AOI Subset and varying results

[Muhammad Murtaza Nadeem]

Hi all, 

I am working to do a landslide timeseries analysis. Have done preprocessing using asf hyp3, my question is regarding selecting the AOI subset in mintpy config, because changing the subset size is slightly changing the results. 

To confirm my assumption, I checked my mean velocity in the red outlined area, and it was slightly changed as I increased my AOI subset size. 

Also, what is the ideal choice for adding a buffer around my study area (What should be the minimum size of AOI subset in which results can be considered true/most accurate?) 

Thanks

[Steven Brooks]

Hi, I usually use ISCE2 for processing interferograms before mintpy and in that case minimum is 30km x 30km and when I do subsets in mintpy nothing changes. 

My suggestion is to work on a bigger area and then do zoom in with tsview.py and save the outputs.

--
You received this message because you are subscribed to the Google Groups "MintPy" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mintpy+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/mintpy/601f5808-c465-4f3f-b433-524254608fa8n%40googlegroups.com.

[litao xu]

My habit is to use areas with high coherence (3x3 rasters) in the study area, and INSAR inverts the relative displacement relative to the reference point, so you should ensure that mintpy can choose a relatively stable reference point within your AOI, usually rock, bare soil. Based on this, generally 30kmx30km is the ideal range, or if you use HYP3's multi_burst products, you can not crop it.

[Muhammad Murtaza Nadeem]

Thanks alot, 
I manually selected the reference point in valley (having houses) and it was same and present in all AOI's, yet there was difference in mean velocity.