Bauxite in Gilgit-Baltistan: What the Grade Sheets Actually Say

Most buyers who call us about bauxite assume Pakistan only has the Khushab and Muzaffarabad belts. That's the textbook answer. It's also incomplete.

We hold concessions covering bauxite-bearing horizons in Gilgit-Baltistan that don't appear in the standard GSP summaries from the 1980s. Some of it is karst bauxite sitting on Permian-Triassic carbonates. Some is lateritic, weathered out of aluminous shales along the northern margin of the Indian plate where it shoulders into the Kohistan arc. The grade story is mixed — and I want to be honest about that before anyone wires money.

What we're actually pulling and what it assays

Our sampling programme across two concessions (one near the Chilas-Astore corridor, one further north on the Hunza side) has returned Al₂O₃ values ranging from 41.2% to 53.8%. The better intervals — and I'm talking channel samples, not grab — sit between 48% and 52% Al₂O₃, with SiO₂ between 4.1% and 7.9%, and Fe₂O₃ between 14% and 22%.

That's metallurgical-grade territory for most of it. Not premium. Not Guinea CBG. But honest Bayer-process feedstock if the silica behaves.

The reactive silica is the thing I watch. Total silica is one number. Reactive silica — the kaolinite and the fine quartz that actually consumes caustic in the digester — is what determines whether a refinery wants your tonnes or not. On our Chilas-side samples reactive SiO₂ runs around 3.2 to 5.6%. Workable. On the lateritic horizons further north it climbs, and some intervals we've already written off for Bayer feed and are looking at as refractory or cement-grade material instead.

LOI sits around 24-27% on the gibbsitic portions, which is what you'd expect.

Honestly, I got the mineralogy wrong at first. I assumed everything would be boehmite-dominant because of the depth of burial and the structural history. Turns out the karst pockets are gibbsite-rich, which is good news — gibbsite digests at lower temperatures and refineries in the Gulf and China can handle it without re-tooling.

Tonnage, geology and what's still inferred

I won't quote a JORC or NI 43-101 figure because we don't have one yet. What we have is mapped surface expression across roughly 14 km of strike on one concession, with thicknesses in the bauxite horizon ranging from 3 m to (in one karst pocket) about 19 m. Exploration target — and I'm calling it that deliberately, not a resource — is in the range of 8 to 14 million tonnes across the two main concessions, at grades that should average above 46% Al₂O₃ after selective mining.

A proper drilling programme would tighten this fast. We're in conversation with two parties about funding that, and the JV structure we use for the copper assets translates cleanly to bauxite (covered that elsewhere, won't repeat it here).

The geology itself is interesting because the bauxite-bearing carbonate sequence has been folded but not metamorphosed to a degree that destroys the ore. You can walk the contact in several places. For anyone who's worked Mediterranean karst bauxites — the Greek deposits, parts of southern France, Montenegro — the setting will feel familiar. Same paleo-weathering story, different tectonic overprint.

Why this matters for aluminium feedstock buyers right now

Guinea supplies roughly 23% of seaborne bauxite. China imports something like 158 million tonnes a year and the concentration of supply makes their refiners nervous — I've had three separate Chinese trading houses tell me as much in the last eighteen months. The Indonesian export ban on raw bauxite (still in force, despite the back-and-forth) hasn't helped.

Gulf refiners — EGA in particular — have been quietly diversifying. EU buyers under the Critical Raw Materials Act are looking at bauxite and alumina with fresh eyes after it was added to the strategic list.

We're not pretending Pakistan competes with Guinea on scale. We can't. But for a buyer who wants 200,000 to 800,000 tonnes a year of mid-grade gibbsitic bauxite from a non-aligned source with reasonable shipping economics to Chinese east coast refiners or Gulf smelters, the conversation is worth having.

Getting it to a ship

Here's the part everyone asks about and nobody wants to think through carefully.

Gilgit-Baltistan is landlocked and the road down to Karachi via the Karakoram Highway is roughly 1,600 km. That's the honest constraint. Bauxite is a low-value, high-volume commodity. You can't move it like copper concentrate and expect the economics to work.

So we've looked at three routes:

1. KKH down to Karachi for sea export — works for trial cargoes and for higher-grade selective material, doesn't work for 500kt/year unless rail upgrades happen.
2. KKH north into Xinjiang via Khunjerab — politically simple under CPEC framing, seasonally limited (the pass closes November to April), and trucking costs aren't trivial.
3. Beneficiation and partial calcination at site or at an intermediate plant near Havelian, shipping a higher-value product. This is where I think the real answer sits for any serious off-take volume.

A buyer who's willing to co-invest in a calcination or washing plant changes the unit economics completely. We've modelled it. The numbers work above roughly 300,000 tpa of feed.

Licensing-wise, GB mineral concessions are issued by the GB Mineral Department under the 2003 rules (amended). Export of bauxite is permitted — there's no ban, unlike chromite in certain periods — and royalty rates are published. Foreign equity in the JV is straightforward; the structures we use for copper apply.

If you're a refiner or trader sitting on a desk in Shanghai, Dubai, Rotterdam or Tokyo and bauxite supply concentration is on your risk register, send me your spec sheet. Reactive silica tolerance, Al₂O₃ minimum, available-alumina target, and the volume you actually need. We'll tell you which of our horizons match and which don't.

And if they don't match — I'll say so.