Borehole · Logging
Which log answers which question, and the window for asking
A borehole is the most expensive thing on most site investigations and the shortest-lived. Once it is cased or backfilled, most of what it could have told you is gone.
Borehole geophysics gets described as though it were one thing. It is not. It is a set of instruments that answer different questions, need different conditions, and stop being available at different moments. The decision that matters is rarely which log is best. It is which ones are still possible by the time anybody thinks to ask.
The window closes faster than people expect
A geotechnical borehole typically exists as an open, fluid-filled hole for a matter of hours. Then it is cased for monitoring, grouted, or backfilled and abandoned. Each of those closes off a different set of measurements, and none of them can be reopened without drilling again.
The common sequence is that a project drills, logs the samples, backfills, and then three weeks later a question arises — about bedrock depth, about a soft horizon between sample intervals, about site class — that the hole could have answered for a fraction of what it cost to drill. By then the answer costs another borehole.
What each log answers, and what it needs
| The question | The log | What the hole has to be |
|---|---|---|
| Shear and compression wave velocity against depth; site class | Downhole seismic | Open or cased — but casing must be well grouted to the formation, or the arrival is not worth picking |
| Clay and shale content; correlating horizons between holes | Natural gamma | Anything. Logs through steel or PVC — the one tool that survives casing |
| Formation character, water quality changes, permeable horizons | Resistivity and spontaneous potential | Open and fluid-filled. Casing ends it |
| Hole condition, washouts, fractures, intersected voids | Caliper | Open hole |
| Inflow zones, water quality stratification | Fluid temperature and conductivity | Open and fluid-filled |
Read the right-hand column as a schedule rather than a specification. Everything except natural gamma has to happen before the hole is cased, and three of the five need fluid in it.
Run the caliper first
The caliper looks like the least interesting log on the list and it is the one to run first. Hole diameter against depth is what tells you whether the rest of the logs can be believed.
A washout — a section where the hole has caved to well over gauge — puts standoff between every other tool and the formation it is supposed to be reading. A resistivity anomaly over a washed-out interval may be the formation or may be a foot of drilling fluid. Without the caliper there is no way to tell, and the interpretation quietly inherits an error nobody can see. It also finds the fractures and open voids the hole passed through, which on karst ground is frequently the thing the whole investigation was about.
What casing costs you, and what it does not
Casing a hole is not the end of logging, which is the useful half of this. Natural gamma reads straight through steel and PVC, so clay content and horizon correlation remain available indefinitely — including in holes drilled years ago, where the gamma log is often the only way to tie an old investigation to a new one.
Downhole seismic also survives casing, with a condition attached. The geophone clamps against the casing wall, so the casing has to be coupled to the ground around it. A properly grouted installation logs well. One with voids behind the casing, or a loose annulus never properly tremied, gives shear arrivals that cannot be picked with confidence. That is worth specifying at the time the hole is installed rather than discovering on the day somebody turns up to log it.
What casing does end is everything that needs electrical or hydraulic contact with the formation: resistivity, spontaneous potential, caliper, fluid logging. Those are open-hole measurements and there is no way around it.
Why a logged hole is worth more than a drilled one
Surface geophysics gives continuous coverage with uncertain depth. A borehole gives certain depth at a single point. Neither is sufficient alone, and the value of logging is almost entirely in joining them — the log supplies the velocity or resistivity tie that anchors the surface model, and the surface model carries the borehole across the site.
That is the difference between a depth interpretation with a real uncertainty band on it and one a designer can build to. One logged hole will often do it for an entire site.
What to ask, and when
Ask before the drill rig is scheduled, not after it has left. Ask whether the hole will be cased, when, and whether the annulus will be properly grouted — because that single decision governs most of the table above. If the hole is going in for monitoring anyway, ask what it would cost to log it open before the casing goes in, which is usually a small fraction of the drilling already being paid for.
And if the holes on your site were drilled years ago and cased, ask anyway. Gamma still works down there.