. Thus, the savings are really limited to amortization of mobilization costs; trading increased volume at lower unit prices.The annual sedimentation rate of 5,300 m3 is a comparable volume with the March 2016 operation. Assuming a comparable unit price of $130/m3, annual dredging expenditures of $689,000 annually or $2.1 million over a 3-year period could be expected. Assuming 3 years’ worth of material would be removed in one operation would result in 15,900 m3. This is comparable with the February 2017 dredging operation and would suggest unit prices around $77/m3 and total 3-year project costs of $1.2 million, representing significant savings of about 40%. Inefficiencies of the deposition basins likely require annualized volumes greater than the annual sedimentation rate (i.e., 5,300 m3/year). The question then becomes: how much more volume can the project absorb at a lower unit price while still resulting in a lower total cost? Using the historical unit prices, the offshore depositions basins could pump up to 65% more total dredge volume and still provide some, albeit nominal, savings when compared with annual dredging. The modeled offshore deposition basins presented herein represent a total dredge volume of approximately 24,000 m3, or roughly 50% more total volume. Assuming that the dredging interval is extended to 3 years, this would provide a savings of 11% or $219,000. At 25% more volume (approximately 20,000 m3), the savings still represent savings on the order of $540,000 over 3 years (26%).ConclusionsThe model results and observations indicate that the sediment transport near Fortescue Inlet is dominated by wave action with significant ebb shoal bar bypassing moving in both north (summer) and south (winter) directions. The site experiences pronounced seasonal variations in the dominant wave direction. Sediment transport occurs well offshore (approximately 300 m offshore) and beyond the existing jetty. The results indicate that structural modifications (i.e., relocation or lengthening) of the existing jetty do not improve the sedimentation rate within the channel. Instead, changes to the operational dredging or what the authors have termed broadly as “adaptive dredging techniques” appear to deliver the most promising improvement. Therefore, the recommended, preferred alternative is the construction of offshore deposition basins located both north and south of the existing channel.The viability of offshore deposition basins at this site is attributed to the relatively shallow depths extending well offshore. The depths outside of the channel are well above the closure depth, with pronounced shoals/bars that fluctuate with the seasonal variation in wave direction. Sediment transport, and therefore channel infilling, occurs well offshore and outside the immediate influence of the existing impoundment jetty. While not definitive, initial sensitivity testing unsurprisingly suggests that the efficacy of the offshore deposition basin is directly related to the size. At minimum, the guidance from this case study suggests that the planform of the deposition basins should extend cross-shore for a distance of at least the width of the ebb shoal(s). Dredge volumes should be the same order of magnitude or larger than the estimated fill volumes. The use of offshore deposition basins will likely increase the volume of dredge material but can provide potential savings through amortization of dredge mobilization costs, with savings estimated to be greater than 10% over existing strategies.References Bijker, E. W. 1971. Littoral drift computations on mutual wave and current influence. Delft, Netherlands: Delft Univ. of Technology. Booij, N., R. C. Ris, and L. H. Holthuijsen. 1999. “A third-generation wave model for coastal regions: 1. Model description and validation.” J. Geophys. Res.: Oceans 104 (C4): 7649–7666. CERC (Coastal Engineering Research Center). 1984. Shore protection manual. Vicksburg, MS: Waterways Experiment Station, Corps of Engineers. Chen, J. L., D. K. Ralston, W. R. Geyer, C. K. Sommerfield, and R. J. Chant. 2018. “Wave generation, dissipation, and disequilibrium in an embayment with complex bathymetry.” J. Geophys. Res.: Oceans 123 (11): 7856–7876. CIRES (Cooperative Institute for Research in Environmental Sciences). 2014. Continuously updated digital elevation model (CUDEM)—1/9 Arc-second resolution bathymetric-topographic tiles. Washington, DC: NOAA, National Centers for Environmental Information. Deltares. 2020a. Delft3D-WAVE user manual. Delft, Netherlands: Deltares. Deltares. 2020b. D-Morphology, 1D/2D/3D, user manual. Delft3D FM suite 2020. Delft, Netherlands: Deltares. Elias, E. P. L., and J. E. Hansen. 2013. “Understanding processes controlling sediment transports at the mouth of a highly energetic inlet system (San Francisco Bay, CA).” Mar. Geol. 345: 207–220. Escoffier, F. F. 1940. “The stability of tidal inlets.” Shore Beach 8 (4): 3. Escoffier, F. F. 1977. Hydraulics and stability of tidal inlets. Fort Belvoir, VA: USACE. Fernández-Fernández, S., C. C. Ferreira, P. A. Silva, P. Baptista, S. Romão, Á Fontán-Bouzas, T. Abreu, and X. Bertin. 2019. “Assessment of dredging scenarios for a tidal inlet in a high-energy coast.” J. Mar. Sci. Eng. 7 (11): 395. Flynn, M. J. 1999. Sediment characterization of the New Jersey shoreline: Comparison of median grain size distribution from 1950 to 2011. Galloway, NJ: The Richard Stocton College Coastal Research Center. Gibbs, R. J., M. D. Matthews, and D. A. Link. 1971. “The relationship between sphere size and settling velocity.” J. Sediment. Res. 41 (1): 7–18. Jackson, N. L. 1999. “Evaluation of criteria for predicting erosion and accretion on an estuarine sand beach, Delaware Bay, New Jersey.” Estuaries 22 (2): 215–223. Jarrett, J. T. 1976. Tidal prism-inlet relationships. Vicksburg, MS: US Army Corps of Engineers. Keulegan, G. H. 1967. Tidal flow in entrances. Committee on tidal hydraulics. Vicksburg, MS: U.S. Army Corps of Engineers. Lemke, L., M. S. Janssen, and J. K. Miller. 2020. “Mitigation of channel shoaling at a sheltered inlet subject to flood gate operations.” J. Mar. Sci. Eng. 8 (11): 865. Lesser, G. R., J. A. Roelvink, J. A. T. M. van Kester, and G. S. Stelling. 2004. “Development and validation of a three-dimensional morphological model.” Coastal Eng. 51 (8–9): 883–915. Miller, J. K., L. Lemke, and M. S. Janssen. 2020. Fortescue inlet sediment management study. Prepared for New Jersey department of transportation. Hoboken, NJ: Stevens Institute of Technology. Miller, J. K., A. Mahon, and T. Herrington. 2009. Development of the Stevens dynamic underwater coastal surveying (DUCKS) system. Davidson Laboratory Technical Report. Hoboken, NJ: Coastal Protection Technical Assistance Service. Moeini, M. H., and A. Etemad-Shahidi. 2009. “Wave parameter Hindcasting in a lake using the SWAN model.” Sci. Iran. 16 (2): 156–164. Mulligan, R. P., A. J. Bowen, A. E. Hay, A. J. van der Westhuysen, and J. A. Battjes. 2008. “Whitecapping and wave field evolution in a coastal bay.” J. Geophys. Res. 113 (C3): 1–16. Petti, M., S. Bosa, S. Pascolo, and E. Uliana. 2020. “An integrated approach to study the morphodynamics of the lignano tidal inlet.” J. Mar. Sci. Eng. 8 (2): 77. Reeve, D., A. Chadwick, and C. Fleming. 2012. Coastal engineering: Processes, theory and design practice. New York: Spon Press. Reyes-Merlo, M. A., M. Ortega-Sanchez, M. Diez-Minguito, and M. A. Losada. 2017. “Efficient dredging strategy in a tidal inlet based on an energetic approach.” Ocean Coastal Manage. 146: 157–169. Ris, R. C., L. H. Holthuijsen, and N. Booij. 1999. “A third-generation wave model for coastal regions: 2. Verification.” J. Geophys. Res.: Oceans 104 (C4): 7667–7681. Rutgers University. 2020. NJ weather network. Piscataway, NJ: Office New Jersey State Climatologist. Shaeri, S., R. Tomlinson, A. Etemad-Shahidi, and D. Strauss. 2017. “Numerical modelling to assess maintenance strategy management options for a small tidal inlet.” Estuarine Coastal Shelf Sci. 187: 273–292. Van der Westhuysen, A. J. 2007. Advances in the spectral modelling of wind waves in the nearshore. Delft, Netherlands: Delft Univ. of Technology. Zarzuelo, C., A. López-Ruiz, and M. Ortega-Sánchez. 2019. “Evaluating the impact of dredging strategies at tidal inlets: Performance assessment.” Sci. Total Environ. 658: 1069–1084. Zimmerman, T., K. Jansen, and J. Miller. 2020. “Analysis of UAS flight altitude and ground control point parameters on DEM accuracy along a complex, developed coastline.” Remote Sens. 12 (14): 2305.

Source link

Leave a Reply

Your email address will not be published. Required fields are marked *