east-west and up-down time series

This topic contains 4 replies, has 5 voices, and was last updated by  matusbakon 3 months, 3 weeks ago.

  • Author
  • #2780


    Hello everybody, I would like to estimate east-west and up-down time series from ascending and descending displacement time series. The time series function in the Multi-sensor module is not complete so is anybody aware of other tools (qgis) or techniques that can estimate the time series without extensive coding?

    Thank you!

  • #4082


    Hello Dear All,
    I have the same question as mentioned by @adugna. IS there any way to analyze the Sentinel-1 data in the east-west direction in Sarproz. If there is, could you share documentation about that. Thanks

  • #4087


    At the moment you can project the velocities using the Multi-Sensor module
    Here a bit of description
    In the next future we will implement also the projected time series export

  • #4088


    Hi all,
    I don’t know if there is a QGIS tool to compute the up-down and E/W component, let me know if someone knows.
    Anyway, it is possible to follow this procedure: 1) starting from geocoded results, resample the measurement points (MPs) using the same grid for both ascending and descending; 2) for each cell, separately for asc and desc, compute the average velocity (to obtain a finer estimation you can use a weighted function accounting for the distance of the MPs from the cell center); 3) apply the vector composition to estimate up-down and E/W vectors. The LOS direction is in the SiteDescription.txt file (i.e. AcqHeading <<the angle with respect to the North>> AcqIncAngle: <<incidence angle with respect to the vertical direction>>)

  • #4089



    There’s simple approach for LOS decomposition published here (see Appendix):

    The strategy is further described and applied for example here (Section 4. Results, end of page 4):

    Although, east-west component refers to “azimuth look direction” here and only 2 orbit tracks are used, for obtaining east-west component and more robust transformation 3 or more satellite tracks should be used (some details can be found here: Obtaining vertical component should be even easier as you just need to multiply LOS displacement with cosine of an incidence angle.



You must be logged in to reply to this topic.