By ADELINA PASTOR
“We want to detect planets in the habitable zone of a star like the Sun”
“Kepler has found quite a few planets with radii between 1 and 2 Earth radii, situated in the habitable zones of their stars”
“We are going to be able to see the formation and evolution of planets in a routine way”
Coincidences occur, in science to, and Phil Gregory knows all about this. During a visit to Cambridge in a sabbatical year this radioastronomer became interested in statistics. “On a sabbatical in the early 80’s I was visiting Cambridge and started a collaboration with Steve Go. He used a very significant programme which he and his students had developed over a period of 10 years. I did not comprehend all the details, but part of it was based on Bayesian inference. I decided that I needed to retrain myself and one of the areas I needed was advanced statistical analysis. One day I walked into a washroom and someone had left a book there by Larry Bretthorst. It was perfect timing for me. There was no name on the cover so I took it home for the weekend and it was like reading a thriller, I just could not put it down. As a physicist I am very drawn to unifying principles and the Bayesian inference which I learned from the book gives that unifying
approach which allowed me to optimize the answer to any of the scientific problems that interested me. Si it was clear that this was the way to go. Emeritus Professor at the University of British Columbia, still dedicated to research, but now retired from teaching except on special occasions such as this XXVI Winter School of the Astrophysics Institute of the Canaries on “Bayesian Astrophysics”, Gregory is also the author of a textbook on Bayesian analysis, which he wrote in a collaboration with his students and was published in 2005.
Question: Do you remember the title of the book you found?
Answer: Bayesian Spectrum Analysis and Parameter Estimation, by Larry Bretthorst. I tried to find out who had put it there, and discovered that a colleague had been with Bretthorst’s supervisor, Ed James, at a conference, and that he had given out copies of the book. That year I was in charge of the colloquia, so I invited him, and spent four days with him (laughs). He invited me to his next meeting, and there I met Tom Loredo (also a lecturer at this XXVI Winter School), with whom I have collaborated ever since. I have been able to apply Bayesian statistics to all sorts of interesting projects which I could not have resolved using the traditional approach.
Q: What kind of problems?
A: One of the most interesting, a genuine intellectual tour de force, on which I worked together with Tom, was how to disentangle periodic signals which are hidden in noise, but whose you are not sure of. This is a problem in any field of science, not only in astronomy. Is there an optimum method to look for those signals, whose nature you don´t know, but which are periodically repeating. We were able to find a solution in 1992, and we published a number of articles about this until around 2000, when I began to become interested in extrasolar planets, which has been my field for the past decade.
Q: We know some 1,800 extrasolar planets. What are the current challenges in looking for them?
A: As well as the 1,800 confirmed planets, o those of which we are very sure “Kepler” has identified some 3,000 further candidates. Many of them are really far away so that it is difficult to confirm them by independent observations which use other techniques, so that the plans are for another mission to focus specifically on more nearby stars, so that we can study them in greater detail. The planets detected have either been found by Kepler, or by using measurements of radial velocity, a technique in which we measure the velocity of the star, and look for small changes which indicate the presence of planets. Up to now we have been able to measure velocities similar to that of a baby wheeled in its pram, around 1 metre per second, but there are much more ambitious plans to obtain a great increase in precision. The European project “Espresso” and the US project “Express” aim at measuring velocities down to around 10 cm per second. Present day accuracy lets us detect planets of about the mass of the Earth but what we really want to do is to detect these in the habitable zone of a star similar to the Sun. This is what we hope to be able to do with the next generation of spectrometers, which are now being developed. In practical terms the main problem is due to the activity of the star itself. Starspots (equivalent to sunspots) and active regions can, under certain circumstances, produce a signal similar to that produced by a planet. Thus in the area of data analysis the challenge is to find a way to distinguish between these two types of signals. In fact we have an international competition under way following a previous meeting in Portugal, and the results will probably be presented next July at the University of York.
Q: And what is the prize?
A: Probably a bottle of Port. We would prefer it to be a million dollars, but unless Google or some similar company gets interested… (laughs again).
Q: Going back to exoplanets, the main challenge is to find planets similar to the Earth. How many have we found so far?
A: “Kepler” has found quite a lot of planets with a radius between 1 and 2 Earth radii which are in the habitable zones of their stars. In many cases we even know their masses. Statistical analyses of these data show that around 22% of the stars in our Galaxy similar to the Sun have planets similar to the Earth, with an uncertainty of 8%. So we expect to find a large number of them in the new projects. Kepler was the first mission to find these roughly Earth sized planets, which are just on the edge of our present measurements using radial velocities, but around smaller stars, not those of the size of the Sun. So we are clearly going to find a very large number of earth-like planets (billions of them). We have no reason to suspect that we are unique. A very recent discovery, using ALMA ( the large radiotelescope situated in Chile), is that around a very young star “HLTau”, which is less than one million years old, is a rotating disc of gas and dust, which is where planets are formed. The astronomers found that the disc is made up of rings, and the dynamics suggests strongly that the gaps between them are caused by newly formed planets. There are seven rings, corresponding to seven planets. We are going to be able to do this routinely in the future, seeing planets in formation, and evolving.
Q: But the big question is still whether we will find life, and whether it will be intelligent.
A: We will have to look a very large number of planets. Earth is four and a half billion years old. If we present this scaled down, as a time span of one year, then the time during which intelligent life has existed would be scaled down to 30 seconds. Basing our thinking on this, it is much more likely that we will discover a planet with dinosaurs, which existed on Earth for hundreds of millions of years, than intelligent life as we understand it. And this is for planets with ages similar to that of the Earth, because there are both younger and older planets. Future missions will be able to analyse the atmospheres of planets to look for indicators of biological activity, so that we can centre our attention on those… that is if life has evolved on other planets similarly to the way it has evolved on Earth. We may be in for many surprises.
Organizing Committee: Andrés Asensio Ramos, Íñigo Arregui, Antonio Aparicio y Rafael Rebolo.
Secretary: Lourdes González.
Contacts: Andrés Asensio Ramos (IAC): aasensio [at] iac.es (aasensio[at]iac[dot]es) y 922605238 Íñigo Arregui (IAC): iarregui [at] iac.es (iarregui[at]iac[dot]es) y 922605465
Press: Carmen del Puerto: prensa [at] iac.es (prensa[at]iac[dot]es) y 922605208
Previous press release: http://www.iac.es/divulgacion.php?op1=16&id=897
Programme of the Winter School: http://www.iac.es/winterschool/2014/pages/about-the-school/timetable.php
Further information: http://www.iac.es/winterschool/2014/
