posted on Apr, 26 2015 @ 02:29 PM
Just chiming in because I've had cause to examine audio for unrelated reasons in the past. An easy trick to check whether something has a mono
component is to simply invert one of the channels and mix the result together. If you do this then you discover that only the bass frequencies of the
sound in question are cancelled, leaving the mids untouched, which is consistent with a slightly off-axis sound source.
The left channel hiss isn't too hard to account for. It's likely that this was recorded on a handheld device, and the placement of fingers near a
microphone can create quite pronounced resonant effects which boost high frequencies. An alternative is asymmetry in the construction of the recording
device, or even poorly matched or faulty left/right microphone preamps. There does seem to be some wind noise at lower frequencies in the right
channel, so the environmental sounds aren't strictly constrained to the left. Also, if you mute the left channel, there are still traces of the
high-frequency content, only attenuated when compared to the left.
The audio recording therefore doesn't seem to contain any real red flags.
As for the noise itself, several people have suggested a train and I'm inclined to agree that this is a strong possibility. The sound is noticeable
rhythmic at a frequency of around 6.3 Hz. Rail cars frequently have flat spots on their wheelsets caused by too much braking effort resulting in a
slide which grinds down the tyre. A wheelset rotating at 6.3 Hz, depending on the tyre diameter, corresponds to a speed of anywhere between 30 and 50
mph (using wheel diameters from
here as an example). This is a perfectly
reasonable range within which a freight train might travel.
My best guess is that the sound is that of a freight train crossing a bridge or other solid, resonant structure. This would explain the louder,
rhythmic sections (as a flatted wheelset crosses the bridge), as well as the quieter single knocks which could be caused by wheelsets crossing a rail
joint.
Apparently the sound lasted for 3-5 minutes. So we have a range of times from 3-5 minutes and a range of speeds 30-50mph. That gives a range of train
length between 1.5 and ~4 miles. Given that the longest train in the US is 3.6 miles, it doesn't seem such an unreasonable prospect.