Analysis of Earth's past ice ages, tracked across one million years, has shed new light on the mystery of the planet's natural climate cycles, enabling improved projections of future climate, finds a new study co-authored by UCL researchers.
Published in Science, the international team of scientists led by Cardiff University identified the specific contributions of various subtle changes in Earth's orbit that cause the climate to shift between warm periods and glacial periods every 100,000 years.
The researchers matched the repeated expansion and contraction of ice sheets across the Northern hemisphere with variations in the shape of Earth's orbit of the Sun, its wobble and the angle on which its axis is tilted, which change the amount of solar energy that is needed to melt ice.
Co-author Professor Chronis Tzedakis (UCL Geography) said: "In 2012, our UCL-led study had suggested that the relative timing of changes in orbital parameters determines the length and shape of warm periods. The new study provides a rigorous test of these predictions and creates a comprehensive and cohesive picture of the natural cycles in Earth's climate and what drives them."
The analysis also suggests that the onset of the next ice age could be expected within the next 11,000 years, in the absence of any human modification of natural climate.
Predictions of a link between Earth's orbit around the Sun and fluctuations between glacial and interglacial conditions have been around for over a century but were not confirmed by real-word data until the mid-1970s.
Since then, scientists have struggled to identify precisely which orbital parameter is most important for the beginning and ending of glacial cycles because of the difficulty in dating climatic changes so far back in time.
The team overcame this problem by looking at the shape of the climate record through time.This allowed them to identify how the different parameters fit together to produce the climate changes observed.
Lead author Professor Stephen Barker from Cardiff University, said: "We were amazed to find such a clear imprint of the different orbital parameters on the climate record. It is quite hard to believe that the pattern has not been seen before.
"The pattern we found is so reproducible that we were able to make an accurate prediction of when each interglacial period of the past million years or so would occur and how long each would last."
Co-author Professor Lorraine Lisiecki of the University of California, Santa Barbara, added: "This is important because it confirms the natural climate change cycles we observe on Earth over tens of thousands of years are largely predictable and not random or chaotic.
"And because we are now living in an interglacial period - called the Holocene - we are also able to provide an initial prediction of when our climate might return to a glacial state."
Co-author Dr Gregor Knorr from the Alfred Wegener Institute said: "But such a transition to a glacial state in 10,000 years' time probably won't happen because human emissions of carbon dioxide into the atmosphere have already diverted the climate from its natural course, with longer-term impacts into the future."
The team plans to build on their findings to create a baseline of the Earth's natural climate for the next 10,000-20,000 years by calibrating past changes.
Used in combination with climate model simulations, researchers hope to quantify the absolute effects of human-made climate change into the far future.
Professor Barker added: "Now we know that climate is largely predictable over these long timescales, we can actually use past changes to inform us about what could happen in the future.
"This is something we couldn't do before with the level of confidence that our new analysis provides, and is vital for better informing decisions we make now about greenhouse gas emissions, which will determine future climate changes."
Professor Tzedakis said: "The new study is a major step towards a unified theory of glacial cycles."
