two possible explanation (probably both apply)
- both give you deformation along the line of sight (LOS), so ascending and descending can be different, especially when there is vertical and horizontal motion. There are approaches which try to combine both results (Vertical Velocity Displacement along Ascending and Descending tracks) and we plan to implement this in SNAP as well in the future
- unwrapping errors in one (or both) of the results can cause large differences over the image. It depends on the degree of coherence and the quality of the interferograms how reliable the results are