A weather report for the hot Jupiter HD 189733b II.

The weather forecast of HD 189733b is further analyzed by the LEAP team, this time led by Graham Lee. They report further cloud coverage and dynamics in their newly published paper in Astronomy and Astrophysics (Lee et al. 2017).  Learn more about their results from Graham’s summary below.

In part I of this series, we coupled a radiative-hydrodynamic (RHD) atmospheric model to a microphysical cloud formation module. We simulated the atmosphere of the hot Jupiter HD 189733b and found that the planet is likely to host a highly inhomogeneous cloud structure dependent on latitude, longitude and depth of the atmosphere. In our next paper in the series, we compare our results to available Hubble Space Telescope and Spitzer Space Telescope observations of the exoplanet. We do this by post-processing our large scale cloudy RHD results using the Monte Carlo radiative-transfer (MCRT) method.

Figure 1 Images of the simulated planet at 1 micron wavelengths and at a planetary phase of 60o (eastern terminator region). Left: Incident stellar scattered light. Right: Emitted thermal light. The cloud structures are more prominent in the left image and a twilight effect (scattering past the day-night terminator) is present. In the right image the hotter equatorial jet is very recognizable compared to the cooler higher latitude regions and the nightside of the planet. (Lee et al. 2017)

Monte Carlo radiative-transfer in a nutshell

A particular method for radiative-transfer is the Monte Carlo radiative-transfer (MCRT) technique. A MCRT simulation is a 3D stochastic, statistical sampling method of determining the local radiation field in an environment. We track the 3D random walk (or path) of millions of simulated “photon packets” through our 3D RHD simulation results. We do this by sampling a random number which govern the packets interactions with the surrounding medium. For example, if a photon packet undergoes a scattering event with a cloud particle, the new propagation direction of the photon packet is determined by sampling a random number and comparing that number to the “phase function”, which is the probability of the packet being scattered at any one angle (e.g. a random number of 0.1 could correspond to a new path slanted 10 degrees from the original direction). By simulating the path and interactions (scattering, absorption) of many millions of packets, we can count the total amount of energy escaping towards a certain direction and produce synthetic observable properties of our large scale atmospheric model. Figure 1 shows an example output from the MCRT simulations applied to our RHD model.

Figure 2  Albedo of our simulated planet compared to the observational data obtained by Evans et al. (2013). Our model is reasonably consistent with the B-band data but does not predict the drop in albedo at the V-band. Our simulation is able to decompose the effect of the different scattering material present in the atmosphere. Most scattering comes from cloud particles with a small H2 Rayleigh scattering component. (Lee et al. 2017)

Albedo/reflection spectra of HD 189733b simulation

When a transiting planet passes behind its host star, both the emitted light from the planetary atmosphere and reflected incident starlight disappear from the total luminosity of the star and planet system, resulting in a dip in observed flux of the system during the secondary eclipse of the transit. The albedo of a planet is defined as the fraction of incident starlight that is scattered back into space. For example, a planet with an albedo of 0.2 at a wavelength of 300 nm, returns 20 % of the incident star light at 300 nm back into space. The albedos of many hot Jupiter planets have been measured by the Kepler space telescope, and some by the Hubble Space Telescope. Figure 2 shows the albedo of our HD 189733b simulation produced from the MCRT simulation compared to the geometric albedo observations using the Hubble Space Telescope by Evans et al. (2013). We also combine our albedo and emission spectra to compare to dayside observations of HD 18973b from Hubble and Spitzer Space Telescopes (Fig. 3).

Figure 3  Dayside luminosity ratio of our simulated planet compared to HST and Spitzer measurements (Evans et al. 2013, Crouzet et al. 2014, Barstow et al. 2014 and Knutson et al. 2012). Our simulation compares well to the observed trends in the dayside emission spectra. (Lee et al. 2017)

We find that our model is generally consistent with the observations but has some offsets with some data points. However, we do not fit the “V-band” (Visible wavelength range) dip in albedo as observed by the HST measurement. This suggests the presence of an absorbing material present in the atmosphere of HD 189733b which was not accounted for in our modelling. This absorber has been noticed in other hot Jupiter planets and it is currently a mystery what causes these dips in albedo at these wavelengths. Overall though, this general agreement with the observations suggests that the cloud coverage and temperature structures of our large scale simulation are probably a reasonable expectation of the dayside properties of HD 189733b.

Figure 4  TESS and CHEOPS predicted dayside albedos and bandpasses (dashed line). CHEOPS has a larger albedo due to the effect of photons scattered by clouds at optical wavelengths, which TESS is not sensitive too. (Lee et al. 2017)

Predictions for TESS and CHEOPS

We also produce predictions of the geometric albedo and phase curve behaviour of HD 189733b for the upcoming NASA Transiting Exoplanet Space Survey (TESS) and ESA CHaracterising ExOPlanets Satellite (CHEOPS). Figure 4. shows our results shows our TESS and CHEOPS predictions, and includes a Kepler prediction (even though HD 189733b will not be observed by Kepler) for comparisons. We found that our simulated dayside atmosphere would be 10% brighter when viewed by the CHEOPS mission compared to the TESS instruments. This is due to the fact that CHEOPS is more sensitive to the optical scattering component (cloud cover), as TESS is more sensitive to infra-red thermal emission from the planet. By comparing the albedos and phase curves between the CHEOPS and TESS instruments, an idea of the extent of the cloud coverage on the observed planet may be inferred. This may be useful for helping decide which planets to observe with more time limited spectroscopic instruments such as the James Webb Space Telescope (JWST).

Figure 5  RGB channel GIF of the incident scattered light varying with planetary phase. The real planet is probably slightly bluer than the brownish colour presented here due to an unknown absorber at visible wavelengths (See Fig. 2; Lee et al. 2017).

RGB channel phase curves

Lastly, we produced a RGB channel phase curve output using the albedo results. This is provided in GIF form in Fig. 5. The real planet is likely to be slightly bluer than the reddish brown colour presented due to the unknown V-band reducing the R channel’s magnitudes.

For more details check out the original paper on ADS:

G. Lee, K. Wood, I. Dobbs-Dixon, A. Rice, Ch. Helling, 2017, A&A, 601, A22

The LEAP Group can be found here:

http://leap2010.wp.st-andrews.ac.uk/

And finally, don’t forget to like us on Facebook:

https://www.facebook.com/leap2010

A weather report for the hot Jupiter HD 189733b I.

One of the most prominent and visual features of weather phenomena on Earth are water clouds which govern a large part of Earth’s climate. In Lee et al. (2016), the first of a series of papers, we investigated what the atmosphere of the hot Jupiter exoplanet HD 189733b may look like, in particular what the properties of clouds that maybe present in the atmosphere are. We used a computer simulation of the atmosphere of the hot Jupiter exoplanet HD 189733b including, for the first time, a microphysical description of 3D cloud particle behaviour in the atmosphere.

Simulating a cloudy exoplanet atmosphere.

A major tool for the exploration of the properties of Earth, Solar System and Exoplanet atmospheres are the use of Radiative-Hydrodynamic (RHD) or Global Circulation Models (GCM). These models simulate the temperature, pressure and other important properties of an atmosphere by evolving the equations of geophysical fluid dynamics, representing the large scale motions or “winds” of the atmosphere. The motions of the atmosphere depend on numerous factors such as the rotation rate, which affects the fluid flows primarily through the Coriolis force; pressure gradients, causing flows of gas. These models also take into account the heating from an external source onto the atmosphere, primarily radiation from the host star of the exoplanet. For hot Jupiter atmospheres, the intense effect of heating on dayside of the planet has large consequences on the specific weather properties, driven primarily from the large temperature differences between dayside and nightside regions of the planet.

The number density of cloud particles at 1 mbar pressure in the HD 189733b simulation. A build up of material at the equator near 270 degrees longitude is seen. Arrows show the direction and strength of the atmospheric flow. The dynamics is dominated by the strong super sonic equatorial jet which efficiently transports cloud particles around the globe.

Figure 1. The number density of cloud particles at 1 mbar pressure in the HD 189733b simulation. A build up of material at the equator near 270 degrees longitude is seen. Arrows show the direction and strength of the atmospheric flow. The dynamics is dominated by the strong super sonic equatorial jet which efficiently transports cloud particles around the globe (Lee et al. 2016).

In our study we used the RHD model from Dobbs-Dixon & Agol (2013) of the hot Jupiter HD 189733b and coupled it to our microphysical based cloud formation model, the details of which can be found in our numerous other blog posts (here and here). We found that our simulated cloud structures varied significantly in latitude, longitude and depth of the atmosphere. Cloud particles were more numerous of the nightside of the planet compared to the dayside regions, primarily due to the fast winds flowing from the dayside to the nightside slowing down as they approached the western terminator region. This creates a “traffic jam” effect, where cloud particles pile up on the western terminator nightside (Fig. 1).

The average cloud particle size in micron at 10 mbar pressure in the HD 189733b simulation. A large difference in cloud particle sizes between dayside and nightside as well as latitude is apparent.

Figure 2. The average cloud particle size in micron at 10 mbar pressure in the HD 189733b simulation. A large difference in cloud particle sizes between dayside and nightside as well as latitude is apparent (Lee et al. 2016).

Cloud particle sizes were also very different from the dayside to the nightside. On the dayside, the hotter temperatures evaporate the more volatile Magnesium containing materials MgSiO3 and Mg2SiO4, while these materials are stable on the nightside of the planet. This leads to different particles sizes, smaller particles of ~nm sizes on much of the dayside and ~μm sized particles on the nightside (Fig. 2).

Latitudinal differences in cloud particle size are also seen, with the hot equatorial super-sonic jet containing no Mg containing materials, and only the very stable TiO2 and SiO2 minerals. Higher latitudes, where it is cooler, contained ~90%+ Mg bearing materials. In our simulations we see bands of different materials dependent on the atmospheric properties, similar to how the bands of different materials on Jupiter colour different bands.

A collage of atmospheric properties at the equator of the simulated planet. The radial numbers denote the atmospheric pressure in log_10 bar. Numbers on the outside of the ring denote longitude. Top row: Gas temperature, cloud particle number density, cloud particle mean radius. Middle: TiO2 abundance, SiO abundance, SiO2 abundance. Bottom: MgSiO3 abundance, Mg2SiO4 abundance. Clear differences in the cloud properties can be seen in longitude and depth. For example, the dayside of the planet contains a large region of TiO2, while the nightside has more silicate materials present (Lee et al. 2016).

Figure 3. A collage of atmospheric properties at the equator of the simulated planet. The radial numbers denote the atmospheric pressure in log_10 bar. Numbers on the outside of the ring denote longitude. Top row: Gas temperature, cloud particle number density, cloud particle mean radius. Middle: TiO2 abundance, SiO abundance, SiO2 abundance. Bottom: MgSiO3 abundance, Mg2SiO4 abundance. Clear differences in the cloud properties can be seen in longitude and depth. For example, the dayside of the planet contains a large region of TiO2, while the nightside has more silicate materials present (Lee et al. 2016).

Overall, our results show that we can expect very cloudy, dynamic weather on HD 189733b, with cloud particles evaporating and condensing constantly as they are pushed along with the winds from dayside to nightside. However, unlike the temperate conditions on Earth, these cloud particles would be travelling at super-sonic speeds, dwarfing by magnitudes the most violent hail storms found on Earth.

In part II of our weather report, we will look at simulating the observable properties of our RHD modelling results and comparing our modelling efforts to Hubble Space Telescope and Spitzer Space Telescope data of HD 189733b.

For more details check out the original paper on ADS:

G. Lee, I. Dobbs-Dixon, Ch. Helling, K. Bognar, P. Woitke, 2016, A&A, 594, A48

The LEAP Group can be found here:

http://leap2010.wp.st-andrews.ac.uk/

And finally, don’t forget to like us on Facebook:

https://www.facebook.com/leap2010

Carbon planets and diamond clouds

Could diamond planets exist? Probably not, but what about carbon-planets in more general terms? So far, almost all discovered planets were oxygen-rich. The quest about carbon-rich planets touches fundamental question of astrophysics, including how stars formed in the early universe and how planets form in disks around young stars. Ultimately and somewhat surprisingly perhaps, we are required to understand what kind of clouds form in atmospheres of carbon-rich planets in order to disentangle existing observations or to predict possible observations, because clouds do act like an optical and a thermal blanket to the planet below it. Therefore, Helling, Tootill, Woitke & Lee (2016) set out to model cloud formation in carbon-rich atmospheres. They find that the material composition and, hence, the appearance of carbon-rich planets (possibly WASP-12b or HR 8799 b, c, d, e) drastically differs from the well-studied oxygen-rich giant gas planets (like HD 189733b or HD 209458b). They show, that only the very inner parts of a carbon-cloud could possibly be made of diamond crystals, making it a rather darkish type of planet.

Stars come with all sorts of chemical ‘tastes’: Those with more oxygen than carbon (called ‘oxygen-rich’), and those with more carbon than oxygen (called ‘carbon-rich’). Stars in the present universe are born as oxygen-rich and only change to carbon-rich at later stages in their life-time when they become giant stars with diameters of hundreds of stellar radii. Therefore, extrasolar planets are expected to be oxygen-rich.

Observations start to indicate that some extrasolar planets maybe carbon-rich, or at least have enough carbon-molecules in their atmospheres so that our telescopes can see them. For example, WASP-12b was suggested to be carbon- rich (Madhusudhan et al. 2011) but a second observation could not confirm that finding (Kreidberg et al. 2015). Four planets of the planetary system HR 8799 have tentative detections of C2H2, CH4, and CO2, all rather large carbon-binding molecules (Opennheimer et al. 2013).

Planets only form in disks around young stars and young stars are oxygen-rich. Stars only turn carbon-rich at later ages. So, why should carbon-rich planets exist at all?

Figure 1: The top panels show the material composition that we predict from our cloud models in per cent: 1.0 means that the cloud particles are made to 100% of this material (e.g. Al2O3 – light blue, carbon – grey) and 0.2 means that the cloud particles contain 20% of that material (e.g. Mg2SiO4 or MgSiO3 – yellow/orange). Different lines represent different materials. The lower pictures visualize how the cloud would look like if we were able to observe it more closely: An oxygen-rich giant planet (left) would sparkle in different colours while a carbon-rich planet would look rather dull and dark-grey.

Figure 1: The top panels show the material composition that we predict from our cloud models in per cent: 1.0 means that the cloud particles are made to 100% of this material (e.g. Al2O3 – light blue, carbon – grey) and 0.2 means that the cloud particles contain 20% of that material (e.g. Mg2SiO4 or MgSiO3 – yellow/orange). Different lines represent different materials. The lower pictures visualize how the cloud would look like if we were able to observe it more closely: An oxygen-rich giant planet (left) would sparkle in different colours while a carbon-rich planet would look rather dull and dark-grey (Helling et al. 2016).

Here are three ideas how this could work:

1) Extremely carbon-rich stars formed in very early universe (Mashian & Loeb 2016). These carbon-rich stars were young at some point and may therefore have formed carbon-rich planets. The problem with that hypothesis is that our Sun is oxygen-rich and not that old. The Sun is a young star compared to the stars we know of in our present universe. It is therefore highly unlikely that any presently observable carbon-rich planet has formed in the early universe.

2) Carbon-rich stars in our present-day universe bath their originally oxygen- rich planets in a strong wind that carries large amounts of carbon-rich gas and dust. As this wind continues to hit the planets atmosphere, it increases slowly but steady its carbon content and appear as carbon-rich at some point. The obvious conclusion is that all carbon-rich stars with strong mass losses should have carbon-rich planets. This has not been observed so far.

What are we left with? We need to take a closer look at the star and planet formation processes and see whether chemical niches appear that allow us to explain the possible emergence of carbon-rich planets or episodes of carbon- enrichment. In other words, we need to study how the local chemistry changes in planet-forming disks that emerge around young stars in order to conserve the angular momentum of the collapsing gas-cloud that will form a young star. This local chemistry will play a large role in determining what kind of planets will form in such disks. Hence, the beauty lies in the detail.

Knowing about the problem, we have merged our expertise in atmosphere modelling (Christiane Helling) and planet-forming disk modelling (Peter Woitke). Through this work, published in the Life journal in 2014, we realized that we need to understand cloud formation a lot better in order to be able to say if the chemical composition of a planetary atmosphere would be carbon-rich or oxygen-rich.

In Helling, Tootill, Woitke & Lee 2016 we worked out a model that would allow us to predict how clouds form in carbon- rich atmospheres. The oxygen-rich case was well studied already and we could build on our experiences with it.

Figure 1 visualises our main results. The left-hand panel shows the results for the cloud layer that forms in an oxygen-rich atmosphere of a planet, the right-hand side shows the cloud forming in the carbon-rich planet. Here are the three main observations from Figure 1:

a) Clouds in oxygen-rich atmospheres are made of 40% silicates like Mg2SiO4 and MgSiO3, 20% SiO, 10% iron, with the remaining 30% being a mix of other materials.

b) Clouds in carbon-rich planets would be made of a minimum of 60% carbon with the remaining 40% (or less!) being a mix of other carbon-binding materials and iron.

c) While silicate clouds in oxygen-rich atmospheres are semi-transparent at the top and dark further inside, carbon-clouds are graphite-black at the top and diamond-clear further inside.

It appears that an oxygen-rich planet with strong winds should be much more beautiful to observe because many mineral cloud particles could appear in many colours. Carbon-rich planets would look rather dull and grey in comparison.

For more details check out the original paper on ADS:

Helling, Tootill, Woitke & Lee 2016, eprint arXiv:1612.01863

The LEAP Group can be found here:

http://leap2010.wp.st-andrews.ac.uk/

And finally, don’t forget to like us on Facebook:

https://www.facebook.com/leap2010

EXO-LIGHTNING Part I: what can we learn from the Solar System?

Is lightning a phenomenon only occurring on Earth? Or is it universal? How can the knowledge we learnt from Solar System lightning help with discovering lightning on exoplanets and to understand these very different worlds? The next two entries will be devoted to the work on extraterrestrial lightning carried out by Gabriella Hodosán, LEAP PhD student, under the supervision of LEAP PI Dr Christiane Helling, in collaboration with various LEAP group members. In Part I we apply results of lightning surveys of several Solar System planets, including Earth, to different groups of extrasolar planets. Part II will be about a specific planet, HAT-P-11b and the possibility of lightning detection in its atmosphere.

Lightning is one of the most spectacular phenomena on Earth. It has interested not just scientists but the general public for thousands of years. However, it is not a unique phenomenon to Earth. It has been observed before on several Solar System planets, such as Jupiter, Saturn, or Uranus and Neptune. Spacecraft like Cassini, Galileo, New Horizons or the Voyagers provided us with breath-taking images of the outer Solar System, including images and measurements of lightning occurring on the gas giant planets (Fig. 1).

Lightning on Jupier (right top and bottom) and Saturn (left bottom). (Credit: NASA/JPL-Caltech/SSI, NASA/Galileo)

Figure 1. Lightning on Jupiter (right top and bottom) and Saturn (left bottom). (Credit: NASA/JPL-Caltech/SSI, NASA/Galileo)

Since the late 90s, thousands of exoplanets have been discovered. These exoplanets show a large diversity (Fig. 2) in sizes, masses, even distances to the host star, much different to our Solar System planets: Jupiter-size planets orbiting other stars at the distance of Mercury; planetary systems with several planets inside the orbit of Mercury; terrestrial planets several times bigger than Earth, but still rocky and not made of gas. Could these planets host lightning in their atmospheres? Let’s look at Earth and Saturn. They have different composition, different sizes, masses, different atmospheres. And still, they both show lightning activity. So why couldn’t it occur on exoplanets?

Diversity of exoplanets and brown dwarfs on a mass-radius and distance-density plot (Hodosán et al. 2016).

Figure 2. Diversity of exoplanets and brown dwarfs on a mass-radius and distance-density plot. The lines on the top plot indicate the potential chemical composition of the bodies based on their obtained mass and radius  (Hodosán et al. 2016).

In our work, we were focusing on the statistical side of lightning occurrence on Solar System planets, then extrapolated to extrasolar objects. Lightning climatology explores the spacial and temporal distribution of lightning. It uses the quantity of flash rate [flashes/unit time, e.g. flashes/hour] or flash density [flashes/unit time/unit surface area, e.g. flashes/hour/km2] to quantify this distribution. It is a tool to estimate the lightning activity on the surface of an object. This is important in order to estimate the total energy released of lightning flashes, and to determine whether the signatures produced by lightning would be observable from Earth.

Earth

Lightning observing networks net the whole surface of the Earth and satellites continuously look for lightning flashes from near Earth orbits. Lightning monitoring is important because of the hazards (e.g. forest fires, large scale power outage, fatalities) it causes. Measurements from Earth provide the largest data sets we can work with. In our study we analyzed data from the Lightning Imaging Sensor (LIS)/Optical Transient Detector (OTD), which are optical instruments on board of satellites, and from two ground based radio networks, the Sferics Timing and Ranging Network (STARNET) and World Wide Lightning Location Network (WWLLN). Figure 3 shows an example of lightning climatology maps produced from LIS/OTD data. It shows an average of lightning occurrence over the period of 1995-2013. It shows clear trends of more lightning over continents than over oceans, and more lightning over lower latitude regions than higher latitude regions.

Figure 3. LIS/OTD lightning climatology map averaged from 1995-2013 (Hodosán et al. 2016).

Figure 3. LIS/OTD lightning climatology map averaged from 1995-2013 (Hodosán et al. 2016).

Lightning occurs not only in thunderclouds but in volcano plumes as well (e.g. see work by our group, and the eruption of Eyjafjallajökull). Based on literature research we collected flash densities from two volcano eruptions: Mt Redoubt in 2009 and Eyjafjallajökull in 2010. Some interesting fact: both eruptions show orders of magnitude larger flash densities than what thunderclouds produce. On average the LIS/OTD data showed a 2×10-4 flashes/km2/h flash density while Eyjafjallajökull showed 0.1 – 0.32 flashes/km2/h and Mt Redoubt produced up to 2000 flashes in an hour over a square kilometer.

Venus, Jupiter and Saturn

Is there lightning on Venus? The long debated question has not yet been fully answered, however, more and more clue indicate the existence of such phenomenon on Earth’s sister planet. If lightning does exist on Venus, it is probably not very energetic and it appears deep within the atmosphere, since no optical observation of it has been made to date. Radio data from the Venus Express mission, however, shows a possible ~10-11 flashes/km2/hour flash density.

Figure 4. Lightning distribution on Jupiter. Triangles: Galileo data from 1997. Circles: New Horizons data from 2007.

Figure 4. Lightning distribution on Jupiter. Triangles: Galileo data from 1997. Circles: New Horizons data from 2007.

Jupiter and Saturn are more interesting, since, apart from Earth, these are the only planets in the Solar System where lightning has been observed directly. We used published data from the Galileo (1997, Jupiter), New Horizons (2007, Jupiter) and Cassini (2009, 2011, Saturn, e.g.) spacecraft. Figure 4 shows and example lightning distribution map for Jupiter. It shows and increased lightning activity around the +- 50-degree latitude regions, most probably due to the increased affect of internal heating on convection and cloud formation. Until 2009, only radio signals of lightning were observed on Saturn. The so-called SEDs (Saturnian Electrostatic Discharges) are short and strong radio bursts from lightning, detected by e.g. the Voyagers and Cassini spacecraft. The two giant gas planet, on average, shows a flash density of 10-6 – 10-7 flashes/km2/hour.

Exoplanets

Now the most exciting part: How does this all relate to exoplanetary research? The idea behind our paper is to use the statistics from Solar System planets to estimate a possible lightning occurrence on exoplanets with similar environments to those described in the previous sections. E.g. let’s take possible ocean planet: Kepler-62f. Then apply the flash density derived for Earth for above oceans: ~5 x 10-5/km2/hour. Combining this, we estimate the flash density on Kepler-62f, throughout its whole surface, to be the same: ~5 x 10-5/km2/hour. Another example is to apply the flash densities in order to estimate the total lightning occurrence during the transit of a planet. 55 Cancri e is a close-in rocky planet, most probably cover by lava. How much lightning could occur during its transit on the projected disk of the planet, if we assume continuous flashing from volcano activity? Depending on which eruptions we consider as the template, 55 Cnc e could produce as many as 108-1012 flashes on its whole disk during its 1.5-hour transit. This enormous amount of lightning would produce radio emission that might be observable during the transit.

In Part II we introduce our original idea of estimating whether a weak and tentative radio signal observed on the mini-Neptune, HAT-P-11b, could have been caused by lightning or not.

If you have an opinion, please leave a comment below.

For more details check out the original paper on ADS:

 Hodosán, G.; Helling, Ch; Asensio-Torres, R.; Vorgul, I.; Rimmer, P. B., MNRAS, 461, 3927, 2016

The LEAP Group can be found here:

http://leap2010.wp.st-andrews.ac.uk/

And finally, don’t forget to like us on Facebook:

https://www.facebook.com/leap2010

The mineral clouds on the extrasolar giant gas planets HD 209458b and HD 189733b

The giant gas planets HD 189733b and HD 209458b are the two most studied extrasolar planets today. Both have been observed by several research groups with varies telescopes including the Hubble and the Spitzer Space Telescopes, and the super-high-precision HARPS spectrograph mounted on the 3.6 m telescope in La Silla in Chile. These extensive observational programs have reviled details about the atmospheres of these planets, like for example the presence of CO and CH4 in HD 189733b and CO and H2O in HD 209458b. Observations have further established that both gas giants form clouds inside their atmospheres (e.g. Sing et al. 2016). Are these clouds similar to clouds on Earth? What are they made of? Why does HD 209458b seem to have more water vapour than HD 189733b? How different are clouds between the two planets?

We use results from 3D radiative-hydrodynamics simulations of the atmospheres of HD 189733b and HD 209458b to answer the above questions and to derive cloud characteristics. We apply the same ideas about cloud formation as described in our Drift-Phoenix post. First, condensation seeds form with a certain efficiency. Once they are present, many solid materials (e.g. MgSiO3[s], Fe[s], SiO[s], TiO2[s], [s] meaning solid phase) can condense on these numerous but small surfaces. As these cloud particles grow, they fall into the atmosphere (gravitational settling). These raining cloud particles will encounter changing ambient conditions because the gas temperature and the gas pressure increase inwards the atmosphere. On their way, the cloud particles change in size but also in composition.

We now probe the atmospheric cloud formation in HD 189733b and HD 209458b by calculating the cloud structure for the vertical atmosphere at different longitudes and latitudes shown on Fig. 1.

Figure1. Points in the atmosphere where cloud formation was probed

Figure 1. Points in the atmosphere where cloud formation was probed

We find that both planets have the smallest cloud particles near the top of the cloud and the largest cloud particles at the bottom of the cloud, which is far inside the atmosphere beyond observable heights. This can be seen from the black solid line in all the panels in Fig 2.

Figure 2 demonstrates the vertical cloud structures for the daysides of the giant gas planets HD 189733b (top) and HD 209458b (bottom). We show how the material composition changes with height for different latitudes (Φ=270°, 315°, 0°, 45°, 90°) along the equator. The material composition is visualized by the lines of different colours, representing one material each: TiO2[s] – solid dark blue, Al2O3[s] – solid blue, CaTiO2[s] – solid purple, Fe2O3[s] – dashed light green, Fe[s] – dotted green, SiO[s] – dashed brown, SiO2[s] – solid brown, MgO[s] – dashed dirty orange, MgSiO3[s] – dashed orange, Mg2SiO4[s] – solid orange. The contribution of the different materials to the volume of the cloud particles, Vs/Vtot, is shown in percentage. For example, VMg2SiO4/Vtot=0.3 means 30% of the cloud particle is made of Mg2SiO4.

Figure 2. Dayside cloud particle material composition (colour coded, left axis) and mean grain sizes (black, right axis) for both exoplanets. For colour codes refer to the original paper at Helling et al. 2016, fig. 7 , or to the text above

Figure 2. Dayside cloud particle material composition (colour coded, left axis) and mean grain sizes (black, right axis) for both exoplanets. For colour codes refer to the original paper at Helling et al. 2016, fig. 7, or to the text above

To the left in Fig. 2, where the gas pressure is low, is the upper part of the atmosphere and the cloud. Here, the cloud particles are small (10-2 μm) and made of a rich mix of materials indicated by many coloured lines appearing in the plots of Fig 2. Letting your eyes wonder more to the right shows that most of the lines disappear, because these materials evaporate (like Fe2O3[s], MgO[s]). Most of the cloud particles are now made of MgSiO3[s], Mg2SiO4[s], SiO2[s] and a bit of Fe[s]. When moving further inwards the atmosphere where the temperature increases beyond thermal stability of the silicate materials, a larger fraction of cloud particles will be made of Fe[s].

Inspecting Fig 2 a bit closer by comparing the results for HD 189733b and HD 209458b shows that the cloud particles at the inner rim of HD 189733b are more Fe[s] rich than for HD 209458b. The cloud particles in the upper atmospheric regions appear rather similar in material composition: they are made of silicates and oxides with only very small contribution form iron.

A major result of our work is that all cloud properties are interlinked and that it is extremely difficult to guess correct combinations of cloud particle-sizes and their material composition that will occur at a certain place inside the atmosphere of an extrasolar planet.

Why would HD 209458b show more water absorption than HD 189733b? Hence, why does HD 209458b seem to have more water vapour than HD 189733b according to observations? Water is the most abundant absorbing species in the gas phase and maybe one would not expect any differences between two relatively similar planets like HD 189733b and HD 209458b. However, our research shows that a considerably larger portion of the atmosphere of HD 209458b is affected by the cloud than for HD 189733b. The clouds in HD 209458b reach into regions of lower atmospheric pressure, hence lower gas densities, compared to HD 189733b. It should therefore be more difficult to observe water on HD 209458b than on HD 189733b. Therefore, the more a cloud layer extends into the low-density upper part of the atmosphere, the shallower the gas-absorption features will be in the observed spectrum (Fig. 3).

Figure 3. More water molecules accumulate above clouds at higher atmospheric pressure. Clouds on HD 209458b form lower in the atmosphere (right), therefore have more water molecules above them resulting in a deeper water absorption feature in the spectrum, than on HD 189733b (left)

Figure 3. More water molecules accumulate above clouds at higher atmospheric pressure. Clouds on HD 209458b form lower in the atmosphere (right), therefore have more water molecules above them resulting in a deeper water absorption feature in the spectrum, than on HD 189733b (left)

 

For more details check out the original paper on ADS:

 Ch. Helling, G. Lee, I. Dobbs-Dixon, et al. 2016, MNRAS, 460, 855

The LEAP Group can be found here:

http://leap2010.wp.st-andrews.ac.uk/

And finally, don’t forget to like us on Facebook:

https://www.facebook.com/leap2010

Mapping the sparkling clouds of the extrasolar planet HD 189733b

How do clouds look like on alien worlds? Graham Lee and Christiane Helling in collaboration with Ian Dobbs-Dixon (New York University Abu Dhabi) and Diana Juncher (University Copenhagen) took the first steps in modelling the formation of clouds for the giant gas planet HD 189733b, a member of a class of exoplanets commonly called ‘hot Jupiters’.

Observations (e.g. Pont et al. 2013; Sing et al. 2015, ) suggest that many hot Jupiters contain a large dust cloud component in their atmosphere because they obscure the absorption signatures of the atmospheric gas underneath the cloud layers. These clouds are made of mineral compounds such as TiO2[s], MgSiO3[s], SiO[s], Al2O3[s], Fe[s] (‘s’ meaning solid particles) (see previous blog post), and not of water like on Earth.

Inspired by previous research which proved clouds exist in brown dwarf atmosphere (see our post on DRIFT-PHOENIX Atmosphere Models) we set out to investigate if the same family of clouds could reside in hot Jupiter atmospheres.

We applied our cloud formation model to a 3D radiative-hydrodynamic simulation (RHD) of HD 189733b (Dobbs-Dixon & Agol 2013), to prove that the temperature and pressure conditions on these planetary atmospheres are suitable for cloud formation. We took temperature, density and pressure data in 1D “slices” of the 3D simulation as input for our cloud formation model (Figure 1). This is like using an atmospheric probe to sample the local conditions of the atmosphere during descent. The combination of a sophisticated 3D RHD atmospheric model and our 1D cloud formation model allowed us to create cloud “maps” of the HD 189733b atmosphere.

Left: Illustration of the sample trajectories (black points) taken from the 3D radiative-hydrodynamic (RHD) model atmosphere of HD 189733b (Dobbs-Dixon & Agol 2013). Right: Input temperature and pressure profile for the cloud formation model at the equator of the 3D RHD model.

Left: Illustration of the sample trajectories (black points) taken from the 3D radiative-hydrodynamic (RHD) model atmosphere of HD 189733b (Dobbs-Dixon & Agol 2013).
Right: Input temperature and pressure profile for the cloud formation model at the equator of the 3D RHD model. The dayside profiles are φ= 0°, 45°, 315°, nightside profiles φ = 135°, 180°, 225° and day-night terminator regions φ = 90°, 270°.

Our results show how cloud properties change between different regions of the planet. First we noticed that the size of cloud particles changes with the location on the globe. Grains found on the dayside generally grow faster and larger than those on the nightside. However, because of the lower temperatures on the nightside, more grains form on the nightside. This leads to an cloud structure where numerous small grains reside on the nightside while larger (but less abundant) grains reside on the dayside face of the planet.

We converted the cloud properties across the globe into a map of global cloud properties: Figure 2 depicts the mean particle size at an atmospheric pressure of 10-2 bar across the globe of HD 189733b, where the difference between nightside and dayside is most apparent.

With our simulations, we show that the maximum reflectivity of mineral clouds correspond to the 8 micron Spitzer global flux maximum observed by Knutson et al. (2007). Our results therefore suggest that clouds can significantly contribute to the infrared flux from these planets by scattering photons back into space.

We further found that the clouds on the hot Jupiter HD 189733b reflected more efficiently in blue than red spectral range. This suggests that the clouds on this planet will appear midnight blue in colour if viewed with human eyes. Figure 3 shows an RGB colour estimation for HD 189733b clouds  by interpolating the light scattering result. Evans et al. (2013) presented observations with the Hubble Space Telescope suggesting a bluish appearance of HD 189733b which our work now supports on the basis of detailed cloud formation modelling.

RGB scale and colour estimate of the cloud particles on the dayside face of the planet. Hubble Space Telescope observations found the planet to be a deep blue colour.

RGB scale and colour estimate of the cloud particles on the dayside face of the planet. Hubble Space Telescope observations found the planet to be a deep blue colour.

Could these clouds sparkle? Mineral cloud particles are thought to form crystalline structures as they travel through the atmosphere (Helling & Ritmeijer 2009). This means that the mineral particles that form the clouds on HD 189733b are likely to “sparkle” similar to gemstones on Earth, such as sapphire.

 

 

 

 

For more details check out the original paper on ADS:

Lee, G., Helling, Ch., Dobbs-Dixon, I., Juncher, D. 2015, A&A, 580, 12L

The LEAP Group can be found here:

http://leap2010.wp.st-andrews.ac.uk/

And finally, don’t forget to like us on Facebook:

https://www.facebook.com/leap2010