New tool maps Earth's continental drift over 320 million years

The ground beneath your feet has traveled farther than you might imagine. A rock formation in the Netherlands today may have crystallized thousands of miles away, under a completely different climate, in a world that no longer exists. That simple fact—that continents move, that latitude shifts, that the past cannot be read from present-day geography alone—has long complicated the work of scientists trying to understand ancient climates and the fossils locked inside ancient stone.

Now an international team led by Earth scientist Douwe van Hinsbergen at Utrecht University has built a tool to solve that problem. Paleolatitude.org, published this week in PLOS One, lets anyone click on nearly any location on Earth and see where that place sat millions of years ago—back to roughly 320 million years, when the supercontinent Pangea dominated the planet. The site rests on what the team calls the Utrecht Paleogeography Model, a reconstruction of how tectonic plates have moved through deep time, where continents have drifted, and which ocean basins have opened and closed. For paleoclimate researchers and paleontologists, the implications are substantial. Latitude controls the angle of sunlight hitting Earth's surface, which determines whether a place tends toward ice, desert, rainforest, or tropical sea. Get the latitude wrong, and you misread the entire climate story locked inside a fossil or a sediment layer.

Consider a concrete example from Winterswijk in the eastern Netherlands, where scientists study plants and animals that lived 245 million years ago in conditions resembling today's Persian Gulf—desert heat, tropical sea. The obvious question: Was the entire planet dramatically hotter then, or was that part of Europe simply much closer to the equator? Earlier work by the Utrecht group had already suggested the second explanation. The new model aims to answer such questions with greater confidence than older tools allowed. Previous reconstructions could capture broad continental movements but often stumbled when dealing with smaller tectonic plates, heavily deformed mountain belts, and fragments of crust that no longer exist as intact landmasses. This upgrade folds many of those missing pieces into a single global system, including so-called lost continents and microcontinents such as Greater Adria, Argoland, and the Tethys Himalayas, whose remains are now preserved as folded rocks in the Mediterranean, the Himalayas, and Indonesia.

Building such a reconstruction requires two main steps. First, geoscientists work out how tectonic plates moved relative to each other—a process that in deformed mountain belts means effectively unfolding rock packages that were bent, stacked, shortened, or shifted by long episodes of collision and subduction. The team drew on detailed regional reconstructions from the Mediterranean, Iran, the Himalaya, Tibet, Southeast Asia, the Caribbean, and parts of China and Indochina. Then comes the second step: placing that entire reconstruction at the correct latitude on the ancient globe. Scientists rely on paleomagnetism, the magnetic information preserved in rocks. Because Earth's magnetic field meets the planet's surface at different angles depending on latitude, magnetic minerals can preserve a record of where a rock formed. Combined with dating methods, this gives researchers a way to estimate the latitude of ancient rocks and, by extension, the movement of the plates carrying them through time. The updated paleomagnetic reference frame, called gAPWP25, adds 32 entries to the earlier database—about a 10 percent increase—and generally reduces uncertainty by drawing on more data.

One of the larger advances in this version is that it goes far beyond the rigid interiors of the biggest tectonic plates. Many important fossils and rock records come from mountain belts, not quiet continental interiors. These vast zones where crust gets compressed, stacked, stretched, or accreted preserve rich records of oceans that closed and continental fragments that collided. Yet they are also among the hardest places to reconstruct cleanly. The Utrecht model divides those regions into many rigid polygons representing crustal blocks with shared origins. In some intensely deformed places, the polygons are only tens of kilometers across. That lets the system estimate paleolatitude for rocks in geologically messy regions such as the Alps, the Zagros, the Himalaya, Japan, New Zealand, and parts of California and Alaska—though not every disputed or poorly resolved region is included yet.

The public-facing website allows users to click on a map and generate a paleolatitude curve for a location back through time. Up to ten curves can be shown at once. Users can also enter coordinates manually, choose ages or age ranges, and download graph outputs or Excel files. For researchers, a bulk option for processing large fossil datasets may be one of the most important additions. It makes it easier to recalculate paleolatitudes for thousands of specimens while also estimating uncertainty for each point. Earlier paleogeographic tools often gave users a paleolatitude estimate without a clear error range. This version combines uncertainty in the paleomagnetic reference frame with uncertainty in fossil or rock age, opening the door to more careful comparisons across time and space.

The implications extend well beyond tectonics. Emilia Jarochowska, a Utrecht paleontologist and co-author, argues that better paleolatitude estimates could help paleontologists ask more precise questions about how biodiversity shifted with climate. The improved tool may help scientists test long-standing ideas about mass extinctions and recovery—which latitudes became uninhabitable first, which became refuges, which species migrated, which adapted, which went extinct. The paper illustrates that idea using fossil datasets, including about 34,000 Upper Jurassic marine fossils used to examine the latitudinal diversity gradient, the long-observed pattern in which species richness tends to be higher at lower latitudes. The authors acknowledge that simplifying geologically complex mountain belts into a global 2D model inevitably loses detail, and that small georeferencing errors may place a coordinate in the wrong tectonic polygon. Some regions remain absent pending better reconstructions—about 1,000 fossils from Alaska and the Canadian Cordillera were discarded because those areas are not yet included in the model. Even so, the new system marks a significant move toward a more detailed and more transparent global framework. The team says the model will eventually be extended back to roughly 550 million years ago, reaching the Cambrian explosion of complex life.