Vast new fossil record challenges story of Earth’s ‘big five’ mass extinctions

Large language models (LLMs) such as OpenAI’s ChatGPT are trained to recognize patterns in written texts, helping them to choose a statistically probable word to complete a sentence. Palaeontologists have now trained a similar model on fossil records instead of words, then used it to reconstruct a timeline of changes in Earth’s biodiversity over hundreds of millions of years in unprecedented detail.

This big-data approach to interpreting the rock record — the field known as stratigraphy — could spur a rethink of major upheavals in the history of life, the model’s developers say. For instance, the model suggests that three of the “Big Five” mass-extinction events were gradual declines in biodiversity, rather than the sudden cataclysms described by other reconstructions.

Supercomputer scours fossil record for Earth’s hidden extinctions

The model is “a real revolution for studying stratigraphy and past diversity change”, says Shuzhong Shen, a palaeontologist at Nanjing University in China, who presented the research at a meeting of the Geological Society of America in Denver, Colorado, on 12 October.

Other researchers say that although the model’s reconstruction — called OneStrata — seems to be the most detailed global record so far, scientists already have high-resolution reconstructions of changes in biodiversity at specific locations, including many that shed light on mass extinctions.

“We could do some extra things with it, but I don’t know if it would be Earth shattering,” says Matthew Clapham, a palaeontologist at the University of California, Santa Cruz.

Strata struggles

The history of life is written in the fossil record. Researchers can reconstruct the grand story of adaptation and extinction across Earth’s history by marking the depth at which species occur in layers of sedimentary rock, called strata; depth corresponds to when a fossil was deposited in geological time. However, it is a steep computational problem to combine sequences from many different sites involving tens of thousands of species into a single coherent timeline of changing biodiversity.

To simplify this task, stratigraphers divide sequences of fossils for each species into time intervals of millions of years. This makes it easier to assemble many sequences into a unified whole. But it produces a blurry view, obscuring changes in biodiversity that occur over shorter time intervals.

For the past two decades, Shen and his colleagues have worked to get a sharper picture by compiling a huge data set of fossils and other geological data called OneStratigraphy. This includes data on the depth of millions of fossil specimens in sequences of rock from tens of thousands of sites around the world.

Large fossil model

In 2020, Shen and his colleagues ran an algorithmic method on a supercomputer to piece together thousands of these sequences, enabling them to estimate when a species appeared and went extinct with a resolution of 26,000 years — much higher than the resolution of previous estimates. They could then reconstruct a high-resolution timeline for 11,000 marine species over about 300 million years¹. However, including more species at shorter intervals requires an impractical amount of computational resources.

Biodiversity theory backed by island bird data

To get around that computational limit, Shen’s colleagues at the Zhejiang Lab — an AI-focused research institution in Hangzhou, China — used data from OneStratigraphy to train a machine-learning model on statistical patterns in the fossil record. Just as an LLM uses its training to predict the next word in a sentence, the OneStrata model uses those patterns to predict when a particular fossil is most likely to have been deposited.