STAMPS Standard Deviation Export Puzzling Variation

Dear STEP community,
I’ve found some intrigging results in the STAMPS Standard Deviation export, for what I can’t find any explation yet.

I’m using an algorithm adapted from @thho StaMPS-Visualizer manual.

It follows:

%%% the ‘v-doa’ parameter is an example you can change it to your needs

ps_plot(‘v-dao’,‘a_linear’,‘ts’);

%%% a new window will open
%%% in the new window select a radius and location of the radius center to select the PS to export
%%% the ‘v-dao’ parameter is an example you can change it to your needs
%%% but be sure that you use the same paramters as above in the ps_plot()!

ps_plot(‘v-dao’,‘a_linear’,-1);

%%% loading phase displacement velocity from v-dao plot

load ps_plot_v-dao.mat;

%%% concatenation of phase velocity with time series usin lat lon
lon2_str = cellstr(num2str(lon2));
lat2_str = cellstr(num2str(lat2));
lonlat2_str = strcat(lon2_str, lat2_str);
lonlat_str = strcat(cellstr(num2str(lonlat(:,1))),cellstr(num2str(lonlat(:,2))));
ind = ismember(lonlat_str, lonlat2_str);
disp = ph_disp(ind); %% mean displacement velocity
disp_ts = ph_mm(ind,:); %% displacement time-series

%%% leaving one column with zeros to fill with STD results (next)
%%% 1#id 2-3#lon-lat 4#vel 5#stdev 6-n# time-series

export_res_vel = [linspace(1,length(ind),length(ind))’ lonlat(ind,1) lonlat(ind,2) disp zeros(length(ind),1) disp_ts];

%%% To see time table (uncomment next line)
%%% datetime(day,‘ConvertFrom’,‘datenum’)

%%% now lets load Time-series STANDAR DEVIATION for previous selected region
ps_plot(‘vs-dao’,‘a_linear’,-1);
load ps_plot_vs-dao.mat;

%%% variable with STD values
export_res_vel(:,5)=ph_disp(ind);
export_res = [[zeros(1,5) m2xdate(day)'];export_res_vel];
export_res = table(export_res);
writetable(export_res,‘ts_peneda_Asc_20210111_20230218.csv’)

Every time I repeat this procedure for the following PS Time Series:

Using a 50 000 m radius and selecting the same reference lat lon (given by the small balck asterisk (*) in next image):

I always get different STD values, although the TS Velocities are always the same:

FIRST RUN results:
Export4 (is TS mean Velocity) and Export 5 (is Standard Deviation)

SECOND RUN results:

THIRD RUN results:

Does anyone have any guess about what could be trouble here? I really don’t have any idea whats is the issue :thinking:

could it simply be the order of the PS which slightly changes? You could check by calculating the sum of each column and row and see if these are identical.

Hi @ABraun !
Thank you for your suggestion. Unfortunately thats not the case. For instance I have sumed up std of RUN01 and RUN02, and got the following results:

  • sum(STD)run1 = 57838,732001 (mm)
  • sum(STD)run2 = 57317,8350085 (mm)

:thinking: intriguing