Chapter 13 — Slope Stability Analysis

Slope stability governs the design of railroad embankments, cut slopes, and side-hill fills across all nine railroad client systems. An embankment failure — whether a rotational slide in soft foundation clay, a translational wedge along a shale bedding plane, or a surficial slump after heavy rainfall — can result in track loss, derailment, emergency speed restriction, and weeks of service disruption. This chapter develops the complete slope stability analysis framework: the Bishop Simplified method for circular failure surfaces, the Janbu Generalized method for non-circular surfaces, pseudo-static seismic analysis, tension crack depth, progressive failure in strain-softening clays, and cut slope design in rock. Minimum factors of safety follow USACE EM 1110-2-1902 as the primary criterion, supplemented by client-specific requirements from CSXT ES, NSR ESM, BNSF EI, UPRR EIM, CPKC TS, Amtrak ES, G&W EB, CN ES, and AREMA MRE Chapter 8. Three worked examples address an NSR Appalachian embankment over residual clay, a BNSF staged construction embankment over soft foundation soil, and a CPKC prairie cut slope in glacial till. A full slope stability analysis checklist, representative design tables, and five practice questions with complete answers are provided.

Slope failures impose direct costs on railroad operations through emergency repair, traffic rerouting, crane mobilization, and environmental spill response. They impose indirect costs through locomotive speed restrictions, customer delivery failures, and regulatory scrutiny. The Federal Railroad Administration does not prescribe slope stability factors of safety directly, but 49 CFR Part 213 mandates that track geometry be maintained within class tolerances — a failed or deforming slope that moves the track out of tolerance triggers an immediate slow order or line closure under FRA authority.

Each railroad client's approach to slope stability risk reflects its operational philosophy and geographic exposure:

• CSXT ES: Requires FS ≥ 1.5 for long-term static stability of mainline embankments; FS ≥ 1.3 during staged construction with pore pressure monitoring. Coastal and Piedmont territory includes frequent soft clay and organic soils.
• NSR ESM: Mandates slope stability analysis for all embankments > 15 ft high and all cut slopes in the Appalachian coal territory (WV, VA, KY). NSR requires peer review by its Geotechnical Engineering group for FS < 1.7. Shale rock cuts are a specific concern in the Pocahontas and Clinch districts.
• G&W EB: Low-speed branch line operations reduce the consequence of slow-rate creep failures; G&W accepts FS = 1.3 for non-critical sidings with annual monitoring. Active slide areas require FS ≥ 1.5.
• UPRR EIM: Requires FS ≥ 1.5 static for all mainline embankments and fills supporting intermodal facilities. Loess embankments in Nebraska and Iowa are subject to rapid drawdown from river flooding — FS must be checked for drawdown condition.
• BNSF EI: Transcon and northern mainlines require FS ≥ 1.5 static and FS ≥ 1.1 pseudo-static (seismic). Montana and Idaho mountain territory has documented rockfall and debris flow hazard — slope hazard rating required per BNSF EI Appendix D.
• CPKC TS: Prairie cut slopes in glacially overconsolidated clays are subject to delayed failure — progressive failure along residual strength surfaces is the design concern, not peak strength. CPKC requires residual φ'_r analysis per Skempton (1985) for all cut slopes in Cretaceous clay shale.
• Amtrak ES: NEC embankments and retaining walls require FS ≥ 1.75 static due to high-speed operations and minimal access for emergency response. Amtrak requires SLOPE/W or SLIDE analysis submitted electronically with every slope stability deliverable.
• CN ES: Muskeg and quick clay in northern Ontario and Quebec require undrained stability analysis with Bjerrum-corrected VST strengths (Chapter 7); FS ≥ 1.4 for staged construction fills.
• FRA/AREMA: AREMA MRE Chapter 8 recommends FS ≥ 1.5 for new construction embankments; emergency repair slopes may be temporarily accepted at FS = 1.1–1.2 pending permanent stabilization.

Reference graphics

Railroad slopes exhibit four primary failure modes that drive the selection of analytical method:

1. Rotational Circular Slide: The failure surface approximates a circular arc through a homogeneous or layered cohesive material. Typical in embankments over soft clay foundations, or in cut slopes through uniform residual soils. Analyzed by Bishop Simplified or Spencer methods.
2. Translational (Planar) Slide: The failure surface is planar, controlled by a weak layer — bedding plane, shale parting, geosynthetic interface, or loose fill lift. Common in NSR and CPKC rock cut territory and in BNSF prairie fills over weak bentonitic shales. Analyzed by the infinite slope method or Janbu with a specified non-circular surface.
3. Compound/Non-Circular Slide: Failure surface transitions from rotational through a weak layer to translational. Common in overconsolidated clays with a thin weak stratum. Requires Janbu Generalized or Morgenstern-Price analysis.
4. Progressive Failure: In strain-softening clays and clay shales, the mobilized strength progressively reduces from peak to residual along the failure surface. A slope that was stable at FS = 1.5 (peak) may eventually fail at FS = 1.0 (residual) without any change in loading. Critical for CPKC prairie clay shale cuts and NSR Appalachian weathered shales.

Reference graphics

Supplemental figures · LEVI Library

Slides from downhill and from uphill: field categories

Table 13-5. Slide categories used in the field course

Category	Field signature	Typical response
1.1 Scarp far below track	Tension crack on lower slope; shallow creep shown by leaning poles, fences, trees	Walk the slope; monitor and plan mitigation
1.2 Scarp at shoulder	Crack in or at the toe of ballast; surfacing and cross-level defects	Instability retrogressing toward the track; investigate and act
1.3 Scarp directly below track	Main slide body at ballast level; slow or rapid	Drainage/subdrainage, geometry (flatten, berm, divert) or structure (walls, nails)
1.4 Scarp uphill of track (very deep)	Cross-level drops toward the high side	Large toe berm about 1/3 of slide height; walls do not work
2.1 Uphill toe falling on or pushing track	Crack far above; toe bulge in ditch	Gravity or anchored wall; dewatering
2.2 Uphill toe heaving track	Track lifted and shifted sideways	Gravity or anchored wall; dewatering
2.4 Slide taking track down	“Whole hill moving”; up to about 60 ft deep; reactivated old slides	Very difficult; major engineering

Figures · click to zoom

After Geotechnology for Railroaders Program, Module 5; §03–07. Full reference under Supplemental Sources.

For railroad slopes in seismically active regions (UPRR western network, BNSF Cascades/Rockies, CPKC Rockies, Amtrak NEC through New York/Connecticut), a pseudo-static horizontal force is applied to the failure mass to represent earthquake loading.

USACE EM 1110-2-1902 (2003) provides the standard minimum FS criteria used across all nine railroad clients as a baseline:

Client-specific deviations from USACE baseline reflect the consequence of failure on high-density operations. Amtrak's FS = 1.75 for long-term static reflects the catastrophic consequence of a slope failure under a 125 mph Acela Express.

In overconsolidated clays and clay shales, the post-failure (residual) strength is significantly lower than peak strength. The residual friction angle φ'_r is measured by ring shear test (ASTM D 6467) or repeated direct shear at large displacement.

Reference graphics

Supplemental figures · LEVI Library

For rock cut slopes along NSR Appalachian routes, BNSF Montana/Idaho territory, and CPKC Rocky Mountain corridor, kinematic analysis of joint and bedding plane orientations governs slope geometry over soil mechanics methods.

Reference graphics

Rockfall management: stabilize, protect, warn

Table 13-6. Rock-slope management strategies and indicative costs

Strategy	Methods	Notes
Stabilizing	Drainage, trimming, scaling, shotcrete, bolting; retaining walls; stabilizing berms; horizontal drains	Highest safety and reliability impact; cost varies; scaled-rock disposal can be a problem
Protecting	Catchment nets; barrier walls; lock blocks; rock sheds (~$1–10M); tunnels (~$3–50M); flumes (~$600K); rock mesh (~$100–250K)	Nets for confined chutes and unscalable sources; keep catch ditches and wall accumulation zones clean
Warning	Slide-detector fences; mercury and tip-over switches; temporary radio units	A warning layer, not a fix: about 90% of trips prove passable; maintenance burden and false positives

A Jasper near-miss shows the limit of warning devices: the fence tripped as designed, but the train could not stop on the grade. Use engineering-grade concrete for lock blocks and source matching keys. Mark already-documented fallen rocks with an X.

Figures · click to zoom

After Geotechnology for Railroaders Program, Module 6.4; §02–04. Full reference under Supplemental Sources.

UPRR Missouri River territory and BNSF river crossing embankments are subject to rapid drawdown — when a river in flood recedes quickly, the phreatic surface within the embankment lags behind the external water level, creating a destabilizing outward seepage gradient.

The three-stage approach per USACE EM 1110-2-1902 Appendix F:

1. Analyze embankment at full pool (phreatic surface at top of bank); compute total and effective stresses.
2. Instantly lower the pool; the phreatic surface remains at its high position.
3. Analyze with consolidated-undrained (CU) strength of the embankment materials after consolidation at full pool, but with the external water pressure removed.

For materials with k > 10⁻³ ft/day (sands and gravels), drained strengths can be used at the lowered pool because drainage occurs faster than drawdown. For clays with k < 10⁻⁵ ft/day, CU strengths apply throughout.

Rapid drawdown in the field: the 12–24 hour window

While a pond is high, its weight helps hold the slope despite saturation. When it drops quickly, that support disappears while the fill is still saturated, and seepage reverses outward. Field guidance is that the slope stays at its least stable for about 12 hours after drawdown, recovering over 12–24 hours as the bank drains. Monitor for at least that long, and for at least three days where the drop is 2 ft or more. Beaver-dam or culvert releases create the same condition. The Nakina derailment of 1992, in which two employees were killed, was a rapid-drawdown failure beneath a moving train.

Figures · click to zoom

After Geotechnology for Railroaders Program, Module 7.1, Module 4, Module 6.5; Module 7.1 §06; Module 4 §10. Full reference under Supplemental Sources.

Case study: 13.16.1 Flood drawdown failures along the Brisbane River (Look 2023), in §13.16 Case Studies.

1. Figure 13-1: Rotational failure geometry — slice numbering, weight W_i, base angle α_i, piezometric line.
2. Figure 13-2: Bishop Simplified iteration convergence — FS vs. iteration number chart.
3. Figure 13-3: Non-circular failure surface through weak clay layer — Janbu slice geometry.
4. Figure 13-4: Tension crack diagram — cohesive slope crest with water-filled crack, force vectors.
5. Figure 13-5: Pseudo-static seismic force application — horizontal F_eq on failure mass centroid.
6. Figure 13-6: USACE FS vs. loading condition matrix — graphical summary of Table 13-1.
7. Figure 13-7: Progressive failure concept — peak to residual strength along failure surface.
8. Figure 13-8: Planar rock slide geometry — tension crack, water uplift U, slide block weight W.
9. Figure 13-9: Rapid drawdown stages — pool level, phreatic surface at drawdown, seepage force.
10. Figure 13-10: Staged construction approach — fill lifts, pore pressure dissipation, FS vs. time.

Reference graphics

1. Search the critical failure surface systematically: The minimum FS may not occur at the toe; for embankments over soft clay, the critical circle often exits through the foundation rather than the embankment slope face. Use a grid search over at least 200 trial centers and 5 radii per center in SLOPE/W or equivalent.
2. Never use peak strength for clay shale or Appalachian weathered shale: Identify pre-existing slickensides in the boring core before selecting design strength. A single set of slickensides on a joint dipping toward the cut is sufficient to require residual-strength analysis. CPKC TS and NSR ESM both require written confirmation of slickenside inspection in the geotechnical report.
3. Always include the tension crack in cohesive slope analyses: A water-filled tension crack at the crest reduces FS by 5–15% and is the most commonly omitted factor in routine slope analyses. USACE EM 1110-2-1902 requires it; so do CSXT ES and Amtrak ES.
4. Staged construction requires real-time instrumentation and a trigger plan: Specifying staged construction in a geotechnical report without a corresponding instrumentation plan and construction trigger criteria (maximum pore pressure ratio, minimum settlement rate before next lift) is incomplete and creates construction-phase liability. BNSF EI and CN ES require the trigger criteria to be included in the geotechnical special provisions (GSP).
5. Rapid drawdown is often the governing condition for river-crossing embankments: UPRR Missouri River and BNSF Platte River embankments have been subject to drawdown failures after spring flood recession. The rapid drawdown case per USACE EM 1110-2-1902 Appendix F must be analyzed for any embankment adjacent to a river or reservoir with seasonal water level fluctuation > 5 ft.
6. Report FS for all loading conditions — not just the critical one: CSXT ES and NSR ESM require the geotechnical report to tabulate FS for end-of-construction, long-term static, steady-state seepage, and seismic (where applicable) — even when some conditions clearly are not critical. This demonstrates due diligence and aids the client's Bridge/Track Engineering review.
7. Coordinate with drainage design: Most railroad slope failures occur after prolonged rainfall that raises pore pressures above the design phreatic surface. The slope stability analysis should be accompanied by a drainage design recommendation (ditch grades, blanket drains, horizontal drains) that maintains pore pressures at or below the assumed steady-state seepage condition.

Reference graphics

The slope remediation toolbox, in order

Consider remedial measures in this order: (1) drainage and subdrainage; (2) geometry, meaning a toe berm, flattening or moving the track; (3) structural support; (4) improving soil properties. Warning devices supplement these.

• Slopes flattened to 2H:1V are generally very stable. A toe berm about 1/3 of slope height and 12–15 ft wide adds substantial stability; a 30-ft slope needs a berm about 10 ft high. Fill left “hanging” on the shoulder reduces stability.
• Never leave riprap or fill at the top of an unstable slope. Place riprap from the toe upward. Where a slope has failed into muck, key a rock berm into virgin ground first and add a subdrain (“ballast sausage”).
• Air-dump fill at a safe point and push it into the slide with an excavator. Never dump directly onto unstable ground.
• If there is room uphill, shifting the track about 6 ft and flattening the old shoulder can be the whole fix.
• Use a smooth-drum compactor on silt and clay backfill; a sheepsfoot roller can liquefy it (Camrose 56.4). Clear ice-plugged culverts before building any structural fix.
• Never go it alone: check with all agencies and follow environmental rules. Where a third party causes the instability (a hydro dam on the St. Maurice Sub), the cost may shift to them.

PG Sub Mile 445.7 (April 2013) is typical of most slides seen along the track. Snowmelt ran down small gullies, disappeared into the fill, raised pore pressure and took out the embankment. The repair keyed in a toe, rebuilt a 2:1 slope about 50 ft high from air-dumped fill, and added culverts.

Figures · click to zoom

After Geotechnology for Railroaders Program, Module 6.1, Module 7.1; Module 6.1 §02–04; Module 7.1 §05. Full reference under Supplemental Sources.

Project management · Swearingen (n.d.)

PERT Three-Point Estimates and the Probability of Finishing a Slope Repair Before a Deadline

PERT handles uncertain durations by asking for three estimates per activity: optimistic a, most likely m and pessimistic b. A beta distribution gives the expected time t = (a + 4m + b) ÷ 6 and variance σ² = [(b − a) ÷ 6]². A forward pass gives the early start and finish (ES = largest EF of the predecessors; EF = ES + t), and a backward pass gives the late times (LF = smallest LS of the successors; LS = LF − t). Slack is LS − ES; activities with zero slack form the critical path. Project variance is the sum of the variances of the critical activities. Treating completion time as normally distributed, the probability of finishing by deadline D follows from Z = (D − T) ÷ σ.

Table 13-PM1. Worked example time estimates, weeks (after Swearingen n.d., slide PMT-17)

Activity	a	m	b	t	σ²	Slack
A	1	2	3	2	0.11	0
B	2	3	4	3	0.11	1
C	1	2	3	2	0.11	0
D	2	4	6	4	0.44	1
E	1	4	7	4	1.00	0
F	1	2	9	3	1.78	6
G	3	4	11	5	1.78	0
H	1	2	3	2	0.11	0

In the slides’ example, a foundry has 16 weeks to install a pollution-control system. The critical path A–C–E–G–H gives T = 15 weeks with variance 3.11, so σ = 1.76 weeks. Then Z = (16 − 15) ÷ 1.76 = 0.57, which corresponds to about a 71.6% chance of meeting the deadline.

For railroad work: slope repairs often have to be finished before a wet season, a freeze-up or a planned outage. Use wide pessimistic estimates for activities controlled by subsurface conditions, such as soldier pile or anchor drilling in uncertain rock, dewatering, or excavation below the slide plane. Use narrow ranges for routine tasks. If the computed probability is low, change the plan (more rigs, a second shift, a staged repair that restores track first) instead of only reporting the risk. The method assumes a single critical path. Where a near-critical path also carries large variance, the true chance of finishing on time is lower than the calculation shows, so check those paths as well.

Source: Swearingen, J. (n.d.). The Basics of Project Management Techniques [lecture slides PMT-1 to PMT-40; author from document properties], slides PMT-13 to PMT-34. Summarized and paraphrased; figures reproduced for internal reference. Full reference under Supplemental Sources.

Project management · Pratt (2011)

Quantifying a Slide Repair: Section and Grid Methods for Cut and Fill

Pratt sets two rules for measuring earthwork. First, measure in bank measure: the actual in-place dimensions, with no allowance for swell or compaction. Those factors are applied later, when the work is priced. Second, keep each excavation category and each material separate, and measure disposal of surplus as total excavation minus common backfill.

He gives two manual methods for volumes. The grid method takes the cut or fill depth at each grid node and weights it by the area that node represents: one quarter-cell at a corner, two on an edge, and four in the interior. The section method suits long, narrow work such as road and railroad construction. Cross sections are plotted at regular stations, the cut and fill areas are measured at each section, and the volumes come from the average-end-area relation:

V = station spacing × (A₁/2 + A_N/2 + sum of the intermediate areas)

His example uses five stations 20 ft apart. The cut areas are 108, 40, 73, 91 and 0 ft², and the fill areas are 245, 248, 195, 195 and 328 ft². That gives 5,160 ft³ (191 CY) of cut and 18,490 ft³ (685 CY) of fill. All the cut is reused, so about 494 CY of fill must be imported. Accuracy depends mainly on how faithfully the section profiles are plotted, and closer station spacing improves it.

For railroad work: a slide or washout repair is a short, long-and-narrow job of exactly this kind. A reasonable procedure is:

1. Cut sections perpendicular to the track at a regular spacing from the survey or LiDAR surface. Add extra sections at the head scarp, the toe and any culvert.
2. On each section, plot the existing ground, the interpreted slip surface, and the repair geometry: removal limits, keyway, benches and buttress.
3. Tabulate the areas separately for removal of slide debris, excavation into competent material for the key, granular buttress or rock fill, and select embankment fill.

Each material has its own swell and compaction behavior and its own haul. Report the volumes in bank measure, state the section spacing and the survey date, and keep the section sheets in the calculation appendix. A slide that is still moving changes the existing-ground line, so the quantities will need updating before the bid.

Source: Pratt, D. J. (2011). Fundamentals of Construction Estimating, 3rd ed. Clifton Park, NY: Delmar Cengage Learning, pp. 57, 59–64. Summarized and paraphrased; figures reproduced for internal reference. Full reference under Supplemental Sources.

Question 8.1 Define the four primary failure modes for railroad slopes and describe which analysis method is most appropriate for each.

Answer: (1) Rotational circular slide — homogeneous or layered cohesive soil; Bishop Simplified or Spencer method for circular surface search. (2) Translational planar slide — failure along a discrete weak plane (bedding, shale parting, geosynthetic); infinite slope method (shallow, parallel to slope) or Janbu with a specified planar surface. (3) Compound/non-circular slide — failure surface transitions through weak layer; Janbu Generalized or Morgenstern-Price for full rigor. (4) Progressive failure — strain-softening clay or clay shale reduces mobilized strength from peak to residual over time; use Bishop or Janbu with residual strength parameters (c'_r, φ'_r from ring shear test). USACE EM 1110-2-1902 requires that the analysis method be justified in the geotechnical report based on the failure surface geometry expected for the site conditions.

Question 8.2 A railroad embankment has s_u = 500 psf in the soft foundation clay, γ_fill = 118 pcf, and a proposed fill height of 10 ft with 2H:1V slopes (β = 26.6°). Using Taylor's stability number N_s = 5.52, compute the undrained FS at end of construction. Does this meet BNSF EI's EOC minimum of FS ≥ 1.4?

Answer:

FS = Ns · su / (γfill · H)

= 5.52 × 500 / (118 × 10)

= 2,760 / 1,180

= 2.34 ≥ 1.4 ✓

Yes, FS = 2.34 comfortably meets the BNSF EI minimum. The critical check occurs if the foundation s_u were lower; for FS = 1.4, minimum s_u = 1.4 × 1,180 / 5.52 = 299 psf. CPTu soundings should confirm s_u > 300 psf throughout the foundation zone before assuming this slope geometry is adequate.

Question 8.3 Explain the difference between peak and residual shear strength and why CPKC TS requires residual strength analysis for all clay shale cut slopes.

Answer: Peak shear strength is the maximum strength the soil can mobilize before significant displacement has occurred, corresponding to the intact structure of the clay mineral fabric. After large displacements (typically > 0.5–2 in. for clay shales), clay platelets reorient parallel to the failure plane, reducing strength to the residual — the absolute minimum drained shear strength. For clay shales, φ' drops from 20°–28° (peak) to 6°–14° (residual). A cut slope analyzed at peak strength may appear stable at FS = 1.8, yet fail within months to years as swelling and progressive straining reduce mobilized strength to residual and FS drops below 1.0. CPKC's Saskatchewan mainline passes through extensive Cretaceous clay shale belts where this mechanism has caused repeated cut slope failures. CPKC TS mandates residual strength design to ensure that slopes will remain stable through their design life without progressive failure, not just immediately after excavation.

Question 8.4 A 30-ft high railroad cut slope in saturated clay (c' = 150 psf, φ' = 18°, γ = 122 pcf) has a tension crack at the crest. Compute the theoretical tension crack depth and the water pressure force per foot of slope width if the crack is fully water-filled.

Answer:

Tension crack depth (Eq. 13-8 for general case):

Water pressure force in tension crack (Eq. 13-9):

This 354 lb/ft horizontal force is added to the driving force in the slope stability analysis. For a 30-ft cut, this adds approximately 354/(122×30) = 0.10 to the driving stress ratio — not trivial. USACE EM 1110-2-1902 and NSR ESM both require this to be included for cohesive cut slopes.

Question 8.5 What is the pseudo-static horizontal coefficient k_h for a site with a_max = 0.25g, and what is the corresponding horizontal seismic force on a failure mass weighing 45,000 lb/ft? Does the resulting FS = 1.05 meet BNSF EI's seismic stability requirement?

Answer:

USACE EM 1110-2-1902 recommends k_h = 0.5 × a_max/g for pseudo-static analysis:

kh = 0.5 × 0.25 = 0.125

Feq = kh × W = 0.125 × 45,000 = 5,625 lb/ft

BNSF EI requires FS ≥ 1.1 for pseudo-static seismic analysis. The computed FS = 1.05 < 1.1 — this does not meet BNSF EI's requirement. Mitigation options include: (a) flatten the slope from the current geometry to increase resistance; (b) install a rock-fill toe berm to extend the failure surface and add resistance; (c) install deep drainage to reduce pore pressures and increase effective stress; or (d) install ground anchors or soil nails (Chapter 16). The geotechnical report must identify the governing seismic condition and recommend the most cost-effective mitigation for BNSF's review.

Field case studies by Burt G. Look, from Earthworks: Theory to Practice – Design and Construction (CRC Press/Balkema, 2023), chosen for this chapter’s topics. Each is summarized from the book with its key figures and a closing note on railroad application added by the authors of this manual. Credit for the case material belongs to Look; see item S10 under Supplemental Sources.

Case Study 13.16.1 — Flood drawdown failures along the Brisbane River

Case study · Look (2023) §9.7

Related section: 13.8 Rapid Drawdown Analysis

Look and Thorley (2011) studied riverbank failures in Brisbane during the 2011 floods. The annual probability of flood-induced riverbank failure was 2.7 × 10⁻³, with a further 1.25 × 10⁻³ for erosion-related failures. Because the floods submerged much of the alluvial bank, drawdown failures should recur at a similar rate with each comparable event. No deaths or injuries came from riverbank failures in 2011, or in the comparable 1974 flood.

The mechanism was sudden drawdown after the flood peak: excess pore pressure remained in the bank while strength softened, so the failure line fell as the load rose.

A practical lesson came from the council’s lead engineer. Traffic was stopped on the six-lane road along the top of the banks to unload the crest, and the emergency crews then parked their vehicles on the empty road, replacing moving traffic with static load.

For railroad work: the post-flood period is when river-bank and embankment-toe slides develop, so inspections should continue through the recession, not stop at the peak. If a crest is being unloaded, the instruction has to cover equipment staging, and train restrictions should not be replaced with parked work equipment.

Source: Look, B. G. (2023). Earthworks: Theory to Practice – Design and Construction. CRC Press/Balkema. Case study §9.7, pp. 302–304. Summarized and paraphrased; figures reproduced for internal reference. Full reference under Supplemental Sources.

Authors: Daniel P. Colwell and Mario Ruel. This chapter was written and compiled by Daniel P. Colwell and Mario Ruel for Railway Ground Hazard Risk Management.

Technical content is drawn from the published sources below, and credit for that material belongs to their authors and publishers. Tables, equations and figures taken from a specific source are also cited where they appear.

Primary sources

• USACE EM 1110-1-1804, Geotechnical Investigations (2001)
• USACE EM 1110-2-1902, Slope Stability
• AREMA MRE Chapter 1 — embankment design

Supplemental sources (LEVI Library)

• U.S. Army Corps of Engineers (2003). Slope Stability. EM 1110-2-1902
• Stark, T. D. et al. (2016). Development and Use of Fully Softened Shear Strength in Slope Stability Analyses, Third Draft. USACE / ASCE FSS Subcommittee
• Mott MacDonald (2011). The Effects of Railway Traffic on Embankment Stability. RSSB Research Project 1386 (Revised), Rail Safety and Standards Board
• Wyllie, D. C. and Mah, C. W. (2004). Rock Slope Engineering: Civil and Mining, 4th ed. Spon Press
• Amend, F. Slope Stabilization with High Tensile Wire Mesh. Geohazards in Transportation in the Appalachian Region Conference, Geobrugg North America
• Wang, S., Hagan, P. C. and Cao, C. (2016). Advances in Rock-Support and Geotechnical Engineering. Butterworth-Heinemann
• Look, B. G. (2023). Earthworks: Theory to Practice – Design and Construction. CRC Press/Balkema. Case studies §9.7

File locations for these documents are listed under Supplemental Sources — LEVI Library.

Project management sources

• Swearingen, J. (n.d.). The Basics of Project Management Techniques [lecture slides PMT-1 to PMT-40; author from document properties]
• Cretu, O., Stewart, R. and Berends, T. (2011). Risk Management for Design and Construction. Hoboken, NJ: John Wiley & Sons
• Jacob, D. and Müller, C. (Eds.) (2017). Estimating in Heavy Construction: Roads, Bridges, Tunnels, Foundations. Berlin: Ernst & Sohn
• Pratt, D. J. (2011). Fundamentals of Construction Estimating, 3rd ed. Clifton Park, NY: Delmar Cengage Learning