[Go-essp-tech] QC L2 Checks on MOHC cmip5 data

[martina.stockhause]

 Dear Karl,

 the plots show on the upper part the global mean, minimum and maximum 
 values of the parameter checked and on the bottom plot the error code. 
 An error code of zero means no errors detected. The "pass" or "failed" 
 of the QC check is written into the experiment log file.

 Your question of an available QC criteria description is to be 
 answered: We are currently working on that. Since the results of all 
 atomic datasets belonging to an experiment have to be viewed in order to 
 decide if the data of this experiment is assigned QC Level 2, we work on 
 a suggestion how to deal with errors in the QC checks for single atomic 
 datasets. We will our suggestion to the list for discussion as soon as 
 we are ready.

 We started to enhance our documentation, too: 
 https://redmine.dkrz.de/collaboration/projects/cmip5-qc/wiki . It is 
 still work in progress. If there are any demands on more documentation 
 and improvements or questions to the given descriptions, please let us 
 know.

 Maybe we will even have the opportunity to discuss about the QC checker 
 and criteria for the QC Level 2 assignment here in San Francisco.

 Best wishes,
 Martina


 On Mon, 13 Dec 2010 12:02:48 -0800, Karl Taylor <taylor13 at llnl.gov> 
 wrote:
> Dear all,
>
>  Is the test that determines whether the QC criteria has been met or
> not described anywhere?  The plots show max, min, and mean, but how
> is it determined if these are "in bounds".
>
>  thanks,
>  Karl
>
>  On 12/3/10 9:31 AM, kevin.marsh at stfc.ac.uk [1] wrote:
>
> Hi All,
>
> I’ve run a test of the QC code over some actual MOHC HADGEM2-ES
> data we have at the BADC in the DRS directory structure, as well as 
> on
> an example “areacella” file produced by the MOHC. More details and
> some plots of the results are available at: 
> http://proj.badc.rl.ac.uk/go-essp/wiki/CMIP5/QualityControl [2]
>
> Thanks,
>
> Kevin