Re: grib2 fie read

I am not sure what you mean by "correctly". As I stated below, there is indeed a bug in NCL 6.1.0-beta (and previous versions), that has been fixed. The issue with the parameter number is not fixable in general because neither NCEP nor ECMWF have an entry for parameter 3 in this particular table. All this means is that the variable has unknown units and name. It still will display correctly. If you know what the variable is supposed to represent you could add an entry in your own copy of 4.2.2.4.table. The format should be fairly self explanatory if you look at the entries for some other parameters.
I am attaching a copy of the plotting output from this data file, using a slightly modified version of the example script lcnative_2.ncl. I'll attach the script as well, though you will not be able to get the same results without NCL 6.1.0. If you need a binary with the fix immediately we can get you one, but the new release is due out very soon now.
 -dave

On Sep 24, 2012, at 10:06 PM, Luo, Chao wrote:

> Hi Dave,
>
> Thanks for your email and help! I was wondering if there is a way to read theses GRIB2 files correctly. I think these GRIB2 files may not be in error.
>
> Thanks much,
>
> CL
>
> From: "David Brown" <dbrown@ucar.edu>
> To: "Chao Luo" <chao.luo@eas.gatech.edu>
> Cc: "ncl-talk" <ncl-talk@ucar.edu>
> Sent: Monday, September 24, 2012 5:54:35 PM
> Subject: Re: grib2 fie read
>
> HI LC,
> The warning message has nothing to do with the problem you encountered with out-of-range coordinate values. The message simply means that NCL cannot find a table entry for the variable. In this case that is because there is no published entry for parameter 3 in table 4.2.2.4. The GRIB file is either in error or is using an unpublished parameter value. I do think there may have been an earlier version of this table that had indexes 1,2, and 3 instead of 0, 1, and 2.
>
> Regarding the out-of-range coordinate values, that turns out to be a bug in NCL's generation of lambert conformal coordinates. This file has the Y coordinates going from north to south, which is rather unusual (for lambert conformal grids at least). Anyway, the problem has now been fixed. The new code will be available in NCL 6.1.0.
> -dave
>
>
> On Sep 21, 2012, at 2:42 PM, Luo, Chao wrote:
>
> Hi,
>
> I am reading biomass burning area of NOAA satellite grib2 data by using ncl-6.1.0. I print out the latitude of the grib2, and found maximum of latitude is 1.e+12, which is wrong. The error message when I reading the file is:
>
> warning: Entry (3) not found in code table file /data11/cluo/GAfire/ncl-
> 6.1.0/lib/ncarg/grib2_codetables/ncep/4/4.2.2.4.table
>
> I was wondering if this the problem caused by this warning. The NCL script I used is very simple:
>
> begin
> f = addfile("../NOAA_HMS/march_april/grib2/burned_area_20120316_18_23_12km.grib2","r")
> lat = f->gridlat_0(:,:)
> lon = f->gridlon_0(:,:)
> area = f->VAR_2_4_3_P31_GLC0(:,:)
> printMinMax (lat, True)
> printVarSummary (lat)
> end
> ~
> and output is :
>
> Variable: lat
> Type: float
> Total Size: 2987436 bytes
> 746859 values
> Number of Dimensions: 2
> Dimensions and sizes: [ygrid_0 | 723] x [xgrid_0 | 1033]
> Coordinates:
> Number Of Attributes: 11
> corners : ( 45.7454, 40.18576, 1e+12, 1e+12 )
> long_name : latitude
> grid_type : Lambert Conformal (secant, tangent, conical or bipolar)
> units : degrees_north
> Latin2 : 60
> Latin1 : 20
> Dy : 12
> Dx : 12
> Lov : 264
> Lo1 : 161.7229
> La1 : 45.7454
> (0)
> (0) latitude: min=36.144 max=1e+12
>
>
> Thanks very much,
>
> CL
> _______________________________________________
> 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

_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Tue Sep 25 12:11:14 2012