[Dart-dev] asking for input about DART obs_seq_netcdf files

Hui Liu hliu at ucar.edu
Sat Nov 6 10:10:14 MDT 2010


Tim:

   I like the ascii version of obs_seq files a lot. I can make som 
echanges to them easily. So, it would be great the "ascii" version
could still be option.
hui

On Fri, 
5 Nov 2010, Tim Hoar wrote:

> I've gotten a request asking to change the storage order of the 'copies' in
> the netcdf format version of the DART obs_sequence files. The copy order is
> described with strings in the CopyMetaData variable. My strategy has always
> been
> to never assume anything about the storage order - this won't affect me or
> the matlab
> scripts (see the get_copy_index(), get_qc_index() matlab functions).
> 
> For people who like to hardwire magic numbers into scripts (and who
> doesn't!)
> this makes a potentially profound change. The NCO operators are also geared
> to using magic numbers (as far as I can tell), so the time to speak up is
> NOW.
> 
> Presently, the obs error variance is always 'last'. If 
> filter_nml:num_output_obs_members > 0, these ensemble members get preserved
> right after the 'mandatory' copies:
> 
> NCEP BUFR observation  (for example)                                        
>                                                                  
> prior ensemble mean                                                        
>                                                      
> posterior ensemble mean                                                    
>                                                      
> prior ensemble spread                                                      
>                                                      
> posterior ensemble spread                                                  
>                                                      
> prior ensemble member      1                                                
>                                                     
> posterior ensemble member      1                                            
>                                                     
> prior ensemble member      2                                                
>                                                     
> posterior ensemble member      2                                            
>                                                     
> prior ensemble member      3                                                
>                                                     
> posterior ensemble member      3                                            
>                                                     
> ...
> observation error variance
> <then the QC values follow>
> 
> The proposed change is to make it such that the copy number of the
> observation_error_variance is invariant to
> the filter_nml:num_output_obs_members setting:
> 
> NCEP BUFR observation  (for example)                                        
>                                                             
> prior ensemble mean                                                        
>                                                      
> posterior ensemble mean                                                    
>                                                      
> prior ensemble spread                                                      
>                                                      
> posterior ensemble spread    
> observation error variance                                                  
>                                                  
> prior ensemble member      1                                                
>                                                    
> posterior ensemble member      1                                            
>                                                    
> prior ensemble member      2                                                
>                                                    
> posterior ensemble member      2                                            
>                                                    
> prior ensemble member      3                                                
>                                                    
> posterior ensemble member      3                                            
>                                                    
> ...
> <then the QC values follow>
> 
> If you're still reading, keep the following in mind: The observation error
> variance is probably ALWAYS present, even if you convert obs_seq.out files
> to
> netCDF. In that case (I believe, but am unwilling to check at Friday at
> 4:30)
> that the storage order is:
> 
> NCEP BUFR observation  (for example)                                        
>                                                            
> observation error variance                                                  
>                                          
> <then the QC values follow>
> 
> So in some ways, I'd like to move the observation error variance to copy 2
> for complete consistency. If the prior/posterior means/spread exist, they
> get appended after that. This would cause the world as we know it to
> cease to exist. I know of lots of "other people's scripts" that REQUIRE the
> prior_ensemble mean copy 2.
> 
> As you may have surmised, I have no opinion on the matter ... I'm happy
> to follow the group decision. This is a democracy. I'd like to be a
> benevolent
> dictator, but those pesky minions keep proving me wrong.
> 
> Tim Hoar, Associate Scientist
> National Center for Atmospheric Research
> thoar at ucar.edu
> 303 497 1708
> 
> 
>


More information about the Dart-dev mailing list