A Project of Constant Improvement

Related Resources • Length-of-Day Studies • Techniques/Instruments    Elsewhere on the Web • GWR Gravimeters • Global Geodynamics Project

An important part of science is studying the constants—because in reality almost nothing is constant. The length of the day, for instance, is different every day, if only by a millisecond or so. The Sun's light output, often called the solar constant, is constant enough for a cat on the windowsill to rely on, but it actually varies over the years by a percent or so. These small variations in what we assume to be unvarying things are information about something new and unknown.

Earth's gravity is one of these constants. For all practical purposes, gravity is the same everywhere on Earth's surface, but for scientific purposes surface gravity is full of subtle variation—that is, information. Over the last 200 years, we've uncovered layers upon layers of information from gravity readings. And today, a new generation of gravity meters is taking the quest to the next level.

The gold standard in gravity instruments is the absolute gravimeter, an apparatus that basically repeats Isaac Newton's legendary observation of the falling apple: a prism is dropped in a vacuum chamber and its velocity is measured with a light beam. There are only a relative handful of these delicate and expensive machines in the world. They're monitored and calibrated and checked against each other constantly.

Those machines are in demand by lots of people besides Earth scientists, so geologists make do with cheaper spring gravimeters for most traditional gravimetry. These rely on the position of a weight on a spring in a carefully controlled chamber to determine gravity. The best instruments are so sensitive that if you lifted one just 3 centimeters, it could detect the change in gravity. That's good enough for mapping the deep structure of the Earth—because dense bodies of rock exert more gravitational force locally—or detecting slow rises and falls of the ground.

Both kinds of gravimeters are slow machines because they must make many repeated measurements to arrive at their precise averages. And the raw readings must be adjusted for various factors. Everything matters—for instance, the quality of the power supply, the accuracy of temperature control, the age of the electronics in the box, tilting of the ground it sits on, earth tides and ocean tides, the amount of rainfall and its effect on the local groundwater, the weight of the air above as it changes with the weather. Spring gravimeters even have "morning sickness" and "travel sickness."

A faster machine with cleaner data could detect interesting things that happen in minutes, like slow earthquakes, deep volcanic activity, and the natural ringing of the Earth. Such a machine, the GWR superconducting gravimeter, whose operation is particularly elegant. A sphere of niobium metal is levitated by magnetic fields in a vacuum chamber, with no springs and no falling objects. The levitating magnets are powered by a permanent electric current in a superconductor, the nearest thing we have to a perpetual-motion machine. Motion of the spheres is measured through the response of the electric current. The latest version uses two niobium spheres at once, for a data stream with double-checked accuracy. It's so sensitive that if you crawled underneath the machine, it would detect the gravitational attraction of your body.

In 1997 a 10-year program began, the Global Geodynamics Project, which placed these gravimeters at 17 sites in the first world network of its type, to see what can be seen. No doubt that many of the questions we will put to the gravity data will have a surprise answer. It's an ancient truism in science (one that goes back at least to Galileo) that every advance in instruments has opened a new world to us.