Using cesm output to drive CLM offline ( I - compset) solar variables

[boyle5@llnl_gov]

The CLM manual indicates that setting histaux_a2x3hr and histaux_a2x3hrp to true will generate output from the coupler every 3hr suitable for driving the CLM offline.
This works fine but the resulting files contain no solar forcing data.
In the file ( cesm1_1_1/models/atm/datm/bld/datm.template.streams.xml) describing the data models for the CLM there is an example of exactly what I want to do.
The stream data source : CPLHIST3Hrly.precip corresponds to that produced by setting histaux_a2x3hrp to True.
The stream data source : CPLHIST3Hrly.nonSolarNonPrecip corresponds to that produced by setting histaux_a2x3hr to True.
But there is an additional stream for this data model:
CPLHIST3Hrly.Solar which has the solar forcing every 3h. By setting histaux_a2x equal to true I can get the same variables from the coupler as the CPLHIST3Hrly.Solar stream
but every time step - I cannot see how to get the solar every 3h easily.
Of course, I can compute the necessary solar values from the time step information but the CLM manual seems to say that this should not be necessary.
I suspect that the way the data model files are broken out that the solar is special and perhaps this extra step is needed.

So, is there an easy way to get the solar? I am missing something obvious?

 

[slevis]

I did this kind of run a long time ago. At the time all the variables that I needed appeared in the same files, e.g., b40.rcp8_5.1deg.006.cpl.ha2x3h.2005-01.nc. Are you sure this is not the case for you?

Sam Levis

 

[boyle5@llnl_gov]

The files of the type "cpl.ha2x3h' from my run do not have the rain/snow or solar variables.
I did an ncdump and looked carefully.

I note that the CPLHIST3Hrly data streams for the atmospheric data model to drive the CLM are broken into
CPLHIST3Hrly.Solar, CPLHIST3Hrly.precip, CPLHIST3Hrly.nonSolarNonPrecip but all three streams point to a single file
of the type 'cpl.ha2x3h' . So in the past the cpl.ha2x3h might well have had all the data as the CLM user guide indicates.

The cesm version I am running is from the developers svn repository tag cesm1_1_beta15.
So perhaps the ha2x3h output is broken in this pre-release. Something to look for in the release.


I was just hoping that there was something that was common knowledge that I overlooked.
It is going to be a bit of work to make the land forcing.

Thanks for your help.

 

[nick]

Hi, I'm also having this problem. The manual (examples 4.8 and 4.9) suggests that setting only histaux_a2x3hr to .true. will give you the variables required, but this isn't the case. Has anyone gotten over this yet?

 

[nick]

After conversing with some of the CLM staff I've been told this is a bug (identified a while back in CESM1.1 but not 1.0.x), the report and fix for which can be found herehttp://bugs.cgd.ucar.edu/show_bug.cgi?id=1733 The CESM1.1 userguide also mentions this bug.http://www.cesm.ucar.edu/models/cesm1.1/clm/models/lnd/clm/doc/UsersGuide/x9068.html