Snow and ice are in the headlines as the East Coast of the United States digs itself out from an historic blizzard. In an era of rapid human-caused climate change, last week’s storm was a reminder that ice is an important part of our planet’s past.

Over the last 800,000 years, Earth has been through 11 major ice ages. The last ice age reached its peak about 20,000 years ago. Massive ice sheets, mostly more than one kilometre thick, covered much of Northern North America, Europe, and Asia. The global average temperature was about 5°C less than it is today, and the large quantity of water locked into the ice sheets meant the sea level was about 150m lower.

Are we headed for the next ice age? Will climate change delay the next ice age? A new paper in Nature - by Ganopolsky and colleagues - concludes that the answers to these questions and no and yes. We are not headed for a new ice age, and human-caused climate change will, among its many impacts, delay the start of the next ice age until 50,000 to 100,000 years into the future.

Discovering past ice ages and understanding their causes and consequences has been one of the most exciting scientific stories of the past 200 years. By the middle of the 19th Century, many geologists reported landscape features that could only be explained if, at some point in the past, there were glaciers in areas that are currently ice free and, in many cases, thousands of kilometers from current glaciers. Fingerprints of glacial activity range from the distinctive glacial landscapes of California’s Yosemite Valley, to giant boulders moved large distances, to deep deposits of temperate soils composed of rock particles pulverized by glaciers.

The cause of the sequence of ice ages and warm interglacials was completely mysterious until the 1920s, when a Serbian mathematician and astronomer, Milutin Milankovitch, realized that periodic “wiggles” in Earth’s orbit produce small but systematic changes in sunlight distributed between the Southern and Northern Hemispheres. Because there is much more land in the Northern Hemisphere and because land temperature is more sensitive to variations in solar radiation, Milankovitch argued that periods of lower Northern Hemisphere sunshine might lead to ice ages, and periods of higher Northern Hemisphere sunshine to warm interglacials.

Evidence for Milankovitch’s theory was thin at first, but by the 1970s and 1980s, detailed long-term climate records from ocean sediment and ice cores documented the overwhelmingly strong synchronization of past ice ages with the sunshine variations calculated from Milankovitch's theory. Since then, evidence for overall control of ice-age and interglacial cycles by Milankovitch’s mechanism has become increasingly stronger, specific, and multi-threaded. Still, the full suite of conditions sufficient to trigger an ice age remain unclear.

Ganopolsky and colleagues used a “pattern-matching” approach to refine the understanding of conditions sufficient to trigger an ice age. The approach uses a climate/ice model combination with many variations equally consistent in simulating current conditions. The final analysis uses only the configurations that accurately simulate the onset of past ice ages. The investigators were especially interested in two periods, about 400,000 and 800,000 years ago. These periods had sunshine distributions similar to the present and marked the start of ice ages.

What they found is that the present is actually quite different. Compared to the period 400,000 years ago, Northern Hemisphere sunshine is higher, substantially too high for an ice age. Compared to 800,000 years ago, the sunshine is about the same, but the level of atmospheric carbon dioxide, even at the start of the industrial revolution, has been too high for an ice age. A shivering conclusion is that, if carbon dioxide were only 20% lower, we would be moving toward an ice age over the next 1500 years. The reason for the low carbon dioxide 800,000 years ago or for the high value at the start of the industrial revolution is not known. But there have been intriguing suggestions that land clearing and farming expansion prior to the industrial revolution could have contributed to the difference.

Armed with a family of model configurations that accurately simulate past ice ages, the researchers asked - where we are headed? Their conclusion is that carbon dioxide emissions to date preclude another ice age for the next 50,000 years.

How does this relate to our current fix, with human-caused climate change leading to problems around the world? The story of ice ages is a fascinating window on how climate works and how well we understand it. The difference in global average temperature between a cold ice age and a warm interglacial is about 5°C. This is the magnitude of warming we can expect with business-as-usual emissions of heat-trapping gases in the 21st Century. But an ice age unfolds over thousands of years, not decades. And finally, the new study says no, there is no possibility that ice-age-type cooling will provide even a tiny bit of relief from human-caused warming in the 21st Century.