CAM level information in history files

[bidyut]

Hi All,

I ran a CAM test run with (compset=F2000climo). The output files are having level information as given below:

double lev(lev) ;
lev:long_name = "hybrid level at midpoints (1000*(A+B))" ;
lev:units = "hPa" ;

lev = 0.00802796150392404, 0.00779076150022784, 0.00755145382403144,
0.00731010321640146, 0.00706677497108843, 0.00682153491686288,
0.00657444939970631, 0.0063255852648626, 0.00607500983875275,
0.00582279091076254, 0.0055689967149023, 0.00531369591134656,
0.00505695756786018, 0.00479885114111267, 0.00453944645788773,
0.00427881369619243, 0.00401702336627252, 0.00375414629153525,
0.00349025358939137, 0.00322541665201326, 0.00295970712702243,
0.00269319689810577, 0.00242595806556778, 0.00215806292682663,
0.00188958395685368, 0.00162059378856672, 0.0013511651931809,
0.00108137106052142, 0.000811284379303112, 0.000540978217386523,
0.000270525702008362, 3.38174281594439e-05 ;

These values surely do not look like in hPa units.

Can anyone please comment what am I missing here?

Thank you.

 

[cacraig]

I agree that these values look suspect. A recent run that I made had lev ranging from 5.9e-6 to 992.5.

Unfortunately you did not give us enough information to help you debug your problem. Please include the details listed here: Information to include in help requests

 

[bidyut]

Hi Cheryl,

Thank you for your reply.

I did some digging and I found out something funny. I had actually copied my output files from where it was archived to another location in the machine and the file did not get copied properly because of which the level values were showing unexpected values. Apparently, my test run is all good and the level issue which I had reported in my post is not related to any error in my run. It is, most probably, related to some memory crunch in my machine. I could see the expected level values when I checked the output files in the archived location.

Nonetheless, I am attaching the the version details and the log files in the attachment for your reference. Kindly go through the attachments and please confirm if my run looks OK to you. JFYI, I had killed the job at the end as I did not want it to run any longer.

Thanks you.

 

[cacraig]

As I normally am concerned with the atmosphere, I usually look at the atm.log file (and the cesm.log if a run crashes). You neglected to attach that file. In a typical run, after the initial startup phase, I usually see a few lines of text in the atm.log file for every timestep with occasional information during the history files writes. If that is what you see, then there is no need to attach your atm.log file.

Looking at your version_info.txt, it looks like you are running a standard checkout with minimal changes to support your machine. If you have followed and validated your results using the steps detailed at: 6. Porting and validating CIME on a new platform — CIME master documentation, then your port should be okay.

Since I am a software engineer, I am unable to tell you other than that if your run is OK or not. I am not qualified to validate your results. If you have concerns, I suggest you run one of the supported compsets and plot out key variables and confirm for yourself that they are what you expect.

 

[bidyut]

Got it. Thanks.