adjtomo
parameter sort_by = "abs_azimuth" behavior
In recsec, parameter sort_by = “abs_azimuth” brings in unwanted additional normalization to the seismograms. The zip file setup from #156 will still work. Run the script to generate a record section. Then change sort_by to abs_azimuth from azimuth, and see the difference.
@aakash10gupta
Can you clarify what you mean by additional normalization? I'm not sure if I'm recreating the issue you're seeing but when I use your data and plot both by azimuth and abs_azimith I do see some amplitude differences but I would expect that given the y-axes are different. Let me know if I'm missing something
recsec --pysep_path observed_data/AK_20250130_0_200/seismograms --syn_path synthetic_data/OUTPUT_FILES_ISO/seismograms --source synthetic_data/DATA/CMTSOLUTION --stations synthetic_data/DATA/STATIONS --scale_by normalize --xlim_s 0 100 --min_period_s 8 --max_period_s 16 --sort_by azimuth --amplitude_scale_factor 5
recsec --pysep_path observed_data/AK_20250130_0_200/seismograms --syn_path synthetic_data/OUTPUT_FILES_ISO/seismograms --source synthetic_data/DATA/CMTSOLUTION --stations synthetic_data/DATA/STATIONS --scale_by normalize --xlim_s 0 100 --min_period_s 8 --max_period_s 16 --sort_by azimuth --amplitude_scale_factor 5
Since the amplitude units are not shown on the plot it was hard to say if the seismograms are getting scaled further or not. By additional normalization I was referring to the visual scaling that you see on comparing the two plots. I'm requesting the same visual scaling for the two plots by default, but I see the challenge. Also, the parameter y_axis_spacing cannot be used in this case of sorting by "absolute" quantities. So something needs to be provided to control the visual scaling in the sort_by "absolute" quantity case.
Yeah that would be difficult, because the y-axes are completely different in these two cases. Would amplitude_scale_factor work to scale the abs_azimuth case to something closer to the azimuth?
I just tested the parameter amplitude_scale_factor and it seems to do the job. So the issue was not a bug but rather a query on the behavior which I think is clear to me now. I have updated the issue title to reflect that. You might want to update the label and close the issue.