B08_Attachment_5_-_Geotech_Report_-_VICK_310725_324137.pdf

i

TABLE OF CONTENTS

1.0 INTRODUCTION AND PURPOSE

1.1 Project Description

1.2 Purposes

2.0 PREVIOUS FIELD INVESTIGATIONS

3.0 SUPPLEMENTAL FIELD EXPLORATION

3.1 General

3.2 Drilling and Sampling Methods

3.3 Field Classifications, Sample Preservation, and Borehole Abandonment

4.0 LABORATORY TESTING

4.1 General

4.2 Strength Tests

4.3 Classification Tests

4.4 Water Content Tests

5.0 GEOLOGY

6.0 SUBSURFACE CONDITIONS

6.1 General

6.2 General Soil Conditions at Station 1+25

6.3 Groundwater

7.0 ENGINEERING ANALYSES

7.1 General

7.2 Proposed Landslide Remediation

7.3 Calculation Package

7.4 Soil Parameters

7.5 Slope Stability Analyses – General

7.6 Slope Stability Analyses Using Drained Strengths

7.7 Undrained and Seismic Stability Analyses

7.8 Design of Soldier Piles, Lagging, and Anchors

7.9 Application of Design Cross-Sections to Develop Wall Profile

7.10 Potential for Conflict between Soldier Piles and Anchors and the

Existing Soil Nails

7.11 Pavement

8.0 CONSTRUCTION RECOMMENDATIONS

8.1 Anchored Soldier Pile and Lagging Retaining Wall Construction

8.2 Site Preparation Before Fill Placement

8.3 Earthwork Construction

8.4 Subsurface Drainage Features

8.5 Groundwater Control During Construction

8.6 Quality Control and Quality Assurance During Construction

8.7 Maintenance

9.0 REPORT LIMITATIONS

ii

FIGURES

1) Plan of Existing Conditions

2) Synopsis of USCS

3-4) Boring Logs

5a) Cross-Section: Existing Conditions at Station 1+25

5b) Cross-Section: Existing Conditions at Station 1+90

6) Plan of Proposed Conditions

7) Cross-Section: Proposed Conditions at Station 1+25

8) Cross-Section: Proposed Conditions at Station 1+90

9) Recommended Retaining Wall Profile

10) Suggested Drain Details

APPENDICES

A) Logs of Borings from Previous Investigations

B) Data Sheets for UU Compression Tests

C) Site Geology Information

D) Engineering Calculations

E) Slope Stability Analyses for Existing Conditions, Sta. 1+25 and 1+90, Drained Strengths

F) Slope Stability Analyses for Proposed Conditions, Sta. 1+25 and 1+90. Drained Strengths

G) Slope Stability Analyses for Proposed Conditions, Sta. 1+25 and 1+90, Undrained Strengths

H) Slope Stability Analyses for Proposed Conditions, Sta. 1+25 and 1+90, Seismic

I) ITR Comments and Responses

1.0 INTRODUCTION AND PURPOSE

1.1 Project Description

On February 13, 2020, a landslide occurred at the National Cemetery at the Vicksburg

National Military Park (VNMP). The approximate limits of the landslide are shown on Figure 1.

The landslide is approximately 200 ft by 200 ft in plan dimensions and resulted in the formation of an approximately 15-ft high head scarp and migration of slide debris to the bottom of the hillside. The landslide displaced graves that were located at the top of the slope and on a terrace located in the lower portion of the slope and a portion of a cemetery roadway referred to as Given

Hill Road. Slide debris flowed across and covered a number of graves located on a terrace near the bottom of the slope. A steepened approximately 30-ft high mound, referred to as the Indian Mound, is located upslope of the scarp. The Indian Mound would be impacted if the head scarp migrates upslope more than about 10 ft towards the east.

In 2021, Burns Cooley Dennis, Inc. (BCD) performed a geotechnical exploration at the site of the landslide, installed field monitoring instrumentation, performed laboratory testing, and performed calculations for the design of a slope stabilization system. The proposed stabilization system consists of an anchored soldier pile and lagging wall, subsurface drainage improvements, and a buttress at the toe of the slide. Our final geotechnical report was submitted on December 14, 2022.1

This current study supplements the information presented in the 2022 report (Ref 1), and modifies the remedial design to include the following:

a) reconstruction of the terrace located on the lower portion of the slope so that graves can be placed back in their approximate original locations;

b) reconstruction of the failed roadway (Given Hill Road); and

c) addition of a surface drainage system to collect and direct stormwater runoff/drainage from the existing soil nail wall.

BCD is tasked with developing a revised landslide stabilization system that accounts for Items a) and b). Item c) is being addressed by Stantec.

1 BCD “Final Geotechnical Investigation Report and Slope Stabilization Recommendations, National Cemetery, Vicksburg National Military Park, Vicksburg, Mississippi” prepared for Stantec, dated December 14, 2022.

1.2 Purposes

The specific purposes of this exploration were:

1) to further explore subsurface conditions in the landslide by advancing two additional borings (B-5 and B-6) at the locations shown on Figure 1.

2) to verify field classifications and evaluate physical properties of soils encountered by means of visual examination of soil samples and testing in the laboratory;

3) to prepare design calculations for a stabilization wall that account for the reconstructed terrace and roadway;

4) to develop pavement recommendations for the reconstructed roadway; and

5) to prepare a report documenting the results of the supplemental geotechnical exploration and laboratory testing programs, and presenting our recommendations and the design calculations for the revised landslide remediation system.

2.0 PREVIOUS FIELD INVESTIGATIONS

The USACE advanced Boring MS-4-97U at the southern end of the landslide in 1997 as part of an investigation of slope instability along Mint Springs Bayou. The approximate location of the boring is indicated on Figure 1.

BCD advanced two borings, Boring 4 (2002) and Boring 8 (2002), within the landslide area in September and October of 2002. The approximate locations of these borings are indicated on Figure 1. These borings were made as part of a study of the brick wall/fence around the perimeter of the cemetery. The wall originally followed an alignment east from the entrance archway to the top of the slope, but the portion along the slope had been destroyed by previous landslide activity.

BCD advanced four borings, Boring 1 through Boring 4, in the landslide area in March of

2021. The locations of these borings are indicated on Figure 1. These borings were made as part of BCD’s geotechnical exploration of the landslide as documented in the 2022 report (Ref 1).

3.0 SUPPLEMENTAL FIELD EXPLORATION

3.1 General

Subsurface data from previous explorations were supplemented by means of two additional borings advanced at the site on November 12, 2024. The approximate locations of Borings 5 and

6 are presented in Figure 1.

All soils were classified in general accordance with the Unified Soil Classification System

(USCS). A synopsis of the USCS is presented on Figure 2 along with symbols and terminology typically used on graphical soil boring logs. Graphic logs for Boring 5 and Boring 6 are presented on Figures 3 and 4. The graphical logs illustrate the types of soil encountered with the depth below ground surface at the individual boring locations.

3.2 Drilling and Sampling Methods

Boring 5 was advanced to an exploration depth of 25 ft using a 6-inch diameter short-flight auger. Boring 6 was initially advanced to a depth of 7 ft using a 6-inch short-flight auger and then to a completion depth of 35 ft using the rotary wash method. Relatively undisturbed Shelby tube samples were obtained by pushing a 3-inch outside diameter (O.D.) thin wall tube sampler approximately 2 ft into the soils. Disturbed samples were obtained from the borings by driving a

2-inch O.D. split-spoon sampler 18 inches into the soil with a 140-lb hammer falling freely a distance of 30 inches. The standard penetration test (SPT) blow count resulting from the split-spoon sampling is recorded under the “Blows Per Ft” column of the graphic log. Shelby or split-spoon samples were generally obtained at 3-ft to 5-ft intervals in the borings. Observations were made continuously during auger drilling to detect free water entering the open boreholes.

3.3 Field Classifications, Sample Preservation, and Borehole Abandonment

All soils encountered during drilling were examined and classified in the field by a geotechnical engineering technician. Each undisturbed Shelby tube sample was extruded from the sampling tube in the field. An approximate 6-inch-long portion of each Shelby tube sample was sealed with paraffin in a cylindrical cardboard container to prevent moisture loss and structural disturbance. An additional portion of each Shelby tube sample and representative portions of the split-spoon samples were sealed in jars to provide material for visual examination and classification testing in the laboratory.

At completion, the Boring 6 was backfilled with cement-bentonite grout. Boring 5 was backfilled with auger cuttings.

4.0 LABORATORY TESTING

4.1 General

An evaluation of the strengths and classifications of the subsurface soils encountered in the borings was considered to be of primary importance for this investigation. These properties were evaluated by visual examination and from the results of the laboratory tests described in the following paragraphs.

4.2 Strength Tests

The shear strengths of the soils encountered in the borings were estimated by means of correlations with index properties, from the results of standard penetration tests and from the results of laboratory strength tests. The undrained shear strengths of the fine-grained soils encountered in the borings were investigated by means of unconsolidated-undrained triaxial compression tests (UU). The tests were performed on selected undisturbed Shelby tube samples.

The results of the UU compression tests in terms of cohesion are plotted as small open triangles in the data section of the graphic boring logs. The data sheets for the UU compression tests are presented in Appendix B. The water contents and dry densities were also determined for each compression test specimen. The water contents are plotted as small shaded circles in the data section of the logs. The dry densities are tabulated to the nearest lb per cu ft under the “Dry

Density” column of the graphic boring logs.

4.3 Classification Tests

The classifications of the soils encountered in the borings were investigated by means of visual examination and by Atterberg liquid and plastic limit tests. The numerical difference between the liquid limit and plastic limit is defined as the plasticity index (PI). The liquid limit and the plasticity index are indicators of the effective (drained) shear strength of clayey soils. The proximity of the natural water content to the plastic limit is an indicator of the undrained strength.

The results of the liquid and plastic limit tests are plotted as small crosses interconnected by dashed lines in the data section of the logs.

To assist in classifying sandy soils, tests were performed to determine the percent fines passing the No. 200 sieve. The percentage of fines resulting from each test is tabulated at the appropriate depth under the far-right column of the graphic log.

4.4 Water Content Tests

Additional water content tests were performed to corroborate field classifications and to extend the usefulness of the strength and plasticity data. The results of the water content tests are plotted as small shaded circles in the data section of the graphic boring logs.

5.0 GEOLOGY

Geology of the project area is presented on the Vicksburg, MS-LA quadrangle of the

Waterways Experiment Station Technical Report No. 3-480, Geological Investigation of the

Yazoo Basin, Lower Mississippi River Valley. A portion of the quadrangle and a geologic section illustrating the subsurface stratigraphy are presented in Appendix C.

The uppermost geologic unit in the park is the loess. Although several hypotheses have been presented for the formation of loess, it is generally recognized that loess forms from the deposition of silt sized sediment by wind. At the time of loess deposition in the Vicksburg area, the Mississippi River was chocked with glacial sediment and was a wide and shallow braided stream of many small channels, not the meandering single channel as it is today. Loess deposits in

Vicksburg actually consist of at least three separate geological formations ranging in age from approximately 120,000 to 11,000 years.

The loess is underlain by Pleistocene age pre-loess Terrace deposits of sands and gravels with occasional strata of clays. The pre-loess Terrace deposits are underlain in descending order by clays and sands of the Catahoula formation, silts and clays of the Bucatunna formation, fossiliferous marls of the Byram formation, limestones and marls of the Glendon formation, fossiliferous marls of the Mint Spring formation, and carbonaceous clays and silts of the Forest

Hill formation. The Bucatunna, Byram, Glendon, and Mint Spring formations are collectively referred to as the Vicksburg Group. Reported ranges in thickness of the formations are:

Byram/Bucatunna 90 ft to 100 ft; Glendon 25 ft to 65 ft; and Mint Spring 5 ft to 25 ft2.

The sands and clays of the Catahoula formation were deposited in coastal and delta environments along Mississippi’s southern coast as the coastline moved south out of the

Mississippi Embayment during the Miocene Epoch. The Catahoula formation is absent within portions of the park and, when present, exists as a remnant within the higher elevation portions of the park. Based on information presented in the Mississippi State Geological Survey Bulletin for

Warren County1, the total thickness of the Catahoula formation exceeds 100 ft in Warren County.

The Vicksburg Group consists of sedimentary materials deposited in a shallow Oligocene

Sea. The uppermost indurated ledge in the Glendon is an ideal stratigraphic marker and it has been a common practice by oil companies to construct structural contour maps showing the top of the

Glendon. These structural contour maps show anomalies that are caused by deep seated movements in the subsurface that in some cases can trap oil or gas. The top of the Glendon is usually relatively flat except in areas where it has been disturbed by faulting or pushed upward by salt domes. Within the limits of the park, the top of Glendon generally slopes down towards the east at about 1 percent (1 ft over 100 ft).

The contact between the loess deposits and pre-loess Terrace deposits and the contact between the pre-loess Terrace deposits and the Catahoula formation (or Vicksburg Group where the Catahoula is absent) are considered disconformities in the sense that the surface of the older deposits was sculpted by erosion prior to deposition of the more recent deposits.

6.0 SUBSURFACE CONDITIONS

6.1 General

Two additional borings (Boring 5 and Boring 6) were made for this exploration to better define the subsurface conditions near the north end of the landslide area and to update the cross-section at Station 1+25. General descriptions of the soil and groundwater conditions encountered in the two new borings are provided in the following subsections. Graphical logs for Borings 5 and 6 are presented in Figures 3 and 4. Graphical logs from previous subsurface investigations are presented in Appendix A. Stick logs of Borings 5 and 6 and previously drilled Boring 4 are

2 Mellen, F. F., (1941), Bulletin 43, Warren County Mineral Resources, Mississippi State Geological Survey.

presented on the cross-section of existing conditions at Station 1+25 on Figure 5a. Stick logs for previously drilled Borings 1 and 3 are presented on the cross-section of existing conditions at

Station 1+90 on Figure 5b. Tabulated adjacent to the stick logs are Atterberg liquid and plastic limits, water contents, dry densities, undrained cohesions, percentages of fines, and SPT blow counts.

6.2 General Soil Conditions at Station 1+25

Medium dense to dense silty sand (SM) fill materials were encountered from the surface to a depth of 4 ft at Boring 5. The fill materials are underlain by stiff silty clays (CL) to a depth of

8 ft. The silty clays (CL) appear to be either fill materials or talus/slide debris. The silty clays (CL) are underlain by medium stiff clay (CH) marl that extends to the 25-ft terminal depth of Boring 5.

Talus/slide debris was encountered from the existing ground surface to a depth of about 4 ft at Boring 6. The talus/slide debris includes medium dense silts (ML) and stiff silty clays (CL).

Stiff to hard clay marls (CL, CH) were encountered in Boring 6 from a depth of 4 ft to 16.5 ft.

These clay marls are interpreted to be part of the Byram formation. Medium dense to very dense clayey fine sands (SC) with shell fragments were encountered from a depth of 16.5 ft to 23 ft, and limestone was encountered from a depth of 34 ft to the 35-ft terminal depth of Boring 6. The clayey sand marl and limestone are interpreted to be part of the Glendon formation.

6.3 Groundwater

Free water was encountered at an approximate depth of 18.5 ft during auger drilling for

Boring 5. Drilling was suspended for about 15 minutes, and the water level was measured at a depth of 17.7 ft. Free water was not observed during auger drilling for Boring 6, or during the auger drilling of Boring 4 in 2021.

Vibrating Wire Piezometers (VWPs) were installed in Boring 2 during the 2021 exploration. Data collected from the VWPs is presented and discussed in the 2022 report (Ref 1).

7.0 ENGINEERING ANALYSES

7.1 General

The National Cemetery at the VNMP consists of a series of manmade terraces that step up the western slope of the cemetery. Each terrace contains a grid of burials. The 2020 landslide impacted the south end of the southernmost lower terrace. Historical surveys and aerial imagery indicate that landslides before 2020 had likely covered burials near the south end of the terrace with landslide debris. After the design of the landside mitigation system described in Ref 1 was completed, the National Park Service determined that the terrace should be restored to its original configuration so that burials in that area could be replaced. The location of the terrace is shown in

Figure 6. The reconstructed terrace will be 45 ft wide and will extend south to about Baseline

Station 1+90.

To restore the terrace, the slope between the east side of the terrace and the stabilization wall will also need to be regraded to provide adequate factors of safety for slope stability. This required regrading between the restored terrace and the stabilization wall increases the height of the stabilization wall significantly, which in turn reduces the global stability factor of safety of the stabilization system and requires the stabilization wall to provide higher resisting forces. Cross-sections that illustrate the revised grading are presented in Figures 7 and 8.

The purpose of the engineering analyses presented in this supplemental report is to update the stabilization system design to account for the revised grading plan west of the proposed anchored soldier pile and lagging wall. The updated design is based on slope stability analyses performed for cross-sections at Station 1+25 and Station 1+90.

7.2 Proposed Landslide Remediation

Several slope stabilization methods were initially presented to the NPS in 2021. After considering the advantages/disadvantages of the remedial options and potential impact to cultural resources, the NPS chose the remedial option that consists of construction of an anchored soldier pile and lagging retaining wall located parallel to and offset about 40 ft to 50 ft west of the landslide scarp and construction of a stability berm along the lower portion of the slope. The area between the retaining wall and head scarp will be backfilled with compacted fill. Internal drainage features will be included within the backfill to reduce the potential for saturation of the backfill. Two subsurface drainage features are planned. A combination blanket drain/underdrain in the wall backfill begins at about Sta 0+70 and extends south to about Sta 2+30. A French drain will be located below the wall from about Sta 0+00 to about Sta 2+30. The pre-stressed anchors for the retaining wall will be bonded in the Glendon formation limestone and marls. Large 12-strand anchors with maximum test loads up to 281 tons have been successfully bonded in Glendon formation limestones and marls for anchored walls in the Vicksburg area. The soldier piles for the retaining wall will also be socketed into the Glendon formation limestones and marls below the slide-prone Byram clays. A stability berm will be constructed at the toe of the slope to raise the factor of safety of potential failure surfaces in the talus downslope of the retaining wall. The original design of the stabilization system is presented in the 2022 report (Ref 1).

The proposed stabilization system has been revised to account for creating a 45-ft wide terrace about midway down the slope below the retaining wall as shown in Figure 6. The grading in the area east of the proposed retaining wall will be similar to the 2022 remediation plan, except that the roadway will be reconstructed. The revised remediation plan still incorporates a stability berm along the lower portion of the slope. Cross-sections depicting the currently proposed remediation at Station 1+25 and 1+90 are presented in Figures 7 and 8.

7.3 Calculation Package

Calculations performed to develop the revised remedial design are presented in Appendix

D. The calculations supplement the calculations presented in Ref 1. The calculations include the following:

Table 1 - Organization of Appendix D - Calculation Package

Section Description Page

A Description of Additional Subsurface Exploration and Survey 1

B Back-Analyses of Existing Conditions for Cross-sections at Station 1+25 and 1+90 2

C Slope Stability for Proposed Conditions 3

D Design of Soldier Piles and Anchors 4-13

E Design of Lagging 14-15

F Design of Anchor Seat 16-24

G Design Anchor Bond Length 25

H Axial Capacity of Soldier Pile 26

7.4 Soil Parameters

The shear strength parameters used in the slope stability analyses are summarized in Table

2. It is our opinion that the landslide occurred under drained conditions. Back-analysis of the existing conditions was performed using drained strengths, and the stabilization system design was performed using drained strengths from the back-analysis. At the request of USACE, analyses of the proposed remediation have also been performed using undrained strengths. In addition, pseudo-static slope stability analyses were performed to check seismic loading of the proposed remediation. The seismic analyses were performed using undrained strengths. Undrained shear strength parameters for the soils comprising the slope were developed considering the results of laboratory triaxial compression tests, standard penetration tests and our experience with these soil types.

The drained strengths of the talus and Byram formation clays (CH) were back-calculated assuming a slope stability factor of safety equal to or slightly above 1.0 for the existing conditions.

Drained shear strength parameters for design analyses at Station 1+25 were based on the back-analysis of the existing conditions at Station 1+25 considering the additional subsurface data from

Borings 5 and 6. Drained shear strength parameters for the design analyses at Station 1+90 were based on back-analysis of the existing conditions for Station 1+90. The slope stability analyses for back-calculation of the drained strengths are presented in Appendix E.

Table 2 - Material Properties used in Stability Analyses

Material Description γ (pcf)

Drained

Parameters

Undrained

Parameters φ' (deg) c' (psf) φ (deg) c (psf)

Fill 125 30 0 12 500

Talus 120 25 30 16 250

Loess 115 32 460 21 700

Upper Byram - Station 1+90 118 16 0 0 920

Lower Byram - Station 1+90 118 23 0 0 1500

Byram - Station 1+25 118 18 50 0 920

Glendon/Mint Springs 130 20 1,000 0 3000

Forest Hill 118 20 200 0 3000

Weathered Marl Station 1+25 116 20 30 0 900

7.5 Slope Stability Analyses - General

Slope stability analyses were performed for cross-sections at Stations 1+25 and 1+90.

Station 1+25 represents the northern portion of the landslide where less displacement occurred in

2020 and where no slide debris was displaced beyond the abandoned road as shown in Figure 1.

Station 1+90 is located within the central portion of the landslide where the scarp is the tallest and where the slide debris extends beyond the toe of the slope. The stability of the slope at Stations

1+25 and 1+90 will be reduced as a result of the excavation/degrading required to restore the terrace, so the stabilization wall design has been updated based on analyses at these locations.

Global stability analyses were performed using the computer program SLOPE/W by Geo-

Slope International. Factors of safety were computed for circular and noncircular failure surfaces using Spencer’s procedure. Optimization schemes in the program were utilized to identify potential failure surfaces producing the minimum calculated factors of safety. Global stability calculations were conducted for the existing conditions and for the proposed conditions (subsequent to construction of the landslide stabilization system).

7.6 Slope Stability Analyses Using Drained Strengths

The results of the analyses of the existing conditions at Stations 1+25 and 1+90 are presented in Appendix E. Back-calculated drained strengths for the talus/slide debris and Byram were used for the analysis of the proposed conditions.

The results of the analyses of the proposed conditions at Stations 1+25 and 1+90 are presented in Appendix F. The resistance provided by the anchored soldier pile and lagging wall was varied until a minimum factor of safety of 1.4 was achieved. Based on these calculations, the retaining wall at Station 1+25 has a design load of 29.2 kips/ft, and the retaining wall at Station

1+90 has a design load of 39.5 kips/ft. The cross-section at Station 1+90 is located at the south end of the proposed terrace where the free height of the retaining wall will be the greatest (≈ 19 ft).

The cut slope between the retaining wall and the restored terrace should not be steeper than 4H:1V.

The location of the toe of the stability berm was shifted east along the approximate northern two-thirds of the berm at the request of the NPS. In order to maintain the 45-ft width of the restored terrace, the location of the berm crest was not shifted, so the slope between the crest and toe of the berm was steepened. In areas where the berm slope is 2.5H:1V or flatter, the minimum calculated factor of safety of the berm is 1.4 or greater. Where the berm slope is 2H:1V, the calculated factor of safety of the berm is less than 1.3. We requested guidance from the NPS on the minimum acceptable factor of safety for the berm, and we were directed to design for a factor of safety of

1.3. In areas where the berm slope is steeper than 2.5H:1V, geogrid reinforcement will be required in the lower berm slope in order to provide a factor of safety of 1.3 (see Figure 8).

7.7 Undrained and Seismic Stability Analyses

Additional analyses were performed to evaluate the stability of the proposed conditions at

Station 1+25 and 1+90 using the undrained soil strengths presented in Table 2. The results of these analyses are presented in Appendix G. The calculated factor of safety are greater than 1.3 for all potential slide surfaces.

As an additional check, the proposed conditions were analyzed for seismic loading using the pseudo-static slope stability method. The VNMP lies within a relatively low seismic activity region according to the seismic zone mapping referenced in the International Building Code. The site adjusted Peak Ground Acceleration (PGA) was estimated using the computer program

“AASHTO Seismic Design Parameters” version 2.10 developed by the U.S. Geological Survey.

A site classification D was conservatively assumed based on the soil profile encountered in the borings and anticipated for the area based on our experience. A seismic coefficient of 0.0395, which is equal to half of the site adjusted PGA, was used in the analyses. Based on guidance provided in Ref (3), a pseudo-static factor of safety greater than 1.0 using a seismic coefficient of half of the site adjusted PGA corresponds to ground displacements of 2 inches or less. The results of the pseudo-static slope stability analyses are presented in Appendix H. All calculated factors of safety are greater than 1.0.

7.8 Design of Soldier Piles, Lagging, and Anchors

The required wall resistance to provide a factor of safety of 1.4 was determined by slope stability analysis. This load was then applied as a uniform distributed load from the top of wall to the elevation of the slide plane. The factored active pressure and the unfactored passive pressure were conservatively assumed to be equal and opposite from the slide plane to the top of the

Glendon formation. A rock mass strength, Sm, of 900 psf was conservatively assigned to the

3 NHI Course No. 130093 and 139903A, “LRFD Seismic Analysis and Design of Bridges”, Publication No. FHWA-

NHI-15-004.

Glendon formation, and the passive resistance was estimated using the equation presented in

Section 3.11.5.6 of the AASHTO LRFD Bridge Design Specifications (9th Edition) for embedded discrete vertical wall elements in rock.

Calculations were then performed to determine the required beam size, spacing and penetration. The results of the calculations are summarized in Table 3 and illustrated in Figures 7 and 8.

TABLE 3 – DESIGN SUMMARY

Station

Wall

Load

(Kips/Ft)

Min. Beam

Size

Beam

Spacing

(ft)

Top of

Soldier

Pile

(El.)

Free

Height of Wall

(ft)

Soldier Pile

Tip

Elevation

(ft)

Soldier

Pile

Encasement

Diameter

(in.)

Anchor

Test

Load

(kips)

Anchor

Design

Load

(kips)

1+25 29.2 HP14x102 8.5 163 14.5 124 30 241.5 181.1

1+90 39.5 HP18x135 8.5 167.6 19.5 124 30 394.1 295.6

The anchors will be bonded in the Glendon formation limestones and marls. Six-strand anchors are required at Station 1+25, and nine-strand anchors are required at Station 1+90. The anchors will be inclined at 25 degrees from horizontal. The six-strand anchors will be installed in

6-inch diameter boreholes, and the nine-strand anchors will be installed in 8-inch diameter boreholes. The anchor grout should have a minimum 28-day compressive strength of 4,000 psi.

An allowable bond stress of 3,000 lbs per sq ft was used to determine the required anchor bond length. The allowable bond stress was based on the strength of the marls in the Glendon formation and is not dependent on the presence of limestone layers, which obviously do occur within the bond zone. We have load tested anchors in Vicksburg that are bonded in the Glendon formation and demonstrated allowable bond stress in excess of the value used in the design. Based on the stratification at the site and considering an allowable bond stress of 3,000 lbs per sq ft, the required minimum bond lengths for six-strand and nine-strand anchors were calculated to be 38.4 ft and 47 ft, respectively. On that basis, we recommend bond lengths of 45 ft and 50 ft be specified for the six-strand and nine-strand anchors, respectively.

Lagging could consist of either precast concrete, cast-in-place concrete or timber. We understand that the NPS prefers to use pressure-treated lumber. Lagging should consist of 6-inch nominal pressure-treated southern pine, commercial grade 1 or better. Timber lagging should be pressure-treated in accordance with the requirements of AASHTO M133, and AWPA Type UC4C heavy-duty ground contact requirements. Lagging should extend a minimum of 1.5 ft below the ground surface in front of the retaining wall.

7.9 Application of Design Cross-Sections to Develop Wall Profile

We recommend that a pile tip elevation of 124 ft be used from Station 1+90 to the north end of the wall. South of Station 1+90, the pile tip elevation can be stepped up while maintaining a minimum socket length of 22 ft. Figure 9 presents our recommended wall profile.

7.10 Potential for Conflict between Soldier Piles and Anchors and the Existing Soil Nails

A soil nail wall is currently located south and east of the slope above Mint Spring Creek.

This wall was apparently constructed around 2011. As discussed in Section 7.9 of Ref (1), there is a potential for conflict between existing soil nails and the four southernmost ground anchors associated with the proposed soldier pile and lagging wall. Because the current design for the southern end of the anchored soldier pile and lagging wall is similar to the southern end of the wall in the 2022 design, there remains a potential for conflict.

Please see section 7.9 and Figure 19 from Ref (1) for further information.

7.11 Pavement

For pavements subjected to a low volume of light passenger and pickup truck traffic, it is our opinion that the proposed new pavement section can be constructed as traditional asphalt pavement supported by low volume change (LVC) natural and/or fill subgrade soils and 6 inches of crushed limestone (No. 610) granular base material. We recommend that a Type V woven geotextile fabric be placed on top of the prepared subgrade to serve as a separator between the subgrade soils and the granular base material. The fabric should comply with MDOT

Specifications for Road and Bridge Construction Section 714.13. The asphalt pavement section should consist of 3 inches of asphalt base course (ST 19mm) and 2 inches of asphalt surface course

(ST 9.5mm). The granular base material should meet the applicable requirement of Section 703 in the 2017 Edition of the MDOT specifications. The asphalt base and surface materials should conform to all applicable specifications for ST 19mm and ST 9.5mm, respectively, presented in the 2017 Edition of the MDOT specifications.

8.0 CONSTRUCTION RECOMMENDATIONS

8.1 Anchored Soldier Pile and Lagging Retaining Wall Construction

Retaining wall construction can proceed subsequent to or concurrently with the construction of the stability berm. Significant loading of the slope should be avoided prior to installation of the soldier piles. Installation of the soldier piles should begin at the north end and progress towards the south. The placement of fill and the operation of heavy equipment along and east of the wall alignment should not be allowed south of the point where the soldier piles have been installed and grouted in place. Excavation equipment, haul trucks and drilling rigs for pre-boring and anchor construction should at all times be positioned east of the installed soldier piles.

The soldier piles should be installed by drilling minimum 30-inch diameter holes to the required tip elevation, installing the soldier piles in the holes and then backfilling the annular space between the soil and soldier piles with tremie placed concrete from the bottom of the pile to the bottom of the lagging. The concrete should have a minimum 28-day compressive strength of 3,500 psi.

At the north end of the retaining wall (as depicted in Figure 7), retaining wall construction should proceed from top down by excavating in front of the wall while concurrently placing wood lagging from the top of wall downward to not less than 1.5 ft below the level of the anchors. The anchors should then be installed, load tested, and locked off. Excavation in front of the wall and installation of lagging should then proceed to the bottom of the wall.

In the main portion of the landslide (as depicted in Figure 8), the retaining wall construction should proceed from the existing ground surface downward by excavating in front of the wall while concurrently placing wood lagging downward to the bottom of the wall. The wood lagging should then be placed above the existing ground surface concurrently with placement of compacted wall backfill to a level of 2 ft above the anchor location. The anchors should then be installed, load tested and locked off. Placing the upper wood lagging and backfilling to final grade behind the retaining wall should then proceed to final grade.

Each anchor should be load tested in accordance with Ref (4). The anchors should be provided with Class I corrosion protection as defined in Ref (4).

4 Post Tensioning Institute (PTI) “Recommendations for Prestressed Rock and Soil Anchors”, PTI DC35.1-14, Copyright 2014.

8.2 Site Preparation Before Fill Placement

Conventional site preparation before placement of fill typically includes the removal of existing structures, pavements, underground utilities or pipes, trees and any other subsurface obstructions that might interfere with earthwork construction and stripping to remove obviously weak and organic-laden surficial soils, debris, brush and roots. We understand that these construction practices might not be acceptable considering the sensitive nature of the site in terms of cultural resources.

During preparation of the 2022 geotechnical report (Ref 1), the NPS specifically asked whether the broken and displaced pavement from Given Hill Road can be left in place. While there were concerns at that time that voids might exist below the displaced asphalt pavement, recommendations were provided to address those concerns during construction. Subsequent to completion of the 2022 report and associated plans and specification, the NPS has cleared the area of human remains and prefers that the displaced pavement be removed.

Organic matter, and broken sections of pavement, curb, drain tile, etc. that are larger than

12 inches in any dimension should be removed before the placement of fill east of the proposed wall alignment. The surface exposed after removal of the rubble should be proofrolled before fill placement. The surface should then be covered with an 18-inch-thick layer of 610 crushed limestone to create an all-weather working surface. 610 Limestone shall meet the requirements of

MDOT Section 703.04- Aggregate for Crushed Stone Courses. Debris removal and excavation along Given Hill Road should progress from north to south as soldier piles are installed. Any excavated areas along Given Hill Road should be backfilled within 24 hours of excavation to the pre-excavation elevation.

We recommend that the NPS review specification Section 31 23 03.00 09 Excavation, Fill and Backfill and provide specific guidance regarding acceptable construction practices in the project area. As indicated previously in this report, we anticipate that close observation and possibly guidance by NPS archaeologists and cultural resource specialists will be necessary during construction.

In areas to be paved, there is often some delay between completion of earthwork operations and placement of the pavement structure materials, possibly resulting in deterioration of subgrade conditions. Thus, we recommend that the density and stability of the subgrade soils be confirmed or reestablished immediately prior to construction of the new asphalt pavement.

8.3 Earthwork Construction

Select fill should consist of soils classified as ML, CL, SM or SC in accordance with the

Unified Soil Classification System (ASTM D 2487). All select fill shall have a liquid limit less than 45 and a plasticity index (PI) less than 25. Locally available loess silts (ML) and clayey silts

(ML) can be utilized as select fill, but it should be realized that stringent moisture control will be required during compaction of the silts (ML), because these soils are very susceptible to pumping at higher water contents.

Fill materials should be compacted from lifts not exceeding 9 inches in loose thickness to not less than 95 percent of standard Proctor maximum dry density (ASTM D 698). Silts (ML) and silty sands (SM) should be compacted at moisture contents within 2 percentage points of the optimum water content. All other fill materials should be compacted at moisture contents within 3 percentage points of the optimum water content. Where hand-controlled mechanical tampers are utilized to compact fill materials, the loose lift thickness should be limited to 5 inches. Stability must be evident during compaction of each lift before any subsequent lifts of fill material are added.

Special care will be required to properly compact backfill adjacent to and inside the flanges of the soldier piles. Backfill should be placed in maximum 5-inch-thick loose lifts and compacted using hand-controlled mechanical tampers.

The work will be performed within the limits of an active landslide, within a National

Cemetery, within a National Military Park. The entire work site is considered a cultural resource.

The Contractor should select and use the smallest equipment that will accomplish, in a practical way, the required compaction while minimizing disturbance to the site.

Portions of the berm where the final slope will be steeper than 2.5H:1V should incorporate geogrid reinforcement to achieve the desired safety factor. The geogrid should have a Long-Term

Design Strength (LTDS) ≥ 1100 lb/ft. A small temporary excavation will be required at the toe of the slope to allow for installation of the geogrid (see Figure 8). This temporary excavation and reinforced soil slope construction should not be performed until after the soldier piles are installed and after the excavation west of the retaining wall is accomplished.

In areas where new fill is placed against an existing slope, benching into the existing slope is required in order to place and compact the fill in horizontal layers. The vertical face of the existing slope resulting from the benching operation should be a minimum of 1 ft in height but not more than 2 ft in height. We understand that benching, particularly at the scarp, might not be acceptable due to the need to avoid disturbing graves and to protect cultural resources.

Laboratory classification tests, including Atterberg limit determinations and grain-size analyses, should be performed on the fill soils initially and routinely during earthwork operations to check for compliance with the recommendations provided herein. Field moisture/density tests should be performed at a rate of not less than one test for each 250 cubic yards of material placed but not less than one test in each 12-inch thickness of compacted fill.

It should be noted that loess silts (ML) and clayey silts (ML) are very susceptible to erosion.

In order to reduce erosion, we recommend that the final surface in fill areas be plated with a minimum 12-inch thickness of select fill having a PI of 6 or greater. The quick establishment of turf will be critical on any freshly graded and filled areas. The quickest and most effective means of establishing turf will be to place sod.

8.4 Subsurface Drainage Features

Two subsurface drainage features are recommended to collect and remove groundwater in the vicinity of the retaining wall. The locations of these features are shown on the typical sections on Figures 7 and 8. A combination blanket drain/underdrain in the wall backfill should begin at about Sta 0+70 and extend south to about Sta 2+30. The sand blanket drain should be placed adjacent to the scarp as shown on Figure 8. The blanket drain should initiate above an underdrain located near the base of the scarp. An outlet pipe should extend from the underdrain passing through the lagging to a discharge structure/apron on the slope below the wall. Details of the blanket drain/underdrain are shown on Figure 10.

A French drain should be constructed in front of the wall from about Sta 0+00 to about Sta

2+30. A detail of the French drain is shown on Figure 10.

The drains are designed as two-stage filter systems. Perforated/slotted pipe should be encapsulated in No. 57 coarse aggregate. The No. 57 coarse aggregate should be encapsulated in concrete sand. The No. 57 stone should be encapsulated in a layer of MDOT Type V geotextile.

The filter layer thicknesses and configurations shown on Figure 16 are the minimum values. Layer thicknesses can be increased for ease of construction. However, the bottom elevations of the drains should be as shown on the drawings.

Placing of the filter material by methods which will tend to segregate particle sizes should not be permitted. Filter material should be maintained free of muddy water and foreign materials.

Filter material found to contain foreign materials or muddy water should be removed and replaced with satisfactory material.

8.5 Groundwater Control During Construction

Zones of saturated silts (ML) and silty clays (CL) were encountered in previous borings advanced at the site, and piezometers indicate multiple levels of perched groundwater exist within the Vicksburg Group. Saturated zones might be encountered in temporary open excavations made for installation of the French drains. The quantity of flow from saturated silts (ML) and silty clays

(CL) is expected to be relatively low and can likely be managed with sumps and pumps. Perched groundwater will likely be encountered during pre-drilling for the soldier piles. The use of temporary casing might be required to maintain a stable borehole through saturated zones of silt and sand. The Contractor should be solely responsible for managing any groundwater encountered during construction.

8.6 Quality Control and Quality Assurance During Construction

Quality Control and Quality Assurance (QC/QA) observation and testing during construction will be critical to the successful construction and long-term performance of the project. The specifications should require the contractor to submit detailed QC plans for each aspect of construction, including, but not necessarily limited to, soldier pile installation, earthwork construction, drain construction, and ground anchor construction. An appropriate level of QA testing should be performed by the owner. QA observation of the construction should be full time.

The NPS should develop a QA plan specific to the protection of cultural resources.

8.7 Maintenance

The site soils are highly erodible, and the stability of the slopes at the site is sensitive to increases in groundwater levels. The NPS is well aware of these characteristics of the site soils and has an aggressive and consistent maintenance program in place to monitor and react to adverse changes in site conditions. Regular inspections of the completed work and timely maintenance when issues are identified will be important to ensure satisfactory performance.

Surface and subsurface erosion are the most common and frequently observed wasting processes in the park. Surface erosion generally initiates as shallow rills, but can quickly progress to deep rills and/or gullies if left unattended. Repairs to surface erosion should also include an evaluation of surface drainage to ensure that unintended concentrations in flow have not developed. If concentrated flow has developed outside of intended drainageways, modifications should be made to either redirect water into designed drainage features or to redistribute drainage into sheet flow. The entire work area will be susceptible to surface erosion and should be subjected to regular inspections after the project is completed, particularly during the first year while permanent ground cover is being established.

Subsurface erosion generally expresses at the surface in the form of dropouts or sinkholes.

Subsurface erosion often initiates where there is buried debris, decaying organic matter, or tension cracks that create avenues for surface water to enter the ground and initiate subsurface erosion.

The entire work area will be susceptible to subsurface erosion and should be subjected to regular inspections after the project is completed.

The area between the south end of the soldier pile wall and the existing soil nail wall is of particular concern. The existing ground surface is relatively steep, and it is likely the area will be disturbed to some degree during the proposed construction. No specific stabilization measures are planned in this area as it is outside of the limits of the landslide. The area should be monitored during and after construction for evidence of developing erosion and/or sloughing.

9.0 REPORT LIMITATIONS

The analyses, conclusions, and recommendations discussed in this report are based on conditions as they existed at the time of our field investigation and further on the assumption that the exploratory borings are representative of subsurface conditions throughout the area investigated. It should be noted that actual subsurface conditions between and beyond the borings might differ from those encountered at the boring locations. If subsurface conditions are encountered during construction that vary from those discussed in this report, Burns Cooley

Dennis, Inc., should be notified immediately in order that we may evaluate the effects, if any, on our analyses. Burns Cooley Dennis, Inc. cannot assume responsibility or liability for the adequacy of recommendations if we do not observe construction.

This report has been prepared for the exclusive use of Stantec for specific application to the geotechnical aspects of remediation of slope failure at the Vicksburg National Military Park

National Cemetery in Vicksburg, Mississippi. The only warranty made by us in connection with the services provided is we have used that degree of care and skill ordinarily exercised under similar conditions by reputable members of our profession practicing in the same or similar locality. No other warranty, express or implied, is made or intended.

FIGURES

Sta. 1+25

B-6

B-5

Sta. 1+90

FIGURE 1

B-5

B-6

2024 BORING LOCATION

B-4

B-1

B-3

2021 BORING LOCATION

2002 BORING LOCATION

1999 BORING LOCATION

GMGM

0+

0+

0+

0+

0+

1+

1+

1+

1+

1+

2+

2+

2+

2+

2+80

3+00

3+20 3+25

B-2/P-1

FIGURE 2

MAJOR DIVISIONS

C

O

AR

SE

-G

R

AI

N

ED

S O

IL

S

FI

N

E- G

R

AI

N

ED

S O

IL

S

GRAVELS

SANDS

WELL GRADED GRAVEL, GRAVEL-SAND MIXTURE

POORLY GRADED GRAVEL, GRAVEL-SAND MIXTURE

SILTY GRAVEL, GRAVEL-SAND-SILT MIXTURE

CLAYEY GRAVEL, GRAVEL-SAND-CLAY MIXTURE

WELL GRADED SAND, GRAVELLY SAND

POORLY GRADED SAND, GRAVELLY SAND

SILTY SAND, SAND-SILT MIXTURE

CLAYEY SAND, SAND-CLAY MIXTURE

SILT WITH LITTLE OR NO PLASTICITY

CLAYEY SILT, SILT WITH SLIGHT TO MEDIUM PLASTICITY

SILTY CLAY, LOW TO MEDIUM PLASTICITY

SANDY CLAY, LOW TO MEDIUM PLASTICITY (30% TO 50% SAND)

CLAY, HIGH PLASTICITY

ORGANIC CLAY OR SILT OF MEDIUM TO HIGH PLASTICITY

PEAT, HUMUS, SWAMP SOIL

SAMPLE TYPES

(Shown in Sample Column)

Shelby Tube

Split Spoon

No…