Pangaea_Netherlands

Location of The Netherlands (pink) in Pangaea, 258 million years ago | Brown: Current continental crust above water | Light brown: Thinned and reconstructed continental crust, mostly submerged | Dark brown: Cratons/very old continental crust | Light blue: Oceanic crust | Dark blue: Thickened oceanic crust due to volcanism (Credit: Utrecht University)

In A Nutshell

  • A free online tool called Paleolatitude.org has been upgraded to estimate where rocks and fossils were located on Earth for chosen time intervals going back up to 320 million years.
  • Version 3.0 is the first to include mountain belt regions like the Himalayas and Alps, which hold some of the richest fossil records on the planet but were left out of earlier versions.
  • A more precise magnetic reference system reduces uncertainty in the ancient location estimates, drawing on 32 newly added datasets from sites around the world.
  • Researchers tested the tool on roughly 34,000 late Jurassic marine fossils and found its estimates hold up well enough for large-scale scientific analysis.

A fossil found today in Norway might have formed in the tropics millions of years ago. Coal deposits in Antarctica perhaps initially accumulated in what was once a warm, swampy forest. Paleolatitude, a rock’s ancient position relative to the equator, is the single biggest factor scientists use to reconstruct those vanished environments, because latitude controls how directly sunlight strikes the ground and therefore drives climate. Get that number wrong, and the whole picture of what an ancient world looked like can fall apart.

A decade ago, a team of researchers built a free online calculator to help solve that problem. Called Paleolatitude.org, it lets scientists estimate where rocks and fossils originally formed, long before plate tectonics shuffled them to their current locations. Now that tool has received its biggest upgrade yet.

Version 3.0 went live in April 2026 with three major improvements. For the first time, it incorporates what the authors describe as the first global paleogeographic model back to 320 million years that also restores many rock units now stacked inside mountain ranges like the Himalayas, the Alps, and the Andes. It also uses a more precise magnetic reference system and a redesigned interface that lets researchers process entire datasets at once instead of entering locations one at a time.

Published in PLOS ONE, the work was led by Douwe van Hinsbergen of Utrecht University in the Netherlands, with nine co-authors from institutions across Italy, France, Tajikistan, Austria, and the Netherlands.

Earth’s continents have never stayed put. Plates split apart, collide, and rotate. On top of that, the entire solid Earth can wobble relative to its spin axis, shifting all the continents at once relative to the equator and poles. Accounting for all of those motions simultaneously is exactly why the original tool was built. Type in coordinates and a time period, and the calculator returns an estimate of where that spot was, along with a margin of error.

Mountain Belt Rocks Finally Included in Global Paleolatitude Model

Version 3.0’s biggest change is the inclusion of the Utrecht Paleogeography Model, which covers most locations in the model for chosen time intervals over the last 320 million years. For the first time, the model attempts to restore rock units now crumpled and stacked inside mountain ranges, not just the stable continental cores that earlier versions handled.

Mountains form when tectonic plates collide and compress, thrusting slices of rock into towering piles. The Himalayas, for instance, contain rocks that were once part of a distant ocean floor. Earlier versions of the tool left those regions out entirely, which was a significant gap. Mountain belts tend to have far better rock exposures and richer fossil collections than many stable continental interiors, which are often buried under soil and vegetation.

Version 3.0 incorporates detailed reconstructions of deformed zones including the Mediterranean, the Caribbean, the Tibetan Plateau and Himalaya, Southeast Asia, the western United States, Iran, the Scotia Sea near Antarctica, and the continental fragments making up Mongolia, China, and Indochina. Each region was reconstructed using field geology, rock dating, and layering patterns, deliberately without using any climate or biological data, preventing circular reasoning when the tool is later used to study those very subjects.

Each deformed region was divided into thousands of small rigid pieces representing recognizable geological units. When reconstructed backward in time, these pieces may overlap, representing ancient stretching, or separate, representing compression. Overlaps rarely exceed about 100 kilometers, translating to roughly one degree of latitude in uncertainty.

Geo-Paleolatitude
The shift in the latitude of Washington, D.C. over the past 320 million years. Screenshot from Paleolatitude.org (Credit: Utrecht University)

Sharper Magnetic Reference Data Shrinks Paleolatitude Uncertainty

Ancient rocks preserve a record of Earth’s magnetic field like tiny frozen compasses. By analyzing these records from rocks of known ages on stable continents, scientists can track how the magnetic pole appeared to shift over time, which is really a record of moving continents.

Earlier versions of the calculator grouped magnetic data into study-level averages that contained arbitrary numbers of measurements. This introduced reproducibility problems and inflated uncertainty. Version 3.0 uses a reference system called gAPWP25, which works at the level of individual measurement sites, giving equal weight to each reading and producing smaller uncertainty bands.

Thirty-two new datasets, roughly a ten percent increase, were added to the magnetic database, drawing from locations as varied as New Zealand, Iceland, Morocco, Brazil, Argentina, India, Siberia, and England. Despite those additions, the updated reference path differs only slightly from its predecessor, with the largest shifts of about 1.5 to 2.5 degrees appearing in time windows with the sparsest existing data.

Late Jurassic Fossil Test Shows Biodiversity Tool in Action

To show what the upgraded calculator can do, the researchers applied it to roughly 34,000 late Jurassic marine fossils, calculating ancient latitudes while carrying both positional and age uncertainty through to the final result. Most of the biodiversity pattern held up under that more honest accounting, confirming the tool’s estimates are precise enough for large-scale analyses.

Some regions remain outside the model’s reach. Reconstructions of the Canadian mountain belt, Alaska, and parts of northeastern Siberia are not yet included. Rocks on the outermost edges of continental shelves may carry up to about two degrees of additional error. Small mapping inaccuracies could occasionally assign a coordinate to the wrong geological unit. For the vast majority of locations covered by the model, though, the tool now offers among the most precise model-based paleolatitude estimates currently available. And it is entirely free.

For paleontologists, climate scientists, and anyone studying how life responded to a shifting planet, getting the geography right is not a footnote. It is the foundation. Version 3.0 just made that foundation a lot more solid.


Paper Notes

Limitations

Several regions are not yet incorporated into the model, including the Canadian Cordillera, Alaska, and pre-late Cretaceous northeastern Siberia, because no existing reconstruction follows the team’s protocol. Passive margin rifting has not been fully incorporated along all continental edges, meaning rocks on distal passive margins may carry up to about two degrees of positional error if rifting had a north-south component. Within deformed mountain belts, the true spatial distribution and geological structure of rock units cannot be fully captured in a two-dimensional global model. Small georeferencing errors may place tectonic boundaries off by a few kilometers, potentially assigning a coordinate to the wrong polygon and returning an incorrect paleolatitude. Inaccurate recorded sampling locations for fossils or rocks from mountain belts could similarly place them in an incorrect tectonic unit. Potential errors are typically not more than a few degrees but should be considered when interpreting results. The time windows with the sparsest paleomagnetic data, the Late Cretaceous, latest Jurassic, and Early Triassic, showed the largest shifts when new data were added, highlighting sensitivity to data density in those intervals.

Funding and Disclosures

Funding came from the Nederlandse Organisatie voor Wetenschappelijk Onderzoek (NWO) through a Vici grant to Douwe van Hinsbergen and a Veni grant to Lydian Boschman. Additional support came from the HORIZON EUROPE European Research Council through a starting grant to Emilia Jarochowska and two consolidator grants supporting Bram Vaes. Funders played no role in study design, data collection or analysis, decision to publish, or preparation of the manuscript. No authors have competing interests.

Publication Details

Title: Paleolatitude.org 3.0: A calculator for paleoclimate and paleobiology studies based on a new global paleogeography model | Authors: Douwe J. J. van Hinsbergen, Bram Vaes, Lydian M. Boschman, Nalan Lom, Suzanna H. A. van de Lagemaat, Eldert L. Advokaat, Sanne de Baar, Menno R. T. Fraters, Joren Paridaens, Emilia B. Jarochowska | Journal: PLOS ONE, Volume 21, Issue 4 | Published: April 29, 2026 | DOI: 10.1371/journal.pone.0346817 | Data Availability: All data are held in public repositories at https://doi.org/10.5281/zenodo.18183857 and https://doi.org/10.6084/m9.figshare.31021144. | Editor: Carlo Meloro, Liverpool John Moores University, United Kingdom | Received: January 11, 2026 | Accepted: March 17, 2026

About StudyFinds Analysis

Called "brilliant," "fantastic," and "spot on" by scientists and researchers, our acclaimed StudyFinds Analysis articles are created using an exclusive AI-based model with complete human oversight by the StudyFinds Editorial Team. For these articles, we use an unparalleled LLM process across multiple systems to analyze entire journal papers, extract data, and create accurate, accessible content. Our writing and editing team proofreads and polishes each and every article before publishing. With recent studies showing that artificial intelligence can interpret scientific research as well as (or even better) than field experts and specialists, StudyFinds was among the earliest to adopt and test this technology before approving its widespread use on our site. We stand by our practice and continuously update our processes to ensure the very highest level of accuracy. Read our AI Policy (link below) for more information.

Our Editorial Process

StudyFinds publishes digestible, agenda-free, transparent research summaries that are intended to inform the reader as well as stir civil, educated debate. We do not agree nor disagree with any of the studies we post, rather, we encourage our readers to debate the veracity of the findings themselves. All articles published on StudyFinds are vetted by our editors prior to publication and include links back to the source or corresponding journal article, if possible.

Our Editorial Team

Steve Fink

Editor-in-Chief

John Anderer

Associate Editor

Leave a Reply