insarlab
Mean velocity result
@mirzaees I've got some interesting results of landslide about the fading signal issue. So, I've been doing some experiments and trying out different data processing techniques ? the result seems unexpected . The results from each method are surprisingly similar. What would be the reason?
I would just use the default. What matters more is the phase-unwrapping network. You may want to try single-reference and delaunay with different options. Masking matters of course.
@falkamelung Thanks so much for your response. My understanding from [1] is delaunay network is better so I applied that. Then I have time series which is mask according temporal coherence .The western side of the hillslope looks masked in comparision of SBAS which shows high rate of deformation . I applied masked in the following result
Do we expect to end up with a very similar result from the all the phase linking method or they should look different? To what extend we should expect that deformation rate could decrease while we use same delaunay network? Is deformation rate could decrease more than 1 or 2 cm/yr for some point?
1:Mirzaee, S., Amelung, F., Fattahi, H., 2023. Non-linear phase linking using joined distributed and persistent scatterers. Comput Geosci 171, 105291. https://doi.org/10.1016/j.cageo.2022.105291.
I would try single-reference network as phase-unwrap errors don't propagate. If you see the same signal as with Delaunay network you can be confident. If different you may want to try different Delaunay networks.
Thanks once again @falkamelung, what about magnitude of deformation rate? what If I see a signal there but with different magnitude
@mohseniaref, I expect all phase linking techniques have similar results, however the magnitude of deformation rate is slightly different/smaller for sequential mode. please refer to figure 2 in the paper comparing in the simulations. SBW or SBAS on the other hand, are affected by fading signal and have higher deformation rate where it is present
