Take the 2-minute tour ×
Geographic Information Systems Stack Exchange is a question and answer site for cartographers, geographers and GIS professionals. It's 100% free, no registration required.

I am seeking advice on working with world data sets where the GCS has a positive range (0 degrees to 360 degrees). The data I am working with are NetCDF oceanographic data and have positive coordinate values as mentioned. When displayed in a normal GCS WGS84 in ESRI's ArcGIS 10, it is offset from other data that exist in the range of -180 to 180. If I reproject it, gaps occur around the prime meridian (mostly due to its origins, sometimes as far off as 25 degrees W). My thought is to create a custom projection/coordinate system that has the positive values and hope that other world data sets reproject on the fly just fine. Does anyone see any issues with this or have other solutions? Hope this makes sense. I appreciate any help in advance.

** Update ** This is a screenshot of a countries data set with graticule on top to show the normal CRS of GCS. Just importing the NetCDF data as-is brings it in but the values don't start until 20.5 degrees east of the Prime Meridian.

Normal GCS

If I choose to use the NetCDF's CRS (which is actually the same) it will reproject the world data on the fly if it is defined. Note the graticule is undefined so it does not reproject. Sort of reprojected GCS

But then if I try to reproject everything to a normal GCS with negative values, it's like wrapping a flat map of the data around the world and then it disappears when it hits the Prime Meridian. Sort of reprojected GCS

Now, I know there are no values from 0-20.5 as is stated in the metadata. But why can't the values from rest of the data display in the gap? I even tried a Shift in the Raster Tools to no avail.

share|improve this question
    
geografa - If you define a custom projection couldn't you set the document to another crs (one of the other world data sets) and not require them to "on-the-fly". But the netcdf would? –  Brad Nesom May 25 '11 at 21:38
    
Please post a copy of the NetCDF somewhere. There might be a way to do what you want, but it would be easier to describe it to you using a real example. Essentially, it might be possible to do something like in the "Tying it All Together" portion of this presentation: download.osgeo.org/gdal/presentations/…. OR it might be possible to simply create a new worldfile with standard coordinate ranges in it and overload the built-in georeferencing in the NetCDF. Either way, having a data sample would help. –  lagerratrobe Jul 28 '11 at 1:06
    
Realized I CAN reproject the raster. I just had to give it the right offset. I'll post a more succint answer in the next day or two. For data, check data-portal.ecmwf.int/data/d/edit/personal/temporary/netcdf29 . –  geografa Aug 6 '11 at 17:28
    
I'd be interested in setting the right offset here. One thing to note, is that, I believe, 0 and 360 are the same. –  Hairy Aug 8 '11 at 9:52

3 Answers 3

OK, I think the main issue here is that ESRI use the centre of the grid to calculate the position, and when I have written the data, I am using the bottom left value of the cell to calculate the data. How I resolved this, was by creating my netCDF file with an X grid origin of the centre of the first cell, then calculating each cell using that as a starting point. Drop me a note, in chat, if you'd like to discuss this further, as it now works.

share|improve this answer

If needed, you can easily convert your rasters to -180 180 with GDAL

http://eloiribeiro.wordpress.com/2009/07/08/coverter-imagens-0%c2%ba-360%c2%ba-a-180-180%c2%ba-com-gdal/

(in Portuguese, you will need to use Google Translate)

share|improve this answer

This is a known bug now, as we have experienced the same issue. With a bit of tooing and fromming, which ended up with me sending them details of panopoly, to view the netCDF data without the missing line.

Interestingly, this isn't solely limited to the netCDF. If you create points, using the grids inside the netCDF, and then aggregate the points to build polygons, it will not build a polygon over the prime meridian too, it will be separate ones.

The bug is NIM070286: NetCDF Raster Layer does not display data located along prime meridian

I have requested a hot fix from ESRI Inc. I'll keep this as a favourite question and update it all once I hear about it some more. I am on holiday soon, so hopefully this will coem over the next 10 days.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.