I am running Radarsat-2 data through StaMPS (Windows machine, pre-processing in SNAP 7.0). The pre-processing workflow is as follows:
Band-Select --> Stack --> Terrain-Correction --> Output: Terrain Corrected Stack
Stack (not Terrain Corrected) --> Interferogram and Topo Phase Removal --> Terrain-Correction --> Add lat/lon bands using band math --> Output: Terrain Corrected Interferogram Stack with Lat/Long bands
Then, I export the outputs to StaMPS. All the pre-processing is done in gpt.
Unfortunately, after the StaMPS processing finishes, I get an error that says Error using load ‘mean_v’ is not found in the current folder.
I went through all the intermediate products to make sure that there were no errors in the input data and I found the triangles that form due to a low resolution of the DEM, so I increased the percent tile overlap and those went away. Then, I tried to not Terrain Correct to make sure that the Terrain Correction wasn’t causing the problem, and I still get the same error. To test if it was a data volume issue, I used a large stack >20 scenes, a small one with 6 scenes, and a medium sized one with 12 scenes, all with the same result. I also geographically subset a large stack to see if it was an issue with geographic extent (perhaps too many scatterers?).
Does anyone have any recommendations on how I may be able to tackle this?
Edit: I am using the latest version of StaMPS: https://github.com/dbekaert/StaMPS/releases/tag/v4.1-beta