GNSS interpolation

[gshirazinejad]

Hello, Which software should I use to interpolate my GNSS data? Thanks.

[espiritocz]

hi,

perhaps use verde? e.g. https://www.fatiando.org/verde/latest/gallery/spline_weights.html

On Sun, 28 May 2023, 20:05 gshirazinejad, @.***> wrote:

Hello, Which software should I use to interpolate my GNSS data? Thanks.

— Reply to this email directly, view it on GitHub https://github.com/andwatson/decompose_insar_velocities/issues/19, or unsubscribe https://github.com/notifications/unsubscribe-auth/AADYYKVMCQKO2IUEZHTQDE3XIOOYJANCNFSM6AAAAAAYSAOFZM . You are receiving this because you are subscribed to this thread.Message ID: @.***>

[andwatson]

Hello, Exactly how to interpolate the GNSS velocities is a bit down to person preference, but a couple options are:

• spline interpolation (as suggested by espiritocz), this can also be done in Matlab or GMT.
• Kriging interpolation (as done in "Large-Scale Interseismic Strain Mapping of the NE Tibetan Plateau From Sentinel-1 Interferometry"), be careful with the underlying assumptions though. There's a kriging python package which can do this, and Matlab probably has something for it as well.
• The method in "Optimal Interpolation of Spatially Discretized Geodetic Data".

[gshirazinejad]

Thank you for your respond. Which one did you use for article ''Interseismic Strain Accumulation Across the Main Recent Fault SW Iran From Sentinel‐1''?

[andwatson]

For that paper I used a 2nd order polynomial to interpolate the GNSS velocities, however, I wouldn't recommend that in hindsight. We've recently submitted a new paper for a full-Iran InSAR velocity field and for that I used Velmap (Wang & Wright 2012). Alternatively, Qi Ou used kriging interpolation in this paper: https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2022JB024176 Which worked well but does require a number of assumptions to be valid.

[gshirazinejad]

Thank you very much.

[gshirazinejad]

Can we use Velmap to interpolate GNSS data and then tie them to InSAR velocities? As I understood, there is difference between interpolated GNSS data (in the stage that we tie InSAR velocities to them) and GNSS velocities that are produced using Velmap. Is there any difference between these 2 sets of interpolated GNSS data?

[andwatson]

Sorry not sure I follow your question.

The InSAR velocities get tied to the interpolated GNSS velocities. Velmap is one way of generating the interpolated GNSS velocities that are then used for the InSAR reference and the decomposition. Its actually two-stages of interpolation, the first using a triangular mesh and a smoothing factor to produce smoothed velocities at each vertex, and a second fine interpolation to match those vels to the unified InSAR grid. The advantage of this method is that uncertainties in the GNSS velocities are smoothed out, and we can achieve reasonable velocities in areas with low GNSS coverage.

[gshirazinejad]

Thank you very much for the respond Andrew.