Skip to main content

Understanding the physics of first-year sea-ice growth using theoretical and computational fluid dynamics

Final Report Summary - SEA ICE CFD (Understanding the physics of first-year sea-ice growth using theoretical and computational fluid dynamics)

SEA-ICE-CFD: Final Project Report

Sea ice provides key feedbacks on climate processes, weather, and polar biogeochemistry. Observations show that young sea ice (less than 1 year old) makes up an increasingly large fraction of the sea ice cover.

This project aimed to understand processes relevant to the growth of young sea ice, by developing state-of-the-art models of the relevant fluid dynamics. As part of the broader career development of the Marie Curie Fellow, it also pursued related research questions on the impact of changing ocean conditions on the melting and stability of the Greenland and Antarctic ice sheets with implications for sea level rise.


We tackled the following questions:

(i) how do ice crystal dynamics impact the initial growth of sea ice in a turbulent ocean? Such granular ice makes up a significant fraction of Antarctic sea ice, but the formation rate is poorly understood.

(ii) how rapidly does dense salty water drain from sea ice into the ocean during ice growth? The sinking of dense water masses across the polar oceans forces large scale ocean currents with implications for ocean heat transport and climate.

A further goal was to develop a wider research program on ice and ocean interactions, applying approaches from fluid dynamics to tackle this significant climatic problem. The researcher supported by this proposal has developed a group investigating wide ranging facets of ice ocean interaction, and during the project has been re-appointed to a permanent position as Associate Professor in Physical Climate Science. Ongoing research is tackling the following subsidiary questions:

(iii) How do ocean conditions impact the melting of the Greenland and Antarctic ice sheets, via the coupling between buoyant flow of meltwater and the transfer of heat and salt towards the ice?

(iv) How does convective flow contribute to the transfer of heat in so-called melt ponds on Arctic sea ice? Melt ponds are puddles of meltwater on sea ice that provide a key feedback on energy transfer and ice melting rates


We have developed a range of new models necessary to study these key environmental phenomena.

We have discovered that ice crystals grow in a cold ocean at a rate around 10-100 times faster than had previously been anticipated.

We have developed a hierarchy of models of a mixture of ice crystals in the ocean, to determine the interaction between buoyant ice crystals rising to the ocean surface to form sea ice, and sinking cold and dense water currents that try to push ice crystals deeper down into the ocean. These models range from simple models of a population of ice crystals of different sizes in the ocean mixed layer, theoretical models of the impact of ice crystals on convection, and full numerical simulations of a turbulent flow with embedded ice crystals. The latter modelling approach has provided with a new tool to study interactions between sea ice floes and the ocean mixed layer, which we are pursuing in an ongoing spin off PhD project.

By applying the above models, we have shown that the nucleation rate of ice crystals presents a major control on the initial growth rate of ice, and turbulent mixing can enhance the ice growth rate by promoting collisions, fragmentation and the formation of new ice "seeds".

We have also studied the processes that control when salty water starts to drain from sea ice into the ocean, and found that the rate of ice cooling is a key factor. We have developed theoretical and computational models of the subsequent flow, to determine how rapidly salt drains into the ocean.

A key finding supported by both theoretical and computational analysis, is that the drainage of salty brine from the ice tends to localise near the base of the sea ice. In addition to providing new understanding of the salt fluxes to the ocean as ice grows, this has implications for sea ice ecology, because the resulting flow has important consequences for nutrient transfer and biological growth within the porous sea ice.

Finally, we have translated our ideas to two new areas of ice-ocean interaction. In ongoing work, we have also extended our modelling approach to describe convective flows in melt ponds on sea ice, using these to interpret observational data and predict the heat transfer down into the ice that drives ice melting. This quantifies a key feedback in sea ice evolution.

We have also provided a new method to predict the melting rates of glaciers running out into a stratified ocean, and used meltwater plume models to examine the stability of melting ice shelves. These results can be used in projections of ice sheet melt and sea level rise.