Multiple hydrophones per bout

[dthaler]

Is your feature request related to a problem? Please describe.
Soon we will have 3 hydrophone nodes (Orcasound Lab, DAS, Andrews Bay) within a radius of 1-2 miles, meaning that the same calls on one will generally be heard on the other two at the same time. It would be confusing to create 3 bouts with the same information.

Describe the solution you'd like
A bout should have a set of hydrophones, either by using a list of nodes, or by designating an area. I mention the latter just because that came up in shipnoise discussion of designating pre-defined regions. For example, we might define a region "Haro Strait" for the combination of the three feeds.

[dthaler]

11/22, the facebook thread reported hearing the whales on two hydrophones at the same time, at multiple times, e.g., at
21:34 they were heard on both Andrews Bay and Orcasound Lab. This highlights why we now need bouts that aren't limited to 1 hydrophone node each.

[scottveirs]

I'm going to beg to differ on this one, @dthaler . I don't find having multiple bouts in the same area of the ocean confusing; on the contrary, I think the differences between the geographically-adjacent bouts are fascinating and valuable.

For example, the exceptionally (best ever?) Bigg's bout at Orcasound Lab on 11/26 has a much higher signal-to-noise ration (SNR) then the nearly coincident Andrews Bay bout. Presumably both bouts should be associated with the Bigg's killer whales that transited the area that morning (and the bout tags begin to reflect that, but the relative signal strengths suggest the orcas were much closer to the Orcasound Lab binaural array than the Andrews Bay one).

Similarly, the start and end times of the bouts overlap for less than a minute (if we trust the bout UI time stamps) and therefore allow us to infer that the whales were northbound when they became vocal:

Start time | End time  | Location 
06:26:37  | 06:51:55  | Andrews Bay
06:51:20  | 07:04:03 | Orcasound Lab

Localization of signals or sequences that cross-correlate would further inform the movement of the sources, whether based on a single array, pairs of binaural arrays, the DAS cable, and/or combinations of them... I'll take a listen to the 11/22 event when I get a chance on sabbatical, but I'm pretty sure I'll make the same case even when bouts have substantial or total overlap and many signal sequences can be cross-correlated between network nodes.

It may be that what you are suggesting has more to do with an acoustic "occurrence," rather than a bout. Perhaps we should discuss the nomenclature we're using for acoustic events within Orcasound's data schemes vs the presence/absence information we're trying to capture in the (evolving) Acartia and salishsea.io apps and their underlying databases?

[dthaler]

Someone on the orcasound standup call (maybe Adrian?) suggested the notion of "linked bouts", to make it easy to navigate in the UI from one hydrophone bout to another nearby hydrophone bout. That idea sounded promising to me.