Where can I find a fully spun-up water age tracer for CMIP6 output from ACCESS-ESM1.5 or ACCESS-CM2?
The water age (variable_id = "agessc") in historical CMIP6 output for ACCESS-ESM1.5 and ACCESS-CM2 looks a bit too young, as if it had not had enough spin-up time. See for example the age in January 2000 for r1i1p1f1 ACCESS-ESM1.5, which barely reaches 200yr at 1000m in the Pacific (instead of the expected 1000+ years in the oldest waters of the North Pacific):
I’ve also looked at the piControl output it looks too young as well (after 1000+ years of simulation, r1i1p1f1 member):
There’s also the issue that the historical water age is roughly 1/3 younger than the preindustrial water age but I am assuming this was an issue with the “initial” water age of the historical run being picked somewhere?
Just in case this is useful, this is the code for reproducing historical age plot (I’m omitting the usual setup (xarray, intake, matplotlib, etc.):
agescc_historical_cat = cat.search(
source_id = "ACCESS-ESM1-5",
experiment_id = "historical",
realm = "ocean",
variable_id = "agessc",
member_id = "r1i1p1f1",
file_type = "l",
)
ds_historical = xr.open_mfdataset(agescc_historical_cat.df.path)
age2D_historical_2 = ds_historical["agessc"].sel(time=datetime.datetime(2000, 1, 1), lev=1000, method='nearest')
age2D_historical_2.plot()
PS: The reason I’m looking for this is that I am building ocean transport matrices from CMIP output here and I would like to use spun-up water age output from CMIP as the ground truth for calibrating the diffusivities of the transport matrices, as done by @matthew.chamberlain for ACCESS1.3 (Chamberlain et al., 2019).