cosima/cice4

Clone of The Los Alamos sea ice model (CICE). See http://oceans11.lanl.gov/drupal/CICE

cice4's Issues

Incorrect namelist error checking

The routine for reading in namelists (from cice_input.nml) incorrectly checks for errors in this loop:

https://github.com/OceansAus/cice4/blob/e7549ebd2044690a432cc67c1317c81cb194b750/source/ice_init.F90#L320

As written this does not pick up namelist errors correctly.

Smoothing of coupler fields needs to be optional

Nic created a smoother to fix low resolution atmosphere forcing creating discontinuities in the coupler fields:

https://github.com/OceansAus/cice4/blob/master/drivers/access-om/cpl_interface.F90#L463

This needs to be controlled by a namelist option

Support for 0.1 degree grid

To change resolutions CICE needs reconfiguration and rebuilding.

Model time reset when iced file present

When there is a iced file present in the RESTART directory the model time is read from the file here:

https://github.com/OceansAus/cice4/blob/access/source/ice_restart.F90#L324

This does not reset the value of runtime0. If the restart file is older than the current run this implies that no iced file will subsequently be written, because of the logic here:

https://github.com/OceansAus/cice4/blob/e7549ebd2044690a432cc67c1317c81cb194b750/source/ice_calendar.F90#L340

which determines the value of write_restart, which is tested before dumping a restart file here:

https://github.com/OceansAus/cice4/blob/e7549ebd2044690a432cc67c1317c81cb194b750/drivers/access-om/CICE_RunMod.F90#L438

if no restart is written, the next time the model is run the same thing happens. The effect of this is for a restart file to never be written again.