Method of Prediction：Tracking of sea ice from winter to spring

Figure１：Predicted sea ice cover on September 11, 2017.

Figure 3：Animation of predicted Arctic ice extent from the July 1 to the September 15. Yellow and green lines indicates the ice edge of the same day for 2015 and 2016.

Figure 2：The interannual change of minimum extent since 2003. Value of 2017 is predicted one.

Figure 6：Distribution of particles for April 30, 2016, which are first arrayed over the ice-covered area on December 1, 2015 and are moved based on the satellite-derived daily-ice velocity. The color bar indicates the thickness of the ice on December 1, 2015.

Figure 7：Distribution of particles for April 30, 2017, which are first arrayed over the ice-covered area on December 1, 2016 and are moved based on the satellite-derived daily-ice velocity. The color bar indicates the thickness of the ice on December 1, 2016.

Figure 5：Animation of distributed particles’ movements during December 1, 2015 through April 30, 2016. The color bar indicates the thickness of the ice on December 1, 2016.

[Japanese]

The way of this prediction is based on our recent research (Kimura et al., 2013). First, we expect the distribution of the sea ice thickness in spring from the movement of sea ice from December to April. Then, we predict the summer ice area depending on the assumption that thick ice remains later and thin ice melts sooner than the average.

To calculate the movement of the sea ice, we first distributed particles homogeneously over the Arctic sea ice on December 1, and traced the trajectories of the particles to the end of April by using the satellite derived daily ice velocity (Figure 5). Data from satellite microwave sensors, AMSR-E (2002/03-2010/11) and AMSR2 (2012/13- 2016/17) was used to derive the daily ice velocity. Based on the relationship (linear regression line) between yearly particle concentration on April 30 and ice concentration on a specific day in summer, we predicted the ice area on the day from the particle concentration in April of this year.

In addition, we considered initial ice thickness on December 1 calculated by the method of Krishfield et al. (2014). It is defined as “virtual ice volume” that the sum of thickness of all the particles in 240 kilometer radius. In this prediction, we predicted the summer ice concentration  using this virtual ice volume instead of the particle concentration.

Sea ice extent on September 11, which is the minimal area phase of Arctic sea ice, is expected to be 4.58 million square kilometer, which is 8% larger than the last year.

Sea ice in the Laptev Sea and Chukchi Sea is expected to be thin and retreat quickly. On the other hand, sea ice in the East Siberian Sea will retreat slowly with nearly same speed as normal years. Sea routes of the Russian side will open around August 20, which is almost equal to the last year.

On the Canadian side, sea ice in the Beaufort Sea is expected to be thick and retreat slowly compared with the last year. The sea routes of the Canadian side except for Canadian archipelago will open around the July 20.

※Prediction map is also available in Arctic Data Archive System of National Institute of Polar Research.

References

Figure 4：Animation of the ice edge (30% of ice concentration) for 2003-2016 (yellow contours) and 2017 (predicted: red contour).