The output from the Snap v6.0.4 Reprojection operation extends data beyond the edge of the source data. This does not happen when using Snap v6.0.1 as seen in the attached image.
My source file is a S3 OLCI netcdf file output from NASA’s l2gen. S3tbx is disabled so it’s using the CfNetCdfReaderPlugin product reader.
Thanks,
Andrew
Hi
I’m wondering if anyone has any insight into this issue or perhaps a work around. Is it a bug?
Here’s hopefully a clearer description of the problem which as i mentioned only showed up after v6.0.1.
I’m reprojecting an OLCI L2 NetCDF file generated by SeaDAS l2gen from a Sentinel_3a L1B to a fixed WGS84 UTM N17 300m grid that is only partially covered by the input file. In this example the grid extents north of the granule. I’d expect all cells beyond the extent of the source data to be filled with NaN’s but since v6.0.2 that’s not the case (as seen upper image attached below).
If performing the same operation, but changing the input to the L1B SAFE file, the projected image looks as expected (as seen in the lower image attached below). There is no buffering of the data beyond the extent of source data.
This problem is not limited to just this example, but occurs whenever reprojecting a granule that only partially covers the extent of the grid to which it is being projected.
The source in this example was S3A_OL_1_EFR____20180709T154117_20180709T154417_20180710T220627_0179_033_168_2340_MAR_O_NT_002.SEN3.
Sample reprojection graph:
<node id="reproj_node">
<operator>Reproject</operator>
<sources>
<source>read_node</source>
</sources>
<parameters>
<wktFile />
<crs>PROJCS["WGS 84 / UTM zone 17N",GEOGCS["WGS 84",DATUM["WGS_1984",SPHEROID["WGS 84",6378137,298.257223563,AUTHORITY["EPSG","7030"]],AUTHORITY["EPSG","6326"]],PRIMEM["Greenwich",0,AUTHORITY["EPSG","8901"]],UNIT["degree",0.0174532925199433,AUTHORITY["EPSG","9122"]],AUTHORITY["EPSG","4326"]],PROJECTION["Transverse_Mercator"],PARAMETER["latitude_of_origin",0],PARAMETER["central_meridian",-81],PARAMETER["scale_factor",0.9996],PARAMETER["false_easting",500000],PARAMETER["false_northing",0],UNIT["metre",1,AUTHORITY["EPSG","9001"]],AXIS["Easting",EAST],AXIS["Northing",NORTH],AUTHORITY["EPSG","32617"]]</crs>
<resampling>Nearest</resampling>
<referencePixelX>0.0</referencePixelX>
<referencePixelY>0.0</referencePixelY>
<easting>281282.934996</easting>
<northing>4764308.997997</northing>
<orientation>0.0</orientation>
<pixelSizeX>300.000000</pixelSizeX>
<pixelSizeY>300.000000</pixelSizeY>
<width>1377</width>
<height>639</height>
<tileSizeX />
<tileSizeY />
<orthorectify>false</orthorectify>
<elevationModelName />
<noDataValue>NaN</noDataValue>
<includeTiePointGrids>true</includeTiePointGrids>
<addDeltaBands>false</addDeltaBands>
</parameters>
</node>
</graph>
Thanks,
Andrew
Hi Andrew,
Did you figure out a solution for this? I am experiencing the same issue, reprojecting the outputs of Polymer (https://www.hygeos.com/polymer) to WGS84.
Thanks
James
I see that there is an update notification in the lower right of your SNAP.
It could be that it is fixed with the latest updates for SNAP 7. If not it will be with the next major release of SNAP.
Thanks for looking at this.
I have updated SNAP and the bug has persisted.
That’s great to hear, let me know if you need any files or anything so that you can reproduce the bug.
Thanks
James
Hi dear,
I had this problem with Sentinel-3 SLSTR data.
In order to solve this problem, I have used a trick.
before reprojecting the desired area (for example Paris area), I have subset the larger area than Paris. That is, I subset about 50 kilometers from around Paris.
then, I reproject the Previous step image (Paris with 50 km around it)
finally, I subset the exact area of Paris with coordinates.
This will remove the edge of the image that has been extended due to reprojection from the original image.
Hi you can use this trick i explained
Good luck
Thanks for you solution! I’ll see if I can apply it to my workflow.
Thanks!
