Borehole Correlation Panel Maker

How the correlation lines work

Ties follow the Unit column, never the USCS symbol. Two holes both logging CL are not necessarily the same clay — they may be two clays with a sand between them that neither hole caught — and joining them because the letters matched would be the software inventing a correlation and putting your name on it. The unit label is your interpretation, and it is the only thing the panel will act on. Leave it blank and you get a clean panel of strips with no ties at all, which is a perfectly good figure to start from.

Between each pair of ADJACENT holes, a unit present in both gets two dashed lines: top boundary to top boundary, base to base, with an optional very light wash between them so the body reads as one stratum. Where a unit is present in one hole and missing from its neighbour, the pair converges to a single point half way between the two — the standard pinch-out convention, and the honest one, because all the data supports is that the bed dies out somewhere in that gap.

Ties are only ever drawn between neighbours. If a unit appears in holes 1 and 4 but not in 2 and 3, you get a wedge dying out east of hole 1 and another appearing west of hole 4, and nothing across the middle — rather than one bed sailing over the top of two holes that looked for it and did not find it. Where a unit is absent, the pinch point is placed at the contact between the units that bracket it in that hole, so a bed wedges out exactly on the surface that the two beds either side of it share.

Scale, exaggeration and the things that mislead

A section is read by measuring off the paper, so the two ways to make a misleading one are both about scale, and the figure defends against both in the caption. The vertical exaggeration is printed as an H:V ratio in the title block and again in the footer, because it is almost never 1:1 — a 120 m section of 13 m holes has to be stretched vertically to show anything at all, and a reader who does not know by how much will read a 2° dip as 9°.

The second is spacing. Proportional places every hole at its true chainage, and only then is the drawing a section whose geometry you can measure. Equal spacing produces a fence diagram: much easier to read when one hole sits 400 m from a tight cluster of three, and no longer telling the truth about the horizontal. The footer states which you are looking at every time, in words, because a reader cannot tell by eye. If two holes are so close that proportional spacing would overlap their strips, they are pushed apart to a minimum gap and the footer downgrades the scale statement to "nominal" rather than quietly lying.

Who this is for

Geotechnical and engineering-geology consultants drawing the sections for an interpretative ground investigation report. Site-investigation engineers checking that a set of logs actually hangs together before the factual report goes out — a panel makes a mistyped ground level or a unit called two different things obvious in a way that ten separate log sheets never do. Environmental consultants sectioning a line of trial pits across a plume, and anyone who has drawn one of these by hand in CAD and knows what an afternoon of it costs.

It is not a geological modelling package. There is no gridding, no surface interpolation across a whole site, no block model, no AGS import and no project database. It draws the section you tell it to draw, from a table, and states its own assumptions on the sheet.

Honest limits

Metres only, both axes, and the footer says so. One straight section line per figure — holes are placed by their chainage along it, so an offset from the line is something you resolve before you type the number, exactly as you would drawing it by hand. Ties connect units you have named; nothing is correlated automatically, and nothing on this figure is a proved boundary except the intervals inside the strips themselves. That last point is printed on every export.

Everything runs in this browser tab. No borehole record is uploaded anywhere, which for pre-planning site data on a live acquisition is usually the whole question.

Common questions

The page says Pro — what can I actually do for free?

Everything except enter your own data. The section below is the finished tool running live on a real four-hole dataset: change the vertical scale, switch between proportional and equal spacing, turn the ties, the shading and the water line on and off, widen the strips, edit the title block, and export as SVG or PNG (watermarked, like every free export on this site). What Pro buys is the right to replace the sample holes with yours. There is no trial timer and no account, and the demo does not expire.

How do I tell it which strata to correlate?

With the Unit column. Give the intervals you believe are the same body the same unit name — "Alluvial Sand", "Till", "Unit B" — and adjacent holes carrying that name are tied together. Matching ignores case and spacing, so "alluvial sand" and "Alluvial Sand" are one unit. Nothing is correlated on the USCS symbol, deliberately: that decision is yours to make and to sign.

What happens to a unit that is missing from one hole?

It pinches out. The tie pair converges to a single point half way between the hole that has it and the one that does not, at the contact between the units that bracket it in the hole that is missing it — so the wedge closes exactly on the surface those two neighbouring beds share. If that hole logs nothing either side of the missing unit, the point falls back to the mid-level of the unit in the hole that does have it.

Do my holes have to be in order in the table?

No. They are sorted by chainage before anything is drawn, and the figure carries a note telling you the table order and the drawn order differ. Rows for one hole can also be anywhere in the table — they are grouped by the Hole column, not by being adjacent.

Why do I have to repeat the chainage and ground level on every row?

Because the alternative is worse. They are properties of the hole rather than of the interval, but a set of per-hole option fields that has to grow every time you paste a fifth borehole is a far clumsier thing to use than a column you fill down in a spreadsheet. If two rows of the same hole disagree about its ground level, the panel uses the first and prints a warning naming the row — which is how a typed-twice level gets caught rather than silently moving a hole.

What vertical exaggeration should I use?

Whatever makes the strata readable, and then say so — which the figure does for you, in the title block and the footer. Sections of this kind commonly run between 2:1 and 10:1. The ratio here follows from the vertical scale you pick and the length of your section, so if you need a specific number, adjust the vertical scale until the printed ratio is the one you want.

Does it draw the same hatch patterns as the borehole log?

The identical ones — the two figures share a single set of pattern definitions rather than each having its own copy, so a clay cannot look one way on the log and another way on the section. Patterns follow FGDC-STD-013-2006 §37 and the group symbols ASTM D2487, both cited in the footer.