Hi!

I have a student who wants to look at the sensitivity of sea ice in the model. I can't seem to find any examples in the tutorials I have looked at, and I'm afraid that's one part of the model I've never messed much with.

Can someone think of a quick/easy change to the model or initial conditions that would be a good sensitivity test for a student to do?

Thanks much,

Natalie

—

Hi Natalie,

We do have some samples in the tutorial of sensitivity experiments with the CICE model. I'm not sure what you are thinking of exactly, but one common thing we do is tweak the surface albedos. With the newer radiation code, it is actually the inherent optical properties of snow, bare ice, and ponds that we tweak. For example, with the snow we have the following:

1. R_snw

This is the number of standard deviations away from the default nonmelting snow grain effective radius where:

Rsnw_nm = R_snw_base - R_snw*Std_snw

with the constraint Rsnw_fresh <= Rsnw_nm <= Rsnw_melt

where R_snw_base = 500um, Std_snw = 250um, R_snw_fresh = 100um and R_snw_melt is another tunable parameter. Out of the box, R_snw = 1.5 so the default nonmelting snow grain radius is 125um. The bounds on Rsnw_nm mean that values of R_snw larger than about +/- 1.6 do nothing. The larger the effective radius, the lower the dry snow albedo.

2. Rsnw_melt_in, dT_mlt_in

There is a linear dependence of the snow grain radius on temperature. dT_mlt determines the temperature at which snow melt starts and Rsnw_melt is the maximum snow grain radius.

rsnw = Rsnw_nm + (Rsnw_melt-Rsnw_nm)*fT

fT = -min((Tmelt-Tsfc)/dT_mlt - 1, 0)

So, out of the box, dT_mlt = 1.5 and Rsnw_melt = 1500um. This means fT = -min((0-Tsfc)/1.5 -1, 0) i.e. fT is 0 at Tsfc = -1.5C and 1 at 0C. This implies that rnsw is Rsnw_nm at -1.5C and Rsnw_melt at 0C.

There are similar parameters of R_ice and R_pnd for bare ice and ponds respectively, but the snow parameters are the most powerful in terms of the sea ice sensitivity. Some of this is in the Holland et al. 2012 special issue paper in the Journal of Climate. Everything here is in the NCAR Tech Note, Briegleb and Light 2007.

Regards,

Dave

Thanks: that's just what I needed: sorry I missed it in the tutorial!

Natalie