Improved key process in representing Arctic warming (D3.5)

In the work presented in this deliverable the Blue-Action teams focused on improving the representation
of some of the most important physical processes which contribute to Arctic warming within the climate
models used by the consortium. The two processes we addressed were the effect on the atmospheric
state due to the presence of leads in the sea ice cover and turbulence under strongly stable thermal
stratification.

The work done: We first analysed the results of previously performed large eddy simulations which
resolved the turbulence over leads to determine the effect leads have on sensible heat flux from open
water. Because of the effect of three-dimensional structures in the turbulent mixing above leads, the
heat flux coming from leads can be amplified compared to the fluxes one would get from open water
under the same air-sea temperature difference. The amplification effect strongly depends on the width
of the lead, with the largest effect occurring for leads of widths around 1.4 km. We assessed the
functional sensitivity of this amplification effect to key parameters used in the turbulence-resolving
model, including the length scale for the convective boundary layer, which characterizes the background
stability in the atmosphere.

We combined this relation between the amplification effect of heat fluxes as a function of lead width
with observed distributions of lead widths. These were taken from the peer-reviewed literature.
Together, this gives us a scheme to describe how the presence of leads affects the surface sensible heat
flux, and this depends upon the concentration of sea ice and the background atmospheric stability. We
implemented this scheme in four climate models (NorESM, EC-Earth3, IPSL/ LMDZ6A and CAS-ESM/
IAP4) and tested the scale of the effect using multiple single-model ensemble simulations of historical
climate.

The key findings: The presence of leads in sea ice dramatically alters the surface energy balance in the
Arctic. There is a large seasonal cycle to the effect of the presence of leads, because the flux from the
leads depends strongly on the background stability in the atmosphere. In the winter when the
atmosphere is often strongly stably stratified, the leads greatly amplify the surface sensible heat flux
coming from open water. In the summer there is the opposite effect and the generally weaker
atmospheric stability reduces the flux coming from leads. The net effect is to increase near-surface
temperatures over sea ice in winter, with little or no change in the summer. Therefore this scheme may
be used to address the long-standing winter cold bias over ice in many contemporary global climate
models, as shown in CMIP6.

Linked Zenodo records: Guidelines for improving the representation of surface heat flux in the Arctic https://zenodo.org/record/4728073