W912QR-50146144_Specs_Vol2_AppA-C-0000.pdf

Solicitation For Multiple Award Task Order Contracts for Construction Design Bid Build Rivers and Marine Activities within the Great Lakes and Ohio River Division Mission Boundaries.

NE CONSTRUCTION MATOC

Covington Emergency Levee Repair Covington, Kentucky

P2: 511693 Design-Build

Specifications - Vol. 2 - APPENDIX A-C Certified Final Design

05 February 2025

W912QR-50146144

MATOC

W912QR21D0056-21D0060, 21D0085-21D0086

All requirements of the base MATOC Volume 1 of 2 shall apply to this Task Order unless noted otherwise.

ARIMS: 1200 Disposition: Permanent

W912QR-50146144_Specs_Vol2_AppA-C-0000

A achment 3‐ Exis ng Subsurface Data

COVINGTON EMERGENCY LEVEE REPAIR

COVINGTON, KENTUCKY

Request For Proposal Specification 01 02 10.00 0: Section 5

Appendix A Geotechnical History and Design Guidance

OCTOBER 2024

P2# 511693

U.S. ARMY CORPS OF ENGINEERS LOUISVILLE DISTRICT

Covington Emergency Levee Repair Request For Proposal Covington, Kentucky Specification 01 02 10.00 0 Section 5, Appendix A

LEVEE SAFETY AND GEOTECHNICAL DESIGN

1. GEOLOGY

The regional bedrock of Covington, Kentucky is composed of interbedded soft shales and thin beds of limestone, which were deposited in shallow marine environments approximately 450 million years ago during the Ordovician period (Figure 1-1). The area was covered by a shallow inland sea, which left behind significant layers of carbonate rocks interbedded with finer-grained shales. The regional stratigraphy of the Ordovician period includes formations such as the Kope, Fairview, and Bellevue, which are visible in the cliffs and exposures along river valleys in the project vicinity. The Ordovician rocks in Covington are relatively flat-lying, with only minor undulations caused by regional folding and faulting.

Figure 1-1: Geology of Kentucky

Covington is positioned on the northeastern flank of the Cincinnati Arch, which is a broad, domal uplift that gently dips away from the city towards the east and west. The presence of this arch has influenced the current drainage patterns and the regional landscape, contributing to the development of the Ohio River valley. The Ohio River has incised deeply into these rocks, creating a steep-walled valley with elevations that vary from around 460 to 500 feet above sea level at the valley floor to about 540 to 550 feet on the adjacent terraces.

The surficial deposits in Covington are primarily composed of outwash and river-deposited silts, silty clays, sands, and gravels (Figure1-2: Surficial Geology Near Covington). The alluvial materials were transported and deposited by the Ohio River and its tributaries, while the outwash deposits were deposited by glacial meltwater. The thickness of these unconsolidated sediments can vary significantly, with bedrock often encountered at depths greater than 50 feet beneath the ground surface in terrace areas.

Figure1-2: Surficial Geology Near Covington

The subsurface conditions along the levee system are complex, owing to a combination of natural and man‐ made deposits. There is documented presence of loose fill material throughout the floodplain, consisting of cinders and other urban debris, particularly in areas that have been subject to historical development and industrial activities. This artificial fill material is underlain by a natural surface layer of clay‐silt with minor sand content. Beneath these surface layers, bedrock is encountered at varying depths. Refer to Geotechnical Appendices A and B for geotechnical boring information and cross-sections in the project vicinity.

Flooding along the Ohio River has shaped the region’s hydrologic and geomorphic character. During major flood events, the river overtops its banks, inundating the flat-lying floodplain areas. The levee system in Covington, constructed as part of the broader flood protection efforts, aims to mitigate the impacts of these floods by redirecting and controlling floodwaters. The levee system in Covington is underlain by a complex stratigraphy of unconsolidated floodplain deposits and bedrock formations.

Along the floodwall, the surface deposits transition from recent alluvial materials to older Pleistocene terraces, which reflect changes in river flow patterns over time. The Pleistocene deposits include stratified sands and gravels that are interbedded with silts and clays, indicating periods of active channel migration and floodplain aggradation.

The groundwater conditions in Covington, Kentucky, especially along the Licking River, include notable artesian characteristics in certain areas. Artesian conditions occur when groundwater is under pressure, allowing it to rise above the level of the aquifer without the need for pumping.

According to data from the USGS Groundwater Data for Kentucky, the region’s geology can create confined aquifers that contribute to artesian pressures in some locations. The historical presence of wells along the Licking River, particularly those that tap into deeper sandstone aquifers, have shown fluctuations in water levels indicative of these conditions.

Additional information from the USGS Water Data for Kentucky highlights that artesian flows have been observed in wells where the natural pressure exceeds that of the surface elevation, leading to a rise of groundwater in certain low-lying areas, particularly near the riverbanks and alluvial deposits. This phenomenon is influenced by the hydrogeological setting of Covington, where the interaction between the Ohio and Licking Rivers contributes to variations in groundwater pressure.

Artesian conditions were encountered in two (2) borings drilled after the 1956 slope failure, as well as the drainage pits installed in 1958/1959. However, the artesian conditions were not detected in the borings drilled for the original levee construction.

2. HISTORY OF SLOPE CONDITIONS

A brief history of the levee slope events is summarized below. The events discussed include: 1953 Sewer Trench Event, 1956 Slope Movement, and Other Slope Movements (1983, 2012, 2019, and 2022). See Figure 2-1 below for the general location of the past levee events, as well as additional nearby features. See Geotechnical Appendices B through E for geotechnical boring data (Appendix B) and cross-sections (Appendix C), selected USACE geotechnical documents (Appendix D), and engineering reports performed by others for this area of the Covington levee system (Appendix E).

Figure 2-1: Location of Past Levee Events

2019 NMonoliths Replaced

1964/1965 Sewer Line

Drainage Pits (Typ.)

1953 Sewer Excavation

Collapse2024

Note: Locations are approximate

Gatewell 16

(creep)

1953 Sewer Trench Event The first indication of possible slope stability issues was reported on June 8, 1953, between Stations 176+00 to 178+00 during the excavation for a Sanitation District No. 1 sanitary sewer trench near the toe of the constructed levee. The sewer installation (by a Sanitation District No. 1 contractor) was not part of the levee construction contract but coincided with the levee construction. The sewer line and the toe of the embankment diverge as the stations increase from Station 176+00 to 178+00, with the sewer line crossing the levee embankment toe near Station 176+00 and increasing to more than 100 feet beyond the toe of the embankment with increasing stations.

The sanitary sewer was 33 inches in diameter. The excavation was estimated to be on the order of 13 to 15 feet deep based on the as-builts reviewed and assumed to be 5 to 6 feet wide based on typical sewer construction for the pipe diameter. The bottom of the excavation reportedly heaved up during installation, with obvious movement of the landside (uphill) excavation face into the sewer trench. The movement was attributed to insufficient bracing procedures. The stockpiling of excavated material on the ground surface adjacent to the landside excavation edge of the trench was believed to have contributed to the localized trench failure. The slippage reportedly took place in a brown clay layer near the contact with the underlying gray clay.

It was concluded that the effect of this slide on the existing protection, which included concrete T-wall, was insignificant, likely due to the limited nature of the failure and the increasing distance to the floodwall. The available documentation stated that the movement was localized and could be repaired by the removal and replacement of the affected soil using ordinary construction means.

Backfill of the trench with compacted fill and the use of shoring in future excavations were also recommended.

The start of repairs was delayed until August 10, 1953 (approximately 2 months) because the Sanitation District No. 1’s engineer wanted to confirm no additional movement of the trench was occurring before backfilling the trench. Excavation of the sewer continued in August 1953 once the Sanitation District’s engineer was satisfied that no additional movement was occurring. Movement of the T-wall from Station 173+22 to 177+87 was noted in a September 3, 1953, letter, and sewer excavation operations in the area of the wall reportedly were sped up and were to be finished in the “next couple of days.” The movement between Monoliths 39/40, 48/49, and 49/50 was described as a “perceptible increase” since the writer’s last observation on July 28, 1953 (approximately 1 month prior), with “definitive transverse shear movement of about 1-inch occurring between the tops of Monoliths 46/47.” It is important to note that it was later discovered that the floodwall was constructed on 10 to 12 feet of uncontrolled fill that was concealing a steep natural drop-off. Differential settlement or movement due to the likely poorly compacted, poorly benched uncontrolled (not engineered) fill below the tall floodwall (total wall height of 20 feet, with 15 to 19 feet exposed) being founded on a steep natural slope could easily account for the wall movement observed, instead of the wall movement being related to the sewer excavation. The sewer excavation was only 13 to 15 feet deep and was located at least 65 feet away from the edge of the wall foundation at the closest point (with the intervening distance increasing quickly). The minimum distance, which was more than 4 times the depth of the excavation, generally would be a sufficient distance from the wall that the excavation activities would not be expected to impact the wall in the absence of unusual adverse conditions.

It was recommended in a September 3, 1953, letter that “frequent periodic measurements of the wall and riverbank be continued for at least another 3 months or longer if movement persists.”

The toe of the embankment was “cut back to good fill” and backfilled concurrently with the sewer. The 1953 letter writer concluded that the backfilling operations being undertaken would minimize further movement of the wall. According to the Remedial Measures Report dated September 8, 1959, the floodwall reportedly “suffered only minor settlement following construction of the sanitary sewer as evidenced by a series of wall measurements which were discontinued in 1954 when it became apparent that all movement had stopped and the slope became stabilized.” Again, as noted previously, the movement detected may have been the movement (settlement/translation) that naturally resulted from the dead weight of the large floodwall on the uncontrolled fill founded on a steep, natural drop-off.

1956 Slope Movement No more issues were noted with the slope until slope movement between Stations 171+00 to 178+00 was observed following high water in the spring months of 1956. River stage during this event reached EL 483 feet (all elevations in this section are NGVD 29) versus the Licking River normal pool elevation of EL 441.8 feet (before construction of Markland Dam). For reference, the wall extended to a depth of EL 485.1 feet. The verbal description included in Design Memorandum (DM) No. 1, dated December 31, 1956, suggested a deep, large- scale failure involving the riverside levee embankment and the riverbank was developing as a result of the event (e.g., tension cracks were observed on the levee embankment, additional tension cracks were observed at the river bank, trees rotated, earth flowed around Manhole No. 37A, and the sewer had displaced sufficiently that line-of-sight observations were unsuccessful for the 33-inch pipe). This slope movement, described as “the most critical movement to date”, is what precipitated preparation of DM No. 1.

The following possible contributing factors to the 1956 flood-related movement were identified in DM No. 1: 2 to 9 feet of debris/fill had been placed on the slope between the toe of the embankment and the water’s edge since the pre-construction topography was obtained and several feet of the embankment toe at the water’s edge had been eroded or experienced material removal by other physical means, such as barge maneuvering. DM No. 1 concluded “that the submergence and subsequent drawdown along with the conditions noted above caused the slide” (referring to the 2 to 9 feet of fill and removal of the toe). It is important to note two things. First, removal of the debris/fill (i.e., unloading the slope) did not restore slope stability as the slope continued to move with removal of this material. Second, and more importantly, subsequent investigations revealed other more likely causes of the movement than the two sources identified in DM No. 1. Specifically, artesian conditions and a steep natural drop-off directly below the wall being obscured with fill were identified as the causes of the movement in later documents. There was an erroneous belief at the time of the 1956 DM No. 1 that the water build-up behind the well-compacted sanitary sewer backfill may have contributed to elevated hydrostatic pressures in the slope in the vicinity of the sewer line. However, subsequent subsurface investigations revealed that the source of the excess water was more likely due to a buried gravel stratum tied to an uphill source, not due to the well-compacted fill at the sewer line.

The Louisville District Office (LDO) recommended the following measures to stabilize the slope in DM No. 1:

− Provide relief of the hydrostatic pressures in the embankment and foundation with installation of horizontal drains on 100-foot center that would extend through the sewer backfill and penetrate into the embankment and foundation. These drains were to consist of a series of seven drains evenly spaced along the slide area with 8-inch perforated metal pipes to be placed in conventional trenches surround by a filter-graded sand and gravel backfill. It is noted that the drainage pits eventually installed in 1958 were much different than the drainage system described in DM No. 1, likely based on the new source of artesian water identified during excavation of the exploratory pits that eventually were converted to the drainage pits based on a field conference in August 1958.

− Flatten slope at the drop-off formed between the toe of the constructed embankment and Licking River to 3H:1V.

− Restore embankment previously constructed as part of the flood protection.

− Deferred construction consisting of the addition of underwater bank protection and riprap if the toe of the slope at the water’s edge continued to erode.

After inspection of the site by Ohio River Lakes & River Division (LRD) revealed that the cracking and sliding were more advanced than at the time the data was obtained for the 1956 DM No. 1 and because the reasons for the failure were not definitively established (per LRD) and the proposed corrective measures were of questionable effectiveness (per LRD), the design memorandum was returned to the district office for reconsideration and resubmittal per a letter dated March 21, 1957.

As a result, a Revised DM No. 1 dated September 8, 1957, was issued. The Revised DM No. 1 appeared to be the same as or substantially similar to DM No. 1, except several letters that provided additional background on the failure were appended to the front of the report and the stabilizing rock toe that was identified as “deferred construction” in DM No. 1 was included as part of the Revised DM No. 1 recommendations.

LRD approved the revised design memorandum prepared by the district office in an endorsement letter dated August 22, 1957, except it was recommended that the rock toe be omitted in the remedial construction until the need for it was more definitively proven after the reshaped river bank had passed through at least one high water season, primarily because it was surmised that navigation interests would object to the narrowing of the channel, especially at Acme Newport Steel Company, which had a loading and unloading terminal with mooring cells on the opposite bank of the river, slightly upstream of the slide. The cost of the rock toe also was mentioned along the hope that the unloading of the riverbank as proposed would reduce the tendency of the natural toe of the slope to move out (i.e., the bank would be stabilized adequately against low level slides or failures without the added support of the rock toe). The endorsement recommended investigation of the feasibility of steel sheet pile protection of the toe, either cellular or I-wall type, in the event it was determined that toe protection was necessary. In an endorsement letter dated September 13, 1957, Headquarters (HQ) approved the recommendation of LRD with respect to the revised design memorandum in view of the undesirable effect of encroaching on the navigable channel that the placement of the rock toe protection would have.

As a result, in the winter of 1957/1958, the riverbank slope was unloaded as outlined in the Revised DM No. 1 (i.e., the excess 2 to 9 feet of debris/fill that reportedly had been placed on the slope between the toe of the embankment and the water’s edge since the pre-construction topography was removed). Unfortunately, these efforts were unsuccessful (likely because, as mentioned previously, this excess soil was not the underlying issue causing the slope problems). The slope continued to move, and the contract was terminated.

Since unloading the riverbank slope was unsuccessful, the decision was made to excavate large-scale exploratory pits to better understand the subsurface (both soil and water) conditions. These exploratory pits were subsequently converted to drainage pits in a field conference on August 27, 1958, based on the source of the elevated hydrostatic head observed in the pits, a “deep” pervious gravel stratum underlying the floodwall and levee that was assumed to connect with or extend into the adjoining hill section of Covington. These pits were installed between November 20, 1958, and March 30, 1959. The drainage pits implemented were different than the finger drain-type drainage proposed in the Revised DM No. 1, likely to address the difference in the assumed source of water (i.e., a “deep” pervious gravel stratum versus water trapped behind the surficial sewer line backfill).

The drainage pits consisted of 14.5-foot square open pits created by sheet piles driven to approximately EL 458 to 460 feet (11 to 38 feet below grade).

An Interim Remedial Report dated April 3, 1959, was prepared after the drainage pits were constructed. The drainage plan was predicated on the drainage of the pervious gravel stratum below the floodwall and levee that resulted in artesian heads of 12 to 14 feet. The gravel stratum was encountered in 6 out of the 7 drainage pits constructed. The gravel stratum was not encountered in the sixth northernmost pit and was only encountered in the landward one-third of the northernmost pit, resulting in the addition of lateral drains extending 40 to 120 feet landside to improve drainage at these two locations. Flow from the pits varied, but the Interim Remedial Report concluded that the results of the drainage pits were very encouraging since there was no vertical or horizontal movement of the floodwall after Drainage Pit (DP) No. 1 became effective despite two periods of high water (January 26, 1959, and February 18, 1959). Based on soil profiles associated with the drainage pits, the gravel stratum appeared “blocked” or “trapped” since it was encountered landside of the drainage pits but not riverside of the drainage pits based on draft “Wall Repair - Sections” drawings dated August 1959.

A Remedial Report dated September 8, 1959, was prepared summarizing the progress since installation of the drainage pits. The amount of flow noted in the drainage pits was about the same as during the Interim Report, indicating a constant flow into the aquifer from an uphill source (since a leaking water main source was eliminated). The 1958 exploratory borings confirmed the effectiveness of the drainage pits, with a reported 21-foot reduction in head in one piezometer near the drainage pits and a 6-foot reduction in head in one piezometer 380 feet from the southernmost drainage pit. The drainage pits were considered “successful.” The report included the following conclusion: “The removal of the excess heads permits the redevelopment of higher grain-to-grain contact within the soil skeleton and hence the utilization of the some of the natural strength of the soil has been achieved. Here again, the time lapse becomes critical. It cannot be definitively proved that any major structural increase in strength has developed, but all indications are that with further time, which will permit vertical percolation of the entrapped moisture, there will be a strength increase in the wall foundation.”

One of the implied intentions of the drainage pits was to achieve thixotropy, which is the phenomenon in which soil may gain strength with the passage of time mainly due to the gradual reorientation of water molecules in the adsorbed water layer and due to re-establishment of chemical equilibrium. This recovery of the strength is not 100 percent. Soil cannot achieve its original strength before remolding (i.e., the strength before failure, or remolding) because strength loss due to the destruction of the soil structure cannot be recovered with time. Only strength loss due to disturbance of water molecules can be regained. The time element required to regain the lost strength was a matter of speculation, but it generally was agreed at that time that as much as 6 months would be required. For this project, they also assumed consolidation of the soil also would contribute to increased strength. Additional borings were drilled in 1959 to determine by laboratory testing if there had been any physical change of the soil due to drainage and consolidation as would be indicated by a comparison of moisture contents, unit weights, and shear strength. The test results reportedly showed the trend of a general improvement in the strength characteristics via thixotropy, although it was noted that the individual tests when viewed objectively were inconclusive.

In addition to the source of the artesian conditions destabilizing the slope, the 1959 Remedial Report also provided insight into another contributing source to the significant wall movement observed: “the wall monoliths that moved were founded directly over a steep natural drop-off”.

The subsurface conditions encountered in the 1959 borings indicated the wall was “perched on top of a steep natural drop-off which had been modified by the development of the area long before the adoption of the flood control alignment.”

It was noted that “the holes drilled for the alignment foundation studies in 1950 did not locate the trouble area.” It was concluded that (1) the wall was founded in part on an unnatural shelf created by 10 to 12 feet of random fill below the wall surcharged on the old, steep natural drop-off; (2) the position of the failed walls indicated the center of gravity of the failed mass was immediately riverward of the wall; and (3) the mass will not completely stabilize until the downhill or riverbank slope is satisfactorily redeveloped to positively restrain any movement of the upper slope. The uncontrolled fill provided poor support conditions for the wall, likely resulting in higher-than-normal wall settlement. In addition, the steep, natural drop-off created a preferential sliding plane.

The following relevant conclusions were provided in the 1959 Remedial Report:

− The drainage pits were successful, and the flow had become constant at a constant head in the piezometer.

− The left bank of the Licking River was in a state of “uneasy equilibrium” with active movement opposite Drainage Pits 1 and 2 (downslope of the drainage pits between the drainage pits and the Licking River). A tension crack on the lower slope reportedly grew from about 1-inch to 8 inches in width over 41 days, with the crack underwater for long periods of time during the winter, spring, and early summer months.

− The horizontal movement of the ruptured floodwall is slight since the drainage pits were installed.

− The vertical settlement of the floodwall is negligible since the drainage pits became operational.

In addition to these conclusions, the two important facts included in the 1959 Remedial Report were the identification of the source of the artesian conditions and the foundation conditions that contributed to significant damage to the wall. In the absence of the slope movement from the 1956 event, the foundation conditions likely would not have resulted in the magnitude of movement observed in the wall.

The repairs outlined by the district office for approval in the 1959 Remedial Report included the following five (5) main features:

− Construction of a rock toe approximately 60 feet in width (measured perpendicular to the Licking River) during the low water period.

− Removal of the abandoned 33-inch sewer and backfill with selected borrow “to eliminate any possible water pockets that might develop”.

− Construction of riprap protection over the exposed slope downstream from Drainage Pit No.

5.

− Further observation of the failed section with replacement of eight (8) monoliths if the bank became definitively stable. It is noted that only six (6) monoliths ultimately were replaced in 1964/1965 instead of the eight (8) recommended in this report. Pressure grouting, instead of removal and replacement, was conducted below the other two monoliths.

− Fill drainage pits with pervious material and cap with 2 feet of filter material (to prevent migration of impervious material from above into the pervious material below) and 3 feet of impervious material.

After discussion between the district office and LRD in October and November 1959, LRD endorsed the recommendations included in the September 8, 1959, Remedial Report in an endorsement letter dated December 1, 1959, including construction of the rock toe. However, in a letter dated December 18, 1959, HQ questioned whether or not the proposed rock toe would in fact stabilize the lower slopes due to the low strength soils under the rock toe and since the bottom of the rock toe was at or near the riverbed (i.e., near the assumed failure surface versus well below it). HQ stated that the need for stability was important and that the funds already spent on this project were large. Furthermore, HQ stated remedial measures would be very expensive to construct if the rock toe was in-place but was ineffective. HQ recommended full consideration be given to positive stabilizing measures consisting of steel pile cells partly backfilled with sand and gravel. The top of the cells were to be above the future pool once Markland Dam was constructed, and the sheet piles possibly were to be driven to EL 390 feet where the subsurface materials showed definite improvement in strength (estimated top of rock). The measure also included a generous sand and gravel drain landward of the cells. As a result, LRD issued a letter dated December 23, 1959, to the district office requesting preliminary plans, design analyses, revised cost estimates, and a time estimate for construction of the revised remedial measure consisting of a cellular sheet pile wall.

Stabilizing cellular sheet pile wall drawings were prepared in 1960. The cellular sheet pile wall design extended from Station 169+50 to 180+00, which (not coincidentally) is the same area that has exhibited problems throughout the subsequent years.

In December 1959, the LDO, LRD, and HQ approved the recommended changes to the design to better accommodate the newly identified subsurface conditions (e.g., changing from flat-bottomed T-walls to sloped T-walls bearing at greater elevations to address the steep, natural drop-off), and the recommended addition of a stabilizing feature at the riverbank toe. Furthermore, and most importantly, it was clear that all parties agreed that additional stabilizing measures above and beyond the drainage pits were warranted and intended.

The cellular sheet pile wall planned at the toe the slope was never constructed, nor was the alternative rock berm. No further correspondence in the records was located to indicate why the cellular sheet pile solution (or the toe berm) was not constructed.

Structural modifications to the walls were implemented in 1960 until a permanent solution to the slope stability issues was implemented. In an endorsement letter dated April 23, 1964, from LRD to the district office, it was concluded that the wall movement readings confirmed that “the bank has stabilized and that the floodwall can be reconstructed safely”. Six (6) of the eight (8) wall monoliths were replaced in 1964 and 1965. According to “Tilting Floodwalls in Covington, Kentucky”, by Terry Sullivan, 2021, the six (6) flat-bottomed T-Wall monoliths (Monoliths 41-46) were replaced with new sloping base T-Wall monoliths. The new monoliths had keys that extended to a depth of over 16 feet below finished grade, compared to the original 12-foot-deep shear keys.

Additionally, the bases of the new walls were 25’-8” wide, as compared to only 21’-8” for the original walls. Over-excavation of the foundation to EL 480 for each new monolith was required as well. Pressure grouting below the two adjacent flat- bottomed T-Wall Monoliths 40 and 47 was included in the scope to ensure all voids had been filled.

Other Slope Movements (1983, 2012, 2019, and 2022) A slope failure involving the riverside levee embankment occurred in June 1983 from approximately Station 179+00 to 180+00. This failure followed an extended period of elevated river levels that generally were between EL 475 to 485 feet in April and May versus normal pool around EL 456 feet.

The embankment failure reportedly extended 80 feet down the face of the levee slope and included a visible toe bulge. This failure reportedly appeared to be confined to the levee embankment (i.e., a relatively shallow failure limited to the levee slope instead of a deep-seated failure that extended to the riverbank). A letter dated November 8, 1983, prepared by the district office recommended the failed material be excavated from the levee, a granular drainage blanket installed, and the excavated material replaced. The embankment was repaired using Federal funds in 1984. During construction, finger drains were added to the remediation plans.

A deep-seated slope failure involving the riverside levee embankment and riverbank from Station 169+50 to 172+50 was reported by the Sponsor in January 2012. It is noted that significant flooding occurred in the Ohio River Valley in 2011 between March and May (water elevation reached EL 485 feet and included a rapid drawdown event between the two events). This major flood event was followed by sustained elevated water levels again starting in December 2011 with a brief rapid drawdown event in mid to late January. The levee was repaired by the City of Covington using deep soil mixing between late 2015 and mid-2016. The repair of the embankment slope was performed in conjunction with the removal and replacement of a 24-inch pressurized steel gas main.

A deep-seated slope failure involving the riverside levee embankment and riverbank was reported by the Sponsor between approximate Stations 174+00 to 177+00 in August 2019. However, the exact date of failure is unknown. Numerous highwater events between EL 475 to 485 feet occurred between mid-December 2018 and March 2019 followed by an elevated event in June 2019 around EL 473 feet. Inclinometers installed in the area of this failure indicate a deep-seated failure that likely involves most of the slope. The failure surface generally is approximately 20 to 26 feet below existing grades at both the toe and near mid-levee slope according to the most recent inclinometer readings provided by ECS Midwest, LLC. This failure has continued to expand to the north to Station 171+00 and to the south to Station 180+50.

In addition, in 2022, from approximately Station 164+00 to 169+00, Duke Energy installed a sheet pile tieback wall to protect their existing gas pipeline, which is located at the crest of the riverbank, just past the toe of the levee system. This area is beyond the known historical problem area, but the area has shown evidence of minor creep beginning to develop.

3. RECENT SLOPE MOVEMENT AT GATEWELL 16

There is currently on-going instability in the area of Gatewell 16 based on tension cracks, scarps, and slides that were first noted in 2019 and that have continued to progress. Additional tension cracks and scarps were noted in 2024 site visits, indicating that the slopes are continuing to move. On April 5, 2024, Bill Matteoli, Covington Public Work Assistant Director, informed USACE that a maintenance crew member discovered a “sinkhole” earlier that week between Gatewell 16 and the Licking River (Figure 3-1:

"Sinkhole” Near Gatewell 16 (April 5, 2024)).

Figure 3-1: "Sinkhole” Near Gatewell 16 (April 5, 2024)

At the time of the USACE site visit on April, 9 2024, there was approximately two feet of freeboard between the lowest elevation of the rim of the stoping subsidence (sinkhole) and the current river level at that time, with an approximate two-foot rise in the Licking River anticipated in the following days.

Subsequent review of USGS stream gage data indicated a peak gage reading of 51.69 feet (elevation

479.74 feet NAVD88) on April 7, 2024. The Sponsor later confirmed that river levels approached the rim of the hole elevation but did not reach the projected elevation. The precise stage/elevation of the subsidence rim was not known; however, it is anticipated that river levels will reach the rim at approximately Stage 53 (elevation 481 feet NAVD88). Review of the provided notes and photographs from the levee Sponsor and observations from the USACE site visit indicate that the subsidence was retaining water a few feet below the existing ground surface elevation at the apparent elevation of the Licking River.

The rim of the hole measured approximately five (5) feet in the east-west direction and seven (7) feet in the north-south direction on April 9th, 2024. The hole appeared to be directly in-line with the pipe crossing the levee centerline at Station 172+87 (gatewell station). The landside rim of the hole was measured to be approximately 23 feet from the waterside face of Gatewell 16. Based on the as-built plans for the sewer crossings at Station 172+87, the horizontal distance of the subsidence from the gatewell correlates to a depth of approximately 13 feet from the surface elevation to the top of the 30-inch cast iron pipe. Basic depth measurements in the field with a survey rod indicated that the cave-in depth of the hole ranged from approximately seven (7) to ten (10) feet below the current ground surface elevation.

While on-site, USACE personnel identified scarps and tension cracks, visually and with the assistance of a probe rod, along the embankment in the area of the subsidence and gatewell structure (Figure 3-2).

The scarps and tension cracks identified included both known pre-existing features and apparent new features, including tension cracks along the slope directly through the subsidence (Figure 3-3:

Stoping, Scarp, and Tension Cracks April 9, 2024). In the following months, the tension cracks that passed through the subsidence were monitored for movement. In the first few weeks, total movements of less than 1/8-inch were measured and reported by the Sponsor. After the initial movements were recorded, additional movements were not observed in the following weeks and months.

Figure 3-2: Overhead Image with Mapped Tension Cracks, Scarps, and Stoping

Note: Image prior to April 2024

Figure 3-3: Stoping, Scarp, and Tension Cracks April 9, 2024

There has been apparent past subsidence of soil surrounding the gatewell structure based on the exposed non- painted area of concrete below the mural on the south side of the gatewell, and there are scarps and tension cracks present on all sides of the gatewell structure. There were no apparent indications of visual damage to the gatewell structure at the time of the site visit, and there have been no reports that the sluice gate housed in the structure is inoperable. It appeared that the sluice gate was currently in the closed position at the time of our site visit based on the observed position of the valve stem.

Pipe video inspections, dated December 1, 2015, were reviewed by LRL Levee Safety. The video inspection included portions of the pipe on both sides of Gatewell 16. The portion of the 30-inch cast iron pipe that was on the west side of the gatewell appeared to be in acceptable condition. Unanticipated conditions were observed in the portion of pipe on the east side of the gatewell. According to the as-built drawings, a 30-inch cast iron pipe was installed from Gatewell 16 to Manhole 37A and beyond to the headwall at the Licking River. However, based on the distances indicated on the inspection video, the cast iron pipe transitions to corrugated metal pipe (CMP) of a similar diameter approximately 19 feet past the gatewell. Manhole 37A and the outlet headwall were not present in the video. Based on this information, it appears that a portion of the cast iron pipe, Manhole 37A, and the corresponding headwall were damaged and replaced with CMP. No record of this repair or alteration has been located.

At the transition between the cast iron pipe and CMP, the CMP immediately slopes downward towards at an apparent consistent slope discharging into the Licking River. In the provided video (2015), the CMP appears corroded along the length of the pipe with some distortion and dark staining of the CMP and leakage within the apparent transition zone. On April 12, 2024, Sanitation District No. 1 attempted to perform a follow-up inspection of the pipe from Gatewell 16 to the Licking River in an effort to determine the current condition of the pipe.

Review of the conditions of the pipe was difficult since a significant portion of the pipe was filled with water and debris. Based on the distance measurements provided on the video, it appears the blockage in the pipe (soil/grass/debris) started approximately 20 feet past the gatewell structure. The distance of the transition from cast iron to CMP in the December 1, 2015, video and the location of the blockage in the pipe encountered in the April 12, 2024, inspection video correlate well with the field measured location of the subsidence. For more information on the 2024 event, refer to the April 9, 2024, Trip Report in Geotechnical Appendix D.

In May/June 2024, Sanitation District No. 1 installed a temporary repair for the damaged section of the pipe (deteriorated and separated CMP). The temporary repair consisted of installing a 12-inch diameter HDPE pipe between the undamaged cast iron portion of the pipe and the bank of the Licking River. The temporary HDPE pipe was installed directly over the existing CMP. Therefore, the permanent repair will require the removal of the temporary 12-inch pipe and the underlying 30-inch pipe.

4. GROUNDWATER AND ARTESIAN CONDITIONS

Designs must properly account for elevated hydrostatic conditions. USGS groundwater data and geologic data suggests that artesian conditions are present within the bedrock along the Licking River due to confined aquifers within the hills surrounding the Licking River valley. See Figure below, which shows the project location, the Licking River valley, and the hillsides surrounding the area that contribute to the artesian conditions. These confined bedrock aquifers, when directly connected with confined granular layers associated with the river deposits, can create artesian conditions in the overburden soils along the river. At the southern portion of the portions of the levee and floodwall (approximate Stations 177+00 to 180+00) and at the plateau beyond the crest of the levee system, bedrock (generally shale interbedded with limestone) was encountered at much shallower depths than bedrock was encountered in the areas to the north (EL 447-449 ft versus EL 389 to 391 ft). In addition, the gravel layer was in direct connection with the bedrock in this area, which would allow the artesian conditions in the bedrock to transmit to the confined gravel layer. An example of the gravel layer in direct contact with the bedrock (in boring DC-636) is shown in the upper righthand corner of Figure .

Figure 4-1: Hills Surrounding Project Location

Additionally, the USGS Soil Survey for Boone, Campbell, and Kenton Counties, Kentucky indicates that seasonal groundwater as shallow as 2 to 3 feet below the ground surface is typical for the Licking-Captina soils that are present along the slopes adjacent to the Licking River. The original and repair work borings confirm shallow groundwater was encountered in some locations near and along the slopes along the Licking River.

At boring locations A-517 and D-627 (approximately midway up the upper slope), which were drilled in response to the 1956 slope failure, groundwater freely flowed from the borings during and after drilling. At boring A-517, groundwater reportedly flowed for approximately 6 months.

Approximately one year later, water was observed at the ground surface near that boring location.

The ground surface at these two borings locations was EL 478.3 feet (A-517) and 478.6 feet (D- 627).

Drainage pits were constructed in 1958 and completed in 1959 to address the elevated hydrostatic conditions that were identified in the exploratory pits excavated. During the installation, artesian conditions of 12 to 14 feet were encountered in most of the exploratory pits when the excavations encountered a sand and gravel layer at an elevation of approximately Elevation 461 to 465 feet (i.e., water rose to EL 473 to 479 feet). It is important to note that the river levels were not elevated when the drainage pits were excavated, so these water levels were not necessarily representative of the worst conditions that could be expected to occur.

Long term monitoring of piezometers in the area (generally on the plateau immediate west of the project location) showed that the water levels in the piezometers could be connected with river levels during high water events. Water levels in these piezometers tended to remain above river levels (i.e., generally between EL 476 feet to EL 487 feet) when the river was at “normal” levels (Note: These measurements were before the drainage pits were installed), but the water levels tended to rise dramatically during flood events to either match the river level or be within a few feet of river level when the river water levels were higher than the “normal” water levels encountered in the piezometers.

The granular layer encountered in the drainage pits appeared to be more or less continuous in the north-south direction (i.e., parallel with the levee crest) and the east-west direction (i.e., perpendicular to the levee crest) based on the subsurface conditions encountered in borings from the design phase borings, borings from the remediation phase, and the drainage pits. The sand and gravel layer, where encountered, generally was at least several feet thick, but generally was not more than 10 to 12 feet thick.

The design of the remediation system must account for the groundwater conditions observed.

Given the published information and the site conditions encountered, for normal conditions, the factor of safety shall be based on the worst of the following 2 conditions: (1) groundwater modeled at the original ground surface (prior to fill placement for the levee) or (2) artesian pressures in the gravel layer modeled at EL 485 feet. For flood conditions, the factor of safety shall be based on the worst of the following 2 conditions: (1) groundwater at the original ground surface (prior to fill placement for the levee) or (2) artesian pressures in the gravel layer modeled at the river level or EL 485 feet if the river level is less than EL 485 feet. Furthermore, any design must include provisions (spacing of improved soil columns, gaps in sheet piles, etc.) to prevent trapping groundwater behind the remediation system.

5. WEAKENED SOILS

The history of slope failures and ongoing slope movements along the Licking River have been well documented. In several areas, both the lower slope (the riverbank area between the gravel service road and the Licking River) and the upper slope (beyond the gravel service drive and including the levee prism, or to the floodwall foundations) have failed. Based on historical data and site observations, it appears that the lower slopes are failing more frequently. The apparent higher occurrence of slope failure is likely the result of the frequency in which the waters of the Licking River inundate the lower slopes. Numerous failures of the upper slopes have occurred as well;

however, these failures have not been observed to be occurring as frequently. Refer to Geotechnical Appendix G for inclinometer data to better understand the depth of the failure in areas adjacent to the gatewell. A summary of the apparent depths of movement at various inclinometers near the gatewell are depicted in Figure 5-1.

Figure 5-1: Inclinometer Indicated Failure Depths – Inclinometers near Station 172+87

There are likely numerous preferential slip planes within the previously failed areas of the upper and lower slopes. Therefore, a significant portion of the lower slope soils and portions of the upper slope soils are in a weakened state due to the past and on-going slope failures. Due to the extent and number of past slope failures, all soils that have not otherwise been remediated, that are above the failure surface, must be modeled using their residual strength. Assuming a fully-softened strength (typically applied to first-time failures) for those regions will not be acceptable.

Additionally, portions of the lower slopes appear to be eroding at the riverbank along the Licking River. The apparent rate of loss of riverbank in those areas has not been studied, and the rates of removal are not known. This loss could impact the overall slope stability.

6. PAST REMEDIAL WORK

Techniques such as tie-back walls and deep soil mixing have been used to improve the marginally stable slopes. All of which have demonstrated some amount of improvement to the overall slope stability along the Licking River. Recent efforts to improve the slope stability include deep soil mixing (2012) and a tied-back sheet pile wall (2022) appear to be performing adequately. However, long term performance of those systems has yet to be established. See Geotechnical Appendix F for information regarding the 2012 and 2022 measures implemented.

7. PROPOSED REPAIR STABILIZATION SYSTEM

It is anticipated that a wall system with tieback anchors, similar to the one used during the 2019 gas line stabilization between approximate Stations 164+00 to 169+00, will be utilized for the emergency stabilization repair in the area of Gatewell 16, except that the wall may be comprised of continuous sheet piles or drilled soldier beams piers with a permanent facing system. See Figure 7-1 below for general examples of these types of wall systems. Other slope stabilization measures and/or wall systems that meet the design requirements are acceptable.

Figure 7-1: Potential Wall Systems

The following minimum requirements will be incorporated into the project requirements:

1. The specialty contractor designing and installing the proposed stabilization system shall have the following minimum qualifications/provide the following information:

– Conducted at least 3 slope stabilization projects of similar size and scope in the past

5 years (provide a list with a summary of the project details – type of wall system, length stabilized, depth of failure – along with owner name and contact information).

– Conducted at least 10 permanent grouted tie-back installations over the last 5 years.

– Designer shall be a registered engineer in the Commonwealth of Kentucky with at least 10 years of continuous experience designing slope stabilization projects (at least 1 project per year – experience in other slope stabilization methodologies permissible).

– Provide a list of at least 3 owners, along with contact information, for projects similar to the proposed remedial system.

2. The proposed repair must restore stability to the gatewell and ensure it continues to function as designed into the future, allowing no future movement associated with past slope stability issues.

3. The emergency repair shall extend a minimum of 75 feet on either side of the gatewell and be located immediately downslope of the gatewell (within 10 feet).

4. The DB Contractor shall provide all calculations, assumptions, and basis for assumptions for the proposed stabilization system for review and approval by USACE.

5. Local stability in the immediate area of the gatewell, as well as global stability for the entire levee system and associated riverbank within the area of the proposed repair system, shall be evaluated for the design cases outlined in the most current edition of EM 1110-2-1913 and shall include at a minimum the following: stability analysis for the…