Re: area_conserve_remap_Wrap

I tried the "ESMF_regrid" program and I get the same problem as "area_conse=
rve_remap_Wrap". It runs fine with NLAT = 96 and crashes with NLAT = 9=
5 with the following errors:
20140718 181231.614 INFO PET0 Running with ESMF Version 5.2.0rp2
20140718 181232.290 ERROR PET0 ESMCI_Regrid_F.C:146 c_esmc_regrid_create=
() Invalid argument - Dst contains a concave cell
20140718 181232.290 ERROR PET0 ESMCI_Regrid_F.C:282 c_esmc_regrid_create=
() Invalid argument Internal subroutine call returned Error
20140718 181232.290 ERROR PET0 ESMF_Regrid.F90:383 ESMF_RegridStore Inva=
lid argument - Internal subroutine call returned Error
20140718 181232.290 ERROR PET0 ESMF_FieldRegrid.F90:771 ESMF_FieldRegrid=
StoreNX Invalid argument - Internal subroutine call returned Error

Thanks,
MO
-------------------
Mohammad Omary
Institute for the Environment
University of North Carolina
http://www.ie.unc.edu/cempd
voice:951-789-6284
omary@unc.edu<mailto:omary@unc.edu>


From: Dennis Shea <shea@ucar.edu<mailto:shea@ucar.edu>>
Date: Friday, July 18, 2014 3:02 PM
To: Mohammad Omary <omary@email.unc.edu<mailto:omary@email.unc.edu>>
Subject: Re: area_conserve_remap_Wrap

Please send all responses to ncl-talk. THX
==========================
============
Really, you should use ESMF regridding to get high quality' conservative re=
mapping.
Your input grid is rectilinear and your target grid is rectilinear so it sh=
ould be easy.

Save the weight file and reuse it multiple times.


The '"area_conserve_remap_Wrap' was written for The Finite Volume climate m=
odel.






On Fri, Jul 18, 2014 at 3:34 PM, Omary, Mohammad <omary@email.unc.edu<mailt=
o:omary@email.unc.edu>> wrote:
Dennis,
I have met data for MOZART that has the lat as:

 lat = -90, -88.10526, -86.21053, -84.31579, -82.42105, -80.5263 ….

When I run the NCL program with NLAT = 96 I get

lat = -90, -88.125, -86.25, -84.375, -82.5, -80.625, -78.75, -76.875 …

To get lat the same as the met (the first lat above) I need to run the NCL =
with NLAT = 95

But the program "area_conserve_remap_Wrap" crashes.


Thanks,
MO
-------------------
Mohammad Omary
Institute for the Environment
University of North Carolina
http://www.ie.unc.edu/cempd
voice:951-789-6284<tel:951-789-6284>
omary@unc.edu<mailto:omary@unc.edu>


From: Dennis Shea <shea@ucar.edu<mailto:shea@ucar.edu>>
Date: Friday, July 18, 2014 2:27 PM
To: Mohammad Omary <omary@email.unc.edu<mailto:omary@email.unc.edu>>
Cc: "ncl-talk@ucar.edu<mailto:ncl-talk@ucar.edu> USERS" <ncl-talk@ucar.edu<=
mailto:ncl-talk@ucar.edu>>
Subject: Re: area_conserve_remap_Wrap

[1]
No idea what you mean by "...but the lat interval do not match the lat inte=
rvals in the met file." To which 'met file' are you referring?

[2]
Since area_conserve_remap_Wrap is located in contributed.ncl, you can see t=
he code being used:

%> less $NCARG_ROOT/lib/ncarg/nclscripts/csm/contributed.ncl<http://www.=
ncl.ucar.edu/Document/Functions/Contributed/contrib.shtml>

Search for: area_conserve_remap_Wrap

The most pertinent part is ....

undef ("area_conserve_remap_Wrap")
function area_conserve_remap_Wrap (xi[*]:numeric,yi[*]:numeric, fi:numeric =
\
                                  ,xo[*]:numeric,yo[*]:numeric, opt)
---
[snip]
  fo&$fo!(nDim-2)$ = yo ; create coordinate var
  fo&$fo!(nDim-1)$ = xo ; two rightmost dimension=
s
^^^^^^^^^^^^^^^^^^^^^^^^
==================
These latter two lines are directly assigning what *you* sent in as latitud=
e ('yo') and longitude ('xo').

[3] Whenever a question a sent to NCL, please include the output from 'prin=
tVarSummary(...)'


On Fri, Jul 18, 2014 at 2:50 PM, Omary, Mohammad <omary@email.unc.edu<mailt=
o:omary@email.unc.edu>> wrote:
All,

I am remapping 1x1 degree (180x360 nlat/nlon ) global aircraft emissions. =
 When I remap them to 96x144 (nlat/nlon) the program "area_conserve_remap_W=
rap" runs fine and I get nc output but the lat interval do not match the la=
t intervals in the met file. To get the correct lat intervals I need to ru=
n "area_conserve_remap_Wrap" with 95x144 (nlat/nlon). In this case the "la=
tGlobeFo" gives me the correct lat, but the program "area_conserve_remap_Wr=
ap" crashes" with:

 Error in "cremapbin":
 Could not map global lat array into grid array
Abort


Any help out there ?

Thanks,
MO
-------------------
Mohammad Omary
Institute for the Environment
University of North Carolina
http://www.ie.unc.edu/cempd
voice:951-789-6284<tel:951-789-6284>
omary@unc.edu<mailto:omary@unc.edu>


_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk

_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk


Received on Fri Jul 18 10:31:38 2014