The Gold Standard of Archaeology | Seismic Reflection, Velocity Pull-Up and Migration Artefacts

1. Begin with an impedance contrast

Back to contents

A reflection arises where density and seismic velocity combine into a change in acoustic impedance. A wall, bedrock surface, filled channel or stratigraphic boundary may reflect energy if the contrast is large enough and the geometry is resolvable. Absence of a reflection is not absence of archaeology; weak contrast, attenuation, destructive interference or poor coupling can hide a target. Define the expected physical contrast before choosing a source, receiver spacing or processing sequence.

2. Keep travel time separate from depth

Back to contents

The raw record measures arrival time. Converting two-way travel time to depth requires a velocity model and the path the wave travelled. For a flat reflector and constant velocity, depth is approximately velocity multiplied by two-way time divided by two, but shallow ground is rarely uniform. Moisture, compaction, fill and weathered rock change velocity laterally and vertically. Label time sections honestly and show how depth estimates depend on measured or assumed velocities.

3. Design acquisition around the shallow target

Back to contents

Source bandwidth, geophone frequency, spacing, offsets and line direction control what can be resolved. Very shallow reflections need close spacing and careful suppression of direct, refracted and surface waves. Lines crossing a target obliquely can distort its apparent width. Record topography, coupling, surface materials and cultural noise. A dense array cannot recover frequencies the source or ground never transmitted, and a powerful source is not automatically a high-resolution one.

4. Build common-midpoint gathers before stacking

Back to contents

Common-midpoint processing groups traces that sample roughly the same subsurface point at different offsets. Normal-moveout correction uses a chosen velocity so a reflection aligns before stacking. A wrong stacking velocity leaves residual moveout, smears the event and can favour noise. Inspect gathers and velocity panels rather than accepting a smooth stack as proof. Muting must remove stretched or interfering energy without erasing the shallow reflection of interest.

5. Recognise velocity pull-up and push-down

Back to contents

A high-velocity body above a reflector shortens travel time, so the deeper reflector can appear pulled upward on a time section. A low-velocity zone delays energy and can push it downward. These are travel-time effects, not necessarily real relief. Compare nearby gathers, refraction or surface-wave velocity information, boreholes and alternative velocity models. If an apparent mound or hollow follows the overlying velocity anomaly, interpret structure only after testing the velocity explanation.

6. Worked example: a buried channel beneath compact fill

Back to contents

A profile shows a bright horizon rising by four milliseconds beneath a compacted track. Using 600 metres per second everywhere gives an apparent 1.2 metre rise because depth equals velocity times two-way time divided by two. Refraction data indicate that the track fill is much faster than adjacent soil, so energy returns earlier without the underlying boundary being shallower. Depth migration with the measured near-surface model flattens most of the rise. Excavation then finds the horizon nearly level and a narrow channel beside it. The first image was useful, but its geometry was not literal.

7. Understand what migration is trying to repair

Back to contents

Unmigrated reflections are plotted at recording positions even when energy came from a dipping boundary or diffraction point. Migration uses a velocity model to move energy toward its probable subsurface origin, collapse diffraction hyperbolas and improve structural position. It does not manufacture certainty. Too low, too high or laterally incomplete velocities can under-migrate, over-migrate or focus noise into false edges. Compare unmigrated and migrated sections and retain the parameters used.

8. Changed-context transfer check: a possible chamber

Back to contents

A diffraction pattern beneath a temple courtyard is interpreted as a chamber corner. On a second line the event changes shape, and migration focuses it at a utility trench crossing rather than a closed void. The transferable method predicts how a real three-dimensional target should appear on differently oriented lines, tests alternative velocities and checks surface infrastructure. Reflection geometry becomes archaeological evidence only when it is consistent across acquisition directions and independent constraints.

9. Know the limits and common misconceptions

Back to contents

A strong continuous reflector is not automatically a wall, and a gap is not automatically a doorway. Multiples, air waves, ground roll, side reflections, ringing and processing choices can imitate structure. Vertical and lateral resolution differ, especially for thin beds and steep edges. Two-dimensional lines assume limited out-of-plane energy. Migration may sharpen a wrong interpretation. Report ambiguity and avoid converting every colour boundary or high amplitude into a named archaeological feature.

10. Audit the profile from source to interpretation

Back to contents

Archive raw traces, geometry, source and receiver specifications, timing checks, topography, field notes and every processing parameter. Preserve intermediate gathers, velocity picks, mute windows, filters, stacks and migrations. Reprocess plausible alternatives and document which features persist. Tie depth to boreholes, excavations or independent geophysics where possible. An interpreter should be able to distinguish a field observation from a processing decision and a geological inference.

11. Frequently asked questions

Back to contents

Does reflection directly show depth? No; it records travel time until a velocity model is applied. What is pull-up? An apparent upward shift caused by faster overlying material. Does migration always improve the image? Only when geometry, velocity and dimensional assumptions are adequate. Can it find small artefacts? Usually the method is better suited to interfaces or substantial features than isolated small objects. Why keep the unmigrated section? It reveals how much geometry the processing changed.

12. The gold standard treats the section as a model

Back to contents

A seismic reflection image is a disciplined reconstruction from recorded wavefields, survey geometry and a velocity model. The gold standard defines the target contrast, preserves gathers, tests velocities, explains pull-up or push-down and compares migration choices. It uses excavation or independent geophysics to anchor interpretation. The goal is not the sharpest-looking picture; it is a subsurface claim whose travel-time and processing assumptions remain visible.