18-Geol-A7 Applied Geophysics · May 2015
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
National Exams — May 2015 — 04-Geol-A7 Applied Geophysics. Three-hour, closed-book exam; no calculator permitted. The NOTES state that SIX questions constitute a complete paper (the first six as they appear in the answer book), but the printed paper offers a choice of six of the following nine questions, and every question requires an essay-format answer with no numeric data, formula sheet or figure supplied — this is an all-essay paper. All nine questions are answered below.
Reference texts: Telford, Geldart & Sheriff, Applied Geophysics (2nd ed.) — the primary reference for every method touched in this paper (gravity, magnetics, seismic reflection/refraction, resistivity, IP, EM, radiometrics, well logging); Kearey, Brooks & Hill, An Introduction to Geophysical Exploration (3rd ed.) — survey planning, data display and case-history context; Blakely, Potential Theory in Gravity and Magnetic Applications — magnetic anomaly shape and reduction-to-pole theory (Q6); Simpson & Bahr, Practical Magnetotellurics (Q3); Selley & Sonnenberg, Elements of Petroleum Geology (Q4, Q8 well-logging context).
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.
Profile (wiggle-trace / line) plots. Individual survey-line data plotted as amplitude vs. distance (or, for seismic, as stacked wiggle or variable-density traces vs. time). Assists: preserves the finest detail of a single line, essential for careful first-break/reflection picking and for spotting instrument noise or spikes. Hinders: gives no sense of the spatial (map-view) continuity of a feature between lines, so a real, laterally continuous target and an isolated one-line artifact can look identical.
Contour maps. Gridded data (from an interpolation such as minimum curvature or kriging) plotted as contour lines. Assists: reveals map-view trends, strike direction and closure of an anomaly at a glance, which a set of individual profiles cannot show directly. Hinders: the gridding/interpolation algorithm and contour interval both impose an artificial smoothness that can create or destroy apparent closures, and sparse or irregular line spacing can produce gridding artifacts (bull's-eyes, line-direction "corrugation") that look geological but are not.
Colour-shaded (image) maps. The same gridded data rendered as a continuous colour scale rather than discrete contour lines, often combined WITH contour lines. Assists: the eye picks out subtle amplitude gradations and spatial patterns from colour far more readily than from contour-line spacing alone, especially for a broad reconnaissance survey. Hinders: the choice of colour palette and of a linear vs. histogram-equalized (or other non-linear) colour stretch can visually exaggerate or suppress real amplitude differences, so the same data can be made to look dramatically different depending on the display choice — a poorly chosen stretch can hide a real, subtle target or manufacture a false one.
Shaded-relief (sun-illumination) maps. The gridded surface is artificially "illuminated" from a chosen azimuth and inclination to create a pseudo-3-D shaded appearance. Assists: dramatically enhances subtle LINEAR trends (faults, dykes, lithological contacts, structural fabric) that are easy to miss in a plan contour or colour map. Hinders: features trending PARALLEL to the illumination azimuth are strongly suppressed (an illumination-direction bias), so a single illumination azimuth can create the false impression that structural trends only exist in one direction — the standard remedy is to generate the shaded-relief map at several different azimuths and compare.
Vertical and horizontal derivative filters. The first (or higher-order) spatial derivative of the field, computed in the frequency domain. Assists: derivatives amplify short-wavelength (shallow, high-frequency) signal relative to long-wavelength (deep, regional) signal, sharpening edges and helping locate the boundaries/contacts of near-surface sources with much better spatial resolution than the raw field. Hinders: the same amplification acts on high-frequency NOISE, so a derivative map of noisy or coarsely sampled data can be dominated by artifacts rather than real geology, and it suppresses genuinely deep sources that a geologist may still care about.
Upward continuation. A frequency-domain filter that mathematically projects the field upward to a higher observation level. Assists: attenuates short-wavelength near-surface noise and shallow "clutter" far more than long-wavelength deep signal, so it isolates and enhances REGIONAL (deep) trends — the complementary operation to a derivative filter. Hinders: irrecoverably smooths out and can completely eliminate genuine shallow targets, so it must be used with a clear purpose (regional-residual separation) rather than as a generic "de-noising" step.
Analytic signal / total horizontal gradient maps. As introduced in Q6, these locate source edges independent of magnetization or field direction. Assists: gives a latitude- and remanence-independent structural (edge/contact) map, useful for combining magnetics data acquired across a range of latitudes or for bodies with unknown remanence. Hinders: loses the sign/polarity information of the original field, and peaks broaden for deep sources, reducing resolving power between closely spaced features.
3-D/2.5-D model or inversion sections. A cross-section or volume showing an inverted physical-property model rather than the raw field. Assists: the most direct link to a geological cross-section, showing inferred depth extent and geometry rather than a field pattern the geologist must mentally invert. Hinders: inversion is inherently non-unique, so the displayed model is only ONE of many that fit the data equally well, and an unwary viewer can mistake the smoothness/sharpness imposed by the inversion's regularization for real geological detail.
