An Interview with Sean Solomon

Photo: Lamont-Doherty Earth Observatory

Sean C. Solomon, the Director of Columbia University’s Lamont-Doherty Earth Observatory since July 2012, has been on the science teams for the Magellan, Mars Global Surveyor, and GRAIL missions. He served as Director of the Department of Terrestrial Magnetism at the Carnegie Institution of Washington for 19 years. He is a member of the National Academy of Sciences and a past president of the American Geophysical Union. As the MESSENGER Principal Investigator, he had overall responsibility for all aspects of that mission.[1]  He has just co-edited a new book entitled Mercury:  The View after MESSENGER, published by Cambridge University Press and available at

I had the chance to interview him this April in his LDEO office.  What follows has been edited for clarity and brevity.

Eyepiece:  Your book on MESSENGER’s discoveries has just been published.  How did the MESSENGER mission first come about?

SS:  Mercury had previously been visited by only one spacecraft , Mariner 10, which flew past the planet in 1974 and 1975.  And then after that, there were no known trajectories to deliver a spacecraft to an orbit around Mercury using existing propulsion systems.  That changed in 1985, when a JPL mission design expert, Chen-Wan Yen, discovered the multiple gravity-assist trajectories. Multiple flybys of Venus and Mercury could take enough energy out of a spacecraft’s orbit so that the relative positions and velocities of the spacecraft and the planet would be close enough at a later encounter for the propulsion system to be able to achieve orbital insertion.  Unfortunately, the Challenger disaster was one year later, and no new launches of spacecraft took place until Galileo in 1989.  In fact, Galileo ended an 11-year period with no planetary mission launches, so there was no possibility to consider a Mercury mission.  What then enabled NASA to undertake a mission was a new program for smaller, non-flagship missions, the Discovery Program.  In 1994, ’96 and ’98, there were a total of five missions to Mercury proposed under this program.  The first MESSENGER mission concept was judged too high-risk, so we did additional development and testing to retire the risk, and in the intervening time progress was made in materials that could withstand the solar heating and radiation, so that the project was finally approved.  Even so, we were really pushing the edges of the Discovery Program. When we launched, the spacecraft was 55% propellant by mass!  We needed all that propellant to do the flyby maneuvers and the orbital insertion.

Eyepiece:  And you eventually ran out of propellant.

SS:  That was understood, we were quite accurate in being able to forecast the propellant costs of each of our maneuvers.  We actually saved some propellant along the way, when our navigation and control team learned how to apply solar sailing. That happened just before the first of three Mercury flybys when we discovered we could do this by adjusting the position of the spacecraft and solar arrays.  The spacecraft wasn’t designed to do solar sailing, but we were close enough to the Sun so that solar radiation was a big effect.  During the two Venus flybys, we used propellant to make small course corrections. But after we discovered solar sailing, we didn’t use propellant at all for corrective maneuvers.

Eyepiece:  So what were the major findings of MESSENGER?

SS:  There were many, but let me list a few.  Of course, we saw more than half the planet at close range for the first time.  But I would list the top three discoveries of MESSENGER as the following:

First, we did remote chemical sensing of surface materials and discovered that contrary to all theories of how Mercury was assembled in a manner that led to a high fraction of metal to rock, the planet was rich in volatile elements, meaning elements and compounds that should have been removed by some period of high-temperature processes.  And we did know that high-temperature processes had been invoked for Mercury’s high metal fraction:  either condensation only of certain materials that are close to the parent star in the solar nebula,  or vaporization of the outer silicate shell of a differentiated planet in an extraordinarily hot solar nebula,  or a giant impact of an object nearly the same size as Mercury that collides with Mercury and ejects most of the rocky fraction of a differentiated protoplanet into trajectories that don’t re-accrete.  Those were the leading ideas for how to make Mercury metal-rich, but all three predicted that Mercury should be deficient in volatiles.  We sent several instruments for chemical analysis, and we measured the surface concentrations of such elements as sulfur, potassium, sodium and chlorine, which were not only present on Mercury but more abundant than on the Earth, and as abundant as on Mars, which is often regarded as the most volatile-rich of the inner planets.  So that threw out all those theories and forced everybody back to the drawing board for how to make the inner planets and particularly how to make one with Mercury’s high metal fraction.  I don’t think we yet have an answer.

We also showed that Mercury is extraordinarily chemically reduced, that is to say, the activity of oxygen is very low, and the compounds which are stable under those conditions are different from those that characterized Mars, Venus and the Earth.  So Mercury had to be assembled from materials that came from a different part of the solar nebula.  Lots of new ideas are now being discussed for how to make a Mercury, and some of the studies of extrasolar planets are finding objects with Mercury-like densities…but I would say there is no favored numerical model that captures in a physical-chemical simulation all that we know about Mercury yet.  The jury is still out.

Number 2:  Mariner 10 carried a magnetometer and discovered that Mercury had an intrinsic  magnetic field.  And that was a bit of a surprise.  At that point we knew that neither Venus nor Mars had a global magnetic field.  And the expectation was that the smaller the planet, the easier it was for it to lose internal heat, and the core would solidify.  An Earth-like mechanism for a magnetic field involves convective motions in a fluid around the core as a necessary condition for a global magnetic field.   So it was a surprise that Mercury, with only 5% of the Earth’s mass, today has a magnetic field.  It was merely a detection by Mariner 10, and what was needed was to understand what created it.  All of the ideas for what might make the magnetic field made predictions for what the geometry of the field would be today.  Those ideas could be distinguished by putting on board a magnetometer, so we did that, and got an answer that no one had predicted. Mercury has an Earth-like field that’s dominantly dipolar, with the dipole in line with its spin axis, which is the case on the Earth if you average over thousands of years.  But the difference on Mercury is that the magnetic field is offset by about 20% of its radius.  None of our dynamo models had predicted the offset There are some ideas that are coming out now, but they all invoke some cause for the asymmetry that we still don’t understand.

Number 3:  In 1992 a pair of radar studies found that Mercury had deposits in its polar regions which were highly reflective to radar and would de-polarize a polarized signal.  That combination was not seen elsewhere in the solar system other than on icy surfaces, and it was seen only in the polar regions of Mercury, so  it was hypothesized that the deposits consist of ice which is stable because it is in permanent shadow within polar impact craters.  The situation is like that on the Moon except that the polar deposits on Mercury are much more extensive. The first thing we did was to show that all the deposits were in regions of permanent shadow and that not all of these regions had ice deposits.  We also took an instrument to look for hydrogen: the neutron spectrometer, which showed that in the north polar region there was an excess of hydrogen that was just what one would predict if all these deposits were H2O.  But the flux of the most energetic neutron did not match the predictions if the water ice were present all the way to the surface, but rather required a model in which most of the polar deposits were covered to a depth of tens of centimeters by another material. What our models showed us was that the temperatures were such that water ice was stable at the surface for polar deposits near the pole, but as soon as you got a few degrees away from the pole, the surface temperature was too high for water ice to be stable over geological time periods. If the ice is buried below an insulating layer 20 cm thick, however, you can preserve water ice indefinitely.  What MESSENGER showed was that the polar deposits were very dark off the poles but very bright right near the poles.  And the brightness near the poles was consistent with water ice at the surface, but the darkness off the poles was darker by a factor of 2 than the darkest areas elsewhere on Mercury, so that meant it was not typical Mercury soil. The hypothesis is that this insulating layer is dark organic material, most likely delivered to Mercury by the same comets and other objects that also delivered the water ice.  What makes it even more interesting is that we think that this process may also have contributed to the water and organic inventory of the early Earth.

Eyepiece:  So what are your next projects?

SS:  Well, the Lamont-Doherty Earth Observatory takes up a good deal of my time.  We have some wonderful scientists, and they don’t need a director telling them what to do.  But I spend a good deal of my time recruiting and managing.  I’m not currently involved heavily in another spacecraft mission, but I’m peripherally involved in some mission concepts that might be in competition for selection a few years down the road.

Eyepiece:  Can you tell us a bit about your plans for LDEO?

SS:  We have a well-developed strategic plan which includes five scientific strategic initiatives.  One of them is called Changing Ice, Changing Coastlines, which is the study of how the evolving polar regions are losing ice to the oceans and how vulnerable the ice sheets are to acceleration of those losses in a warming climate.  We’ve got an initiative in Extreme Weather and Climate, including draughts and severe precipitation events and changes to precipitation patterns driven by climate change.  We have an initiative called Anticipating Earthquakes. We’re in a good position to ask whether we can understand the earthquake cycle to a better degree than we do now.  There’s an initiative called Real-Time Earth, which takes advantage of two revolutions, one in platforms and sensors which can go where we couldn’t otherwise go and take measurements with much higher resolution, and the other in data modeling, looking at cross-disciplinary ways to manage and interrogate the ever-growing new data sets.  Climate is a recurring theme which runs through a lot of what we do (

Eyepiece:  Well thank you so much for your time.  LDEO is indeed a wonderful place.

SS:  I agree.