Report_6_Hydraulic.pdf

Report of Phase III: Hydraulic and Scour Analysis

2D Hydraulic Analysis for FOPU 10(3)

Fort Pulaski Bridge over South Channel of the Savannah River

Savannah, Georgia

Prepared for:

Prepared By:

April 2015

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | i

Table of Contents

Section 1 ‐ Introduction

1.1 Background

1.2 Design Criteria

1.3 Quality Assurance

Section 2 ‐ Data Collection

2.1 Field Investigation

2.2 Tidal Data

2.3 Storm Surge Data

2.4 Mapping and Other Model Data

Section 3 ‐ Hydrology

3.1 Riverine Flows

3.2 Storm Surge Hydrograph

Section 4 ‐ Hydraulic Modeling

4.1 Model Mesh Development

4.2 Model Calibration

4.3 Model Boundary Conditions

4.4 Model Simulations

Section 5 ‐ Scour Analysis

5.1 Soils Data

5.2 Scour Components

Section 6 ‐ Wave Force Analysis

6.1 Design Cases

6.2 Submersible Superstructures

6.3 Fetch and Wind Speed

6.4 Wave Forces Impacting the Superstructure

6.5 Wave Forces Impacting the Substructure

Section 7 ‐ Results and Conclusions

7.1 Hydraulics

7.2 Bridge Scour

7.3 Wave Forces

Table of Contents

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | ii

Section 8 ‐ Appendices

List of Tables

Table 1: Storm History

Table 2: Storm Surge Frequency

Table 3: Model Boundary Conditions

Table 4: Storm Surge Frequency

Table 5: Wave Parameters

Table 6: Preliminary Hydraulic Results

Table 7: Preliminary Scour Results

Table 8: Forces on Exposed Piles

List of Figures

Figure 1: Project Aerial Photo (Google Earth)

Figure 2: Tidal Data Collection Locations

Figure 3: Tidal Gauging at Fort Pulaski Bridge (from S&ME Report)

Figure 4: Tidal Gauging at A‐12 (from S&ME Report)

Figure 5: Tidal Water Surface Elevations

Figure 6: Tidal Velocities

Figure 7: Tidal Discharge

Figure 8: LIDAR Coverage (green contours)

Figure 9: Single Design Hydrograph

Figure 10: FESWMS Model Mesh

Figure 11: Existing Bridge Piers

Figure 12: Model Land Cover

Figure 13: Water Surface Elevation vs. Channel Velocity

Figure 14: Wave Forces Impacting the Superstructure

Figure 15: Wave Forces Impacting the Substructure

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 1

Section 1 ‐ Introduction

Preliminary hydraulic analysis results are presented here for the proposed Fort Pulaski Bridge over the south channel of the Savannah River (FOPU 10(3), based on TS&L6), located in Chatham County Georgia

(see Figure 1). The work was performed by CDM Smith under contract with Eastern Federal Lands

Highway Department (EFLHD). Detailed assumptions and analysis will be presented in this preliminary report to assist in the evaluation and assessment of the proposed replacement structure.

1.1 Background

The following information regarding the overall project is excerpted from the project contract:

The existing Fort Pulaski Bridge (Structure Number 5420‐001P) is located on Fort Pulaski Entrance Road within the Fort Pulaski National Monument. The existing bridge is 1,287 feet long and has two travel lanes with a railing on each side. The bridge is comprised of 63 composite timber/concrete spans. The 64 substructure units (pile bents) are each comprised of timber piles and a timber beam cap. There are a total of 360 timber piles supporting the structure. The existing timber/concrete composite deck is the original deck that was built in 1939. The bridge is experiencing deterioration of the timber piles, steel beams and bearings, composite timber/ concrete spans, and embankment. The project will replace the existing bridge with a 1,295‐foot long concrete structure on new foundations constructed upstream and next to the existing bridge.

The deck of the new bridge will be raised by about 4.5 feet to match the existing bridge’s low steel elevation and the existing navigational channel opening. Project will include AASHTO Type IV girders for the new bridge. The spans will be 105 feet long, and there will be 12 pile bents (reduced from 64). Additionally, approximately 700 feet of the Fort Pulaski Entrance Road will be reconstructed. The existing bridge will be used by the public during construction and will be removed after the new bridge is operational. Project horizontal and vertical datum is US State Plane Georgia east NAD 1983 (Conus) & NAVD 88 (Geoid 03). Eastern Federal Lands Highway Division will design the project.

Fort Pulaski Bridge over Savannah River is located on Fort Pulaski Entrance Road. The bridge provides access to Cockspur Island and the Fort Pulaski National Monument and is in a tidal environment. The proposed bridge is a 1,295‐foot long 13‐span concrete structure on new foundations constructed upstream next to the existing bridge. The deck of the new bridge will be raised by approximately 4.5 feet from the existing vertical alignment. The bridge will be a two lane vehicular bridge and will accommodate two 11‐foot vehicle travel lanes (two‐lane, two‐way), and one 5‐foot shoulder on the downstream side of the bridge. The bridge approaches on both ends tie into Fort Pulaski Entrance Road owned by the Partner Agency. The road and channel survey, proposed alignment of Fort Pulaski Entrance Road, approved bridge type, size and layout (TS&L), and Geotechnical survey have been completed by Government and are included in the Government Provided Data. The as‐builts and bridge inspection report for the existing Fort Pulaski Bridge are also included in the Government Provided Data.

Introduction

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 2

Figure 1: Project Aerial Photo (Google Earth)

1.2 Design Criteria

Hydraulic Design should meet the requirements of the following standards:

EFLHD PDDM Chapter 7

HEC‐25

HEC‐18

AASHTO Guide Specifications for Bridge Vulnerable to Coastal Storms

1.3 Quality Assurance

CDM Smith uses a system of checklists and red/yellow/green checks for quality assurance of design work, according to its companywide quality assurance/quality control (QA/QC) procedure manual. The

QA Plan developed for this project is the basis for quality checking. Modeling QC matrices were developed for this project to ensure quality procedures were followed in accordance with the requirements of this contract.

Main Channel

Savannah River

South Channel

Atlantic Ocean

Fort Pulaski

US Highway

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 3

Section 2 ‐ Data Collection

The data collection for this project included both evaluation and incorporation of existing data as well as collection of field data. The government provided data included:

Existing bridge plans and inspection reports

Proposed bridge plans (TS&L)

Channel digital terrain model

Geotechnical data

EFLHD Standards

Further data collection for the project included:

Tidal data collected at three locations in the south channel

Mapping: county LIDAR terrain data

USGS gauge data with record tidal information

Other reports

2.1 Field Investigation

A field inspection was conducted by the data collection team to assess the site and determine the acceptable data collection locations. Three locations were determined to be most logical with respect to time constraints, accessibility, and desired results.

Permission was obtained from National Park Service (NPS), US Coast Guard, and a private dock owner to place gauges in various locations along the south channel of the Savannah River. These entities were very helpful in accommodating the data collection effort. After reviewing various potential locations with both NPS and Coast Guard personnel, the two most likely locations under their control that would fit our needs were the existing Fort Pulaski bridge and channel navigational aid (piling) A‐12 near the

Elba Island Bridge. Efforts to obtain permission to attach to this private bridge were not successful, however. At the third location, permission was granted by the owner to attach to an existing dock, scheduled to be demolished, just east of Lazaretto Creek on the south channel (see Figure 2).

Tidal data measurements were made in December 2014 by our subconsultant S&ME at three locations in the south channel of the Savannah River for a one week period. Useable data was reduced for a time period of approximately 36 hours. The autonomous Greyline Stingray gauges were attached to each site with removable non‐destructive banding and were set to record stage and velocity on 10‐15 minute intervals. The deployed gauges were field surveyed to establish the elevations on the NAVD88 datum.

Both water surface elevation and flow velocity were obtained at two locations, while the A‐12 location recorded only water surface elevations. See Appendix B for raw data and S&ME report.

Data Collection

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 4

Figure 2: Tidal Data Collection Locations

Figure 3: Tidal Gauging at Fort Pulaski Bridge (from S&ME Report)

Channel

Marker A‐12

Fort Pulaski

Bridge

Abandoned

Dock

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 5

Figure 4: Tidal Gauging at A‐12 (from S&ME Report)

2.2 Tidal Data

Tidal data was obtained for the Fort

Pulaski gauge at the Savannah River

Corps of Engineers dock for the time period the field data was collected.

As with all data used for this study, the gauge data at these locations was obtained with respect to the

NAVD88 datum.

Tidal data (water surface elevation) from the three data collection sites and two existing continuous recording gauge stations is shown in

Figure 5. Note the close correlation of the data for the time period shown.

Similarly, tidal data (velocities) are shown in Figure 6 below for two data collection sites and two existing continuous recording gauge stations. This comparison depicts the

SUMMARY OF GAUGES ON THE SAVANNAH RIVER NEAR FORT PULASKI

021989773 SAVANNAH RIVER AT USACE DOCK, AT SAVANNAH, GA

AVAILABLE DATA PERIOD: 10/01/2007 TO PRESENT

PARAMETERS: GAUGE HEIGHT, PRECIPITATION, STREAM VELOCITY & DISCHARGE

LOCATION: SOUTH OF US17N/GA25ALT

02199000 SOUTH CHANNEL (SAVANNAH RIVER) NEAR SAVANNAH, GA

AVAILABLE DATA PERIOD: 10/01/2007 TO PRESENT

PARAMETERS: GAUGE HEIGHT, PRECIPITATION

LOCATION: ELBA ISLAND (SOUTH CHANNEL) ATTACHED TO BRIDGE

0219897993 SAVANNAH RIVER AT ELBA ISLAND, NEAR SAVANNAH, GA

AVAILABLE DATA PERIOD: 8/16/2013 TO PRESENT

PARAMETERS: GAUGE HIEGHT, PRECIPITATION

LOCATION: ELBA ISLAND (NORTH CHANNEL)

02198980 SAVANNAH RIVER AT FORT PULASKI, GA

AVAILABLE DATA PERIOD: 10/01/2007 TO PRESENT

PARAMETERS: GAUGE HEIGHT, PRECIPITATION (10/01/2007 TO PRESENT)

STREAM VELOCITY, DISCHARGE (6/03/2009 TO PRESENT)

LOCATION: FORT PULASKI, ATTACHED TO COAST GUARD PIER (NORTH CHANNEL)

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 6

significant difference in velocities between the north and south channels. Velocities are much lower in the south channel (generally by a factor of two).

Tidal discharge is shown in Figure 7 below for two existing continuous recording gauge stations. The discharge readings at the Fort Pulaski gauge are assumed to include flow in both the north and south channels.

Figure 5: Tidal Water Surface Elevations

Figure 6: Tidal Velocities

‐5

‐4

‐3

‐2

‐1

12/16/2014 12:00 12/17/2014 0:00 12/17/2014 12:00 12/18/2014 0:00 12/18/2014 12:00 12/19/2014 0:00

W S

EL

EV

A

TI

O N

FT

N A V D

DATE/TIME

A12 Dock Bridge Ft Pulaski Corps Dock

‐4

‐3

‐2

‐1

12/16/2014 12:00 12/17/2014 0:00 12/17/2014 12:00 12/18/2014 0:00 12/18/2014 12:00 12/19/2014 0:00

V

EL

O C

IT

Y

FP

S)

DATE/TIME

Corps Dock Ft Pulaski Dock Bridge

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 7

Figure 7: Tidal Discharge

2.3 Storm Surge Data

FEMA FIS for Chatham County provides a storm history that describes the damages due to hurricanes along the coastal areas of the county as shown below.

Year Flood Height

NAVD

Hurricane

1881 15.6 Un‐named

1893 16.1 – 18.6 Un‐named

1940 6.5 – 11.5 Un‐named

1944 5.0 Un‐named

1947 7.0 Un‐named

1959 6.4 – 11.0 Gracie

Table 1: Storm History

Storm surge stage‐frequency relationships were obtained for the entrance to the Savannah River from the USACE publication Coast of South Carolina Storm Surge Study. The following table can also be found in Appendix A of that publication. The vertical datum referenced in this publication is mean sea level

(msl).

‐300000

‐200000

‐100000

100000

200000

300000

12/16/2014 12:00 12/17/2014 0:00 12/17/2014 12:00 12/18/2014 0:00 12/18/2014 12:00 12/19/2014 0:00

D

SI

C H A R G E (C

FS

DATE/TIME

Corps Dock Ft Pulaski

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 8

Table 2: Storm Surge Frequency

2.4 Mapping and Other Model Data

Mapping for development of the ground data was obtained from county LIDAR (NOAA digital coast) and navigational charts. The data was manipulated to provide a scatter set suitable for use in the SMS‐

FESWMS preprocessor. The coverage is shown in Figure 8 below.

Figure 8: LIDAR Coverage (green contours)

Other models and reports were obtained from public sources, but due to insufficient density of data they were of little use in this study.

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 9

Section 3 ‐ Hydrology

The hydrology data for this bridge is governed by storm surge and tidal prism rather than by riverine flows, even for significant rainfall events, as discussed below. The storm surge data was developed from

ADCIRC information provided in HEC25 and the tidal prism data was developed from field collection and existing gage data.

3.1 Riverine Flows

The Savannah River has several USGS gauges located in and around Savannah monitoring riverine and tidal flows on the Savannah River. The USGS gauge 02198500 on the Savannah River near Clyo, Georgia has been in place since 1929. The gauge measures riverine discharges and gauge heights as it is located over 60 miles from the mouth of Savannah River at the Atlantic Ocean. Other USGS gauges located downstream are gauges 021989773 and 02198980. Gauge 021989773 is located south of US 17N at the

USACE dock in Savannah, GA. This gauge is located approximately 12 miles above the mouth of the

Savannah River. USGS gauge 02198980 is located on the Savannah River at Fort Pulaski near the mouth of the Savannah River. Both gauges have been in these locations from 2007 to present.

A comparison of the gauges was required to determine if riverine flows affect the discharge and gauge height at Fort Pulaski. A search revealed the maximum discharge during the time period the two gauges were in place occurred in July of 2013. On July 18th of that year, the Savannah River peaked at elevation

18.38 (at Clyo), over 7 feet above floodstage, with a peak discharge of 58,000 cubic feet per second (cfs).

The gauge at the USACE dock (021989773) and the gauge at Fort Pulaski (02198980) appeared to have a rise in gauge height several days after the peak in Clyo, but upon further inspection the low tides also had extremes during this time period. The fluctuations in tides were attributed to a moon phase rather than riverine discharges. Another peak occurred in January of 2014, when the recorded discharge at the

Clyo gauge was 38,000 cfs with a gauge height of 16.5 feet (5.5 feet above floodstage). The gauge at the

USACE dock and the Fort Pulaski gauge showed no increase in gauge height or discharge on that date.

The USGS gauge at Fort Pulaski also records rainfall precipitation. At the beginning of September in 2007 the gauge recorded a rainfall event of 6 inches. The rainfall was widespread, measuring 4 inches in downtown Savannah. The gauge was unaffected by the rainfall event. It appears the Fort Pulaski gauge is unaffected by riverine flows and is totally influenced by tides.

3.2 Storm Surge Hydrograph

In keeping with normally accepted practice, the Single Design Hydrograph (SDH) method, as described in

HEC‐25, was used to generate the design hydrograph that provided the ocean boundary conditions for model runs. This method requires the use of the 50 percent values for radius of maximum winds and forward speed of surge producing storms. The 50 percent values are very similar to the values derived from analysis of historical storm surge hydrograph, and are regarded as the most probable values given a surge height of any recurrence interval (100‐year for instance). The 100‐year single design hydrograph

(SDH) is provided in Appendix C.

Hydrology

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 10

The SDH method attempts to combine the relevant driving forces of water surface into an easily applied hydrograph that can be used to simulate the boundary conditions in modeling applications. The SDH for this project was developed using a tidal amplitude of 3.75 feet, tidal period of 12 hours, radius of maximum winds (50%) of 25 miles, a forward speed (50%) of 10 miles per hour, and a time to peak surge of 24 hours. These variables yield the hydrograph shown in Figure 9 below, which was utilized to derive boundary conditions at the ocean boundary of the project.

Figure 9: Single Design Hydrograph

Once the SDH was established, a correlation was made between the instantaneous rate of change in the hydrograph and the measured discharge at the Fort Pulaski gage. This correlation (regression equation) was then used to extrapolate an estimated discharge for the falling limb of the surge hydrograph, which was found to have the highest rate of change and therefore the greatest discharge. This discharge was used as the upstream boundary condition for the 2D modeling.

-10

-5

0 10 20 30 40 50

E le va ti o n

, f t

(N A

V D

Time (hr)

Daily Tide Surge Tide + Surge

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 11

Section 4 ‐ Hydraulic Modeling

4.1 Model Mesh Development

CDM Smith prepared a 2D finite element hydraulic model using the SMS pre‐processor and the FESWMS fst2dh333 finite element algorithm. The effects of the 100‐year storm tidal surges were modeled for the existing and proposed bridge configurations. Model development included creating a suitable mesh to accurately represent the horizontal aspects of the riverine/estuarial system, and a scatter data set to provide the elevations needed for the mesh.

Development of the appropriate mesh requires assumptions based on modeling experience regarding the placement of the modeling boundaries. The side boundaries are normally closed and limited to no‐ slip conditions, while the end boundaries are open and typically described by a bounded section such as a channel. Boundaries in coastal environments can be more challenging given the potential for complete submergence within the model domain, making the selection of boundaries difficult. The objective of boundary selection should then be to provide enough separation between the point of interest and the boundary such that the boundary has limited effect on the solutions at the point of interest. In this study, the side (closed) boundaries are set with slip conditions to eliminate shear stress at these locations, attempting to mimic overland flow as closely as possible by disallowing cross boundary flow at these boundaries and allowing only tangential flow.

The open boundary upstream was evaluated at two locations. The first location was just upstream of

Elba Island where the north and south Channel diverge from a single channel. The second location was about 2 miles upstream of the bridge location. The second location was selected as providing the most stable boundary. Similarly, two open boundaries at the ocean were compared, the first is about a mile offshore and extending between south Tybee Island and Daufuskie Island to the north, and the second being the mouth of the Savannah River at the north end of Tybee Island. The second boundary was selected for stability. The resulting mesh is shown in the figure 10. This mesh provides approximately 5 miles of modeling along the Savannah River with the Fort Pulaski Bridge approximately in the middle of the model. Mesh elements range from 20 feet in size around the bridge to approximately 200 feet in size at the outer boundaries.

Figure 10: FESWMS Model Mesh

Hydraulic Modeling

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 12

Figure 11: Existing Bridge Piers

Bridge piers were added to represent the existing and proposed bridge configurations. Adding bridge piers to the model, simplified the output is obtained for scour and wave force evaluation. Because the bridge deck is submerged in storm surge conditions, added shear stress is imposed on the flow by the bridge deck. This additional shear stress must be represented in the mesh elements below in a depth averaged model. The bridge does not create large backwater differences during either normal tidal flow

(when the bridge effectively spans the channel) or extreme surge events (the surge flow is above and around the bridge upstream and downstream). Therefore, an adequate means of transferring the additional shear stress caused by the obstructed area of the bridge deck to the mesh, is to convert the obstructed area to an equivalent pier size (width and height) that will be seen as an increased bed shear stress by the model. This equivalent pier width concept is useful in evaluating the impacts due to the proposed structure when compared to the existing condition.

Following development of the physical model characteristics, land cover was established along with other model execution parameters and the process of spinning down the model was begun. Spindown typically requires lengthy iterations to reach convergence tolerances for all the model elements and extensive debugging of the model to provide stable model runs. Much of this effort is an attempt to balance the mesh resolution with computing time to achieve solution convergence by adjusting controllable parameters such as number of iterations, computation interval, and initial boundary conditions.

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 13

Figure 12: Model Land Cover

4.2 Model Calibration

Once spindown was achieved, the stable model was calibrated to known measured conditions within the system. Specifically, the water surface elevation and velocity at the existing bridge were adjusted to calibrate the model to a point on the measured tidal cycle. By adjusting the initial values of Mannings

“n” and eddy viscosity, the model was forced to achieve the velocity and water surface elevation as measured in the field through repeated manual adjustment and model execution. Once calibrated, the model could be used with some assurance of reliability for predicting values outside the range of measured data.

4.3 Model Boundary Conditions

To drive the hydraulic simulations, parameters at the model boundaries were selected to match the values anticipated from a storm surge. Typically the upstream (channel) boundary is provided with a discharge and the downstream (ocean) boundary is provided with a water surface elevation. Estimating these values for the storm surge conditions is made possible by extrapolating the known values of discharge, velocity and stage from the calibrated model and recorded gauge data. The parameters used for this study are shown below.

Simulation ID Storm Surge

Hydrograph Location Upstream Boundary

Downstream Boundary

9 Peak 75000 17.0

9e Falling Limb 550000 15.2

9d Falling Limb 350000 11.2

9f Falling Limb 350000 7.0

Table 3: Model Boundary Conditions

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 14

4.4 Model Simulations

While it is ideal to run continuous simulations for tidal (or unsteady) environments, after several attempts steady state modeling was selected due to the long run times required for continuous simulation and the short performance time allocated for this project. It was possible to run several steady state models in the same time as a single continuous simulation in order to evaluate different scenarios. Steady state runs were also useful in confirming that the assumption of worse‐case condition on the falling limb of the SDH was valid.

It is generally accepted that the ebb of the storm surge provides the greatest stress on a bridge because the ebb flow occurs over a shorter time frame, thereby indicating a steeper hydrograph, as shown on the SDH plot. This assumption is used to identify simulation points on the SDH that can help bracket discharge characteristics with steady state simulations. Model simulations therefore were selected and executed for various boundary conditions, described above, based on the point of inflection on the SDH hydrograph. The four models presented are:

near the peak water surface elevation on the falling limb of the SDH (ID 9) at the point of maximum rate of water surface change on the falling limb of the SDH (ID 9e) near the midpoint of the falling limb of the SDH (ID 9d) at an elevation on the falling limb of the SDH that is consistent with the overbank elevation near the bridge (ID 9f)

The flow characteristics for these simulations were then used to identify worst case conditions for force computation purposes. Simulation one was also executed for the existing and proposed conditions to provide a comparison that can be used to determine impacts of the replacement bridge on the surrounding flow regime.

When comparing existing to proposed models, changes to the velocity are minor with most variations close to the new bridge structure being less than 0.1 ft/s. Differences in water surface elevation are not discernable.

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 15

Section 5 ‐ Scour Analysis

The proposed bridge over the south channel of the Savannah River at Fort Pulaski has approximately the same opening as the existing bridge. The proposed bridge is approximately 15 feet longer with far less piers. The existing bridge has 62 interior bents compared to 12 interior bents for the proposed bridge.

While the spans for the proposed bridge are much longer and deeper the low chord remains very close to the same elevation due to the raise in grade for the proposed bridge.

The scour analysis for the proposed bridge was performed in accordance with HEC‐18. Long term scour was evaluated utilizing bridge inspection reports dating back to 1977. The bridge inspections reports provide an excellent resource to predict long term aggradation and degradation of the stream bed. The bridge inspection reports include 15 measurements at all 64 bents from 1977 to 2014. The measurements definitely indicate an alluvial channel that aggrades and degrades with time but indicates no long term trends. Pier 47, for example has the greatest change in groundline elevation with a change of 8 feet. Even though there is a maximum change of 8 feet, the 1977 groundline measurement is within

1.1 feet of the 2014 measurement. Pier 46, located just 20 feet from pier 47, only has a maximum change in groundline elevations of 4.9 feet and the 1977 measurement compared to the 2014 measurement is within 0.1 foot of each other. When comparing the average 1977 measurements at all

64 bents against the 2012 measurements at all 64 bents, the data indicates 43 bents have a groundline that has aggraded, 20 have a groundline that has degraded, and one shows no change. The average

1977 groundline measurements at all 64 bents compared to the measurements at all 64 in 2014 indicate almost the opposite of 2012; the groundline has aggraded at 24 bents, has degraded at 37 bents, and 3 bents have no change.

The channel appears to be alluvial but stable. The changes in groundline elevations at the shallower piers near the existing bridge abutments did not experience much change and therefore no lateral migration is anticipated.

5.1 Soils Data

The soil borings provided for the site were used to evaluate sediment transport. Twelve soil borings were provided for locations within the stream channel. An average D50 equal to 0.48mm was determined after analyzing all 12 boring locations, describing coarse to medium, non‐cohesive sands.

The critical velocity for this sediment size is 2.3 feet per sec (ft/sec). Analyzing the velocities from the hydraulic model, the maximum velocity is 1.74 ft/sec and an average stream velocity is 1.57 ft/sec suggesting clear‐water scour. The average D50 may be somewhat misleading since sediment sizes vary so much at the streambed. Six of the 12 borings have a D50 less than or equal to 0.2mm and three of those six are less than 0.15mm. The critical D50 for the maximum velocity of 1.74 ft/sec is 0.2mm and the critical D50 for the average velocity of 1.57 ft/sec is 0.15mm. This suggests that live‐bed scour also exist in the stream channel. The average Froude number for the channel is equal to 0.05. The stream is definitely in the lower regime flow.

Scour Analysis

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 16

5.2 Scour Components

Pier scour was calculated at individual bents for the proposed structure. The proposed bridge has 12 bents located in the channel, with five 18‐inch‐square piles at each bent except for bent 6 and 7. Bents 6 and 7 have two rows of five piles, battered at a 1.5 to 12 slope. All bents have very minimal angles of attack except for bent 1, where the angle is 18 degrees. Bent 1 was the only bent where the K2 value was adjusted to account for skew. Pier scour was calculated at a water surface elevation equal to 11.2 feet. The discharge at this water surface elevation provided the combination of velocity and depth to produce the maximum scour.

Contraction scour and abutment scour were not calculated due to the geometry of the proposed bridge.

The constrictions of flow caused by the proposed bridge are so minimal that scour variables leading to abutment and contraction scour are minimal. Contraction scour is typically included in scour estimates;

however, at this bridge any overbank flow (at the design event) is not forced back into the channel, so the ratios leading to contraction scour are muted. Therefore, contraction scour is estimated to be zero for this bridge configuration.

Pressure flow scour was computed for the bents expected to submerge during the design storm as a routine part of the scour analysis. Pressure flow scour is predicated on full submergence of the bridge

(usually this is a bridge that has a relatively level grade), creating a significant separation zone on the downstream side of the deck. Because the FOPU deck is inclined, and the velocities are relatively slow for the extreme events, there is very little difference in water surface elevation upstream and downstream of the bridge, essentially drowning out the separation. This correlates to very little vertical contraction due to pressure flow. We know that the separation zone is very conservatively estimated as well in the calculation, so the over‐prediction is compounded for cases like FOPU. Therefore, we did not include the pressure flow scour in the table.

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 17

Section 6 ‐ Wave Force Analysis

Wave forces were estimated for the proposed bridge using the methodologies in the AASHTO Guide

Specifications for Bridge Vulnerable to Coastal Storms (2008). The methods developed for calculating the wave forces are based on numerical simulation of the state of pressure, velocity, and acceleration within the water as a wave passes through the bridge. Physical wave tank test were used to develop various coefficients involved in the calculation process. Limited field data from recent storms within the Gulf of

Mexico were used to compare results. Hurricane Ivan, Rita, and Katrina which caused significant damage to multiple bridges across the Gulf of Mexico and were used in the development of the manual. This section will go into detail the methodology used to estimate the forces the proposed Fort Pulaski Bridge will experience during coastal storms.

6.1 Design Cases

Two different design cases were evaluated in the wave force analysis. Design Case I involves the maximum vertical force along with the associated horizontal force. Likewise, Design Case II involves the maximum horizontal force with an associated vertical force. Due to the topography of the site and the proposed bridge design, the recommended vertical clearance above the design 100‐year water surface could not be met for all 13 spans. The spans in close proximity to the end bents of the proposed bridge experience submergence for the higher frequency storm events. In light of these conditions, wave forces acting on each individual span will vary significantly based upon the degree of submergence the span’s superstructure experiences. Initially CDM Smith performed a preliminary wave force analysis on each span for the 100‐year storm event, and determined the vertical uplift force for Design Case I to be the dominant condition in the analysis. The results for the preliminary wave force analysis were provided to

EFLHD, and through discussions with CDM Smith it was determined that the proposed bridge design will be unable to handle the 100‐year design forces for all of the 13 spans. EFLHD requested that CDM Smith provide an analysis of each span for various storm frequencies to determine appropriate designs for each span. The table below lists the various parameters that are modified on a design storm frequency basis. Surge elevations were obtained from ADCIRC and wind speeds were obtained from ASCE wind charts and converted to appropriate design speeds. Channel velocities were determined by evaluating the multiple FESWMS models output results to determine a channel velocity based on the water surface elevation. The polynomial graph below demonstrates the methodology used to define velocities based upon water surface elevation. Once the water surface reaches an elevation outside the bank, the floodplain increases and the cross sectional area grows exponentially, resulting in slower velocities in the channel.

Results and Conclusions

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 18

Figure 13

6.2 Submersible Superstructures

As mentioned previously in this report, some of the proposed spans will have superstructures that will be submerged in higher frequency storms. Through research it was determined that once the water surface elevation reached a certain height above the bottom of the girder, the methodologies and equations in the manual are no longer applicable and the maximum vertical uplift force reaches a limit.

This maximum vertical uplift force occurs when the design wave trough reaches an elevation equivalent to the bottom of the girder. Each span was analyzed with this limit and appropriate forces were determined.

6.3 Fetch and Wind Speed

A fetch is defined as a wind path that will travel across water to a single point. In this case the single point referenced is the Fort Pulaski Bridge. Multiple fetches were evaluated to determine the appropriate fetch length and depth that resulted in the greatest wave height. Depths across the fetch

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8

W at er S u rf ac e El ev at o in N A V D f t)

Velocity ft/s

Water Surface Elevation vs Channel Velocity

Storm Frequency (yrs)

Surge (ft)

Stillwater Elevation

(NAVD 88)

Channel Velocity (ft/s)

Wind speed (mph) 3 second gusts

10 6.7 6.9 1.3 90.0

25 9.2 9.4 1.6 117.7

50 12 12.2 1.6 130.0

100 14.5 14.7 0.9 139.1

Table 4: Storm Surge Frequency

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 19

were determined using county LiDAR along with Nautical Charts for the Savannah River and surrounding water bodies. Due to the water surface evaluation of the multiple storm frequencies being analyzed, the fetch depths were adjusted appropriately to the design storm surge elevation. The design wind speed was determined using the ASCE 50‐year 3 second gust wind speed charts (ASCE, 2005). The 3 second wind gust speeds were converted to the appropriate design storm frequency along with an average wind speed for the 10‐minute and one‐hour duration. Through an iterative process “t” (the time duration required to develop a fetch limited wave) is calculated and a significant wave height, wave length, maximum wave height, and design wave crest can be established.

6.4 Wave Forces Impacting the Superstructure

Design Case I and Design Case II as previously mentioned in the report are shown in the figure below.

Design Case I is used to determine the maximum vertical uplift force the bridge will experience during a coastal event. The maximum quasi‐static vertical force, FV‐MAX is calculated in one thousand pounds per foot (kip/ft) per unit length of span. For all spans except spans 1 and 13 this unit length is 105 feet. For these other spans the proposed length is 70 feet. Trapped air factor was calculated to determine the severity from the impact of trapped air the bridge may experience as the storm surge levels rise.

Trapped air in between the girders of the bridge causes an additional uplift force due to the buoyancy forces that will act on the bridge. Furthermore the vertical slamming force Fs, measured in kip/ft per unit length of span contributes to the uplift forces on the bridge as well. The vertical slamming force is the repetitive force the bridge’s superstructure experiences as the design wave crashes into it. The figure below illustrates the design wave crest passing through the bridge cross section along with the various parameters required for the wave force calculation. The horizontal force component for Design Case I is known as the associated horizontal quasi‐static force, FH‐AV, in kip/ft per unit length of span. Using the vertical and horizontal forces calculated for Design Case I, an associated moment, MT‐AV in kip*ft/ft per unit length of span can be determined for the trailing edge of the bridge. For Design Case II the maximum horizontal force, FH‐MAX is determined rather than the maximum vertical force that was calculated in Design Case I. The associated vertical force, FV‐AH along with the slamming force will be combined with maximum horizontal force to calculate the associated moment due to the maximum horizontal force on the bridge.

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 20

Figure 14 Wave Forces for Design Cases I and II

6.5 Wave Forces Impacting the Substructure

The proposed bridge design calls for five 18 inch‐square piles to be utilized at each of the bents. Bents 6 and 7 will consist of ten 18 inch‐squared battered piles. Bents 6 and 7 are located at the approximate edges of the channel. The forces acting on the exposed piles was determined using the Morison equation. F is calculated in kips per unit length due to non‐breaking waves. The figure below displays the force applied across the exposed piles. The Morison equation requires that certain parameters be calculated using the appropriate wave theory. A Java applet that allows calculations of these values accessed at http://coastal.udel.edu/faculty/rad/streamless.html. Using this tool the maximum horizontal acceleration of water particles along with the maximum horizontal velocity of water particles was calculated for the various design waves. The maximum horizontal particle velocity occurs at the crest of the wave whereas the maximum acceleration occurs near the stillwater elevation along the wave. The values for the maximum acceleration obtained from the Java applet are listed below.

Wave Parameters

Span 1 2 3 4 5 6 7 8 9 10 11 12 13

Du/Dt (ft/s2) 9.8 11.2 11.6 11.6 11.5 11.5 11.8 11.8 11.5 11.5 11.6 11.6 9

U (ft/s) 6.9 11.4 11.5 11.5 11.2 11.2 11.2 11.2 11.2 11.2 11.2 11.2 8.9 Table 5 Maximum Wave Acceleration Values

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 21

Figure 15 Wave Force applied Across Exposed Piles

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 22

Section 7 ‐ Results and Conclusions

7.1 Hydraulics

Based on this hydraulic modeling effort, no recommendations for adjustments to the proposed bridge configuration appear to be necessary to achieve conformance with the desired criteria for hydraulic capacity. While the proposed bridge does not meet requirements for freeboard, due to the low elevation of the bridge ends, the opening is adequate for the anticipated discharge with no adverse impacts to the surrounding area. Since the proposed bridge’s low chord is at least as high as that of the existing bridge, no navigational clearance issues are assumed to exist with the proposed structure.

Hydraulics Hydraulics Hydraulics Hydraulics

Bent

Water Surface Elev. (ft)

Velocity (ft/s)

Water Surface Elev.

(ft)

Velocity (ft/s)

Water Surface Elev.

(ft)

Velocity (ft/s)

Water Surface Elev.

(ft)

Velocity (ft/s)

1 17 0.15 15 0.72 11.2 1.2 7 0.88

2 17 0.018 15 0.79 11.2 1.4 7 1.14

3 17 0.19 15 0.83 11.2 1.6 7 1.25

4 17 0.2 15 0.86 11.2 1.7 7 1.32

5 17 0.2 15 0.81 11.2 1.6 7 1.19

6 17 0.21 15 0.8 11.2 1.6 7 1.15

7 17 0.22 15 0.82 11.2 1.6 7 1.2

8 17 0.21 15 0.82 11.2 1.6 7 1.2

9 17 0.21 15 0.83 11.2 1.6 7 1.2

10 17 0.2 15 0.78 11.2 1.5 7 1.05

11 17 0.2 15 0.78 11.2 1.5 7 1

12 17 0.2 15 0.83 11.2 1.6 7 1.17 Elevations are referenced to North American Vertical Datum of 1988 (NAVD88)

Table 6 Fort Pulaski Bridge Preliminary Hydraulic Results

7.2 Bridge Scour

The scour computations indicate that the long term potential for scour varies between aggradation and degradation, with no single trend being predominant according to the data taken over the life of the bridge. Therefore, the historical maximum scour at the pier with the most variation in bed elevation was

8 feet, which includes both long term and pier scour components. Total scour estimates are presented in Table 7 as a combination of appropriate superimposed components. These scour depths are recommended for inclusion in the structural design as ultimate depths for design.

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 23

Scour Hydraulics

Bent

Long Term (ft)

Contraction (ft)

Local ‐ Pier (ft)

Total Scour (ft)

Scour Elevation (NAVD,ft)

Water Surface Elev. (ft)

Velocity (ft/s)

1 2 0 4.8 6.8 ‐11.8 11.2 1.2

2 2 0 3.4 5.4 ‐22.4 11.2 1.4

3 2 0 3.4 5.4 ‐22.9 11.2 1.6

4 5 0 3.5 8.5 ‐28.5 11.2 1.7

5 5 0 3.6 8.6 ‐30.6 11.2 1.6

6 5 0 5.1 10.1 ‐31.1 11.2 1.6

7 5 0 5.5 10.5 ‐31.5 11.2 1.6

8 5 0 3.6 8.6 ‐32.1 11.2 1.6

9 5 0 3.6 8.6 ‐33.6 11.2 1.6

10 2 0 3.6 5.6 ‐26.6 11.2 1.5

11 2 0 3.4 5.4 ‐23.4 11.2 1.5

12 2 0 3.2 5.2 ‐10.2 11.2 1.6 Elevations are referenced to North American Vertical Datum of 1988 (NAVD88)

Table 7 Fort Pulaski Bridge Preliminary Scour Results

Because the end‐of‐bridge embankments are so minimal, abutment scour was not computed in light of the velocities present. Standard riprap revetment with toe‐in to natural ground is recommended for this installation, and rock riprap should be greater than D50= 1.0 foot. GDOT (805.2) Type 1 riprap or larger is recommended due to wave action (See Appendix E for Typical Abutment Riprap Placement detail).

7.3 Wave Forces

The wave forces computed for the Fort Pulaski Bridge produce significant loads on the bridge superstructure. The accompanying charts in Appendix F provide curves that represent the anticipated horizontal and vertical forces estimated for various storm surge frequencies. Not all spans were analyzed for the 100‐year design storm. Spans located near the center would not have any wave forces acting on the superstructure so this analysis was eliminated. Design Case I typically was the dominant condition in the analysis. The vertical uplift force the proposed bridge experiences during the rising surge elevations is the major design concern. Using the tables in Appendix F, a design storm frequency can be estimated based upon a selected force. The forces on the exposed piles for the substructure were calculated for the 100‐year design storm and the results are shown in the table below. These forces are the maximum distributed force the wave will enact on the exposed piles. These distributed forces decrease fairly significantly at depths below the design wave.

Forces on Exposed Piles

Span 1 2 3 4 5 6 7 8 9 10 11 12 13

F(max) 0.31 0.43 0.44 0.42 0.43 0.43 0.46 0.43 0.43 0.45 0.43 0.45 0.29

F=kips per unit length due to non‐breaking waves resulting from the maximum wave height, Hmax Table 8 Forces on Exposed Piles for 100‐Year Design Storm

FOPU 10(3) FORT PULASKI BRIDGE HYDRAULIC STUDY | 24

Section 8 ‐ Appendices

Appendix A

Data Provided

• Existing Bridge Inspection Report

• Existing Bridge As-built Plans

• Proposed Bridge TS&L

BRIDGE INSPECTION REPORT

FORT PULASKI BRIDGE

ENTRANCE ROAD OVER SOUTH CHANNEL-SAVANNAH RIVER

FORT PULASKI NATIONAL MONUMENT

LN

STRUCTURE NO. 5420-001P

INSPECTION TYPE: ROUTINE

INSPECTION DATE: 5/20/14

Report prepared by: WU

Federal Highway Administration Federal Lands Bridge Office Bridge Inspection and Management Program 21400 Ridgetop Circle Sterling, Virginia 20166 http://www.efl.fhwa.dot.gov/FLHBO/BIP http://www.efl.fhwa.dot.gov/FLHBO/BIP

STRUCTURE SUMMARY STRUCTURE NO. 5420-001P

FORT PULASKI BRIDGE INSPECTION DATE: 5/20/14

BRIDGE IDENTIFICATION AND DESCRIPTION

MAIN SPAN MATERIAL: ....... STEEL NO. OF MAIN SPANS:

MAIN SPAN TYPE: ............... NO. OF APPR. SPANS: ........ STRINGER/MULTI-BEAM or GIRDER 62

# OF BEAMS, MAIN SPAN: .. 5 STRUCT. TOTAL LENGTH: .. 1,283.0 FT (391.1 M)

DECK TYPE: ......................... CONCRETE CAST-IN-PLACE MAX. SPAN LENGTH: .......... 40.0 FT (12.2 M)

WEARING SURFACE: .......... BITUMINOUS DECK WIDTH: ...................... 24.8 FT (7.6 M)

ABUT. FOUNDATIONS: ........ PILE BENT CURB TO CURB WIDTH: ..... 20.0 FT (6.1 M)

PIER FOUNDATION(S): ........ PILE BENT YEAR BUILT:

ABUTMENT / PIER MAT'L: ... WOOD OR TIMBER YEAR RECONSTRUCTED: .. 1965

PROJ./PLAN REF. NO: ......... *1------A1 / 3014A SKEW ANGLE: ..................... 0 °

NPS ROAD CLASS: .............. 1 - PRINCIPAL PARK ROAD FRACTURE CRITICAL: ........ NO

STRUCTURE HISTORY: ...... 1984-JOINT REPAIR. 1996-PILE JACKETING, RIPRAP PLACED. 1999-JOINT REPAIR.

2008-PILE REPLACEMENT/REPAIR, PILE JACKETING, JOINT REPAIR.

LOCATION

FORT PULASKI NATIONAL MONUMENT MILE POINT: ..................... 0.16 (KM PT 0.26)

STATE: ................................. GEORGIA LOCATION: 1.0 mile from Fort Pulaski.

FACILITY CARRIED: ............ ENTRANCE ROAD

FEATURE INTERSECTED: .. SOUTH CHANNEL-SAVANNAH RIVER

STRUCTURE EVALUATION SUMMARY

See next page for Structure Evaluation Summary.

INSPECTION TEAM: J. THOMAS, T. POW ELL, M. HAND (VOLKERT, INC.)

CONDITION ASSESSMENT:

JOE THOMAS, P.E., FIELD TEAM LEADER

REPORT REVIEW :

MARCUS MILLER, P.E., SENIOR INSPECTION ENGINEER 703-404-6252

REPORT APPROVAL: DATE: 9/24/14

JORGOMAI CEESAY, PH.D., P.E., BIP SUPERVISORY ENGINEER 703-404-6244

STRUCTURE SUMMARY STRUCTURE NO. 5420-001P

STRUCTURE EVALUATION SUMMARY CONTINUED

This bridge provides access to Fort Pulaski from U.S. Route 80. The bridge was constructed in 1938 and underwent rehabilitation projects in 1965, 1996, and 2008. The superstructure is comprised of 62 composite timber/concrete "corduroy" spans, and one steel beam span over the main channel. The substructure units (pile bents) are each comprised of 5 timber piles with a timber or concrete beam cap. The existing concrete/timber composite deck is part of the original 1938 construction. The bridge is currently posted for weight restrictions.

Several measures have recently been taken to keep the bridge operating in a satisfactory manner. Extensive structural repairs were made to the substructure in 2008 under Project PRA-FOPU 10(2). The scope of work for the repair project included replacement, splicing, and/or jacketing of a number of timber piles that had exhibited significant deterioration, and replacement of deteriorated cross-bracing members. Since the previous inspection, a new horizontal timber brace has been installed on the south face of Bent #32. In addition, the rails and curbs have been painted.

However, serious problems with the bridge persist, including extensive decay of the timber decking adjacent to the concrete bent caps, where many of the deck timbers and timber shims have lost bearing. In addition, the steel beams and bearings in the main span are exhibiting widespread rusting with substantial section loss in some key locations, and deterioration of the remaining unjacketed pile timber is rapidly progressing. It should also be noted that the existing riprap on the embankment slopes is undersized and gradually washing away, and the concrete bank protection is partially undermined. Due to the age of the structure, the widespread deterioration, and high cost of continued repairs, the bridge has been scheduled for replacement in FY 16 under Project PRA-FOPU 10(3).

RECOMMENDATIONS AND ESTIMATED COSTS STRUCTURE NO. 5420-001P

A (Critical) - Structure is closed, in imminent danger of collapse, or presents a critical safety hazard.

B (Major) - Structure is seriously deficient or presents a safety hazard, but can remain in service at reduced loads or with frequent inspections.

C (Moderate) - Structure is deficient, functionally obsolete, or requires substantial repairs to prevent a serious deficiency or unsafe condition.

D (Minor) - Structure is generally in good condition. Preventative maintenance may be required.

E (Special) - Structure is under construction or reconstruction.

B226 PRIORITY OF IMPROVEMENT

ITEM

NO.

MAINTENANCE RECOMMENDATIONS

(GENERALLY NOT FLHP ELIGIBLE)

QTY. UNIT COST

UNIT

ESTIMATE

COST

PRELIM.

58.8 REALIGN LEANING UTILITY POLE AT SOUTHWEST APPROACH. 1 EA

PRELIM. MAINTENANCE TOTAL :

ITEM

NO.

REHABILITATION/RECONSTRUCTION

(TYPICALLY NEEDED WITHIN 5 YEARS)

QTY. UNIT

UNIT

COST ESTIMATES

PRELIM. COST

REHAB. RECONSTR.

PRELIM. REHABILITATION/RECONSTRUCTION TOTAL :

COST ESTIMATE SUMMARY

PRELIM. MAINTENANCE TOTAL:

PRELIM. REHABILITATION TOTAL:

PRELIM. RECONSTRUCTION TOTAL: ..

TOTAL ITEMIZED COSTS:

PRELIMINARY ESTIMATE TOTAL* :

(*Includes typical contract costs for mobilization, traffic control, and design/construction engineering; but does not include detour bridges, work beyond the bridge limits, or work not listed above. Estimates are approximate and are for preliminary planning purposes only; they should not be used as a complete or final engineers estimate.)

REMARKS: Bridge is scheduled for replacement in FY 16 under Project PRA-FOPU 10(3).

Bridge was repaired in 2008 under Project PRA-FOPU 10(2). South approach has been reposted for the original 18T-22T-34T load limit, reflecting the current Inventory Ratings for the H, Type 3, and Type 3S2 trucks respectively.

North approach remains posted for 18 tons.

Maintenance cost estimate is based on historical data, including costs for similar projects when available.

FIELD INSPECTION REPORT STRUCTURE NO. 5420-001P

G = GOOD (CODES 7 - 9) F = FAIR (CODES 5 - 6) P = POOR (CODES 0 - 4)

CONDITION RATING CODE EQUIVALENTS FOR SUB-ELEMENTS

SERIOUS CONDITION - loss of section, deterioration, spalling, or scour have seriously affected primary structural components. Local failures are possible.

Fatigue cracks in steel or shear…