How LoderaIQ interprets satellite, drone and spectral data with AI to prioritise where your field teams should walk, sample and drill next.
The Exploration layer of LoderaIQ is a target-generation engine. It ingests wide-area satellite coverage and, where flown, high-resolution drone imagery, then applies spectral analysis and pattern recognition to surface the places most worth a boots-on-the-ground visit. Every anomaly it flags is written to the same shared registry, map and audit trail as the rest of the platform, so an exploration lead can trace exactly which pixels, indices and dates produced a given target. This page explains the mechanism: what the imagery measures, how targets are ranked, how surveys are planned around them, and the firm line between an AI target and a defined resource.
Request a briefingOne registry, three data streams, one interpretation pipeline
Exploration draws on three complementary streams. Satellite imagery gives wide-area, repeat awareness across a whole licence or programme, cheaply and frequently. Aerial drone survey, now rolling out, adds centimetre-scale ground truth over the priority zones the satellite layer flags. Spectral data from both sources is where the geology emerges: the way surfaces reflect and absorb light across visible, near-infrared and shortwave bands carries a signature of the minerals present. AI does not invent geology from this; it applies established band-ratio and index methods at a scale and speed no manual workflow can match, and hands the results back as candidate targets for a geologist to judge. Because every layer shares one registry, a spectral flag, a later drone pass and an eventual field assay all attach to the same map feature with a full history.
What spectral analysis actually measures
Reflectance across the spectrum indicates surface mineralogy and alteration. It indicates; it does not assay.
Different minerals absorb and reflect light in characteristic ways across the visible-to-shortwave-infrared range. LoderaIQ computes band ratios and spectral indices from that reflectance to highlight four things that commonly accompany mineralisation, and reports them as indications to be checked, never as confirmed grade.
Surface mineralogy
Band ratios respond to iron oxides, clays, carbonates and other rock-forming and secondary minerals exposed at surface, mapping their likely spatial distribution across the licence.
Hydrothermal alteration
Argillic, phyllic and other alteration haloes often ring a mineralised system. Their spectral signatures can be mapped from imagery, helping outline the footprint of a potential system rather than a single point.
Vegetation stress
Where cover obscures bedrock, subtle stress and geobotanical anomalies in the vegetation canopy can hint at underlying geochemistry, giving an indirect read in areas the naked rock never reaches.
Structural context
Imagery also resolves lineaments, faults and contacts that control where fluids moved and minerals concentrated, so anomalies can be weighed against the structural setting rather than in isolation.
From raw anomaly to ranked target
How the platform decides what deserves a field visit first
A single bright pixel is not a target. LoderaIQ builds and ranks targets by combining multiple lines of evidence, so the list your team receives is ordered by strength of indication and ease of follow-up, not by chance.
Detect
Spectral indices are computed across the full coverage area and multiple acquisition dates, flagging pixels and clusters that depart from the local background.
Corroborate
Candidate anomalies are cross-checked against alteration mapping, structural features and, where available, prior geochemistry, so single-source artefacts drop out and multi-evidence zones rise.
Rank
Surviving targets are scored on the coincidence and strength of their indicators and their spatial coherence, producing a prioritised, defensible queue rather than an undifferentiated heat map.
Plan
Top-ranked targets become the basis for survey design, feeding directly into where drone lines are flown and where sampling traverses are laid out.
Planning the ground and aerial follow-up
Targets are only useful if they route field effort efficiently
Ranked targets drive a concrete survey plan. The platform turns priority zones into aerial and field campaigns that spend scarce time and budget where the evidence is strongest, and records the plan and its results against the same map features for a continuous exploration history.
Drone survey design
For high-priority zones, the aerial layer, now rolling out, will plan flight coverage end to end, route, waypoints, altitude and per-sensor triggers, to acquire centimetre-scale imagery and finer detail than orbit allows. One integrated payload will carry it: high-resolution cameras, a magnetometer designed to map magnetic anomalies, structures and alteration that help target mineralisation, and LiDAR to capture precise terrain and volumes, with live in-flight control to retask any sensor on the fly.
Sampling traverses
The platform suggests where physical sampling and field-portable XRF should be concentrated, so geochemical effort clusters on the strongest indications rather than a uniform, wasteful grid.
Revisit and re-rank
Because coverage is delivered as a continuously updated service, programmes can revisit targets as new imagery arrives and re-rank them over time instead of committing to fixed infrastructure.
Full traceability
Every target, flight line and sample point lives in one registry with a dated audit trail, so the path from first spectral flag to field result is auditable end to end.
A target is not a resource, and a resource is not a reserve
The most important boundary on this page
AI-identified anomalies are target generation for follow-up. They are not resource definition, and nothing on this platform changes that. Spectral imagery indicates surface mineralogy and alteration; it is not an assay. A defensible grade and tonnage come only from physical sampling and field-portable XRF, tied to a properly designed sampling programme. LoderaIQ outputs are inputs to the exploration decision, and they support the work of a competent person or qualified person under JORC and NI 43-101-style codes; they do not replace it and cannot be reported as if they did.
An AI-ranked anomaly
A location where coincident spectral, alteration and structural indications justify ground follow-up. It carries no grade, no tonnage and no code-compliant status.
Established by sampling & assay
A mineral resource is established by physical sampling, assay and geological interpretation, classified under a recognised code and signed off by a competent or qualified person. Imagery contributes to targeting; it does not define the resource.
The mineable, modified portion
A reserve is the economically mineable, modified portion of a resource, demonstrated through technical and economic study. It sits well beyond the reach of any imagery or AI output and is not something this layer claims to produce.
Who this layer is built for
This page explains the mechanism. See the audience page for the business case.
The Exploration layer is designed around the workflows of mineral explorers and their competent-person sign-off. For how it changes the economics and cadence of a real exploration programme, follow the link below.
See the exploration workflow on your ground
We can walk your team through how satellite, drone and spectral data become ranked targets over a licence area, and where the platform stops and your competent person begins. The aerial drone-survey layer is rolling out now and can be scoped into the conversation.
Request a briefing