The area of the Arctic covered by ice is set to reach a record low this year* – and much of the ongoing decline is down to man-made global warming. That’s what recent research into the role of three potential natural causes of change in ice coverage done by Jonny Day from the University of Reading, UK, and his colleagues suggests. In a paper published online in scientific journal Environmental Research Letters last month, they studied these three existing sea and wind movement patterns in five different climate models. They found that one in particular, the Atlantic multi-decadal oscillation (AMO), could be responsible for between one-twentieth and three-tenths of the decline in sea ice since the 1970s. “Our work suggests that the AMO has contributed to getting us to this low sea ice state, but does not tell the whole story,” Jonny told Simple Climate. “It is still likely that man-made global warming is the major contributor to the dramatic decline in sea ice.”
Arctic sea ice loss has surprised researchers, as it’s faster than predicted by leading climate models gathered together by the UN Intergovernmental Panel on Climate Change (IPCC). That made Jonny and his co-workers from the Yokohama Institute for Earth Sciences and University of Tokyo in Japan want to look at why this was. In particular, they wanted to see how much might be down to hard-to-predict ‘internal’ climate patterns, rather than problems with measurements of the ice or models. “The large mismatch between climate model projections of sea ice and observations in the last IPCC report was concerning,” Jonny said “My colleagues and I thought that internal or ‘natural’ climate variability may play a role.”
The possible sources of that natural variability include the Arctic Oscillation (AO), which varies between stronger and weaker winds circulating the Arctic. The other two sources come from the ocean, where the AMO is a long-term cycle of temperature change. The final source, the Atlantic meridional overturning circulation (AMOC), is a large-scale ocean flow pattern that carries warm upper waters north and returns cold, deep water south. But a limited set of measurements available for the AMO and AMOC poses researchers a problem. “The oceanic causes of variability we discuss are not well observed, making such analysis difficult,” Jonny said.
Clearer Arctic forecasts
Instead, Jonny’s team’s work relied on simulating the impact of these patterns in five climate models in conditions typical before humans started emitting the warming-causing greenhouse gas CO2. To gather enough information to produce a reliable result, they ran the models for at least 500 years. From the results of their simulations they could compare the changes in each against the sea ice area the changes produced.
The researchers didn’t see a relationship between the AO and sea ice area. But they did find that the AMOC and AMO’s effects on sea ice are closely linked, and can be thought of as two ways to look at the same changes. And together, they both are strongly linked to how Arctic ice cover varies. Armed with the information on the strength of that effect and direct measurements of the AMO and sea ice area since 1979, they estimated that how much of the reduction in coverage in September, when the ice is at its smallest, the AMO might have caused. Since 1979, sea ice has shrunk by one-tenth every decade. By contrast, Jonny’s team calculated that the AMO alone at most would have been responsible for sea-ice area shrinking by just 3 per cent per decade. “This range of decline is arguably due to natural variability and may have happened regardless of the influence of man-made climate change,” Jonny explained.
While this does underline our responsibility for changes in the Arctic, that isn’t Jonny’s main aim. Instead, his interest comes largely because Arctic climate predictions are highly valuable to the people that live and work there, and also for informing important decisions politicians make about the region. “I am continuing to work on sea ice variability and predictability with a view to improving sea ice forecasting systems as part of the APPOSITE project,” Jonny explained. “The mechanism provided in the paper could possibly lead to improved short term predictability of sea ice, and helps us to interpret both observations and model forecasts of sea ice.”
* – The record minimum was reached the day after this blog entry was posted.
J J Day, J C Hargreaves, J D Annan and A Abe-Ouchi (2012). Sources of multi-decadal variability in Arctic sea ice extent Environmental Research Letters DOI: 10.1088/1748-9326/7/3/034011