A team of researchers led by Dartmouth College has recently used an innovative approach to solve the longstanding debate about what caused the mass extinction of dinosaurs and many other species 66 million years ago: Did a colossal asteroid strike or volcanic eruptions lead to their demise?
To remove potential biases, the experts took the scientists out of the equation and put the decision-making in the hands of computers.
The novel modeling technique they used employed interconnected processors to sift through extensive geological and climate data autonomously.
The processors were given the task of retroactively examining the fossil record to identify the sequences and conditions that culminated in the Cretaceous–Paleogene (K–Pg) extinction event. This event set the stage for mammals, including primates and eventually early humans, to rise.
“Part of our motivation was to evaluate this question without a predetermined hypothesis or bias,” said lead author Alex Cox, a graduate student in Earth Sciences at Dartmouth.
“Most models move in a forward direction. We adapted a carbon-cycle model to run the other way, using the effect to find the cause through statistics, giving it only the bare minimum of prior information as it worked toward a particular outcome. In the end, it doesn’t matter what we think or what we previously thought – the model shows us how we got to what we see in the geological record.”
This computational model tested over 300,000 potential scenarios spanning a million years around the K–Pg extinction, factoring in variables like carbon dioxide and sulfur dioxide emissions and biological productivity.
Relying on a machine learning technique known as the Markov Chain Monte Carlo, the processors functioned collaboratively yet independently, continuously refining their findings until a conclusion was reached that aligned with fossil records.
The traces left behind in fossils distinctly portray the catastrophic environment during the K–Pg extinction. There were significant ecological disruptions as food chains crumbled amidst an atmosphere that oscillated drastically due to airborne pollutants.
While the evidence of this devastation is clear, the root cause remains a topic of debate. Initial theories that pointed to volcanic eruptions have been overshadowed by the discovery of the Chicxulub impact crater in Mexico, attributed to a massive asteroid believed to be the primary catalyst for the dinosaur extinction.
However, recent findings suggest the asteroid might have struck an Earth already destabilized by powerful volcanic eruptions in India’s Deccan Traps.
Yet, there is still uncertainty about the relative contributions of each event to the extinction. This drove Cox and his advisor, Brenhin Keller, a Dartmouth assistant professor of Earth Sciences and co-author of the study, to “see what you would get if you let the code decide.”
Their model indicates that the emissions from the Deccan Traps could have alone instigated the global extinction. The Deccan eruptions began approximately 300,000 years before the asteroid’s impact, releasing staggering amounts of gases over nearly a million years.
“We’ve known historically that volcanoes can cause massive extinctions, but this is the first independent estimation of volatile emissions taken from the evidence of their environmental effects,” Keller explained.
“Our model worked through the data independently and without human bias to determine the amount of carbon dioxide and sulfur dioxide required to produce the climate and carbon cycle disruptions we see in the geologic record. These amounts turned out to be consistent with what we expect to see in emissions from the Deccan Traps.”
While the model recognized a significant decline in organic carbon accumulation in deep-sea regions around the Chicxulub impact timeframe, it revealed no surge in gas emissions during the period, implying that the asteroid’s role in the extinction wasn’t primarily linked to gas emissions.
Comparing historical events to modern times, Cox noted that between 2000 and 2023, fossil fuel burning released roughly 16 billion tons of carbon dioxide annually, a rate 100 times more intense than the Deccan Traps’ highest yearly emissions. However, it would still take millennia for current emission levels to equal the total output of these ancient eruptions.
“Most heartening is that the results we achieved are broadly physically plausible, which is impressive given that the model could have technically run completely wild without stronger prior constraints,” Cox said.
The integration of processors sped up the data analysis, trimming what could have taken years down to mere hours. This method developed by Cox and Keller can be applied to other earth systems models, enabling researchers to delve deeper into well-documented geological events where the outcomes are clear but the preceding conditions remain elusive.
“This type of parallel inversion hasn’t been done in earth sciences models before. Our method can be scaled up to include thousands of processors, which gives us a much broader solution space to explore, and it’s quite resistant to human bias,” remarked Cox.
“So far, people in our field have been more fascinated by the novelty of the method than the conclusion we reached. Any earth system for which we know the effect but not the cause is ripe for inversion. The better we know the output, the better we’re able to characterize the input that caused it.”
The study is published in the journal Science.