Why You Hear What You Hear


Home | MAX patches | Software | Using Falstad applets | Recording Audio| Links

Supplements for Chapter 28

Sound Outdoors


Sound in the sea: http://www.dosits.org/


This website is well worth spending some time with. Although it doesn't immediately jump off the page, there are many gems.


Battle of Gaines Farm Dead Zone Simulation (Acoustic Shadow)

Using temperature gradient 2 in (Harvard-Falstad) Ripple, and moving the source near a ground level which has been put in using filled shapes, draw rectangle, draw walls, and a reciever at the right, it can be verified that almost no sound reaches the reciever.


Although all sound waves travel at the same speed through the air, low frequency sound, and especially infrasound, looses much less energy doing so than higher frequency, shorter wavelengths do. (Witness the booming street thumper-you still hear the lowest whomps when the offending car is half a mile down the road.) Eventually all that remains of a big explosion is the infrasound, traveling for hours and sometimes days on end through the atmosphere.

The explosion of Krakatoa in 1883 in the Dutch East Indies is still the largest explosion in recorded history. Just for starters, after breaking eardrums 20 miles away and windows 200 miles away, sound traveled 4 hours and more than 2500 miles across the Indian Ocean basin to Rodriguez Island off of Africa’s eastern coast, where it was heard loudly - there was plenty of audible sound. The infrasound component kept traveling around the earth at least 7 more days, or over five times around the earth. It was so "loud" (even though inaudible once only the infrasound remained) that barometers picked up the pulses, and that's how we know about it.

Explosions aren't the only source of infrasound. Ocean waves crashing on shore can generate infrasound too. Wind oscillating behind buildings or even mountains produce infrasound, as do thunderstorms, and earthquakes.

Elephants produce infrasound and use it for communication, as was discovered by Katy Payne in 1998 observing elephants in the Washington Park Zoo, Portland, Oregon. She thought she felt rumbles too low in frequency for humans to hear. An elephant’s call is audible to humans for about 800 meters, but the infrasonic component travels 10 kilometers or more where it can be picked up by another elephant (they can hear well below our threshold of 20 Hz.)

Infrasound, homing pigeons, and another shadow effect?

Homing pigeons may use ocean wave generated infrasound (sound too low in frequency for us to hear) to navigate.This interesting article on their navigation methods.

"Microbaroms", known also as the "voice of the sea," are caused by waves colliding in the open ocean, head-on collisions typically found at the edge of storms. The collisions cause the sea surface to rise and fall like a giant loudspeaker, launching infrasound, which can the be heard round the world (by sensitive infrasound microphones). Their frequency is about 0.14 Hz, or about 6 oscillations per second, and its "heard" for hundreds of miles if you are sensitive to it. As pointed out recently by Dr. Jonathon Hagstrum, homing pigeons hearing down to 0.05 Hz may be listening to the voice of the sea even hundreds of miles away, explaining their amazing homing ability (which must to some extent be learned).

Dr. Hagstrum has a really slick reason, too, why you don't release your pigeons from Jersey Hill, 74 miles from their loft at Cornell. They don't get back, flying in random directions from Jersey Hill, as from almost nowhere else. Dr. Hagstrum may have figured it out with a sophisticated computer ray tracing program that can follow the sound pathways. The paths carrying the ocean wave infrasound sound don't make it to Jersey Hill - the sound skips over the hill; Jersey Hill is in what is called a sound shadow. Birds released there can't hear their beacon sounds, and they head of in random directions, as years of data confirms.

Pool bottom effect:

Sound "twinkling" or scintillation due to refraction through pockets of moving air, or temperature fluctuations, and possibly humidity fluctuations.

Below is a sonogram (time on horizontal axis) of a jet passing some distance away (miles) on a fair day with no clouds in the sky, but solar heating (and no doubt rising air columns) in the late afternoon is present. Soundfile is here.

The vertical streaks are loudness scintillations - why are they vertical? Because, all frequencies travel at the same speed in and are refracted along the same ray paths, so that the sound is loud (when many rays arrive at the microphone from the jet) or soft (when fewer ray paths are arriving at the microphone) at all frequecies at the same time.

Noise near roads


This will come as a surprise to no one, but it is still interesting to see how closely noise levels track proximity to roads, as revealed in this detailed study of noise in part of the Dublin area. Maps like this were required by the European Union, and would be incredibly expensive to produce if actually measured down to such fine detail. In fact, due to variations usually beyond the control of those recording the sound levels, the computer models are more accurate in certain senses.

You can see the effect of trees and open fields (as compared to buildings) on highway traffic noise levels in the outlined area for example, that is reproduced in Google Earth below. According to their model, buildings help significantly in reducing noise propagation far from the highway. This makes sense: they reflect the sound waves partially back toward the source, and also partially upward due to diffraction and the trend of rays from road surface to higher elevations; these rays reflect off vertical walls of buildings and continue upward. We should not forget too that the siding of such buildings absorbs some of the sound. The result is that if you are going to live three blocks from a highway, it is better to have lots of buildings between your house or apartment and the road (and better to live on lower floors!) if road noise tends to bother you.

For further information, see this website.

An area near the center of the black outline in the noise map above, from Google Earth. Note that the louder sound (reddish) propagates further near open fields than it does with lots of buildings surrounding, look along the road colored blue.

Upwind vs. Downwind

Here we schematically illustrate the difference between shouting upwind vs. downwind. We did this with a Ripple simulation in Why You Hear What You Hear. The top panel illustrates shouting upwind. The hatched area that the listener is standing in is nearly free of sound in the upwind case. The effect results from the fact that waves travel perpendicular to their wavefronts - and those wavefronts are bent because of the retardation of their progress at higher altitude in the case of upwind propagation, and advancement of their progress at higher altitude in the case of downwind propagation.

The hatched "acoustic shadow" has no sound in it if we believe ray theory, but we know that ray theory does not predict that we can hear pretty well behind a solid wall, due to diffraction of waves. So this is why the sound level being so low in the acoustic shadow is surprising - what happend to the diffractive corrections?

The refraction happens in reverse downwind; reciprocity is definitely broken by a wind gradient (faster wind aloft). The sound is actually louder downwind than if there were no wind.

Acoustic Shadows

The erie silence in defiance of the raging battle of Gaines Farm , only a mile or two away, is remakable. We learn from the 1990 Ken Burns PBS Series "The Civil War" that several such occurences were noted by multiple withesses at different times and places during the Civil War. The sound must have been refracted upward, this can occur under wind speed gradients or temperature gradients with altitude. The lowest sound rays (in a ray tracing study) would be seen climbing above the heads of the listeners.

What is so remarkable is the complete attenuation of the sound in the acoustic shadow. Diffraction is not very important apparently - total silence in the face of that many guns and canon? The attenation may have been over 100 dB.

Here is the reason: Sharp objects like a brick wall edge correspond to sudden changes of impedance,and as we discuss in Why You Hear What You Hear they diffract about half a wavelength's worth of incident power - quite a bit.

But if the rays are slowly bent and a dark shadow region is created, the impedance changes have been gradual (although with significant net effect - the bending of the ray paths). The shadow region can be VERY dark.



Some readers might be interested in the subject of Archaeoacoustics. What sounds are implied by ancient objects, soundspaces, etc.? In this book we mention two sound phenomena that residents of Chichen Itza must have heard coming from the steps of the temple.

Some people maintain that ancient "accidental" recordings of human singing or conversations may exist, say through action of sound perturbing a stylus passing through wet mud on a pottery wheel. This is so unlikely that we hope too much focus on it doesn't sully the larger field of archaeoacoustics, that is an interesting one indeed. Perhaps most plausible accidental "recording" is mentioned in this chapter, namely the screeching sound of a chisel described by Galileo that left precise marks on a metal plate. If only the plate still existed... but it still wouldn't be a recording in the modern sense,

Stanford University, in connection with the CCRMA, supports the Chavín de Huántar Archaeological Acoustics Project.