Skip to main content
Mining module

GeoMine Scientist

Resource Estimation and Geological Modelling on a Certified Compute Foundation. GeoMine Scientist is the mining module of the platform. It takes logged drillholes through desurveying, compositing, implicit geological modelling, ordinary kriging, and sequential Gaussian simulation to block models, grade shells, exceedance probabilities, and grade–tonnage reports. It runs on a separately certified numerical foundation, and every artifact it produces is content-addressed: the same job always returns the same result, and the result always says what produced it.

Request DemoExplore Mining Solutions
A kriged block model in the 3D viewer — banded grade classes with block counts, merged block edges, a scale bar, and an axis triad.

Product positioning

Resource Estimation and Geological Modelling on a Certified Compute Foundation

Drillhole data, implicit geological models, kriged block models, conditional simulation, and grade–tonnage reporting — every result carrying the job, parameters, and inputs that produced it.

Key capabilities

Drillhole desurveying from collars, surveys, and logged intervals
Sample compositing and content-addressed input preparation
Implicit geological modelling of contacts and stratigraphic columns
Isosurface extraction into watertight grade and boundary shells
Ordinary kriging with anisotropic variograms and declared search neighbourhoods
Sequential Gaussian simulation with realization stacks
Exceedance probability models against a stated cut-off
Grade–tonnage reporting with P10–P90 uncertainty bands
3D block, mesh, scalar-field, and drillhole visualization with sections and A/B comparison
Content-addressed artifacts with full lineage back to the source records
Structured, four-category refusals instead of silent bad numbers
Reproducibility guarantees — the same job returns the same result identity

Target users

Resource geologists
Mining geostatisticians
Exploration geologists
Mine planning engineers
Competent Persons and technical reviewers
Corporate technical services teams
Auditors and due-diligence reviewers

Component ecosystem

One chain, from logged hole to reported tonnage

GeoMine Scientist connects the stages that normally live in separate desktop packages — and keeps the artifact each stage produced, so the next one consumes evidence rather than a re-keyed copy.

Drillhole desurveying

Compute traces from collars, downhole surveys, and logged intervals — and refuse to extrapolate past the deepest surveyed station.

Compositing & inputs

Composite assays to a declared support and content-address the payload before any computation runs.

Implicit modelling

Fit a scalar field through logged contacts for a declared stratigraphic column, with anisotropy read from the geologist's interpretation.

Isosurfaces & shells

Contour the field into watertight shells with recorded volume, surface area, and Euler characteristic.

Kriging & variance

Ordinary kriging with an anisotropic variogram, returning an estimate and its kriging variance per block.

Conditional simulation

Sequential Gaussian simulation producing a stack of equally probable realizations addressed as one artifact.

Grade–tonnage reporting

Tonnage above cut-off as P10, P50, and P90, with contained metal computed per realization.

Provenance & lineage

Content-addressed results whose manifests record kernel, inputs, parameters, actor, and toolchain pins.

Connected resource workflow

Each stage consumes the artifact the last one produced

No stage starts from a re-keyed spreadsheet, and every result names the inputs it came from.

Step 1

Log

Step 2

Desurvey

Step 3

Composite

Step 4

Model

Step 5

Estimate

Step 6

Simulate

Step 7

Report

Key capabilities

Resource estimation and geological modelling

GeoMine Scientist is the mining module of the platform. It takes logged drillholes through desurveying, compositing, implicit geological modelling, ordinary kriging, and sequential Gaussian simulation to block models, grade shells, exceedance probabilities, and grade–tonnage reports. It runs on a separately certified numerical foundation, and every artifact it produces is content-addressed: the same job always returns the same result, and the result always says what produced it.

Start from the drilling, not from an export

Collars, surveys, logged intervals, and assays are read from the governed project database. Desurveying returns traces in the mine's own coordinate reference and vertical datum, and the exporter refuses rather than guesses — a station with no bearing on a deviated hole, or lithology logged below the deepest survey, stops the run instead of quietly becoming an assumption.

Make the geological reading geometric

Logged contacts become anchors, an implicit field fits the declared stratigraphic column through them, and isosurfaces contour that field into shells. Nested shells from one field render as layers of a single model, ordered by isovalue — because artifacts that declare identical inputs are lineage siblings, not unrelated objects that happen to share a project.

Estimate, then quantify what the estimate hides

Ordinary kriging returns estimate and variance per block inside a declared search neighbourhood. Sequential Gaussian simulation then produces a realization stack, and an exceedance model converts it into the probability each block clears a stated cut-off — so a single kriged number is never the only thing on the table.

Report a quantity, and say that it is one

Grade–tonnage curves carry a P10–P90 band, contained metal is averaged across realizations rather than multiplied from two means, and the report states explicitly that it is not a resource classification. Classification under a reporting code stays a Competent Person's judgement, and no category label is emitted.

Hand a reviewer something they can check

Every artifact is identified by the hash of its own content and carries the kernel digest, input identities, parameter hash, randomness root, acting identity, and toolchain pins. Lineage is walked from manifests alone, stored results are re-verified on every read, and an invalid job is refused with a code, a category, and a stage rather than answered with a plausible number.

What it deliberately does not do

Being explicit about the boundary is part of what makes the rest defensible.

It does not classify

No Measured, Indicated, or Inferred label is emitted. Classification under a reporting code is a Competent Person's judgement.

It does not interpret for you

The stratigraphic column, contact polarity, and anisotropy orientation are the geologist's reading, recorded as their claim.

It does not guess

Missing assumptions, uninformed blocks, and unsurveyed ground produce a structured refusal, not a plausible number.

A module of the platform

GeoMine Scientist runs on the shared geoscience foundation

GeoMine Scientist is one module of the SpatialTechSolutions AI-powered geoscience platform — sharing the same governed data foundation, map, and AI layer as the GIS, groundwater, geotechnical, and mining modules.

Explore the platform

Product screenshots

GeoMine Scientist in action

A closer look at the working application — the screens your team will use day to day.

Three nested grade shells contoured from one implicit scalar field, rendered as layers of a single model with drillhole traces in the same scene.
Desurveyed drillhole traces in 3D, coloured by the unit logged at each station — drawn between the stations they were logged at, with nothing interpolated.
A grade–tonnage report with a P10–P90 uncertainty band across realizations — a stated quantity, explicitly not a resource classification.
An exceedance model — the probability that each node exceeds a stated cut-off, computed across a stack of conditional simulations.
The kernel catalog — every governed compute kernel rendered from its own manifest, with determinism class, execution class, schemas, and the assumption parameters it will not default for you.
The job builder for ordinary kriging — assumption parameters are required, have no pre-filled default, and say why: a silent default is a decision nobody made.
The artifact inspector — what ran, on what inputs, with which parameters, by whom, and under which toolchain pins, read from the artifact's own manifest.
Provenance that reaches back to the database: the tables read, the canonical project, the analyte, and the drilling behind the estimate — labelled as capture evidence, not a claim the kernel makes.
Refusals cross the boundary as structure — code, category, stage, kernel — so an invalid job is explained and actionable rather than answered with an invented number.

Enterprise evaluation

How teams should evaluate GeoMine Scientist

A production evaluation should prove technical fit, governance, deployment model, data readiness, and workflow value before procurement.

Deployment fit

Review cloud, self-hosted, ArcGIS-connected, and open-source GIS integration options before implementation.

Governance model

Validate authentication, data access, audit logging, and review controls with technical stakeholders.

Workflow proof

Walk through desurveying, implicit modelling, kriging, simulation, grade–tonnage reporting, and a full provenance and reproducibility review.

Adoption plan

Define users, training needs, rollout sequence, success metrics, and implementation responsibilities.

Make your resource model reproducible

Talk to SpatialTechSolutions about product fit, implementation scope, integrations, and deployment support.

Request Demo

Request demo

Make your resource model reproducible

Request a tailored walkthrough for GeoMine Scientist, GeoTech Scientist, Hydro GeoScientist, or Plato GIS — resource modelling, ground engineering, mine water, remote sensing, Data Hub workflows, and governed AI.

Request DemoExplore Platform