[gradsusr] set shp -poly

Wed Jan 4 19:52:47 EST 2012

The data projection is Lambert Conformal with central meridian of 265 and
reference latitude of 25.

It's a GRIB 2 data set, but there are some characters before 'GRIB' key
word. In my opinion, a standard GRIB data file should be start with 'GRIB'.
My question is if it is a standard GRIB data file?

On Thu, Jan 5, 2012 at 7:20 AM, Jennifer Adams <jma at cola.iges.org> wrote:

> Jeff,
>
> On Jan 4, 2012, at 12:19 PM, Jeff Lake wrote:
>
> This is exactly the problem ...
> the dbf is also in a format I nor GDAL has ever seen before,
>
> In GrADS, you can issue the 'q dbf' command and it will spew the contents
> of the dbf file. The utility 'dbfdump', which comes with the shapefile
> library, also has no trouble reading the dbf file. There is very little
> structure to the dbf file, it's just a list of attributes and the values
> that go with them for each element in the file.
>
> could be this temp, but yet it could be this???
>
> I'm sorry, but I don't understand this statement.
>
>
> this is the GRIB file I'm using
>
> http://weather.noaa.gov/pub/SL.us008001/ST.opnl/DF.gr2/DC.ndgd/GT.rtma/AR.conus/RT.01/ds.temp.bin
>
> the RT.01 determines the cycle hour
>
> GrADS version 2.0.1,
> CentOS 5.7 64bit
>
>
> and the script snippet
>
> 'clear shp'
> 'set lat 20 55'
> 'set lon -150 -45'
> 'set gxout shp'
> 'set mproj latlon'
> 'set ccols 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22'
> 'set clevs -40 -35 -30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30 35 40 45 50
> 55 60 65 70 75 80 85 90 95 100 105 110 115 120 125 130 135'
> 'set shp -poly rtma_temp'
> 'd (TMP2m-273.16)*9/5+32'
> 'quit'
>
>
> I grabbed the file, used g2ctl, and drew a grfill plot with GrADS,
> overlaying a shapefile with the state boundaries (instead of the default
> map plotted by GrADS). Here is the result:
> Next I used 'gxout shp' and created a shapfile based on this data. Then I
> import the shapefile I created into Qgis and draw it using no outline and
> coloring in the polygons according to the classification field MIN_VALUE
> and then overlay the same shapefile (statesp020) and get this:
>
>
> There is no re-projecting going on, the plots are the same, other than the
> color scheme -- it's a pain to define colors in Qgis, so I used this
> blue/green palette. As far as I can tell, there is nothing wrong with the
> shapefile created by GrADS.
>
>
>
> I would let GrADS do my mapping, but I'm some what of a stickler about
> my maps looking the same
> I cannot for the life of me get GrADS to match my projection
> http://contours.michiganwxsystem.com/forecasts/gfs/gfsmos_d1_maxt.png
> I got GEMPAK to match (and they say that is harder), but not GrADS :(
>
>
> As for the projection in your michiganwxsystem plot, it doesn't appear to
> be drawn with GrADS, but it looks like a polar stereographic projection. In
> GrADS, use 'set mproj nps' and then display your temperature plot and it
> will look like this:
>
>
>
> This looks like the projection in the image on your website. As for
> getting Qgis to draw in a polar stereographic projection … I have no idea
> how to do this. Perhaps this is another user in the forum who will know.
> --Jennifer
>
>
> -Jeff Lake
> MichiganWxSystem.com
> WeatherMichigan.net
> TheWeatherCenter.net
> GRLevelXStuff.com
>
>
> On 1/3/2012 7:43 PM, Yaqiang Wang wrote:
>
> It seems GrADS does not form real 'polygon' from contour lines by
>
> topology analysis. This is the source of the problem.
>
>
> On Wed, Jan 4, 2012 at 7:17 AM, Jennifer Adams<jma at cola.iges.org>  wrote:
>
> Jeff,
>
> The shapfile interface in GrADS creates a shapefile that contains polygons
>
> that fill out the areas outlined by the shaded contours. Because of the
>
> rules in the shapefile spec regarding polygon geometry, the area inside a
>
> shaded contour is split up into many pieces so that individual polygons do
>
> not overlap and do not contain holes (no donuts). That is why you are
> seeing
>
> all the vertical lines. If you color in the polygons and use the same color
>
> to draw the outline, you won't see the vertical lines and the plot will
> look
>
> more normal (like the shaded contour plot in GrADS).
>
>
> Also, before you create the shapefile, you must be in the latlon projection
>
> (set mproj latlon). No other projection will work. It should be possible to
>
> use your GIS tool to re-project the data in the shapefile using the
> metadata
>
> in the file provided by GrADS. From the attachement you sent, it's
> difficult
>
> to tell what the problem is -- better to draw a plot of a variable such a
>
> land mask that would clearly show whether the boundaries in the data are
>
> lining up with the political boundaries (and color them in and use matching
>
> outline color). If there is an issue in the way Qgis is re-projecting the
>
> shapefile, then that may or may not be a GrADS problem.
>
>
> The polygons 'connect to the outside' because they are filling in the
>
> lat/lon block defined by your dimension environment. You can't mask your
>
> shapefile according to political boundaries, but you can choose to not draw
>
> a certain contour level by using -1 instead of a color number in the 'set
>
> ccols' command.
>
>
> I can't help you any further without more specific information about what
>
> version of GrADS you're using, what the descriptor file looks like for your
>
> data, the source of your data file (so I can grab a sample), what your
>
> commands were to generate the shapefile, and what you did inside Qgis to
>
> create the display in your attachment.
>
> --Jennifer
>
>
>
>
>
> On Jan 3, 2012, at 4:22 PM, Jeff Lake wrote:
>
>
> I believe there is a minor, okay major, flaw in the conversion to shape
> file
>
> here is the shape file I get when converting the RTMA GRIB over (see
>
> attachment)
>
> the resulting polygons all 'connect' to the outside
>
> this is a cint set from -30 to 140 by 5 degree's
>
> this is a better result then another piece of software, but still not
>
> correct..
>
>
> anybody got suggestions ??
>
>
> --
>
> -Jeff Lake
>
> MichiganWxSystem.com
>
> WeatherMichigan.net
>
> TheWeatherCenter.net
>
> GRLevelXStuff.com
>
>
> <rtma.jpg>_______________________________________________
>
> gradsusr mailing list
>
> gradsusr at gradsusr.org
>
> http://gradsusr.org/mailman/listinfo/gradsusr
>
>
>
> --
>
> Jennifer M. Adams
>
> IGES/COLA
>
> 4041 Powder Mill Road, Suite 302
>
> Calverton, MD 20705
>
> jma at cola.iges.org
>
>
>
>
>
> _______________________________________________
>
> gradsusr mailing list
>
> gradsusr at gradsusr.org
>
> http://gradsusr.org/mailman/listinfo/gradsusr
>
>
>
>
> _______________________________________________
> gradsusr mailing list
> gradsusr at gradsusr.org
> http://gradsusr.org/mailman/listinfo/gradsusr
>
>
> --
> Jennifer M. Adams
> IGES/COLA
> 4041 Powder Mill Road, Suite 302
> Calverton, MD 20705
> jma at cola.iges.org
>
>
>
>
> _______________________________________________
> gradsusr mailing list
> gradsusr at gradsusr.org
> http://gradsusr.org/mailman/listinfo/gradsusr
>
>

-- 
*************************************************
Dr. Yaqiang Wang
Chinese Academy of Meteorological Sciences (CAMS)
46, Zhong-Guan-Cun South Avenue
Beijing, 100081
China

yaqiang.wang at gmail.com

http://www.meteothinker.com
**************************************************
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://gradsusr.org/pipermail/gradsusr/attachments/20120105/a41dce10/attachment-0003.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/png
Size: 49473 bytes
Desc: not available
Url : http://gradsusr.org/pipermail/gradsusr/attachments/20120105/a41dce10/attachment-0012.png 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/png
Size: 172637 bytes
Desc: not available
Url : http://gradsusr.org/pipermail/gradsusr/attachments/20120105/a41dce10/attachment-0013.png 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/png
Size: 101587 bytes
Desc: not available
Url : http://gradsusr.org/pipermail/gradsusr/attachments/20120105/a41dce10/attachment-0014.png 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Image00075.png
Type: image/png
Size: 233355 bytes
Desc: not available
Url : http://gradsusr.org/pipermail/gradsusr/attachments/20120105/a41dce10/attachment-0015.png 