Strange results for SBAS (GMTSAR + StaMPS)

Hello all, I have just finished processing SBAS using GMTSAR + StaMPS and I get this very strange result:

ps_plot(‘u’) shows the same problem:

Does anyone recognize this anomaly?

Using the following params in Stamps with amplitude dispersion index of 0.6:

                          Created: '01-Mar-2021'
                       clap_alpha: 1
                        clap_beta: 0.3000
         clap_low_pass_wavelength: 800
                         clap_win: 32
                     density_rand: 2
                   drop_ifg_index: []
                 filter_grid_size: 50
                 filter_weighting: 'P-square'
         gamma_change_convergence: 0.0050
             gamma_max_iterations: 3
               gamma_stdev_reject: 0
                          heading: -13.5919
                  insar_processor: 'gmtsar'
                           lambda: 0.0555
                    lonlat_offset: [0 0]
                     max_topo_err: 20
              merge_resample_size: 0
               merge_standard_dev: Inf
                          n_cores: 8
                     percent_rand: 1
                plot_color_scheme: 'inflation'
                 plot_dem_posting: 90
            plot_pixels_scatterer: 3
              plot_scatterer_size: 120
             quick_est_gamma_flag: 'y'
                ref_centre_lonlat: [0 0]
                          ref_lat: [-Inf Inf]
                          ref_lon: [-Inf Inf]
                       ref_radius: Inf
                     ref_velocity: 0
               sb_scla_drop_index: []
                      scla_deramp: 'n'
                  scla_drop_index: []
                      scla_method: 'L2'
                   scn_deramp_ifg: []
                 scn_kriging_flag: 'n'
                     scn_time_win: 365
                   scn_wavelength: 100
                    select_method: 'DENSITY'
          select_reest_gamma_flag: 'y'
                  shade_rel_angle: [90 45]
                          slc_osf: 1
              small_baseline_flag: 'y'
                      subtr_tropo: 'n'
                     tropo_method: 'a_l'
                     unwrap_alpha: 8
                unwrap_gold_alpha: 0.8000
                unwrap_gold_n_win: 32
                 unwrap_grid_size: 200
          unwrap_hold_good_values: 'n'
             unwrap_la_error_flag: 'y'
                    unwrap_method: '3D_QUICK'
               unwrap_patch_phase: 'n'
            unwrap_prefilter_flag: 'y'
    unwrap_spatial_cost_func_flag: 'n'
                  unwrap_time_win: 730
                   weed_max_noise: Inf
                  weed_neighbours: 'n'
                weed_standard_dev: Inf
                    weed_time_win: 730
              weed_zero_elevation: 'n'

Hi Stefan,

Have you already found the problem? Yes, it’s a kinda strange result for unwrapped. Can you do ps_plot(‘w’) and ps_plot(‘usb’)? Do they also look strange?

You might also check the amplitude difference dispersion (ADD) from scatter_SB.ps inside the raw folder.

Hi Stefan!

Sorry for out of topic. Do you have any documents to do the GMTSAR+STAMPS?

thanks

No, but if you have any problems I would check on the GMTSAR Forum.

1 Like

Unfortunately I moved on to another orbit as I couldn’t trust these results.

I just have an irrelevant question if you wouldn’t mind. How could you get snap or Matlab to plot this Acquisition time vs Perpendicular baseline chart ?

If you are using GMTSAR then preproc_batch_tops.csh will generate a .ps file with time x baseline graph.

Unfortunately my study area isn’t available yet for GMTSAR data base. I am using Snap and Matlab_operated StaMPS

In SNAP you can generate a baseline graph from your InSAR Stack. Is this what you are looking for?

I have checked this out earlier and this method requires a single stack file of all the input co-registered scenes, while the automated snap processing produces multiple stack files of co-registered master-slave pairs which isn’t a valid input for the baseline chart tool in SNAP. I do, however, appreciate attempting to help :blush:

Are you doing SBAS? (Multi-Master?

PSI single master

I’m not too sure how I can help in that case. A custom python script would probably be suitable, you could use matplotlib to generate your graphs. Good luck.