What really ended dinosaurs: Computer simulation sheds light on their final days

The mass extinction of the dinosaurs has triggered the curiosity of scientists for decades. Now, new technology allows them to investigate in more depth one of the biggest mysteries of prehistoric times.

Giant reptiles roamed the Earth for 165 million years before their sudden disappearance about 65 million years ago. Since the first discovery of fossils, which told the story of these ancient behemoths, we've asked ourselves the question: “Who killed the dino?”

The discussion among scientists is still ongoing. Some theories posit that gradual climate changes, influenced by volcanic activity, eventually made the planet unlivable for non-bird dinosaurs.

Another hypothesis suggests that an extraterrestrial object collided with the Earth. Its impact is believed to have formed the giant Chicxulub crater on the Yucatan Peninsula of Mexico and threw up enough dust to cause catastrophic climate change.

Recently, scientists at Dartmouth College performed a computer simulation that showed that the asteroid strike was probably incapable of causing a spike in carbon dioxide or sulfur dioxide gasses large enough to wipe out the dinosaurs' population from the surface of Earth.

However, the simulations showed that the amount of gas released into the atmosphere from volcanic activity alone was sufficient to impact the planet’s temperature.

In the research published on September 28th, the team suggests that massive bursts of gas produced by volcanic activity in the western part of moder-day India were capable of causing extinction.

The eruptions, which lasted for a million years, formed a massive accumulation of igneous rocks now known as the Deccan Traps, one of the largest volcanic landforms on Earth.

As reported by Science, other scientists are skeptical of the findings, stating that they don't provide a definitive answer to the long-standing mystery of the extinction. Although the killer asteroid released less gas than volcanic activity, it occurred suddenly as a one-time event, devastating in its impact nonetheless.

Computer simulation

Instead of first searching for the theoretical 'perpetrator' responsible for the extinction of dinosaurs and then looking for supporting evidence, scientists employed a computer to analyze the available shreds of evidence at the time of extinction.

The researchers used data from three cores drilled into deep-sea sediments, each spanning 67 million to 65 million years ago. Sediments contain foraminifera, ocean microorganisms whose shells’ chemical makeup shows the ocean's chemistry and global temperature at the time of its formation.

The shells showed deadly bursts of gas into the atmosphere, particularly carbon dioxide, which contributes to global warming and ocean-acidifying sulfur dioxide. Determining the source of the gas could answer the question of what was to blame for the extinction of the dinosaurs.

Scientists used a statistical model to calculate the probability of different scenarios of gas emissions from various sources. To do the calculations, they employed 128 different processors to run scenarios in parallel. Parallel computing has reduced the time required for computations from a year to just a few days.

AI is shifting the research process

The enigma of the disappearance of dinosaurs is not the only question that computers can assist scientists with. Advancements in computing and AI model development are certainly impacting the workflows of all kinds of scientific research.

“Scientists at McMaster and MIT, for example, used AI to identify an antibiotic that fights what the World Health Organization calls one of the world’s most dangerous drug-resistant bacteria for hospital patients,” said former Google CEO Eric Schmidt in MIT Technology Review.

“The FDA has already cleared 523 devices that use AI, and a Google DeepMind model can control plasma in nuclear fusion reactions, bringing us closer to a clean-energy revolution.”

According to him, the scientific process remains the same; however, AI has the potential to revolutionize how theorizing, experimenting, collecting, and analyzing data will look.

AI tools such as PaperQA and Elicit can already speed up the process of analyzing existing literature and providing citations. Also, AI can cast a wider search net for the hypothesis and speed up the process of narrowing it down to formulate stronger hypotheses. Not to mention the endless possibilities for low-cost simulations that would never be possible in real-world experiments.

“We can build a future where AI-powered tools will both save us from mindless and time-consuming labor and encourage creative breakthroughs that would otherwise take decades,” writes Schmidt.