Artificial intelligence discovers surprising patterns in Earth’s biological mass extinctions

Visualize the history of life

A new study applies machine learning to the fossil record to visualize the history of life and show the impact of major evolutionary events. This shows the long-term evolutionary and ecological consequences of major extinction and speciation events. Colors represent the geological periods from the Tonian, beginning 1 billion years ago, in yellow, to the current Quaternary Period, shown in green. The transition from red to blue marks the end of the Permian mass extinction, one of the most disruptive events in the fossil record. Credit: J. Hoyal Cuthill and N. Guttenberg

The idea that mass extinctions allow many new species to evolve is a central concept in evolution, but a new study using artificial intelligence to examine the fossil record finds that this is rarely true, and there must be another. be statement.

Charles Darwin’s signature opus, About the origin of the species, concludes with a wonderful summary of his theory of evolution: “There is a greatness in this view of life, with its various faculties, originally breathed into a few forms or in one; and that as this planet cycles on according to the fixed law of gravity, from so simple beginnings endless forms of the most beautiful and wonderful are and are evolving. “

In fact, scientists now know that most of the species that ever existed are extinct. This species extinction has generally been more or less balanced by the emergence of new species in Earth’s history, with some major transient imbalances that scientists call mass extinction events. Scientists have long believed that mass extinctions create productive periods of species evolution, or “radiations,” a model called “creative destruction.” A new study led by scientists affiliated with the Earth-Life Science Institute (ELSI) at the Tokyo Institute of Technology used machine learning to investigate the co-occurrence of fossil species and found that radiations and extinctions are rarely related , and thus probably rarely mass extinctions. cause radiations of comparable magnitude.

Creative destruction is central to classic concepts of evolution. It seems clear that there are periods when many species suddenly disappear and new species suddenly appear. However, radiations of a similar scale to the mass extinctions, which this study therefore calls the mass radiations, have undergone much less analysis than extinction events.

This study compared the effects of both extinction and radiation during the period for which fossils are available, the so-called Phanerozoic Eon. The Phanerozoic (from Greek meaning ‘apparent life’), represents the most recent ~ 550 million year period of Earth’s total ~ 4.5 billion years of history, and is important to paleontologists: prior to this period, most organisms that existed microbes that did not easily form fossils, so the earlier evolutionary record is difficult to observe.

The new study suggests that creative destruction is not a good description of how species originated or became extinct during the Phanerozoic, and suggests that many of the most remarkable times of evolutionary radiation occurred when life entered new evolutionary and ecological arenas, such as during the Cambrian. explosion of animal diversity and the Carboniferous expansion of forest biomes. Whether this is true for the previous ~ 3 billion years dominated by microbes is not known, as the scarcity of recorded information on such ancient diversity did not allow a similar analysis.

Paleontologists have identified a handful of the most serious mass extinction events in the Phanerozoic fossil record. These mainly include the five major mass extinctions, such as the end Permian mass extinction, where more than 70% of species are estimated to be extinct. Biologists have now suggested that we may now be entering a “ sixth mass extinction, ” which they think is mainly caused by human activity, including hunting and land use changes brought about by the expansion of agriculture. A well-known example of the previous mass extinctions of the “Big Five” is the Chalk-Tertiary (usually abbreviated “KT”, using the German spelling of Cretaceous) which appears to have been caused when a meteor hit Earth 65 million years ago and wiped out the non-avian dinosaurs.

In observing the fossil record, scientists came to the belief that mass extinctions cause particularly productive radiations. For example, in the KT dinosaur extinction event, it was conventionally assumed that a wasteland was created, allowing organisms such as mammals to recolonize and ‘radiate’, allowing the evolution of all kinds of new mammal species, eventually laying the groundwork for the emergence of humans. In other words, if the KT event of “creative destruction” hadn’t happened, we might not be here to discuss this question.

The new study began with an informal discussion in ELSI’s “Agora,” a large communal space where ELSI scientists and visitors often have lunch and start new conversations. Two of the paper’s authors, evolutionary biologist Jennifer Hoyal Cuthill (now a research fellow at Essex University in the UK) and physicist / machine learning expert Nicholas Guttenberg (now a research scientist at Cross Labs in collaboration with GoodAI in the Czech Republic), both Postdoctoral scientists were at ELSI when the work began, kicking around whether machine learning could be used to visualize and understand the fossil record.

While visiting ELSI, just before the COVID-19 pandemic began to limit international travel, they worked feverishly to expand their analysis to investigate the correlation between extinction and radiation events. These discussions allowed them to relate their new data to the breadth of existing ideas about mass extinctions and radiations. They soon discovered that the evolutionary patterns identified using machine learning differed in significant ways from traditional interpretations.

The team used a new application of machine learning to investigate the temporal coexistence of species in the Phanerozoic fossil record, examining more than a million entries in a huge curated, public database of nearly two hundred thousand species.

Lead author Dr. Hoyal Cuthill said, “Some of the most challenging aspects of understanding the history of life are the vast timescales and the number of species involved. New applications of machine learning can help by enabling us to visualize this information in a human-readable form. This means that we can hold, so to speak, half a billion years of evolution in the palm of our hand and gain new insights from what we see. “

Using their objective methods, they found that the “big five” mass extinction events previously identified by paleontologists were picked by the machine learning methods as one of the top 5% significant perturbations where extinction was stronger than radiation. or vice versa, as well as seven additional mass extinctions, two combined mass extinction radiation events, and fifteen mass radiations. Surprisingly, unlike previous stories emphasizing the importance of post-extinction radiations, this work found that the most similar mass radiations and extinctions were only rarely linked over time, refuting the idea of ​​a causal relationship between the two.

Co-author Dr. Nicholas Guttenberg said, “The ecosystem is dynamic, you don’t necessarily have to tear down an existing piece for something new to appear.”

The team further found that radiation can in fact cause major changes in existing ecosystems, an idea the authors refer to as “ destructive creation. ” They found that, during the Phanerozoic Eon, the species that once formed an ecosystem almost all disappeared 19 million years later. But when mass extinctions or radiations occur, this turnover rate is much higher.

This provides a new perspective on how the modern “sixth extinction” is taking place. The Quaternary period, which began 2.5 million years ago, had witnessed repeated climatic changes, including dramatic changes in the Ice Age, times when locations at high latitudes on Earth were covered with ice. This means that the current “sixth extinction” is affecting biodiversity that has already been disrupted, and the authors suggest it will take at least 8 million years to return to its long-term average of 19 million years. Dr. Hoyal Cuthill notes that “every extinction that takes place under our watch erases a species that may have existed for millions of years until now, making it more difficult for the normal process of ‘new species emergence’ to replace. what gets lost. “

Reference: “Effects of Speciation and Extinction Measured by an Evolutionary Decay Clock” by Jennifer F. Hoyal Cuthill, Nicholas Guttenberg and Graham E. Budd, December 9, 2020, Nature.
DOI: 10.1038 / s41586-020-3003-4