All posts by Pale Red Dot

Opportunities and Obstacles for Life on Proxima b

by Prof. Rory Barnes, University of Washington

The discovery of Proxima b is the biggest exoplanet discovery since the discovery of exoplanets. The planet is not much bigger than Earth and resides in the “habitable zone” of the Sun’s nearest stellar neighbor. This planet may represent humanity’s best chance to search for life among the stars. But is Proxima b habitable? Is it inhabited? These questions are impossible to answer at this time because we know so little about the planet. However, we can extrapolate from the worlds of our Solar System, as well as employ theoretical models of galactic, stellar, and planetary evolution, to piece together realistic scenarios for Proxima b’s history. The possibilities are varied and depend on phenomena usually studied by scientists in fields that are considered distinct, but an integrated perspective — an astrobiological perspective — can provide a realistic assessment of the possibility that life could have arisen and survived on the closest exoplanet.

As an astrobiologist and astronomer at the University of Washington, and a member of NASA’s Virtual Planetary Lab, I have investigated the habitability of planets orbiting red dwarfs for years. My research involves building computer models that simulate how planetary interiors and atmospheres evolve, how stars change with time, and how planetary orbits vary. The discovery of Proxima b has me very excited, but being Earth-sized and in the habitable zone are just the first two requirements for a planet to support life, and the list of requirements is much longer for planets orbiting red dwarfs than for stars like our Sun. If Proxima b is in fact habitable, meaning it possesses liquid water or even inhabited, meaning life is currently present, then it will have traversed a very different evolutionary path than Earth. This difference is frustrating, in that it will make our initial interpretations challenging, but also exciting, as it offers the chance to learn how Earth-sized planets evolve in our universe. Whether Proxima b is a sterile wasteland or teeming with life, we are now embarking on an unprecedented era of discovery, one that may finally provide an answer that age-old question “Are we alone?”.

To evaluate the possibility of life on Proxima b, we must begin with the only habitable planet we know, Earth. Life on Earth has established itself in a stunning diversity of habitats, including acidic hot springs, the deepest reaches of the oceans, microscopic channels in sea ice, and the deepest levels of Earth’s crust.  Regardless of how extreme the environment, all life on Earth requires three basic ingredients: energy, nutrients and liquid water. The first two ingredients are very abundant throughout the universe, as is the water molecule. The limiting factor from an astrophysical perspective is that water must be in its liquid phase. The habitable zone is a map of where liquid water could exist on the surfaces of rocky, Earth-like planets, hence its status as the first requirement for a planet to be habitable. Life also requires sufficient time to originate and evolve, but on Earth it has proven resilient to calamities as trivial as a thunderstorm or as traumatic. The variety and tenacity of Earth-bound life encourages astrobiologists to imagine that life can exist not only on Earth-like exoplanets, but also on strange, exotic worlds.

So what to make of Proxima b? It is at least as massive as Earth, and may be several times more massive. Its “year” is just over 11 days and its orbit may be circular or significantly elongated. Its host star is only 12% as massive as our Sun, 0.1% as bright, and it is known to flare. It may be joined to the stars Alpha Centauri A and B, 15,000 astronomical units (AU) away, by their mutual gravitational attraction. All three stars contain substantially more heavy elements than our Sun, but we know very little of the composition of Proxima b, or how it formed. The new data point toward the presence of a second planet orbiting in the system with a period near 200 days, but its existence cannot be proven at this time. These are the facts we have and from them we must deduce whether Proxima b supports life.

Proxima b was detected via the radial velocity method, which does not provide a direct measurement of the planet’s mass, only a minimum mass. So, the first question we’d like to answer is whether the planet’s mass is low enough to be rocky like Earth.  If the planet is much larger, it may be more like Neptune with a thick gaseous envelope. While we don’t know where the dividing line between rocky and gaseous exoplanets is, models of planet formation and analyses of Kepler planets suggest the transition is between 5 and 10 times the mass of Earth. Only about 5% of allowed orbits place Proxima b’s mass above 5 Earth masses, so it is very likely that this planet is in the rocky range.

The next question to ask is if the planet actually formed with water. Water consists of hydrogen and oxygen, the first and third most common elements in the galaxy, so we should expect it to be everywhere. Close to stars, however, where Proxima b resides, water is heated into its vapor phase while planets are forming, and hence it is difficult for planets to capture it. Planets that form at larger distances can gather more water, so if Proxima b formed farther out and moved to its current orbit later, it is more likely to be water-rich. At this time, we don’t know how the planet formed, but three scenarios seem most probable: 1) the planet formed where it is from mostly local material; 2) the planet formed farther out while the gas and dust disk that birthed the planetary system still existed, and forces from that disk drove the planet in to its present orbit; or 3) the planet formed elsewhere and some sort of system-wide instability rearranged the planets and b ultimately arrived in its current orbit. The first method is how Earth and Venus formed, and so Proxima b may or may not possess significant water if it formed in this way. The second method produces planets that are very water-rich because water is more likely to be in its ice phase farther out in the disk and so the forming planet could easily gather it up. The third method is inconclusive as the planet could have come from an interior orbit and formed without water or farther out and be water-rich. We conclude that it is entirely possible that this planet has water, but we cannot be certain.

Next let us consider the clues from the stars themselves. Computer models of the evolution of our galaxy suggest that stars enriched in heavy elements like Proxima cannot form locally (25,000 light-years from the galactic center) as there just aren’t enough heavy elements available. But closer to the galactic center, where star formation has been more vigorous and transpiring for longer, stars like Proxima are possible. Recent work by Dr. Sarah Loebman and colleagues has found that stars in our local solar neighborhood with compositions like Proxima must have formed at least 10,000 light-years closer to the galactic center. It would seem Proxima Centauri has wandered through our galaxy and this history may have played an important role in the evolution of Proxima b.

Computer models of the evolution of the Milky Way galaxy suggest Proxima Centauri has moved outward at least 10,000 light-years from where it formed, shown by the orange circle. The Sun and Earth probably formed near where they orbit today (blue circle), which is where we find Proxima Centauri, too.
Computer models of the evolution of the Milky Way galaxy suggest Proxima Centauri has moved outward at least 10,000 light-years from where it formed, shown by the orange circle. The Sun and Earth probably formed near where they orbit today (blue circle), which is where we find Proxima Centauri, too.

The orbit of Proxima around Alpha Centauri A and B, assuming they are gravitationally connected, is large compared to other multiple star systems. In fact, it is so large that A and B’s hold on Proxima is weak and the effects of the Milky Way galaxy have shaped Proxima’s orbit significantly. The mass of the Milky Way as a whole causes Proxima’s orbit to vary both in shape and orientation continuously. Proxima is also susceptible to gravitational encounters from passing stars that can change its orbit. Recent simulations by Prof. Nate Kaib have found that these two effects can often lead to close passages between the stars in a multiple star system that disrupt their planetary systems. The disruption is often powerful enough to eject planets from the system and completely rearrange the orbits of the planets that remain. New simulations by Russell Deitrick are revealing that this scenario is a real concern for Proxima, too; there is a significant probability that at some point in the past, Proxima swooped in close enough to Alpha Centauri A and B to cause its planetary system to break apart, hurling Proxima b’s siblings into deep space. If such a disruption occurred, Proxima b may not have formed where we find it today because its orbit would have been affected by this disruption.

Even if Proxima is not currently bound to Alpha Cen A and B, it appears to be travelling with them, and it is very likely the stars formed from the same cloud of dust and gas. If they formed together, they should have similar compositions and nearly identical ages. Connecting their ages is important because it is very difficult to measure the ages of low mass stars like Proxima Centauri. Astronomers can estimate the age of Alpha Cen A via asteroseismology, the study of “starquakes.” Stars bigger than the Sun vibrate with large enough amplitudes that brightness fluctuations can be observed, and careful monitoring of the pulsations can reveal a star’s age. Recent work by Dr. Michaël Bazot has found that Alpha Cen A is between 3.5 and 6 billion years old. This range is larger than we would like, but Proxima is certainly old enough to support life, and Proxima b might even be about the same age as Earth!

Next we turn to clues from the Proxima Centauri planetary system. The vast majority of the energy used by life on Earth comes from our Sun, and small stars like Proxima can produce energy for trillions of years. The host star is almost as small as stars come, so for a planet to receive as much stellar energy as Earth, Proxima b must be about 25 times closer in than Earth is from the sun. This distance is where the habitable zone lies. While Proxima is much dimmer than the Sun, it is still a thermonuclear explosion, and, everything else equal, life seems more likely at larger distances. And indeed the close-in orbit does produce numerous obstacles that life on Earth did not have to overcome. These include a long formation time for the star, short and energetic bursts of energy in UV and X-ray light, strong magnetic fields, larger starspots, larger coronal mass ejections, and gravitational tidal effects that cause rotational properties to change and frictional heating in oceans (if they exist) and the rocky interior.

The history of Proxima’s brightness evolution has been slow and complicated. Stellar evolution models all predict that for the first one billion years Proxima slowly dimmed to its current brightness, which implies that for about the first quarter of a billion years, Proxima b’s surface would have been too hot for Earth-like conditions. As Rodrigo Luger and I recently showed, had our modern Earth been placed in such a situation, it would have become a Venus-like world, in a runaway greenhouse state that can destroy all of the planet’s primordial water. This desiccation can occur because the molecular bonds between hydrogen and oxygen in water can be destroyed in the upper atmosphere by radiation from the star, and hydrogen, being the lightest of the elements, can escape the planet’s gravity.  Without hydrogen, there can be no water, and the planet is not habitable. Escaping or avoiding this early runaway greenhouse is the biggest hurdle for Proxima b’s chances for supporting life.

Figure 2: Proxima Centauri’s habitable zone has moved inward since it formed, which may mean that planet b lost its water shortly after it formed, when the system was 1—10 million years old. The habitable zone, shown in blue, doesn’t arrive at the orbit of planet b until almost 200 million years after it formed. This early brightness may be the biggest obstacle for life to have gained a foothold on Proxima Centauri b.
Figure 2: Proxima Centauri’s habitable zone has moved inward since it formed, which may mean that Proxima b lost its water shortly after it formed, when the system was 1—10 million years old. The habitable zone, shown in blue, doesn’t arrive at the orbit of Proxima b until almost 200 million years after it formed. This early brightness may be the biggest obstacle for life to have gained a foothold on Proxima b.

As the star dims, the water destruction process halts, and so total desiccation is not inevitable. If some water remains, the atmosphere may also contain large quantities of oxygen leftover from the water vapor destruction. While having large amounts of water and oxygen may sound like a good recipe for life, it almost certainly is not. Oxygen is one of the most reactive elements, and its presence in the young atmosphere of Proxima b would likely prevent the development of pre-biotic molecules that require conditions with little oxygen to form. Life on Earth formed when no oxygen was present, and photosynthesis ultimately produced enough oxygen for it to become a major component of our atmosphere. Note that the destruction of only some water leads to the rather surprising possibility that the planet could possess oceans and an oxygen-rich atmosphere, but has been unable to support life!

Another intriguing possibility is that Proxima b started out more like Neptune and the early brightness and flaring eroded away a hydrogen-rich atmosphere to reveal a habitable Proxima below. Such a world was investigated by Rodrigo Luger, myself and others, and was found to be a viable pathway to avoid total desiccation. Essentially the hydrogen atmosphere protects the water. If Proxima b formed with about 0.1-1% of its mass in a hydrogen envelope, the planet would lose the hydrogen but not its water, potentially emerging as a habitable world after the star reached its current brightness.

This wide range of possible evolutionary pathways presents a daunting challenge as we imagine using space- and ground-based telescopes to search for life in the atmosphere of Proxima b. Fortunately my colleagues in the Virtual Planetary Lab, Prof. Victoria Meadows, Giada Arney and Edward Schwieterman, have been developing techniques to distinguish the possible states of Proxima b’s atmosphere, whether habitable or not. Nearly all the components of an atmosphere imprint their presence in a spectrum, so with our knowledge of the possible histories of this planet, we can begin to develop instruments and plan observations that pinpoint the critical differences. For example, at high enough pressures, oxygen molecules can momentarily bind to each other and produce an observable feature in a spectrum. Crucially, the pressures required to be detectable are large enough to discriminate between a planet with too much oxygen, and one with just the right amount for life. As we learn more about the planet and the system, we can build a library of possible spectra from which to quantitatively determine how likely it is that life exists on Proxima b.

While the early brightness of the host star is the biggest impediment to life, other issues are also important. One of the original concerns for the habitability of planets orbiting red dwarfs was that they would become “tidally locked”, meaning that one hemisphere permanently faces the host star. This state is similar to the rotation of our Moon, in which the same tidal forces that raise waves in our ocean have caused the Moon to show only one face to Earth. Because it is so close to its star, Proxima b may be in this state, depending on the shape of its orbit. For decades, astronomers were concerned that such a tidally locked planet would be uninhabitable because they believed the atmosphere would freeze and collapse to the surface on the permanently dark side. That possibility is now viewed as very unlikely because winds in the atmosphere will transport energy around the planet and maintain sufficient warmth on the backside to prevent this freeze out. Thus, as far as atmospheric stability is concerned, tidal locking is not a concern for this planet’s potential habitability.

Although tidal locking is not very dangerous for life, it is possible for tides to provide large amounts of energy to the planet’s atmosphere and interior. This energy is often called “tidal heating” and is a result of the deformation of the planet due to changes in the host star’s gravitational force across the planet’s diameter. For example, if the planet is on an elliptical orbit, when it is closer to the star, it feels stronger gravity than when it is farther away. This variation will cause the shape of the planet to change, and this deformation can cause friction between layers in the planet’s interior, producing heat. In extreme cases, tidal heating could trigger the onset of a runaway greenhouse like the one that desiccated Venus, independent of starlight. Proxima b is not likely to be in that state, but the tidal heating could still be very strong, causing continual volcanic eruptions as on Jupiter’s moon Io, and/or raising enormous ocean waves. Based on the information we have now, we don’t know the magnitude of tidal heating, but we must be aware of it and explore its implications.

The host star’s short, high energy bursts, called flares, are also a well known concern for surface life on planets of red dwarfs. Flares are eruptions from small regions of the surfaces of stars that cause brief (hours to days) increases in brightness. Crucially, flares emit blasts of positively-charged protons, which have been shown by Prof. Antigona Segura and colleagues to deplete ozone layers that can protect life from harmful high-energy UV light. Proxima flares far more often than our Sun and Proxima b is much closer to Proxima than Earth is to the Sun, so Proxima b is likely to have been subjected to repeated bombardments. If the atmosphere could develop a robust shield to these eruptions, such as a strong magnetic field that then flaring could be unimportant. Alternatively if it exists under just a few meters of water. Therefore, flares should not be considered fatal for life on Proxima b.

The concern over flaring naturally leads to the question of whether the planet actually does have a protective magnetic field like Earth’s. For years, many scientists were concerned that such magnetic fields would be unlikely on planets like Proxima b because tidal locking would prevent their formation. The thinking went that magnetic fields are generated by electric currents moving in the planetary core, and the movement of charged particles needed to create these currents was caused by planetary rotation. A slowly rotating world might not transport the charged particles in the core rapidly enough to generate a strong enough magnetic field to repel the flares, and hence planets in the habitable zones of M dwarfs have no atmospheres. However, more recent research has shown that planetary magnetic fields are actually supported by convection, a process by which hot material at the center of the core rises, cools, and then returns. Rotation helps, but Dr. Peter Driscoll and I recently calculated that convection is more than sufficient to maintain a strong magnetic field for billions of years on a tidally locked and tidally heated planet. Thus, it is entirely possible that Proxima b has a strong magnetic field and can deflect flares.

So is Proxima b habitable? The short answer is “It’s complicated.” Our observations are few, and what we do know allow for a dizzying array of possibilities. Did Proxima b move halfway across the galaxy? Did it endure a planetary-system-wide instability that launched its sibling planets into deep space and changed its orbit? How did it cope with the early high luminosity of its host star? What is it made of? Did it start out as a Neptune-like planet and then become Earth-like? Has it been relentlessly bombarded with flares and coronal mass ejections? Is it tidally heated into an Io-like (or worse) state? These questions are central to unlocking Proxima’s potential habitability and determining if our nearest galactic neighbor is an inhospitable wasteland, an inhabited planet, or a future home for humanity.

The last point is not as rhetorical as it might seem. Since all life requires an energy source, it stands to reason that, in the long term — by which I mean the loooong term — planets like Proxima b might be the ideal homes for life. Our Sun will burn out in a mere 4 billion years, but Proxima Centauri will burn for 4 trillion more. Moreover, if a “planet c” exists and slightly perturbs b’s orbit, tidal heating could supply modest energy to b’s interior indefinitely, providing the power to maintain a stable atmosphere. If humanity is to survive beyond the lifetime of our Sun, we must leave our Solar System and travel to the stars. If Proxima b is habitable, then it might be an ideal place to move. Perhaps we have just discovered a future home for humanity! But in order to know for sure, we must make many more observations, run many more computer simulations, and, hopefully, send probes to perform the first direct reconnaissance of an exoplanet. The challenges are huge, but Proxima offers a bounty of possibilities that fills me with wonder. Whether habitable or not, Proxima b offers a new glimpse into how planets and life fit into our universe.

Thanks to Victoria Meadows, Edward Schwieterman, Giada Arney, and Peter Kelley.

Editorial note. This is an outreach article based on the scientific report “The habitability of Proxima b I : Evolutionary scenarios” , which was submitted to the Journal Astrobiology on Aug 25th. Proxima’s b putative habitability assessments are crucial to interpret the significance of the detection of Proxima b, design follow-up observations and even reshape instruments and space missions. The Pale Red Dot team contacted two expert groups in advance to provide these early habitability assessments at the time of the announcement. Prof. Rory Barnes led one of the teams. The results from the other team (led by I. Ribas + M. Turbat) are summarized at: and are also technically explained on two research papers. More studies are surely underway.


About the author. Rory Barnes is a professor of astronomy and astrobiology at the University of Washington in Seattle, USA. He obtained his Ph.D. in astronomy from the University of Washington in 2004. After a post-doctoral position at the Lunar and Planetary Laboratory at the University of Arizona in Tucson, he returned to the University of Washington and NASA’s Virtual Planetary Lab in 2009, joining the UW faculty in 2013. He has studied exoplanets through computer models, initially focusing on the orbital dynamics, but has now broadened his investigations to include the roles of the Milky Way galaxy, stellar evolution, atmospheric effects and the thermal and magnetic evolution of terrestrial planet interiors.

Proxima b is our neighbor… better get used to it!

It is true. We are convinced that there is a planet orbiting Proxima now. The evidence goes as follows : a signal was spotted back in 2013 on previous surveys (UVES and HARPS). The preliminary detection was first done by Mikko Tuomi, our in-house applied mathematician and his Bayesian codes. However, the signal was not convincing as the data was really sparse and the period was ambiguous (other possible solutions at 20 and 40 days, plus a long period signal of unknown origin). We followed up Proxima in the next years but our two observing runs were 12 days, barely sufficient to secure a signal which ended up being 11.2 days. So the Pale Red Dot was designed with the sole purpose of confirming or refuting its strict periodicity, plus carefully monitor the star for activity induced variability. We got very lucky with the weather so we obtained 54 out of 60 observations. The photometric monitoring telescopes (ASH2 and several units of Las Cumbres Observatory Global Telescope network), worked flawlessly so we could see the effect of spots, flares and rotation of the star, which also had a footprint on the spectra. However, nothing indicated that spurious variability would be happening at 11.2 days.

This plot shows how the motion of Proxima Centauri towards and away from Earth is changing with time over the first half of 2016. Sometimes Proxima Centauri is approaching Earth at about 5 kilometres per hour — normal human walking pace — and at times receding at the same speed. This regular pattern of changing radial velocities repeats with a period of 11.2 days. Careful analysis of the resulting tiny Doppler shifts showed that they indicated the presence of a planet with a mass at least 1.3 times that of the Earth, orbiting about 7 million kilometres from Proxima Centauri — only 5% of the Earth-Sun distance.
This plot shows how the motion of Proxima Centauri towards and away from Earth is changing with time over the first half of 2016. Sometimes Proxima Centauri is approaching Earth at about 5 kilometres per hour — normal human walking pace — and at times receding at the same speed. This regular pattern of changing radial velocities repeats with a period of 11.2 days. Careful analysis of the resulting tiny Doppler shifts showed that they indicated the presence of a planet with a mass at least 1.3 times that of the Earth, orbiting about 7 million kilometres from Proxima Centauri — only 5% of the Earth-Sun distance.

So that’s basically it : the Pale Red Dot campaign also detects the same period, and confirms that the signal has been in phase for the 16 years of accumulated observations. This is a requirement for a proper Keplerian orbit. Features like starspots are more short lived plus affect the velocities in the time-scales of the rotation of the star, which is now confirmed at ~83 days.

The combination of all the data produces this periodogram


which leaves little doubt to the reality of the signal. The peaks in a periodogram tells us where a significant period is spotted, plus give us information about its significant. The horizontal lines correspond to False Alarm Probabilities of 10%, 1% and 0.1%. Our signal is now well beyond that. The probability of a statistical false positive is smaller than one of ten millions!

So what we know? We know the period and the size of the radial velocity wobble. From that we derive a minimum mass of 1.3 masses of the Earth. With the period and the mass of the star, we know it orbits at 5% of an astronomical unit (this is 20 times closer than Earth is from the Sun), which combined with the luminosity of the star tells us that the planet is warm and can currently support liquid water on its surface. Beyond this, all is mostly speculative. But one can do simulations and educated guesses. If you want to learn more about them, follow forthcoming articles at

and a contributed one to this website by Rory Barnes.

We had a press release event at ESO today. We want to thank everyone for the passion and effort shared in this project, including the Breakthrough Starshot foundation and its chair Pete Worden for giving us their support. We hope to reach the stars, there is a foundation to promote technological advancements, and now we have a target. The sky is the limit!

Peer review — or how an experiment becomes scientific literature

What is happening now?

Now that the data collection and  analysis are complete and the results written in a paper, the next step is for the paper to be verified by the scientific community before going public. Peer review is the process the scientific community uses  for quality control of results. While a new exoplanet or supernova might have little impact on our immediate life, mistakes in some scientific disciplines (eg. biomedical research, chemistry, climate change,.. ) can have very serious consequences. Requests for research funding, patents, space missions and even new medicines are generally not accepted unless they rely on publicly available, peer reviewed research.

An important component of the peer review process are the scientific journals. Some journals will publish anything as long as it is scientifically correct, while some others will only publish results that are deemed novel or represent a very significant advance.

Who decides what it is correct and significant?

For each paper, there are at least two key people that are responsible for assessing correctness and significance. They are the editor and the referee(s). To understand how peer review works, it is better to explain the life cycle of a scientific paper.

Flow chart of the peer review process.
Flow chart of the peer review process. The approximate status of our paper as of July 1st, is marked with the red dot.


The authors must choose to submit their paper to a journal of their choice. Once the journal receives the manuscript, a scientific editor is assigned to it. This editor manages and supervises the process. Editors are respected senior scientists that work full-time for the journal, or work at a University and part-time for the journal. Papers can be rejected at this stage because the editor considers there is not sufficient original science in the result, or because the article does not match the philosophy of the journal.

Paper sent to review

After a preliminary quality assessment, the editor will search for experts to provide a more detailed revision.  These experts (called referees) are scientists not involved in the result but are experts in the field to which the paper relates. One or more referees can be assigned to a paper, and they are asked to submit a report within a  few weeks.

Referees’ opinions have a lot of leverage over the fate of a scientific result. Since referees are likely to be working on a related topic, conflicts of interest can arise and it is the editors job to carefully monitor the process. For example, if a reviewer is exceedingly enthusiastic, aggressive (or even careless), editors can search for additional referees or ignore a review. Referees are asked to follow strict ethical rules and confidentiality. The identity of the referees is not revealed to the authors to protect their independence.

First revision

After a while referee reports are sent to the editor and s/he then decides whether or not to proceed with the publication. Passing first revision is an important milestone because serious show stoppers are often identified at this stage. If the referee reports are not negative, the editor forwards them to the authors, and they are given some time to address comments and criticisms. Typical requests consist of providing additional data, analyses, adding references to previous work, and providing better discussion on obscure points of the original manuscript.

This is where we are with our Proxima paper!

After implementing the changes, the authors re-submit the article together with responses to the referee reports. The editor forwards all this information to the referees, and the process is iterated until the editor accepts it.


At acceptance the editor has become convinced that the paper meets the quality standards of the journal. They then write an acceptance notification which is met with great delight by the authors.

We hope to reach that point soon!

… but it is not over yet

Acceptance only concerns the content. At this stage authors might need to remake plots, prepare final tables and even rewrite some small parts of the paper. This process is done in collaboration with the production teams of the journal and can take from a few days to a few weeks. Final editing is performed in collaboration with professional writers who take account of English language and style.

As in any other professionally published work, the last editorial step consists of sending the paper in its very final format (commonly called  ‘galley proofs’) to the authors for their final approval. When this is done, a publication date is assigned and the peer review process is complete.


Scientific results can also be presented in conferences or other media, but these are not considered valid references unless they are published in a peer review journal. Alternative peer review procedures are being tested, but still the vast majority of scientific production goes through this classic peer review system.

… reaching the public!

It is becoming increasingly important to raise awareness of new scientific (peer reviewed) discoveries, and to be clear of what they mean to all of us. Scientists often don’t have time nor the skills to do that, so this falls into the hands of outreach, press offices, science writers and science communicators in general. When a significant result is achieved, the information needs to be transformed from the dry rigour of a scientific paper to something non-specialised audiences can digest. This includes the so-called general public, but also companies, governments and policy makers who might need to decide on crucial matters based on the most updated evidence.

So, if you are a scientist and once the paper is accepted for publication, it’s a good time to contact your outreach department and work together on how to best bring the new results to the public.

Farewell, Pale Red Dot #1

The Pale Red Dot team now goes back to their daily duties. A research paper has been written and submitted to a research journal. The review process can take anytime between a few weeks to a few months. Fingers crossed! The web articles and posts in social media will remain available for your enjoyment.

A second phase of Pale Red Dot project might start soon, with more articles and further details on what the data tells us. Do not delete us from your favourite lists just yet!

Cheers, and don’t forget to look at the sky from time to time!


Pale Red Dot team

Science and edition; Guillem Anglada-Escude (editor-in-chief), Gavin Coleman, John Strachan (QMUL/UK), Cristina Rodríguez-López, Zaira M. Berdinas, Pedro J. Amado (IAA/Spain), James Jenkins (UChile/Chile), Mikko Tuomi (Herts/UK), Christopher J. Marvin, Stefan Dreizler (U.Goettingen/Germany), Julien Morin (U.Montpellier), Alexandre Santerne (CAUP/Portugal), Yiannis Tsapras(Heidelberg/Germany).

Support; Matthew McKinley Mutter (English language editor, QMUL/UK), Predrag Micakovic (web & IT support, QMUL/UK), Silbia López de Lacalle, Ruben Herrero Illana (Editorial support and spanish translations, IAA/CSIC), Radek Kosarzycki (media partner, polish translations)

Observatories; Oana Sandu, Lars Lindberg Christiansen, Richard Hook (European Southern Observatory, Education and Public Outreach Department), Edward Gomez (, Scientist & outreach officer), Nicolás Morales (Research scientists, SPACEOSB-San Pedro de Atacama Celestial Explorations)

Biosignature Gases: A Needle in a Haystack

By Sara Seager, Massachusetts Institute of Technology (MIT)

Somewhere out there, a living, breathing world peacefully orbits its star. Inhabiting the world is a dynamic ecosystem full of thriving bacteria. The life itself has no consciousness or intelligence, but the planet as a whole is an active world, connected through cycles of geophysics, chemistry, and biology in a landscape with liquid water oceans, continents, mountains and volcanoes. We imagine there could be millions or even billions of such planets in our Galaxy. Is Proxima Centauri lucky enough to host such a planet?

To find out whether or not a planet has life, it is not enough to know that the planet is rocky and if the planet orbits in the host star’s habitable zone. We must be able to investigate the planet’s atmosphere. For Proxima Cenaturi, as in most cases, we will need a to use a different telescope than one used to discover the planet (see post by Ignas Snellan). More specifically we want to observe the planet atmosphere to first assess the greenhouse power of the atmosphere to estimate whether or not the surface temperature is suitable for life. Next we want to determine if there are gases that indicate the planet is habitable. Most challenging but most exciting, we want to know if the exoplanet atmosphere contains gases that might indicate if the planet is possibly inhabited.

How to measure the chemical signatures in the atmosphere of a transiting exoplanet. The total light measured off-transit (B in the lower left figure) decreases during the transit, when only the light from the star is measured (A). By subtracting A from B, we get the planet counterpart, and from this the “chemical fingerprints” of the planet atmosphere can be revealed. Credits: NASA/JPL-Caltech.

By habitable I mean a planet with surface liquid water oceans, since all life as we know it needs liquid water. But, since oceans are hard to identify from afar, we will look for water vapor as an indicator of water oceans. By possibly inhabited I mean the identification of signs of life by way of biosignature gases—gases that are produced by life and can accumulate in an exoplanet atmosphere to be detectable from afar. Even if we are lucky enough to detect a biosignature gas on a rocky planet orbiting Proxima Centauri, we will not know if the suspect gases are produced by tiny microbes, enormous animals, or intelligent humanoids. We also will not know if the life producing the gases is carbon based or something more exotic. We focus on what life does—life metabolizes and produces byproduct gases, not on what life is.

Oxygen is our most compelling biosignature gas. Oxygen fills Earth’s atmosphere to 20% by volume. But, without plants or photosynthetic bacteria, Earth’s atmosphere would have virtually no oxygen. Oxygen, and its photochemical byproduct ozone, have strong spectral features at a range of wavelengths, accessible with future ground- and space-based telescopes that might be able to study atmospheres of any planets discovered orbiting Proxima Centauri. If we detect oxygen, astronomers and the public alike will be absolutely ecstatic. But does an oxygen detection mean we have found alien life? No. Unfortunately the attribution of oxygen—or any gas—to life is an in depth, complicated, and somewhat subjective process. The reason is that there are many ways oxygen can be produced, and accumulate, in an exoplanet atmosphere that has nothing to do with life. We must be able to rule out all other possibilities of oxygen generation by non-biological processes. Even then, we will only be able to claim a strong suggestion of life detection not a robust detection.

A flurry of recent activity has detailed a number of different oxygen-producing scenarios that are not related to life. Most of the scenarios have to do with a lack of oxygen “sinks”. If oxygen is not destroyed, then even small quantities of oxygen can accumulate over a large amount of time. One of the more compelling oxygen false positive scenarios is related to the ultraviolet (UV) radiation of exoplanet host stars. The UV radiation splits apart molecules in the planet atmosphere, setting off a chain of chemical reactions that produce byproducts that can destroy oxygen. A major player, OH, is nicknamed the “garbage eater of the atmosphere” because of its power to destroy oxygen and other gases. M dwarf stars typically have a high far-ultraviolet radiation flux (< 200 nm) and a lower near-ultraviolet radiation flux (200-300 nm) compared to our Sun. Any exoplanet orbiting an M star will therefore be subject to different photochemistry than Earth’s atmosphere. Specifically, the chains of reactions that produces OH are weaker, owing to strong far-ultraviolet radiation. With a much smaller amount of OH compared to Earth’s atmosphere, abiotic oxygen can accumulate. To identify this false positive scenario we would need to be able to measure Proxima Cenaturi’s far-UV and near-UV radiation. Other oxygen false-positive scenarios include planets with a carbon dioxide-dominated atmosphere but little volcanic emission, an M star that took a very long time to reach a stable hydrogen burning phase, a planet undergoing a transient ocean evaporation from a runaway greenhouse effect, and more. If we are so lucky to find oxygen on Proxima Centauri, we will have a lot of further observations and atmosphere modeling work to do to understand if the oxygen can be attributed to life or if it might be a false positive.

Beyond oxygen, astronomers also consider a wide range of other biosignature gases, including methane, nitrous oxide, dimethyl sulfide, and others. Despite a growing list and detailed studies, I worry that the list of gases may be too limited, or that the types of planets modeled—usually small deviations from an Earth twin—are not broad enough to anticipate the range of what planet types are out there. If Proxima Cenaturi has a rocky planet in its habitable zone, we should do all we can to make sure we don’t miss a sign of life, just because we were too constrained in our thinking.

Life on Earth produces literally thousands of gases. Most are produced in too small quantities to accumulate to any reasonable level in Earth’s atmosphere. In addition, most are produced for highly organism-specific reasons—such as stress and signaling—that appear to be whims of evolution. Some molecules could be produced in larger quantities on another planet and/or accumulate in an exoEarth atmosphere to high levels, depending on the exoEarth ecology and surface and atmosphere chemistry. In other words, there is a possibility that any gas might be a biosignature gas, if it is present in very high quantities in an exoplanet atmosphere and can’t otherwise be explained away.

Motivated by this reasoning, my team spent a few years constructing and curating a list of all molecules that exist in gas form in a planet atmosphere with a similar temperature and pressure to Earth’s. We both combinatorically constructed lists and also exhaustively searched the literature and found about 14,000 molecules. About 2500 of these are hydrocarbons. We plan to work through this list in classes of molecules to understand their atmospheric and surface chemistry, photochemistry, and spectral properties. From this we can select both promising chemical candidates, and promising ways to search the spectrum that could capture the most diverse range of such candidates.

F7.large (1)
Schematic for the concept of considering all small molecules in the search for biosignature gases. The goal is to generate a list of all small molecules and filter them for the set that is stable and volatile in temperature and pressure conditions relevant for exoEarth planetary atmospheres. Further investigation relates to the detectability: the sources and sinks that ultimately control the molecules’ accumulation in a planetary atmosphere of specific conditions as well as its spectral line characteristics. Geophysically or otherwise generated false positives must also be considered. In the ideal situation, this overall conceptual process would lead to a finite but comprehensive list of molecules that could be considered in the search for exoplanet biosignature gases. Credit: S. Seager and D. Beckner.

Does this sound like a lot of work for a library of gases even though the study of atmospheres of any planets found to orbit Proxima Cenaturi and others lie a decade or more in the future? It is. But it will take a long time to fully prepare so we don’t miss out on a biosignature gas detection.

Despite an exhuberant realization that the search for and detection of biosignature gases is within reach, there is a long road ahead. Nonetheless the coming decades are opportune for extensive progress in finding and characterizing other Earths, and full of hope for biosignature gas detection. I remain as hopeful as ever as I plan to devote the rest of my career to the search for life on exoplanets.

Professor Seager

About the Author

Sara Seager is an astrophysicist and planetary scientist at MIT. Her science research focuses on theory, computation, and data analysis of exoplanets. Her research has introduced many new ideas to the field of exoplanet characterization, including work that led to the first detection of an exoplanet atmosphere. Professor Seager also works in space instrumentation and space missions for exoplanets, including CubeSats, as a co-I on the MIT-led TESS, a NASA Explorer Mission to be launched in 2017, and chaired the NASA Science and Technology Definition Team for a “Probe-class” Starshade and telescope system for direct imaging discovery and characterization of Earth analogs. Professor Seager was elected to the National Academy of Sciences in 2015, is a 2013 MacArthur Fellow, and in 2012 was named in Time Magazine’s 25 Most Influential in Space.

Interview to Didier Queloz — ‘From 51 Pegasi to the search for life around small stars’

Interview to Prof. Didier Queloz at University of Cambridge/UK, by Guillem Anglada-Escude

In the early 90’s the search for extra-solar planets was not even a research topic. What can you tell us about those first days?

At the end of the 80’s and early 90’s, exoplanets were not fashionable at all. I was involved in the design and building of a new type of instrument specifically designed to find planets around other stars. Our team were very successful in making key design decisions, so as soon we had the instrument on the telescope, we quickly identified one with quite a different variability from the others. It was 51 Peg.

The spectrograph concept was developed by a team under the direction of Prof. Michel Mayor. Who created the optical design? I heard that a French professor called Andre Baranne was a key person at that stage…

Yes, in any instrument, there is always an expert in precision optics. The person for that project was Prof. Andre Baranne. He was the creator of the so-called ‘white-pupil’ design, which is now adopted by most high resolution spectrometers. Before Andre’s work, spectrometers were huge, photon-eating devices. Thanks to that improvement, instruments became compact and efficient. He was close to retirement but he became very active in the project. The spectrometer was build at Observatoire de Haute Provence (OHP). In those days they had very sensitive cameras for faint objects, but a lot of telescope time could not be used because of background contamination by the moon. This is when Michel Mayor came forward offering a high resolution spectrometer for stellar astrophysics that, at the same time, would be able to detect radial velocities with unprecedented precision. Because it was a joint effort of Micheal’s team and the observatory, quite a lot of people were behind the design of the numerous subsystems.

The ELODIE spectrograph ready for operation at the 193 cm Telescoep of l'Observatoire de Haute Provence. Image credit : CNRS / OHP
The ELODIE spectrograph ready for operation at the 193 cm Telescoep of l’Observatoire de Haute Provence. Image credit : CNRS / OHP

You and Micheal Mayor were at the Geneva Observatory at the time but the spectrograph was made by OHP?

Yes, OHP built it but most participating astronomers were from Geneva. Michel already had a working instrument at OHP called CORAVEL, so it was a natural choice for him to to build the new one with them. The deal was the following; OHP would build two spectrometers, and the second one would be installed at the Swiss telescope at la Silla in Chile (CORALIE). For a number of reasons, the OHP one -ELODIE- was at the telescope first, which is where I spent most of my PhD time testing the new hardware, detectors, optical fibres, wavelength calibration using Thorium-Argon lamps and simultaneous tracking. These are obvious things to do today, but they were completely new concepts at the time. ELODIE was the first of a series of instruments that led to HARPS.

World-renowned Swiss astronomers Didier Queloz and Michel Mayor of the Geneva Observatory are seen here in front of ESO’s 3.6-metre telescope at La Silla Observatory in Chile. The telescope hosts HARPS, the world’s leading exoplanet hunter. Image credits : L. Weinstein/Ciel et Espace Photos/ESO
World-renowned Swiss astronomers Didier Queloz and Michel Mayor of the Geneva Observatory are seen here in front of ESO’s 3.6-metre telescope at La Silla Observatory in Chile. The telescope hosts HARPS, the world’s leading exoplanet hunter. Image credits : L. Weinstein/Ciel et Espace Photos/ESO

So what was the key element that made possible the breakthrough of finding the first planet in 1995?

Two really important things. We had enough telescope time to look at a meaningful sample of stars. And second, of course, we also had the machine to do it. We could regularly obtain data with a precision better than 10 m/s, which had not been possible before… and the signals were just there. Once you have done the really hard work of getting that kind of precision, the planets come for free (‘almost’). The previous precision was 50–100 m/s with instruments similar to CORAVEL, and even some first results reported by G. Marcy’s team , were in the 20–30 m/s level. When Marcy & Butler managed to get down to 5–10 m/s level, the planets started to show-up in their data too. The same for us. This new machine started delivering better than 10 m/s since the beginning, so with all this hard work done you can only start finding those planets.

How was finding 51 Peg, and more importantly, how sure were you that it was a planet? Lots of people were skeptical those days, arguing that it was an instrumental error? astrophysical artifact?..a binary?

In a sense, people were right to be skeptical. We were as well. You have to realize there were no known exoplanets in those days. It was a rather special situation. Today is very different. You can now publish, or claim detections of planets, even if you are not 100% sure because there are many of them so one more or less is not that transcendental. That was not the case back then. You REALLY needed to be sure. In our case, it was a new instrument and nobody was expecting to find a planet at such short period. I was the first not to expect it, and the same for Michel Mayor. Michel was on sabbatical, so I started the observation program more or less alone. Quite early on I picked up a strange object. It was weird, that star was clearly not stable above those 10 m/s, but it was known to be a very non-active sun-like star too. I kind of felt responsible for the operation of the spectrograph and all the software, so I became completely obsessed with it. I observed 51 Peg much more often than was planned. Consequently I found that there was a periodicity to the signal. Then I took quite some time to convince myself first that the signal was a planet without telling anybody. Convincing myself implied reviewing all the data-processing, the way the velocity was measured, that the period was not related to some instrumental issue and review the other stars in the sample. Once there was no more to check, I sent a fax to Michel who was in Hawaii. “Michel, I think I have found a planet with this period”. Michel responded “Yes, ok… maybe, I’ll see when I come back”. He was really puzzled. We then reviewed everything from the start again, thinking there might be a bug somewhere… even what we knew from the star itself; star-spots on it could create a signal.

Radial velocity curve of 51 Peg as measured by ELODIE. The radial velocity variations follow an amplitude of 59 m/s and have a period of 4.23 days. Source : OHP
Radial velocity curve of 51 Peg as measured by ELODIE. The radial velocity variations follow an amplitude of 59 m/s and have a period of 4.23 days. Source : OHP

It’s kind of funny for me, because most of what has been done later—looking at activity features and comparing it to the orbits of the possible planets—we did all these in that first paper too. I suspect nobody understood the reason for all those tests and complexity (read about the reasons in X. Dumusque’s article here). The detection of the signal was the easy part! The hard part was to be completely sure that it was a planet, and nothing else. When we had all this, we submitted the paper, and it barely got accepted. It felt a bit like magic because it was shaking the currently held theory. In a way, when we announced it at the Florence meeting, we were lucky that G.Marcy & P. Bulter were at the telescope at that very moment. G. Marcy later confessed that he thought the signal was a complete fraud, so they were also really surprised when they could confirm the signal after only a few days. This was kind of the key point of my PhD, and a big relief. That meant that the data was fine, the spectrograph was working and the period was also fine. Then we had to struggle a lot with the community. For example, many argued this could not be a planet but the atmosphere of a star changing over time. In science when you make a big claim you typically get heavily attacked, and if you survive you come back even stronger. So it took us a couple of years to convince everybody, but the final blow came in 2000 when the first of these planets was found to transit in front of the star.

51 Peg, and the planets we familiarly know as hot-Jupiters, are still a mystery and a challenge. We know a lot about these hot-Jupiters, we probe their atmospheres, we can see if their orbits are aligned with the star. But it is still a mystery how they fit in the big picture of how we think planets should form. We now know that those planets are relatively rare (about <2% of the stars have them). But with these odds, you pick up 50 stars at random and this is what you get. True enough, there was only one hot-Jupiter in our sample. In a sense, you need to be lucky to find a planet. You need the right instrument and the right strategy, and the planet needs to be there.

But one needs to push his luck…

Sure, what we were really ‘lucky’ about is that the other team didn’t get it first! Geoff Marcy started 2 years before so they could have found it two years earlier.

There were issues with resources if you ask Paul Butler (see story here!)… Are there other discoveries after 51 Peg that you feel proud of as well?

Well, I think the discovery of 51 Peg was the key to this threshold—it changed the whole game, it opened up the field of exoplanets. So I came out in this strange situation, my best ever result and highest impact paper is that first one. I mean, we created the field with 51 Peg in 1995. Before it was a weird topic, after ’95 it was a scientific topic, and the theme has been made broader because it is related to the search for life in the universe. 51 Peg was key. Of course, I have been doing lots and lots of other things, and working on other techniques like transit searches and astrometry.

What is driving your research these days?

Oh, this is simple. We have a long list of questions now. 51 Peg was the entry point. There are numerous scientific questions to answer, and a handful that are really important and deep ones. For example, the formation of our Solar System in the context of other planetary systems. We need to detect lots of planets and characterize their atmospheres to understand how planetary systems form and evolve…

…but the real question that is driving my efforts is looking for life in the universe. After finding the first planet, this is the next big thing. From a practical point of view; can we define a robust and affordable strategy to do this? I am getting more and more convinced that a step-by-step process is realistic, but it will require out-of-the-box thinking in terms of support of the science. So now I invest a lot of time to try to explain to people that the Victorian division of the sciences like Chemistry, Physics, Astrophysics, Biology doesn’t make sense in this context anymore. The question of life in the universe is a multi-disciplinary problem that needs to be tackled in a different way. I try to convince agencies, and the universities, that all the work I have been doing is about promoting this new kind of work. I might not be doing it myself because I am getting too old, but I really think that the task of the next generation of scientists won’t be searching for the planets, it will be about figuring out whether there is life on these planets.

From all the proposals to search for evidence of life around exoplanets, do you have a favorite one?

There are plenty of ways to look for evidence of life on other planets. The difference is in the practicalities. It will be enormously difficult to detect and characterize an Earth-analog around a star like the Sun. It will be done, I am pretty sure, we will eventually have pictures of such a world, we will see continents, rotation… that will happen, I am confident, nothing will stop. It is just that, being realistic, the technology we need is not there. As scientists we want to think big and far, but we also need to look at what the technology of today can achieve. Along these lines, there are a number of experiments that allow us to push pretty far in the understanding of exoplanets (post by Don Polacco). The transit technique gives potential access to the atmospheres, so we need to work on that. And the direct imaging method has finally made great progress and soon will be providing abundant information about the atmosphere of planets (gas giants first).

Can we do well enough to be able to find life? This is where we need to go back to the books. People have been thinking about this for a long time. What would an Early-Earth atmosphere look like. What about the early UV and X-ray fluxes? All the assumptions made so far were very simplistic and the habitable zone concept much tied to the Earth’s… you add some hydrogen into the atmosphere and the possible climates change completely. We need studies at telescopes, but also in the lab. My idea is being as open-minded as I can. The real drive of the field has been finding and reporting the unexpected. We really need to get away from being over-simplistic.

Today, there are kinds of stars where we might be able to do it, because it is easier. These are very very small stars (like Proxima). With the available technology of today, there are realistic chances of finding the first hints of life in planets around them. This is an amazing field of research. It is extremely exciting to begin the transition from exoplanet detection towards the search for life. These planets must be very different than Earth. Nobody has thought much about taking an Earth and putting it so close to the star. The amount of UV fluxes, tidal interactions, the nature of the atmosphere and climates… all can be so different! We have to go to the drawing board and broaden our expectations. In this sense, I think Pale Red Dot is the kind of project that is opening up where these planets are, it can lead to the new science that will explode soon. There will be some chance of seeing hints of organic activity, but let’s make it more simple… let’s look for something that tells us that an atmosphere is out-of-balance. Life takes the Earth atmosphere out of balance. This is something that cannot happen without an active agent on the planet surface. So, let’s search for signs of these atmospheres being out of balance. This will be a new big window that can potentially open the field as the first planet did. I’m willing to invest time enabling this new era.

We all have high hopes of that… so how do you see the mid-term future? Do you see a large class mission in space anytime soon?

I have experience with space missions. Careful! Space business is about minimizing risk. Space missions and agencies run away from doing new technology. On the other hand, you can do many more technological cycles from the ground. The low-mass stars can be done from the ground. And this is the problem. There is no big experiment systematically preparing to investigate planets around these very low-mass stars. There are small attempts but we really need more. The one program I am aware of is SPECULOOS, and there can be many more of these programs. But these are on small class telescopes and the goal is finding them, not characterizing them… Is there a plan for the big telescopes? No, there isn’t! We can do it and we should do it. Infrared, stabilized spectrographs on the VLT do not need a 100M investment. So a lot can be done from the ground.

Technical sketch of the SPECULOOS bservatory. Source :
Technical sketch of the SPECULOOS bservatory. Source :

Space is great, but space is not the place for innovation and development. You need to first to have the technology, show that it will definitely work, and setup long and expensive technology development programs. The European Space Agency (ESA) is not good at that. The budget is really limited compared to larger agencies like NASA. For example, ESA could not launch something like JWST. Given that this is our working framework, we should be promoting and strongly developing our ground based facilities. We could be world-leading, and we are not doing that. There are exoplanet detection programs attached to some instrument developments but, given the weight and influence of the field, we don’t have enough. We are not investing enough to go for the big challenge that is the search for life. I will be happy to change my mind if a revolutionary idea (and resources) show up. But we need to be very careful in thinking that space is the solution to all our needs.

For example, look at the gravitational wave experiments. It took 30 years to build up and refine the experiments needed to finally be successful, and they might also get a space mission. We are now in a similar situation. I think we need a bit more progress. We should be looking for life around these low-mass stars. Once we find it (or evidence for it), that will completely change the field (as 51 Peg did) . The current designs of big missions are not appropriate to search for evidence of life. People designed the missions to detect planets orbiting G,K and early M-stars. That is not what is needed in the most immediate time-frame to move forward in the search for life. My hope? When we start detecting and investigating these planets around low-mass stars, we will realize we haven’t built the right instrument and we will react to it.

A paradigm change then…

Yes, I think with experiments like Pale Red Dot and SPECULOOS it will become obvious these planets are probably there in large numbers; and then we won’t be looking for the planets themselves, we will start looking for life. The experiments and the field become different. I don’t want to minimize the importance of other questions like origins and formation of planetary systems. It is crucial to understand how the solar system started and put it in context. But if you really want to look ahead, the goal is to search for life, nothing else. By finding hints of life around these small stars, the argument will become strong and solid enough to promote and narrow-down the design of THE space mission that will address the question of life in the universe in a broader context.

Taken during the European Southern Observatory 50th anniversary gala, held in the Residenz, Munich, on October 11, 2012. Image credits : M.McCaughrean (ESA)/ESO
Taken during the European Southern Observatory 50th anniversary gala, held in the Residenz, Munich, on October 11, 2012. Image credits : M.McCaughrean (ESA)/ESO

About the author. Didier Queloz was a Ph.D. student at the University of Geneva when he and Michel Mayor discovered the first exoplanet around a main sequence star. Queloz performed an analysis on 51 Pegasi using radial velocity measurements (Doppler spectroscopy). He worked on and lead several large instrumentation projects including ground based interferometers and space-missions. He was appointed as faculty member at Geneva University in 2003, and in 2008 he became full professor. During his career he has received numerous awards and recognition and he has recently taken a Professor position at the Cavendish Laboratory at University of Cambridge (UK), where he is also a fellow member of the prestigious Trinity College.

Planetary System Dynamics

By Francisco J. Pozuelos, Instituto de Astrofísica de Andalucía (IAA-CSIC)

The ultimate goal of exoplanetary research is to place ourselves in the universe. Are we just the result of normal evolution? That is to say, does life tend to appear almost everywhere, which means that the emergence of intelligent life is just a matter of time. Or, on the other hand, are we unique? Something that just happens in a few places in the vast universe? This question has haunted the humankind since consciousness and for the first time in the history we are close to answering it. We are living in exciting times.

When the next generation of telescopes and instruments point to the sky, we will be able to observe planetary systems as never before. Super-Earths, exotic planets, planetary systems under extreme conditions… We do not know what we are going to find, but for sure it’s going to be surprising.

Artistic impression of the PLATO spacecraft searching for exotic exoplanetary systems. Credits: DLR (Susanne Pieth).

However, it is necessary to take into account that this new technology just offers us a picture frozen in time. To understand what we observe it is mandatory to develop dynamic studies on the order of the life time of the system, from a few million to billions of years. This is possible thanks to the great advances in computational science in the last decades, which allows us to investigate what causes the planetary systems to become as we see them today and how they are going to evolve in the future. It is also necessary to understand that we need to study the planetary system as a whole, taking into account other planets, planetesimal disks, even the evolution of the host star. Here we comment on a few examples of planetary dynamics which will help us to increase our knowledge about formation and evolution:

  • Planet-Planet interaction and Migrations. It seems that multi-planet systems tend to have more circular orbits. This fact decreases the influence of planets on each other, resulting them being stable for very long periods of time. On the other hand, those planetary systems with planets in eccentric orbits generate a chaotic and unstable scenario where the bodies can collide and even be thrown out of the system. In addition, during the first steps of evolution after the formation process, planets can suffer so-called “migrations”. Due to this mechanism planets can evolve to outer or inner orbits; such a scenario can explain the existence of hot Jupiters.
Simulation showing the evolution of the Solar System. Left:  early configuration of the outer planets and planetesimal belt before the Jupiter and Saturn 2:1 resonance . Center:  scattering of planetesimals into the inner Solar System after the orbital shift of Neptune (dark blue) and Uranus (light blue). Right:  final configuration after ejection of planetesimals by planets. Credits:  R. Gomes et al.
  • Tidal interactions. Some of the observational techniques used to detect exoplanets are more sensitive to those planets whose orbits are close to the host star. These planets will experience significant tidal forces as a result of this proximity. The relevance of the tides in the evolution of planets in close-in orbits was apparent with the discovery of 51 Peg b, whose semi-major axis was established as only 5% of the Sun-Earth distance. Since then, the tidal interaction between host stars and close-in planets is considered to be the cause of many effects. For example, these tidal forces are generally expected to lead to the alignment of their rotation axes, synchronization of their rotation and orbital periods, a reduction in orbital ellipticity (tidal circularisation), an accompanying reduction in semi-major axis, and a conversion of orbital energy into tidal heating of the planet. This effect of tidal heating in rocky or terrestrial planets and exo-satellites may have significant implications for habitability. For example, in our Solar System, the cool satellite Europa is a rocky body covered by 150 km of water ice crust, for which the tidal heating may maintain a subsurface water ocean. Or, in the case of the Jovian satellite Io, where the extreme violence of the tides provoke intense global volcanism and rapid resurfacing, ruling out any possibility for habitability. Thus, a correct treatment of the tidal interactions is necessary to determine if the planet was/is/will be habitable, and for how long. Of special interest will be those planets in close-in orbits, classified as terrestrial planets and hosted by M stars, where the habitable zone is expected to be in the region where the tides are acting.
Evolution of the semimajor axis (a), eccentricity (e), and distance of a synthetic planetary system composed by a Jupiter-like and Earth-like planets in presence of tidal interactions. Credits: Francisco J. Pozuelos.
  • Debris Disk-Planet interactions. Debris disks, with a qualitative similarity to the main asteroid belt and Kuiper belt in the Solar System, have been observed in various exoplanetary systems. These debris disks are composed of second-generation material, and their presence implies the existence of a significant planetesimal population. The impact range of these minor bodies and the planets is especially interesting for those planetary systems with planets on the habitable zone. First it is assumed that they are an important source of water and organics once the formation process is finished. On the other hand, a large impact removes any possibility for habitability. This fact was understood with the breathtaking impact of Shoemaker-Levy 9 with Jupiter in 1994, when the impact of two objects in the Solar System was observed for first time.

All these studies will complement the information obtained from telescopes giving us a better idea of how planetary systems evolve. We are going to be able to determine how rare the Solar System and our planet are.


About the author

Francisco J. Pozuelos is a postdoctoral researcher at Instituto de Astrofísica de Andalucía, Spain. He finished his PhD in 2014 about the relationship between comets activity and their dynamical evolution. These studies provide unique clues to determine how the Solar System was formed, and how the gravitational interactions between minor bodies and planets have affected the current Solar System configuration. Since 2014 Pozuelos is collaborating with PLATO 2.0-ESPAÑA at the University of Granada (Spain) where he is developing computational codes to study the evolution of planetary systems, taking into account processes as such as stellar evolution, tidal interactions, and the stellar wind.

Magnetic Open Cluster Stars of a Peculiar Kind observed with HARPS-POL

by James Silvester, Uppsala University

Magnetic fields play a fundamental role in the atmospheric physics of a significant fraction of stars on the Hertzsprung–Russell diagram. The magnetic fields of the chemically peculiar magnetic A and B type stars (Ap/Bp) have quite different characteristics than, for example, cooler stars like the Sun. In these magnetic Ap/Bp stars the large-scale surface magnetic field is static on time-scales of at least many decades, and appears to be “frozen” into a rigidly rotating atmosphere. The magnetic field is globally organised, permeating the entire stellar surface, with a high field strength (typically of a few hundreds up to a few tens of thousands of gauss—by comparison the sun has a polar field strength of one to two gauss). These stars are so called “chemically peculiar” as a result of having peculiar abundances (amounts) of certain chemical elements compared to what is seen in the Sun or other solar type stars.

Shows a comparison between a solar abundance model spectrum (red dashes) and typical peculiar abundance spectrum (solid green) for a 13000 K star.
This figure shows a comparison between a solar abundance model spectrum (red dashes, T=5800 K) and typical peculiar abundance spectrum for a hotter star (solid green, T=13000 K).

It is thought that the presence of this magnetic field strongly influences energy and mass transport, and results in strong chemical abundance non-uniformities within the atmosphere. These uniformities can take the form of large abundance structures in certain layers in the atmosphere.

Originally the magnetic field geometries of these chemically peculiar Ap/BP stars were modelled in the context of a simple dipole field (think of a bar magnet stuck in the star). However, with the acquisition of increasingly sophisticated data, it has become clear that the large-scale field topologies exhibit important differences from a simple pure dipole model. Through the advent of high-resolution circular and linear polarisation spectroscopy we have found the presence of strong, small-scale complex field structures, which were completely unexpected based on earlier modelling.

How do we measure magnetic fields of Ap/Bp Stars

In 1897 Dutch physicist Pieter Zeeman discovered that in the presence of an external magnetic field; light is polarised circularly if viewed parallel to the direction of the magnetic field and is plane (or linearly) polarised if viewed perpendicular to the magnetic field. In addition, spectral lines in the presence of such a field can be split into discrete levels—the so called Zeeman effect—where the strength of the magnetic field is proportional to the width of the splitting within the spectral lines.

The figure below illustrates how both a spectral line and how the linear and circular polarisation signature can change in the presence of a strong magnetic field, in the case of when the field is either perpendicular or parallel to the line of sight of the observer.

Spectral line shapes in different polarization states.
Spectral line shapes in different polarization states.

We use these effects to study the magnetic field of stars with the aid of a spectropolarimeter. The form of the polarised light we receive tells us about the direction of the field on the surface of the star with respect to the observer, and the level of Zeeman splitting within a given spectral line or set of lines allows us to determine the strength of the magnetic field.

In cases where the polarisation signal is too weak to effectively measure in individual spectral lines,  we can use line averaging techniques to improve the signal by averaging all the lines in the spectrum showing polarisation signatures into one line.
The recent advances in tomographic imaging techniques and the new generation of spectropolarimeters such as ESPaDOnS (at the CFHT), NARVAL (at the TBL, Pic du Midi) and HARPS-pol (ESO 3.6 at La Silla) offer the opportunity to improve our understanding of the magnetic field of Ap/Bp stars by allowing us to map the magnetic field and chemical surface structure in quite some detail.  (For the finer details of how we map magnetic fields see the great article by Élodie Hébrard and Rakesh Yadav).

The extended magnetic field topology map for the Ap star HD 32633, the number in the right hand corner is the phase of rotation.
The extended magnetic field topology map for the Ap star HD 32633, the number in the right hand corner is the phase of rotation.

Even though we are able to very successfully map the magnetic fields of Ap/Bp stars, there are however some questions that remain open.  Notably the origin of these magnetic fields is not fully understood and importantly neither is the evolution of such magnetic fields and the atmospheric chemical structures with time. This is where Ap/Bp stars in open clusters comes in.

The Current Project – Observations of Cluster Ap/Bp Stars

The question of the evolution of the magnetic field and chemical surface structures in Ap/Bp stars can be investigated by studying these stars in open clusters.  An open cluster is a group of gravitationally bound stars, and whilst it is very difficult to get an precise age for an individual star, it is however possible to get a more precise age for an open cluster, because you have an ensemble of stars which are thought to be of a similar age. Therefore if you can confirm that a star is a true member of a cluster,  you have a much more reliable age for that star than if it was an individual star.

By studying the magnetic fields of stars in different open clusters with different ages, we can in essence look at Ap/Bp stars at different stages of evolution.  This allows us to investigate if the magnetic field complexity, or the form the magnetic field takes, varies as a function of age,  e.g does the magnetic field structure of an Ap/Bp star evolve with time?

The cluster NGC 6475 / M7. Image from the Wide Field Imager on the MPG/ESO 2.2-metre telescope at ESO’s La Silla Observatory in Chile. Image credits : ESO
The cluster NGC 6475 / M7. Image from the Wide Field Imager on the MPG/ESO 2.2-metre telescope at La Silla Observatory in Chile. Image credits : ESO

Our team, including astronomers based in Sweden, Canada and France, has begun to obtain observations of cluster Ap/Bp stars using the HARPSpol spectropolarimeter. By measuring the circular polarisation of these magnetic Ap/Bp stars and by obtaining measurements at all phases of rotation, we will be able to create magnetic and surface chemical maps for all the stars we observe.   It is hoped that the resulting maps from our target stars will give us insight into how the magnetic field geometries and chemical surface structures of Ap/Bp stars vary with age.   Having more information about the evolution of the magnetic field will also provide a powerful constraint for stellar evolution models.


About the author. James Silvester is a postdoctoral researcher at Uppsala university in Sweden. He completed his BSc in Astrophysics at the University of Hertfordshire in 2004 and then moved to Canada to do an MSc (2007) and his PhD (2014) at Queen’s University in Kingston, Ontario.  The main focus of his research is to understand the magnetic fields of intermediate mass stars. James and his team use HARPS combined with a recently installed polarimeter (also known as HARPS-POL) to measure magnetic fields. His program overlapped with Pale Red Dot, and he was one of the observers that helped obtaining those Proxima spectra by the end of each night. Astrophotography is also among his hobbies.

Magnetic Fields: those troublemakers!

By Rim Fares, Osservatorio Astrofisico di Catania, Italy

I am sure that you are already convinced of this statement if you have followed the previous contributions to this blog. But let’s make sure things are clear: isn’t everything relative? Magnetic fields do trick planet hunters, but their study gives us insights into stars and planetary environments, which help us better understand exoplanetary worlds.

So why do we worry about magnetic fields if we want to detect Earth-like planets?

They can mimic a planet’s signature and make our goal of detecting Earth-like planets much tougher.

In cool stars (don’t be mislead by this name; these stars still have surface temperatures up to 7,000 K), magnetic fields manifest on the stellar surface in different ways. They can emerge in the form of dark spots. These regions have strong magnetic fields (up to a few thousand Gauss in the solar case). As Xavier Dumusque explained in his contribution, the effect of these spots can be greater than the effect of an Earth-like planet on the radial velocity variations of a star.

This is not everything; stars have large-scale magnetic fields too. While spots are small areas on the surface relative to the size of the star (but can be the size of a planet!), large-scale fields are distributed on the scale of the star itself. The star can, for example, act as a huge dipole. This dipolar field contributes to the large-scale field.

As Elodie Hebrard and Rakesh Yadav told us in their contribution, stars have magnetic personalities. Extensive studies of large-scale fields show that stars with similar masses and rotation periods seem to have magnetic fields that share similar properties (field strength, configuration). To detect very small planets, we need to understand the contribution of the large-scale field to the radial velocity variations of a star. But to make things complicated, stars can have magnetic cycles. Which means that both small-scale (e.g. spots’ emergence) and large-scale fields vary with time.

Cristina Rodriguez told us in her contribution that stars beat by waves’ propagation inside the stars. These beats can mimic planetary signals. Over a cycle, the magnetic field changes, and so does the music of the star.

But magnetic fields are not as evil as you think. Have you ever dreamt of watching an aurora, wondered about the ice age, the safety of astronauts? Well, there is a magnetic field contribution to all that. They play an important role from the birth of a star to its death (they affect stellar rotation, stellar wind, mass loss of the star, …). They contribute to the interactions between stars and their surrounding planets. Magnetic fields of planets protect them from stellar winds and are probably important for habitability.

Figure 1. Artist impression of the interaction between the Solar wind and the earth’s magnetosphere. The magnetic field of the Earth (presented to the right) form a magnetosphere that protects us from the Solar wind (coming from the Sun – to the left). A bow shock, highlighted in purple, form from such interactions. Credit: NASA/ESO/SOHO

While it is a tough job to detect Earth-like planets, it is much easier to detect massive planets that are very close to their stars (because they produce a much stronger effect on the radial velocity of the star, or on the light-curve when they are transiting). These planets are called hot-Jupiters, because like Jupiter they are very massive, but more than 50 times closer than Jupiter is to the Sun, so they are pretty hot. Their discovery was a big surprise to the community, because we do not have such planets in our own solar system (the massive gaseous planets are the furthest from the Sun). It raised questions about how they form, and how they migrate from further distances to very close to the star. At these distances, being massive, these planets interact with their stars through tidal interactions (like the Earth-Moon case, and its effect on the oceans’ tides). The large-scale magnetic field of the star and the stellar wind also interact with these planets (remember the auroras I talked about? They result from the interaction between the solar wind and our atmosphere, and the amazing colours they produce are due to the molecules we have in the upper atmosphere).

These interactions affect the planet but they might also affect the star. In fact, some observations suggest there are spots on the stellar surface that follow the planet on its orbit, instead of rotating with the star. When we observed a sample of stars that have hot-Jupiters, we discovered the first polarity flip of the magnetic field for a star other than the Sun. This star is Tau Bootes, a star visible to the naked eye in the Bootes constellation. 

Figure 2. Artist impression of the system Tau Bootes. The star has a large-scale field that flips on yearly basis. The exoplanet is very close to its parent star. Credit: Karen Teramura, University of Hawaii Institute for Astronomy

Imagine this star being a huge dipole (Figure 2), the north-south poles flip every year. This discovery was very surprising because the Sun flips polarity every 11 years. Tau Bootes’ flips are very fast compared to the Sun. When we discovered that, we though that it might be due to the tidal interactions with its very massive planet. This star and the planet are synchronised, which means that the star rotates with the same period of the planet on its orbit. This system was the only one observed with such a characteristic, and it was the only one for which a fast magnetic flip is detected. Coincidence or causality? We are still puzzled about that. We did observe polarity flips in other stars more recently, but for now Tau Bootes is the only star observed with regular flips over a long time period.

Back to the planet—imagine it close to the star, and bombarded by the stellar wind. It can interact with this wind and produce signatures at different wavelength (that’s cool because if you observe the same system with different instruments, you can have the whole picture of the physics that is happening up there). It can for example produce radio emission and signatures in the UV that can be due to a bow shock formation, or to the evaporation of the atmosphere.

We know that these hot-Jupiters are not habitable, so why do we care? Because they are easy to observe and characterize, but also because they give us hints to our future studies on habitable planets. Actually, Earth-like planets we are trying to detect in the habitable zone around cool red dwarfs are subject to a more aggressive environment than that of our own Earth. Red dwarfs can have strong magnetic field, and frequent ejections of particles toward the planet. The planet needs to be protected from the stellar wind in order to keep its atmosphere and be habitable. Consider the example of a volcano. Here in Catania where I currently work, we have an amazing view of Mount ETNA, the highest active volcano in Europe. The soil is fertile and farmers are happy. But the volcano is active, it can erupt and lava flows can destroy villages (luckily though, the flows are usually slow). The stellar ejections can have similar effect as volcanoes on earth, the exoplanet needs to get a protection from them: a planetary magnetic field (that will help deviate the stellar particles from hitting the planet). For now, we cannot detect planetary magnetic fields directly, we have to use indirect techniques, for example detecting a bow shock in the ultra-violet. If we study the stellar magnetic field and wind, and detect these bow shock (or radio emissions), we can calculate the magnetic field of the planet. Currently, we are trying to detect these effects on hot-Jupiters. This will shed light on the strength of the planetary magnetic fields.

The coming years will see many space and ground based instruments, and we are all excited about getting new data and making new discoveries. Among these instruments, SPIROU-–a spectropolarimeter for the study of Earth-like planets in the habitable zones—will help us study the magnetic field while having very good radial velocity precision. SPIROU also has another goal: helping to understand planetary formation and migration by observing young stars and the hot-Jupiters that are formed around them. Exciting times ahead!


About the author. Rim Fares is a research fellow at the Osservatorio Astrofisico di Catania, Sicily, since February 2015. Her main research interests are cool stars’ magnetism, the interactions between exoplanets and their cool stars, and exoplanets’ environment. Before moving to Italy, she was a research fellow at the University of St Andrews, Scotland. She has a PhD from Paul Sabatier University, Toulouse, France.


‘A brief personal History of Exoplanets’, by Paul Butler

by R. Paul Butler, Staff Scientist at the Carnegie Institution for Science


I began working on exoplanets in 1986.  At the time there were no known planets beyond the solar system.  An exoplanet meeting could have been held in a phone booth, of which there were still
many. When asked by other astronomers, “What are you working on?”, one could not respond, “I am searching for extrasolar planets.”
Depending on the person, they might laugh in your face, or
slowly move away from you like you were pitching a new age
religion or alien conspiracy theories.  For most of the previous hundred years sensible planet search programs had relied on the astrometric technique, looking for a nearby star to wobble relative  to background stars. Astronomers were aware of the possibility of  detecting planets by the Doppler velocity method; they knew that Jupiter caused the Sun to wobble by a velocity of ~10 m/s.  They  were also aware that achieving Doppler velocity measurement precision better than 1,000 m/s was difficult, and achieving precision better than 300 m/s was impossible. In 1973 Griffin & Griffin wrote a seminal paper in which they identified several of the most important sources of measurement uncertainty, and  challenged the community to improve velocity precision down to the undreamed of level of 10 m/s.

In the late 1970s Bruce Campbell and Gordon Walker conceived the idea of using a gas absorption cell inserted in the beam of the telescope.  The starlight is collected by the primary mirror, and passes through the gas absorption cell just prior to entering the spectrometer. The spectrum of the gas vapor in the absorption cell is imprinted on the starlight, and provides a reference spectrum against which to measure the Doppler shift of the star.  The reference spectrum is essentially a measuring stick.  Campbell and Walker spent 8 years solving a myriad of problems. Along with their small team, they achieved the critical breakthrough of improving Doppler velocity measurement precision from 300 m/s to 13 m/s.

Bruce Campbell and Gordon Walker, pioneers in precision Doppler spectroscopy via use of a Hydrogen-Fluoride absorption cell, which is both a corrosive and a highly poisonous gas.

The Campbell and Walker gas absorption cell was filled with hydrogen-fluoride (HF) vapor, an extremely dangerous gas that slowly eats glass.  Another disadvantage of HF is that it only provides a few reference lines over a limited wavelength range. With so few stellar and reference lines, they were forced to take hour-long exposures at the telescope, which limited their survey to about 20 stars.

The advent of CCD detectors, and improving computer speed and storage, led to the development of modern echelle spectrometers in the early 1980s.  Arguably the first modern echelle spectrometer, the Hamilton, was designed by Steve Vogt and built in the Lick Observatory optical shop in the early 1980s. This spectrometer remains in use on the 3-m Shane telescope at Lick, and can also be fed by the 24-inch CAT (coude-auxillary-telescope).

Geoff Marcy was Steve Vogt‘s graduate student during much of the time that Vogt was designing and building the Hamilton spectrometer.  In 1986 Geoff Marcy was an assistant professor at  San Francisco State University.  I was a Master’s physics student, with an undergraduate degree in chemistry.  For my Master’s thesis we agreed to work on improving Doppler velocity precision—with the goal of detecting extrasolar planets.

Marcy was aware of the great strides in measurement precision made by Bruce Campbell and Gordon Walker.  I followed Campbell and Walker’s idea of observing stars through an absorption cell. For sun-like stars, most of the velocity information is in the visible portion of the EM spectrum, so I began looking for gases that absorb light in the visible.  In essence I was looking for a colored gas. A major problem that emerged was that most colored gases are either explosive, deadly poisonous, or both. After 6 months in chemistry libraries, chemistry laboratories, and day-time tests at the Hamilton spectrometer, we settled on using molecular iodine (I2).  Iodine vapor is a shade of violet, and produces thousands of absorption lines from 5,000 Angstroms (green) to 6,200 Angstroms (red).  Along with Mylan Healy, the SFSU chemistry glass blower, I constructed the first precision velocity Iodine cell in May 1987.

The iodine cell was first used to take stellar data with the Hamilton spectrometer on the evening of June 10 1987.  I completed my physics Masters thesis at SFSU in August 1987. I then moved to the University of Maryland to pursue my PhD.

Original Iodine cell for Lick Observatory
Original Iodine cell for Lick Observatory.

We ran into a host of problems along the path to obtaining precision velocities, many of them the same problems that Campbell and Walker had faced.  Spectrometers are composed of real stuff; lenses, mirrors, gratings made of different types of glass, separated and held in place with components made of different metals and other materials.  Each of these materials expands and contracts at different rates with changes in temperature.  Imperfections and jitter of the telescope drive cause the starlight to wander on the entrance slit to the spectrometer.  The smearing function of the spectrometer varies with changes in temperature, air pressure, and telescope guiding. This is why prior to Campbell and Walker, Doppler measurement precision had been stalled at 300 m/s for decades.

On short timescales—less than an hour—we were quickly able to  achieve a precision of 5 m/s.  But night-to-night and month-to-month, the precision was 100 m/s or worse.  It took 5 years to achieve long term precision better than 20 m/s. The key breakthrough was suggested by Jeff Valenti, a PhD student at Berkeley at the time.  Valenti was also using the Hamilton spectrometer, with the goal of measuring magnetic signatures in the spectrum of stars, a subtle effect.  Valenti suffered from many of the same problems, in  particular the variable smearing function of the spectrometer. Valenti suggested that the spectrometer smearing function could be directly determined by observing a stable, known spectrum. He  suggested making observations of the Sun, either during the day, or by observing the moon or an asteroid, which reflect sunlight.

We realized that we had a known spectrum embedded in every observation we took, the molecular iodine from the iodine absorption cell.  In 1991 we took the Lick Observatory iodine cell to the McMath Solar telescope on Kitt Peak in Arizona. The McMath had a very special type of spectrometer, a Fourier Transform Spectrometer (FTS).   FTS spectrometers provide extraordinarily high resolution—a factor of twenty or better than high resolution astronomical echelle spectrometers.  Astronomers don’t use FTS spectrometers at the telescope because they require more light than telescopes can provide. FTS spectrometers are typically used by physicists in atomic spectroscopy labs, and at solar telescopes. A detailed comparison of the FTS spectrum of the iodine absorption cell with the iodine cell as observed with the Hamilton echelle spectrometer allows for the Hamilton smearing function to be modeled and accounted for.  I wrote the first software that could model and account for the spectrometer smearing function in early 1992.  A major problem was the speed of early 1990s computers. Observations that took 5 minutes at the telescope required more than 6 hours to analyze on the computer.

Though we could now achieve precision of 15 to 20 m/s, we  continued to use all of our limited computer power in an effort to improve their nascent Doppler velocity reduction software. This was motivated by the results of the Canadian program, which stopped taking data in 1992. With 12 years of data covering 21 stars at a precision of 13 m/s, they did not find any planets (Walker et al. 1995).  Based on this result we decided that precision of 5 m/s or better was needed to make progress. In January 1993 I completed my PhD at the University of Maryland, and moved back to California where I began a postdoctoral position at SFSU and UC Berkeley.  Over most of the next 3 years I worked on improving the velocity reduction software package.

Echelle spectrum as it would have shown in the display of the Hamilton Spectrograph back in the 90’s.

The next big breakthrough in the project was led by Steve Vogt. Vogt had recently completed the design and construction of the HIRES  echelle spectrometer for the Keck 10-m telescope.  Based on a number of advances made over the previous decade, he went back to work on the Hamilton echelle.  In November 1994 he replaced the spectrometer camera with a new design that he and Harland Epps invented.  The new design dramatically improved the resolution of the Hamilton.  In addition he replaced the old CCD detector with a next generation detector that was 6 times larger, significantly increasing the amount of spectrum that could be analyzed.

Steve Vogt with the Hamilton Spectrograph inside the Shane 3-meter dome. Image credits : Laurie Hatch 2003 (c)
Steve Vogt with the Hamilton Spectrograph inside the Shane 3-meter dome. Image credits : Laurie Hatch 2003 (c),

After the Hamilton spectrometer upgrade, my effort was focused
on the newly emerging higher quality data.  By May 1995 the
upgraded hardware and software were producing 3 m/s precision.
Computer speed continued to be a problem.  We had two
computers between us. The 8 years of data we had collected
would require several years of computer time to analyze.


At a meeting in Italy during the first week of October 1995 Michel  Mayor and Didier Queloz announced the discovery of a very strange planet.  51 Peg b has a mass similar to Jupiter, but orbits its host star in 4 days.  While these “hot Jupiters” are now known to be common, at the time nobody had suggested that such planets could exist. Much of the astronomical community as well as the press were skeptical of the claim. We had already been assigned 4 nights of precious time on the Lick Observatory 3-m telescope beginning on the evening of October 11.  We observed 51 Peg multiple times each night. I reduced just the 51 Peg data each day, which was all our  computers were capable of handling.  The observing run concluded on the morning of Sunday October 15.  The first 3 nights of data were consistent with the discovery announcement from Mayor and Queloz, but we wanted to see the final night of data before going public. After 4 nights on the mountain, we drove back to Berkeley and crashed. It took our two computers all day to reduce the  51 Peg observations from the final night.  We met back at our Berkeley office at midnight.  Within a half hour we were able to confirm the discovery of the first extrasolar planet.  We put a plot of the 4 nights of data, along with the orbital fit, on the then brand new World Wide Web.

Doppler measurements of 51 Peg from observations were made at Lick Observatory between Oct.11, 1995 and Dec. 1996.
Doppler measurements of 51 Peg from observations were made at Lick Observatory between Oct.11, 1995 and Dec. 1996.

The discovery of 51 Peg b marked two major changes for the Lick
Planet Search Program.  No longer was the primary target
Jupiter-analogs with 12 year orbital periods.  Planets could
be found at any orbital period, and could already be embedded
in the raw data taken over the previous 8 years.  The second
change was that the field of extrasolar planets had suddenly
become very hot. In the wake of the newspaper and TV publicity
that followed the discovery of 51 Peg b, several research groups
at UC Berkeley offered the loan of research computers.
Shortly thereafter SUN Microsystems made a grant of additional
research computers to the Lick Planet Search Program.

Snapshot of hardworking scientists at San Francisco State University with up-to-date computing facilities of the early 90’s.

In late October 1995 I finalized the Doppler velocity reduction analysis, and began analyzing the 8 year backlog of  data on an armada of computers that finally topped out at more than 20 machines.  Clearing out the backlog of 8 years of data took until June 1996.  Analyzing all the observations of a single star could take from half a day to several weeks, depending on how many observations had been taken.  Observations taken after Steve Vogt’s upgrade in November 1994 are internally referred to as “post-fix”.  The data from the first 7  years is “pre-fix”.  These are two separate data sets.  Upgrading the camera and the CCD detector made the Hamilton a completely new spectrometer, requiring a completely new Doppler analysis  package.  Stitching together the “pre-fix” and “post-fix” data sets was a major problem.  This problem has re-emerged on most of my subsequent Doppler surveys. By mid-December 1995 hints of planet signals were emerging from the data.  At 8 a.m. on the morning of Sunday December 31, I walked into the deserted Berkeley astronomy department to check on the armada of computers.  A few jobs had finished, so I loaded the available computers with new stars and looked at the recently analyzed data.  The bright nearby star 70 Vir had a whopping signal, the star was being tugged back-and-forth by several hundred meters per second.  Within 5 minutes I had fit the data with a Keplerian planetary orbit indicating a 7 Jupiter-mass planet in an 116 day orbit.  The signal was so overwhelming that there could be no doubt.  This was the first definitive planet to be discovered by the Lick Planet Search Program. After 9 years of working toward this moment, I was stunned, silent. I closed my eyes for several minutes, then looked back at the  computer screen.  The signal was still there.  I did this several times to make sure that the signal did not vanish.  In the absolute quiet of a New Years eve Sunday morning I sat for the next hour looking at the signal.  For a long time I had the sense that Johannes Kepler was standing over my shoulder, looking at the same signal.

Doppler signal of 70 Vir b, the first planet detected by the Lick Planet Search Program with the Iodine cell technique.

Over the next two weeks the case for a planet around 47 UMa firmed up. I solved the problem for putting together the pre-fix and post-fix data.  The improved precision of the post-fix data sat on the pre-fix planet prediction like pearls on a string. We announced the planets around 70 Vir and 47 UMa at the the winter meeting of the American Astronomical Society in San Antonio Texas on January 17, 1996.  The story received significant press coverage, including the front page of the NY Times, Washington Post, and the cover of Time magazine.

Over the next 5 months the armada of computers ground through the 9 years of Lick observations.  The next 4 planets quickly emerged, including the planets around rho 1 Cnc, tau Boo, nu And, and 16 Cyg B. The Lick Planet Search Program continued to churn out planets over the next decade.  A few highlights include the discovery of the planet around an M (red) dwarf star (GL 876) in 1998, and first multiple planet system orbiting nu And in 1999. Our planet search programs at the Lick, Keck, Anglo-Australian and Magellan Observatories have subsequently found hundreds of planets, including the first transiting planet, the first Saturn-mass planet, the first Neptune-mass planet, and the first Super-earths.

The camera is positioned near the 2.4-meter primary mirror in the dome of the Automated Planet Finder Telescope at Lick Observatory. At upper right is the secondary mirror. APF is fully robotic, and equipped with a high-resolution spectrograph (designed by Steve Vogt) optimized for precision Doppler measurements. Laurie Hatch 2009 (c),
The camera is positioned near the 2.4-meter primary mirror in the dome of the Automated Planet Finder (APF) Telescope at Lick Observatory. At upper right is the secondary mirror. APF is fully robotic, and equipped with a high-resolution spectrograph (designed by Steve Vogt) optimized for precision Doppler measurements. Laurie Hatch 2009 (c),

After a decade of work, teams at the Carnegie Observatories (Stephen Shectman, Jeff Crane, Ian Thompson) and Lick Observatory (Steve Vogt) have inaugurated the first two purpose built precision velocity spectrometers used with iodine cells.  These instruments (PFS at Magellan and APF at Lick) are producing precision of 1 m/s.  Our Lick and Carnegie teams are enthusiastic about the next decade.  Over the past 25 years the iodine absorption cell has become a standard tool for measuring stellar Doppler velocities. Teams from many institutions  such as University of Texas, Penn State, Yale, Harvard; and national and international facilities from Japan, China, Australia, the European Southern Observatory (VLT/UVES), have adopted the iodine technique.

The most exciting discoveries are yet to come.



About the author. Paul Butler is a Staff Scientist at the Department of Terrestrial Magnetism of the Carnegie Institution of Washington.  Previously he served as a Staff Astronomer at the Anglo Australian Observatory in Sydney Australia and as a Research Fellow at UC Berkeley. Butler’s work has focused on improving the measurement precision of stellar Doppler velocities.  He designed and built the iodine absorption cell system at Lick Observatory which resulted in the discovery of 5 of the first 6 known extrasolar planets.  Thanks to its simplicity of implementation and demonstrated performance, the iodine cell technique has been instrumental in many posterior planet search programs worldwide, such as Keck and the Anglo-Australian Telescope.  Butler’s work has resulted in the discovery of most of the first 200 extrasolar planets, including many of the known multiple planet systems, and many firsts like the first Neptune-mass planet, and the first few-Earth mass objects. This work has been featured in PBS documentaries, front page articles in the New York Times and Washington Post, as well as a TIME magazine cover story. He has received the Bioastronomy Medal from the International Astronomical Union, and the Henry Draper Medal from the National Academy of Sciences.  He has served as a Centennial Lecturer for the American Astronomical Society, and he has been named Space Scientist of the Year by Discover Magazine. Butler received his Ph.D. from the University Maryland under the supervision of Dr. Roger Bell.  He received his BS in chemistry and MS in physics from San Francisco State University.