Tricky Sea Ice Predictions Call for Scientists to Open Their Data
With sea ice levels in the Arctic at record lows this month, a new report comparing scientists’ predictions calls for caution in over-interpreting a few weeks worth of data from the North Pole.
The Sea Ice Outlook, which will be released this week, brings together more than a dozen teams’ best guesses at how much sea ice will disappear by the end of the warm season in September. This year began with a surprise. More sea ice appeared than anticipated, nearing its mean level from 1979-2007. But then ice levels plummeted through May and into June. Scientists have never seen the Arctic with less ice at this time of year in the three decades they’ve been able to measure it, and they expect below average ice for the rest of the year.
But looking ahead, the ultimate amount of sea ice melt is hard to determine. Some trends, like the long-term warming of the Arctic and overall decreases in the thickness of sea ice, argue for very low levels of sea ice. But there are countervailing factors, too: The same weather pattern that led to higher-than-normal temperatures in the Arctic this year is also changing the circulation of sea ice, which could keep it in colder water and slow the melting.
“For this date, it’s the lowest we’ve seen in the record, but will that pattern hold up? We don’t know. The sea ice system surprises us,” said Mark Serreze, director of the National Snow and Ice Data Center.
The loss of summer sea ice over decades is one of the firmest predictions of climate models: Given the current patterns of fossil fuel use and the amount of carbon dioxide already in the atmosphere, sea-ice-free summers in the arctic are a virtual certainty by the end of century, and possibly much sooner. As the globe heats up, the poles are disproportionately affected. Warmer temperatures melt ice, revealing the dark sea water that had previously been covered. That changes the albedo, or reflectivity, of the area, allowing it to absorb more heat. That, along with many other feedback loops makes predicting change in the Arctic immensely difficult.
In 2007, the extent of sea ice in the Arctic declined rapidly. The drop from the previous year was so precipitous that it garnered worldwide attention and media coverage. In the last couple of years, the extent of sea ice in the Arctic, measured by the amount of square miles it covers, has recovered. This series of events, which underscored the year-to-year variability of the measurement, has made researchers cautious about describing events in the Arctic.
“In hindsight, probably too much was read into 2007, and I would take some blame for that,” Serreze said. “There were so many of us that were astounded by what happened, and maybe we read too much into it.”
Some good may have come out of the astonishing ice loss that year, though. It was in the wake of that shocking summer that the Arctic science community came together to try a new approach to climate science. All the big groups working on modeling the sea ice system would reveal their methods, and make predictions, allowing scientists to learn from each and see what worked.
“When this started in 2007, it was pretty scary for a lot of the scientists, putting these numbers out there,” said Helen Wiggins, program coordinator of the Arctic Research Consortium of the United States. “It’s a different way to do science. It’s more a community synthesizing exercise.”
With climate modelers increasingly under attack over the past year, the partnership to create the Sea Ice Outlook, which is organized through Study of Environmental Arctic Change, could be a model for other groups of scientists. Data-heavy, model-dependent fields exposing the mechanisms of the science to scrutiny may be squirm-inducing, but it has already yielded good results. “There has been cross-pollination with datasets and different types of data,” Wiggins said.
“It’s been a very valuable exercise because you put your cards on the table and see who is going to get it right,” Serreze agreed.
All but one of the predictions from the Sea Ice Outlook expect the minimum coverage of sea ice to fall between 4.2 and 5.7 million square kilometers (1.6 and 2.2 million square miles). One group predicts just a million square kilometers, which would easily break the 2007 record minimum of 4.3 million square kilometers. In 2009, the sea ice minimum was 5.3 million kilometers.