How Gold Forms: From Stellar Collisions to Minable Deposits

The gold in a wedding ring was forged in the wreckage of a stellar collision, billions of years before the Earth existed. It travelled across space, arrived on a young planet by asteroid impact, then waited hundreds of millions of years more while geological processes concentrated it into rock worth digging up.

That strange timeline matters more than most investors realise. Exploration companies publish drill results constantly, and the temptation is to judge them on grade alone: a big number looks good, a small number looks weak. Formation type, deposit geometry, and metallurgy determine whether that number means anything at all.

Understanding how gold gets into the ground is prerequisite knowledge for reading those results intelligently. After this, you will know why the same headline grade can mean opposite things depending on deposit type, the difference between a Carlin-type intercept and a vein intercept, and what geological conditions actually put mineable gold in the ground in the first place.

Gold did not form on Earth at all

Gold is a heavy element, and ordinary stars cannot make it. The fusion reactions that power a normal star run out of energy long before they reach elements this heavy, so gold requires something far more violent to exist.

That violence comes from rapid neutron-capture, known as the r-process, a form of nucleosynthesis that occurs only in the most extreme environments the universe can produce. Atomic nuclei absorb neutrons so quickly they build into heavy elements before they can decay. Gold, platinum, and other heavy metals are the result.

Scientists now point to two dominant sources for this material:

The 2017 observation of GW170817 was a turning point for r-process science, but the broader story of cosmic gold creation extends to collapsars and the competing models that still debate each source’s relative contribution to the universe’s heavy-element budget.

• Neutron-star mergers: collisions between the ultra-dense remnants of dead stars, confirmed as gold factories by direct observation
• Collapsars: massive, rapidly spinning stars that collapse into black holes, with some models suggesting they supply more than 80% of the universe’s r-process material
• Supernovae: once thought to be a primary source, now understood to contribute only a minor fraction of the cosmic gold budget

The clearest proof arrived in 2017.

The binary neutron-star merger GW170817 was observed directly, and the ejected matter confirmed that these collisions synthesise heavy r-process elements including gold and platinum. It moved gold’s cosmic origin from theory to observation.

None of that explains how gold reached a depth you can drill. During Earth’s formation around 4.5 billion years ago, gold and other dense metals sank toward the core through gravitational differentiation, leaving the crust stripped of them. The gold accessible today is widely attributed to late accretion: metal-rich chondritic material delivered after the core had already solidified.

The classical model holds that asteroid bombardment roughly 3.8 to 4.1 billion years ago delivered 0.3-1% of Earth’s mass in this material to the crust and mantle. The science is genuinely contested. Hybrid models now suggest the mantle’s gold reflects both retained core-formation metal and a more modest late accretion, rather than bombardment alone.

The practical takeaway holds regardless of which model wins. Earth’s crustal gold budget was fixed billions of years ago. Exploration can locate and define that endowment, but it cannot expand it, which is the foundational scarcity underpinning everything gold is worth.

How tectonic forces turn parts-per-billion traces into mineable ore

Gold is everywhere and nowhere. Spread evenly through average crustal rock, it sits at concentrations measured in parts per billion, so dilute it is effectively invisible and utterly worthless to extract.

An economic deposit needs that background concentration multiplied by factors of hundreds to thousands. That is the concentration problem, and solving it is what separates a mine from a mountain of ordinary rock. Geology took millions to hundreds of millions of years to do it.

What makes hydrothermal gold precipitate

The mechanism is hydrothermal circulation, driven by heat. When a body of magma intrudes into the crust and begins to crystallise, it heats surrounding groundwater and releases its own metal-bearing fluids.

Gold travels in these hot fluids as aqueous sulfur complexes, chemical structures written as Au(HS)₂⁻ or AuHS, in which gold atoms bond with sulfur and hydrogen to stay dissolved. The fluid can carry gold for long distances while conditions stay stable.

Precipitation happens when that stability breaks. The complexes fall apart, and gold drops out of solution. The triggers are specific:

• Temperature change as fluids cool
• Pressure drops during ascent through the crust
• Boiling and vapour loss
• Sulfidation, where fluids react with iron in the surrounding rock
• Fluid mixing with cooler groundwater
• Direct reaction with wall rocks

Precipitation is a disruption event, not a slow drift. That is why ore zones tend to be geometrically discrete rather than diffuse: gold drops where the chemistry breaks, and nowhere else.

Hydrothermal vein formation is not a single process but a family of related mechanisms, and the differences between pressure-driven decompression, boiling, and fluid mixing produce measurably different ore geometries, metal zoning patterns, and grade distributions within a single deposit.

This is also why genuine discoveries are rare. The right fluid chemistry, the right rock chemistry, and the right structural pathway all had to coincide in the same place at the same geological moment. When you understand that ore requires multiple conditions to align, you evaluate exploration results differently. A high grade in the wrong tectonic setting raises questions that a modest grade in the right setting does not.

The three deposit families and what each means for miners

Geology solved the concentration problem in more than one way. Three broad deposit families dominate global gold production, and each is a genuinely different solution with its own geometry, grade profile, and economic logic.

The differences matter to you because they mean the same headline grade signals very different things depending on which family produced it.

The first family is vein and orogenic gold. Hydrothermal fluids deposit gold inside fracture networks, following pre-existing faults rather than forming broad tabular bodies. The result is high grade but geometrically awkward.

Orogenic deposits, a major subtype, form in metamorphic belts during ancient mountain-building episodes and hold some of the world’s largest gold endowments. As deep fluids rise and pressure drops, they separate into vapour and brine, destabilising the gold complexes and precipitating gold in quartz-carbonate veins.

The second family is placer and paleoplacer. This is mechanical concentration, not chemical. Gold weathers out of bedrock, streams carry it, and because it is exceptionally dense, it settles in stream bends and bedrock hollows while lighter material washes away.

Active stream placers triggered the great gold rushes and still act as pointers to bedrock sources upstream. Ancient placers buried and hardened into rock become paleoplacers, and the defining example is staggering.

The Witwatersrand Basin in South Africa, a conglomerate-hosted paleoplacer that formed roughly 2.7 to 3.1 billion years ago, produced approximately 50,200 tonnes of gold from 1887 to 2019, an estimated 30-40% of all the gold ever mined. A similar amount is thought to remain underground or in tailings.

Disseminated and magmatic-hydrothermal systems

The third family is the most economically counterintuitive. Carlin-type disseminated deposits contain gold at a submicroscopic scale, locked inside sulfide minerals like pyrite and arsenopyrite, invisible to the eye and spread through enormous volumes of rock.

Named after the Carlin Trend in Nevada and formed roughly 35 to 40 million years ago, these deposits do not rely on boiling. Precipitation is driven by sulfidation, where fluids react with iron in sedimentary wall rocks, and by fluid-rock reactions that alter carbonate host rocks into traps. They demand bulk-tonnage, open-pit mining because no single spot is rich enough to chase selectively.

Porphyry and epithermal systems complete the picture. Porphyry deposits form at depth around large intrusions, with gold as a secondary metal alongside copper. Epithermal deposits form at shallow levels, and a single magmatic event can produce both on one vertical profile, which is why gold clusters along the subduction zones of the Pacific Ring of Fire.

Subduction zone gold formation drives the clustering of porphyry and epithermal deposits along the Pacific Ring of Fire, where repeated cycles of magma ascent, volatile exsolution, and hydrothermal circulation have stacked multiple deposit generations on the same vertical profiles.

For scale, the Carlin Trend had produced 98.4 million ounces by the end of 2022, up from 82.4 million ounces at the end of 2013, with industry consensus estimates indicating it passed a lifetime total of 100 million ounces by 2025. Knowing which family a deposit belongs to tells you more about its likely economics than the grade number does, because geometry, mining method, and metallurgical difficulty are all baked in at formation.

What deposit geology actually tells an investor reading drill results

Now the geology becomes a tool. A drill result is a headline grade attached to a width, and without deposit context, that pairing is close to meaningless.

Start with grade thresholds, because what counts as economic depends entirely on the mining method the deposit demands.

The single most useful habit is to stop reading grade in isolation and start reading gram-metres, the grade multiplied by the intercept width. Below 10 g·m is speculative territory. Above 50 g·m is significant.

A continuous 0.5-1.5 g/t intercept over tens of metres in a Carlin-type setting can hold far more contained gold than a 30 g/t hit over less than half a metre in a vein. The eye-catching number is not always the better result.

Then there is metallurgy, which can quietly destroy otherwise attractive economics. Carlin-type and disseminated ores are frequently refractory, meaning the gold is locked in sulfides or carbonaceous material and ordinary cyanide leaching does not work. Double-refractory ores are worse still, because the carbon actively adsorbs dissolved gold during processing.

Refractory ore needs pre-treatment before it can be leached, and that pre-treatment is expensive:

• Pressure oxidation at Nevada Gold Mines’ Goldstrike facility runs approximately US$34 to US$52 per tonne, averaging around US$44 per tonne
• Roasting averages around US$33 per tonne
• Bio-oxidation heaps cost an estimated US$4-6 per tonne, cheaper than autoclaves but still a premium over simple oxide leaching

A 1-2 g/t Carlin-type intercept that looks fine on paper can become uneconomic once those processing costs are applied. So when the next drill result lands, work it in three steps:

1. Identify the deposit type and geometry to understand what you are actually looking at.
2. Apply the correct grade threshold for the mining method that deposit requires.
3. Check for metallurgical complexity flags, refractory or double-refractory mineralogy, that force a discount.

A grade without that context is an incomplete piece of information. You now have the framework to ask the questions that turn a headline number into a genuine signal.

Reading the ground beneath the gold price

The whole chain connects. Gold synthesised in stellar collisions through r-process nucleosynthesis, delivered to a young Earth by late accretion, concentrated over geological time by hydrothermal systems, and given its final geometry by tectonic setting and deposit type.

That last link is where formation science becomes investment analysis. Deposit type determines mining method, mining method sets the grade threshold, and the threshold decides whether a drill intercept is signal or noise.

The geography of production makes the point concrete. Per USGS estimates for 2025, the leading producers are:

• China, approximately 380 tonnes
• Russia, 310 tonnes
• Australia, approximately 280 tonnes
• Canada, 200 tonnes
• United States, 160 tonnes

Global mine production estimates diverge, with the USGS putting 2025 output at roughly 3,300 tonnes and the World Gold Council’s preliminary figure at 3,672 tonnes, a gap that reflects timing and revision rather than certainty. That production is concentrated in ancient cratons and along active volcanic arcs, not by accident but because those are the places where the right conditions for each deposit family once existed.

As known deposits deplete and new discovery grows harder, understanding what geological conditions produce a viable deposit becomes more valuable, not less. Connect tectonic setting to deposit type to production geography, and jurisdiction risk stops being a headline and starts being something you can reason about.

For readers wanting to extend this framework into the supply picture, our full explainer on global gold reserves examines how reserve estimation methodologies interact with deposit type, price assumptions, and depletion rates to shape the long-run supply outlook.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions.