maja_vilkas@gmail_com
New Member
HiI am failing to add addiontal ocean variables (e.g velocities) with daily frequency. Would you be able to help me with that?
-
The www2.cesm.ucar.edu website is currently down for maintenance
adding daily ocean output
- Thread starter maja_vilkas@gmail_com
- Start date
ashields@astro_ucla_edu
New Member
Hi, I hope it's ok that I responded to this reply you posted for another user. I posted an issue on the CAM forum that relates to this topic, and I wondered if you might be able to help me with it. I read the document in the link you provided above, but I'm having trouble applying the output frequency info to my specific situation of simulations of planets with non-Earth like rotation periods. Here is what I posted on the other forum:I'm running simulations with CAM4 for planets with different rotation periods, and want to output two fields (surface pressure and downwelling top-of-atmosphere shortwave flux) at an interval such that I have data over a semi-diurnal and diurnal time period. For a 24-hr rotation period, following this website - http://www.cesm.ucar.edu/models/ccsm4.0/ccsm_doc/x2500.html - I've modified the file cam.buildnml.csh to output this data hourly, every 24 hours, like this:
fincl2 = 'PS:I','FSDTOA:I'
nhtfrq = 0,-1
mfilt = 1,24
This works for a 24-hr (Earth-like) rotation period. I get hourly data, and each day's .nc file contains 24 hours worth of data, which I can plot over a semi-diurnal (12 hours) and diurnal period. But for simulations at rotation periods of 25 days, 50 days or 100 days (for example), I'm stuck. How should I modify 'nhtfrq' and 'mfilt' to make sure I have data output over a full rotational period, if the periods are longer than one Earth day? If I leave it as above, is CAM taking the relevant length of the day into account (I've specified the length of the day in Earth days in the SourceMods/src.cam/zenith.F90), and dividing 25 days (for example) by 24, making every "hour" equal to 1+ day? Or should it be modified to be something like this:fincl2 = 'PS:I','FSDTOA:I'
nhtfrq = 0,-24 (output every 24 hours)
mfilt = 1,25 (write 25 time samples to each history file)
I've tried the above, but am not sure I'm getting data over a full "day" (25, 50, or 100 Earth days). I'd appreciate any help trying to figure this out.Thanks,Aomawa
fincl2 = 'PS:I','FSDTOA:I'
nhtfrq = 0,-1
mfilt = 1,24
This works for a 24-hr (Earth-like) rotation period. I get hourly data, and each day's .nc file contains 24 hours worth of data, which I can plot over a semi-diurnal (12 hours) and diurnal period. But for simulations at rotation periods of 25 days, 50 days or 100 days (for example), I'm stuck. How should I modify 'nhtfrq' and 'mfilt' to make sure I have data output over a full rotational period, if the periods are longer than one Earth day? If I leave it as above, is CAM taking the relevant length of the day into account (I've specified the length of the day in Earth days in the SourceMods/src.cam/zenith.F90), and dividing 25 days (for example) by 24, making every "hour" equal to 1+ day? Or should it be modified to be something like this:fincl2 = 'PS:I','FSDTOA:I'
nhtfrq = 0,-24 (output every 24 hours)
mfilt = 1,25 (write 25 time samples to each history file)
I've tried the above, but am not sure I'm getting data over a full "day" (25, 50, or 100 Earth days). I'd appreciate any help trying to figure this out.Thanks,Aomawa
You could try to specify nhtfrq in number of timestep (instead of number of hours). When NHTFRQ(1) = 0, the file will be a monthly average. Only the h0 file series may be a monthly average.
When NHTFRQ(i) > 0, frequency is input as number of timesteps.
When NHTFRQ(i) < 0, frequency is input as number of hours.
When NHTFRQ(i) > 0, frequency is input as number of timesteps.
When NHTFRQ(i) < 0, frequency is input as number of hours.
ashields@astro_ucla_edu
New Member
Thank you for this suggestion. I thought though that the timestep, as specified in Buildconf/cam.buildnml.csh by the field "dtime", was 1800 seconds, which is 0.5 hr. How would specifying nhtfrq in number of timesteps help me to cover the full planet day more completely? Or is the timestep something different than the "dtime" parameter (which I haven't been comfortable messing with in the simulation)?