CICE does not set date correctly when use_restart_time = .false.

[nichannah]

As mentioned in COSIMA/libaccessom2#22 it looks like the driver part of CICE is not setting the year correctly when use_restart_time = .false..

The current year is being set using the forcing date rather than the experiment date.

[aekiss]

Reopening this issue.

Commit e3200e8 (merged in ab47343) is not conditional on use_restart_time = .false. and prevents restarting with use_restart_time = .true., so was reverted in d3e8bdf.

So as yet we don't have a neat way to deal with multiple forcing cycles (issue COSIMA/access-om2#149) or setting date correctly with use_restart_time = .false..

Should the first argument passed to input_data here

cice5/drivers/auscom/CICE_InitMod.F90

Line 136 in d3e8bdf

call input_data(accessom2%get_forcing_start_date_array(), &

be conditional on use_restart_time (i.e. pass get_forcing_start_date_array if use_restart_time is true, get_cur_exp_date_array if false)?

Slack discussion is here: https://arccss.slack.com/archives/C6PP0GU9Y/p1576623326004300

[nichannah]

I have fixed this. More explanation below.

In the case where use_restart_time = .false. the current experiment time will be set entirely by libaccessom2 (according to user input). When use_restart_time = .true. only the start year of the experiment (CICE calls this year_init) is determined by libaccessom2 and the current experiment year is calculated using this and the time-keeping values within the restart file. To handle this libaccessom2 needs to provide something different depending on use_restart_time.