[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[IDV #OVR-601731]: IDV Problems Using CF Compliant WRF Files

> Yuan,
> 
> Without waiting for confirmation on my question of July 9 ["is the 
> z(bottom_top=51) actually equivalent to the ZNU variable"], I went ahead and 
> assumed that the z coordinate variable that you mentioned was indeed 
> equivalent to the ZNU variable since the z long_name says "eta values from 
> variable ZNU" and since z units are non-dimensional (""). So I set up my NCL 
> script to produce a .nc file that yields metadata for the z and z_stag 
> coordinate variables exactly as you specified in your July 9 response to me.
> 
> Below is a relevant snippet of the ncdump metadata so you can see how I 
> matched your specs from your July 9 response. It also shows the metadata for 
> the variable QVAPOR, which I am using for testing and whose metadata is 
> IDENTICAL to the QVAPOR metadata in the ORIGINAL  .nc file which CORRECTLY 
> vertically scales the QVAPOR variable in IDV. (It is no surprise that the 
> QVAPOR metadata is identical in both files, since NCL copies all variable 
> attributes when reading and then writing variables.)
> 
> dimensions:
> bottom_top = 51 ;
> bottom_top_stag = 52 ;
> variables:
> double z(bottom_top) ;
> z:units = "" ;
> z:long_name = "eta values from variable ZNU" ;
> z:positive = "down" ;
> z:_CoordinateAxisType = "GeoZ" ;
> z:_CoordinateAliasForDimension = "bottom_top" ;
> z:_CoordinateZisPositive = "down" ;
> double z_stag(bottom_top_stag) ;
> z_stag:units = "" ;
> z_stag:long_name = "eta values from variable ZNW" ;
> z_stag:positive = "down" ;
> z_stag:_CoordinateAxisType = "GeoZ" ;
> z_stag:_CoordinateAliasForDimension = "bottom_top_stag" ;
> z_stag:_CoordinateZisPositive = "down" ;
> float QVAPOR(Time, bottom_top, south_north, west_east) ;
> QVAPOR:FieldType = 104 ;
> QVAPOR:MemoryOrder = "XYZ" ;
> QVAPOR:description = "Water vapor mixing ratio" ;
> QVAPOR:units = "kg kg-1" ;
> QVAPOR:stagger = "" ;
> QVAPOR:coordinates = "XLONG XLAT" ;
> 
> Unfortunately, I get the exact same error message and an erroneously-scaled 
> plot (see attached). I am at a total loss why IDV has the vertical scaling 
> problem with my new .nc file that I produced by essentially COPYING data from 
> an ORIGINAL .nc file to a NEW .nc file using an NCL script, given that the 
> variables from the ORIGINAL .nc file scale CORRECTLY. I need to use the NCL 
> script to de-perturb, de-stagger, and produce a few new derived variables.


Could you upload your new nc file for me? I need more information.


Yuan
> 
> Can you help me understand why IDV is failing to recognize units from the 
> NCL-copied .nc file and how to get it to work?
> 
> In case it helps, I attached my NCL script which produced the .nc file that 
> is problematic for IDV. It printed the z and z_stag arrays so I could confirm 
> that the one-dimensional arrays indeed contain the ZNU and ZNW values that 
> range from 0.9965 to 0.4213 from bottom to top.
> 
> Thanks!
> 
> Jim
> 
> -----Original Message-----
> From: address@hidden <address@hidden>
> Sent: Monday, July 9, 2018 9:41 PM
> To: address@hidden
> Subject: RE: [IDV #OVR-601731]: IDV Problems Using CF Compliant WRF Files
> 
> Hi Yuan,
> 
> Given that the capital Z variable in my file is Geopotential Height as listed 
> in the IDV System Aliases list (last line of attached list) and calculated by 
> NCL code
> 
> Z = wrf_user_getvar(inpfile,"z",-1)
> 
> it indeed yields the 3D spatial variable for Z as you noted
> 
> float Z(Time=1, bottom_top=51, south_north=1001, west_east=1001)
> 
> I checked the values of Z and they do vary in all 3 spatial dimensions and 
> vary between about 27 and 6500 meters, which is reasonable for a 3D height 
> grid for the WRF domain.
> 
> So is the z(bottom_top=51) actually equivalent to the ZNU variable and the 
> z_stag(bottom_top_stag=52) equivalent to the ZWU variable?
> 
> Thanks!
> 
> Jim
> 
> -----Original Message-----
> From: Unidata IDV Support <address@hidden>
> Sent: Monday, July 9, 2018 5:36 PM
> To: address@hidden
> Cc: address@hidden; address@hidden
> Subject: [IDV #OVR-601731]: IDV Problems Using CF Compliant WRF Files
> 
> Hi Jim,
> 
> In your file, the Z is not defined correctly:
> 
> float Z(Time=1, bottom_top=51, south_north=1001, west_east=1001);
> :coordinates = "XLONG XLAT ZNU";
> :stagger = " ";
> :units = "m";
> :description = "Height";
> :MemoryOrder = "XYZ";
> :FieldType = 104; // int
> 
> The correct Z should be like the below, and you need to have z_stag variable 
> too:
> 
> double z(bottom_top=51);
> :units = "";
> :long_name = "eta values from variable ZNU";
> :positive = "down";
> :_CoordinateAxisType = "GeoZ";
> :_CoordinateAliasForDimension = "bottom_top";
> :_CoordinateZisPositive = "down";
> 
> double z_stag(bottom_top_stag=52);
> :units = "";
> :long_name = "eta values from variable ZNW";
> :positive = "down";
> :_CoordinateAxisType = "GeoZ";
> :_CoordinateAliasForDimension = "bottom_top_stag";
> :_CoordinateZisPositive = "down";
> 
> 
> Yuan
> 
> Ticket Details
> ===================
> Ticket ID: OVR-601731
> Department: Support IDV
> Priority: High
> Status: Open
> ===================
> NOTE: All email exchanges with Unidata User Support are recorded in the 
> Unidata inquiry tracking system and then made publicly available through the 
> web.  If you do not want to have your interactions made available in this 
> way, you must let us know in each email you send to us.
> 
> 
> 
> 


Ticket Details
===================
Ticket ID: OVR-601731
Department: Support IDV
Priority: High
Status: Open
===================
NOTE: All email exchanges with Unidata User Support are recorded in the Unidata 
inquiry tracking system and then made publicly available through the web.  If 
you do not want to have your interactions made available in this way, you must 
let us know in each email you send to us.