I compared the signals between the INF04s located in the NAOJ room and located outside of the Analysis building.

In the infrasound signal, there are large glitches due to human activities in the "in room" data, but not in the "outdoor" data.
The actual acoustic signal, our target,  is larger in the "outdoor" data.
(around 3:00 UTC -> dam draining?  c.f. [SEIS signal])

In the barometer signal, the glitch level is almost the same.

Device temperature was changed a lot in the "outdoor" data. However, its effect doesn't affect the barometer data.