The same problem in 2.4 with idealized wrf-fire data. (In 2.4 it fails on wrf2vdf/2nd step in VDCWizard; in 2.5 fails on wrfvdfcreate/1st step in VDCWizard). Unable to open directly through gui in either version.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
We took a look at one of the files and found that the XLAT and XLON variables seemed to be corrupt. Vapor needs those variables to be in the range of +360 to -360, and be monotonically increasing.
We found the XLON variable had the following values:
Vapor should still not be hanging indefinitely if this is the case though. We should be reporting an error message.
Is there another file that you would like us to look at? If another file does have XLAT and XLON values that conform to the above requirement, then this bug might need to be updated.
Thanks for reporting this.
-Scott
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Scott, thank you for looking into this.
WRF idealized simlations generally have arbitrary values for XLAT XLON, because they don't correspond to coordinates. It's simply a grid of user-defined width and length (represented by XLON and XLAT), with a step corresponding to user defined grid size.
This didn't seem to cause issues in earlier versions.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
It looks like we must have introduced a regression in the WRF data translator when dealing with idealized data. We will get this fixed, but most likely until our next release later in the year. Hopefully you have a workaround by using an older version of VAPOR? Apologies for the trouble, and appreciate your patience.
Scott, thank you for looking into this.
WRF idealized simlations generally have arbitrary values for XLAT XLON, because they don't correspond to coordinates. It's simply a grid of user-defined width and length (represented by XLON and XLAT), with a step corresponding to user defined grid size.
This didn't seem to cause issues in earlier versions.
[bugs:#1259] wrfvdfcreate hangs when converting WRF-FIRE data
Status: open
Group: 2.5.0
Created: Thu Apr 28, 2016 03:41 PM UTC by Scott
Last Updated: Wed May 04, 2016 02:03 PM UTC
Owner: nobody
wrfvdfcreate hangs, possibly going into an infitite loop, when converting WRF-FIRE data.
To reproduce:
1) wrfvdfcreate /glade/p/DASG/VAPOR/Data/Bugs/1259/wrfout_d01_0001-01-01_00-02-00 atoossa.vdf
2) Terminal will hang
--- old+++ new@@ -1,7 +1,7 @@
wrfvdfcreate hangs, possibly going into an infitite loop, when converting WRF-FIRE data.
To reproduce:
-1) wrfvdfcreate /glade/p/DASG/VAPOR/Data/Bugs/1259/wrfout_d01_0001-01-01_00-02-00 atoossa.vdf+1) wrfvdfcreate /glade/p/DASG/VAPOR/Bugs/1259/wrfout_d01_0001-01-01_00-02-00 atoossa.vdf
2) Terminal will hang
Reported by Atoossa Bakhshaii
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
This relates to the above discussion (all idealized WRF simulations have XLON/XLAT in meters corresponding to the defined grid, rather than geographical coordinates - so this is not invalid).
Will there be a fix for idealized WRF runs?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
How does one differentiate the ideal case from the cylindrical equidistant case. Both use map projection = 0. The test file provided lists the units for XLONG as degrees, not meters:
floatXLONG(Time,south_north,west_east);XLONG:FieldType=104;XLONG:MemoryOrder="XY ";XLONG:description="LONGITUDE, WEST IS NEGATIVE";XLONG:units="degree_east";XLONG:stagger="”;
This relates to the above discussion (all idealized WRF simulations have XLON/XLAT in meters corresponding to the defined grid, rather than geographical coordinates - so this is not invalid).
Will there be a fix for idealized WRF runs?
[bugs:#1259] wrfvdfcreate hangs when converting WRF-FIRE data
Status: closed-fixed
Group: 2.5.0
Created: Thu Apr 28, 2016 03:41 PM UTC by Scott
Last Updated: Thu Nov 03, 2016 10:57 PM UTC
Owner: clynejp
wrfvdfcreate hangs, possibly going into an infitite loop, when converting WRF-FIRE data.
To reproduce:
1) wrfvdfcreate /glade/p/DASG/VAPOR/Bugs/1259/wrfout_d01_0001-01-01_00-02-00 atoossa.vdf
2) Terminal will hang
Yes, the variable description (or map projection, for that matter) is not accurate for idealized runs. I am sure you'l see a 60m dx in those test files, though.
There is typically a wrf netcdf attribute SIMULATION_INITIALIZATION_TYPE, which should show 'IDEALIZED DATA'. I guess another test would be to see if LON/LAT step is exactly equal to DX/DY.
Last edit: Nadya Moisseeva 2016-11-04
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Yes, the variable description (or map projection, for that matter) is not accurate in for idealized runs. I am sure you'l see a 60m dx in those test files, though.
There is typically a wrf netcdf attribute SIMULATION_INITIALIZATION_TYPE, which should show 'IDEALIZED DATA'. I guess another test would be to see if LON/LAT step is exactly equal to DX/DY.
[bugs:#1259] wrfvdfcreate hangs when converting WRF-FIRE data
Status: closed-fixed
Group: 2.5.0
Created: Thu Apr 28, 2016 03:41 PM UTC by Scott
Last Updated: Thu Nov 03, 2016 10:57 PM UTC
Owner: clynejp
wrfvdfcreate hangs, possibly going into an infitite loop, when converting WRF-FIRE data.
To reproduce:
1) wrfvdfcreate /glade/p/DASG/VAPOR/Bugs/1259/wrfout_d01_0001-01-01_00-02-00 atoossa.vdf
2) Terminal will hang
That would be fantastic! WRF ideal cases are largely user-designed so they are somewhat all over the place with output, unfortunately. Thanks again for looking into this!
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
This occurs on windows, linux, and mac. Directly importing the file also causes vaporgui to hang.
Atoossa reports that Matlab and nctoolbox cannot open these files either.
The same problem in 2.4 with idealized wrf-fire data. (In 2.4 it fails on wrf2vdf/2nd step in VDCWizard; in 2.5 fails on wrfvdfcreate/1st step in VDCWizard). Unable to open directly through gui in either version.
We took a look at one of the files and found that the XLAT and XLON variables seemed to be corrupt. Vapor needs those variables to be in the range of +360 to -360, and be monotonically increasing.
We found the XLON variable had the following values:
30, 90, 150, 210, 270, 330, 390, 450, 510, 570, 630, 690, 750, 810, 870,
930, 990, 1050, 1110, 1170, 1230, 1290, 1350, 1410, 1470, 1530, 1590,
1650, 1710, 1770, 1830, 1890, 1950, 2010, 2070, 2130, 2190, 2250, 2310,
2370, 2430 ;
Vapor should still not be hanging indefinitely if this is the case though. We should be reporting an error message.
Is there another file that you would like us to look at? If another file does have XLAT and XLON values that conform to the above requirement, then this bug might need to be updated.
Thanks for reporting this.
-Scott
Scott, thank you for looking into this.
WRF idealized simlations generally have arbitrary values for XLAT XLON, because they don't correspond to coordinates. It's simply a grid of user-defined width and length (represented by XLON and XLAT), with a step corresponding to user defined grid size.
This didn't seem to cause issues in earlier versions.
Hi,
It looks like we must have introduced a regression in the WRF data translator when dealing with idealized data. We will get this fixed, but most likely until our next release later in the year. Hopefully you have a workaround by using an older version of VAPOR? Apologies for the trouble, and appreciate your patience.
cheers - jc
On May 4, 2016, at 2:29 PM, Nadya Moisseeva nadya512@users.sf.net wrote:
John Clyne
National Center for Atmospheric Research
303.497.1236 (w), 303.809.1922 (c)
clyne@ucar.edu
Related
Bugs:
#1259Workaround for current version is to create the files manually using ncdfvdfcreate:
ncdfvdfcreate -timedims Time -stagdims bottom_top_stag:south_north_stag:west_east_stag -vars U:V:W:QVAPOR:GRNHFX wrfout file.vdf
Thanks for the update. That is probably a better workaround.
On May 5, 2016, at 4:57 PM, Nadya Moisseeva nadya512@users.sf.net wrote:
John Clyne
National Center for Atmospheric Research
303.497.1236 (w), 303.809.1922 (c)
clyne@ucar.edu
Related
Bugs:
#1259Diff:
Fails with VAPOR3 as well
The data set contains invalid longitude coordinates. In particular, the first view values of the NetCDF XLONG coordinate variable are:
XLONG =
30, 90, 150, 210, 270, 330, 390, 450, 510, 570, 630, 690, 750, 810, 870,
930, 990, 1050, 1110, 1170, 1230, 1290, 1350, 1410, 1470, 1530, 1590,
1650, 1710, 1770, 1830, 1890, 1950, 2010, 2070, 2130, 2190, 2250, 2310,
2370, 2430,
This relates to the above discussion (all idealized WRF simulations have XLON/XLAT in meters corresponding to the defined grid, rather than geographical coordinates - so this is not invalid).
Will there be a fix for idealized WRF runs?
How does one differentiate the ideal case from the cylindrical equidistant case. Both use map projection = 0. The test file provided lists the units for XLONG as degrees, not meters:
thanks!
jc
John Clyne
National Center for Atmospheric Research
303.497.1236 (w), 303.809.1922 (c)
clyne@ucar.edu
Related
Bugs:
#1259Yes, the variable description (or map projection, for that matter) is not accurate for idealized runs. I am sure you'l see a 60m dx in those test files, though.
There is typically a wrf netcdf attribute SIMULATION_INITIALIZATION_TYPE, which should show 'IDEALIZED DATA'. I guess another test would be to see if LON/LAT step is exactly equal to DX/DY.
Last edit: Nadya Moisseeva 2016-11-04
Well, there is no SIMULATION_INITIALIZATION_TYPE attribute, but I think we can cobble something together that won’t result in incorrect results.
thanks for your help.
jc
John Clyne
National Center for Atmospheric Research
303.497.1236 (w), 303.809.1922 (c)
clyne@ucar.edu
Related
Bugs:
#1259That would be fantastic! WRF ideal cases are largely user-designed so they are somewhat all over the place with output, unfortunately. Thanks again for looking into this!
Logic has been added to the WRF reader to detect idealized runs by looking for unreasonable coordinate values expressed in degrees.