[ivca-cz]

Hello,

if I understand it well and no unwrapping is performed (in small area processing), how is it possible that residue values reach up to plus minus 8? Especially for some dates – but generally, there are very few dates for which all residues are smaller than pi. I did not tick anything (model, lin trend etc.) before the export. How were the higher values achieved and what do they mean?

Thank you for the answer,
Ivana

[periz]

can you send us some grab-screens?
in particular, about the time series you mention and about the options you used for processing
please include also the graph you used for the analysis

[ivca-cz]

I am sending matlab-parsed histograms of residues for few dates (each date separately). Often, the higher residues correspond to high values of ktemp (I want to know if the possible correlation is real or is caused by unwrapping). Linear deformation trend was not estimated.

Computed approx. a month ago, in an older version (after Ktemp was added into CSV). Now I am only reloading the computations.

Attachments:

You must be logged in to view attached files.

[periz]

we’d need to know how u computed the time series…
if u estimated and included the thermal expansion, then it’s reasonable…

[ivca-cz]

I do not understand. I computed it in the small area processing, as in the screenshot.

Nothing is ticked in the “data series” line. Do I understand well, that if I do not tick anything, I get only the residues (if I tick temp, the temperature effect is added etc.), and the same applies to the CSV export?

[periz]

i cannot see what u did from the screenshots…
if, as u say, nothing is ticked, u should get only residues…
next question: are u using the star graph or something else?

[ivca-cz]

I am using star graph. What else can be wrong?

The time series as you can see was generated right now, from saved data, with nothing ticked.

[periz]

hi ivana I cannot see it from what you have attached
you should send pictures with (taken from the small area module):
– multi image insar analysis window so that we see the options
– geocoding windows
– example of time series with problems
isn’t it that it’s in mm?

[ivca-cz]

OK, thank you. I ve just discovered that the residue values in the CSV are also in mm, therefore they go up to plus minus lambda/4.

Ktemp (in the CSV) is in rad/deg or mm/deg? That could be the reason of problems.

[anarizka]

Dear Prof.Daniele

I got results from CSV file, but I cant find residual Height value.
What mean of “HEIGHT WRT DEM and SIGMA HEIGHT” in CSV column?
If Residual Height = Height – External DEM, Could I say that Residual Height = HEIGHT – HEIGHT WRT DEM (based on CSV column)

Besides that, in SARPROZ, how I can generated and get data “phase residual”. I want to know the value before and after APS processing.

Best Regards

Ana

Attachments:

You must be logged in to view attached files.

[ivca-cz]

Hello Ana,

(estimated) residual height value is the HEIGHT WRT DEM.

SIGMA HEIGHT is the estimated standard deviation of the HEIGHT (resp. HEIGHT WRT DEM, they are the same), for its background, see another discussion thread. If you are interested, ask more, as it was not yet finished but I do not have time and concentration for it now.

The phase residuals are the values corresponding to each date (the second (right) half of the table). They are not in rad, but in mm. And exporting to CSV, you should be careful what is ticked in the line “data series – mov – temp – zstart” etc. If you do not tick anything, you get the (original) phase values just after residual height component subtraction, but if you tick “mov”, you get the phase value after residual height and movement subtraction etc.

Regards, Ivana

[anarizka]

Dear Ivana,

Thanks for advise and explanation, I get it

Best Regards

Ana

[Author]

Posts