Marble and Granite Exploration in Pakistan: What ASTER and DEM Data Actually Tell You

Last March I spent four days driving from Skardu to a marble quarry near Shigar that two of my partners were ready to invest 80 million rupees into. The rock looked beautiful on the surface. Creamy white, decent grain. But the ASTER scene I'd pulled the week before told a different story — the carbonate index was inconsistent across the body, and the DEM showed the deposit sitting on a structurally weak ridge with serious fracture density.

We walked away. Saved them roughly 80M PKR on a deposit that would've cracked into unsellable blocks within the first cutting season.

That's the gap satellite data fills for dimension stone in Pakistan. Most marble and granite buyers in this country are still scouting deposits the way their grandfathers did — word of mouth, a few field visits, maybe a small test cut. And then they're surprised when 60% of what they extract ends up as gravel instead of slabs.

Why ASTER Is Built for Carbonate and Silicate Rocks

ASTER has 14 spectral bands. For marble and granite work, we care about the thermal infrared (TIR) bands the most — bands 10 through 14. Marble is essentially metamorphosed limestone, so it has a strong calcium carbonate signature. Granite is silica-dominant, quartz and feldspar. Both rocks scream their composition in the thermal range if you know how to listen.

The carbonate index (ratio of band 13 to band 14) lights up marble bodies clearly. We've mapped large carbonate zones across Buner, Mohmand, Khyber, and parts of Chitral using nothing but free ASTER data and a decent processing pipeline. For granite, the quartz index (a more complex ratio involving bands 11, 12, and 13) does similar work.

But here's the thing — finding the rock isn't the hard part. Pakistan has marble and granite practically falling out of the mountains. The hard part is finding commercial-grade deposits. That means:

• Large continuous blocks with low fracture density
• Consistent color and grain
• Accessible terrain (you can't quarry a 70-degree cliff face profitably)
• Minimal weathering and staining

ASTER alone won't tell you all of this. That's where DEM data comes in.

What SRTM DEM Tells You That Spectral Data Can't

SRTM gives us 30-meter elevation data globally. For free. And honestly, I think most explorationists in Pakistan underuse it badly.

When we process a DEM for marble or granite targeting, we're looking at four things:

Slope analysis. Anything above 45 degrees is essentially uneconomic for block extraction in Pakistan because of access, drilling angles, and block recovery rates. We mask out steep zones immediately.

Lineament density. This is the big one. A lineament is essentially a fracture or fault line visible in the topography. High lineament density means the rock is broken up underneath — useless for dimension stone, fine for crushed aggregate. Low lineament density over a confirmed carbonate or granite spectral signature? That's where you drill your test holes.

Drainage patterns. Dendritic drainage suggests homogeneous bedrock. Rectangular or trellis patterns suggest heavy structural control — which means fractured rock.

Accessibility modeling. We calculate cost-distance from existing roads. A perfect marble deposit 14 km from the nearest road through a 2,800m pass is worth less than a B-grade deposit 800m off the Karakoram Highway.

How We Actually Combine This for Pakistani Deposits

Our workflow at GeoMine looks roughly like this for a marble or granite target area:

First, ASTER TIR processing to identify carbonate or silicate anomalies across the area of interest. We're typically scanning blocks of 500 to 2,000 square kilometers at this stage. The output is a heatmap of probable lithology.

Second, Sentinel-2 visual confirmation. ASTER tells us composition, Sentinel-2 (with its 10m resolution) tells us what the surface actually looks like — color, exposure, vegetation cover that might be hiding something.

Third, DEM-based structural analysis. We overlay the lineament density map on the spectral anomalies. The sweet spots are where high carbonate or quartz index meets low fracture density.

Fourth, accessibility and slope filtering. We drop anything uneconomic.

What's left is usually 3 to 8 high-priority targets out of an initial area that might have shown 40+ spectral hits. That's the difference between a useful report and a 200-page document nobody reads.

I'll admit — I got the weighting wrong on our first dozen marble reports. I was over-indexing on spectral purity and under-indexing on structural integrity. Two clients drilled targets I'd marked as A-grade and came back with heavily fractured rock. Embarrassing. We rebuilt the scoring model after that, and now structural data carries almost equal weight to spectral data for dimension stone work.

Where This Actually Matters in Pakistan

Buner alone has marble reserves estimated above 3 billion tons. Most of it has never been properly surveyed. Same story in Mohmand, in Chiniot's granite belt, in the Nagar Parkar granites of Sindh, and across huge stretches of Gilgit-Baltistan where I work personally.

The industry standard right now is to lease land, dig a test pit, hope for the best. And then watch margins evaporate when block recovery comes in at 18% instead of the 35% needed for profitability.

A proper satellite-based pre-screen costs less than the diesel for one week of test extraction. I don't understand why more operators haven't figured this out yet — though I suspect it's because most geological consultants in Pakistan still treat remote sensing as something academic rather than something you'd bet a quarry lease on.

If you're looking at a marble or granite property right now and the only data you have is what the seller told you, you're gambling. Pull the ASTER scene. Run the DEM. Or have someone do it for you before you sign anything.

What's the actual cost of finding out after the fact that your deposit is structurally compromised?