Sound pressure radiates spherically so it must cover the surface of a sphere (formula 4PiR2) so it goes down with R2 (radius squared). So not at all surprising.
Published Sensitivity Ratings, meant to be comparable model to model, are taken 1 meter away from face of speaker.
Formula: 1 watt of signal input/_____ db? sound output measured 1 meter away.
If curious you might adjust for 84/108 as you had it, then measure 1 meter away and see what you get
Answer won't be scientifically publishable, but it will be fun and informative.
Due to the noise setting one’s Spl (A+Weighted) delivers greater range. That’d conclude my personal belief. Strolling railroad "tracks" are not always intended for any specific individual during any given Tuesday, and engineering design flaws are indeed inevitable.
24 dB of difference under the conditions described seems reasonable to me.
I’m not going to make a comprehensive analysis of all the factors that may be in play, and will be making some simplifying assumptions, but I think this covers the dominant factors:
Sound pressure level falls off by 6 decibels for every doubling of distance under anechoic conditions.
If "right at the speakers" is about 2.5 inches from the drivers, then 7 feet (84 inches) is about five doublings of distance way. 6 dB per doubling x 5 doublings = 30 dB, which is 6 dB more than the measured difference of 24 dB.
The room is not anechoic, so we would expect the in-room reflections to be making a greater and greater relative contribution as the microphone distance increases. The relative amount of contribution the room’s reflections make increases with distance because the direct sound is falling off by 6 dB per doubling of distance while the SPL of the reflections remains approximately the same throughout the room. The distance at which the direct and reflected sound have the same SPL varies with the specifics, but it’s often around 4 or 5 feet from the speakers.
So we would expect the measured SPL to be higher at 7 feet distance than anechoic theory would predict. In this case the measured SPL is 6 dB higher than anechoic theory predicts, and imo a 6 dB contribution from the room’s reflections at this distance is arguably in the ballpark.