Bill Chaplin is an excellent communicator, as his students at University of Birmingham (United Kingdom), where he teaches Solar and Stellar Physics, would certainly attest. In recent years, this data analysis expert has worked on BiSON (Birmingham Solar Oscillations Network), a network of six observatories monitoring low-degree oscillation modes in the Sun. Not long ago, he made the transition from the Sun to other stars and is currently leading a task group working on solar-type stars for the Kepler mission. In this interview he recalls the origins of Helioseismology and clarifies some of the great current questions in this and its sister discipline, Asteroseismology.
Could you tell us how Helioseismology started? How do the latest observations from Kepler compare to the very first helioseismic observations?
The story of Helioseismology dates all the way back to the late 1950s when Bob Leighton and colleagues at CalTech began a study of the "boiling", "bubbling" patterns of convection at the solar surface. What they found led to the birth of a new field of Astronomy, the discovery that patches of the solar surface were gently breathing in and out (i.e., oscillating) at a period of around 5 minutes. This first phase of observational discovery for Helioseismology was completed in the mid-to-late 1970s by a team of physicists from Birmingham and the IAC, here in Tenerife, who showed that the gentle pulsations were truly global in nature. The whole of the Sun was oscillating due to sound trapped throughout its interior.
The first data we got from Kepler carry many advantages over those first ground-based observations of the Sun, the most obvious being that the Kepler data are almost continuous (with no breaks due to the day-night cycle), and stretch over periods lasting one month and longer. We also have the tremendous benefits of over 30 years of accumulated knowledge, from Helio- and Asteroseismology, on how to approach analysing the data.
What is the benefit of looking at many stars simultaneously, as Kepler does, rather than focusing on individual stars?
The biggest leap forward that Kepler has so far given is the ability to collect exquisite data on over 500 solar-type stars. Before Kepler, we had oscillations data on only about 25 solar-type stars. This fantastic leap forward allows us to do proper studies of trends, on a sample of stars that is big enough for us to take statistical effects properly in account. We also have the possibility to follow the "Sun in time". We can pick out a sequence of stars having the same mass as the Sun, but different ages, opening direct windows on the past and the future of our own star.
We will also be devoting special attention to the very best Kepler stars. Somewhere between 50 and 100 solar-type stars will have their oscillations monitored continuously throughout the entire mission, giving datasets of a length we have so far only been able to collect on the Sun.
Theoretically speaking, low-mass stars, substellar objects, brown dwarfs and gaseous planets are defined by their ability to produce energy through nuclear reactions. Might it be possible to identify them by means of asteroseismic observations?
There are several important aspects to determining whether or not we can detect oscillations. The intrinsic properties and internal characteristics of a body will fix the strength of its oscillations (e.g., the fractional size of the changes in brightness that result from the oscillations). The distance of the body from us fixes its apparent brightness: the further away the object, the less light we receive from it, and the harder it is to detect oscillations of a given strength (the data are then "noisier"). That problem is compounded for bodies that are intrinsically very faint (like the ones above). It is a combination of these factors that makes it very hard to detect oscillations in these bodies. For example, there have been long-standing attempts to detect oscillations in Jupiter, and it is only recently that those attempts may have met with success.
Can seismological oscillations be detected in all solar-type stars?
Our ability to detect the oscillations depends to some extent on how far away a star is (see above). If an object is close enough, it then comes down to the combination of intrinsic properties. I would be happy to accept the hypothesis that all stars pulsate at some level (i.e., that they are a ubiquitous characteristic of stars), but there may be interesting physics limiting the strength of the oscillations in some stars, thereby making detection very hard. Some of our new results from Kepler show that strong magnetic fields (which produce the likes of sunspots on our own Sun) are doing just that!
Are there any solar-type stars that don't act as such? For instance, stars with an unexpected radius and mass?
I'm going to answer this question from what might seem like a slightly unusual angle: even little differences are exciting! Differences do not need to be dramatic to be significant. Science would be very boring if all our theories proved correct right from the start (in fact that would be a very unrealistic world...), and so we thrive on challenging theory with new observations. The Kepler data are good enough to allow us to test physics under the exotic conditions found in stellar interiors, and we know already that several aspects of theory need to be improved to explain the small differences we see for many stars. New, better data drive improvements to our theories, as freshly revealed features of the data place new, more stringent demands on our understanding.