How it works

From scattered datasets to ranked targets

We start from what already exists: public surveys, state data and your own archives. Then we clean it, model it jointly and rank targets with an honest measure of uncertainty.

Workflow

Five steps, one deliverable

  1. STEP 1Select a regionProspective geology for critical minerals, active claims and good data coverage.
  2. STEP 2Compile and cleanGather every public survey, then harmonize coordinates, resolution and errors.
  3. STEP 3Invert jointlyPhysics-based 3D models that honor all methods at once.
  4. STEP 4Add geologyDeposit models, structure, geochemistry and remote sensing.
  5. STEP 5Rank with uncertaintyValidated AI prospectivity and competing explanations.
  6. DELIVERYRegional intelligence briefPreliminary findings, GIS layers and 3D models for companies working the region.

An anomaly is not an orebody. We say which one you are looking at.

Many AI exploration claims skip steps. We report results at the level the evidence supports. Where drill and assay data are thin, we deliver geological favorability, not deposit predictions, and we say so in writing.

  1. 1Geophysical anomaly
  2. 2Geological interpretation
  3. 3Mineral-system favorability
  4. 4Evidence-supported drill target
  5. 5Confirmed mineralization: only the drill decides

Methods

Six methods behind one question: what else can the existing data tell us?

01

Regional data discovery and reprocessing

We search federal, state and published sources for every survey over a region, then QC, re-level and re-grid it and document what each dataset can and cannot resolve.

Output: a clean regional data compilation and gap report.

02

Gravity and magnetic joint inversion

Structurally coupled 3D inversion that recovers density and susceptibility together, so intrusions, basins and alteration zones agree across methods.

Output: 3D density and susceptibility models with depth sections.

03

EM, IP and MT interpretation

Conductivity and chargeability models from airborne and ground EM, induced polarization and magnetotellurics, including IP effects recovered from existing airborne EM.

Output: 2D and 3D resistivity and chargeability models.

04

AI mineral prospectivity mapping

Machine learning trained on known deposits and mineral-system criteria, with spatial cross-validation and checks for sampling bias.

Output: favorability and uncertainty maps as GIS layers.

05

Remote sensing mineral mapping

Alteration and lithology from Sentinel-2, Landsat, ASTER and airborne hyperspectral data, tied back to the subsurface model.

Output: alteration and mineral maps with interpretation.

06

Integrated 3D geological models

Inversion results, mapped geology, geochemistry and drill holes combined into one model a team can query and update.

Output: 3D models, sections and a technical report.

Working in a region we should study?

Tell us where your ground is and what you are looking for. We will tell you what existing data covers it and whether a brief is available or worth building.