30 Aug 2016

Exoplanet Safari

This is the talk I gave at Brighton Astro. It’s a tour of six exoplanets I liked the look of.


I’m going to share what I know about exoplanets. These are planets way outside our solar system. Planets that have been detected around other stars. To give them their full name they are extra solar planets, but that’s exoplanets for short.

I’ve picked six planets that are important in some way, or that I found interesting. And the idea is to say what we know about them, and as we go along explain how we know it.

We’ll spend a bit of time on the first couple, but then we’ll zip through the rest, and hopefully have fun speculating wildly about what it’d be like to visit them.

There are over 3,000 discoveries to pick from. To give you the context, here’s our galaxy.

Follow along with the slides at https://speakerdeck.com/d6y/exoplanet-safari

The sun is the red mark in the middle. Each of these bright white points are the stars we know harbour planets. So we’ve only uncovered a fraction of our galaxy so far. There’s plenty more to explore.

The furthest is 27,000 light years away —a distance I can’t begin to understand. And it’s mind boggling to think we can detect a planet at that kind of distance.

The closest was reported last week at the European Southern Observatory. An exoplanet right at the next star along, four light years away. Give or take. Well get to that one later.

Start with the first one

We’re going to start with what’s considered to be the first exoplanet. That’s 51 Peg b.

Here’s the artist impression. There are no images of exoplanets anything like this. And we’re going to be seeing a few artist impressions tonight… so don’t confuse this with real images.

And a confession: it’s not really the first exoplanet. Earlier exoplanets were found, but are huge, more like “failed star”, or orbit around stars very different to our own, such as pulsars.

But this planet was discovered in 1995, is the first… exoplanet around a sun-like star. And this kick-started the discovery of exoplanets. Which is why we are starting with it.

What do we know about this planet?

Let’s start with the name. The convention used is to add a letter on to the name of the star. This star is 51 Pegasi. I thought stars had a name, and that was it. But oh no. A star can have lots of name, depending on what star catalog you’re using. It’s not uncommon to see a star with 4, 5…10 names.This particular name is star 51 in the constellation of pegasus. It’s a naming scheme that started in 1700s.

For the planet, you add a letter. The “a” is reserved for the star itself, but then the first planet discovered is “b”. So “51 peg b” is the first planet discovered around 51 peg.

What else do we know? I’ve not memorised all the facts. I’ve made myself these Exoplanet TopTrumps cards:

Exoplanet top trumps

We know some things about the star, 51 Peg, is about 50 light years away. It’s about the same size as the Sun.

The planet is big. It’s drawn here to look a bit like Jupiter, I think. Its radius is estimated at over 1.5x Jupiter.

And it orbits the star every four days, very close in. The distance of 0.05 AU… much closer to the star than Mercury is to the Sun. Here’s a diagram of the system, and this is where Mercury’s orbit would be. Really close, so it’s hot too. Getting on for 1000 degrees C.

The mass of the planet — how much stuff it’s made off — is probably only half that of Jupiter. But I’ve said it’s over 1.5 times the radius of Jupiter. The interpretation is that the atmosphere is heated and expanded out much more than we see for the much cooler Jupiter.

Not a place we want to visit, not a place you’d first look for life.

But it’s interesting. This is nothing like our solar system: this is a hot Jupiter. Before this planet, we thought gas giants like Jupiter or 51 Peg b would be much further out, where they are in our solar system. And they would get big from hoovering up all the gas pushed away from the Sun. And the rocky planets would be close in, in smaller orbits gathering heavier material.

Finding a Jupiter-sized planet this close didn’t fit with then then theory of how planets form; so the theory was wrong. Or, at least incomplete. Maybe planets migrate in: form far from the sun, then move closer. Maybe Jupiter’s can form this close.

So this was a pretty exciting find, and shook things up.


So how do we know all of that?

Well there are a whole bunch of ways exoplanets are being detected. This diagram is trying to explain that. It’s from the Exoplanet Handbook. What it’s showing is a branch for each different kind of detection technique. And the length of the line indicates how accurate it is. For example, this line here is showing the first method we’ll be looking at, called Radial Velocity (RV). And you can see that it goes all the way down here from being able to detect planets 10x the size of Jupiter, to Jupiter sized, to 10x Earth size and then this dotted line showing it’s getting to Earth-sized planets. And generally, there’s this progression of all these techniques getting more and more sensitive and technology progresses.

So we have all of these tools. Most of them I know nothing about. We’ll be focussing today on RV, and this line over here, Transits measured by space craft. Disappointingly, this line, showing imaging, looks like it’ll take a while to get some really nice photos of exoplanets. It’s tricky because stars are very bright (of the order a billion times brighter than a planet), but even this is being worked on.

51 Peg b was discovered by a technique called radial velocity. Let’s see how that works out.

It was a team from the University of Geneva, in 1995, and they used this telescope in southern France. It’s a 1.93 m telescope.

They didn’t directly see the planet. They were recording the spectrum of light from the star, and looking at how it changed over time.

Here’s how that trick works.

When we talk about, say, the moon orbiting the Earth, what really happens is the Earth and the Moon orbit around a common centre of mass. And the same is true for a star and an exoplanet. Here we see a planet going around a much bigger star. And notice the star is moving. It’s wobbling around this centre point.


So if you imagine observing that star, what you’d see is the star coming towards you, and going away, then coming towards you, and going away. And that makes a change in the light: it’s the doppler effect, and when the star is coming towards us, the light is “blue shifted”, and when it’s going away it’s “red shifted”.

The comparison is to the sounds of police siren increasing in pitch as it gets closer to you, and the pitch decreasing as it goes away. The explanation is that the sound waves are getting squashed together as they approach, which we hear as higher pitch. And stretched out as they go away, which we hear as lower pitch.

The same kind of thing is happening with light. So imagine we’re recording the light from the star. As it comes towards us, it is shifted to the blue end of the spectrum; as it goes away, it’s shifted to the red end.

And the interpretation you make is: more blue, is faster towards you; more red, is faster away from us. You’d end up recording a curve like this. The y axis is velocity which goes up to 50 meters a second when the star is approaching as fast as possible here. It’s zero here, where the star is just moving sideways. Then it’s going away super fast at this point. Then zero again here. Then speed increases as it comes back around.

Three facts: RV

Let’s focus in on the graph.There are three key things we get from this graph.

The first is that these changes repeat, which tells us the period of the orbit. You and I would call it a year, even thought it’s four days long in this case.

That also tells us how far the planet is from the star. That’s due to Kepler’s Law — which tell us the duration of the orbit is proportional to the distance from the star. The further away from the star, the longer it takes to go around. If we know the properties of the star, we can work out how much energy is hitting the exoplanet, which will give us information about the temperature.

We also know the mass of the planet. The height of this curve gives us that. More mass has a larger impact, and this peak is higher. Or looking at it another way, a tiny world would have no detectable effect on the wobble of the planet.

We do not get the radius of the planet from this method. Now, 51 Peg b has been around for a while, and we know from follow-up work using different methods, not RV. And generally, I think that’s what you want: throw as many techniques as you can at a planet, and get confirmation from different techniques.

So that’s what we know about 51 Peg b. It’s very different from our solar system, shook up how we thought planets form, and it did seem to kick start exoplanet discovery.

It didn’t happen overnight — back then computers were slower, it took a day to process a single light sample, and it was difficult getting money for this kind of work. But researchers realised there was stuff to detect, and as you can see from this graph, the number of discoverers per year started to grow. This huge spike here is from the Kepler mission, and we’ll look at that next.

Earth 2.0

The next planet is the safari is this one: strikingly different from 51 Peg b.

So you know from the name that this star is Kepler 186, and we’re looking here at “f” — the fifth exoplanet discovered around it. You can see the other four closer in on this artist impression.

This star is over 500 light years away. Which goes to show the improvements in detection. This was reported in 2014, so 20 years after the first exoplanet, we’re finding smaller worlds 10 times further away. Because this is an Earth-sized planet. Orbits the star every 112 days.

We don’t know the temperature, but the estimates are it’ll be something between the Earth and Mars.

Which makes this another first: this is the first Earth-like exoplanet in the habitable zone. The HZ is the region around a star where liquid water could exist.

We can get a better idea of that from this diagram. This is screen shot from the Exoplanet app. This green area is the calculation of where the temperature could allow liquid water to exist.

Inside the green area, you’re close to the star, and it’s too hot. Outside the green area, it’s too cold. This solid line here is the orbit of Kepler 186-f, right in the middle of the HZ.

You can see the other planets of the system are in too close and will be probably too hot. This outer line is Mercury again, for comparison.

But the star itself is a red dwarf. That’s smaller than the Sun, and a kind of star that occurs frequently. Something 70% of all stars in our galaxy are like this.

And these stars are cooler than the Sun, which is why the HZ is closer in.

For comparison, here’s our own solar system and the same calculation of the HZ.

So note Mercury is close in and, very hot. Venus, just missed out. Earth… kind of on the toasty side of HZ. We probably don’t want to screw with our climate too much.

Mars, according to this diagram, ought to have liquid water. The search is on to find that, maybe it appears in small quantities in the summer, maybe it’s under the soil.

But this goes to show that just being in this HZ doesn’t mean you will have oceans of liquid water. Mars is quite a bit smaller than the Earth, probably couldn’t hold on to an atmosphere… lower gravity, weaker magnetic field to protect it, not enough volcanic activity to pump CO2 into the atmosphere… so it probably lost its water and ended up with a very thin atmosphere.

This is, btw, one of the things I love about exoplanets. I’m just describing the properties of exoplanets, but each planet leads on to hundreds of other questions. Things like: The importance of geology, plate-tectonics, atmospheric science, global warming, erosion, volcanism, magnetic fields… the properties of stars: how old are they, how far away are they, how hot are they…how do planets form… So many topics, and all of these these are not areas I thought I’d be interested in, but they all tied up in understanding exoplanets. And for me, that’s part of the joy of exoplanets, that it touches some many topics.

So bare in mind: being in the HZ doesn’t mean habitable. Mars is in the right place, but is just the wrong size.

What’s exciting about 186-f is that it is in the HZ and Earth sized. And from what we know about the ingredients for like, that’s a promising and exciting combination. The red-dwarf star is good for life in that they live for trillions of years, giving life a chance to get started. On the other hand, they are volatile stars, which can dim dramatically and let of powerful solar flares, which are bad for life.

And we can speculate what it’d be like there.

NASA have a web site called the Exoplanet Travel Bureau. And in this poster, the suggestion is that the planet life on this planet — if there was any — might be influenced by the redder light from the star, and end up looking quiet different from plant colours on Earth.

Exoplanets and speculation go hand in hand.


This planet was detected using the Kepler spacecraft. This is a telescope, trailing behind the Earths in a similar orbit, that went up in 2009.

To give you some scale, a person is about that big.

And what it did is stare at one area of the sky. The telescope has these 21 detectors, and it covered area with lots of stars in it, about 150,000 useful stars. Although they are generally quite far away. Up to 3,000 light years.

The job of this telescope was to find Earth like planets in the HZ, and 186-f was the first confirmed one of these. And it does this staring at this patch, and looking for a dip in the light as a planet passes in front of the star. That’s called a transit. Transits might only last a few hours, so it has to stare for as long as it can… for years.

And it’s been hugely successful: over 2000 confirmed planets, and 3000 unconfirmed candidates that need to be followed up on… and the reckoning is that 90% will turn out to be confirmed planets. So there’s lots more finds in this data.

I’m going to use a simulator to show the basic idea of how transits work.

On the left we have a star and a planet. And on the right we have the measured brightness. Ignore everything else.

I’m going to look at this graph at various points, and see how it corresponds to where the planet is. At first, the planet is not in front of the star, so we see 100% of the light. Then it starts to creep in, and we get dip in the light. As more of the planet moves over we get less light… and then there’s this flat part where the whole of the planet is trundling in front of the star. And then we see the reverse as the planet moves on its orbit.

That’s the basic idea. If the planet is bigger, we see a bigger dip. If it’s smaller we see a smaller dip, or maybe we don’t measure anything at all. We could twiddle more parameters around the size of the star, and have hours of fun.

Of course, that’s an ideal curve. The real world is a bit noisier.

Here’s the plots for the five exoplanets in the Kepler 186 system. And you can see the variation is very noisy, but if you get enough observations, you can piece together the pattern. This is why it’s so important that Kepler stares at the same place for so long, so you can collect this data.

If you want to see how hard this really is… you can have a go.

There’s a web site called Planet Hunters. And this presents you with Kepler data, and asks you to report any transits you see. The nice example they give you shows this nice noisy line of brightness, and these sudden drops that signify a transit. And Planet Hunters have found planets, which is super cool. You can sit at home, and contribute to this search from your laptop.

I’ve tried this, and I find it incredibly hard. The brightness of stars vary in all sorts of ways. I don’t think I’ve been presented with a single example that looks anything like this. They are often wavy, lines because some stars are variable, or suddenly the brightness shoots up. Have a go and see what you think.

Three facts: Transits

As with RV, this graph, this light curve, gives us three pieces of information. This is how we know what we know about this exoplanet.

This graph is the light from a star (it’s not Kepler 186, because I thought this was a clearer example). And you can see light is mostly up here at 100%, and there are these transit dips.

We know the time between the dips is the orbit period — the length of a year. And again, because of the Kepler Laws, that gives us the distance from the star. That’s just the same as for RV.

With transits, this depth here tells us the diameter of the planet.

So we know how big it is, but we don’t know the mass of the planet.

RV gives us the mass; Transit gives us the diameter. So ideally you want both methods.

In the case of Kepler 186-f we do not know the mass of the planet. It’s too far away for a follow up using the technology we have today. But our theory or models of planets tells us that planets that are the size of Earth, are probably rocky.

So that’s Kepler 186-f. An important planet. Starting to look more like home. And telling us that Earth like planets are out there, and they are out there in a place that could support liquid water, and maybe life.

That’s two planets we covered: quite different, with different detection methods. That’s all the hard stuff done with. Let’s move on to the big news…

Prox b

Did anyone see the press conference last week? It was pretty big news. The European Southern Observatory announced a new exoplanet. An Earth-sized exoplanet, in orbit around the closest star to the Sun: Proxima Centauri. Not only that, it’s also in the HZ. And that’s big news because, well, it’s close, so hopefully we’ll throw every instrument we have at it over the next few years. There might be other planets there. We’ll find out soon enough. Keep your eye on this one.

Prox C is bit over 4 light years away, and to put that in context, here are the distances involved. The numbers on this diagram are in astronomical units. Here’s 1, the distance between the Sun and Earth. And then each step goes up 10 times. So about 10 AUs to Saturn, 100 out to where Voyager 1 is, then a fair few hops out to the next star, the Alpha Centauri system.

Alpha Centauri is a system of two stars, and a third orbiting those two called Proxima Centauri which is the closest to our sun.

Here’s the artist impression: Prox b, with Prox C here, and these points, if you can make them out, is the pair of Alpha Centauri A and B. They orbit each other every 80 years. And we think Prox C orbits them.

Prox b was found via the RV method, looking for the wobble of the star. So the mass we know is at least 1.3x Earth, and we don’t know the radius. Not yet. It is presumed to be rocky.

This is same method as the first planet we looked at, 51 Peg. That was a huge planet in comparison, and if you remember the velocity curve, it was showing something like a 50m/s effect on the star. That’s fast, that’s a big wobble. The instruments back then could detect something like 13m/s (30 miles per hour). Prox b is much smaller and makes the star wobble by about 1.4m/s, so you can see how the sensitivity is improving. That’s described as walking pace. The aim is to get that down to 0.1m/s sensitivity.

Like Kepler-186, the star is a red dwarf, so the planet can be in close — orbits every 11 days —and be in the HZ.

Temperature is about -40C. That is, if you were to put a lump of rock there it’d be about -40C. If there was no atmosphere. With an atmosphere, it’d be above zero.

Being close, there’s two very exciting possibilities. One is a space journey within a generation. And if we got there what might it look like? I can’t resist another artist impression…

Another is maybe getting images of the planet. That is also being considered as a possibility.

So… it’s very exciting to have something so close. And as I said, you’re going to hear a lot about this in the next weeks, months, and years. It doesn’t get better than a HZ planet on the closest star to the Sun.

Ridiculous interlude

…Unless there’s one closer.

Have you heard of Planet 9 (or Planet X)? This is a predicted planet in our solar system, but far out way beyond any of the planets we know about.

It’s not been detected, but would explain the orbit of these other minor planets that have been detected. If Planet 9 was out here, it’d need to be 10 times the size of the earth, it’d take 15,000 years to go around the Sun. It might get as close as 200 AU, and as far out as 1200 AU. To go back to a previous diagram, that would put it way out here.

I’m mentioning it because there’s been some work looking at whether it is a captured planet from a passing star. The idea being, much earlier in the history of our Sun, another star would passing close enough for us to be able to steal a giant planet from the orbit of the other star.

Which would, I guess, make it an exoplanet inside our solar system, which would make a bit of a mess of the definition of “exoplanet”.

The research puts this at 1% chance at best. So… if Planet 9 is there, it’s probably not an exoplanet. But it’s fun to speculate what it might mean if it was, as it’s much closer that Prox b.

I have three more exoplanets to show you. And they have some really interesting properties. But we’ll rattle through them… they are food for thought.

Strange orbits and exomoons

The first is HIP 57050b.

What drew me to this planet this eccentric orbit. It’s not circular, and ranges right across the HZ. There are exoplanets with even more extreme orbits, but they end up going too close to the Sun. But this one stays inside the HZ. What might that be like? I had to write an essay on this one, and calculate the temperature, which varies from about -50c to +25c. Which is tough, but that is across an orbit of about 40 days. So maybe not so terrible.

The sky would be amazing: the star itself would go from twice the size the Sun appears to four times in the course of the 41 day year.

But this is a Saturn-mass planet. Probably a gas planet, so you can’t walk on it. But our experience of these kinds of plants in our Solar System is that they have moons. Lots of moons. So another area of exoplanet research is looking for exomoons. That might be a rocky world in orbit around a gas exoplanet in orbit around a distant star. In this case about 36 light years away.

And that would be amazing. You’d have a huge gas giant in view all the time. Plus this star coving a huge area of the sky.

No-one has yet found an exoplanet. But the hunt is on. And that’s why I wanted to show you HIP 57050b.

Ancient systems

Another Kepler discovery for us. Kepler-444 is 117 light years away, and has five rocky planets, all smaller than Earth. But they are all in really close to the star and will be hot. So I’m not so interested in these planets specifically.

But: this star is 11 billion years old. Compare that to our Sun, which is 4.6 billion years old. So this is an ancient system.

There’s been plenty of science fiction talking about super advanced aliens. As dreams or nightmares. And here’s some science fact showing us that planets (probably rocky) do form when the universe was just 20% of its current age. And if there are ancient habitable planets, with life — maybe not this specific system — they could be literally billions of years in advance of us. Providing they haven’t killed themselves. And that’s the trick: can civilisations survive for that sort of time. No-one knows.

Yet another Hot Jupiter?

The last one is another gas giant really close it to the star. We’re back where we began: more or less the same thing as 51 Peg b: 62 light years away, Jupiter sized, 800c, 2 day orbit.

Even this is a first. Detected via transit, but then also detected in x-rays transit for the first time. The same transit method we’ve seen, but with x-rays rather than visible light. That’s the x-ray image up there… you can’t see the planet in it, those other two dots are a nearby star and a object much further away.

The reason I mention this one, is, even in visible light, the transit is big. It’s almost a 3% dip in the light. Which means… we could detect it. And we know that because someone has.

Here on the right is a DSL camera, with a 300 mm telephoto lens. And a home made star tracker… a motor and a timer to move the camera round in time with the earth rotation and keep the star in focus. You connect this up to a laptop to record the light, and using a piece of software you can convert that into a chart showing the brightness you see, against time.

On the left you can make out this dip in the brightness. It’s very noisy, but it’s there.

How cool is that? Being able to go out and record these things ourselves?

Pretty cool I think,

OK, I’ll stop now.

In summary

We’ve seen a bunch of planets, and learned a bit about how we know they are there.

We can speculate about what they could be like to visit or live on.

We can get involved, with planet hunters or even trying to spot a transit ourselves.

And the next few months and years are going to be amazing.

A ton of science is going to be thrown at our closest star. Kepler data still has thousands of planets to reveal.

And hopefully soon, maybe end of 2017, TESS will launch. This is looking for exoplanets across the whole sky, and close to us. That means discoveries can be more easily followed up and studied, maybe looking at planet atmospheres to see if there are signs of life in there.

Thank you.


Keynote presentation and pdf: https://github.com/d6y/Brighton-Astro-Exoplanets