The challenge of making sea ice predictions

With the summer extent of Arctic sea ice shrinking, there has been an increased interest in the potential to forecast the scale of the annual September sea-ice minimum. Sea ice forecasts might, for example, enable people to better plan for open-water vessel operations in the Arctic.

But a study conducted by scientists from the National Snow and Ice Data Center, University College London, the University of New Hampshire and the University of Washington has found that sea-ice forecasts are not yet reliable.

“We found that in years when the sea ice extent departed strongly from the trend, such as in 2012 and 2013, predictions failed regardless of the method used to forecast the September sea ice extent,” said Julienne Stroeve, a senior scientist at NSIDC and professor at University College London.

A report from the study, published in Geophysical Research Letters, says that since 2008 a sea ice outlook program has been collecting sea-ice forecasts from a variety of sources. Each year, starting at the end of May, people have made monthly forecasts of the year’s sea-ice minimum using parameters such as the ice concentration, the ice type, ocean temperatures and atmospheric conditions. Forecasts have used a wide variety of prediction strategies, including statistical analysis and sea-ice modeling.

But the report says that, regardless of the prediction method used, the actual ice extent in September came quite close to predictions in years where the annual ice loss followed the long-term trend, but differed wildly from the predictions in off-trend years.

“Short-term predictions are achievable, but challenges remain in predicting anomalous years, and there is a need for better data for initialization of forecast models,” Stroeve said. “Of course there is always the issue that we cannot predict the weather, and summer weather patterns remain important.”

—Alan Bailey