get_gage_data typo

[Iz-0]

I think there's a typo in the get_gage_q function in the Validation module (validation_confluence.py, line 217). If there is more than one index (which I'm guessing is rare if it happens), and there is a model_q value, you use the length of the time series to compare to this model_q, rather than the mean_q values (you compare it to glt rather than gmq). The line is:
index=np.array(index[np.argmin(np.abs(np.array(glt)-model_q))])
But I think it should be:
index=np.array(index[np.argmin(np.abs(np.array(gmq)-model_q))])

[cjgleason]

Hey Izzy- is this still true with the SVS?

[Iz-0]

Hi Colin - I think it is still the same (line 228 now) in the version I can see, unless there is another version that has been updated?

[efried130]

Hi all, thanks for bringing this back to attention. It looks like this is a problem with the SOS gauge pull (not the second SVS pull used for the Apr-May 2026 run) when multiple gauges are present for the same reach. A single gauge match will perform correctly but as Izzy has flagged, a multi-gauge match will be silently dropped.