Seismic · Deliverables
What a site class report should contain
Thin reports are not obviously thin until a plan reviewer asks a question you cannot answer. Here is what should be in the deliverable before that happens.
Most engineers commission their first shear-wave velocity survey without a clear picture of what a good one looks like. The reports vary enormously. The difference rarely shows up at delivery; it shows up months later, in review.
The number, and the profile behind it
The V̄s30 and the resulting site class, stated against ASCE 7-22. Then the velocity profile it came from — the layered model with depths and velocities, not only the averaged figure.
Because the average is a harmonic mean, a soft near-surface layer influences the result far more than its thickness suggests. Without the profile you cannot see that, and you cannot explain it to anyone who asks.
The method and the acquisition parameters
MASW, ReMi, or both. Geophone spacing, array length, source type, record length, the number of shots or recordings. A reviewer may reasonably want to confirm the array was long enough to resolve to the depth being averaged over.
Where both methods were run, both profiles, and a sentence on whether they agree.
The dispersion curve and the inversion
This is what thin reports omit. The picked dispersion curve, the inverted model, and some indication of fit.
Inversion is non-unique. Several velocity models can fit one dispersion curve, and the choice among them is interpretive. A report that presents the profile without the inversion is asking you to take an interpretation on trust, and it leaves you with nothing to show when someone asks how the number was arrived at.
Where the array actually was
A site plan with the array marked, and coordinates. Velocity varies across a site. A reviewer asking whether the measurement represents the structure’s footprint needs to see where it was taken, and so does the next engineer who picks up the project.
The limitations
Any honest geophysical report states what it could not resolve. Shallow refusal, interference, an array shortened by site constraints, a deepest layer extrapolated rather than measured.
A report listing no limitations at all describes either a perfect site or an incomplete account of an ordinary one.
And someone to ask
A named person who can answer questions about the interpretation when the review comes back, which may be a long time after the invoice.
Four questions worth asking first
- What depth will the array resolve to, given the access at this site?
- Which method, and why that one here?
- Do I receive the dispersion curve and the inverted model, or only the result?
- Who answers questions after delivery?
Where a quote is materially cheaper than the others, these questions usually explain why.