[gdal-dev] Questions on working with HYCOM data in netcdf format

Kurt Schwehr schwehr at gmail.com
Mon Jun 6 10:25:05 PDT 2016


+gdal-dev

Some hackish test code I had laying around from last year...

https://gist.github.com/schwehr/01f6604afc7757ea0a676f0eb28be582

You might also be able to just write 80 layers of u-v, pick out pairs for
each depth layer and have the rendering system do the conversion(s).  You
could use whatever you want once you have a geotiff.  In this screenshot, I
made an 80-layer geotiff of u and v alternating layers and uploaded that to
Earth Engine.  It's been a long time since I heavily used mapserver, but
you could probably do the same there.

https://www.flickr.com/photos/schwehr/26895009754/

On Mon, Jun 6, 2016 at 7:51 AM, Stephen Woodbridge <woodbri at swoodbridge.com>
wrote:

> On 6/5/2016 2:04 PM, Kurt Schwehr wrote:
>
>> Stephen,
>>
>> Take a look at these two discussions for starters on working with hycom:
>>
>>
>> http://gis.stackexchange.com/questions/167155/how-can-i-make-geotiffs-of-individual-depth-layers-in-the-hycom-ocean-circulatio
>>
>
> Hi Kurt,
>
> In the link above you mention:
>
> I've got some code written the does things like calculate the speed
>> inm/s from the UV and creates a geotiff, etc
>>
>
> Can you share this code, it sounds like exactly what I need for the
> current vectors? And I'm interested in figuring out how to read/write and
> work with the netcdf files.
>
> I think for now I'll work with the "u" datasets as they seem that they
> will be much easier place to start.
>
> Thanks,
>   -Steve
>
>
>
>> http://gis.stackexchange.com/questions/170882/how-can-a-make-a-netcdf-with-subgroups-smaller
>>
>> You should start with the regular grids.  e.g.
>>
>>
>> ftp://ftp.hycom.org/datasets/GLBu0.08/expt_91.1/data/hindcasts/2015/hycom_glb_911_2015093000_t000_uv3z.nc
>>
>> Make an 80 layer geotiff from the u/v and then use mapserver as Even
>> suggested.
>>
>> -kurt
>>
>>
>> On Sat, Jun 4, 2016 at 6:29 AM, Even Rouault <even.rouault at spatialys.com
>> <mailto:even.rouault at spatialys.com>> wrote:
>>
>>     Hi,
>>     >
>>     > I want to convert some HYCOM data into GTiff format and I'm a little
>>     > lost on how to get started, whether I can do what I need just using
>> a
>>     > vrt or if I'm going to have to write code to process the data.
>>     >
>>     > To start with I have placed gdalinfo on one file here:
>>     > http://imaptools.com:8080/dl/hycom-gdalinfo.txt
>>     >
>>     > The data comes from http://hycom.org/dataserver/glb-analysis
>>     > and I will need to be using GLBa0.08 dataset.
>>     >
>>     > I have two use cases:
>>     >
>>     > 1) The simple one is to extract say Band 1 into a georeferenced
>> GTiff
>>     > and apply color to using color ranges without loosing the NODATA. I
>>     > think I can do later using .vrt, correct?
>>
>>     gdaldem color-relief can output to a VRT, but given that this netCDF
>>     file is
>>     georeferenced through a geolocation array, you'll likely have first
>>     to run
>>     gdalwarp -geoloc to have something useful at the end.
>>
>>     >
>>     > I'm not sure how to deal with the georeferencing, because the
>> download
>>     > site says: "Native hycom .[ab] data converted to NetCDF on native
>>     > Mercator-curvilinear HYCOM horizontal grid" but that does not seem
>> in
>>     > sync with the Metadata in the file.
>>     >
>>     > 2) The more complicated case is that there are two files:
>>     >     a)  eastward_sea_water_velocity (aka: u, u-velocity)
>>     >     b)  northward_sea_water_velocity (aka: v, v-velocity)
>>     > that I would like to use to create a grid of vectors with arrowheads
>>     > using u and v to define the vector and where the color of the
>> vector is
>>     > related to the magnitude of its length. Ultimately this data will
>> get
>>     > rendered via mapserver so I'm wondering if I can do this via a
>> GTiff or
>>     > if I will have to resort to creating a massive point shapefile with
>>     > attributes of angle and magnitude and use a symbol. It seems like it
>>     > would be best if I can keep the data as a GTiff and then sample the
>>     > points and render the vectors on the fly for when you zoom in/out.
>>
>>     See
>>     http://mapserver.org/input/vector/vector_field.html
>>
>>     You don't need to create a vector file.
>>
>>     You'll have to preprocess your file to get the u and v bands in the
>> same
>>     raster.
>>
>>     Even
>>
>>     --
>>     Spatialys - Geospatial professional services
>>     http://www.spatialys.com
>>     _______________________________________________
>>     gdal-dev mailing list
>>     gdal-dev at lists.osgeo.org <mailto:gdal-dev at lists.osgeo.org>
>>     http://lists.osgeo.org/mailman/listinfo/gdal-dev
>>
>>
>>
>>
>> --
>> --
>> http://schwehr.org
>>
>
>
> ---
> This email has been checked for viruses by Avast antivirus software.
> https://www.avast.com/antivirus
>
>


-- 
--
http://schwehr.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/gdal-dev/attachments/20160606/f7197c68/attachment.html>


More information about the gdal-dev mailing list