TerraSentinel

The problem

A widening gap between mining and the water downstream of it.

Global freshwater demand is on track to outstrip supply, and mining is one of the sharpest, most immediate versions of that problem across Southern and Central Africa. In March 2025, a tailings dam failure at a copper mine in Zambia's Copperbelt released an estimated 50 million litres of acidic wastewater into the Kafue River basin — not an isolated event, but one point in a documented pattern of Copperbelt water contamination. No accessible system today ties ground-truth data, dam instrumentation, and downstream water signals together in one place; each exists in isolation, often in specialist tools smaller operators, regulators, and downstream communities never see.

40%

projected global freshwater demand–supply gap by 2030 (UN / World Economic Forum)

4bn people

already live in water-scarce areas, per World Bank estimates

$6.7trillion

water infrastructure financing gap through 2030 — against <2% of global public spending

The approach

Four layers, correlated — not one signal trusted alone.

Ground truth is the foundation everything else is checked against. Each layer above it extends how far that ground truth reaches, and an alert only means something when layers agree — or when they disagree in a way that's worth a physical look.

04 Drone ground-truth sampling Confirms actual water and soil chemistry — pH, copper, cobalt, cyanide. Satellites can flag that something changed; only a physical sample says what it is.

03 Satellite optical / radar imagery Watches downstream turbidity and vegetation-stress signals — early indicators that something in the water has changed, prioritising where a physical sample is needed most.

02 Satellite radar (InSAR) Tracks tailings dam surface deformation — a real, active early-warning technique that can detect ground movement months before a structural failure — cross-validated, never used alone.

01 Ground truth: boreholes, cores & instrumentation Exploration borehole logs and geotechnical instrumentation — piezometers, inclinometers, seismographs — are the scientific backbone: the actual subsurface picture, not something a satellite can infer alone.

An AI layer correlates all four and raises alerts automatically when they disagree with each other or with expected baselines — aligned with the Global Industry Standard on Tailings Management (GISTM), not a replacement for the Engineer of Record it requires.

What exists today

A working pilot dashboard — not yet an automated one.

A field data-entry tool and a live pilot dashboard already run today, backed by a real database — site data (borehole logs, instrumentation readings, water samples) gets logged by hand and shows up on the dashboard immediately. What's not built yet is the automated layer on top: satellite feeds, drone-triggered sampling, and the AI correlation engine that would flag anomalies without someone entering the numbers first.

This is deliberately not for sale or fundraising yet. It's a search for the right technical and scientific collaborators to help take these pilot tools from manual entry to a fully automated system — even if that means the timeline runs to 2030.

Not a replacement for core drilling, geotechnical instrumentation, or a qualified Engineer of Record.
Not a claim that AI can interpret subsurface geology on its own.
The aim: make existing ground-truth data reach further — flagging where a new sample or reading is needed sooner, especially at smaller, less-resourced sites.

Founder

Built on years of reading what comes back up the hole.

Jean Venter

Founder, TerraSentinel

Also founder — Barter Istokvel

Years of core drilling across the DRC, Zambia, and Botswana — reading what comes back up the hole to understand ground formations, water tables, and rock behaviour below the surface — sit behind TerraSentinel's starting premise: no amount of satellite data replaces the judgment that decides where a monitoring borehole actually gets placed.

Alongside that, ongoing self-directed study spans wildlife ecology, forest management, soil science, sustainable development, and physical geology. TerraSentinel isn't a first attempt at taking an idea from concept to a working product — Barter and Istokvel are both live already.

Contact

Talk through where the gaps are.

Open to a conversation with anyone whose expertise touches this problem — hydrogeology, geotechnical instrumentation, remote sensing, or mine-site operations. Happy to share more detail, or just talk through where this is still thin.

jean@terrasentinel.co.za