As Stellar Observations Improve, Earth’s History and Future Get Fuzzier

Regardless of what stock market analysts, political pollsters and astrologers might say, we can’t predict the future. In fact, we can’t even predict the past.

So much for the work of Pierre-Simon Laplace, the French mathematician, philosopher and king of determinism. In 1814, Laplace declared that if it were possible to know the velocity and position of every particle in the universe at one particular moment — and all the forces that were acting on them — “for such an intellect nothing would be uncertain, and the future, just like the past, would be the present to it.”

Laplace’s dream remains unfulfilled because we can’t measure things with infinite precision, and so tiny errors propagate and accumulate over time, leading to ever more uncertainty. As a result, in the 1980s astronomers including Jaques Laskar of the Paris Observatory concluded that computer simulations of the motions of the planets could not be trusted when applied more than 100 million years into the past or future. By way of comparison, the universe is 14 billion years old and the solar system is about five billion years old.

“You can’t cast an accurate horoscope for a dinosaur,” Scott Tremaine, an orbital dynamics expert at the Institute for Advanced Study in Princeton, N.J., commented recently in an email.

The ancient astrological chart has now become even blurrier. A new set of computer simulations, which take into account the effects of stars moving past our solar system, has effectively reduced the ability of scientists to look back or ahead by another 10 million years. Previous simulations had considered the solar system as an isolated system, a clockwork cosmos in which the main perturbations to planetary orbits were internal, resulting from asteroids.

“The stars do matter,” said Nathan Kaib, a senior scientist with the Planetary Science Institute in Tucson, Ariz. He and Sean Raymond of the Laboratoire d’Astrophysique de Bordeaux in France published their results in Astrophysical Journal Letters in late February.

The researchers discovered that a sunlike star named HD 7977, which currently lurks 247 light-years away in the constellation Cassiopeia, could have passed close enough to the sun about 2.8 million years ago to rattle the largest planets in their orbits.

That added uncertainty makes it even harder for astronomers to forecast more than 50 million years into the past, to correlate temperature anomalies in the geological record with possible changes in the Earth’s orbit. That knowledge would be useful as we try to understand climatic changes underway today. About 56 million years ago, Dr. Kaib said, the Earth evidently went through the Paleocene–Eocene Thermal Maximum, a period lasting more than 100,000 years during which average global temperatures increased as much as 8 degrees Celsius.

Was this warm spell triggered by some change in Earth’s orbit around the sun? We may never know.

“So I’m no expert, but I think that’s the warmest period in, like, the last 100 million years,” Dr. Kaib said. “And it’s almost certainly not caused by the Earth’s orbit itself. But we do know that long-term climate fluctuations are tied to Earth’s orbital fluctuations. And so if you want to figure out climate anomalies, it helps to be confident in what Earth’s orbit is doing.”

Dr. Tremaine noted, “The simulations are carefully done, and I believe the conclusion is correct.” He added, “This is a relatively minor change in our understanding of the history of the Earth’s orbit, but it is a conceptually important one.”

The really interesting story, he said, is how chaos in Earth’s orbit could have left a mark in the paleoclimate record.

The ability to track the movements of stars just beyond the solar system has been dramatically improved by the European Space Agency’s Gaia spacecraft, which has been mapping the locations, motions and other properties of two billion stars since its launch in 2013.

“For the first time we can actually see individual stars,” Dr. Kaib said, “project them back in time or forward, and figure out which stars are close to the sun and which ones haven’t come close, which is really cool.”

According to his calculations, about 20 stars come within one parsec (about 3.26 light-years) of the sun every million years. HD 7977 could have come as close as 400 billion miles from the sun — about the distance to the Oort cloud, a vast reservoir of frozen comets on the edge of the solar system — or remained a thousand times as distant. Gravitational effects from the closer encounter could have rattled the orbits of the outer giant planets, which in turn could have rattled the inner planets like Earth.

“That is potentially powerful enough to alter simulations’ predictions of what Earth’s orbit was like beyond approximately 50 million years ago,” Dr. Kaib said.

As a result, he said, almost anything is statistically possible if you look ahead far enough. “So you find that, for instance, if you go forward billions of years, not all the planets are necessarily stable. There’s actually about a 1 percent chance that Mercury will either hit the sun or Venus over the course of the next five billion years.”

Whatever happens, chances are we won’t be around to see it. Stranded in the present, we don’t know for certain where we came from or where we are going; the future and the past recede into myth and hope. Yet we press forward trying to peer past our horizons in time and space. As F. Scott Fitzgerald wrote in “The Great Gatsby”: “So we beat on, boats against the current, borne back ceaselessly into the past.”