question about SST/ICE dataset for CAM run

[huihuojia@163_com]

The dataset I used for CAM run is "sst_HadOIBl_bc_1.9x2.5_1850_2012_c130411.nc", and this is a monthly dataset. When I do the forecasting, for example, run from 00Z 01 JAN to 00Z 11 JAN with the time step dt=1800, Will the SST/ICE be persistent in the whole forecast, or it will be interpolated into each timestep? and if it is interpolated, which method does it use?

Are the variables ice_cov/SST_cpl will be used in the forecast other than ice_cov_prediddle/SST_cpl_prediddle?

Thanks

 

[hannay]

The SST and ICE will inerpolated between the mid-month values by a linear interpolation.The variables used will be ice_cov/SST_cpl. They are obtained used the icesst tools. The Hurrell dataset is specifically created for use with CAM and include the "diddling" procedure that ensures that the monthly mean of the daily-interpolated SST anomalies matches the monthly SST anomalies. This follows the protocol for AMIP simulation described in: Taylor, K.E., D. Williamson, and F. Zwiers (2000): The sea surface temperature and sea-ice concentration boundary conditions for AMIP II simulations. PCMDI Report No. 60 and UCRL-MI-125597, Lawrence Livermore National Laboratory, Livermore, CA, 25 pp.
pdf file available at  http://www-pcmdi.llnl.gov/publications/pdf/60.pdf

  

 

[dwhittle@mit_edu]

I'm running a prescribed ice/sst simulation and have found that my monthly averaged SSTs (from CAM history files) consistently overestimate SSTs in the marginal ice zones (vs. the pre-diddled presribed SSTs). Am I correct that this is probably a result of approximations made during the diddling procedure in the grids with partial ice cover (i.e. the nonlinear newton-raphson approach)?Thanks,Dave

 

[hannay]

HI Dave When you construct your dataset with prescribed SSTs (from CAM history files), are you using TS from CAM or are you using the real SST? If you are using TS, you can have problem along ice edge and around land.  I'm running a prescribed ice/sst simulation and have found that my monthly averaged SSTs (from CAM history files) consistently overestimate SSTs in the marginal ice zones (vs. the pre-diddled presribed SSTs). Am I correct that this is probably a result of approximations made during the diddling procedure in the grids with partial ice cover (i.e. the nonlinear newton-raphson approach)?

 

[dwhittle@mit_edu]

Hannay,Thanks for the response. I'm using the CAM SST field, the values over land match the prescribed SST file.Dave