land subsidence and DEM

[naima aqid]

Hello,

I'm trying to study land subsidence from sentinel data 1, I'm using SANP software for processing. my problem is the displacement map I got a high values ​​around 3 to 5 mm per month, that's a lot for moving. I spoke with the expert who knows approximately the displacement values. Also I did DEM from Sentinel 1 and I got de same problem, around 45m in comparison with SRTM DEM 30 m, or with others DEMs.

Here are all the steps of my treatment:

1- download of sentinel 1 data of my study area (I chosed SLC product and IW mode)

2- I applied the file orbit,

3- Co-Registration 

4- interferogram

5- deburst

6- merge

7- Topographic phase removal 

8-Multilooking

 9-Goldstein flter

10- SNAPhu export

11- SNAPHU Import

12-Range doppler

13 phases to displacement

i did  the Steps 1 to 5 for each IW (1,2,3) and after I have merged all IW1IW2IW3 and I continue from Step 6 to the end.

I need your help as soon as possible.

[Todd Burns]

I'm new to this but, perhaps the values seem high because the displacement values are LOS (line of sight) thus are the hypotenuses of the vertical only competent you are probably interested for subsidence.

In case that is the case, to convert the displacement of LOS to vertical you can use Band Math expression editor to create a band with calculated vertical displacement values.

Since SAR_LOS_displacement = (unwrap_phase * wave_length) / (-4 * PI)

the following expressions have given me reasonable results for Sentinal-1 images:

Vertical_displacement_in_m = (unwrap_phase * 0.0555) / (-4 * PI * cos(rad(incident_angle)))

Horizontal_displacement_in_m = (unwrap_phase * 0.0555) / (-4 PI * sin(rad(incident_angle)))

[naima aqid]

thank you very much for your suggestions, I will try to convert LOS, and I will let you what are my results.

Op woensdag 27 juni 2018 12:18:17 UTC+2 schreef naima aqid: