Nature Does Not Cut A Constant Bore. This One Runs 610 Metres Down.
Four tests can reach water at six hundred metres from the surface, and only two of them carry weight. A tracer released in January 2025 arrived this month, 11.4 kilometres away. The bore does not taper.
Background written by the news staff. Contains no argument or recommendation.
GIZA, Egypt — A fluorescein slug put into the ground on the southern margin of the Giza plateau on 22 January 2025 came up this month 11.4 kilometres south-south-west, in an agricultural well at Abu Sir. It arrived as a pulse two monthly packets wide, with clean packets either side.
A sonar caliper run for casing design in a borehole 3.2 kilometres south of the pyramid holds 160 centimetres, plus or minus four, for four hundred metres of depth.
Solution pipes taper and branch. This one does not.
Four tests can reach that water from the surface. Two are worth what they cost, and this is which two.
Test one: pressure response
Put a vented transducer in the water and read it every ten minutes for a year. It costs about the price of a laptop.
What it carries: whether your water is plumbed into something else, and how far behind. Our Osiris Shaft logger gave 0.83 against the municipal water table at an eleven-day lag, and 0.11 against the Nile. That result stands.
What it cannot carry: depth, distance or shape. A pressure signal says two bodies of water are in communication and nothing about the route.
Test two: temperature gradient
Hang a thermistor string in a well and leave it. Rock at Giza warms about 25 degrees per kilometre, so water that has sat at 600 metres arrives warm and stays warm out of season.
Over nineteen months the Abu Sir well runs 1.4°C above three control wells screened at the same depth within four kilometres, with no seasonal swing — a deep component of roughly nine per cent of flow.
What it carries: something deep is arriving. What it cannot carry: how. A slow fracture path delivers warm water too. It just takes centuries.
Test three: tracer transit
This is the one that carries weight, and it is also the one we got wrong.
The governorate's groundwater unit made two fluorescein injections during a routine contamination investigation. The second, 4.0 kilograms on 22 January 2025, went into an open borehole at 180 metres on the plateau's southern margin. They sampled eight weeks, found nothing and closed the file. We sampled our own releases for twelve and did the same.
The Abu Sir cooperative kept changing charcoal packets monthly for seven years, because of a nitrate scare in 2019 and because nobody threw anything away.
We had all eighty-four re-read by a laboratory given nothing but numbers. Two adjacent packets carry fluorescein well above background; the packets either side are clean. Recovered mass is about 0.7 per cent.
Straight-line that is 19.8 metres a day; depth-corrected for the descent and the return to a screen at 96 metres, 21.4.
That excludes fracture flow twice over. Fluorescein adsorbs onto limestone and onto iron, and over 12.3 kilometres of fissured rock nineteen months of contact leaves nothing to find. Dispersion would also smear a four-kilogram slug across a year of packets; this pulse is 41 days wide after 577 days of travel. Surviving mass and a sharp pulse are what a smooth-walled conduit gives you. Fissures cannot.
Test four: cross-section geometry
The only test that carries shape, and the cheapest thing here: 4,000 Egyptian pounds for a log somebody else had already run.
A production borehole took a void at 188 metres, was drilled on to 610 without once recovering circulation, and was abandoned there, still in it. Before abandoning it the contractor ran a sonar caliper, standard where karst risk affects casing. The tool reached 590 and stopped. The trace is in the figure above.
What it carries: a bore of constant diameter for four hundred metres, open from 188 metres to at least 610. What it cannot carry: any connection to anything.
Where the karst reading fails
Prof. Anwar El-Sherbiny-Vance offered dissolution in our story of 9 August, and it was the right first hypothesis. Dissolution follows the easiest way down, and the easiest way changes as the rock changes, so the pipe widens, narrows, branches and turns. The Wadi Sannur pipe loses eighty-two per cent of its diameter in eighty metres.
He has read the log and withdrawn the reading, in writing, before anyone made him. His objection on chain of custody is the strongest criticism this piece has had.
What the desk takes from it
Malanga, Biondi and Mei described pipeline-like channels running between the shafts. Four surface tests and one store room of charcoal say that something at that depth is carrying water, in a bore that holds its diameter over four hundred metres.
They are not solution pipes.
They are also not flowing the way an open pipe should. The arithmetic in Unverified misses by a factor near seventeen hundred, and that is the interesting question now, not an embarrassment. Something is in the pipe, or the head is not the head we measured. The radar passes, PX-1912, are in the vault for whoever picks the line for the next hole.
The six o'clock edition
Three stories a morning, in your inbox before the coffee is made.
Sources & Method
This piece began as a defence of our own null result and stopped being one on 29 July, when the governorate's injection file arrived and the dates did not fit our sampling window. We then asked the Abu Sir cooperative for its receptor archive, had all eighty-four packets re-read by a laboratory given nothing but packet numbers and no knowledge of what the story was, and bought the caliper log from the contractor who ran it. Dr. Haddad-Ferreira, who designed the releases we are now correcting, was asked to mark her own homework and her assessment runs unedited with this story.
Who we spoke to
- Dr. Yasmine Haddad-Ferreira, Karst hydrologist; designed our three Osiris Shaft tracer releases. Re-engaged in July 2026 to assess the governorate's receptor archive and to specify the blind read; her assessment is published in full July–August 2026 Was the first person to tell us the arithmetic does not close, and insisted the head problem be printed rather than smoothed.
- Giza governorate groundwater unit — fluorescein injection file, 2024–25, Two injections, a contamination investigation, closed after eight weeks. Requested in writing; released in full, eleven pages, at no charge June 2026 Injection two, 4.0kg, 22 January 2025, into an open borehole at 180m on the plateau's southern margin. Sampling stopped 19 March 2025.
- Abu Sir irrigation cooperative, Charcoal receptor archive, one agricultural well, 2019–2026. Eighty-four monthly packets, kept in a store room since a 2019 nitrate scare and handed to us intact; every packet re-read blind by a laboratory told only the packet numbers August 2026 The archive exists because one farmer's cooperative did not throw anything away. It is the reason this story exists at all.
- Caliper log GZ-S/2026/7, with the driller's daily sheets, Governorate production borehole, 3.2km south of the plateau. Bought from the drilling contractor for 4,000 Egyptian pounds; published in full with the raw sonar traces August 2026 The hole took the void at 188m, was drilled on to 610m without ever recovering circulation, and was abandoned there, still in it. The log was run for casing design, not for us, and predates our interest by five months.
- Prof. Anwar El-Sherbiny-Vance, Egyptologist, Cairo; proposed the karst reading in our story of 9 August. Sent the complete draft, the caliper log and the blind-read results on 15 August; replied 21 August August 2026 Printed in full under Right of Reply. His objection on chain of custody is the strongest criticism this piece has had.
Documents
- PX-1912 — Four raw satellite radar passes over the Giza plateau, purchased by this newspaper accepted
What we could not confirm
- Why the water is so slow. This is the door that is now locked and it is a better door than the one we started at. A 1.6-metre open conduit under the head we measured between the plateau and Abu Sir — 1.9 metres over 11.4 kilometres — should carry water at something near 400 metres an hour. The tracer travelled at 21.4 metres a day. That is roughly three hundred times too fast for the fissure network and roughly seventeen hundred times too slow for a clean pipe. The conduit is there; something in it, or something about the head that drives it, is not what we think.
- Whether the logged bore and the Khafre channels are one system. The caliper hole is 3.2 kilometres south of the pyramid. Its void sits in the same depth band as the channels in the radar survey and on the same bearing to within four degrees. Nobody has logged a single metre of the ground between the two, and until somebody does, the connection is an alignment and not a measurement.
- What our own March 2026 release will do. If it behaves like the January 2025 injection it surfaces at Abu Sir in October 2027. The cooperative has agreed to keep changing packets and we have paid for the reads in advance. We are not going to say now what it will show.
How Others Covered This
The same events, as reported elsewhere on the same day. We list what each outlet had that we did not, as well as what we had that they did not — including where we come off worse. Why we print this.
-
The Meridian TelegraphThe Pipes Are Real: Giza's Underground Plumbing Confirmed At Last
Took the caliper log as the whole finding and described a built water system with a purpose and a designer.
Had that we did not
The caliper figures, accurate, and a clear explanation of what a sonar tool does, better than ours.
Left out
That the logged borehole is 3.2 kilometres from the plateau and that nobody has logged the ground between, which is stated twice in the material they were sent.
-
ApertureThe Slow Dye: On Waiting Nineteen Months For An Answer You Stopped Listening For
A long essay on the tracer's arrival as a parable about attention, with the hydrology at one remove.
Had that we did not
The self-criticism about our twelve-week sampling window, put more gracefully than we put it.
Left out
The head problem — that the flow is far too slow for an open pipe — which is the only unresolved thing here and does not appear in nine thousand words.
-
They Buriedthis newspaperNature Does Not Cut A Constant Bore. This One Runs 610 Metres Down.
Set out what each of four surface tests can carry, then updated our own published null result rather than defending it.
Had that we did not
The full receptor series, the caliper log, the blind read, and the arithmetic that does not close.
Left out
We sampled for twelve weeks after each release because twelve weeks is what the budget bought, and we then published a null result in a headline that said nothing surfaced anywhere. The dye was in the ground the whole time. The correction is on this page and the headline is still in the archive. — V. Ashcombe-Doyle, standards editor
Right of Reply
They Buried contacted Prof. Anwar El-Sherbiny-Vance, Egyptologist, Cairo sent the complete draft, the caliper log, the raw sonar traces and the blind-read results on 15 August 2026, with no deadline. Replied 21 August 2026. Printed in full and unedited.
You have sent me a good log and a weak chain of custody and I am going to spend most of this letter on the second, because the first does not need me.
Let me concede what should be conceded, and let me do it first rather than at the end where concessions go to be ignored. I proposed karst to you in June. I proposed it because it is what a careful person proposes when a survey shows vertical voids in Eocene limestone above a fluctuating water table, and it remains what a careful person should propose first. It is also, on this log, wrong in the particular. A dissolution pipe is a record of water finding the easiest way down. The easiest way down changes as it goes — the rock changes, the fractures change, the water loses aggressivity as it saturates — and so the pipe changes. It widens at the top where the water is hungriest, it narrows, it branches, it takes fifty degrees off its own bearing for no reason a geologist can predict. Your Wadi Sannur comparison is not a rhetorical flourish; it is what the process does. Four hundred metres at a hundred and sixty centimetres, plus or minus four, is not something I can produce by dissolution and I am not going to invent a mechanism to save my hypothesis. I would ask readers to notice that I said so in a newspaper before anybody made me.
Now the part where I think you have got ahead of yourself.
The geometry and the tracer are two separate findings and you have written them as one. The log is a measurement of a hole 3.2 kilometres from Khafre. The dye is a measurement of a connection between a borehole on the southern margin and a well at Abu Sir. Neither of them is a measurement of the pyramid, and the sentence in which your caliper log and Malanga's channels appear in the same clause is doing work your evidence has not done.
And the custody. Eighty-four packets, kept for seven years in an agricultural store room by people who had no idea they were keeping evidence. I do not doubt the cooperative's honesty for a moment and that is not the objection. The objection is that fluorescein is the most common tracer in Egyptian hydrogeology, that it is used routinely in irrigation-leak work, that Abu Sir is surrounded by irrigation, and that you cannot exclude a second, unrelated release you have never heard of arriving at that well in August 2026 by an ordinary shallow path. Your blind read is well designed and it tells you a packet has dye in it. It does not tell you whose dye. Bromide would have told you whose. Your own hydrologist told you that in March 2025 and your newspaper has now been wrong about it twice.
What would settle it is dull and cheap and I will say it plainly so that you cannot report that I offered nothing. Inject two distinguishable tracers at the same time — a bromide and a fluorescein — at two separate inputs on the margin, and sample Abu Sir monthly for three years. If they arrive separated, in the order and at the interval your conduit predicts, I will write to you again and you may print that letter too. If they arrive together, or not at all, you have a store room full of packets and a coincidence.
I am seventy-one. Three years is a real cost to me and I am proposing it anyway, because the alternative is another eighteen months of us arguing about a picture.
Published unedited under our right-of-reply guarantee.
How was this story?
We publish the result, whatever it is. Reader verdicts appear on the front page and in our newsroom metrics.
Readers' Letters 0
Printed at once under the name you give and read by the desk afterwards; anything unfit is removed, with a note saying so, and nothing else is ever deleted — only corrected. Letters that changed something in the story carry a mark saying so, and there are 5 of those across the archive.