Chapter 3 — Track Foundation Elements — Railway Ground Hazard Risk Management

Railway Ground Hazard Risk Management

Primary sources: AREMA Manual for Railway Engineering (MRE), Chapter 1 (Roadway and Ballast) and Chapter 8 (Concrete Structures and Foundations); FRA Track Safety Standards, 49 CFR Part 213; FRA Bridge Safety Standards, 49 CFR Part 237; Selig, E.T. and Waters, J.M. (1994), Track Geotechnology and Substructure Management; Li, D., Hyslip, J., Sussmann, T. and Chrismer, S. (2016), Railway Geotechnics; applicable ASTM standards as cited; and Chapters 2, 11–13 and 16–17 of this manual.

New chapter drafted September 2026 for the Railway Ground Hazard Risk Management reorganization. Verify client-specific values against the governing client standard before use on a project.

Apps for this Chapter3

Apps on dirtbert.com that support this chapter. Each one runs in a web browser on a desktop or a phone, with nothing to install. How to use the apps (I.3).

NANCY pages ask you to accept the Use and Disclaimers once in each browser. MINA is a monitoring aid and is not for operational dispatch decisions.

3.0 Executive Summary

Chapter 2 follows the load path from rail to the top of subgrade. This chapter picks up where that path ends: the foundation elements that sit below the subgrade surface, or that interrupt it, and on which long-term track performance ultimately depends. In a ground hazard program these elements are where most failures originate — an old end-dumped embankment that was never compacted, a soft clay foundation that keeps consolidating under heavier cars, a culvert that pipes backfill into its barrel, or a bridge approach where stiffness changes abruptly.

The chapter defines each foundation element, the function it performs for the track, the governing references (AREMA MRE Chapters 1 and 8, 49 CFR Parts 213 and 237, and client engineering standards), what typically goes wrong, and what an investigation must capture to evaluate it. It closes with a consolidated investigation matrix keyed to ASTM test methods, performance criteria, practical recommendations, a checklist, and practice questions.

The governing idea: the track is only as good as the least-capable element beneath it. Ballast and subballast can be renewed in a work window; embankment bodies, foundation soils and buried structures cannot. Identify and characterize them early, because they drive both the risk rating (Chapter 1) and the cost of any fix (Chapter 17).

3.1 Defining Track Foundation Elements

For this manual, the track substructure is divided into the track bed (ballast, subballast and the prepared top of subgrade, covered in Chapter 2) and the foundation elements beneath and alongside it. The distinction matters because the two groups are maintained, inspected and funded differently.

Foundation Element	Function for the Track	Governing References	Typical Geotechnical Concern
Embankment (fill) body	Raises the formation to grade; carries track load to the foundation	AREMA MRE Ch. 1 (Roadbed); client earthwork specs	Uncompacted historic fill, ballast pockets, side-slope instability
Cut section and cut slopes	Lowers formation to grade through high ground	AREMA MRE Ch. 1 (Roadbed)	Seepage, softening of cut floor, slope raveling, rockfall
Foundation soils beneath fills	Supports the embankment and train load	AREMA MRE Ch. 1; Chapters 11–13	Soft clay, organics, sensitive or collapsible soils, karst
Drainage elements (ditches, underdrains, culverts)	Keep water out of and away from the substructure	AREMA MRE Ch. 1 (Culverts); 49 CFR 213.33	Blockage, piping along barrels, corrosion, overtopping
Bridge approaches and transition zones	Transition between ballasted track and a rigid structure	AREMA MRE Ch. 1 and Ch. 8; 49 CFR Part 237	Differential settlement ("bump"), stiffness change, backfill loss
Structural foundations adjacent to track	Support bridges, walls, retainers, signal and catenary structures	AREMA MRE Ch. 8; client standards	Surcharge from train load, scour, settlement, lateral movement
Crossings, turnouts and special trackwork	Carry concentrated and impact loads	AREMA MRE Ch. 1 and Ch. 5; client standards	Poor drainage, pumping, rapid geometry loss
Utility and pipeline crossings	Third-party facilities within the substructure	AREMA MRE Ch. 1 (Pipelines); client utility-crossing rules	Settlement or voids from leaks, trench backfill, casing failure

The consultant's responsibility is to identify which of these elements are present at a site, which one governs performance, and whether its condition is changing. A site that "just needs surfacing" every year is rarely a ballast problem — it is usually one of the elements in this table.

Ballasted track and slab track

Ballasted track is a flexible support; slab track (concrete slab or asphalt over a ballast-concrete layer and an anti-frost layer) is an inflexible support. Ballasted track remains the norm because it tolerates differential settlement, costs much less to build and gives good lateral resistance even at high speed. Slab track is most effective in tunnels, where it allows a smaller cross-section and reduces maintenance, but it is noisier.

The choice turns on construction cost (much higher for slab), maintenance cost (much higher for ballast) and the technical needs of the site. German figures from 2008 put slab track at €575–680 per meter against €365 per meter for ballasted track, before the extra earthwork and subgrade work that slab track needs. Slab-track maintenance was reported at about 10% (Germany) to 20–30% (Japan) of the ballasted-track figure. For a slab the subgrade must be essentially settlement-free, which moves the geotechnical risk from maintenance into design and construction.

Figures · click to zoom

Figure 3-1. Ballasted track (a) and slab track (b): layer sequence from tie to subsoil. Profillidis Fig. 7.4. — **Figure 3-1.** Ballasted track (a) and slab track (b): layer sequence from tie to subsoil. Profillidis Fig. 7.4.

After Profillidis (2014), §7.3, §17.8, Fig. 7.4. Full reference under Supplemental Sources.

3.2 Embankments

Most North American main lines were built between 1850 and 1920 on embankments that were end-dumped or wagon-placed with little or no compaction, then raised repeatedly by adding ballast. The result is a layered, variable fill that bears little resemblance to a modern engineered embankment.

3.2.1 Characteristics of Historic Railroad Embankments

Composition: Locally borrowed soil, often from side ditches, with inclusions of cinders, slag, ash, timber (old trestles buried in place), boulders and debris.
Ballast pockets and "ballast bulbs": Decades of surfacing drive ballast down into soft fill, forming water-holding pockets that soften the surrounding soil (Chapter 2, Section 2.5).
Low density: SPT N-values in uncompacted cohesive fill are commonly low and erratic; do not assume the body is uniform from a single boring.
Buried structures: Old culverts, abandoned trestle bents and timber cribbing may exist inside the fill and are a recognized source of sudden subsidence as they decay (Chapter 1, Section 1.3).

3.2.2 New Embankments and Widening

New fill for sidings, second main track or grade raises should be placed and compacted to the client specification. Typical practice, to be confirmed against the governing client standard:

Loose lift thickness typically 8 in. (thinner for hand-operated equipment).
Compaction to a specified percentage of maximum dry density by ASTM D698 (standard Proctor) or ASTM D1557 (modified Proctor), with higher density usually required in the upper few feet beneath the subballast.
Moisture within a specified range of optimum, verified by ASTM D6938 (nuclear gauge) or ASTM D1556 (sand cone) testing at a stated frequency (Chapter 15).
When widening an existing embankment, bench the new fill into the existing slope so the new material does not sit on a smooth, weak interface. Remove vegetation, topsoil and loose sloughed material first.
Match or improve on the drainage characteristics of the existing fill. A less permeable widening placed against a free-draining old fill can trap water and raise the phreatic surface in the original embankment.

3.2.3 Embankment Stability and Deformation Checks

Evaluate embankment side-slope stability using Chapter 13 methods, with the design train surcharge applied (Cooper E-80 converted to an equivalent strip load, per client criteria). A minimum static factor of safety of 1.5 is the common target (Chapter 1, Section 1.2); confirm with the client. Evaluate settlement of new fill on the foundation using Chapter 12 methods, and design ground improvement where needed using Chapter 17, including the Section 17.16 technology library.

Embankment slopes, benching and overbuild

Commonly used embankment side slopes (horizontal : vertical) quoted from Bowles: 1.5–2 : 1 for usual soils and 2 : 1 for gravel and sand and for erosion-prone soils. Slopes must follow from the geotechnical study, and the foundation beneath the embankment must be evaluated as well as the fill.

Where the original ground slopes steeper than 1 : 10, bench (step) the base before placing fill (Figure 3-2).
Overbuild the initial width and height to allow for post-construction compaction of the fill (Figure 3-3).
For very tall embankment sides, use a retaining wall or reinforced soil designed for earth pressure and train surcharge (Figure 3-4; see Chapter 16).
Compact a formation layer, where used, to 100% Standard Proctor and the embankment body to 95% (Chapter 2, §2.4).

Figures · click to zoom

Figure 3-2. Benching (stepping) the embankment base on steep ground. Profillidis Fig. 9.14. — **Figure 3-2.** Benching (stepping) the embankment base on steep ground. Profillidis Fig. 9.14.

Figure 3-3. Increased initial width (Δb) and height (Δh) to allow for compaction of the fill. Profillidis Fig. 9.15. — **Figure 3-3.** Increased initial width (Δb) and height (Δh) to allow for compaction of the fill. Profillidis Fig. 9.15.

Figure 3-4. Retaining wall at the toe of a very tall embankment side. Profillidis Fig. 9.16. — **Figure 3-4.** Retaining wall at the toe of a very tall embankment side. Profillidis Fig. 9.16.

After Profillidis (2014), §9.12.2, Figs. 9.14–9.16; slope values after Bowles (2001). Full reference under Supplemental Sources.

Century-old fill: what is actually under the track

Most mainline track is over 100 years old. It was often built without sub ballast, for lighter and slower traffic, by observational methods with a low factor of safety and compaction below modern standards. The core of the original fill can still be loose. Because train loads act mainly to about 5–10 ft below the ties, only the upper few feet reflect later work, and the rest is as built. Construction history is therefore worth recovering before planning a subgrade repair.

Figures · click to zoom

Figure 3-13. Original embankment construction: the core of many mainline fills. — **Figure 3-13.** Original embankment construction: the core of many mainline fills.

Figure 3-14. Horse-drawn construction methods, well over a century ago. — **Figure 3-14.** Horse-drawn construction methods, well over a century ago.

Figure 3-15. Track construction in the transcontinental era. — **Figure 3-15.** Track construction in the transcontinental era.

After Geotechnology for Railroaders Program, Module 3; §03. Full reference under Supplemental Sources.

Case study: 3.16.1 Thirty-year-old expansive clay embankments: where the moisture really settles (Look 2023), in §3.16 Case Studies.

3.3 Cut Sections and Cut Slopes

Cuts expose natural ground at formation level, where it is subject to groundwater seepage and repeated loading. Common issues:

Softening of the cut floor. Overconsolidated clays and clay shales swell and lose strength when unloaded and wetted. The track may perform well for years and then deteriorate once the floor softens.
Seepage at the toe of cut slopes. Groundwater daylighting at the toe feeds the ditch and the subgrade. Where ditches are shallow or blocked, water ponds at formation level.
Rock cuts. Rock cut slopes drive rockfall hazard (Chapter 1, Section 1.4). Weathering of shale and interbedded rock produces raveling that fills ditches and reduces drainage.
Ditch maintenance undercutting the toe. Ditch cleaning that steepens or removes support at the toe of a cut slope is a well-known trigger of slope movement (Chapter 4, Section 4.7).

Investigation in cuts should include borings or test pits at formation level. Also measure groundwater levels over time (Chapter 9), and test swell potential where the soil is plastic (ASTM D4546; ASTM D4829 for expansion index).

Cut slope values and protection

Before excavating a cut, study the geological formations along it, particularly the joint pattern (“diaclases”), so as to disturb the existing equilibrium as little as possible. Design should balance safety, cost and fit with the surrounding landscape. Commonly used cut slopes (horizontal : vertical) quoted by the book are:

Table 3-1. Commonly used cut slopes (H : V)
Material	Slope
Silt	2.5–3 : 1
Muddy (silty) sand	2–3 : 1
Gravel	1.5–2 : 1
Stable clay	1–1.5 : 1

Indicative only. The governing slope comes from the site geotechnical study and the stability analysis in Chapter 13.

Protect cut slopes with shrubs or trees, which also blend the work into the landscape, and provide drainage on the slope to prevent softening.

After Profillidis (2014), §9.12.1; slope values after Bowles (2001). Full reference under Supplemental Sources.

Seepage into cuts and cutoff ditches

Where the subgrade in a cut sits at or near the water table, seepage softens it and causes uneven settlement. Gravel trench drains, with or without perforated pipe, lower the table; sometimes a permeable sub-ballast layer is enough. In deep soil cuts, cutoff ditches or benches intercept slope runoff and control erosion that would otherwise fill the side ditches. Keep them at a 1–2% longitudinal slope so water moves without accumulating above the cut slope.

Figures · click to zoom

Figure 3-20. Seepage into a cut where the subgrade is near or below the water table. — **Figure 3-20.** Seepage into a cut where the subgrade is near or below the water table.

Figure 3-21. A saturated side ditch in a cut: a visible clue to what is happening below grade. — **Figure 3-21.** A saturated side ditch in a cut: a visible clue to what is happening below grade.

Figure 3-22. Typical side ditch and cutoff ditch geometry. — **Figure 3-22.** Typical side ditch and cutoff ditch geometry.

After Geotechnology for Railroaders Program, Module 4; §05, §07. Full reference under Supplemental Sources.

Case study: 3.16.2 Four materials in one 1,300 ft cutting (Look 2023), in §3.16 Case Studies.

Case study: 3.16.3 Earthworks tender assessment for a balanced cut-to-fill job (Look 2023), in §3.16 Case Studies.

3.4 Foundation Soils Beneath Embankments

The foundation is the natural ground on which the embankment rests. Its behavior controls long-term settlement, deep-seated stability, and the response to new loading such as heavier cars, a second track or a grade raise.

Foundation Condition	Why It Matters	Key Tests	See
Soft to medium clay	Consolidation settlement; undrained failure under new fill	ASTM D2435 (consolidation), D2166 / D4767 (strength), field vane D2573, CPTu D5778	Chapters 12, 13
Organic soil, peat, muskeg	Large primary and secondary compression; "muskeg punching"	ASTM D2974 (organic content), D2435	Chapters 1, 12
Sensitive and quick clays	Brittle failure with little warning; retrogressive slides	Field vane (peak and remolded), D4648 fall cone	Chapters 1, 13
Loose saturated sands and silts	Liquefaction and cyclic softening	SPT D1586, CPT D5778, fines content D1140	Chapter 14
Collapsible soils (loess, some alluvium)	Sudden settlement on wetting	ASTM D5333 (collapse potential)	Chapter 4
Karst (soluble bedrock)	Sinkholes, raveling voids above bedrock	Geophysics (ASTM D6429 guide), probe borings, rock coring	Chapters 4, 7
Undermined ground	Mine subsidence (sag and pit)	Mine maps, borings to mine level, monitoring	Chapter 4

Borings for embankment evaluation should extend through the embankment and at least through the compressible or weak foundation layers into competent material. Use the depth criteria in Chapter 5.

Embankments on peat, muskeg and soft clay

A peat bog forms as vegetation fills a dying lake, and part of it may float on lacustrine soil. Its limited strength comes from intertwined fibers, so ditching or cable plowing that cuts them can turn a quiet section into a problem site. Old track over muskeg often sits on a corduroy mattress of crosswise logs; where it has decayed, peat pumps up through the gaps under traffic. Organic soils can compress by up to about 40%: 12 ft of peat can mean 5 ft of settlement. Excavate thin peat; build over thick peat and let it settle.

On soft clay or peat, an embankment can fail at its base as soft soil squeezes out from under the load. On peat, a stabilizing berm should be about twice as wide as the peat is deep, built in thin lifts over geogrid or woven geotextile, and never placed after removing the peat crust.

Figures · click to zoom

Figure 3-16. Peat bog formation as a lake fills with vegetation from the fringe inward. — **Figure 3-16.** Peat bog formation as a lake fills with vegetation from the fringe inward.

Figure 3-17. Base failure: soft clay squeezing out from under the embankment. — **Figure 3-17.** Base failure: soft clay squeezing out from under the embankment.

Figure 3-18. Track deflecting over muskeg (illustrative risk rating: High). — **Figure 3-18.** Track deflecting over muskeg (illustrative risk rating: High).

Figure 3-19. Track through deep muskeg near Bala Mile 30. — **Figure 3-19.** Track through deep muskeg near Bala Mile 30.

After Geotechnology for Railroaders Program, Module 2, Module 3, Module 5, Module 6.1; Module 2 §07, §08; Module 5 §06; Module 6.1 §03. Full reference under Supplemental Sources.

3.5 Drainage Elements

Water is the common factor in almost every substructure failure. FRA Track Safety Standards require that drainage and other water-carrying facilities under or immediately adjacent to the roadbed be maintained and kept free of obstruction (49 CFR 213.33). The geotechnical consultant evaluates whether the drainage system can do its job and what happens to the substructure if it cannot.

3.5.1 Ditches and Underdrains

Side ditches should keep the water table below the subgrade surface. As a practical check, the ditch flow line should sit below the bottom of subballast, with a positive grade to an outlet.
Underdrains (perforated pipe in filter aggregate, or geocomposite edge drains) are used where ditches cannot be deep enough. That includes cuts, tight right-of-way, yards, and platforms. Filter compatibility and outlet maintenance govern performance (Chapter 2, Section 2.8).
In multiple-track territory, intertrack drainage is frequently the weak point. Water trapped between tracks saturates the shared subgrade.

Build the drainage first

On new construction and rehabilitation, finish ditches, culverts, subdrains and other earthwork near the track before final subgrade preparation begins. Working in that order keeps equipment off the finished subgrade and soil out of the ballast. It also gives the contractor a site that drains while the track goes in.

Inspect subdrains (pipe, filter aggregate, fabric wrap and outlet) before they are covered.
Check ditch flow-line elevations and cross sections against the plans before track work starts. A ditch that is too shallow or too flat is far cheaper to correct before ballast is dumped.
Direct surface runoff away from geotextile that is placed but not yet covered, so fines do not blind it.

Construction-phase checks for these items are listed in Chapter 15, Table 15-S1.

After UFC 4-860-01FA (2004), ch. 9, §4b and §4c. Full reference under Supplemental Sources.

Choosing the drain type

Embankment side slope. This is the most effective side drain when the top of the subgrade is sloped to it and the ballast toe is close by. It handles outlet only, and the shoulder needs protection from erosion.
Open ditches. They are economical and carry large storms. Set the grade steep enough to avoid sedimentation but flat enough to avoid erosion, and line the ditch where velocity requires it. They are often neglected. A shallow or lined ditch does nothing for subgrade water.
French drains. There are three types: a single fine filter (low capacity), coarse aggregate wrapped in geotextile with AOS ≤ D₈₅ of the protected soil, and a two-zone graded filter. Fabric sized for silt and clay is hard to find.
Pipe drains. The filter D₈₅ should be more than 1.2–2 times the slot width and more than 1.0 times the hole diameter. Otherwise, wrap the pipe.
Geocomposite fin (edge) drains. They install continuously in a narrow slot and discharge to a carrier pipe. Three Southern Railway wet-subgrade sites installed in 1984 were performing well after 2 years.
Cross drains. They are rarely needed in sand and gravel. In fine soils the spacing they need is uneconomic, so raising the track on a granular blanket or replacing the subgrade is usually better.

Cuts with seepage. Make the side drains deep enough to lower the phreatic surface at least 3–6 ft below the subgrade. The depth depends on permeability, fissuring and pervious seams. Intercept ditches above a cut must not leak into the slope or cut across a critical slip surface.

Ballast pockets that hold water. If severe, reconstruct the substructure. If less severe, reshape the subgrade and replace the subballast and ballast.

After Selig & Waters (1994), ch. 13. Full reference under Supplemental Sources.

3.5.2 Culverts

Culverts pass water beneath the embankment and are foundation elements in their own right. AREMA MRE Chapter 1 covers culvert types, installation, bedding, backfill and cover. The geotechnical scope typically includes:

Bearing and settlement of the culvert foundation, especially at the embankment centerline where fill height is greatest.
Bedding and backfill specification and compaction adjacent to the barrel. Poorly compacted backfill beside the haunches is a common cause of track settlement over culverts.
Corrosion and abrasion potential for metal and concrete pipe. Test soil and water for pH, resistivity, chlorides and sulfates (e.g., ASTM G51, G187, D4327, or the AASHTO T 288–T 291 series as the client requires).
Piping and seepage along the outside of the barrel, and through open joints or corrosion holes. Backfill loss into the pipe leaves voids that collapse suddenly under load (Chapter 4, Section 4.5).
Hydraulic adequacy is normally the hydraulic engineer's responsibility. The consequence of overtopping and embankment washout belongs in the ground hazard risk evaluation, though (Chapter 1).

Sizing drainage: runoff methods and a lesson from the TGV Méditerranée

Water that can reach the subgrade must leave quickly. Give the top of subgrade a 3–5% crossfall toward transverse and longitudinal drainage devices. The book distinguishes two flows to design for: runoff from the track itself (in cut or on fill) and runoff from the contributing catchment. By the rational method:

Q_p (m³/s) = c · I · A / 3.6 (I in mm/h, A in km²)

Table 3-2. Runoff coefficients quoted for railway drainage
Surface	c
Track formation	0.85
Embankment slope	0.3–0.4
Impermeable surfaces	0.9–1.1
Cultivated soils	0.4–0.8
Sandy soils	0.3
Forest	0.2

The runoff coefficient is the least certain input and needs judgment. Semi-empirical formulas must be checked against observed flows. For the TGV Méditerranée high-speed line, rainfall-based formulas gave 1–5 m³/s/km² for the 100-year flow, but observed extremes were about 10 m³/s/km². The methods were recalibrated to the observations and the minimum diameter of cross-drainage under the track was set at 39 in.. Because rainfall intensity is increasing, the book suggests return periods well beyond the traditional 50–100 years, up to 1,000 years, with the probable maximum flood as the reference flow.

Figures · click to zoom

Figure 3-5. Drainage devices along the railway subgrade. Profillidis Fig. 9.4. — **Figure 3-5.** Drainage devices along the railway subgrade. Profillidis Fig. 9.4.

After Profillidis (2014), §9.14.2–9.14.3, Eq. 9.12, Fig. 9.4. Full reference under Supplemental Sources.

Cardinal rules of drainage, ditch capacity and culvert inspection

Water enters the substructure in three ways: precipitation, surface water (slope runoff and blocked or flooded ditches and culverts), and subsurface seepage. Poor drainage can increase surfacing frequency up to tenfold, increase track buckling 10–30% through loss of shoulder strength, and multiply slow orders. It is deferred because its effects are gradual and are easy to blame on rail, tamping or “soft track.”

Move water across the track at every opportunity and away from the right-of-way quickly; minimize how far it flows in upslope ditches.
Keep culvert inlets clear for at least 20 ft, and keep ditches free of vegetation and debris.
Maintain ditch gradients of at least 0.5%; keep cutoff ditches at 1–2%.
Leave vegetation on slopes; large trees drain and bind the ground.
Act on ditch and culvert obstruction long before it looks full: a quarter- or half-blocked channel has already lost much of its capacity.

Culverts are the most failure-prone drainage element. Check the inlet for scour, small slope failures, sand fill and open joints; the outlet for erosion, ice plugs and seepage; and the barrel for obstructions (beavers, sand, snow, trees) and sink holes over open joints. Open joints in concrete pipe let soil migrate out, leaving a void that collapses as a sink hole (“turtle hole”). Stone-box and blind-stone drains fail the same way through uncemented joints. Crawling the barrel is the only reliable condition check. Older 100-year sizing is being superseded; the course recommends sizing closer to a 500-year storm at many locations and never using pipe smaller than 36 in.

Figures · click to zoom

Figure 3-23. The three entry paths: precipitation, surface water, subsurface seepage. — **Figure 3-23.** The three entry paths: precipitation, surface water, subsurface seepage.

Figure 3-24. Side ditches intercept and divert water and lower the water table. — **Figure 3-24.** Side ditches intercept and divert water and lower the water table.

Figure 3-25. Progressive ditch obstruction: capacity falls well before the ditch looks full. — **Figure 3-25.** Progressive ditch obstruction: capacity falls well before the ditch looks full.

Figure 3-26. What to check at a culvert inlet, barrel and outlet. — **Figure 3-26.** What to check at a culvert inlet, barrel and outlet.

Figure 3-27. A sink hole opening beneath the ties over a culvert (illustrative risk rating: High). — **Figure 3-27.** A sink hole opening beneath the ties over a culvert (illustrative risk rating: High).

Figure 3-28. Joint separation and deterioration inside a concrete culvert. — **Figure 3-28.** Joint separation and deterioration inside a concrete culvert.

Figure 3-29. An old masonry stone culvert with uncemented joints. — **Figure 3-29.** An old masonry stone culvert with uncemented joints.

After Geotechnology for Railroaders Program, Module 4, Module 2; Module 4 §02–04, §07, §09; Module 2 §06. Full reference under Supplemental Sources.

3.6 Bridge Approaches and Transition Zones

Where ballasted track meets a bridge, a culvert with little cover, a grade crossing, or a direct-fixation or slab track section, support stiffness changes abruptly. Wheel-rail dynamic forces increase at the change and settlement accumulates on the softer side. The result is the familiar "bump at the end of the bridge," which demands frequent surfacing.

Causes: settlement of approach fill and foundation, loss of backfill behind abutments, poor drainage at the abutment, and the stiffness mismatch itself.
Geotechnical controls: select free-draining, well-compacted abutment backfill; drainage behind the abutment with positive outlets; treat or remove compressible foundation soils under the approach well ahead of track placement; and limit post-construction settlement of the approach fill.
Track-side transition treatments (coordinate with the track designer): approach slabs, graduated support stiffness (longer ties, tie spacing, under-tie pads, ballast mats), asphalt underlayment, geocell-confined granular layers, and cement- or lime-treated backfill wedges.
Bridge approach embankments fall within the bridge owner's inspection program under 49 CFR Part 237. Record approach settlement and backfill condition in the geotechnical report so it can be carried into bridge inspection records.

Stiffness transitions: ballasted track to slab track

Slab track is stiffer and less flexible than ballasted track, and ride quality cannot change abruptly at a single point. The book’s example is the transition for the Rheda slab system (Figure 3-6):

the transition zone has a slab section and a ballasted section;
under the transition slab, the ballast-concrete layer is thickened from 12 in. to 20 in.;
in the transition ballasted section the ballast is bonded; over A–B the ballast-concrete layer extends and partly replaces the subballast, and over B–C the subballast is extended;
two auxiliary rails run inside the running rails through the zone;
the anti-frost layer extends well into the transition.

The same principle (graded stiffness over a length, not a step) applies at bridge approaches and at tunnel portals, where the book notes that rubber pads of suitable thickness are placed at the entrance and exit to soften the change felt by passengers.

Figures · click to zoom

Figure 3-6. Transition between ballasted and slab track (Rheda system): plan and longitudinal section. Profillidis Fig. 17.11. — **Figure 3-6.** Transition between ballasted and slab track (Rheda system): plan and longitudinal section. Profillidis Fig. 17.11.

After Profillidis (2014), §7.3, §17.7, Fig. 17.11. Full reference under Supplemental Sources.

The bump at the end of the bridge

A bridge is stiffer than the track bed leading to it, so approach track deflects and settles more. The low spot that forms at the transition generates extra dynamic load, which deepens the low spot: a self-reinforcing cycle like the mud-spot snowball. Manage it from both sides. Stiffen the approach (for example with geogrid reinforcement) and soften the bridge (a more flexible deck, a ballast mat on a ballast deck, or under-tie pads on an open deck). Aim to minimize the difference in deflection rather than to eliminate settlement. Water management at the approach matters as much as the structural transition; mud pumping between the ties there points to a drainage problem below.

Figures · click to zoom

Figure 3-30. Settlement behavior of an approach embankment, a ballast-deck bridge and an open-deck bridge. — **Figure 3-30.** Settlement behavior of an approach embankment, a ballast-deck bridge and an open-deck bridge.

Figure 3-31. A bridge approach, where stiffness mismatch shows up first. — **Figure 3-31.** A bridge approach, where stiffness mismatch shows up first.

Figure 3-32. Mud pumping at a bridge approach. — **Figure 3-32.** Mud pumping at a bridge approach.

Figure 3-33. Internal drainage on a ballast-deck bridge. — **Figure 3-33.** Internal drainage on a ballast-deck bridge.

After Geotechnology for Railroaders Program, Module 3, Module 4; Module 3 §10; Module 4 §05, §09. Full reference under Supplemental Sources.

3.7 Structural Foundations Supporting or Adjacent to the Track

Bridges, retaining walls, ballast retainers, signal and communication structures and overhead catenary supports all have foundations within the zone influenced by train loading. AREMA MRE Chapter 8 covers the design of concrete structures and foundations for railway loading, and client standards add their own requirements.

3.7.1 Foundation Types

Foundation Type	Typical Railroad Use	Key Geotechnical Inputs	See
Spread footings	Abutments and piers on competent soil or rock; walls; signal bases	Bearing capacity, settlement, frost depth, scour depth	Chapters 11, 12
Driven piles (steel H, pipe, precast concrete, timber)	Bridge abutments and piers; trestle replacement	Axial and lateral capacity, drivability, downdrag	Chapter 11
Drilled shafts	Piers, catenary and signal foundations, walls	Side and tip resistance, lateral response, groundwater and caving	Chapter 11
Micropiles and helical piles	Limited access and under-track work; retrofit of existing structures	Grout-to-ground bond, buckling in soft soils	Chapter 17 (17.16)
Sheet piles and soldier piles with lagging	Ballast retainers, temporary and permanent shoring	Earth pressure with train surcharge, embedment, drainage	Chapter 16

3.7.2 Train Surcharge and the Influence Zone

Any excavation or structure close to an operating track must be designed for train surcharge. Class I railroads publish shoring guidelines that define a live-load influence zone projected from the track. Excavations and temporary works within it require railroad-reviewed design and monitoring (e.g., the BNSF–UPRR Guidelines for Temporary Shoring). Treat the client's current guideline as governing. Lateral earth pressure from the surcharge is evaluated using the methods in Chapter 16.

3.7.3 Scour at Structure Foundations

Scour at piers and abutments is one of the leading causes of bridge failure. Foundations in waterways must be evaluated for total scour (long-term degradation, contraction and local scour) following FHWA HEC-18 and the bridge owner's scour program. The geotechnical deliverable should state the soil or rock erodibility, the scour depth used, and whether the foundation remains stable at that depth.

Overhead catenary pole foundations

Catenary poles (cast or galvanized steel, prestressed or reinforced concrete) are spaced 164–246 ft, depending on pantograph oscillation, locomotive sway and climate. Excavation and backfill must be done carefully to keep ground settlement to a minimum (Figure 3-7).

On good subgrade the pole is set in an embedded block whose overturning moment is resisted by lateral soil pressure, M = c · B · L³, where B is the block width, L the embedment depth and c a coefficient that depends on the soil (Figure 3-8). On poor subgrade the poles are founded on a concrete slab, typically 6.6 ft × 11 ft at a depth of 3.6–4 ft.

Pole foundations sit in the shoulder, often on embankment edges. They should be included in slope-stability and settlement checks when a shoulder is steepened or a fill is widened.

Figures · click to zoom

Figure 3-7. Erection of electric traction poles at a double track in cut and on fill. Profillidis Fig. 20.15. — **Figure 3-7.** Erection of electric traction poles at a double track in cut and on fill. Profillidis Fig. 20.15.

Figure 3-8. Pole foundation on good-quality subgrade: lateral pressure distribution and M = c·B·L3. Profillidis Fig. 20.16. — **Figure 3-8.** Pole foundation on good-quality subgrade: lateral pressure distribution and M = c·B·L³. Profillidis Fig. 20.16.

After Profillidis (2014), §20.9, Figs. 20.15–20.16. Full reference under Supplemental Sources.

3.8 Grade Crossings, Turnouts and Special Trackwork

These locations concentrate dynamic load and trap water, so they fail sooner than open track on the same subgrade.

Grade crossings: The crossing surface and pavement restrict evaporation and cross-drainage. Water collects beneath the crossing panel, and pumping follows. Provide subsurface drainage through the crossing, and extend subballast and any geosynthetic separator beyond the crossing limits.
Turnouts and diamonds: Frog and switch impacts produce higher dynamic loads. Longer switch ties also span a wider zone of subgrade. Investigate subgrade strength and drainage specifically at turnouts, not by assumption from the adjacent open track.
Yard and industry track: Often built on minimal subballast with poor drainage, and heavily loaded by switching moves. Pavement-subgrade methods in Chapter 18 may apply where tracks are embedded in paved areas.

Turnouts as a foundation element

A turnout lets a vehicle pass from the main track to a diverging track without interrupting its movement. Its parts are the stock rails (fixed), the switch or tongue rails (moving), the frog at the intersection point, set at a frog angle stated as a tangent such as 1:9 and usually made of manganese steel, and the check rails opposite the frog. Turnouts must carry the line’s axle load, impose the fewest possible speed restrictions, resist wear and require little maintenance.

Keep vertical curves at least 16–33 ft clear of the beginning and end of switches and crossings.
Turnouts are levelled with light (portable) tamping equipment, which needs sound, clean ballast. A fouled or poorly drained bed under a turnout cannot be maintained to geometry, so the substructure under special trackwork deserves the same investigation and drainage attention as a bridge approach.
Turnouts on bridges, sharp curves and similar locations may also need locally increased lateral resistance (special tie shapes, roughened seats, transverse anchors).

Figures · click to zoom

Figure 3-9. A simple turnout in service (photo). Profillidis Fig. 15.1. — **Figure 3-9.** A simple turnout in service (photo). Profillidis Fig. 15.1.

Figure 3-10. Components of a turnout: stock rails, switch rails, mathematical point, frog angle, frog, check and wing rails. Profillidis Fig. 15.3. — **Figure 3-10.** Components of a turnout: stock rails, switch rails, mathematical point, frog angle, frog, check and wing rails. Profillidis Fig. 15.3.

After Profillidis (2014), §13.6, §14.11.2, §15.1–15.2, §16.8, Figs. 15.1, 15.3. Full reference under Supplemental Sources.

Asphalt underlayment at crossings, turnouts and bridge approaches

Its main North American use is on short problem sections: diamonds, turnouts, bridge approaches, tunnel floors and highway crossings. There are two forms:

Underlayment replaces the subballast only. The ballast stays above it and can still be tamped.
Full depth replaces both the ballast and the subballast.

Typical underlayment.

A 4–8 in. asphalt mat with 5–10 in. of ballast between the tie bottom and the mat. Thickness depends on axle load, tonnage and subgrade stiffness; see NAPA IS-99 for design.
The mix is a dense-graded highway base with slightly more binder and fines, 1–3% air voids and a maximum aggregate size of 1–1½ in.
The ballast cover protects the asphalt from oxidation and temperature extremes.

Full-depth limitations. Full depth reduces lateral and longitudinal resistance and resilience. It cannot be tamped, which is a serious limitation if the subgrade moves. It is also hard to rebuild. Use it only where those losses are acceptable and geometry adjustment is essentially not needed.

Drainage. Asphalt blocks upward drainage from the subgrade. Where seepage is expected, provide a granular drainage layer beneath the mat. That layer also serves as a construction platform.

After Selig & Waters (1994), ch. 11. Full reference under Supplemental Sources.

3.9 Utility and Pipeline Crossings

Pipelines and utilities crossing beneath the track are third-party foundation elements. AREMA MRE Chapter 1 and client utility-crossing standards govern casing, depth of cover, installation method and inspection. From a ground hazard standpoint:

Leaking water and sewer lines erode fines and create voids. This is a recognized cause of sudden subsidence beneath track (Chapter 1, Section 1.3).
Poorly compacted open-cut trench backfill settles and produces a local geometry defect.
Trenchless installations (bore and jack, horizontal directional drilling, microtunneling) can cause heave or loss of ground if face pressure, overcut or grouting are not controlled. Require settlement monitoring points on the track during installation (Chapter 15).

A fiber-optic installation done wrong

A fiber-optic cable installation along a shoulder destabilized the slope by breaking nearly every rule at once. Crews excavated at the toe, overloaded the shoulder with spoil, and then left the ditch unmaintained because the cable running through it discouraged cleaning. Support later placed only at the shoulder, with none at the toe, made it worse. For any utility work in the right-of-way: do not excavate at the toe, do not overload the shoulder, keep ditches drained whatever runs through them, and do not let water seep into the shoulder or fill.

Figures · click to zoom

Figure 3-34. Fiber installation: toe excavation and shoulder overload both apply. — **Figure 3-34.** Fiber installation: toe excavation and shoulder overload both apply.

Figure 3-35. A problematic fiber-optic cable slot along the shoulder. — **Figure 3-35.** A problematic fiber-optic cable slot along the shoulder.

Figure 3-36. Riprap at the shoulder with none toward the toe: the wrong way around. — **Figure 3-36.** Riprap at the shoulder with none toward the toe: the wrong way around.

After Geotechnology for Railroaders Program, Module 5, Module 6.1; Module 5 §10; Module 6.1 §03. Full reference under Supplemental Sources.

Case study: 3.16.4 As-constructed drawings that were wrong (Look 2023), in §3.16 Case Studies.

3.10 Investigation Matrix for Foundation Elements

Use this matrix together with the investigation planning criteria in Chapter 5 and the laboratory program in Chapter 10.

Element	Field Exploration	Laboratory Tests (ASTM)	Deliverable Focus
Embankment body	Borings through crest and shoulders (SPT D1586), CPT D5778, test pits; GPR for bed layering (ASTM D6432 guide)	D2487 / D2488 classification, D4318 limits, D6913 gradation, D2216 moisture	Fill composition, ballast pockets, water in fill, stability
Cut floor and slopes	Borings or test pits at formation, groundwater observation wells	D4318, D4546 swell, D4829 expansion index, D3080 direct shear	Softening potential, seepage, slope stability
Foundation soils	Borings to competent layer; undisturbed sampling (D1587); field vane D2573; CPTu	D2435 consolidation, D2166 / D4767 strength, D2974 organic content, D5333 collapse	Settlement, deep stability, improvement need
Culverts	Borings at inlet, outlet and centerline; camera inspection of barrel; void detection (GPR or probing)	Classification; pH, resistivity, chloride and sulfate (G51, G187, D4327)	Bearing, backfill condition, corrosion, piping
Bridge approaches and structures	Borings at each substructure unit; rock coring (D7012 for strength); scour-related sampling	Strength, consolidation, rock strength and RQD	Foundation type and capacity, approach settlement, scour
Utility crossings	Locate and verify depth; borings along the alignment; monitoring points	Classification, moisture, groundwater	Settlement and heave limits, monitoring plan

3.11 Performance Criteria

Foundation elements are ultimately judged by their effect on track geometry. FRA 49 CFR 213.63 sets surface (profile, cross-level and warp) limits by track class. A foundation that causes geometry to exceed those limits between normal maintenance cycles is not performing, whatever its calculated factor of safety.

Criterion	Common Target (confirm with client)	Where Evaluated
Static slope stability, embankments and cuts	FS ≥ 1.5 with design train surcharge	Chapter 13
Subgrade and foundation bearing under cyclic load	FS ≥ 3.0 on undrained bearing (Chapter 2 method)	Chapters 2, 11
Post-construction settlement of new fill	Limited so that geometry remains correctable by routine surfacing; client-specific	Chapter 12
Differential settlement at transitions	Minimized; evaluate over the transition length rather than as a single number	Chapters 2, 12
Culvert and structure foundations	Per AREMA MRE Ch. 8 and the client's structure standards	Chapter 11

European track-geometry limits for comparison

European practice sets three tiers of geometry limits: intervention values that trigger scheduled maintenance, acceptance values allowed to remain after maintenance, and emergency values that require an immediate speed restriction. FRA Track Safety Standards (49 CFR Part 213) and client standards govern in North America; the European figures are given to show how tight track-support performance has to be.

Table 3-3. Emergency values of track defects and maximum permitted speed
Longitudinal defect LD (in.)	Horizontal defect HD (in.)	Maximum speed (mph)
≥ 0.71	≥ 0.59	87
≥ 0.79	≥ 0.67	75
≥ 0.91	≥ 0.75	62
≥ 1	≥ 0.87	50
≥ 1.1	≥ 1	37
≥ 1.2	≥ 1.1	25

33-ft measurement base. LD = surface (profile), HD = alignment.

Table 3-4. Limits on track-gauge variation, European interoperability specifications
Speed (mph)	Minimum gauge deviation (in.)	Maximum gauge deviation (in.)
V ≤ 50	−0.35	+1.4
50 < V ≤ 75	−0.35	+1.4
75 < V ≤ 99	−0.31	+1.4
99 < V ≤ 143	−0.28	+1.1
V > 143	−0.2	+1.1

The European interoperability specifications also limit twist to 0.084 in./ft for speeds up to 124 mph and 0.06 in./ft above, adjusted for measurement base length (Figure 3-12), and limit the difference between nominal and actual cant to ±0.79 in..

Figures · click to zoom

Figure 3-11. Intervention limits for longitudinal, transverse and horizontal defects, twist and gauge, by speed, for DB, SNCF, NS, BR and UIC. Profillidis Table 16.2. — **Figure 3-11.** Intervention limits for longitudinal, transverse and horizontal defects, twist and gauge, by speed, for DB, SNCF, NS, BR and UIC. Profillidis Table 16.2.

Figure 3-12. Twist limit vs. measurement base length, conventional and high-speed track. Profillidis Fig. 16.7. — **Figure 3-12.** Twist limit vs. measurement base length, conventional and high-speed track. Profillidis Fig. 16.7.

After Profillidis (2014), §16.6, Tables 16.2, 16.4, 16.5, Fig. 16.7. Full reference under Supplemental Sources.

3.12 Practical Recommendations

Treat recurring geometry defects at the same location as a foundation problem until proven otherwise. Ask the client for geometry-car history and surfacing records before scoping the investigation.
Never assume a historic embankment is uniform. Space borings to capture variability, and use GPR or CPT to fill gaps between borings.
Walk the drainage path from the track to the outlet on every site visit. Blocked outlets are the cheapest foundation problem to fix.
Identify buried structures (old culverts, trestles, timber cribbing) from valuation maps and track charts before drilling. They explain many "unexplained" sinkholes.
At bridge approaches, evaluate the approach fill and the foundation beneath it as part of the bridge project, not as an afterthought.
Document each foundation element and its condition in the report. The ground hazard inventory (Chapter 1) is only as good as the site descriptions feeding it.

Project management · Kelly et al. (2004)

Whole-Life Costing of Remediation Alternatives

Kelly, Male and Graham hold that options generated in a value study should be compared on whole-life cost. Whole-life cost is the present value of all relevant costs over the investor’s time horizon. Relevant costs include investment (design, construction, fees), operation and maintenance, unplanned repair, planned component replacement, and residual or terminal values at the end of the horizon. Some costs are excluded: trivial amounts, sunk costs already spent, and costs that are the same for every option. Future amounts are discounted with

P = A / (1 + i)ⁿ

where A is the future cost, i the discount rate per period and n the number of periods. Inflation is handled with a real rate, i = (1 + bank rate)/(1 + inflation) − 1. In their example, a component costing £6,000 today that will fail in 10 years has a present value of £4,053 at 4 percent with no inflation. With 2 percent inflation the real rate is 1.96 percent and the present value is about £4,941, because the replacement will by then cost about £7,314.

For railroad work: Track foundation remedies often trade first cost against recurring maintenance. Examples are lining a deteriorated culvert versus replacing it, flattening a cut slope versus installing rockfall mesh and patrols, and full subgrade rebuild versus periodic undercutting and surfacing. List each alternative’s recurring items over the railroad’s planning horizon: surfacing cycles, ditch cleaning, drain flushing, mesh inspection and repair, slow-order delay costs where the client supplies them, and expected replacement dates. Discount them using the rate the railroad’s finance group specifies, and present the present-value totals with a sensitivity check on rate and cycle length. Exclude money already spent on earlier repairs at the site. Sunk-cost thinking is a common reason a chronic location keeps receiving the same short-lived fix.

Source: Kelly, J., Male, S. and Graham, D. (2004). Value Management of Construction Projects. Oxford: Blackwell Science, pp. 311–314. Summarized and paraphrased. Full reference under Supplemental Sources.

3.13 Foundation Elements Checklist

Plain text version

[ ] Valuation maps, track charts and as-builts reviewed for fill history, buried structures and culverts
[ ] Geometry-car history and maintenance (surfacing, undercutting) records obtained
[ ] Each foundation element present at the site identified and described (Table in Section 3.1)
[ ] Borings extend through embankment and weak foundation layers to competent material
[ ] Groundwater levels measured and, where relevant, monitored over time
[ ] Drainage path traced to outlet; ditch flow line elevation compared with subballast base
[ ] Culverts inspected for joint separation, backfill loss and corrosion; corrosivity samples tested
[ ] Train surcharge and influence zone applied to all excavations and structures near track
[ ] Scour evaluated at waterway structures
[ ] Stability, bearing and settlement checked against client criteria
[ ] Findings entered into the ground hazard inventory with risk rating (Chapter 1)

3.14 Cross-Reference Table

Topic	See Also
Ground hazard classification and risk rating	Chapter 1
Ballast, subballast and subgrade design; stress at top of subgrade	Chapter 2
Failure mechanisms and triggers	Chapter 4
Investigation planning and boring depths	Chapter 5
Groundwater characterization	Chapter 9
Laboratory testing program	Chapter 10
Bearing capacity and deep foundations	Chapter 11
Settlement analysis	Chapter 12
Slope stability analysis	Chapter 13
Construction monitoring and compaction testing	Chapter 15
Retaining structures and train surcharge	Chapter 16
Ground improvement methods and technology library	Chapter 17; Section 17.16

3.15 Practice Questions

Q1. A main line shows a cross-level defect that returns within months after each surfacing at the same location over a 60-year-old fill. List three foundation elements you would investigate, and the first piece of client data you would request.

Answer:

Investigate the embankment body (ballast pockets, uncompacted fill), the foundation soils (soft or organic layers), and drainage elements (blocked ditch or a culvert losing backfill). Request geometry-car history and surfacing records to confirm the defect's extent and recurrence rate.

Q2. Why should new fill be benched into an existing embankment when a second track is added?

Answer:

Benching removes the smooth, often vegetated and weak surface of the old slope and keys the new fill into it. That prevents a continuous weak plane along which the widening could slide or separate.

Q3. A corrugated metal culvert under 25 ft of fill shows track settlement directly above it. Name two mechanisms that could cause this, and the tests or inspections that would distinguish them.

Answer:

(a) Backfill loss through corrosion holes or separated joints. Distinguish it with a camera inspection of the barrel and corrosivity testing (pH, resistivity, chlorides and sulfates). (b) Poorly compacted haunch backfill or foundation settlement. Distinguish it with borings beside the barrel and a check of the invert profile for sag.

Q4. What FRA regulation establishes the drainage maintenance requirement relevant to foundation elements, and what does it require?

Answer:

49 CFR 213.33 requires that drainage and other water-carrying facilities under or immediately adjacent to the roadbed be maintained and kept free of obstruction to accommodate expected water flow.

References

AREMA, Manual for Railway Engineering, current edition — Chapter 1, Roadway and Ballast; Chapter 8, Concrete Structures and Foundations.
Federal Railroad Administration, Track Safety Standards, 49 CFR Part 213 (incl. §213.33 Drainage; §213.63 Track surface).
Federal Railroad Administration, Bridge Safety Standards, 49 CFR Part 237.
BNSF Railway and Union Pacific Railroad, Guidelines for Temporary Shoring (current edition as issued by the railroads).
FHWA, HEC-18, Evaluating Scour at Bridges, current edition.
Selig, E.T. and Waters, J.M. (1994). Track Geotechnology and Substructure Management. Thomas Telford.
Li, D., Hyslip, J., Sussmann, T. and Chrismer, S. (2016). Railway Geotechnics. CRC Press.
ASTM D698, D1557, D1556, D1586, D1587, D2166, D2216, D2435, D2487, D2488, D2573, D2974, D3080, D4318, D4327, D4546, D4648, D4767, D4829, D5333, D5778, D6429, D6432, D6913, D6938, D7012, G51, G187 — current editions.

3.16 Case Studies

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 S5 under Supplemental Sources.

Case Study 3.16.1 — Thirty-year-old expansive clay embankments: where the moisture really settles

Case study · Look (2023) §10.15

Related section: 3.2 Embankments

Earthworks for the 4.4 mi Cooroy Bypass, north of Brisbane, were in highly reactive clays (weighted plasticity index 2200–3200). In 1990, time domain reflectometry (TDR) probes were installed in two trenches about 4.3 mi apart in existing 30-year-old embankments of the same clay, built at MDD and OMC to the specification of their day (Look and Reeves, 1992). Findings:

The seasonally active zone reached below 5.9 ft, not the 5 ft assumed. Swell tests showed that movement at that depth is suppressed by overburden, so the “stable zone” is a no-movement zone even with minor moisture change.
In service, stable-zone moisture was 1.2–1.3 times standard OMC, and density had dropped to 92%–94% of standard MDD although the fills were built at MDD. Clay placed drier will wet up, swell and lose strength.
Peak moisture tracked the 50-day moving average of rainfall; the low point of the drying phase tracked the 300-day average.

Recommended design: subgrade CBR 3%; target density 92%–94% of standard MDD with characteristic limits of 90%–96%; target moisture 110%–130% of standard OMC. A three-section trial embankment placed at 1.0, 1.2 and “as wet as possible” times OMC converged to a similar moisture within nine months, again 110%–130% of OMC, confirming the approach. During construction, TDR probes and horizontal profile gauges set the curing period before the pavement went on.

Key lessons: zone the clay, with a 8.2 ft side zone matching the measured movement from the batter; specify an upper density limit so that compaction at or above 100% MDD fails for these clays where rainfall exceeds about 39 in a year; target the equilibrium moisture; and allow curing before final layers.

For railroad work: fat-clay embankments on many North American lines show the same seasonal movement, expressed as recurring surface and cross-level defects. When rebuilding or widening such fills, compacting the clay hard and dry is not conservative.

**Figure 3-LK1.** Moisture in the stable zone at about 6 ft depth in two trenches 4.3 mi apart, against daily rainfall (Look 1996). *Source: Look (2023), Figure 10.15-4, p. 363.*

**Figure 3-LK2.** Trial embankment: all three placement moistures converged to a similar moisture within about 9 months (Look 1996). *Source: Look (2023), Figure 10.15-5, p. 365.*

Source: Look, B. G. (2023). Earthworks: Theory to Practice – Design and Construction. CRC Press/Balkema. Case study §10.15, pp. 360–366. Summarized and paraphrased; figures reproduced for internal reference. Full reference under Supplemental Sources.

Case Study 3.16.2 — Four materials in one 1,300 ft cutting

Case study · Look (2023) §8.9

Related section: 3.3 Cut Sections and Cut Slopes

A cutting of about 392,000 yd³ was to supply embankment fill. Over 1,310 ft its cross sections showed four material types, defined so estimators could plan equipment:

residual soil and extremely weathered rock: easy digging
highly weathered rock: hard digging; seismic velocity typically 3,600–5,250 ft/s
moderately to slightly weathered rock: easy to hard ripping; velocities up to 7,900 ft/s
a microdiorite dike and fresh volcanolithic sandstone at depth: hard ripping with a D11, point load index typically above 580 psi

The preliminary borings no longer lined up with the final alignment. Transport bulking ranged from 20% to 70%, and from 0% to 30% after compaction. Types 1–3 behaved as “soil” once compacted, controlled by density testing. Type 4 was suitable as rock fill and could have been crushed on site for drainage or pavement material, but a nearby quarry made that uneconomic.

For railroad work: the same breakdown suits a rail cut balanced into fill. Re-check the borings against the final alignment and have each material type carry its own excavation method, bulking factor and end use. The tender view of this same cut is in the next box.

**Figure 3-LK3.** Cut 3 cross sections over 1,300 ft; preliminary borings are offset from the final alignment and a microdiorite dike crosses at Ch. 75,550. *Source: Look (2023), Figure 8.9-1, p. 269.*

Source: Look, B. G. (2023). Earthworks: Theory to Practice – Design and Construction. CRC Press/Balkema. Case study §8.9, pp. 268–270. Summarized and paraphrased; figures reproduced for internal reference. Full reference under Supplemental Sources.

Case Study 3.16.3 — Earthworks tender assessment for a balanced cut-to-fill job

Case study · Look (2023) §14.14

Related section: 3.3 Cut Sections and Cut Slopes

Look follows the same Cut 3 into the tender stage of a “balanced” job with six cuttings feeding the embankments. The quantities were bank volumes. Each cut had its own rock type, quantity and quality, and had to be assessed on its own merit from source to destination. For each material type the contractor set out depth, class and the best equipment, from scrapers in the top 6.6 ft to a D9 for rock fill between about 16 and 43 ft.

The assessment ran in three steps: excavation; bulking and transport; and placement. Transport bulking and post-compaction bulking have to be treated separately. The balance quantity for each source, its location, matching sources to nearby fills, and zoning of the embankments all formed part of the assessment.

For railroad work: a rail grading contract that is “balanced” on bank volumes can still come up short once bulking, breakdown and material zoning are counted. Ask bidders to show the assessment by source and material type.

**Figure 3-LK4.** Tender assessment for Cut 3: material types, approximate depths, class and suitable equipment (scraper, D7, D9). *Source: Look (2023), Figure 14.14-2, p. 553.*

Source: Look, B. G. (2023). Earthworks: Theory to Practice – Design and Construction. CRC Press/Balkema. Case study §14.14, pp. 550–552. Summarized and paraphrased; figures reproduced for internal reference. Full reference under Supplemental Sources.

Case Study 3.16.4 — As-constructed drawings that were wrong

Case study · Look (2023) §3.7

Related section: 3.9 Utility and Pipeline Crossings

As-constructed drawings are assembled at the end of a project, from second- and third-hand notes, when funds are nearly spent. Look gives three examples from his early site-investigation career:

Bridge boring. The water authority’s plans and a joint inspection placed a sewer main on the east side of the bridge. A farmer who had worked the adjacent land for 40 years stopped as the rig set up on the west side and pointed out two buried access chamber covers: the sewer bowed to the west and back. His knowledge was better than the asset owner’s.
Empty industrial lot. The owner’s plans showed all services south of his building, and he was sure one 3 in boring could go anywhere in the 330 ft × 197 ft lot. Within 15 minutes the rig hit a high-pressure water main. On area alone the odds were under 0.001%; repairs took half a day.
City site. Records checks, sign-off by every service authority and a locator produced an approved 49 ft service-free triangle. A pre-1950 gas main without a tracer band was not on the plans and could not be located. It was hit within 15 minutes; five blocks were evacuated and the city was closed to traffic for hours.

For railroad work: railroad corridors carry fiber, signal cable and pipelines installed by others over a century or more, often under old permits with poor location data. Treat records as a starting point, pothole or vacuum-excavate before drilling within the right-of-way, and ask track and signal forces what they know about the location.

Source: Look, B. G. (2023). Earthworks: Theory to Practice – Design and Construction. CRC Press/Balkema. Case study §3.7, pp. 90–92. Summarized and paraphrased; figures reproduced for internal reference. Full reference under Supplemental Sources.

Supplemental Sources

Sections on this page marked “After Profillidis (2014)” draw on the document below. Text is summarized or paraphrased; tables are transcribed or reproduced with their original source noted; figures and photos are reproduced from the cited document for internal reference use. Consult the original document before relying on a value for design. Blocks marked “After UFC 4-860-01FA (2004)” draw on item S3 and are rewritten for railroad clients. Blocks marked “After Selig & Waters (1994)” summarize item S4 for railroad clients; tables are condensed from the book and its figures are not reproduced. The Case Studies section summarizes case studies from item S5 by Burt G. Look; its figures are reproduced from the book for internal reference use, and the closing “For railroad work” note in each case study is added by the authors of this manual. Blocks tagged “Project management” summarize methods, examples and figures from the project-management, estimating and construction-management books listed here, applied to railroad work; the closing “For railroad work” note in each block is added by the authors of this manual.

Profillidis, V. A. (2014). Railway Management and Engineering, 4th ed. Farnham, UK: Ashgate. Chapters 7–9, 11–13, 16–17, 20 and 22 (track system, mechanical behavior of track, subgrade, sleepers, ballast, transverse effects, maintenance, slab track, electrification foundations, safety)
Benesch. Geotechnology for Railroaders Program, Modules 1–8 (two-day training course reference: risk management, geology and soils, ballast and sub ballast, drainage, slides, slope and subgrade stabilization, scour, rock falls, beaver dams, winter problems, spring thaw, extreme weather, earthquakes). Field photographs and figures from the course; slope stabilization process slides adapted from material by D. Colwell
U.S. Army Corps of Engineers, Naval Facilities Engineering Command and Air Force Civil Engineer Support Agency (2004). UFC 4-860-01FA, Railroad Design and Rehabilitation. Unified Facilities Criteria, 16 January 2004 (supersedes TI 850-02, 1 March 2000); approved for public release, distribution unlimited. Used on this page: ch. 9 (§4)
Selig, E. T. and Waters, J. M. (1994). Track Geotechnology and Substructure Management. London: Thomas Telford. Used on this page: chs. 11 and 13
Look, B. G. (2023). Earthworks: Theory to Practice – Design and Construction. Leiden: CRC Press/Balkema (Taylor & Francis). ISBN 978-1-032-10470-6. Case studies used on this page: §3.7, §8.9, §10.15, §14.14
Kelly, J., Male, S. and Graham, D. (2004). Value Management of Construction Projects. Oxford: Blackwell Science. Used on this page for project-management material

Author and Sources

Authors: Daniel P. Colwell and Mario Ruel. This part 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.

References

AREMA, Manual for Railway Engineering, current edition — Chapter 1, Roadway and Ballast; Chapter 8, Concrete Structures and Foundations.
Federal Railroad Administration, Track Safety Standards, 49 CFR Part 213 (incl. §213.33 Drainage; §213.63 Track surface).
Federal Railroad Administration, Bridge Safety Standards, 49 CFR Part 237.
BNSF Railway and Union Pacific Railroad, Guidelines for Temporary Shoring (current edition as issued by the railroads).
FHWA, HEC-18, Evaluating Scour at Bridges, current edition.
Selig, E.T. and Waters, J.M. (1994). Track Geotechnology and Substructure Management. Thomas Telford.
Li, D., Hyslip, J., Sussmann, T. and Chrismer, S. (2016). Railway Geotechnics. CRC Press.
ASTM D698, D1557, D1556, D1586, D1587, D2166, D2216, D2435, D2487, D2488, D2573, D2974, D3080, D4318, D4327, D4546, D4648, D4767, D4829, D5333, D5778, D6429, D6432, D6913, D6938, D7012, G51, G187 — current editions.

Case studies

Look, B. G. (2023). Earthworks: Theory to Practice – Design and Construction. CRC Press/Balkema. Case studies §3.7, §8.9, §10.15, §14.14

Project management sources

Kelly, J., Male, S. and Graham, D. (2004). Value Management of Construction Projects. Oxford: Blackwell Science