search

GeoscienceAustralia / PyRate

A Python tool for estimating velocity and time-series from Interferometric Synthetic Aperture Radar (InSAR) data.
https://geoscienceaustralia.github.io/PyRate/
Apache License 2.0
203 stars 71 forks source link

frequency mistmatch during gamma prepifg in slc files #49

Closed basaks closed 8 years ago

basaks commented 8 years ago

Clarify if the frequencies in two gamma_slcs are indeed different, and whether it is safe to ignore the difference after a few decimal place, say 6?

For example, this exited with wavelen == hdr1[ifc.PYRATE_WAVELENGTH_METRES] evaluating as False:

/g/data/dg9/INSAR_ANALYSIS/SURAT/TSX/GAMMA/T140A/pyrate_files/unw_ifms/20150103-20150216_HH_4rlks_utm.unw.

20150103_HH.slc.par has radar_frequency: 9.6499986e+09 Hz

20150216_HH.slc.par has radar_frequency: 9.6499993e+09 Hz

Small differences in wavelengths/frequencies were ignored with the following:

np.isclose(wavelen, hdr1[ifc.PYRATE_WAVELENGTH_METRES], atol=1e-6)

See change in git # 88043383bfd4dec80d6155906d353f48a2e32a48.

mcgarth commented 8 years ago

I am not sure why the radar frequency is changing in the Gamma software output, but it has very little impact on what is being measured when the difference is at the 6th decimal place.

Therefore it is acceptable to use the 1e-6 tolerance. With this you are saying that the wavelength must agree to 1000th of a millimetre. In terms of our measurand this is much much smaller than what we can reliably resolve with the technique (i.e. a tenth of a millimetre at best).