Mercury's Molten Core

Mercury, the solar system’s smallest planet, had long been thought to have cooled and solidified ages ago. So scientists were astounded when, in consecutive flybys by JPL’s Mariner 10 in 1974 and 1975, the planet gave inklings of a magnetic field, albeit a weak one—about 1 percent that of Earth. (Then again, Mercury is only 5 percent the mass of Earth.) This suggested the possibility of a molten core, but various measurements and models yielded an array of possible internal configurations with no conclusive evidence for fluid inside the planet—until now, says Cornell astronomy professor Jean-Luc Margot, lead author of a report published in the May 4 issue of the journal Science. The report shows that Mercury does indeed have a liquid center, although how big will only be determined by further observations.

The idea to examine the state of Mercury’s core began brewing during Margot’s O. K. Earl postdoctoral fellowship at Caltech, from 2001 to 2002, which, says Margot, “came with the freedom to investigate the science problems that I found interesting.” Among them was a hypothesis—posed by physics professor emeritus Stan Peale of UC Santa Barbara, a coauthor on the paper—that the nature and extent of Mercury’s core could be determined via observations from afar. Because Mercury is the closest planet to the sun, its surface temperature is too toasty for the spacecraft of today, so the scientists turned to radar astronomy. Margot began the work at Caltech by designing a way to test Peale’s idea and by gathering preliminary data, and continued it at Cornell.

They applied a technique—derived from ideas first set forth in the 1960s and revived recently by coauthor Igor
Holin of the Space Research Institute in Moscow—called the “speckle displacement effect,” using JPL’s 70-meter antenna at Goldstone, California; the Arecibo Observatory in Puerto Rico; and the Robert C. Byrd Green Bank Telescope in West Virginia. From 2002 to 2006, 21 different radar signals were beamed to the planet from Goldstone or Arecibo, and their echoes were received by two of the three antennas each time. Each echo had a unique speckled pattern, reflecting the planet’s surface roughness, which swept across each receiver in turn like spots of light from a rotating disco ball, allowing Mercury’s spin rate to be determined to within one part in 100,000. To make the measurements, the planet and the receiving antennas had to line up in a configuration that lasts only 20 seconds on any given day. “Everything had to happen within that 20-second time window,” Margot says.

The team found tiny variations in Mercury’s spin rate that could only be explained by the sun’s gravitational influence on a planet that is part liquid. “We have a 95 percent confidence level in this conclusion,” Margot says. The variations, called longitudinal librations, arise as the sun’s gravity exerts varying torques on the planet’s slightly asymmetrical shape. In addition to measuring Mercury’s spin rate, the authors also made a vastly improved measurement of the alignment of the planet’s axis of rotation, showing that Mercury’s spin axis is almost perpendicular to the plane of its rotation around the sun.

Goldstone observations were enabled by coauthors Raymond Jurgens, senior JPL engineer, and Martin Slade, head of the Goldstone Solar System Radar and JPL’s Planetary Radar Group Supervisor. —EN