Wednesday 13. February 2013
GFI, East Wing, 3. floor, BCCR meetingroom
The workshop is to coordinate future activity in Bergen on sea ice, and give a coordinated
input to a national meeting in Tromsø 18-19 February 2013.

Tentative schedule:

09:30 - 09:45 Welcome and purpose of the day - Lars H
09:45 - 10:00 Bergen Polar Ice Group - why and what now ? - Peter M.
Coffe Break, 20 min
10:20 - 10:40  New and old observations - Storfjorden versus the entire Arctic - Lars H.
10:40 - 11:00  The NorESM CMIP5 results and trouble on the Siberian shelves - Mats
11:00 - 11:20  The ROMS (COAWST) system, air-ice-ocean-wave coupling - Stefan
11:20 - 11:40   The TOPAZ system - ice break up and healing ? - Francois 
11:40 - 12:00  Sea Ice reology -  news from the Nemo model development - Sylvain

Lunch 1 hour

13:00 - 14:00 Suggested national input - Lars H & Pierre

Coffe Break
14:30 - 15:00 More discussion and suggestions if needed


Summary : 

We had a good meeting, and about 20 peopele showed up.

A more detailed summary
has been sent out to all participants.

We agreed to change the name of the group to
Bergen Ice Group (BIG),
and create an e-mail list. This is now working and all the
people on the list
can send an e-mail to:

If you would like to be a member tell the coordinator (Lars H. Smedsrud).

Observations: A good range of observations are available, from loss of Arctic

Sea Ice area and how uncertain these are, to CTD profiles from drifting bouys
in the Arctic. Generally the ice remains fairly think north of Greenland, but
north of Svalbard it has thinned, quite likely due to inflow of Atlantic Water.
Process studies from Storfjorden (~50 km wide fjord) on Svalbard
might be of use when working with brine release and polynya dynamics.

NorESM: Many results are 'as good as it gets' for the Arctic, but the
inflow of Atlantic Water seems to be too low, and there is too much
thick ice on the Siberian shelves. Ongoing work has removed some of the
thickness bias, and there are suggestions to use a 'Neptun' velocity
parametrization to improve the flow along contintal boundaries.

COAWST: The multi-coupled model is up and running, and has created
some very interesting results north of Svalbard with ocean=>air fluxes
of about 2000 W/m2. Efforts are ongoing to setup longer runs, and
using the ocean-air coupling "only".

TOPAZ: This model is used for operational forecasts in the Arctic region,
and is also run as 'reanalysis'. Break-up of ice floes due to waves have
been parametrized, but waves also seem to increase melting, and how
this happens is an "open issue". A new type of reology (Elestic-Brittle, EB) has
performed well in a french model system (NEMO) and will be implemented
in TOPAZ. This improves the fracturing properties of the sea ice, and also
works for quite coarse resolution (~100 km grid size).

At the end we had a more general discussion, and Justin sparked some rather BIG questions, so that is included
here as well. Even though we might not get to work on them very soon they were good overall questions:

a) Evaluation shows that CMIP3 models were poor and CMIP5 models are now better, but a large spread remains among models.
What (thermodynamic?) tuning has been applied and how consistent are these with observations?

b) Can applying a range of different stress/strain/rheology fields fundamentally change the simulated Arctic sea ice distribution?
Can the stress/strain be evaluated, and  how does sea ice thickness and deformation compare?

c) How and how much does the atmosphere "care", or respond to changes in sea ice cover ?
How do we best evaluate models with observations of SW, LW and turbulent heat fluxes?
(Different parts of the seasonal cycle and over different "types" of sea ice)