Episode 5: XO XO! - Exoplanets

Abigail James
Mar 2, 2020
16 min read

This blog post is a transcript of Series 1, Episode 5 of my science podcast, Don't touch my Radium! All sources, references and recommended material can be found here.

Our solar system formed when our sun ignited, sweeping microscopic dust particles along a protoplanetary disk allowing for materials to condense and eventually form our planets, moons, asteroids, comets and all other objects contained within the boundary of the Oort cloud. Hubble has spent 30 years searching our cosmos in an effort to learn about our early universe, and gather information on galaxy, stellar and planetary formation. By learning more about how extrasolar systems form, we can learn about potential habitable systems different from our own. It’s hard to make predictions based on our own experience, as this experience is of one habitable planet, in one solar system orbiting one type of star. The Kepler data indicates that there are about 200 billion stars in our galaxy and that each one has on average, at least one planet meaning there are approximately 40 billion exoplanets in their stars habitable region. So how are we going to apply what we know to what is potentially out there? Let’s start by looking at the stars.

Stellar Classification

How do we define a star? A star is defined as any massive self-luminous celestial body of gas that shines by radiation derived from its internal energy sources. This means that it must undergo self-sustaining fusion that will generate enough energy for the thermal pressure to balance gravity. There is a lower limit on the mass that a star must be in order to achieve the necessary internal temperatures for this thermonuclear fusion, and that mass is 80 times the mass of Jupiter. But the stars in our observable Universe have a wide range of masses and it turns out that our star, the sun, is actually pretty normal. A defining characteristic of a star is its luminosity, this is the measure of its ‘intrinsic brightness’, it tells us how much energy is emitted at a moment in time, you can think of it as the ‘power output’. When looking at extrasolar suns we talk about luminosity as a comparison to our sun, so it may have a power of 10 solar luminosities, emitting 10 times more energy than our star.

Bigger doesn’t necessarily mean better, our Sun is of a class of star that will live for maybe 10 billion years before running out of hydrogen, larger mass stars will shine brighter and die sooner. When we study a star we look at both the temperature and the luminosity to determine its class and the Hertzsprung-Russel Diagram to track the stellar evolution over time.

Harvard Spectral Classification:

First we’ll determine it’s spectral class using the Harvard Spectral Classification system. Remember back in episode two when we talked about absorption lines in the colour spectrum and how we can track the emission of radiation by collecting the light that excited electrons send out when they are returning to their rest state. Well if a star has a low temperature, there won’t be that many excited electrons and so the absorption lines for Hydrogen will be pretty weak, but at a medium temperature there’s a lot more energy flying around and so the absorption line will be much stronger but as we get hotter the energy is now so high that they hydrogen is mostly ionized and so there’s hardly any absorption lines visible. So if we track the absorption lines of hydrogen in the spectra of a star we can assign a temperature class to it, this is a letter from the sequence OBAFGKM, going from hottest to coolest.

Note: mnemonic ‘Oh Be A Fine Girl, Kiss Me’.

Within each letter class there are variations in the strength of the absorption line so that they are further broken down into 10 sub-classes from 0 to 9.

Our sun is a white-yellow star with temperature in the range of 5,000 to 6,000K and falls under the spectral class G with absorption line strength 2. So under the Harvard Spectral Classification, it is a G2 star.

Morgan-Keenan Luminosity Class:

Next we need to determine the luminosity class. Stars can have the same temperature while having different luminosities and the spectral class cannot distinguish between these. So we use the Morgan-Keenan Luminosity Class. A roman numeral or a letter is assigned to a star depending on the region of luminoisty it falls under, for example a luminous supergiant is assigned Ia while a ‘normal’ giant is assigned roman numerals III. Our sun is a main sequence dwarf star and so it is assigned the roman numeral V. A stars classification will change as it evolves over time.

Hertzsprung-Russel Diagram:

All stars will go through evolutionary stages that depend on its original mass, internal structure and the process by which it produces energy. To look at how a star lives and dies, the HR diagram plots the temperature against the luminosity and track the changes that these values will undergo throughout the stars life.

The power of all this information means that we can gather the light from extrasolar stars and determine not only what their size, temperature and luminosity is but also what it’s future might hold, and given the time it takes for the light to reach us from some of these stars, that future may actually be its present.

When we look for exoplanets, we are looking for planets orbiting an extrasolar main sequence star, one that burns hydrogen like ours. We want to observe stars that are smaller than our sun normally in the M class of the Harvard Spectral Classification, (kiss me, stars on the coolest end) these are stars that are red dwarfs and have temperatures below 3500K such as Proxima Centauri. These star types are smaller than our sun and the most common star types in our galaxy, their light is fainter so we need larger telescopes to pick up their light signatures but it is also easier to distinguish a nearby planet than looking at stars in a class similar to our own or larger. That’s star classification, what about exoplanets?

Planet Classification

There has been some controversy over the official classification of a planet. The International Astronomical Union (IAU), state that a planet is classified as ‘a celestial body that is in orbit around a sun, has sufficient mass to have a nearly round shape, it is not a satellite of another object and it has cleared away any other objects of a similar size around its orbit’, sorry Pluto! However there are arguments about this definition as it can be said that other planets in our solar system have not cleared their orbit either, Earth included, although to a much smaller magnitude than Pluto.

Based on our own solar system there are four types of planets: terrestrial (rocky), gas giants, ice giants and dwarf planets. Taking the IAU definition, there are 8 planets in our solar system. Four inner rocky planets; Mercury, Venus, Earth and Mars. Two outer gas giants: Jupiter and Saturn and two outer ice giants: Uranus and Neptune. Objects like Pluto, Eris and Makemake are all dwarf planets. For an exoplanet, there are proposals for classification schemes but as of yet there doesn’t seem to be any universally used system in place beyond the four types found in our solar system so we’ll find that a lot of the work will refer to an exoplanet mostly by its comparison to Earth or Jupiter.

The Centre for Astrophysics have recently come up with a new classification system for exoplanets based on data gathered from the Gaia mission. The classification goes as:

First: Small, gas poor planets: Less than 2 Earth-radii

Between 2 and 4 Earth-radii

Ones most likely to have water rich cores

Second: Transitional planets: Between 4 and 10 Earth-radii

Third: Gas giant planets: Larger than 10 Earth-radii

Dominated by hydrogen and helium, including Jupiter analogs and brown dwarf stars.

But classifying an exoplanet by size isn’t enough to tell us about potential life signs. We can see that while life arose on Earth, it is not abundant throughout our solar system and so there is something special about our planet. It turns out that there is a ‘habitable’ zone which is a specific region around a star where a planet could sustain the necessary conditions for life.

Habitable Zone

Investigations into exoplanets have shown that there is an abundance of terrestrial planets out there, in fact, there are more terrestrial ones in our galaxy than there are gas giants. So we apply some criteria that allows us to further classify exoplanets by their potential habitability. We define a suitable orbital zone around an extrasolar star whereby there will be a balance between the amount of stellar radiation and the radiative cooling from the planet, this allows for liquid water to exist on the surface which is the main criteria. It’s sometimes referred to as the ‘Goldilocks’ zone, where the planet is ‘just right’ rather than in the too hot or too cold regions. Secondly this zone presumes that there will be a solid surface and an atmosphere present. This is what we expect of a planet in its habitable zone, but how do we define this region?

Surely it’s not just how far it is from the star? You’d be right! The criteria put in place by the Harvard-Smithsonian Centre for Astrophysics are:

Incident Stellar Flux and its Harvard Spectral Classification

The flux is the amount of the radiative energy that reaches the surface.

The planets eccentricity

This is a way to describe how the planets distance changes during orbit, circular would have an eccentricity of zero while higher numbers show more elliptical orbits.

The planet's reflectivity

This relates to the amount of starlight reflected by the planet and includes the effects of partial cloud cover

Details of the planets atmosphere including:
Greenhouse gas concentration

Greenhouse gas concentration

A greenhouse gas is a gas that traps heat in the atmosphere, how it does this is that it absorbs energy from the infrared region of the em spectrum but it also emits energy in this range. The IR range as we remember, is a version of heat. For Earth, greenhouse gases are: water vapour (H2O), carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O) and ozone (O3). These gases have pros and cons, the pros are that they keep the surface warm enough for life to survive, without them surface temperatures would be around -18 degrees celsius! Instead the atmosphere keeps us at a temperature that keeps our climate primed to support life in many forms. But the cons, are that our industrial and technological advancements lead to manufactured contributes of more gases to the atmosphere from various emissions, causing more heat to be trapped in the atmosphere and warming the surface further with the potential to change this climate and limit the planets capability to support life.

We’ll come back to these gases when we talk about biosignatures in more detail.

The Earth is actually in the middle of our solar system's habitable zone. So how often does this occur? Well, before you get too excited, in 2017 the Harvard-Smithsonian Centre for Astrophysics catalogued the data from Kepler leading to an estimation of the number of exoplanets within their habitable zone which are also within the size range of less than two Earth-radii, there are 20. Now this might have changed in recent years, but given the amount of potential out there, this seems like a really small number of possible Earth-like candidates. If you extend the Habitable zone to a less conservative or what’s referred to as an ‘optimistic’ zone as well as allowing for planets with any radii, this number bumps up to 104. Still pretty small given that we’re expecting billions of exoplanets!

So we’ve classified our star, we’ve categorised exoplanets and set an orbital region to search within. We’ve talked briefly about looking at an exoplanets atmosphere through spectroscopy but what about finding the planet in the first place? How do we actually know when something is there? Let’s take a look at the detection methods used to discover these alien worlds!

Detection Methods

There are five main methods of detection when it comes to searching for exoplanets: radial velocity, transit, direct imaging, gravitational microlensing and astrometry. Each method will use a different technique and so it will gather different information about the object it is observing.

Radial Velocity

This is an indirect method that is normally described as a ‘wobble’ and is detected using doppler spectroscopy. Out of the confirmed exoplanets that have been discovered so far, nearly 800 have been detected using radial velocity. So much of what we discuss is related to the effects of gravity, from our solar and planetary formation to the tidal locking of the moon and the way planets remain in orbit around their star. As we know, each object has a gravitational pull, the more massive the object then the larger the gravitational pull. Taking our solar system as an example, the Sun holds us in place due to its large gravitational pull, but we also have a gravitational effect on the sun a tugging of our own, however given the differences in size; the sun's effect on us is our orbit, our effect on the sun, may not be noticeable. Well it may not be noticeable except for a tiny ‘wobble’ effect, the gravity of the planet is causing the star to move around in space, very slightly almost on it’s own tiny little orbit. A bigger planet will have a larger effect, Earth has a tiny effect whereas Jupiter will have a stronger one.

In order to detect a ‘wobble’, astronomers focus on the Doppler shift of the energy radiating through space. We’ve talked about the em spectrum and that stars emit all forms of radiation as light, the wavelengths of this spectrum can actually be stretched and squeezed depending on the movement of the star they are emitted from. Basically what happens is that when an object that emits waves moves closer to an observer, the waves become a bit more squashed together, whereas when it moves away from the observer, they stretch out. So if you remember Newton's prism and that when he used it to refract white light it showed all the colours of the rainbow, this occurred as each colour has a different wavelength. So when a star is moving away from us, the visible light we collect will appear more red, whereas as it moves towards us the light will appear more blue. This colour change is referred to as ‘redshift’ or ‘blueshift’ and can be tracked by using a spectrometer to look at the way the spectral lines are displaced indicating if there is a wobble present.

And it's not just about the wobble of the star. If we can get a spectral line for the planet this can be used to distinguish the planets radial velocity, which allows for the inclination of the orbit to be determined leading to a measurement of the mass of the planet! It’s like magic but science! Being able to figure this out helps to establish that there is a planet present and so the wobble isn’t a false positive.

We’re mostly talking about M class stars, red dwarfs as these class of stars are more affected by the gravitational tug of planets and also, they rotate more slowly so we can get clearer spectral lines. Despite this method being distance independent, it does require a high signal-to-noise ratio spectra in order to be considered precise and it adds a distance limitation to the region we investigate. It can detect gas giants, the third class of planets, up to thousands of light years away but when it comes to the first and second class, the Earth-like planets, this method is only used to look at low-mass planets in the range of 160 light years from us.

Another limitation is that you can’t observe a lot of stars simultaneously with a single telescope which is becoming more of an important feature in modern instruments. On top of this, multi-planet and multi-star systems coupled with insufficient data from short observations can lead to false signals coming through.

So what do we do? Well couple the method with another one, maybe the Transit method.

Transit

The transit method, referred to as transit photometry, is the most used in the hunt for exoplanets, or maybe its that it’s the most successful. To date this method has found over 3000 planets orbiting extrasolar stars. The method is really very simple, you have a bright star and as a planet passes between the star and your telescope, the light will be dimmed by an amount that can then be measured. Think about when there is a solar eclipse, the moon passes between us and the sun and blocks the light, we can observe it clearly. By measuring the light reaching us from a star we can trace a light curve and the drop in brightness will indicate that something has passed in front of the star, between us and the light. Once this event repeatedly occurs with the same period and for the same amount of time, it can be determined to be caused by an object in orbit. A larger drop in brightness will indicate a bigger orbiting planet and distance from the star is determined by the length of orbit so we can tell how far away the planet is. Given that the transit method can determine the size of the planet and that the radial velocity method can determine its mass, if we combine the two we can figure out its density! Also, when the planet travels behind its star, we can measure the luminous intensity and use this to determine what temperature the planet might be! It can even determine the presence of cloud formations in the atmosphere, although I have no idea how this is achieved. This simple method can tell us a lot. Transit method is actually the method used by Kepler during its mission from 2009 - 2013. One thing to look out for though, a system with multiple planets will have a more complicated light curve and so requires a bit more work to find the different planets.

One of the most important aspects of this method, from this series point of view, is that it can provide us with information about the planets atmosphere and its temperature. We’ve mentioned in a previous episode that when starlight travels through an atmosphere, some molecules will absorb or reflect the light in such a way that we can analyse the spectral lines and identify ones that correspond to specific energy transitions for some molecule/ion/element so that we can learn about what types of molecules are held in the planet's atmosphere.

The drawbacks of this method comes down to the need for perfect alignment, you can only observe a transit if the orbit is perfectly aligned with the observer's line of sight. This happens with only 10% of short orbital periods and gets less and less as orbital periods increase. For this reason, this method alone cannot actually guarantee that there are any planets orbiting a star under observation. But it is effective for surveillance of thousands of stars at a time and so is useful as primary detection but further confirmation should be done by a different method.

Direct Imaging

Direct imaging is pretty much what it sounds like, taking pictures, and to be honest before I started researching this topic I honestly believed that this is the main way we knew about exoplanets. But surprisingly, or not as we’ve learned how difficult it is to take a picture of an object so far away, it only accounts for about 50 discoveries so far but it is still really early on in its life as a detection method. In order to take a picture of an exoplanet we need to remove the glare from the star that it orbits as this light overwhelms the light emitted by the planet. We can then look for the infrared light that is emitted by the exoplanet, infrared is easier to separate than visible as the ratio of intensity is smaller. We talked about this last week when we introduced the use of the coronagraph as an instrument used in telescopes, where it blocks the light from the star using a mask to simulate an eclipse before it reaches the detector but as the light from the exoplanet enters from a different angle it passes through and reaches the detector allowing for the light to be separated.

Coronagraphs aren’t the only method that can be used to block the light, the second method is referred to as ‘starshade’ and it’s a separate space craft that works in conjunction with a space based telescope. The star shade, kind of like a parasol, would position itself at a distance and angle between the telescope and the star blocking the light before it enters the telescope.

The potential for direct observation coupled with spectrographic information can allow astronomers to learn about the composition of the atmosphere and determine potential habitability. It’s also possible to learn about the planets interaction with the stars protoplanetary disk as well as calculating an estimate for a planet's mass. As this method is direct, it doesn’t require a second method to confirm the results but at its current early stages, it works best for wide orbits and gas giants so is not ideal when searching for Earth-like terrestrial planets in the habitable zone. However the main limitation comes in the difficulty to obscure the light from the star so that the planet can be observed so really it’s limited by current technology.

Gravitational Microlensing

This requires some understanding of the curvature of space-time, and since I chose not to take the general relativity course at uni, let’s stick with how NASA explains this method: gravity is a geometric property of spacetime with large objects warping the fabric of space, this causes a distortion in the light being emitted resulting in a direction change when being influenced by the gravity of an object with significant mass. This means that the light from a distant star can be bent and focused by gravity as another star system passes in front, making the background star appear brighter.

Over 80 exoplanets have been observed using this method but it requires surveys of large parts of the sky for long time periods which might seem like a disadvantage, but it is the only one of the detection methods that can find small exoplanets at great distances. This method is possibly the most effective at finding Earth-like planets around sun-like stars. And this way of surveying can be done using ground-based telescopes, watching tens of thousands of stars simultaneously.

However, we’re not talking about planets orbiting their star, we are using background stars from other systems as references, meaning that these events are unique and do not repeat. Once an observation has occurred, this method cannot observe that planet again. Also, as we are looking at great distances, follow up observations are not possible. However, there is a very high signal-to-noise ratio with microlensing so the precision is pretty high and follow-up surveys are not really necessary to confirm the presence of an exoplanet. Despite this, it depends on rare and random events namely that it requires one star to pass precisely in front of another so detections are unpredictable.

What is interesting though, is that quick microlensing events will occur as a result of free-floating planets, so this method can tell us about how common ‘rogue’ planets are in our galaxy.

Astrometry

Astrometry is our final detection method and I think it’s safe to say that it is the most difficult one to achieve. It’s similar to the radial velocity method in that we’re looking for a ‘wobble’ but rather than look at the doppler shift, it’s done by imaging a group of stars overtime and looking for shifts, wobble, in a target stars position relative to the other stars. This method requires extremely precise optics and so attempts from ground based telescopes would be too difficult given atmospheric distortions. Despite being hailed as one of the most sensitive methods to exoplanet detection as well as the oldest detection method, as of today it has only been credited with accurate detection of 1 exoplanet. When they say it’s hard to do, they mean it.

Each method has advantages and disadvantages, and each has telescopes designed to search and report information that all contributes to our understanding of the stars and planets in our galaxy. From here on we’ll only be talking about the detection and investigation of exoplanets using radial velocity, transit and direct imaging. Or most likely, a combination of these methods. Now that we have figured out what stars to look for, decided on what methods to use, and gone out and found us a bunch of planets in their habitable zone, it’s probably time to figure out if there’s life on other planets?!

Still to come this series, how Kepler found so many exoplanets, what exactly is a biosignature, what about life in our solar system?, how do we send probes out to navigate space, what is the future of space telescopes and are we even looking for the right signs?

Thanks for listening. Come back next week if you’d like to learn more. If you want to get in contact you can email dtmyrad@gmail.com or drop by our instagram @dtmyrad.

Articles