2021 Second report complement : Analysis of sea ice age and sea ice prediction based on it
Generally, the older ice becomes thicker (e.g., Tschudi et al., 2016). Thus, the distribution of sea ice age can be a valuable information for understanding sea ice conditions in summer (e.g., Kimura et al., 2020).
The method to estimate sea ice age
Daily sea ice drifting velocity and sea ice concentration data are used to estimate sea ice age. These are derived by processing satellite remote sensing data by the microwave radiometers, AMSR-E and AMSR2. The data is obtained through the ADS (Arctic Data archive System). Sea ice velocity is calculated by the maximum cross-correlation method. Sea ice concentration is calculated by Bootstrap Algorithm（Comiso, 2009）.
The ice age is estimated by tracking sea ice positions up to 4 years back in time. In an example shown in Figure 2, the particles are placed where sea ice concentration is over 15% on May 31st 2016. Then, backward displacement of particles was calculated on one-day time steps using the ice velocity data. Since the velocity is given in 60km square grids as shown in Figure 1, velocity at an arbitrary position is calculated by interpolating the velocity data at surrounding grids. By tracking backwards in this way, the particle reaches to where there is no sea ice, i.e. ice concentration under 15% (Figure 2). At that point sea water starts freezing and generates sea ice. In this manner, sea ice age is estimated to be the period from that date to May 31, 2016, the starting date of backward tracking.
Spring ice age distribution and summer sea ice condition for recent 5 years
Spring ice age distribution of each year significantly reflects the ice age distribution at minimum sea ice season in the previous year. Sea ice which survives in summer moves toward Greenland, the Canadian Arctic Archipelago, the Beaufort Sea, and the Chukchi Sea during the early spring of the following year. This movement varies greatly from year to year.
When looking at the trends of multi-year ice melting, it melted the most in 2016, but not so much from 2017 to 2020. For the Beaufort Sea, the older sea ice tends to remain more at sea ice minimum, except in 2016. On the other hand, like in 2016, sometimes very old ice drastically melts. This means that we need more analysis on the ice melting processes to predict sea ice distribution in summer.
Predicted sea ice condition for summer of 2021
Figure 4: Sea ice age distribution on May 31, 2021. The red to light-blue area shows multi-year ice that has survived the previous summer, while the dark-blue shows first year ice.
As of May 31, very old sea ice covers along the coast from Greenland to the Canadian Arctic Archipelago and from the Beaufort Sea to the Chukchi Sea. This ice area was left unmelted in 2020. Since similar condition is seen in the previous year, 2020 (the bottom pair of Figure 3), old sea ice will possibly remain in the Beaufort sea as it did last year. Even in case old sea ice melts significantly as seen in 2016, thick old ice may remain sparsely in this area.
Reference
- Comiso J.C. Enhanced sea ice concentrations and ice extents from AMSR-E data. Journal of Remote Sensing of Japan 29, 199–215, 2009.
- Kimura N., Nishimura A., Tanaka Y. and Yamaguchi H., Influence of winter sea ice motion on summer ice cover in the Arctic, Polar Research, 32, 20193, 2013.
- Kimura, N., Tateyama K., Sato K., Krishfield R.A and Yamaguchi H., Unusual drift behaviour of multi-year sea ice in the Beaufort Sea during summer 2018, Polar Research, 39, 3617, 2020.
- Tschudi M.A., Stroeve J.C. and Stewart J.S. Relating the age of Arctic sea ice to its thickness, as measured during NASA’s ICESat and Ice Bridge campaigns. Remote Sensing 8, 457, 2016.