Hi all,
I’ve run into some strange behaviour while running the subset operation. It is part of an automated processing chain that we have set up through the cloud and appears similar to some previously posted issues relating to the subset operation in the SNAP graph:
https://forum.step.esa.int/t/error-of-subsetting-in-graph-builder/23515
https://forum.step.esa.int/t/confusion-arises-when-subsetting-sentinel-1-image-based-on-geographic-coordinates-with-subset-tool-using-graph-processing-tool/34985
In our cases, there is an inconsistency in how the subset operation produces results and we are unsure what the root cause is. There are three scenarios with graphics attached below. In each scenario, the black rectangles are the subswath bounds, yellow is the input AOI, and blue is the resultant AOI of the interferograms.
- Input AOI is within the bounds of IW1, generated interferogram is larger in all directions
- Input AOI is within bounds of IW1 and IW2, generated interferogram is larger in the east, west, and northern directions
- Input AOI is within bounds of IW1 and IW2, generated interferogram is shifted to the northeast and only contains ~10% of the original search AOI.
These are all over the same area, using the same path/frame, for the same time period. Any help would be greatly appreciated!
