Pakistan's $6 Trillion Mineral Story: What Satellites Are Actually Telling Us

By Sufyan · 2026-07-29 · 5 min read

Six trillion dollars. That's the number the Ministry of Petroleum keeps repeating, and honestly, I used to roll my eyes at it. Then I started flying drones over my own leases in Gilgit Baltistan and cross-checking what Sentinel-2 was picking up from 786 km overhead. The numbers stopped feeling like a slogan.

Here's the thing most executives miss: Pakistan doesn't have an exploration problem. It has a targeting problem. We've got roughly 600,000 sq km of prospective ground and maybe 200 serious exploration geologists working it. Do the math. A field team can cover 4-6 sq km of detailed mapping in a day if the terrain is kind. In Chagai or Kohistan, it's not kind.

Satellites don't care about terrain.

Why the old exploration model is broken in Pakistan

I spent my first two years in this industry watching companies burn $180,000-$250,000 on a single grassroots exploration season and come back with a report that basically said "looks interesting, need more work." One junior I know spent 14 months getting road access to a target in Diamer before a single hammer hit rock. By the time they drilled, the copper price had moved and their board pulled funding.

That's not a geology failure. It's a sequencing failure.

Satellite mineral exploration Pakistan flips the order. You start with data that's already been collected — Sentinel-2 for iron oxides and vegetation stress, ASTER mineral detection for clay and carbonate alteration, SRTM DEM for structural controls, SAR for surface roughness and fault mapping. You spend maybe $1,500-$4,000 generating a proper targeting report. Then you send the field team to the 3-4 highest-ranked anomalies instead of a 400 sq km block.

The cost ratio is roughly 60:1 in favor of doing satellite work first. And yet.

A surprising number of Pakistani mining executives I talk to still treat remote sensing mining as "something for later, after we prove up the resource." That's backwards. Remote sensing tells you where to prove up.

What the satellites are actually finding

Let me give you specifics because vague claims annoy me too.

In the Chagai arc — the same belt that hosts Reko Diq and Saindak — ASTER band ratio 4/6 lights up argillic alteration like a highway at night. We ran a study on a 240 sq km area south of the main Reko Diq footprint and pulled out 11 alteration zones that weren't on any published geological map. Two of them sit directly on structural intersections that any porphyry geologist would want to walk.

For copper gold lithium exploration in the north — think Kohistan Island Arc, parts of Waziristan, and the pegmatite fields around Skardu — the workflow is different. Lithium-bearing pegmatites are small (often under 100m wide) but they produce distinct spectral signatures because of the muscovite and spodumene content. Sentinel-2 at 10-20m resolution catches them if you know what ratios to use. I won't pretend it's easy. False positives are real. But narrowing 5,000 sq km down to 40 candidate outcrops before you ever pay for a helicopter? That's the game.

Chromite in the ophiolite belts of Muslim Bagh and Waziristan shows up through a combination of low reflectance in visible bands and specific ASTER SWIR responses. Marble and granite mapping in Gilgit Baltistan — which is what pays a lot of small operators' bills — works even better because the outcrops are usually clean and large.

Gold's the tricky one. You don't detect gold directly. You detect the alteration halo, the pyrite oxidation, the sericite, the silicification. AI geological survey work is where this gets interesting because a trained model can weigh 15-20 spectral and structural indicators simultaneously. A human geologist can do it too — just not for 600,000 sq km.

What investors should actually ask before writing a check

I've sat in enough investor meetings to know the pitch deck template. "Highly prospective ground, adjacent to known deposits, experienced team." Fine. But if I'm putting real money into a Pakistan mining investment right now, here's what I want to see:

If a promoter tells you satellites can "see" a deposit at 200m depth, walk out. They're lying or they don't understand the technology. What geo mining data does is dramatically improve your odds on where to drill. It doesn't replace the drill.

Look, I own 15 mines in GB. I've made every mistake in this business — bought a marble lease once where the satellite signature was gorgeous and the outcrop turned out to be a 4m veneer over schist. Cost me about 22 lakhs to learn that lesson. So I'm not selling magic. I'm selling better odds and faster decisions.

Where this goes next

The piece nobody's talking about yet is repeat-pass monitoring. Sentinel-2 revisits every 5 days. That means we can watch illegal mining, track lease compliance, monitor tailings, and — for investors — verify that a company is actually doing the work they claim between site visits. The Punjab and Balochistan mines departments haven't fully caught up to this yet, but they will.

The other piece is scale. Once you've built the pipeline for one lease, running it across 50 leases costs maybe 3x, not 50x. That's the economics that make platforms like GeoMines viable in a country where individual mine owners can't afford a $40,000 consulting engagement but can absolutely afford a $2,000 report.

So when someone quotes you the $6 trillion figure, the honest answer isn't "yes, that's real" or "no, that's hype." The honest answer is: we finally have the tools to find out. Which parts of that number are you willing to go look for?

© 2026 GeoMine AI. All rights reserved.