How Satellite Intelligence is Actually Opening Up Pakistan's $6 Trillion in Minerals
Last Tuesday I got a call at 11:47 PM from an investor in Karachi. He'd just been offered a copper lease in Chagai for PKR 340 million and wanted to know if it was worth it before he flew out on Thursday. Two days. That's what he had.
Ten years ago I would've told him it's impossible. You can't validate a copper deposit in 48 hours. You'd need trucks, geologists, sample bags, a lab in Islamabad, and about six weeks of patience.
But we pulled it up on satellite. Sentinel-2 bands stacked with ASTER SWIR ratios, SAR for structural mapping, SRTM for elevation context. By Wednesday afternoon he had a 34-page report showing exactly where the alteration halos sat and where they didn't. He didn't buy the lease. Saved him roughly PKR 290 million because the mineralization was thinner than the seller claimed.
That's the shift I want to talk about.
The $6 trillion number nobody knows what to do with
Everyone in Pakistani mining quotes the $6 trillion figure. Reko Diq alone holds an estimated 5.9 billion tonnes of copper-gold ore. Saindak is still producing. Balochistan has chromite. Gilgit-Baltistan has emerald, ruby, gold, and some of the cleanest marble on earth (I own 15 mines up there, so I'm not guessing). KPK has copper corridors that nobody's properly drilled.
And yet — here's the thing — most of this reserve estimate is theoretical. It's based on regional geology, old Soviet-era mapping from the 1970s, and GSP surveys that were done when Zia was still president. The number is real. The exploration data behind it is not.
That's the gap satellite intelligence fills. Not replacing drilling. Not replacing geologists. Just answering the question every executive actually asks first: where should I even be looking?
Honestly, I used to think satellite work was a supplement to ground surveys. I got that wrong at first. After running about 1,200 reports across Pakistan, I've realized it's the opposite — ground work should be a supplement to satellite screening. You use pixels to narrow 10,000 sq km down to 40 sq km. Then you send humans.
What the satellites actually see
Let me be specific because the marketing language in this industry is exhausting.
Sentinel-2 gives us 13 spectral bands at 10-20 meter resolution, free, refreshed every 5 days. For iron oxides, vegetation stress over sulfide bodies, and broad lithology mapping — it's the workhorse. We pull band ratios (4/2 for iron, 11/12 for clay minerals) and overlay them on terrain.
ASTER is the one that actually matters for hydrothermal alteration. Its SWIR bands pick up argillic and phyllic alteration zones — the classic fingerprints around porphyry copper systems. If you're looking at Chagai, Waziristan, or the Kohistan arc, ASTER is non-negotiable.
SAR (Sentinel-1) shoots radar through clouds and darkness. In Gilgit-Baltistan where cloud cover kills optical imaging six months a year, SAR is how we map structural controls — faults, shear zones, the plumbing that hosts orogenic gold.
SRTM DEM gives us 30m elevation data. Sounds boring. But drainage patterns and lineament analysis on a good DEM will show you fault intersections that don't appear on any published geological map.
Stack all four with AI classification and you get something a solo geologist would need three months to produce. We do it in about 72 hours.
What this means if you're writing the cheque
Look, I'm not going to pretend satellite reports replace due diligence. They don't. But they change the economics of exploration in a way that hasn't been properly absorbed by the Pakistani mining sector yet.
A traditional exploration program in Balochistan runs USD 180,000 to USD 400,000 for the first-pass regional survey. Six to nine months. Helicopter time, security escorts, geologist per diems, sample logistics, lab fees in Peshawar or overseas.
A satellite-first workflow costs about 3-7% of that for the screening phase. You still spend the drilling money later — but only on targets that survived the pixel-level filter. For an investor evaluating five leases across KPK and GB, that's the difference between committing PKR 2 billion blindly and committing it with a defensible target ranking.
Three things I'd tell any mining executive reading this:
- Ask for the raw band ratios, not just the pretty maps. If your consultant can't show you the ASTER 4/6, 5/6, 7/6 stack, they're selling you a picture.
- Cross-check satellite anomalies against published GSP mapping and — this matters — against old artisanal mining locations. Where villagers have been digging for 200 years, there's usually a reason.
- Don't skip ground truthing. Ever. Satellite gets you to the coordinate. A hammer and a hand lens confirm it.
Where I think this actually goes
Pakistan has roughly 92,000 sq km of prospective mineral terrain that has never had modern exploration. Not underexplored — unexplored. The Karakoram, the Chagai arc, the Waziristan ophiolite, the Chitral belt.
With current satellite constellations and the AI models we've trained on Pakistani geology specifically (not generic global models — that's another thing I got wrong early), you can now do a first-pass mineral prospectivity map of an entire district for less than the cost of a Land Cruiser.
The country's mineral wealth isn't hidden anymore. It's sitting in freely available Copernicus and NASA archives waiting for someone to run the right analysis on it.
So the real question isn't whether the $6 trillion exists. It's who moves first — and whether Pakistani companies do it before the Chinese, Saudis, and Australians finish doing it for us.
What would you do with 48 hours and a satellite feed?