Because of the complexity of GRIB data and the fact that it is not fully supported by Digital Atmosphere yet, Weather Graphics Technologies will not provide assistance in locating, retrieving, or using GRIB data. It is provided as-is. Comments are welcomed but we may not be able to respond to questions. If you need help, please access our public discussion group on our website at www.weathergraphics.com
GRIB is a standard format for binary data interchange between weather agencies. Digital Atmosphere offers the ability to view this data. Although we strive to give you substantial amount of protection from the sheer complexity of the data, a working knowledge of GRIB is important to getting the most from this data (see references below).
Some data sources may store the data in a compressed format, such as *.ZIP or *.BZ2 (BZip2). You will need to decompress the data using a third-party program before it can be used in Digital Atmosphere!
Once Digital Atmosphere has examined the file, you will see a dialog similar to this one:
You must then choose which field you want to see. The specific field you chose will be extracted.
Now you should click on either Contour or Plot in the Toolbox window (see illustration). Now click on "Select field".
|The DIGATMOS.GRD grid definition file|
Advanced users can edit the digatmos.grd file to change the grid, model, and parameter definitions.
The MODEL section defines a model name based on its model number (Column 1-3) and IDENTIFIER of the facility (Column 5-8; see IDENTIFIER section).
The CENTRE section defines a production facility.
The FIELD section defines a product based on its parameter number (Column 1-3) and the IDENTIFIER of the facility (Column 5-8; see IDENTIFIER section). Based on what appears here, Digital Atmosphere selects the Preferences Style type (Column 25-27) to display this field.
|Fixing GRIB problems with DIGATMOS.GRD|
There are two symptoms which indicate a problem with GRIB data:
To resolve this, you need to identify the centre, model, and field causing the problem. Go to File > Preferences > Misc 2 and enable "Debug GRIB Data". When this is enabled it causes full information to be displayed in the field selection panel, rather than the simplified user-friendly format. For example, we recently had display problems with ECMWF data, and the field panel showed "72HR 500 MB ? ECMWF ? CREATED 29DEC06 1200Z". When the debug mode was enabled, we saw this:
TIME:1=72HR LEVEL:100=500 MB FIELD:129= ? CENTRE:98= ECMWF MODEL:128=? CREATED 29DEC06 1200Z
Thus it can be seen that Digital Atmosphere was unable to decode FIELD and MODEL, and only the coded values are known (129 for field and 128 for model).
To determine the missing information, we have to do some Googling. For the model number, we made the query "ecmwf global grib 98 128" in Google. The page http://www.ecmwf.int/products/data/operational_system/evolution/evolution_2006.html tells us "The new GRIB model identifier for the atmospheric model will be 128. The model identifiers for the wave models will remain the same (116 for the global wave model and 216 for the limited area wave model)."
Excellent. Thus we enter the following additions to digatmos.grd in the
MODEL section. Note that we refer to centre 98 (ECMWF) indicating that
these model identifiers 116, 128, and 216 are applicable only to centre 98
116 98 ECMWF/GLOBAL-WAVE
128 98 ECMWF/GLOBAL-ATMOS
216 98 ECMWF/LIMITED-WAVE
Now we need to find what field 129 is. Googling for "ecmwf grib 129
height" (since we suspect 129 is a height field) we get a good link at http://www.cgd.ucar.edu/cms/eaton/cf-metadata/ECMWF.html .
This shows that 129 is indeed a geopotential [height] field. Since
everything in this ECMWF Grib Code list is useful, we should ideally enter
everything, but since ECMWF restricts most of its data, it's smarter to make
only a small number of changes as listed below, in the FIELD block.
Note that we specify that these fields only apply to products from centre 98
1 98 Pa - slp Pressure
2 98 Pa - slp Pressure MSL
7 98 cm - hgt Geopotential height
11 98 K - tms Temperature
41 98 s - vrt Vorticity (absolute)
43 98 s - vrt Vorticity (relative)
129 98 cm - hgt Geopotential height
138 98 s - vrt Vorticity (relative)
We entered "cm" under geopotential height so that isopleths will plot legibly. It is also acceptable to use dm and m.
That's it! Our model data will plot correctly now.
* For full information on how to obtain GRIB data, see Appendix > GRIB data.