Terrain correction producing different sized files for S-1 stack

Microwave Toolbox Interferometry
1

I’ve got a stack of interferograms and they aren’t the same size after terrain correction, despite the unwrapped interfergrams being the same size. I can understand if they are maybe 1 row or sample length different, but not several values off.

SRTM 1Sec HGT and BISINC_21_POINT_INTERPOLATION has been used for Backgeocoding and Terrain Correction

For terrain correction:

  • alignToStandardGrid True
  • standardGridOriginX 0.0
  • standardGridOriginY 0.0

All interferograms have 12 bursts and 3 swaths.

ifg tc_samp tc_lines unw_samp unw_lines
ifg1 11410 10142 8306 8596
ifg2 11410 10142 8306 8596
ifg3 11410 10143 8306 8597
ifg4 11405 10141 8306 8596

I need to use these interferograms in MintPy for time-series processing, but it warns of the different file dimensions.

Can anyone provide any thoughts on why this is?

Thanks

2

as all input products (and interferograms) were acquired with slightly different incidence angles, they represent different areas on the ground. These differences become visible only after geometric terrain correction.

If you need all of them in the same geometry, you can apply the collocation tool with a suitable master product and resampling method.

3

Thanks @ABraun, that makes sense. I’ll modify my workflow to incorporate this.

4

A more convenient way would be to stack all input products in the beginning and create interferograms in the stack. These would have the same geometry from the beginning, but currently only single-master processing is supported.
Multi-master interferograms (from the same stack) will be supported in future versions of SNAP.

5

That’s good to know. I’m using multi-master interferograms, but I can easily incorporate it into my snappy workflow. I’ve already got a ‘master ifg’ set, so I can get the rest to resample to that one.