The Sound Book: The Science of the Sonic Wonders of the World (6 page)

Figure 1.2 Wormit water reservoir (using a very long exposure on the camera).

As James and I chatted, the acoustic immediately revealed itself: a rumble began building up and hung about us like a pervasive fog. Many very reverberant rooms are acoustically oppressive, making it hard to have a conversation. But not this reservoir.
⁴⁶
Surprisingly, we could talk to each other even when we were quite far apart—something that was not possible in the similarly reverberant Hamilton Mausoleum.
⁴⁷
It reminded me of a cathedral, with the great advantage that I could shout and clap. Whooping unleashed the full power of the “preposterous” acoustic; the sound rattled around for ages before dying away.

I had a few balloons with me, which I burst to get a rough measurement of the reverberation time. As in mausoleums, the most impressive values were at low frequency: 23.7 seconds at 125 hertz. For the midfrequencies that are most important for speech, the reverberation time was a more modest 10.5 seconds.

Saxophonist John Butcher made recordings in the Wormit reservoir as part of the
Resonant Spaces
tour. The
Wire
review of the album describes how he “attacks the spaces.”
⁴⁸
In Butcher's piece “Calls from a Rusty Cage,” it is often hard to discern the sound of a saxophone among the strange electronic whistles, breathy squeaks, and blasts, which sound like ship horns. Will Montgomery in
Wire
described how halfway through, Butcher “suddenly leaps into whirling circular breathing with a flamboyant glissando (which . . . recalls the opening to
Rhapsody in Blue
).”
⁴⁹
This is certainly one musical approach to such a reverberant place: accept the dissonant smog created by lingering notes, and play on.

Another approach is that taken by American trombonist and didgeridoo player Stuart Dempster in his album
Underground Overlays from the Cistern Chapel
. The chapel in question is the Dan Harpole Cistern in Fort Worden State Park, Washington State, the place Mike Caviezel described as crazy and disorienting. It looks very similar to Wormit, although it is circular rather than square. It was built to supply about 7.5 million liters (2 million US gallons) of emergency water for extinguishing fires. A few websites and books quote a 45-second sound decay. This means it takes about 3 seconds for a note to become half as loud, and musicians can achieve note separation only when they play incredibly slowly.
⁵⁰
Billboard
magazine described the recording by Stuart Dempster and his fellow musicians as creating an “intensely serene music in which the slightest changes seem cataclysmic, and gradual swells emerge as tidal waves.”
⁵¹
Writing in the
Times
, Debra Craine describes the music as having an “eerie, majestic calm that envelops you with hypnotic elation.”
⁵²
Notes played seconds apart form lush layers on top of each other, requiring a player to think about the interaction of notes played far apart; otherwise, intense dissonance is produced. Stuart Dempster commented, “Usually when you stop for a mistake, the mistake has the decency to stop too, but it doesn't [in the cistern;] it just sits there and laughs at you . . . You have to be a clever composer [or improviser] and incorporate all your errors into the piece.”
⁵³

I listened to the album, enjoying the meditative polyphony, but also listening for the ends of phrases, because after the musicians stopped playing, the sound would naturally ring around the cistern. From these parts of the music, the reverberation time can be estimated. For over a decade, colleagues and I have been developing ways of extracting reverberation time from speech and music. The idea is to make measurements in concert halls, railway stations, and hospitals while they are in use. Conventional reverberation measurements require loud sounds: gunshots or loudspeakers blaring out noise or slow glissandos. These are unpleasant to listen to and can damage hearing. Audience members also have an annoying habit of ruining the results by commenting on the noise—“Wow, that was loud”—as the decay is being measured. But the sound of an orchestra in a concert hall, or the speech of a teacher in a classroom—while imperfect for measurement—does include the room acoustic; the difficult part is finding a way to extract the effects of the room from the music or speech. One of the most exciting areas of research at the moment is the use of computer algorithms to extract information from audio. A well-known example is Shazam, an app that identifies music from a brief recording through a mobile phone's microphone. Other algorithms try to transcribe music automatically or identify the genre of unlabeled audio files.

Applying our algorithm to Stuart Dempster's recording gave an estimated reverberation time of 27 seconds over the low frequency ranges of the trombone and didgeridoo.
⁵⁴
This is a good indication that the American cistern beats the Scottish reservoir. But to be sure, I wanted a conventional impulse response. When creating a new auditorium, acoustic engineers work from graphs and tables of reverberation times and other parameters to check that the hall meets design specifications. However, these scientific charts and parameters mean little to architects, so acousticians are increasingly making audio facsimiles of a proposed auditorium and getting clients to listen. This
auralization
starts with a piece of music that has been recorded in a completely dead space, like an anechoic chamber (described in Chapter 7). In other words, it is the sound of the orchestra without any room. Acousticians then combine this music with a model of how sound will move in the future place. In the past, impulse responses came from scale models of the auditorium at one-tenth or one-fiftieth of the full size, but nowadays they are more often predicted by computers.

Auralization also works with impulse responses measured in real rooms, so it has also been encoded into artificial reverberation algorithms used by musicians and sound designers creating film and game soundtracks. In one of these reverberators I stumbled across a library of impulse responses that included three measured in the American cistern. At low frequency the Dan Harpole Cistern has the same reverberation time as the Wormit reservoir: 23.7 seconds. But at midfrequency the America cistern wins, with a reverberation time of 13.3 seconds. These reverberation times are longer than those found in even the biggest cathedrals in the world.

E
ntering the oil storage complex at Inchindown, near Invergordon, Scotland, felt like walking into a villain's secret lair in a James Bond movie. The 210-meter-long (230-yard) entrance tunnel was narrow, concrete lined, and not much taller than me. And as I walked up the gradual incline from the entrance in the hillside, the daylight steadily faded behind and my torch vainly attempted to illuminate the way. The concrete lining ended, the tunnel became bare rock, and an alcove on the left revealed the entrance to the number one oil tank. But this was not a door, because the only way to get through the 2.4-meter-thick (8-foot) concrete wall and into the gigantic storage tank is via one of the four oil pipes, each only 46 centimeters (18 inches) in diameter. This was no time to worry about claustrophobia, because at the other end of the pipes was hopefully the most reverberant space in the world.

I was in Inchindown nine months after experiencing Wormit, visiting tanks that had once held heavy, crude shipping oil. These reservoirs supplied the naval anchorage in the Cromarty Firth at the bottom of the hill. The tanks were constructed in great secrecy amid concerns about the strengthening of Germany's armed forces during the 1930s and the threat posed by long-range bombers, which is why the tanks were dug deep into the hillside. The vast complex took three years to complete. The whole depot held 144 million liters (38 million US gallons) of fuel—enough to fill up two and a half million diesel cars.

My guide was Allan Kilpatrick, an archaeological investigator for the Royal Commission on the Ancient and Historical Monuments of Scotland. Allan is incredibly passionate about the oil tanks, having learned about the secret tunnels as a local boy. With us were about eight other people, taking advantage of a rare opportunity to see the place, although some never got into the main storage tanks, because they found the entrance too claustrophobic.

I was about to enter one of the big tanks, designed to hold 25.5 million liters (about 7 million US gallons) of fuel. I lay down on a trolley, a narrow sheet of metal about 1.5 meters (5 feet) long, and was pushed into the pipe like a pizza being put into a deep oven. The entrance holes looked even smaller when I was waiting to be dispatched, and as I entered I could feel the walls of the pipe pushing into my shoulders, compressing and squeezing me. The helpers kept shoving, my hard hat fell off, and then I was in. It was an undignified landing as I stopped at an angle, with my feet on the floor of the storage tank and my torso still half in the pipe. I struggled upright with a helping hand from Allan—dressed like a climber and looking at home in the dark underground world. Soon afterward, my acoustic measurement gear was pushed through, all carefully selected to ensure it would fit the narrow pipework.

I now had a few moments to take in the tank. All I had was a front bike light, which was too weak to illuminate much of the vast, barrel-vaulted cavern. It was difficult to get a sense of scale. My initial guess that it was 9 meters (30 feet) wide was spot-on. But how high was the space? That was difficult to estimate in the dark. Allan later told me the ceiling was 13.5 meters (44 feet) high.

Much of the floor was covered in pools of water and oil residue. Boots and gloves lay festering in the foul brown liquid, discarded by workers who had the horrible job of cleaning up the oil tanks when they were decommissioned. Fortunately, there was a dry causeway down the middle of the tank following a spine of slightly higher flooring.

As I wandered down the center line, I sang a few notes that hung in the space and built upon each other. In the Baptistry of St. John in Pisa, Italy, there is a long tradition of guides harmonizing with themselves in the impressive reverberance. In the nineteenth century, author William Dean Howells wrote, “The man poured out in quick succession his musical wails, and then ceased, and a choir of heavenly echoes burst forth in response . . . They seemed a celestial compassion that stooped and soothed, and rose again in lofty and solemn acclaim, leaving us poor and penitent and humbled.”
⁵⁵
My singing in the oil tank was much less poetic, I'm afraid, and I contented myself with seeing how many notes I could get going simultaneously—the audio equivalent of a magician spinning plates. There was time to sing great long phrases as the sound seemed to go on forever, maybe half a minute before it died away. The reverberance dwarfed the sound of Wormit's water cistern.

I carried on walking and started to realize how long the tank was: more than twice the length of a soccer or football field, at about 240 meters (260 yards). Whooping brought this giant musical instrument to life. Never before had I heard such a rush of echoes and reverberation. I was like a toddler sitting at a piano for the first time, thrashing the ivories to see what sounds would come out. Reluctantly, after a few minutes I stopped playing with the acoustics and started preparing for my measurements. I put the instrumentation on old heating pipes (used to get the oil flowing), which were covered in a sticky black residue. I fumbled around in the glow of the bike light—tripods under my arms, cables wrapped around my neck, and expensive microphones delicately positioned between my teeth—in a desperate effort not to ruin the equipment.

Modern acoustic measurements are often carried out on laptops, which in theory should make the process easier. But my laptop developed a sense of comic timing: up popped a dialog box announcing that Windows was updating itself deep inside the hillside. I had to resort to plan B: to record gunshots onto a digital recorder.

Allan fired a pistol loaded with blanks about a third of the way into the storage tank, and I recorded the response picked up by the microphones about a third of the way from the far end. This is a standard technique used in concert hall acoustics; old black-and-white photographs show a gun being fired on the stage of the Royal Festival Hall in London when they tested the acoustics back in the 1950s. Although there are many modern measurement techniques using clever noises and chirps, firing a gun is still a respectable and effective technique.

But measuring such a reverberant space was not straightforward. If either I or Allan made a noise—for example, saying to the other something like, “OK, ready to measure”—we would have to wait a minute or more for the sound to die away before we could fire the gun. We also had to stand completely still and make no noise while the sound was decaying, because otherwise the measurement would be ruined. Since we were standing about a hundred yards apart in the pitch black, hand signals were out of the question. Allan suggested we signal by shining torches on the ceiling.

With communication sorted, Allan walked away from me into the gloom. I saw a dim light on the ceiling and responded likewise to show I was ready. The gun went off, and I felt a quick rush of adrenaline as I fumbled with the recorder. But the sound was far too loud, and my digital recorder overloaded. A simple adjustment and I was ready for the second shot, but then I realized I needed to tell Allan what was going on. As I trudged along the center line to explain what had happened, I made a mental note to bring walkie-talkies the next time.

The second shot was fired, and I listened through my headphones, waiting to turn the recorder off when the sound had disappeared. The recording time ticked up on the dial; 10 seconds, 20, 30, 40—still I could clearly hear the reverberation; 50, 60—this was getting ridiculous. After a minute and a half it was completely silent, and I turned off the recorder.

For the third gunshot, I took off my headphones to appreciate the sound. The familiar crack of the gun was followed by a wave of explosion that washed past me and bounced off the end wall, before returning and bathing me in reverberance from all directions. If the world ends with an apocalyptic thunder crack, this is what it will sound like, with the rumble lingering and forlornly dying away. I wanted to shout with astonishment, but I had to remain silent so as not to ruin the recording.