Question 3 of 4

Question 3 (25 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

(a) Defining risk; a risk-graded investigation model

In the context of a site investigation, risk is the combination of the likelihood that an unfavourable or unanticipated ground condition exists (or that the investigation fails to detect a condition that does exist) and the consequence of that condition to safety, cost, schedule or the environment if it materializes: Risk = Likelihood × Consequence. Both terms are always present — a highly variable ground condition beneath a low-consequence structure may still warrant only a modest investigation, while even a well-understood, uniform site can carry high risk if the structure itself is life-safety critical.

Risk is incorporated at the investigation stage by explicitly grading the scope of exploration to the assessed risk category, rather than applying one fixed program to every project — a model directly analogous to the geotechnical-category concept used in Eurocode 7 and reflected in CFEM's own risk-based investigation guidance:

• Low risk (simple, well-understood ground; low-consequence structure — e.g. a single-storey warehouse on known uniform soil): a limited desk study plus a small number of boreholes/test pits to standard depths, basic index/classification testing, minimal in-situ testing, and a standard-format report; investigation effort is proportioned to confirming the desk-study assumption rather than exhaustively characterizing the site.
• Moderate risk (typical building/infrastructure project on reasonably understood but variable ground): a standard investigation program — boreholes spaced and depth-scaled to the structure footprint and loads, routine in-situ testing (SPT/CPT), a representative laboratory strength/consolidation program, and a full engineering report with design parameters and recommendations; this is the industry-norm scope for most projects.
• High risk (complex or poorly known ground, high-consequence structure, or both — e.g. a dam, hospital, high-rise on variable/soft ground, or any site with a known geohazard): an enhanced program with denser and deeper exploration, specialist in-situ testing (pressuremeter, geophysics, in-situ stress measurement), possibly a staged/phased investigation where early results scope a further round, independent peer review of the interpretation, installed monitoring instrumentation, and, on the construction contract, a Geotechnical Baseline Report to formally allocate the residual risk that even an enhanced investigation cannot fully eliminate.

The category is assigned early (typically at the desk-study/reconnaissance stage) using the structure's consequence class, the site's known geological complexity, and any preliminary hazard flags, and can be escalated if field results reveal conditions worse than assumed — the model is a starting scope, not a fixed one.

(b) Ten priorities when investigation resources are limited

Because a fully comprehensive investigation is rarely affordable, the available budget should be prioritized toward what most directly controls life-safety, structural performance and downstream cost — roughly in descending order:

1. Confirm the presence or absence of critical geohazards (liquefiable soil, sensitive/quick clay, karst, active faulting, deep-seated slope instability) — these carry life-safety consequence and cannot be deferred to a later phase.
2. Establish the founding stratum and its depth/continuity across the structure footprint — the single most direct driver of foundation type selection and cost.
3. Establish groundwater conditions (level, seasonal range, artesian pressure) — governs dewatering, excavation stability and buoyancy design.
4. Obtain samples of adequate number and quality for the strength/consolidation parameters that govern the controlling design case (bearing capacity or settlement, whichever is more critical for the structure type).
5. Cover the structure footprint spatially with a minimum defensible number and spacing of exploration points before adding depth or redundant testing at any single point.
6. Identify contamination or environmental liability early — changes disposal cost and handling requirements and can halt construction if discovered mid-project.
7. Characterize seismic site response where required by code — affects the structural design load, not just the foundation.
8. Flag construction obstructions (boulders, buried structures, very dense/hard layers) that affect means, methods and cost even where they do not change the final design.
9. Document and communicate residual uncertainty explicitly wherever the budget stopped short of full characterization, so the client and design team make an informed decision rather than an implicitly optimistic one.
10. Preserve the ability to add exploration later (retained contingency budget, retained site access, a staged program) if construction reveals conditions outside what the limited investigation characterized.

(c) Correlating information between boreholes; interpretation risk

Individual boreholes are point samples; correlating between them is what converts a set of discrete logs into a usable three-dimensional ground model. The principal methods are: stratigraphic cross-sections, drawn between boreholes along and across the structure footprint or alignment, connecting like units by elevation and matching lithological/index-property signature; structure-contour and isopach mapping, contouring the elevation of a marker horizon (e.g. top of bedrock, top of a specific clay unit) or its thickness across the site to reveal dip, relief or an unexpected buried feature (channel, pinnacle); geostatistical interpolation (kriging or simpler inverse-distance methods), used where enough boreholes exist to estimate a property or horizon elevation between them with a quantified confidence/variance, rather than a purely visual best-fit line; and index/property correlation (grain-size distribution, Atterberg limits, SPT/CPT profile shape) used to group boreholes into the same inferred depositional unit even where a marker horizon is not obviously continuous. Increasingly, this correlation is done digitally in 3-D geotechnical modelling software, which lets the same borehole set be reviewed as sections along any alignment rather than only the few sections drawn by hand.

The risk in this interpretation is that correlation is inherently a projection between widely-spaced point data, and it is easy to draw a smooth, plausible connection between boreholes that in reality straddles a discontinuity the boreholes never intersected — a buried channel, a fault, a lens of soft or organic material, or an erosional unconformity — producing a ground model that looks more continuous and certain than the data actually support. This risk grows directly with borehole spacing and with the actual geological complexity of the site (a glacially-derived or fluvial deposit is far more heterogeneous than a uniform marine clay), and is mitigated by: using geological process knowledge (depositional environment from the desk study) to constrain which correlations are geologically plausible rather than purely fitting the numbers; explicitly flagging and testing alternative correlations at points of ambiguity rather than presenting only the most likely one; supplementing borehole correlation with continuous geophysical coverage between holes where the risk of missing a discontinuity is high; and stating the confidence level of the interpretation (and the corresponding residual risk) explicitly in the report rather than presenting an inferred ground model as a certainty.