CESM2 historical output data "time" variable not matching time period indicated by file naming convention...

[milunita]

Good afternoon!

I am kindly requesting insight into the "time" variable in CESM2 historical output data. While the filename ('psl_Amon_CESM2_historical_r4i1p1f1_gn_185001-201412.nc') indicates that the historical run extends from January 1850 to December 2014, my time conversion using Python's datetime module yields a start date of October 24, 1848. Has anyone else run into a similar issue? The historical file's metadata states that the "time" variable is days since 0001-01-01, which is what I use to calculate the date at each timestep. Any and all help is greatly appreciated. Thanks in advance!

Here's my code:

hist_file = nc.Dataset('psl_Amon_CESM2_historical_r4i1p1f1_gn_185001-201412.nc')
time = hist_file.variables['time'][:]

from datetime import date, timedelta

monthly_dates = []
start = date(1,1,1)
for i in time:
days = i
delta = timedelta(days)
offset = start + delta
monthly_dates.append(offset)
monthly_dates[0]

Output is: datetime.date(1848, 10, 24)

 

[strandwg]

Good afternoon!

I am kindly requesting insight into the "time" variable in CESM2 historical output data. While the filename ('psl_Amon_CESM2_historical_r4i1p1f1_gn_185001-201412.nc') indicates that the historical run extends from January 1850 to December 2014, my time conversion using Python's datetime module yields a start date of October 24, 1848. Has anyone else run into a similar issue? The historical file's metadata states that the "time" variable is days since 0001-01-01, which is what I use to calculate the date at each timestep. Any and all help is greatly appreciated. Thanks in advance!

Here's my code:

hist_file = nc.Dataset('psl_Amon_CESM2_historical_r4i1p1f1_gn_185001-201412.nc')
time = hist_file.variables['time'][:]

from datetime import date, timedelta

monthly_dates = []
start = date(1,1,1)
for i in time:
days = i
delta = timedelta(days)
offset = start + delta
monthly_dates.append(offset)
monthly_dates[0]

Output is: datetime.date(1848, 10, 24)

That version of CESM has a fixed 365 day calendar with no leap years, so your computation needs to take that into account.

The "time" coordinate variable has the attribute "noleap" to signify this property.

 

[milunita]

That version of CESM has a fixed 365 day calendar with no leap years, so your computation needs to take that into account.

The "time" coordinate variable has the attribute "noleap" to signify this property.

Hi Gary! Thanks for your reply - this resolves my question. Thanks again!