Sea ice and waves in the Arctic: Modelling important interactions

Our Sea Ice Modelling group members Guillaume Boutin, Timothy Williams, Einar Ólason, and Pierre Rampal (Nansen Center & CNRS) recently published an article together with Camille Lique (IFREMER, Brest, France). They coupled the sea-ice model neXtSIM to a wave model and present exciting results!

Wind is one of the main drivers of sea ice moving and deforming in the Arctic. But not the entire Arctic is covered with sea ice. Some areas are open water, with the ocean surface being exposed to the winds. The Marginal Ice Zone does not have a continuous sea ice cover but both fragmented sea ice and open water and is particularly interesting for researchers. Naturally, waves occur in the open water areas. The warming of the Arctic is leading to increasingly larger open water areas in recent years as the sea ice extent is shrinking. With more open water where waves can form, the impact of waves on sea ice in the Arctic is becoming more and more interesting to researchers that investigate the dynamics of sea ice.

Why are the interactions between waves and sea ice interesting?

What waves do to a sea ice edge is the following: They push and bend the ice cover floating on the ocean and weaken it. The sea ice cover then starts breaking up into smaller pieces, it fragments. This can happen over long distances, up to dozens of kilometers. The current decrease of Arctic sea ice extent makes this process – sea ice fragmentation through waves – likely even more prevalent in the future. What we don’t know much about is: To what extent is the fragmentation impacting how sea ice drifts? A general assumption is that a sea ice cover that gets fragmented by waves is less resistant to deformation than un-fragmented ice. This was the starting point of the study conducted by researchers from the NERSC sea ice modelling group.

How can our sea ice model help here?

Their tool for the job was our in-house sea ice model, neXtSIM. neXtSIM has been developed at the Nansen Center over the past years, and the latest beneficial change in it is the use of a Maxwell elasto-brittle rheology. Rheology is the study and mathematical description of how hard material is being deformed and flows. The hard material here is sea ice, and this new mathematical description of its behaviour allows our researchers to simulate processes in the sea ice that resemble what happens in nature even more closely. One of the original features of this rheology is the use of a quantity called “damage”. A high level of damage represents the behaviour of sea ice cover with many cracks, that is easily deformed by ocean currents and winds. To assess the impact of waves on sea ice, the authors therefore linked the level of damage to the occurrence of sea ice fragmentation by waves. Another asset of the model is that the new rheology can accurately “remember” where the sea ice has been damaged before, and this is particularly important for the purpose of investigating how waves interact with sea ice.

What else do we need?

A sea ice model alone cannot give an answer here, so the authors coupled a wave model to the sea ice model. When coupling two different models together, they exchange information and influence each other’s results. This is exactly what is necessary for this kind of science. The wave model the authors chose is WAVEWATCH III®. It can use wind speed and direction data to predict the speed, orientation, and height of waves in the ocean. The coupling of a wave and a sea ice model allowed the authors to estimate where and when sea ice is broken – fragmented – by waves along the edge.

What do we learn about the interaction between sea ice and waves?

The authors ran several experiments, and they focused on the Barents Sea as a “test area” for their coupled model. The two videos below visualize how sea ice reacts to storms with high waves hitting the Marginal Ice Zone over the course of ten days. What their results show is that sea ice can be a lot more mobile after storms when it has been fragmented by waves, up to 50% faster. On average, accounting for wave-induced fragmentation increases the sea ice drift speed by 7%. The zone where this applies is the Marginal Ice Zone. And exactly this zone is the one we frequent most often: Ship operations, for example for fishing purposes, in these waters are common in the Arctic. Knowing about the sea ice drift in this area is crucial to make traffic safe and including wave-induced fragmentation into sea ice forecasts will be beneficial in the future.


In short: This study shows that in a coupled sea ice-wave model setup, waves significantly impact how mobile sea ice can be – it drifts faster in the Marginal Ice Zone when taking wave-induced fragmentation into account, on average 7% faster!

Above: Wave height (left) and sea ice concentration (right) during a 10-day period in the Barents Sea north and northeast of Svalbard. Sea ice concentration is the percentage of the ocean covered with sea ice. The purple line indicates the area of sea ice that experienced the most damage due to being fragmented by waves.

Above: Changes in maximum sea ice floe size (Dmax, left), damage quantity (middle) and sea ice drift (right) during the same period. The purple line indicates the area of sea ice that experienced the most damage due to being fragmented by waves. On the right (sea ice drift) you can see that larger drift speeds are observed within this “very damaged” area compared to the rest of the ice cover.


Denne studien viser at i en koblet sjøis-bølgemodell har bølger betydelig innvirkning på hvor mobil sjøis kan være – den driver raskere i den marginale issonen når man tar hensyn til bølgeindusert fragmentering, i gjennomsnitt 7 % raskere!


The Cryosphere:
«Wave–sea-ice interactions in a brittle rheological framework»