look vectors problem

[CUMT530849827]

I processed gmtsar data with kite and found that the look vectors for lifting the rail were incorrect. I processed two earthquakes in a row and found that it was the same in both cases. Could you tell me how this happened? How should I solve this problem https://github.com/pyrocko/kite/issues/73#issuecomment-1648016837
Here are the look vectors for the two earthquakes I worked with The following is the direction of the lift vector given in the case list

[hvasbath]

Yes, so please make sure that the los vector is correct. You need to rotate your heading angle by 180 degree.

[CUMT530849827]

@hvasbath @miili I calculated the enu of the ascending and descending orbits and found that there was no error here. I searched for many reasons and did not find it, but I found https://github.com/hvasbath/beat/issues/93 that the ascending vector in this problem was the same, and I wondered if there was a problem with the drawing script https://github.com/hvasbath/beat/issues/93 The picture in question is below

[hvasbath]

First of all there is no point in showing plots of BEAT. Please open the kite scenes with spool and check the look vectors there. BEAT does nothing but reading from Kite, if there is sth wrong then happens in the import from GMTSAR. I dont understand what the first column of your screenshot is supposed to be?

[CUMT530849827]

I used spool to open the track lifting data, but I don't know how to check their look vectors

[hvasbath]

You can select phi here, but already from the formula in scene_io we can see there is sth wrong.

For ascending track this arrow you marked must look into the exact other direction- so rotated by 180 deg.

So the error is in the equation you marked just here:

Until that is fixed I suggest you load your scene through a python script and add pi.

from kite.sceene import Scene
from numpy import pi

scn = Scene.load("your_scene_name")
scn.phi += pi
scn.save('kite_scene_outname')

Then please check again in spool. And once that is correct- please reimport to BEAT.

[CUMT530849827]

Thank you very much for helping me find the mistake so kindly and carefully