Velocity output seems to be affected by latitude

[AntaraAntariksa]

Hello! I've been trying to use dolphin for the past week. I've run some ALOS-2 data using alosStack.py in ISCE2 to get coregistered images. Then I use the data in the folder dates_resampled as an input. I've run it through dolphin just fine but I seem to get weird result from the velocity output?

However, when i try to open one of the images in timeseries folder (not geocoded yet), it seems fine (?) so I don't know what went wrong

Also, here's what the velocity file looks like before geocoding

For reference, I also use dolphin to run some Sentinel-1 data and the output seems fine

[AntaraAntariksa]

Ah, apparently I have this warning when running dolphin. But I still don't have any ideas how to fix it?

[scottstanie]

Just checking - Is the output geocoded velocity weird because of the red-to-blue gradient? Or because of the shape (more square rather than a long rectangle)?

It's hard to tell if it's a latitude problem without seeing the background map: Did you use any --strides when running dolphin? Or is the output the same size as the input SLCs?

[AntaraAntariksa]

It's weird because the red-to-blue gradient. Below is a comparison between Sentinel-1 output with ALOS-2 output. The output of sentinel-1 is somewhat similar with the actual subsidence that's happening in that area, that's why I call the output of ALOS-2 is kinda weird.

And yes, i use strides for both sentinel and alos, i used --strides 10 5. And both the size of input SLC and output is the same

[scottstanie]

Ok thanks for clarifying!
I agree that's strange... I haven't used ALOS-2 before, so I'm not sure what type of corrections the ISCE2 processor uses. But my first guess would be ionospheric effects, since

• L-band is 16x more sensitive to the ionospheric than C-band (Sentinel-1) and
• the radar-coordinate image looks similar, so it don't think it's a geocoding artifact.

Does the ALOS processor run corrections by default? What's the general study area here?

[AntaraAntariksa]

Really sorry for the late reply.

Yes, I figured that the problem is with ionosphere correction. The correction runs by default but not applied it directly to coregistered SLC. Now, I'm still trying to figure out how to apply the corrections to the data.

The study area is on Borneo Island near Palangkaraya City with the main focus on the peatland area

[AntaraAntariksa]

Well, the only thing I can think of is get corrected interferogram using ISCE2. But I'm really not sure how to continue the process?

I tried preprocessing using alosStack.py in ISCE2, but now I don't know what next...