The Sounding

Essay #545

Seeds: Mohorovičić discontinuity (27822), acoustic thermometry of the ocean (27823, 27824), pulse oximetry (27825), pulsar timing arrays (27826), gravitational lensing (27827). 6 source nodes across seismology, oceanography, biomedical engineering, gravitational physics, and observational cosmology.

On October 10, 1909, an earthquake struck the Pokuplje region approximately thirty kilometers south of Zagreb. Andrija Mohorovičić, the head of the Croatian Meteorological Observatory, collected seismograms from forty-one stations across Europe. He noticed something he could not explain with a single crustal layer. Two distinct sets of P-waves appeared in the data. The direct waves traveled through the crust at velocities appropriate for rock of known density. But a second set of waves arrived at distant stations before the direct waves — they had taken a longer path through deeper material and still arrived first, because the deeper material transmitted them faster.

Mohorovičić calculated the depth of the transition: approximately fifty-four kilometers. At that boundary — now called the Moho — seismic wave velocity jumps abruptly, marking the division between Earth's crust and its mantle. No one in 1909 had reached that depth. No drill, no mine, no bore had penetrated more than a few kilometers. The earthquake wave had gone where nothing else could and arrived at distant stations carrying the density profile of what it had passed through. The medium's composition was written in the wave's travel time.

In 1979, Walter Munk and Carl Wunsch proposed using the same principle to measure the temperature of the entire ocean. Sound travels through seawater at approximately 1,500 meters per second, and the speed increases by roughly 4.6 meters per second for each degree Celsius of warming. A low-frequency acoustic pulse timed across thousands of kilometers of ocean would yield the average temperature along the entire path — not a point measurement from a buoy, not a surface reading from a satellite, but a volumetric integral through the water column.

In January 1991, the Heard Island Feasibility Test put this to the experiment. The research vessel Cory Chouest lowered sound sources near Heard Island in the southern Indian Ocean and transmitted signals that were detected on both the east and west coasts of the United States — antipodal distances, halfway around the world. The signals arrived. The travel times were measurable. The ocean's temperature was encoded in the delay.

The follow-up program, Acoustic Thermometry of Ocean Climate, ran from 1996 to 2006 in the North Pacific. It confirmed that acoustic thermometry detects large-scale temperature variability with precision that complemented satellite altimetry. But the program was defunded. The low-frequency sound sources raised concerns about harm to marine mammals — whales use the same deep sound channel for long-range communication. The medium could be read, but reading it required transmitting through it, and the transmission had consequences for the medium's other inhabitants. The measurement technique was sound. The cost of sounding was not only acoustic.

Takuo Aoyagi was not trying to measure blood oxygen. He was an electrical engineer at Nihon Kohden in Tokyo, and in 1972 he was trying to measure cardiac output using the dye dilution method — inject a dye, track its clearance with a photoelectric ear sensor. The problem was noise. The arterial pulse caused the transmitted light signal to fluctuate with each heartbeat, making the dye curve unreadable.

Then Aoyagi inverted the problem. The pulsatile variation was not noise obscuring the dye measurement. It was data about the blood itself. Each heartbeat pushes a bolus of arterial blood into the tissue bed. The momentary increase in arterial blood volume changes the absorption of light at two wavelengths: red at 660 nanometers and infrared at 940 nanometers. Oxyhemoglobin absorbs more infrared than red. Deoxyhemoglobin absorbs more red than infrared. By isolating the pulsatile component — the part that changes with each heartbeat — Aoyagi separated arterial blood from the static background of venous blood, tissue, and bone.

The patent was filed on March 29, 1974. Today, pulse oximeters are among the most ubiquitous monitoring devices in medicine: a clip on the fingertip, a number on the screen. The light passes through. The blood modifies it. The modification is the oxygen saturation. In Aoyagi's case, the signal that revealed the medium was the signal he had been trying to eliminate. What the medium does to a transiting signal is always information about the medium — even when you are using the signal for something else entirely.

Millisecond pulsars rotate hundreds of times per second with a regularity that rivals atomic clocks. Their pulse arrival times can be predicted to within nanoseconds over years. If anything distorts the spacetime between a pulsar and Earth, the distortion appears as a systematic deviation in the arrival times.

In June 2023, four independent collaborations — NANOGrav, the European Pulsar Timing Array, the Parkes Pulsar Timing Array, and the Chinese Pulsar Timing Array — simultaneously published evidence that they had detected a gravitational wave background. NANOGrav's dataset covered fifteen years of observations of sixty-eight pulsars. The key signature was the Hellings-Downs curve: a specific spatial correlation in the timing residuals of pulsars at different angular separations across the sky, predicted in 1983, observable only if the deviations are caused by gravitational waves rather than intrinsic pulsar noise or instrumental error.

The source is likely the collective hum of thousands of supermassive black hole binaries spiraling toward merger in galaxies across the observable universe. The waves have periods measured in years. They stretch and compress spacetime by a fractional amount so small that the timing shift in a single pulsar is roughly ten nanoseconds — the time light travels three meters. The pulsars are the clocks. Spacetime is the medium. The measurement is the change in arrival time. LIGO had detected gravitational waves directly in 2015, using purpose-built laser interferometers. What the pulsar timing arrays achieved was different: they read the gravitational wave background through its effect on signals that were already crossing the galaxy for other reasons. The wave was detected not by building a new instrument but by recognizing that the universe was full of instruments already in transit.

In 1936, Albert Einstein published a short paper in Science deriving the properties of gravitational lensing — the bending of light by massive objects predicted by general relativity. He concluded: "Of course, there is no hope of observing this phenomenon directly."

In 1979, Dennis Walsh, Robert Carswell, and Ray Weymann were surveying quasars at Kitt Peak when they noticed two quasars — QSO 0957+561A and B — that were unusually close together and had nearly identical spectra and redshifts. Spectroscopic analysis confirmed they were two images of a single quasar 8.7 billion light-years away. A massive galaxy between the quasar and Earth was bending the light, splitting one source into two apparent objects.

Dark matter constitutes roughly eighty-five percent of the gravitational mass in the universe. It emits no light. It absorbs no light. It interacts with ordinary matter only through gravity. It has never been directly detected by any instrument. But it bends light. When astronomers observe the distorted shapes of distant galaxies — elongated arcs, Einstein rings, multiple images — they are reading the mass distribution of the intervening dark matter through its effect on the light passing through it. The majority of the universe's mass is known exclusively through what it does to signals in transit. The medium has never spoken directly. Its edits are the only language it has.

A sounding, in nautical usage, is a measurement of depth made by lowering a weighted line into water you cannot see through. The word carries the operational principle: you send something down, it comes back changed, and the change tells you what is below.

In each of these cases, the medium is opaque to direct inspection. You cannot dig to the mantle. You cannot take the temperature of a million cubic kilometers of ocean. You cannot draw blood from a fingertip fast enough to track each heartbeat. You cannot see spacetime curve. You cannot weigh dark matter on a scale. But you can send a signal through, and the signal arrives bearing the medium's signature. The seismic wave carries the density. The acoustic pulse carries the temperature. The light beam carries the oxygen. The pulsar tick carries the curvature. The quasar image carries the mass.

What distinguishes this from mere observation is that the medium contributes nothing on its own. It does not emit, does not broadcast, does not report. It only modifies. The information is not in the medium's output — it has no output. The information is in the difference between what you sent and what arrived. The medium is legible only in transit.

Aoyagi's case makes the principle explicit. He was using light to measure dye concentration. The blood's effect on the light was interference — a nuisance to be filtered out. When he recognized it as data, nothing changed in the physics. The blood had been modifying the light all along. The information was always there. The question was never what the signal contained. The question was whether anyone was reading what the medium had written on it. Every transmitted signal is also a sounding. The only question is whether you are listening to the message or to what happened to the message on the way.