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

[IDV #CLG-761846]: Re: [idvusers] Problem using NARR-A as aggregated GRIB files in IDV

Hi Jay-

> Full Name: Jason Shafer
> Email Address: address@hidden
> Organization: Lyndon State College
> Package Version: 2.4 build date:2007-12-20 16:34 UTC
> Operating System: Linux
> Hardware: Java: home: /software/idv/IDV_2.4/jre version: 1.5.0_11 j3d:1.3.1
> Description of problem: Hi,
> 
> I'm having problems aggregating the NARR Reanalysis data. IDV does not seem 
> to be
> capable of loading multiple times and aggregating the files. I noticed your 
> solution in a
> previous post below, but couldn't get the solution to work. I think this 
> might have been
> due to this link being dead: 
> http://www.unidata.ucar.edu/namespaces/netcdf/ncml-2.2%22

No, that's just an XML namespace pointer.  It can be anything (e.g. 
http://foo.bar)
and doesn't have to be a valid link.

For the NcML solution I provided Ryan, you need to be running version 2.5 
instead of 2.4.  Try downloading the latest version and use the NcML
solution.

> I was wondering if a solution exists currently, and if not when a solution 
> might exist.

I'm in the process of testing the next version which should handle
this out of the box, but that probably won't be publicly available 
until next month (July).

> Thanks for your time.

Let me know if the NcML doesn't work with version 2.5.

Don Murray

> 
> * To: rboller@xxxxxxxxxxxx
> * Subject: [IDV #CLG-761846]: Re: [idvusers] Problem using NARR-A as 
> aggregated GRIB files in IDV
> * From: "Unidata IDV Support" <support-idv@xxxxxxxxxxxxxxxx>
> * Date: Wed, 05 Mar 2008 09:26:41 -0700
> * Delivered-to: address@hidden by laraine.unidata.ucar.edu (Postfix) with 
> ESMTP id 89BB7CB189; Wed, 5 Mar 2008 09:26:41 -0700 (MST) id 618DFD5116; Wed, 
> 5 Mar 2008 09:26:41 -0700 (MST)
> 
> Hi Ryan-
> 
> Sorry for the delay in responding.  I've been out for the past few
> weeks.
> 
> > I seem to have come across the same problem as I've found on the message
> > board (see below).  While I could use the "Make a time sequence" formula,
> > this quickly becomes old when 20+ files are involved.  I'm using NCEP model
> > data and have similar time tag issues... i.e.,
> >
> > The "fnl_060913_00_00" file:
> >
> > :_CoordinateModelRunDate = "2006-09-13T00:00:00Z"
> > int time(time=1);
> > :long_name = "forecast time";
> > :units = "hour since 2006-09-13T00:00:00Z";
> > :GRIB_orgReferenceTime = "2006-09-13T00:00:00Z";
> > :GRIB2_significanceOfRTName = "Start of forecast";
> > :_CoordinateAxisType = "Time";
> >
> >
> > The "fnl_060913_06_00" file:
> > :_CoordinateModelRunDate = "2006-09-13T06:00:00Z";
> > int time(time=1);
> > :long_name = "forecast time";
> > :units = "hour since 2006-09-13T06:00:00Z";
> > :GRIB_orgReferenceTime = "2006-09-13T06:00:00Z";
> > :GRIB2_significanceOfRTName = "Start of forecast";
> > :_CoordinateAxisType = "Time";
> >
> >
> > I'm thinking that the easiest workaround would be to write a script to make
> > all of the model run times to be the same while varying the
> > orgReferenceTime.  Is this appropriate?   I'm open to tweaking the IDV code
> > if that makes more sense.
> 
> The issue is down in the netCDF Java code and a fix is being worked on
> for the next version.  But that release is several months away.  However,
> in the mean time, you might be able to use NcML to solve your problem.
> Here's an example that works with the IDV (open it using I'm Feeling Lucky
> or with the Grib files Data Source Type):
> 
> <netcdf xmlns="http://www.unidata.ucar.edu/namespaces/netcdf/ncml-2.2";;>
> <variable name="time" shape="time" type="int">
> <attribute name="units" value="hour since 2006-09-13T00:00:00Z"/>
> <attribute name="_CoordinateAxisType" value="Time" />
> <values start="0" increment="6" />
> </variable>
> <aggregation type="joinExisting" dimName="time">
> <netcdf location="fnl_060913_00_00"/>
> <netcdf location="fnl_060913_06_00"/>
> <netcdf location="fnl_060913_12_00"/>
> </aggregation>
> </netcdf>
> 
> basically, you define the time variable and it's values, then use
> the aggregation tag to list the files you want to aggregate.
> 
> Save this in a text file with the extension .ncml (eg. ryan.ncml)
> and place it in the same directory as the files.   Alternatively,
> you can specify the full path name to the file like:
> 
> <netcdf location="file:/C:/data/grids/fnl_060913_00_00">
> 
> > Thanks much for any suggestions.
> 
> Let me know if that doesn't work for you.  I tested this with
> version 2.5b1 (nightly build).
> 
> Don
> 
> > -----
> >
> > Hi Tim-
> >
> > Tim Axelrod wrote:
> > > I'd like to be able to use a set of NARR-A grb files as a data source to
> > > IDV and be able to display them as an animation and/or use the data
> > > probe/time series tool. I'm able to do this with NAM grb files by
> > > selecting "aggregated GRIB files" as the data source type. This fails
> > > with the NARR-A grb files, however. Only a single time gets shown in the
> > > field selector/times window - and that time seems to be randomly selected
> > > from the set of files.
> >
> > This is not possible at present but a solution is being worked on
> > in the underlying netCDF-Java layer. But that won't be available
> > for a few months.
> >
> > The NAM works because they all have the same model run time and each
> > file has an offset forecast hour. Each NARR file has a different
> > base ("model run") time with a forecast hour of 0, so the aggregation
> > doesn't know what to use as a "model run" time.
> >
> > > I'm using IDV 2.3, build data 2007-08-15. Any advice?
> >
> > If you are daring, you could run the nightly build which has
> > formula in it for making a time sequence from a set of single time
> > images or grids. What you would do is select all the NARR files you
> > want to "aggregate" and use the I'm Feeling lucky option in the
> > Files Chooser. Then use the "Miscellaneous->Make a time sequence from
> > single time grids/images" formula. That will pop up a dialog that
> > allows you to select the individual grids. This will be time consuming
> > if you have a lot of files. It's a a new feature (multiple parameter
> > selection), so feedback is welcome.
> >
> > The nightly build is available at:
> >
> > http://www.unidata.ucar.edu/software/idv/release/nightly
> >
> > Don Murray
> > *************************************************************
> > Don Murray UCAR Unidata Program
> > dmurray@hidden_host P.O. Box 3000
> > (303) 497-8628 Boulder, CO 80307
> > http://www.unidata.ucar.edu/staff/donm
> > *************************************************************
> >
> >
> 
> 
> Ticket Details
> ===================
> Ticket ID: CLG-761846
> Department: Support IDV
> Priority: Normal
> Status: Open
> 
> 
> 
> 


Ticket Details
===================
Ticket ID: CLG-761846
Department: Support IDV
Priority: Normal
Status: Open