The pillow lavas look like a bunch of stone beanbags, collapsed one on top of the other where the molten rock collided with cold seawater and hardened on the spot. Those on display in a stretch of the Pilbara in Western Australia have remained there, largely undisturbed, since long before anything on this planet had a leaf or a core. Rock this old is usually a lost cause. Three billion years of heat and pressure alter the lava’s original chemistry until it becomes guesswork. The Pilbara Craton largely dodged that fate, with a study published in Nature Communications describing it as the least altered surviving crust over 2.8 billion years old anywhere on Earth. Cooling joints, gas bubbles and pillow outlines can still be seen on the outcrop, which for a rock of that age is almost miraculous. So an international team led by University of Adelaide geochemist Eric Vandenburg went out with hammers and sampled a ten kilometer-thick slice. What the hammers found The target was the Whundo Group, a pile of volcanic rock deposited between 3.13 and 3.1 billion years ago. Approximately 30 million years of eruptions, preserved in sequence like pages in a book that no one had read carefully. By sampling in small spaces, the team obtained three different families of lava: tholeiites, calc-alkaline basalts and boninites. That third one is the interesting one. Boninite is a rare, water-rich, magnesium-rich lava named after the Bonin Islands in the western Pacific, and today it erupts almost exclusively where one tectonic plate crushes beneath another. The Whundo examples are the oldest known extensive boninites. The number that changed history Water does something specific to the rock in the mantle. It lowers the melting point, pretty much the same way a pinch of salt turns road ice into slush on a day that should be too cold to do so. In a paper in The Conversation, researchers used exactly that comparison to explain why volcanoes cluster in subduction zones and not anywhere else. Work backwards from the chemistry and you can estimate how wet the source rock must have been. The team’s calculations put the mantle that fed the Whundo boninitas at between 0.8 and 1.5 percent water by weight. The primitive mantle is close to 0.11 percent. The depleted mantle is closer to 0.01. The mantle rock beneath a modern-day arc volcano is between 0.1 and 2 percent, putting the old figure squarely within the current range, and the paper emphasizes that its estimates are minimal. Water in that quantity does not appear in the mantle by accident. Subduction without plates So how do you get seawater at that depth without plate tectonics? Conventional subduction was almost certainly not yet available. A hotter young Earth meant a weaker, more ductile crust, and geodynamic models have long held that slabs sliding cleanly beneath each other would have been mechanically very difficult to support so far back. The alternative is known as trickling, and was developed through several modeling studies, including the work of Oliver Nebel and his colleagues at Precambrian Research. Imagine a section of flooded crust that becomes dense enough to sink, then subside, sinking into the warmer mantle below as a twisted mass rather than a rigid descending sheet. No continuous plate limit, no conveyor belt, nothing built to last. Gravity does locally, in short bursts, what plate motion does today globally. As confusing as it may seem, Whundo’s chemical result looks remarkably similar to the real thing. Fluids released by the sinking crust supplied up to 93 percent of certain trace elements in the mantle source that then melted to form the boninites, according to the team’s mass balance calculations. The pattern repeats itself throughout the sequence, meaning that whatever was happening kept happening for tens of millions of years. A piece of rock, an article A study is a study, and this team says it clearly. The work covers a single volcanic package of about 114 square kilometers. None of those involved claim that modern plate tectonics kicked in in the Pilbara 3.1 billion years ago, and the paper begins by acknowledging that the arc volcanism of the Archean eon, the stretch stretching from about four to 2.5 billion years ago, remains controversial evidence for subduction. There is also compromise in modeling. The study clearly points out that their preferred explanation is not the only one that fits, and offers a second scenario that reproduces the same boninite chemistry by a different route. Zenodo contains complete data sets for anyone willing to try a third party. What the evidence supports is more limited and even substantial: at one well-preserved site, surface water reached meltwater depths and formed arc-shaped volcanoes long before the machinery normally attributed to such work existed. The crust that isn’t there Between 50 and 90 percent of the evolved crust that models say the Archean should have produced is missing from the record. Cratons older than 2.8 billion years, the ancient cores of continents, make up about seven percent of today’s continents. The University of Adelaide’s announcement framed the finding as evidence that key planetary processes were already at work early. Vandenburg and his colleagues go further, suggesting that the drip conduit of the constructed crust was thin and chemically primed for destruction, which is precisely why so little survived. Continue to the end and the rock record starts to feel like a survivor bias problem. What is still left here is what was too thick, too floating, or too lucky to be swallowed. Whundo may be less a discovery than a rare witness, and the process he records could have been the normal state of things on a planet that ate almost all of its own evidence. About this articleThis article is for general information and reflection. It is not professional advice. For your specific situation, consult a qualified professional. A personal measure of deep time. Look what has changed in the universe since you were born. Start with a date. Get a personal, detailed journey through everything that kept moving after you arrived. Six measurements obtained. No predictions. No astrology.
