# Example: Facility Siting and Construction Form Driven by Subsurface Conditions (Civil Engineering)

**Status:** Draft
**Phase:** The Bedrock Phase

## What This Example Demonstrates

Context record structure (OE-0003) applied to a construction siting and methodology decision where the governing constraints were not visible at the site surface and were discovered only through targeted investigation, illustrating how environmental conditions that are "unseen" — subsurface, hydrological, climatic — can override every other siting factor (OE-0004, OE-0007).

## The Observation

A regional logistics facility was planned for a 12-hectare greenfield site adjacent to a highway interchange in a semi-arid region. The site scored highest on every surface-level criterion: highway access, proximity to the target delivery radius, land cost, and zoning availability. A preliminary geotechnical investigation (two boreholes to 6m depth) confirmed competent bearing capacity for conventional spread footings. Construction began with conventional slab-on-grade foundations and a steel-framed structure.

During excavation for the stormwater detention basin at the northwest corner of the site, the excavation contractor encountered a layer of expansive clay (plasticity index of 52, well above the 35 threshold for high-expansiveness) beginning at 2.1m depth and extending to at least 8m — the limit of the excavation. This clay had not been encountered in the preliminary boreholes because both were located in the southeastern portion of the site, where the clay layer was absent. Subsequent investigation (six additional boreholes, two to 15m depth) revealed that the expansive clay underlay approximately 60% of the site in an irregular lobe, with thickness varying from 1.5m to over 9m.

The expansive clay meant that conventional slab-on-grade foundations — already under construction on the southeastern portion — would experience differential heave as the clay absorbed moisture from seasonal rainfall and landscape irrigation. Predicted heave ranged from 25mm to over 100mm across the affected area. A structure built on a slab resting on this clay would crack as different parts of the foundation moved at different rates.

## Engineering Translation

This case demonstrates a category of engineering constraint that is fundamentally different from the constraints typically documented in design specifications. Surface-level constraints — load, span, clearance, access — are visible and can be assessed during initial site evaluation. Subsurface and environmental constraints are invisible until specifically investigated, and the investigation must be targeted at the right locations and depths to find them. The preliminary geotechnical investigation satisfied the minimum standard for the project type (two boreholes), but those two boreholes happened to miss the expansive clay lobe entirely. The constraint was real, binding, and present from the beginning — but it was unknown because the observation had not been made at the right location.

The decision that follows is not about whether to address the clay (that is mandatory) but about how, and the "how" has consequences that cascade through the entire construction methodology, schedule, and cost. Each alternative carries a different combination of cost, timeline impact, and long-term risk, and selecting among them requires understanding not just the clay's properties but the project's tolerance for schedule delay and budget overruns. The engineering context record for this decision must preserve not just what was decided but what was unknown at the time of the initial site selection and how that unknown was discovered — because a future facility on a similar site needs to understand that minimum-standard geotechnical investigations carry a risk of missing subsurface conditions that are later found to be the binding constraint on the entire project.

## Context Record

| Field | Content |
|---|---|
| **Decision** | Adopt a post-tensioned slab-on-grade foundation with moisture barriers for the expansive clay zone and conventional spread footings with reinforced slab for the non-expansive zone, with a structural transition beam at the clay boundary. Continue steel-framed construction but modify the stormwater management system to prevent surface water infiltration into the clay zone. Relocate the detention basin 40m east to a non-expansive area identified in the supplemental investigation. |
| **Observation** | (1) Expansive clay encountered at 2.1m depth during detention basin excavation, with plasticity index of 52 (highly expansive per ASTM D4829). (2) Supplemental borehole investigation (8 boreholes, 2 to 15m depth) delineated the clay lobe covering approximately 60% of the site with thickness varying from 1.5m to over 9m. (3) Laboratory swell testing (ASTM D4546) on undisturbed samples showed free swell pressures of 180–340 kPa and vertical swell strains of 3.2–6.8% at the anticipated foundation bearing stress. (4) Seasonal moisture monitoring probes installed over 6 months showed moisture content variation from 18% (dry season) to 31% (wet season) in the clay zone at 2m depth, confirming active seasonal heave potential. (5) The preliminary boreholes (both in the southeastern 40% of the site) were in areas free of expansive clay — the clay was not missed due to insufficient depth but due to insufficient spatial coverage. |
| **Alternatives** | (A) Deep foundations (driven piles or drilled shafts) to bypass the clay entirely — rejected: pile installation to competent bearing below the clay (estimated 12–18m depth) would cost approximately $2.8M for the affected area, compared to $900K for the post-tensioned slab solution. Additionally, the pile caps and grade beams would require significant excavation in the clay zone during construction, exposing the clay to moisture and triggering heave during the construction period itself. (B) Excavate and replace the clay with engineered fill — rejected: the clay extends to over 9m in the thickest areas, requiring excavation and replacement of approximately 180,000 cubic meters of material. The nearest suitable fill source is 35km away, making the haulage cost alone approximately $4.1M. The construction timeline impact (estimated 4-month delay for excavation, hauling, and compaction) would breach the occupancy deadline. (C) Chemical stabilization of the clay in situ using lime treatment — rejected: laboratory testing showed that lime treatment reduced the plasticity index from 52 to approximately 28 (moderately expansive), which was insufficient to eliminate differential heave risk. Achieving non-expansive results would require lime treatment to depths of 4–5m, which is beyond the practical reach of standard mix-in-place methods and would require pressure-injected lime slurry at significantly higher cost than the post-tensioned slab. (D) Relocate the facility to an alternative site — rejected: the alternative sites scored significantly lower on access and proximity criteria, and the land acquisition process would add 8–12 months to the project timeline. The cost of relocating (land premium, design restart, permitting) was estimated at $3.5M plus the sunk cost of the partially completed foundations. (E) Post-tensioned slab with moisture barriers and modified construction methodology — selected. |
| **Constraints** | The facility must achieve occupancy by the contractual deadline (14 months from notice to proceed, now 10 months remaining due to the investigation delay). The foundation solution must control differential heave to less than 15mm across any 12m span to prevent structural distress in the steel frame and rack systems. Construction in the clay zone must minimize moisture intrusion during the construction period — the clay cannot be allowed to swell before the building mass and slab are in place. The stormwater management system must prevent concentrated surface water from infiltrating the clay zone for the 50-year design life. Total foundation cost increase (including the investigation, redesign, and modified construction) must remain below $1.5M to preserve project economics. |
| **Reasoning** | The post-tensioned slab solution works by placing the slab in permanent compression, which counteracts the tensile stresses that expansive clay heave would otherwise induce in the foundation. The slab does not prevent the clay from swelling — it prevents the swelling from damaging the structure by resisting the upward force with pre-compression. The moisture barrier (horizontal polyethylene membrane beneath the slab with vertical barriers extending 1.5m depth around the perimeter) reduces the moisture variation that drives seasonal heave, cutting the predicted heave from 25–100mm to 8–18mm — within the 15mm differential limit for the post-tensioned slab's capacity. The structural transition beam at the clay boundary accommodates the differential movement between the expansive and non-expansive zones without transferring stress from one zone to the other. The solution costs $900K (including moisture barriers, transition beam, and supplemental investigation) against the $2.8M deep foundation and $4.1M excavation-and-replace alternatives. It preserves the schedule because post-tensioned slab construction is a well-understood process that the contractor can execute without specialized equipment or extended lead times. The trade-off is that the post-tensioned slab requires careful attention to tendon placement and stressing sequence, and the moisture barriers must remain intact for the building's life — any breach of the moisture barrier during future utility work or landscaping changes could reactivate the heave mechanism. |
| **Verification** | (1) Post-construction level survey monitoring over 24 months (quarterly surveys at 48 monitoring points) showed maximum differential heave of 11mm across a 12m span, within the 15mm design limit. Seasonal heave amplitude averaged 6mm, consistent with the moisture barrier's predicted 70% reduction in moisture variation. (2) Tendon elongation measurements during post-tensioning confirmed that all tendons achieved design force within 3% tolerance. (3) Visual inspection at 12 and 24 months showed no structural cracking in the slab, transition beam, or steel frame connections in the clay zone. (4) Moisture probe readings at 2m depth showed seasonal variation reduced from 13 percentage points (pre-construction) to 4 percentage points (post-construction with moisture barriers), confirming the barrier's effectiveness. (5) Relocated detention basin excavation confirmed competent non-expansive soils at the new location, validated by two confirmation boreholes. |
| **Lineage** | Builds on the initial site selection and geotechnical assessment (CR-CIV-2022-001, "Distribution center site selection — Highway 41 interchange"). That record documented the preliminary two-borehole investigation and the selection of conventional spread footings based on the observed bearing capacity. The present record extends that lineage by documenting the discovery that the preliminary investigation had insufficient spatial coverage, the conditions that were missed, and the reasoning behind the revised foundation system. The structural design of the steel frame and the building layout from CR-CIV-2022-001 are inherited without modification — only the foundation and stormwater systems changed. |
| **Assumptions** | The moisture barriers will remain intact for the 50-year design life without degradation from soil chemistry, root penetration, or construction activity. The post-tensioned tendons will not experience corrosion within the slab (the tendons are greased and sheathed, but long-term durability in expansive clay environments has a shorter track record than in non-expansive soils). The clay lobe's boundaries are accurately delineated by the 8 supplemental boreholes — the actual boundary may be more complex than the interpolated surface. Future construction on the site (utility trenches, landscape grading) will respect the moisture barrier perimeter and will not create new moisture infiltration paths into the clay zone. The seasonal moisture variation observed during the 6-month monitoring period is representative of long-term behavior — an unusually wet multi-year period could produce greater heave than predicted. |
| **Open Questions** | What is the minimum borehole density required to reliably detect subsurface conditions of this type on sites of similar size? Should the project's geotechnical investigation standard be revised for future sites to require boreholes at a maximum spacing rather than a minimum count? At what point does the cumulative cost of post-tensioned slab maintenance (barrier inspection, tendon monitoring) exceed the lifecycle cost of deep foundations that bypass the clay entirely? |

## Why This Is Different From the Bridge Example

The bridge example (bridge-survey.md) demonstrates a context record where all relevant conditions were known at the time of the decision. This example demonstrates a context record where the binding constraint was unknown at the time of the initial decision and was discovered during construction. The context record must therefore preserve not only the final decision but the sequence of discovery: what was assumed, what was observed that contradicted the assumption, and how the decision was revised in response. This is a more complex form of engineering reasoning — it includes a correction — and it is precisely the form of reasoning that is most frequently lost when only the final design is documented. A future engineer who sees only the post-tensioned slab specification without the context record cannot reconstruct why that specification exists, what it replaced, or what conditions would make the original spread footing design correct for a different site.

## Self-Fading Assessment

This example builds a bridge from the abstract principle that "engineering decisions are shaped by constraints" to a concrete case where the binding constraint was invisible at the site surface and was discovered only after construction had begun. The reader has crossed this bridge when they recognize that the reliability of any engineering decision is limited by the completeness of the observations on which it is based — and that a context record must preserve not only the final decision but the sequence of discovery that led to it, including what was initially unknown. When that recognition is stable and the reader can identify the same pattern in their own work (decisions revised when previously unknown conditions are discovered), this example has served its purpose and can be set aside.