B08_Section_J_Attachment_10_Marine_Survey.pdf

Marine Survey in the Vicinity of Fort Jefferson in Preparation for Major Repairs to the

Counterscarp and Dredging Operations

Dry Tortugas National Park

Monroe County, Florida

Cover Image: A diver examining one of the large previously known anchors in the South Swim Beach survey block.

Joshua Marano, Maritime Archeologist South Florida Parks and Preserve (SOFL)

Homestead, FL

With contributions from

Anne Wright, Archeologist Submerged Resources Center (SRC)

Lakewood, CO

Matthew Hanks, Archeologist Submerged Resources Center (SRC)

Lakewood, CO

DRTO/EVER Cultural Resources Report CRR-XXXX

May 2022

This page intentionally left blank

Contents Figures

Tables

Introduction

Description of Undertaking

Legal Location for the Undertaking

Local Environment

Area of Potential Effect (APE)

Records Checks

Property History

Planning

Construction

Maintenance and Repairs

National Park Service Ownership and Major Repairs to the Counterscarp

Previous Maritime Archeological Investigations

Additional Research Needs

Field Methodology

Marine Magnetometer Survey

Systematic visual and metal detecting survey of the moat

Results of Inventory

Magnetometer Results

Metal Detecting and Visual Survey of the Moat

Photogrammetry of the Moat Wall

Discussion of Results

Native American Consultation

Conclusions and Recommendations

References Cited

Figures Figure 1. APE MAP

Figure 2. Detail from the 1848 sketch of a proposed cofferdam for the counterscarp wall at Fort

Jefferson. This inset image highlights the construction of the cofferdam. Note that this proposed design changes slightly over time. NPS image, TIC#137468

Figure 3. Detail from the 1848 sketch of the counterscarp wall at Fort Jefferson. This inset image highlights an alternative design for the counterscarp wall. Note that this proposed design changes slightly over time. NPS image, TIC#137468

Figure 4. Image highlighting the status of construction of Fort Jefferson in 1854. Note the inset detail images that highlight the specific construction characteristics of the counterscarp. NPS image, TIC#

135172

Figure 5. Detail of the “Plan of the U.S. Naval Station, Dry Tortugas, Florida, Showing Condition on July 1, 1901”. Here, the sedimentation of the moat is highlighted. These blockages were cleared and the shoaling is not shown on subsequent diagrams. NPS image, TIC# 60904

Figure 6. 1938 Garden Key General Plan. NPS image, TIC# 1005

Figure 7. Image taken from a Bastion showing damage to the counterscarp and the extent of shoaling within the moat in 1937. Image on file at Everglades National Park

Figure 8. Image showing the shoaling of the moat at Front 4 in 1937. NPS image on file at Everglades

National Park (EVER 7117)

Figure 9. Workers hand clearing shoals within the moat during the 1938 WPA project. Image on file at

Everglades National Park (DRTO 300946)

Figure 10. Workers hand clearing shoals within the moat during the 1938 WPA project. Image on file at

Everglades National Park (DRTO 300947)

Figure 11. Photograph submitted by park custodian Russel A. Gibbs documenting storm damage to the counterscarp wall in 1946. Image on file at Everglades National Park (DRTO 300734)

Figure 12. Photograph submitted by park custodian Russel A. Gibbs documenting storm damage to the counterscarp wall in 1946. Image on file at Everglades National Park (DRTO 301128)

Figure 13. Moat wall damage in 1951. Image on file at Everglades National Park

Figure 14. Moat wall damage along Front 4. Note the fishing vessels at anchor and Loggerhead Key in the background. Image on file at Everglades National Park (DRTO 301130)

Figure 15. Conditions of the counterscarp, highlighting damage to Fronts 3 through 5 in 1951. NPS image, TIC# 2351

Figure 16. Image highlighting storm damage to the counterscarp in 1957. Note the surviving portions of concrete or tile covering historic base material (coarse coral aggregate). Also note the shoaling to the interior of the counterscarp. Image on file at Everglades National Park

Figure 17. 1952 trail plan highlighting current trails in use at the fort. Note the breaks in the counterscarp and shoaling of the moat. NPS image, TIC# 2007

Figure 18. Damages to the counterscarp, likely at Front 5, in May 1963. Iage on file at Everglades

National Park, DRTO 300796

Figure 19. Heavy equipment utilized in the clearing of sediment from the moat in 1965. Note the depth of the excavation as indicated by the scale of the tractor. This excavation likely extends beyond the original depth of the excavated moat (-5 MLW). It is also likely that a cofferdam was used to allow for the excavation of the moat in this manner. Image on file with Everglades National Park, DRTO 300923. 34

Figure 20. Another image of heavy machinery being utilized in the excavation of the moat in 1965. Image on file with Everglades National Park, DRTO 300925

Figure 21. Details from the 1962 plans to repair the counterscarp wall between Bastions 3 to 5. Note the use of concrete filled bags to form a cofferdam within which tremie concrete is poured directly into rubble. Also note the specific mix of concrete which was not necessarily sympathetic either from a historical or engineering standpoint. NPS image, TIC# 60020

Figure 22. 1962 plans to repair portions of the counterscarp with unstable foundations. Some notes to point out, that the rubble fill used in these repairs would be sourced from the recently demolished

Officer Barracks and the use of rip-rap (in this case, block granite, also likely salvaged from the fort) along the ocean face of the counterscarp wall. NPS image, TIC# 60021

Figure 23. A newly repaired section of the counterscarp at Front 5 in 1964 (north coal dock is in the background). Note the variety of brick, poured concrete walking surface, and the non-symmetrical courses of brick in the coping and façade. Image on file at Everglades National Park, DRTO 300797

Figure 24. Damages to the newly repaired counterscarp wall in 1965 (Top) and 1967 (Bottom). Given the catastrophic failure of these portions of the counterscarp so soon after repair forces the question of whether or not the repairs made between 1962 and 1964 were sound. Images on file at Everglades

National Park, DRTO 300745 (Top) and DRTO 300745 (Bottom)

Figure 25. Detailed repair plans for the counterscarp developed in 1968 during a series of repairs to the seaplane ramp, dock, and other facilities at Fort Jefferson. NPS image, TIC# 41000

Figure 26. Current conditions at the fort in 1978 during planning efforts for additional repairs to the counterscarp identifying the approximate locations of breaks in the structure as well as shoaling within the moat. Note the presence of the “Rubble from demolished buildings” located along the eastern edge of Garden Key, just beyond the counterscarp adjacent to Front 6. NPS image, TIC# D10/134606

Figure 27. 1982 plan for repairs along the counterscarp surrounding Fort Jefferson. NPS image, TIC#

135506

Figure 28. Finished repair work completed along the counterscarp in 1985. Note the finished cross sections include iron reinforcement connecting the poured concrete slab to the counterscarp. Many of these fittings are now visible in damaged sections of the counterscarp along Front 5. NPS image, TIC#

80005

Figure 29. A map showing the early survey areas as described in Murphy 1993

Figure 30. North Swim Beach survey area (indicated in purple) and South Swim Beach survey area

(indicated in green)

Figure 31. Geometrics G-882 cesium vapor marine magnetometer. NPS image

Figure 32. 17-foot Whaler used for survey operations. NPS image

Figure 33. South Swim Beach survey area. Survey lines are indicated in black, with labeled magnetic anomalies

Figure 34. Ladder associated with a 20th century shipwreck identified within the South Swim Beach survey block, NPS photo

Figure 35. One of four large tanks associated with a 20th century shipwreck identified within the South

Swim Beach survey block, NPS photo

Figure 36. SSB_41 Historic anchor previously known to team. NPS image

Figure 37. A diver examining one of the large previously known anchors in the South Swim Beach survey block. Code named “Ship Killer” by the crew given its depth, a survey buffer was included around this and the adjacent reef as hazards to navigation. NPS image by Susanna Pershern

Figure 38. Drawing documenting the anchor that once sat at the entrance to the sallyport. From Murphy

1993:231

Figure 39. An undated photo showing the anchor documented by Murphy in place as a decoration at the sally port. At some point in the past, the anchor was placed in the swim area and given its dimensions, orientation, and associated tackle, is most likely the one referred to as “Ship Killer” by the crew. See

Figure 35

Figure 40. Drawing documenting the unprovenienced anchor discussed by Murphy 1993:231-232 as being located on the north coal dock in the 1990s. At some point this anchor, along with one other on the coal dock, was moved to the swim area. Given the dimensions and shape of this anchor, it is most likely SSB-41 (see Figure 34)

Figure 41. North Swim Beach survey area. Survey lines are indicated in black, with labeled magnetic anomalies. Note the Fort Wreck in the southwest corner of the survey area

Figure 42. Combined confidence modeling of the North and South Swim Beach Blocks. The colors differentiate between high and low confidence that an object with a 136 kg mass was detected

Figure 43. Combined confidence modeling of the North and South Swim Beach Blocks. The colors differentiate between high and low confidence that an object with a 200 kg mass was detected

Figure 44 Magnetic anomalies/material identified during the systematic metal detecting survey along

Front 2

Figure 45. Magnetic anomalies/material identified during the systematic metal detecting survey along

Front 5

Figure 46. Survey team marking the locations of individual magnetic items and anomalies as they are recorded with a total station along Front 5

Figure 47. Wooden sheet piling identified along recently disturbed portions of Front 2.The dimensions of these timbers as well as the manner in which they are constructed are identical to those proposed and described in historic plans produced during the initial construction of the counterscarp and are likely evidence of surviving original portions of the structure. NPS image

Figure 48. A 1962 plan for the “Disposition of Excess Bricks” from the demolition of several structures within Fort Jefferson. NPS image, TIC# 60018

Figure 49. Detail of the excess brick disposal plan highlighting specific storage areas and plans. Of particular interest in this image is the location of a temporary storage area for brick and other aggregate material in the, at the time, silted in moat adjacent to Front 5 that was to be used as aggregate material for repairs to the counterscarp in that location (See Figure 22). This area correlates with a large number of unexplained magnetic anomalies identified during the systematic metal detecting survey of that area.

Figure 50. Images of typical finds adjacent to Front 5 including a trowel (left) and the remains of scaffolding (right) that are directly associated with repairs to the structures at that location. NPS image.

Figure 51. A historic bottle documented in situ near the current sluice gate opening between Fronts 3 and 4. The green bottle had no markings but was formed with a two-piece mold with an applied lip and likely dated from the latter half of the 19th century

Figure 52. A clear embossed bottle found near the modern fuel line. The bottle is marked “Carters Made in U.S.A.” This bottle is a master ink bottle and dates from the first quarter of the 20th century and is likely associated with the U.S. Navy utilization of Fort Jefferson in the early 20th century

Figure 53. An example of the braided cable located at several Bastions that terminated within the moat.

It is hypothesized that the cable is associated with a lighting strike/electrical protection system

Figure 54. A typical photogrammetric (dense cloud representation) section of a portion of the counterscarp collected during the November 2021 field operations

Figure 55. Details of the construction of the counterscarp as represented in the 1854 status of construction (see Figure 4). This representation should be considered the ‘as built’ historical design of the structure. NPS image, TIC# 135172

Figure 56. Representative cross sections of the counterscarp following various repairs as proposed in

1962. Note the extensive use of iron in the foundation and the use of riprap along the ocean facing side of the counterscarp. TIC# 60020 and 60021

Figure 57. Representative cross sections of the counterscarp following various repairs as proposed in

1968. Note the keyed notches connecting the slab to the wall sections as well as the longitudinal and traverse steel reinforcement (#4 rebar in the slab and #6 rebar in the wall). TIC# 41000

Figure 58. Representative cross sections of the counterscarp following various repairs as proposed in

1985. Note the use of rebar pins to secure the poured concrete walking surface to the counterscarp. The remains of this repair are readily visible along Front 5. TIC# 80005

Figure 59. A hand marked delineation of fields of fire and maximum ranges of large armament at Fort

Jefferson in 1879. NPS image, TIC# 134169

Figure 60. Image of a bovine mandible located within the south swim beach survey block while investigating magnetic anomalies in the area. While this item was obviously not the cause for the magnetic anomaly, it is likely eroded from the nearby swim beach where there was once a butcher shop to supply meat for workers at the fort

Figure 61. Google Earth image showing the remains of the fishing Celeste Joan. Areas highlighted in red represent large tanks associated with the wreckage and are threats to navigation. The yellow polygon includes a sparse artifact scatter mostly consisting of steel cable, a ladder, and other small fittings associated with the vessel as well as the larger reef that likely led to its demise. Given the proximity of both the reef and the vessel remains to the surface, this area should be avoided

Figure 62. Map identifying sensitive areas that should be avoided during operations. The orange areas identify specific hazards to navigation (reef/wreck) as well as the location of a historic anchor immediately adjacent to the counterscarp wall. This anchor is not in its original context and could be moved if needed. Red circles denote a 10m buffer around unidentified (buried) anomalies

Tables

Table 1. South Swim Beach anomalies that were located and documented during investigations

Table 2. North Swim Beach anomalies that were located and documented during investigations

Introduction

Dry Tortugas National Park is located approximately 70 miles west off Key West in the Straits of

Florida and encompasses seven islands within its 101-mile boundary at the westernmost extent of the Florida Keys. Garden Key, the second largest island within the park, is the site of Fort

Jefferson, one of the most ambitious and expansive 19th century American coastal fortifications.

Originally designed to protect shipping access to the Gulf of Mexico while providing an outpost to protect United States’ interests throughout South America and the Caribbean, the construction of the fort, which started in 1846 was never completed. Logistical challenges coupled with the onset of the American Civil War delayed work at the fort, which was utilized as a military prison during the conflict (NPS 2017:3-4). While the fort was never formally completed and never fired a shot in aggression, the structure stands as one of the largest masonry structures in the western hemisphere and one of the best unaltered examples of the Third System of Coastal Defenses in the United States.

The extreme isolation of the Dry Tortugas has always plagued any development of the islands there. The lack of any reliable source of freshwater, distance from rapid or convenient resupply, and the harsh marine environment have worked in tandem to hinder nearly every undertaking at the Fort since construction began in 1846. While the design, supply, and construction of the Fort and its ancillary structures were complicated by these factors, the continued maintenance and repair of these structures often proved a greater challenge. This was particularly true when contending with the impacts of major storm events and hurricanes that historically landed on or passed near the Dry Tortugas.

These threats and challenges have not relented and have multiplied with climate change and sea level rise drastically increasing the frequency and intensity of major storms impacting the Fort.

While the fortification and many of the ancillary structures constructed on Garden Key were designed to withstand battle, a lack of maintenance during extended periods as well as constant exposure to the caustic marine environment have degraded and considerably weakened many of the surviving structures. Nowhere can this degradation and subsequent damage be seen more than in the counterscarp and moat surrounding Fort Jefferson.

The design and construction of the counterscarp around Fort Jefferson was vital in both the construction and defense of the Fort. First, the structure was designed to form a moat or ditch around the fortification, providing another passive defense against ground assault. The construction of the counterscarp acted as a cofferdam where sections of the moat could be drained, and the below ground foundations of cisterns could be constructed without being constantly inundated with each tide. Finally, the counterscarp served as a physical breakwater protecting the fort walls from being damaged by waves during storms (Hitchcock and Byrd

2011:69)

Given its designed purpose, it is not surprising that the surviving structure has sustained significant damage since its construction and was under constant repair between 1849-1873. The first reported damage to the structure occurred in 1853, only four years after initial construction began, when several cracks were observed in portions of the newly built counterscarp (Manucy

1961:53). From that time the counterscarp has been subject to considerable damage throughout the years, with several significant breaches forming, usually following major storm events.

While the counterscarp itself has been damaged and even breached numerous times, the structure has continuously been rebuilt, maintaining the same footprint as the original. Overall, these repairs were generally in keeping with the designed construction techniques.

The counterscarp was again significantly damaged during the 2017 hurricane season resulting in one large breach in the counterscarp, the topping and removal of portions of brick facade, and the significant undercutting and weakening of the counterscarp in several locations resulting in visitor safety concerns. This damage has also significantly impacted the experience of visitors traversing the area around the Fort, interrupting pedestrian flow and depriving them of the experience and views offered while circumnavigating the exterior of Fort Jefferson.

In addition to the physical damage to the counterscarp, the major storms impacting the Dry

Tortugas during the 2017 season inundated portions of Garden Key and caused significant sediment movement across the island. While erosion was noted in several areas, the deposition of large quantities of sediment have caused more immediate management concerns. Considerable amounts of sediment were deposited within the moat between the scarp and counterscarp in two locations completely blocking waterflow to a portion of the moat. This lack of waterflow has caused anaerobic conditions within the blocked portion of the moat resulting in the water to become stagnant and filled with the rotting remains of the flora and fauna that once thrived there.

Again, like the early damage noted on the counterscarp, the challenges of maintaining waterflow within the moat is not just a contemporary issue. The moat, which at one time served as drainage for the Fort’s sanitary facilities, had routine issues with sedimentation and stagnation. Additional sediment deposits in the Park’s finger piers where both NPS and private recreational craft are moored have reduced the depth in the slips, making it more difficult for larger vessels to utilize the docks safely.

While this type of damage has occurred in the past, the unprecedented impacts from the 2017 storm season combined with the effects of climate change, sea level rise, and deferred maintenance has culminated in more significant damages that require action. As such, a series of projects are currently in the planning stages to address these issues.

Description of Undertaking From PEPC 86259

Damages are primarily to the Counterscarp and Scarp and include scouring and undercutting of the structure, numerous missing brick veneer, stress fractures, and the total collapse and breach of approximately 50 LF of the wall along Front 3 and a complete removal of the top 18" of the structure along Front 4.

This project will include engineering, design and the repairs necessary for the hardening the

Counterscarp and Scarp against future storm energy, and maintain a significant visitor attraction. This work will also protect more than $20 million recently invested to stabilize the

Scarp along Fronts 2, 3, 4 and 5. Repairs will include 50' long x 8' wide x 10 deep wall repair the 60 LF wall cap which is 8' wide and a 18" concrete cap. Repairs also include an additional 0.35 miles of Counterscarp face and 9200 LF of scouring beneath the wall; and replacing missing brick from the veneer. Also included is rebuilding the foot bridge across the moat. The repair method has not been determined, but past repair of the Counterscarp in the 1970's required the construction of a sheet pile coffer dam so the repair work could be conducted in a dry environment.

From PEPC

This project will accomplish two major objectives: 1) dredge of the slips and finger piers at the Garden Key WaterFront to allow for continued unobstructed recreational and park use of the dock and slips, and 2) and removal significant filling of sand and silt material in the moat completely surrounding Fort Jefferson. A total area of approximately 2,400 cubic yds.s will be dredged between the two areas identified in the project. This project will be accomplished by contract and the recovered sand will be used in the repair of the Fort

Jefferson Counterscarp and Scarp (PMIS 244370).

Legal Location for the Undertaking

Fort Jefferson is located on Garden Key in the Dry Tortugas in Monroe County, Florida. The Dry

Tortugas are located in the eastern Gulf of Mexico approximately 117km west of Key West, Florida (bounded by coordinates 2433’-2444’ N and 8246’-8258’ W) (Murphy 1993:52).

The Fort is not located in an area covered by the Public Land Survey System, but is located on the DRY TORTUGAS OE E (2021) United States Geological Survey Quadrangle map. It is also located on National Oceanographic and Atmospheric Administration (NOAA) charts 11438, 11434, and 11420.

Local Environment

Fort Jefferson is located on Garden Key in the Dry Tortugas National Park. Its 100-square-mile boundary encompasses 64,701 acres and includes seven keys composed of coral bedrock, and sand. The Park is situated in the westernmost part of the Florida Keys, which extend from the keys in the Dry Tortugas east following the Gulf Stream to the east coast of Florida, terminating at Soldier Key near Miami. Dry Tortugas National Park is located near the southwestern edge of the Florida Platform. This broad, flat, carbonate platform is only partially exposed above sea level as the Florida peninsula and has varied in depth over time. Today, the Dry Tortugas are a complex of reefs, shallow sand banks, and low relief islands atop Pleistocene limestone bedrock of the Florida Platform (Multer 1977 in Port 2014).

There are currently seven small keys composed of sand atop coral bedrock. Given their composition, the exact size and shapes of the islands are in constant flux as they are impacted by seas, storms, and currents and some of the smaller keys can be completely inundated at certain times of the year. Garden Key is the second largest key within the park encompassing approximately 20 acres of dry land. Fort Jefferson spans the majority of Garden Key enclosing the parade ground within the fort walls (Hitchcock and Byrd 2011). The lands located outside of the fort walls on Garden Key include modern infrastructure including a marina, dock, north and south historic (U.S. Navy) period coaling docks, buried fuel tanks and utilities, campground, picnic area, and swimming beach.

The benthic environment within the immediate vicinity of Garden Key is predominately populated by a variety of Subsurface Aquatic Vegetation (SAV) including Thalassia testudinum, Halodule wrightii, and Syringodium filiforme of varying coverage densities (Schneider

Engineering and Consulting 2020). Additional limestone outcroppings (patch reefs) host a wide variety of sea fans and corals. The counterscarp itself, as well as the surrounding debris provide habitat for many coral species. Schneider Engineering and Consulting (2020) identified 32 different species of coral forming a total of 4,151 individual coral colonies along the counterscarp. These coral species were evenly split between hard and soft coral types and were overwhelmingly dominated by Pseudodiploria clivosa, Eunicea flexuosa, and Gorgonia ventalina (Schneider Engineering and Consulting 2020:7).

Area of Potential Effect (APE)

While the extent and specific nature of the future repairs are still in development, the project will constitute a significant undertaking with an area of potential effect (APE) extending well beyond the counterscarp itself. Given the fact that the submerged bottomlands surrounding Fort Jefferson had not yet been systematically surveyed by Secretary of the Interior qualified marine archeologists, this project sought to proactively survey the area in preparation for the upcoming repairs. While the APE for the proposed undertaking is limited to the “immediate vicinity” of

Fort Jefferson interpreted as the shallow areas within the swim area extending approximately

100-200 m from the counterscarp, this project offered an opportunity to survey the entirety of the shallow waters north and west of Garden Key.

Given the scope of this undertaking, the APE will extend beyond the footprints of the individual proposed structures and improvements and should include staging areas as well as right of ways, pipelines, and the paths of any new or expanded below surface infrastructure. As such, it is proposed here that the APE include the entire developed area within the Gulf Coast

Administrative Area. The APE, detailed in Figure 1 below, includes the entire developed area and consists of one polygon with a total area of 82,455m2 around the project area, staging area, and dredge spoil location.

Figure 1. APE MAP

Records Checks Record checks were made with the Florida Master Site File (FMSF), Dry Tortugas and

Everglades National Park files, park Geographic Information System databases, and the National

Park Service’s Cultural Resource Inventory System (CRIS). One listed resource was identified within the APE, the Fort Jefferson Complex (08MO00229, DRTO-29). Within this complex, this proposed undertaking specifically addresses repairs to a subsite within the fortification, the counterscarp and moat (HS-11). While there has been extensive documentation and research conducted on the fortification as a complex and/or landscape, very little of the previous archaeological or historical research at the park specifically examined the counterscarp and moat, or the area surrounding it. Documents that include the counterscarp, moat, and the submerged environment, include Manacy (1961), Bearss (1983), Murphy (1993), Lord Aeck Sargent (2006), and National Park Service (2017).

Given these deficiencies, historic research should be undertaken specifically to examine the construction history of the counterscarp and moat as well as any major maintenance issues or significant damages sustained throughout its history. This historiography is vital in understanding the original construction of the features to be repaired as well as significant management actions that may impact its current condition, or the presence/condition of possible archeological deposits within the surrounding APE. Additional information gathered through more targeted research efforts will also be vital in understanding the vulnerabilities of the current structure and will allow for more accurate testing and modeling of the current surviving structure that will in turn produce better informed plans for its repair. Additional archeological survey is also needed of the submerged bottom lands throughout the APE. Given the shallow waters surrounding the fort and the enclosed moat, a variety of non-intrusive survey techniques would identify the presence of significant archeological remains in the APE.

Property History

While the repairs and improvements to the counterscarp and moat proposed in this undertaking may seem relatively benign, a thorough understanding of the history of these structures is required for the development of treatment solutions in keeping with historic preservation standards. Examining the original design and construction characteristics of the structures, management issues associated with them, and a concise overview of major damages and restoration efforts will provide information required to develop a well-informed restoration.

Though an overview of this information is provided in Hitchcock and Byrd (2011:69-71) it is minimal and specifically in reference to the development of the overall cultural landscape at Fort

Jefferson. The major restoration of the counterscarp requires a higher level of understanding of its construction, maintenance, and repair in order to develop a repair strategy with historic preservation in mind. This level of research will additionally provide insights into how to best manage these resources as the impacts of sea level rise and climate change increasingly impact these imperiled resources.

One important note should be made in regard to the changes in numbering of the Fronts and

Bastions over time. The original numbering of the Fronts utilized by the military throughout the

19th century is different than those utilized in the 20th century while Fort Jefferson has been managed by the NPS. This document makes no distinction between the numbering of individual

Fronts historically versus contemporarily (i.e., Front X, historic Front X) as has been done in some management documents focusing on Fort Jefferson unless relevant to the interpretation of historic or archeological features. Discussion of repairs throughout the 20th century will use the modern designation of the Fronts.

Planning

Though the moat and counterscarp are often seen as minor ancillary structures when compared to the complexity and scale of larger features associated with Fort Jefferson, each component served vital roles in both the initial construction and defense of the main facilities. As designed, purpose of the counterscarp was two-fold. First, the counterscarp served to form the outer perimeter of a moat (often referred to as a “ditch”) to obtain “a depth of water too great for fording… and by preventing boats from reaching the scarp, thus strengthening the work against assault” (Wright 1848 in Bearss 1983:97). As such, the area between the counterscarp and scarp would be excavated to -6 ft. below high tide (Bearss 1983:97). Its second purpose was to protect the foundations of the scarp from wave action (ibid). Interestingly, the construction of the counterscarp wall for this purpose was questioned by some as it was thought that cheaper alternatives may have been equally effective. In a letter to Chief Engineer Joseph Totten, Lt.

Horatio Wright of the U.S. Army Corps of Engineers argued that the construction of an

“embankment” in the same approximate location of the proposed counterscarp would accomplish the same goal at a fraction of the price (Bearss 1983:98).

Despite these arguments, Totten opted to proceed with the construction of the counterscarp, which would require the construction of a cofferdam for both the excavation of sediment within it as well as the pouring of a concrete foundation. Wright’s proposed cofferdam consisted of three lines of piles (either square timbers or rough-cut logs) at a distance of 8 ft. from center to center and parallel to the direction of the counterscarp. These openings between the successive walls of the cofferdam would be 11’2” and 9’4” with sheeting designed to “keep out the sand and make smooth water in the interior”, was to be 3-inch plank while the floor of the platform was to be 2-inch plank (Bearss 1983:98, also see Figures 2 and 3). While the components of the cofferdam were originally designed to not extend below the foundations of the structure and would be removed and reused along other portions of the counterscarp, Totten later advised that the outside sheet piles would be cut “off at reference (0) and covered with sand… the others should be either “drawn or cut off” as low as possible” (Bearss 1983:100).

The foundation of the counterscarp would be constructed within the confines of the cofferdam and was originally proposed to be formed of a concrete foundation starting -5 ft. below mean high water that would extend to the waterline, after which the masonry wall would be started.

While there was some discussion of various cost cutting measures, Totten eventually dictated that the upper portions of the counterscarp should be constructed with a vertical wall “facing of very hard bricks, laid in quality cement mortar, in successive courses of 2-1/2 bricks, backed by concrete, coped by blocks of similar brickwork (the blocks divided by, and laid in joints of mastic) would better resist the dynamic faces of the ocean (ibid). While taking into account

Totten’s requests and refining his plans for the foundations of the counterscarp, Wright proposed modifications in the profile of the counterscarp and its foundations in which he left the uppermost portion of the concrete foundation exposed to the battering of the sea (Figure 4).

Figure 2. Detail from the 1848 sketch of a proposed cofferdam for the counterscarp wall at Fort Jefferson. This inset image highlights the construction of the cofferdam. Note that this proposed design changes slightly over time. NPS image, TIC#137468.

Figure 3. Detail from the 1848 sketch of the counterscarp wall at Fort Jefferson. This inset image highlights an alternative design for the counterscarp wall. Note that this proposed design changes slightly over time. NPS image, TIC#137468.

While this was not ideal, Wright reassured the Department that it would be covered with sand

“either through wave action or currents” and wherever this action did not occur naturally, “it was to be secured by short coral piers, perpendicular to the counterscarp” (sand groins). He noted that this was required because where sand did not naturally accumulate along the wall, it would tend to erode and possibly undercut the structure. He also argued that the number and placement of groins needed to trap sand along the wall could be determined by observing the action of the surf and sand as construction proceeded (ibid).

One final note to be made regarding the planning for the construction of the counterscarp that may be relevant to the planning of this undertaking involves the composition of the foundation itself. Wright recommended that the foundation and superstructure be laid in sand and cement mortar without lime. He believed that no lime should be added to the mortar as it had been his experience that when used in areas exposed to water or weather, that it often failed (Bearss

1983:101). There is some level of discourse regarding the use of lime in the original design of the structure, as it was later pointed out that damages to the counterscarp observed shortly after the start of construction (to be discussed later) may in fact be attributed to the presence or absence of lime in the foundation (ibid).

Construction

While general plans were being formed, the actual construction of the counterscarp could not proceed until the dimensions and locations of the Bastion were finalized (ibid). Wright received approval from Totten to commence construction of the counterscarp in late 1848 and construction commenced in February 1849 (Bearss 1983:101, Manucy 1961:212). As the preliminary construction efforts began throughout the winter of 1848-1849, Wright further refined the design for a cofferdam after watching nearby dredge operations at the fort. As an experiment, he observed 112 ft. of sheet piling for the cofferdam on Front “A” along the northern portion of Garden Key were driven. The piles were 5 ft. long, 2 in. thick plank were driven on each side of the interior framing. The area bounded by the pilings was then filled with sand, the space left for the foundation bailed of water, sediment excavated to the desired depth and concrete poured. Material excavated from the counterscarp was placed onto Garden Key in the area to become the parade ground (Bearss 1983:127). This process was attempted as a cheaper alternative to pouring concrete underwater. He specifically highlighted the costs savings as two men with heavy mauls could drive about 100 running ft. of outer sheeting per day. Wright’s experiment was successful and this procedure was adopted for the entire extent of the counterscarp, as well as the scarp (Bearss 1983:102).

By mid-April 1849, the cofferdam on Fronts “A” and “B” were completed though the actual progress on the construction of the counterscarp was minimal due to labor and equipment needs.

By the end of the fiscal year in September 1849, Wright listed the length of the counterscarp finished (except for coping) at 510 ft. The mostly completed portions of the counterscarp included the brickwork atop the concrete foundation with brick facings measuring an average

1’4” in thickness. These facings were laid in three courses of bricks in height of 2-1/2, 2, and 1-

1/2 bricks thick respectively with the mortar used the same composition as the foundation. The

5’4” section core between was filled with the concrete and coral mixture. The composition of the concrete at this stage of construction being one part cement, two parts sand, and four parts coral.

He also noted that the quantity of the concrete laid exceeded the volume of coral by 13 percent and that ocean water was used for mixing the concrete. In addition to the mostly completed portions of counterscarp, by the end of fiscal year 1849 an additional 518 ft. of foundation and

1,010 ft. of counterscarp construction was completed (Bearss 1983:103).

By September 30, 1850 an additional 1,759 ft. of cofferdam and 1,307 ft. of counterscarp had been completed. The newly constructed counterscarp was comprised of 4,356 cubic yards of brick and concrete masonry while 2,080 cubic yards of material had been excavated (an average depth of -6.5 ft. MLW) to facilitate its construction. At this point, the cofferdam extended across three Fronts and two connecting curtains. The following fiscal year construction of the counterscarp extended onto the shoal (Front No. 2) and in the area off Front No. 3 “beyond the shore” (See Figure 4). At this point, two areas were left open for the eventual installation of sluice gates and two others were only completed to within 2 ft. of MLW to provide access for vessels bringing materials into the work area (Bearss 104).

While often hindered by lack of funding, tropical storms, delayed materials, or labor challenges including both the ability to find both enslaved and paid skilled laborers, the construction of the counterscarp steadily proceeded in this manner over the next several years. It was about this time that the first damages were reported to the counterscarp. In 1852, a hurricane had removed a considerable amount of sediment from the counterscarp in Front of Front No. 5 and only 2 ft. of sediment remained before the bottom of the foundation would be exposed and possibly undermined. Efforts were immediately made to replace the sand to a level approximately 8 ft.

wide and level with the top of the foundation. This embankment was constructed along the entire extent of Front No. 5 and for a short distance of Front No. 6 (Bearss 1983:106). While this effort could be seen as the first stabilization effort for the counterscarp, it may have not been overly successful. A little more than a year later in September 1853 the first cracks were identified in the counterscarp. These cracks were considered minor at the time and were at first attributed by

Wright as the result of unequal expansion between the brick and concrete used in its construction. By October 1855, the cracks had increased in size and number and were then attributed to a “slacking of the lime” but remained insignificant (Bearss 1983:106-107). While subsidence later proved to be a significant factor in planning for the construction of many of the larger structures within Fort Jefferson, it was not considered an issue at the time for the counterscarp as it was believed that its weight was insufficient to cause such a settlement (Bearss

1983:106).

It was also during this time that the installation of a vital ancillary system to the fortification began that would have significant impacts to the future maintenance of the moat and counterscarp. While not directly within the APE of this project, the design and construction of

Fort Jefferson’s sanitation system plays a major role in future maintenance issues for both structures. While discussion of the design of the system is beyond the scope of this survey, the approved design included the placement of three drains through the scarp into the moat at three locations. These drains would be flushed by tidal oscillation which would pull waste from privies and vaults within the fort into the moat where it would then be flushed by the circulation of water through the installation of two “sluice gates”.

In theory, the two sluice gates would act to flush the moat of stagnant water by allowing in a rush of water from one side of the moat and pushing water out from a second gate located on the opposite end of the moat.

Figure 4. Image highlighting the status of construction of Fort Jefferson in 1854. Note the inset detail images that highlight the specific construction characteristics of the counterscarp. NPS image, TIC# 135172.

The exact location and construction of these gates were discussed, and the decision was made to make the openings nearly opposite of one another while avoiding their placement along either

Fronts Nos. 1, 5, or 6 as these were the Fronts most exposed to heavy seas, as well as possible naval bombardment (Bearss 1983:105). While the sluice gates were installed as designed, their constant need for repair proved problematic for the maintenance of water flow within the moat, which in turn later impacted the drainage of privies and the public sanitation system within the fort.

Work slowly continued through the 1850s and 1860s as the onset of the American Civil War drastically increased activity at the fort. The progress of the cofferdam, counterscarp foundation, and brick masonry progressed slowly. As entire Fronts of the counterscarps were completed, cofferdams were constructed to drain portions of the moat which was then excavated to the original designed depth of 5-6 ft. In 1863, instructions were relayed to use fill from the excavation of the “ditch” between the counterscarp and scarp to build up the area between the scarp and the breast-height wall of the barbette tier. While there were reservations about using the excavated material which contained little marl, and though it was likely unsuitable for the purpose, the sediment was lifted directly from the moat to the terreplein above where it was utilized in construction.

The following year the excavation of the “ditch” proved extremely difficult. While laborers under the directions of artisans continued to construct the cofferdam at Bastion E, serious difficulties were encountered in excavating the moat at Front No. 6 which had prevented finishing the parapet at that location. Three steam engines, two screw pumps, and four 12-inch lift pumps had been unable to keep the ditch dewatered and thus prevented further excavation

(Bearss 1983:256).

While there was no funding allotted for construction activities at Fort Jefferson during fiscal year

1865 (a recurring occurrence following the end of the war), construction continued with remaining funds and the use of convict labor. While excavation of the moat continued, it was slower than desired. The moat parallel to Front No. 6 was completed on the last day of

November. By 1866, two-thirds of the moat in Front of curtain No. 3 had been fully excavated and the counterscarp constructed for the same distance. More than 80,000 cubic ft. (2,265 cubic meters) of sand had been removed from the ditch and embanked (Bearss 1983:263).

While work slowly progressed, the arrival of yellow fever to the garrison at Fort Jefferson would further highlight the need for a clear moat to ensure functioning sanitary facilities for those living there. Bearss (1983:290) notes that on the 19th of August 1867, a soldier in Company K quartered in the “fungus grown” upper casemates on Front No. 4 overlooking the unfinished ditch was sick. The following months would see hundreds of cases of the disease that resulted in dozens dead. Focus throughout late 1867 shifted from the various construction activities at the fort to caring for the sick and maintaining effective isolation within the confines of the fort.

Subsequent investigation into the cause of the yellow fever outbreak revealed two possible causes related to the condition of the moat. It was noted that the unfinished condition of the ditches on Fronts Nos. 3 and 4 that resulted in its exposed bottom at low tide and the situations of the sewers, which had become clogged and the vaults unable to be flushed as designed as contributing to the cause and severity of the outbreak (Bearss 1983:292).

Following the outbreak of yellow fever in late 1867, more attention was given to those sanitary facilities thought to have contributed to the start and spread of the outbreak. Foul odors permeating through the fort’s sinks in the hospital and officer’s quarters forced an examination of the sewer system. As designed, sinks and privies were supposed to empty into a series of vaults which were to be flushed into the moat, either through tidal oscillation or through the addition of water to the system. Tidal oscillation within the moat, once completed, would then carry waste to sea. It was argued that while the design was sound, the fort remained in a state of constant construction and as such, many of the systems would not function as designed.

As such, efforts were made to rectify the sanitation issue through a renewed effort to complete the moat and attempt to remedy the immediate health hazard. Convict labor was again utilized as obtaining reliable workers elsewhere had been difficult. In addition to completing the excavation of the moat, saltwater pumps were installed in the parade ground to provide salt water to flush the systems within the complex. Another idea was proposed that rather than having the sewers drain into the unfinished moat, pipelines be constructed crossing the moat and emptying beyond the counterscarp in the open ocean. While considered, the additional pipeline was not deemed necessary as long as the moat was eventually completed and flushed as originally designed

(Bearss 1983:297-298).

Given the seriousness of the recent yellow fever outbreak, efforts to rectify the sanitation issues at the fort began in earnest shortly after. Between June 15-30th 1868, 40 laborers nearly completed excavating the moat parallel to Front No. 3 to an average depth of 4’6” until the pump malfunctioned four days into the project. Efforts were shifted to excavating the moat at Front No.

4. In each case, spoil from the excavation was lifted to the top of the adjacent scarp for use on the parapet or terreplein (Bearss 1983:302). Nearing completion, the supervising engineer requested an additional $5,000 in funding to complete the counterscarp, but no funding was allocated for the purpose. Working from previous funding, minimal work on the counterscarp was completed, including continued excavations of the moat at Fronts No. 3 and 4 with the spoil being deposited on the parade or outside of the moat wall. Additionally, during this time a sluice gate was installed on Front No. 6 (Bearss 1983:307).

The lack of funding for the fortification became an increasingly persistent issue, drastically reducing the amount of work undertaken over the next several years. Little work was completed on the counterscarp or moat until March 1871 when a meager appropriation was made for the maintenance of the barracks and seawall. Between 500 and 600 ft. of counterscarp would be built but would require the purchase of new pumps as well as raw material for the job (Bearss

1983:312-313). Work was postponed until after November to avoid quarantine during the summer, when disease was most prevalent. Once resumed, work on the counterscarp focused on the portion of the moat west of the drawbridge. By mid-January 1872, the concrete foundation of that portion of the counterscarp had been laid, and 283 running ft. of superstructure built to the low water mark (Bearss 1983:314).

Just a few months later, the post commander on site urged engineers to resume work on the counterscarp as the surf had washed a large amount of sand through a breach into the moat forming a bar. This blockage prevented the tidal flow from pulling sewage away from the fort

(Bearss 1983:314-315). When work resumed in December 1872,…