Interestingly, the sound intensity will decay as a 1/r law, where r is the distance from the source. (I'm assuming conservation of sound energy as it travels horizontally, i.e., no loss to interactions with atmosphere and terrain).
Compare with the 1/r² laws like the strength of an electric field at a distance r from charge Q: E(r)=kQ/r².<p>In both cases there is a total of something (sound energy or electric field lines) and that total must be split over all possible directions. The total electric field must be split over the surface area of increasingly larger and larger spheres hence the 1/r² behaviour (1/4πr² to be precise). The sound energy is spread uniformly over a circle with circumference C=2πr, hence giving a 1/r decay, over short distances.<p>For longer distances, the curvature of the earth will play a role. Come to think of it, it must have been <i>really</i> loud somewhere diametrically opposite to Krakatoa, 17 hours after the eruption...