Question 2 of 10: Seismic Refraction Survey Design, Execution and Interpretation

Question 2: Seismic Refraction Survey Design, Execution and Interpretation (16.66% of paper)

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.

Seismic refraction exploits the fact that a seismic wave incident at the critical angle on a boundary separating a slow layer over a faster layer travels along that boundary as a head wave, re-radiating energy back to the surface that, beyond a crossover distance, arrives before the direct wave.

Design. The target requires a positive velocity contrast with depth (velocity increasing downward, the "blind-zone" and hidden-layer limitations of the method are both consequences of this requirement). Spread length must be several times the target depth — a common rule of thumb is a spread length of at least 3–5× the depth of interest, since the crossover distance grows with refractor depth and with the velocity contrast. Geophone spacing is chosen to resolve the shallowest layer of interest (typically 2–5 m for near-surface geotechnical work) and shot points are placed at both ends of the spread (a "reversed" shot pair) plus, for multi-layer or dipping-interface work, one or more offset/interior shots to resolve velocity structure that a single forward shot cannot.

Execution. A source (sledgehammer/plate for shallow work, explosives or a weight-drop for deeper targets) generates the wave at each shot location; a string of geophones records first-arrival travel times along the spread. Reciprocal shooting (forward and reverse shots over the same spread) is essential because it lets the interpreter detect and correct for a dipping refractor, which a single-direction shot cannot distinguish from a velocity change.

Interpretation. First-break travel times are picked and plotted on a time–distance ($t$–$x$) graph. Each layer produces a straight-line segment whose slope is the reciprocal apparent velocity ($1/V_n$) and whose crossover/intercept locations feed the depth calculation. For a simple two-layer, flat-lying case, the reciprocal-slope method gives

$$h=\dfrac{x_{cross}}{2}\sqrt{\dfrac{V_2-V_1}{V_2+V_1}}\qquad\text{or equivalently}\qquad h=\dfrac{t_i V_1}{2\cos\theta_c},\ \ \theta_c=\sin^{-1}(V_1/V_2)$$

using either the crossover distance $x_{cross}$ or the intercept time $t_i$. For dipping layers or more than two layers, reciprocal (forward+reverse) shooting and the generalized reciprocal method (GRM) or delay-time methods are used to resolve refractor depth and dip independently at each geophone, and modern surveys increasingly use full tomographic travel-time inversion rather than the layered slope-intercept method alone.

Example application. Refraction is routinely used to map depth-to-bedrock beneath a proposed building or dam foundation, since overburden (till, alluvium) is reliably slower ($V_P\approx$300–1500 m/s) than sound bedrock ($V_P\approx$3000–6000 m/s) — a strong, positive, and usually sharp velocity contrast that is exactly the condition refraction needs. The same survey often also detects a shallow weathered/fractured bedrock zone as an intermediate-velocity layer, directly informing excavatability and blast design.

Would other methods work? Electrical resistivity would likely also succeed here (rock is typically far more resistive than saturated overburden) and is a useful cross-check, but resistivity is more sensitive to pore-water salinity/clay content and gives poorer depth resolution to a sharp, planar interface than refraction's travel-time geometry. Gravity would be far less effective — the density contrast between till and competent bedrock is often small (both roughly 2.0–2.7 g/cm³) compared with the strong velocity contrast, so a gravity survey over the same target would likely produce an ambiguous or negligible anomaly. GPR would fail in a conductive clay-rich till (rapid signal attenuation) even though it can be excellent in resistive sand/gravel. Refraction's strong positive velocity contrast at the till/bedrock boundary is therefore the most diagnostic and reliable of the available methods for this specific problem.