Cracking of a Hillside Retaining Wall Above an Occupied Hostel
A three-storey reinforced concrete hostel with 900 m² of floor area was completed in 2019 on a cut slope at Peradeniya. Uphill of the building, a 4 m high rubble masonry retaining wall with a reinforced concrete capping beam holds back a laterite slope. During a three-day monsoon event that recorded 350 mm of rain, the caretaker reports a 15 mm wide, 2.5 m long vertical crack in the wall, a new 8 mm crack across the ground-floor slab of the adjoining annexe, and a diagonal crack at the head of a ground-floor column. The only two visible weep holes are blocked with clay. The site's surface drains discharge directly onto the top of the wall, and a 150 mm water main serving the campus runs 3 m behind it. The soil profile is 6 to 9 m of residual laterite over weathered gneiss, and no groundwater was struck in the original boreholes to 8 m. The client wants a quotation for crack stitching and repainting before the next monsoon.
1. State the most probable mechanism causing the damage, and identify which observations support it and which do not.
Read the worked answer
The most probable mechanism is a build-up of water pressure behind the wall, not a failure of the masonry as a material.
Supporting evidence: the timing follows 350 mm of rain in three days; the weep holes are blocked, so the wall has no drainage path; surface drains discharge onto the wall, concentrating water exactly where it is least wanted; the 150 mm main behind the wall is a plausible leak source; and the movement has reached the annexe slab and column, which means the failure surface extends beneath the building rather than being confined to the wall.
The mechanism is that saturated laterite backfill loses suction and gains pore pressure, so the lateral thrust rises well above what a 4 m wall is normally designed for. Water at the toe also reduces passive resistance, and laterite softens on wetting.
The apparently contradictory evidence is the absence of groundwater to 8 m in the original boreholes. That does not weaken the diagnosis: it rules out a regional water table, but not a perched water body or a leak from the main, both of which are local and would not have appeared in holes drilled before the leak developed. The absence of groundwater also means the wall was probably designed without any water pressure at all, which makes a drainage failure sufficient on its own to explain the cracking.
2. Set out the immediate actions and the investigation programme you would require before any permanent works are designed.
Read the worked answer
Immediate actions, within 24 hours and in writing:
- Restrict or vacate the annexe ground floor and the zone below the wall; record the decision in a dated site instruction.
- Notify the client in writing of the risk, and inform the local authority, because the building may be unsafe for occupation.
- Isolate and pressure-test the 150 mm main; if it is leaking, shut it and provide a temporary supply.
- Cut off and divert the surface drains away from the wall crest, and rod out or add weep holes with granular filters to relieve water now.
- Install tell-tales and crack gauges, photograph everything to a dated baseline, and start weekly readings.
Investigation before design:
- Topographic and drainage survey of the slope, wall and building.
- Trial pits behind the wall to establish the backfill material, its permeability and the water level at the time of investigation.
- Boreholes with SPT through the laterite to gneiss, plus an inclinometer in the slope to detect a deep slip surface.
- A chartered geotechnical and structural engineer's stability analysis of the wall and the annexe foundations.
- Retrieval of the original wall design, approvals and as-built drawings.
The sequence matters: relieving water and establishing monitoring comes before drilling, and drilling comes before any remedial design.
3. Explain why crack stitching and repainting is not an acceptable response, and outline the remedial options available.
Read the worked answer
Stitching closes a gap that is the visible symptom of a displacement. The wall has moved because it was pushed; unless the water that pushed it is removed, the movement continues and the stitches become the next thing to crack. Repainting is worse than useless: it destroys the crack-monitoring record, conceals a developing failure from the occupants, and transfers liability to the architect who endorsed it.
Remedial options, in the order they should be considered:
- Water control - a cut-off drain at the crest, a geocomposite drainage layer against the back of the wall, weep holes at roughly 1.5 m centres with graded granular filters and geotextile, and permanent repair or rerouting of the main.
- Backfill replacement - remove the saturated laterite behind the wall and replace it with free-draining granular material, which cuts the thrust permanently rather than resisting it.
- Structural strengthening - if the stability analysis shows the wall is inadequate even when drained, options are soil nails or ground anchors through the wall into the gneiss, a mass concrete counterfort or gravity wall in front, or partial reconstruction of the failed section.
- Building remediation - underpinning or a root-barrier solution for the annexe if its foundations have been affected, designed only after the wall is stable.
- Monitoring and maintenance - a written regime of crack readings, drain clearing and inspection tied to the monsoon, funded by the operator.
Separately, advise the client that remedial cost is likely to be an order of magnitude above the stitching quote, and that delay into the next monsoon raises it further.