Incorrect outline on world view/ reprojection problem

[jamesrunnalls]

Hi All,

I am having some trouble reprojecting a file I produced using POLYMER (from S3).

When I view the file in SNAP it looks fine except that in the world view the extent incorrectly extends to the origin.

I have looked manually into the NetCDF file and cannot see where these zero values are coming from as the latitude and longitude variables seem correct. When I try to reproject it the default size in pixels is far greater than the actual width and height of the raster (313, 239). When I run the reprojection it is very slow and produces bad results.

My assumption is that there is zero lat/lon values somewhere that SNAP is reading. Does anyone know how SNAP calculates the bounding box?

Thanks in advance!!

Here is a link to the file in question: example.nc - Google Drive

[gnwiii]

I opened the latitude and longitude bands from example.nc in NASA Panoply. Using “equirectangular regional” projection they look OK, but the file doesn’t conform to Minimal georeferencing requirements for common climate software so not surprising that some software gets it wrong.

[jamesrunnalls]

Thanks for your response! I will try editing the file and see if that makes a difference.