matplotlib-users Mailing List for matplotlib

IMHO that's the most straightforward approach.
He can use masked array for empty blocks (if contour data doesn't
already contain the holes as masked array) and apply inpainting, then
draw the land.
For more details about inpainting: http://stackoverflow.com/a/17125125/992005
On Wed, Jan 29, 2014 at 6:22 PM, Eric Firing <ef...@ha...> wrote:
> On 2014年01月29日 5:41 AM, A Short wrote:
>> Is there any work around so it looks like the below image?
>
> It looks like with any reasonable contouring algorithm, this would
> require interpolating into land regions, contouring, and then plotting
> the land on top. The key is the interpolation, not the plotting. The
> example you show might have been interpolated to a finer grid
> everywhere, not just in the missing value regions.

On 2014年01月29日 5:41 AM, A Short wrote:
> Is there any work around so it looks like the below image?
It looks like with any reasonable contouring algorithm, this would 
require interpolating into land regions, contouring, and then plotting 
the land on top. The key is the interpolation, not the plotting. The 
example you show might have been interpolated to a finer grid 
everywhere, not just in the missing value regions.
I can't comment on the file itself.
Eric
>
> Could anyone confirm that this would be the correct grib file for The North
> Atlantic..?
> ftp://ftpprd.ncep.noaa.gov/pub/data/nccf/com/wave/prod/wave.20140129/nww3.t06z.grib.grib2
>
> Thanks for all the help
>
> <http://matplotlib.1069221.n5.nabble.com/file/n42798/figure_1.png>
>
>
>
>
>
> --
> View this message in context: http://matplotlib.1069221.n5.nabble.com/Plotting-NOAA-grib2-data-in-basemap-tp42698p42798.html
> Sent from the matplotlib - users mailing list archive at Nabble.com.
>
> ------------------------------------------------------------------------------
> WatchGuard Dimension instantly turns raw network data into actionable
> security intelligence. It gives you real-time visual feedback on key
> security issues and trends. Skip the complicated setup - simply import
> a virtual appliance and go from zero to informed in seconds.
> http://pubads.g.doubleclick.net/gampad/clk?id=123612991&iu=/4140/ostg.clktrk
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

On 2014年01月28日 11:40 PM, Ian Thomas wrote:
> On 29 January 2014 03:21, Eric Firing <ef...@ha...
> <mailto:ef...@ha...>> wrote:
>
> On 2014年01月28日 10:01 AM, A Short wrote:
> > Hi - Ive now improved my code and confirmed the use of the right
> grib file
> > but i cant for the life of me figure out the missing data near the
> > coastline..? Could anyone help?
>
> The present contouring algorithm works with rectangular blocks, and if
> any corner has missing data, nothing is filled for that block.
>
>
> This will improve shortly, cutting off the corners of some of those
> empty blocks. I am currently testing the new algorithm for this prior to
> submitting it for others' approval.
Ian,
I'm glad to hear that! One possibility would be to use a temporary 
rcParam (temporary in that it might be phased out after a couple 
releases) to allow switching between the two algorithms. This would 
make it much easier to test, and it would also allow a transition during 
which people could reproduce results obtained with earlier mpl. It 
would also be a safety measure, in case someone hits a corner case which 
the new algorithm doesn't handle but the old one does--not that I'm 
expecting such cases to arise.
Eric
>
> Ian

Is there any work around so it looks like the below image?
Could anyone confirm that this would be the correct grib file for The North
Atlantic..? 
ftp://ftpprd.ncep.noaa.gov/pub/data/nccf/com/wave/prod/wave.20140129/nww3.t06z.grib.grib2
Thanks for all the help
<http://matplotlib.1069221.n5.nabble.com/file/n42798/figure_1.png> 
--
View this message in context: http://matplotlib.1069221.n5.nabble.com/Plotting-NOAA-grib2-data-in-basemap-tp42698p42798.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

On 29 January 2014 03:21, Eric Firing <ef...@ha...> wrote:
> On 2014年01月28日 10:01 AM, A Short wrote:
> > Hi - Ive now improved my code and confirmed the use of the right grib
> file
> > but i cant for the life of me figure out the missing data near the
> > coastline..? Could anyone help?
>
> The present contouring algorithm works with rectangular blocks, and if
> any corner has missing data, nothing is filled for that block.
>
This will improve shortly, cutting off the corners of some of those empty
blocks. I am currently testing the new algorithm for this prior to
submitting it for others' approval.
Ian

On 2014年01月28日 10:01 AM, A Short wrote:
> Hi - Ive now improved my code and confirmed the use of the right grib file
> but i cant for the life of me figure out the missing data near the
> coastline..? Could anyone help?
The present contouring algorithm works with rectangular blocks, and if 
any corner has missing data, nothing is filled for that block.
Eric
>
> `import Nio
> from mpl_toolkits.basemap import Basemap
> import matplotlib.pyplot as plt
> import numpy as np
>
> f = Nio.open_file('nww3.t12z.grib(2).grib2')
> lons = f.variables['lon_0'][:]
> lats = f.variables['lat_0'][::-1] # flip latitudes so data goes S-->N
> times = f.variables['forecast_time0'][:]
> ntime = 5
> data = f.variables['HTSGW_P0_L1_GLL0'][ntime,::-1]
>
> fig = plt.figure(figsize=(16,16))
> m = Basemap(llcrnrlon=-35.,llcrnrlat=42.,urcrnrlon=5.,urcrnrlat=65.,
> projection='lcc',lat_1=10.,lat_2=15.,lon_0=10.,
> resolution ='h',area_thresh=1000.)
>
> x, y = m(*np.meshgrid(lons, lats))
> m.fillcontinents(color='#477519')
> m.drawcoastlines(linewidth=0.5, color='k', antialiased=1, ax=None,
> zorder=None )
>
> m.contourf(x, y, data, np.arange(0,9.9,0.1))
> plt.show() `
>
> Resulting plot is here
> <http://matplotlib.1069221.n5.nabble.com/file/n42790/figure_7.png>
>
>
>
> --
> View this message in context: http://matplotlib.1069221.n5.nabble.com/Plotting-NOAA-grib2-data-in-basemap-tp42698p42790.html
> Sent from the matplotlib - users mailing list archive at Nabble.com.
>
> ------------------------------------------------------------------------------
> WatchGuard Dimension instantly turns raw network data into actionable
> security intelligence. It gives you real-time visual feedback on key
> security issues and trends. Skip the complicated setup - simply import
> a virtual appliance and go from zero to informed in seconds.
> http://pubads.g.doubleclick.net/gampad/clk?id=123612991&iu=/4140/ostg.clktrk
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>