LEYLINE // MINERAL INTELLIGENCE KEN · EASTERN RIFT MARGIN · GRID 0.05° --:--:-- UTC

Every terrane gives itself away differently.

Leyline scans every land cell on Earth, then works region by region to find out which signals actually reveal ore on that ground, and which ones lie.

Request a full scan View the map coming soon
STANDBY 248,137 LAND CELLS SCANNED · 15 SIGNAL FIELDS IN USE 00.000°S 000.000°E
02What we do

The data that finds the next deposit already exists. Nobody reads it together.

What we do

Leyline scans every land cell on Earth for the traces ore leaves behind. Public geoscience carries most of it: magnetics, gravity, lithology, structure, the shape of a drainage network. Some of it comes from places that have nothing to do with mining. Water chemistry. Soil gas. The way vegetation behaves over metal. Colonial survey ledgers written by people who walked the ground a century before anyone flew over it.

Then we narrow to one region at a time and work out which of those traces can be trusted on that particular ground.

Why

Because the ground does not behave the same way twice. A signal that reliably marks gold under the cover of Western Australia can be worth nothing across the copper belts of central Africa. The rock is different, the weathering is different, the plants growing on top of it are different.

Almost every prospectivity model ignores this. One rule gets applied to the whole planet and the map comes out looking confident everywhere. We do the opposite. We test each signal against ground where the answer is already known, region by region, and we keep what fails as carefully as what works. A region we have worked through can be searched fast, with a reason attached to every answer.

Some of the signals we use
Potential-field geophysics Lithospheric architecture Structural geology Landform and drainage Hydrogeochemical Biogeochemical Zoological Microbial Volatile and gas Isotopic Remote sensing Exotic geophysics Archival and historical Archaeological
Which of these hold, and where, is what the codex records.
03The inversion

Everyone else needs to see the rock. Most ground will not show it.

Exposed rock01
Alteration mapping
Under cover02
Alteration mapping

Bedrock in the open. The alteration halo around ore is visible from orbit, and every operator in the industry can see it.

Closed canopy over the same kind of rock. The standard workflow masks this ground out before the analysis starts.

The constraint

Satellite exploration reads alteration minerals on bare bedrock. It works, and it has one hard requirement: the rock has to sit in the pixel. Forest, regolith, sand and ice all defeat it. Whole basins get discarded before anyone looks at them, and exploration drifts toward the places that are easy to see rather than the places likely to hold metal.

The inversion

We treat the cover as the instrument. Roots reach metres down and integrate the chemistry of a soil volume for decades. Streams carry evidence downhill from ground nobody can see. Termites build their mounds out of material excavated from well below the surface. None of that needs an outcrop.

Arid ground, read the rock. Covered ground, read what grows on it and what runs off it. Knowing which one to trust is the part that changes with the terrane.

04The codex

What we learn about a region, we write down.

Every method we test gets an entry. The entry names the region, carries the verdict, and keeps the evidence that produced it.

A verdict is local. The same method can hold on one terrane and fall apart on the next, so the verdict belongs to the pair, the method and the ground, never to the method on its own. Most systems decide once that a technique works, then apply it everywhere.

Failures stay in. Knowing that something has already been tried on your ground and returned nothing is worth as much as knowing what worked, and almost nobody publishes it.

A Full Scan draws on the codex for the ground it covers. Everything it learns goes back in.

01 / 05

05The codex

Somewhere a plant grows only over gold. We are building the thing that knows which one.

BiogeochemicalCentral Africa
A flowering shrub that grows only where soil copper runs to thousands of parts per million
Published
BiogeochemicalWestern Australia
Particulate gold inside living leaf tissue, over ore buried under tens of metres of cover
Published
ZoologicalSouthern Africa
Mounds built from material excavated well below the surface and stacked in the open
Published
LandformGlobal
One sediment sample at an outlet, speaking for every square kilometre draining into it
Published
Everything above is published science. Nothing below it is.
Biogeochemical▪7
Validated here
Hydrogeochemical▪6
Validated here
Volatile and gas▪5
Under test
Zoological▪9
Rejected here
Archival▪6
Validated here
Microbial▪7
Under test
Remote sensing▪5
Rejected here
Archival▪8
Validated here
Biogeochemical▪6
Validated here

The record grows with every region we work. Nothing we measure leaves it.

06The stack

Everyone reads the same few layers. The answer is in the ones nobody bothers to read.

A conventional scan

Standard stack
Airborne magnetics Gravity Regional geology Satellite spectral
Signals read00

The same ground, ours

Full stack
Airborne magnetics Gravity Regional geology Satellite spectral Lithospheric architecture Mapped structure Drainage catchments Hydrogeochemistry Biogeochemistry Zoological Volatile and gas Isotopic Archival records Ancient workings Landform
Signals read00

Four layers and you are still searching a province. Fifteen and you are standing on a licence.

07Where this goes

The list of things that could reveal ore is long. The list anyone has actually tested is short.

Models that read further than any team can propose what to test. We test it ourselves, at continental scale, on ground where the answer is already known. Most candidates die there. The ones that live are ours.

Each region sharpens the next. Far enough down that road, finding a deposit stops being an expedition and becomes a lookup.

08Contact

Tell us about your ground.

Licence areas, whole jurisdictions, or a question about a terrane we have already worked through. We reply to everything.

contact@leylineai.com