A Hidden Buried River Full of Gold in Western Australia

Out here, south of Kambalda, the ground looks completely ordinary. Flat, dusty, scrub-covered. Nothing about it suggests gold. No quartz reefs punching through the surface, no old shafts, no diggings scattered across the landscape. And yet, directly beneath your feet, there’s an ancient river system—long buried, totally invisible—that holds hundreds of thousands of ounces of gold. Not mined out. Not fully explored. In places, barely touched. A gold-bearing lead that the old timers never saw, never worked, and never even knew existed. Because this isn’t a reef you can follow, or a gully you can pan. This is a river that disappeared millions of years ago… and took its gold with it.

Most people think of gold deposits in Western Australia as hard rock—quartz veins, shear zones, big open pits chewing into ancient greenstone. And that’s exactly what dominates the Higginsville region. It sits within the Norseman–Wiluna greenstone belt, one of the most gold-endowed geological provinces on Earth. These rocks are Archaean in age—over 2.6 billion years old—and they’ve been folded, faulted, and mineralised multiple times. When geologists talk about “greenstone belts,” they’re referring to sequences of ancient volcanic and sedimentary rocks that have been metamorphosed—basically cooked and compressed over geological time—and are often rich in gold.

But the palaeochannel deposits at Higginsville don’t form in those rocks directly. They sit above them. Much later in time.

To understand them, you have to jump forward from the Archaean—billions of years ago—to the Tertiary period, which spans roughly 66 million to 2.6 million years ago. By this point, those ancient gold-bearing rocks had already been sitting there for an unimaginable length of time, slowly breaking down under weathering. Rain, groundwater, chemical reactions—all of it working to dismantle the bedrock and release whatever metals it held.

Gold, despite its reputation for being chemically inert, can actually move under the right conditions. In oxidising environments—where oxygen is present—gold can dissolve in groundwater as tiny complexes, especially if other chemicals like chloride or organic compounds are around. This process is called “supergene mobilisation,” which just means the movement of metals near the Earth’s surface due to weathering.

So now you’ve got this landscape where gold is being freed from its original host rocks. And at the same time, rivers are flowing across it.

These weren’t the rivers you see today. They were ancient drainage systems—palaeorivers—that cut across the landscape, eroding, transporting, and redepositing material. Over time, these rivers carved channels into the weathered bedrock. And as they flowed, they picked up gold—either as tiny particles eroded mechanically, or as dissolved gold that later re-precipitated.

Where does gold end up in a river system? At the bottom.

Because gold is dense—much heavier than sand or clay—it settles out in low-energy environments. Places where the current slows down. Inside bends. Depressions. The base of the channel. Over time, this leads to the formation of what’s called a “placer deposit.” That’s a concentration of heavy minerals, like gold, formed by mechanical sorting in water.

At Higginsville, that’s exactly what happened. Gold accumulated at the base of these palaeochannels, within sands, gravels, and conglomerates—coarse sedimentary layers that formed in the river bed. According to the data, the mineralisation is specifically located at the base of the Tertiary sediments, sitting just above the weathered Archaean basement .

But here’s where things take a turn.

These rivers didn’t stay exposed.

Over time, the climate shifted. The landscape changed. Sedimentation increased. And those active river systems were gradually abandoned, infilled, and buried. Layers of clay, silt, calcareous soils, and aeolian (wind-blown) sands built up over the top. In some places, you’re looking at tens of metres of cover—up to around 80 metres before you even hit fresh rock .

So now, you’ve got a gold-bearing river system that’s completely hidden.

No surface expression. No visible gravel bars. No exposed leads. Just flat ground.

And that’s why the old timers missed it.

During the gold rush era, prospectors relied on what they could see. They followed quartz veins. They panned creeks. They sank shafts where gold was already present near the surface. Even when they chased “deep leads”—ancient buried river systems—they usually had some kind of surface indicator. A trail of alluvial gold. A line of workings. Something pointing them in the right direction.

At Higginsville, there was none of that.

The gold is too deep. The overlying sediments are transported—meaning they’ve been moved from elsewhere—so they don’t reflect what’s below. And critically, geochemical sampling at the surface doesn’t reliably detect the gold underneath. In many cases, soil samples taken directly above mineralisation show no detectable gold at all .

So even if you knew exactly where to stand, you still might not see anything.

That’s what makes this system so deceptive.

It wasn’t until modern exploration techniques came into play that the Higginsville palaeochannel deposits were discovered. Companies began using systematic soil sampling—looking for subtle, broad anomalies rather than obvious high-grade hits—and then following up with drilling. Drilling is exactly what it sounds like: physically boring holes into the ground to collect samples from depth.

And that’s when the hidden river revealed itself.

The discovery wasn’t based on someone finding gold in a pan. It was based on data. Patterns. A weak but consistent signal that something was happening beneath the surface, spread over more than two kilometres .

Follow that up with drilling, and suddenly you hit it—the basal gravels, the sands, the gold.

Estimates put the palaeochannel deposits at around 300,000 ounces of gold. Not world-class, but significant. And more importantly, completely invisible until you drilled into them.

So what’s been mined?

This is where the story gets a bit more nuanced.

Yes, the palaeochannel has been mined. But not in the way people imagine when they think of a continuous lead being worked end to end.

Instead, mining has focused on the shallowest, most accessible sections. Where the base of the channel—the gold-bearing horizon—is close enough to the surface to be economically reached by open pit mining. Open pit mining involves removing overburden—the material above the ore—using large-scale earthmoving equipment, then extracting the ore from below.

But when your gold is sitting, say, 40 or 60 metres down beneath clays and sediments, the amount of material you have to move becomes enormous. That’s called the “strip ratio”—the ratio of waste to ore. And once that ratio gets too high, the economics fall apart.

So what happens is selective mining.

Certain sections of the palaeochannel—where the gold is shallow, thicker, or higher grade—get mined. Others are left behind. Not because they don’t contain gold, but because they’re not worth mining under current conditions.

That means the palaeochannel isn’t mined continuously along its entire length. It’s mined in patches. Sweet spots. Windows where geology and economics line up.

And beyond those windows, large portions of the system remain in the ground.

Some have been drilled and defined but not mined. Others may extend further than currently known, still hidden beneath cover. The channels themselves are tributaries—smaller systems feeding into a larger palaeodrainage network beneath Lake Cowan—so the full extent of gold distribution is likely more complex than a single continuous lead.

Even within the regolith—the weathered layer above the bedrock—gold behaves in unusual ways. It can associate with carbonates—minerals like calcite that form in soils—and with ferruginous granules, which are iron-rich particles. Some of these granules contain surprisingly high gold concentrations, acting like tiny reservoirs that slowly release gold into the surrounding soil over time .

But again, none of this produces a clean, reliable surface signal.

Which brings us back to the core idea.

This is a gold system that was never “worked out.” Not because it was exhausted—but because it was never visible in the first place.

The old timers didn’t ignore it. They couldn’t see it.

Modern miners have tapped into parts of it. But even now, they’re only scratching sections where it makes economic sense. The rest remains buried, defined in places, inferred in others, and potentially extending further than we currently understand.

It’s not untouched. But it’s far from finished.

And that’s what makes palaeochannel systems like Higginsville so compelling. They sit right alongside some of the most intensively explored goldfields in the world—and yet, they operate under a completely different set of rules. Hidden, subtle, and in many cases, still holding gold where no one has looked… or where it simply hasn’t been worth digging yet.

Here's the video we made on this on the OzGeology YouTube Channel: