[ncl-talk] Plotting Lat/Lon Points on a map

[Dave Allured - NOAA Affiliate]

Mary,

That sounds good.  I suggest that you aim for fairly high accuracy, better
than this apparent 100 meter discrepancy between GADM U.S. states and
Google Earth.  There seems to be a general trend for higher resolution data
and graphics, and it would help if NCL graphics could keep up with this
curve.  Thanks for working on this.

Perhaps you could switch your built-in outline features to something like a
shapefile-based system, rather than the home grown NCARG system that has
been used for a long time.  I don't know how much work that would be, but
it might make outline updates easier in the long run.

--Dave


On Thu, Jun 1, 2017 at 9:37 AM, Mary Haley <haley at ucar.edu> wrote:

> Hi Dave,
>
> We have a ticket (NCL-2207) for modernizing the map outlines in NCL, and
> we hope to do this through the use of public domain sources, like GADM or
> similar. I just made this ticket a higher priority, and it recently was put
> in our road map as a high priority.
>
> I agree that the use of shapefiles is the best route for now. We have many
> examples at:
>
> http://www.ncl.ucar.edu/Applications/shapefiles.shtml
>
> and a template for adding shapefile outlines to an existing plot:
>
> http://www.ncl.ucar.edu/Applications/Templates/#
> AddingShapefileOutlinesTemplates
>
> --Mary
>
>
> On Tue, May 30, 2017 at 7:32 PM, Dave Allured - NOAA Affiliate <
> dave.allured at noaa.gov> wrote:
>
>> Tim and NCL support,
>>
>> State and county lines in the current NCL database are not accurate for
>> high resolution maps.  Please see the attached figure.  Near Tim's
>> reference point, NCL's built-in NC/VA state line is about 1 kilometer north
>> of the same line in the GADM United States shape file.  This offset is
>> large enough to move the appearance of the reference point into the wrong
>> state.  It appears that the GADM shape file is much more accurate than the
>> NCL database.
>>
>> Can the NCL built-in database be updated with more accurate state and
>> county lines?
>>
>> Tim, for an interim fix I recommend disabling the built-in state
>> outlines.  Instead, plot state and county outlines using an accurate shape
>> file such as the GADM U.S. shape file.  See the attached NCL script for an
>> example.  I recommend against using any manual offsets, because those can
>> lead to abundant hidden mistakes.
>>
>> In addition, I also found discrepancies in state and county lines,
>> between the GADM U.S. shape file and Google Earth.  The state line
>> discrepancy at this Virginia location is about 100 meters, or in general
>> about an "order of magnitude" less than the same discrepancy with NCL.
>>
>> --Dave
>>
>>
>> On Sun, May 28, 2017 at 10:13 AM, Tim Melino <melino33 at gmail.com> wrote:
>>
>>> Thanks Barry and Dennis for your help on this!
>>>
>>> Dennis, I added your  my other question was why is the point in NCL
>>> plotted south of the North Carolina border with NCL? The actual location is
>>> in Virginia, I noticed this was occurring in the image that you sent me
>>> too.
>>>
>>> [image: Inline image 1]
>>>
>>>
<snip>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20170601/fc6eb4e2/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: sample_point.png
Type: image/png
Size: 95010 bytes
Desc: not available
Url : http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20170601/fc6eb4e2/attachment.png