Question 1 of 10: Planning a Geophysical Survey

Question 1: Planning a Geophysical Survey (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.

Planning any geophysical survey is a sequence of linked choices, each of which constrains the ones that follow, and every choice ultimately traces back to one question: which physical property contrasts with the target, and by how much?

Method selection. Once the diagnostic property contrast is identified (density, susceptibility, resistivity, velocity, etc.), the choice of method follows directly — there is little point running a method the target does not contrast in. Platform (airborne, ground, borehole, marine) trades cost and coverage rate against resolution and depth control: airborne surveys are fast and cheap per unit area but coarser and farther from the target; ground surveys are slower but allow tighter station spacing and direct ground coupling; borehole surveys give the closest possible approach to a target already known to exist near a hole. Station and line spacing must be small enough, relative to the target's size and depth, to avoid spatial aliasing — as a rule of thumb the station interval should not exceed about half the target's along-strike dimension or roughly the target's depth, whichever is smaller, and line spacing should be tight enough that at least two or three lines cross any target of interest. Line orientation should run perpendicular to the expected geological strike so that a narrow, elongate target is crossed by as many lines as possible rather than run along its length by one line alone. Frequency/wavelength or the electrode/loop geometry sets depth of investigation directly (e.g. skin depth in EM, or spread length in resistivity) and must be matched to the target's expected depth, not simply maximized. Sensor sensitivity and sampling rate must resolve the expected anomaly amplitude above instrument and environmental noise. Finally, QC design — base-station repeats, tie lines, diurnal/drift corrections, and a pre-survey test line over a known feature — must be budgeted from the outset, not added afterward.

Worked example. Consider a ground electromagnetic conductivity survey planned to map the lateral extent of a landfill leachate plume expected at 5–15 m depth in a sand/silt aquifer, where the leachate is markedly more conductive than the surrounding unsaturated sand. A frequency-domain instrument (e.g. an EM31/EM34-class ground conductivity meter) is chosen because it gives fast, continuous apparent-conductivity readings without ground contact, appropriate parameters being: station spacing 5 m along each line (small relative to the plume's expected width of tens of metres, and comfortably below the instrument's own footprint), line spacing 10 m, lines oriented perpendicular to the inferred groundwater-flow/plume-migration direction (so that every line crosses the plume rather than running parallel to its axis), a vertical-dipole coil configuration and 10 m intercoil spacing chosen to give an effective depth of investigation of about 15 m (matching the target depth), and a base-line repeat measured every 20 stations to monitor instrument drift.

Consequences of inappropriate choices. A station spacing of, say, 20 m on this survey would under-sample the plume: if the true plume were narrower than 20 m at some point, it could be entirely missed between stations (spatial aliasing), or its edge could be mislocated by tens of metres. A line orientation run parallel, rather than perpendicular, to the plume's migration direction would produce a survey that samples along the plume's length without ever crossing its lateral boundary, giving a false impression that the anomaly is uniform or absent. Choosing too high an operating frequency (too shallow a skin depth) would fail to see the 15 m-deep leachate front at all, wasting the whole survey budget on a signal that never reaches the target; choosing too low a frequency would sacrifice near-surface resolution and blur the plume's true edge. Omitting base-station repeats would leave the survey unable to distinguish a genuine spatial trend in conductivity from simple instrument drift over the survey day, undermining confidence in the final map regardless of how well the spatial sampling was otherwise designed.