Earth's history is a roller-coaster of climate fluctuations, of relative warmth giving way to frozen periods of glaciation before rising up again to the more temperate climes we experience today.
What triggers these periods of glaciation, or ice ages, is difficult to pin down, though for some time researchers have strongly suspected quirks of Earth's orbit around the Sun are involved.
New research has demonstrated the precise relationship between past ice ages and each wobble, tilt, and angle of the planet's path, unlocking a new tool for predicting the future fluctuations of our global climate.
"The link between slight changes in axial tilt and orbital geometry and the waxing and waning of continental ice sheets represents one of the oldest mysteries in climate science," Earth scientist Stephen Barker of Cardiff University in the UK explained to ScienceAlert.
"As such, it represents a fundamental gap in our understanding of the climate system. Increasing our awareness of how Earth's dynamic climate system operates is crucial if we hope to be able to predict how climate might change in the future."
Earth's orbit around the Sun isn't perfectly symmetrical. It has a somewhat oval shape referred to as orbital eccentricity, with the Sun off-center in the oval, which means Earth's distance from the Sun changes throughout the year.
And the position of that oval in space shifts a little with each orbit; we call that orbital precession.
Finally, the tilt of Earth's rotational axis, a property known as obliquity, changes as it orbits the Sun.
It's been known for some time that these different properties of our planet's relationship to the Sun result in cycles of warmer and cooler climates, with periodic changes in different aspects of Earth's orbit affecting the seasonal and geographical distribution of sunlight.
Collectively known as Milankovitch cycles, these periodic changes occur roughly every 20,000, 40,000, 100,000, and 400,000 years, but teasing out which aspects of Earth's orbit are involved in fluctuations in climate is not an easy task.
"Earth's climate is an interconnected system of complex processes, all acting together to produce the changes we observe," Barker said. "Modeling these changes over the timescales relevant to glacial cycles requires a lot of processing power, in addition to the processes themselves being difficult to quantify and model independently."
For example, there are two very closely timed cycles; precession at 21,000 years, and the second harmonic of obliquity at 20,500 years. No one has been able to establish a clear link between either of these cycles and the end of an ice age.
In addition, for the last 800,000 years, ice ages have ended every 100,000 years, and scientists have been unable to find the cause of this cycle.
Barker and his colleagues started by looking at changes in oxygen isotope ratios in the deep sea over the last 800,000 years, preserved in the fossilized exoskeletons of tiny marine organisms called foraminifera. These changes can be used to map changes in the volume of continental ice, or ice sheets – a key metric in studying Earth's past glaciation.
With this information, the researchers created a detailed graph of glaciation cycles, against which they compared two idiosyncrasies of Earth's orbit – its precession and obliquity. And an amazing pattern emerged. The critical stages of the transitions between glacial and interglacial periods matched up with a particular relationship between precession and obliquity.
Deglaciation – the end of an ice age – seems intimately linked to a relationship between precession and obliquity; but it's obliquity alone that's responsible for the onset of an ice age.
This, the researchers say, explains the 100,000-year cycle. And it was all right there, hiding in plain sight.
"I was really excited when I saw the relationship between orbital phasing and the shape of the climate curve across these well-known transitions," Barker said.
"The curves we are looking at have been around for decades and have been looked at thousands of times (including by myself) and yet the relationship we found (which is easy to see when pointed out) remained all but hidden before now."
Previous studies, he said, have argued that the timing of ice age onsets is more random. His team's work shows that it's deterministic, meaning we now have a tool that lets us predict when ice ages are going to happen in the future.
Earth's obliquity is currently in the process of declining towards a minimum, which it will reach in 11,000 years or so; according to the team's calculations, the next ice age will kick off before then.
This is vitally important information to understanding the long-term, future effects of current human activity, Barker said.
"According to the latest IPCC reports, humans have already started to alter the course of climate away from its natural trajectory by the emission of greenhouse gases," he explained.
"This means that the decisions we make now will have consequences into the far future. At present, projected future climate change is gauged relative to modern (or pre-industrial) conditions.
"But we believe that to fully appreciate the true magnitude of future changes, these need to be compared with what might have happened in a hypothetical future, free from the influence of mankind. Therefore, we hope to create better predictions of future natural climate variability in order to quantify possible human influence into the coming millennia."
The team's research has been published in Science.