The Habitable Zone Planet Finder

The Unusual Transit of TOI-3884b: A Pole-Spot Crossing Super-Neptune

Posted on May 16, 2023 by Jessica Libby-Roberts

The TESS Mission

For the past five years, NASA’s Transiting Exoplanet Survey Satellite (TESS) has surveyed the sky searching for new planets orbiting our closest stellar companions. To accomplish this colossal task, this satellite stares at hundreds of thousands of stars waiting for a planet to pass in front of it blocking some of the light. This technique is known as the transit method, in which small periodic dips in a star’s light indicate the presence of a transiting object (Fig. 1). The depth of the transit is directly proportional to the size of the planet. The overall symmetric transit shape however, is consistent regardless of planet size – with the notable exception of the TOI-3884 system.

Fig 1. Cartoon of the transit method courtesy of NASA. As the planet passes in front of the star, the star dims slightly giving a symmetric transit curve.

TOI-3884b – A Super-Neptune Around a Low-Mass M-dwarf Star

A potential planet orbiting the star, TOI-3884, every 4 days was first flagged back in 2020. At this point, the only observations of this target relied on 30-minute exposures (imagine opening your camera shutters for 30 minutes at a time) taken by TESS. Because of this only 2-3 points were observed during the 1.5-hour transit. Interestingly, these dips blocked 3-4% of the star’s light – a significant fraction as most planets block less than 1% of light. This large several percent transit depth could only be explained by a Neptune or larger gas giant planet orbiting a small star known as a M-dwarf or red dwarf. The existence of a large planet around a low-mass star like TOI-3884 is surprising (with only three others known: TOI-3235b, TOI-519b and TOI-5205b, also confirmed by HPF).

However, simply observing a transit is not enough to confirm a planet. Small stars and brown dwarfs are also known to create similar deep transits. One reliable way to confirm this object is by measuring its mass. We began following up this planet with radial velocities using the Habitable-zone Planet Finder to derive a mass measurement. From these RV observations, we confirmed this object has a mass of 32 times that of Earth (~2 times Neptune’s mass). By all appearances, we discovered the first super-Neptune orbiting the lowest mass star known to date. A straightforward discovery, until we observed the transit.

A Persistent Asymmetric Transit

To confirm the object was transiting the star in question, we observed 3 transits from the ground-based telescope 0.3 m TMMT in Las Campanas, Chile. While we found the planet did orbit the correct star, the transit shape was…wonky. Instead of a clean symmetric dip (as shown in Fig. 1), the transit appeared skewed, slightly shallower in the first half, deeper in the second half (Fig.2). Odd, but perhaps an issue with the instrument? We observed it again, and again. Same result each time. We then decided to observe it with the 3.5 m Apache Point Observatory (APO) telescope using a fast exposure rate (one image per 20 seconds). Not only do we observe the same skewed transit, there appears clear structure in the transit shape! TESS observations in 2022 confirm this shape. Whatever is distorting the transit must be constant as it appears at the same time across every transit for nearly 2 years.

Fig. 2. Various transits of TOI-3884b observed with TESS (top row), the 0.3 m TMMT (middle row), and the 3.5 m ARC Telescope at APO (bottom row). The expected symmetric transit shape is plotted in red. The light blue points highlight the persistent bump that is apparent in every transit for the past 2 years.

The Culprit: A Large Starspot

Starspot crossing events, or when the planet passes over a star spot during transit, are notorious for creating bumps in a transit (Fig 3.). As the planet passes over the colder and darker spot, it suddenly blocks a little less light thus creating a slight rise in the transit shape. Once the planet reaches the hotter and brighter photosphere, it blocks a little more light creating a deeper shape. How much light is blocked during the spot crossing depends on the temperature of the spot – cooler spots are darker leading to more contrast and a bigger bump. For TOI-3884b, the amplitude of this bump leads to spot temperature 300 K cooler than the average stellar temperature (2900 K versus 3200 K). A value originally predicted by previous models and confirmed with this new data!

Fig. 3. Our transit model when including three different spots (red circles) on the star. The orbit of the planet is shown with the black line. As it passes over a spot, less light is blocked creating a bump in the below transit. In order to match our transit shape the star must be tilted so that we are observing its pole and TOI-3884b must possess a misaligned (polar) orbit.

However, the spot is only part of the story. We needed to explain how the spot appears at the exact same location in every transit for 2 years. This led to three hypotheses. First, the star could be rotating very slowly such that the spot has not rotated far on the surface. HPF spectra however, suggest the star rotates every 10 days. Second, the star’s rotational period could be exactly the same as the planet’s orbital period. This however implies that the spot would be rotating in and out of view every several days and the star would become brighter (no spot) and fainter (with a spot) over the 4 days. TESS shows the brightness of TOI-3884 to be constant over multiple observations ruling out this hypothesis as well (Fig. 4). The most promising hypothesis is if the star is inclined from our line of sight, such that we were observing its pole, and the spot was located on the pole (Fig. 3). Thus, as the star rotated, the spot never rotated in and out of view from our viewpoint. Pole spots on low-mass M-dwarf stars are not uncommon, especially young active stars like TOI-3884. They are also known to exist for months to years, it is feasible to observe this spot over multiple years.

Fig 4. If the spot was on the star’s equator, it would rotate in and out of view creating large changes in the amount of light (flux) we see (red model). However, TESS does not observe these changes meaning the star must be oriented such that most of the spot does not rotate out of view (blue model).

The only way for the planet to cross the spot, however, is if its orbit was significantly inclined from the star’s rotational equator. In other words, unlike our solar system with all the planets essentially lining up along our Sun’s equator, TOI-3884b’s orbit is inclined such that it nearly passes over its star’s pole! How TOI-3884b ended up on this skewed orbit is still a mystery, though TOI-3884b joins the small population of known misaligned Neptunes.

Future Prospects of the TOI-3884 System

Discovering a gas giant orbiting a low-mass star is in itself a surprising discovery! Understanding how TOI-3884b formed and evolved on its misaligned orbit remains an ongoing challenge. The atmosphere of this planet may hold tentative clues surrounding its origin. We show that TOI-3884b is one of the most promising planets for atmospheric characterization using the James Webb Space Telescope especially for planets cooler than 500 K (Fig. 5). The consistent spot crossing event also enables a unique look into the structure of M-dwarf pole spot and their impacts on the observed atmospheric spectrum of a planet.

Fig. 5. Simulation of what the atmospheric spectrum of TOI-3884b could look like with JWST. Water and methane is expected to be the most dominant molecules in the spectrum, and deviations from this could yield new insights into the atmospheres of giant planets and the impact of starspots on these measurements.

The TOI-3884 system is a rare find – a misaligned super-Neptune consistently transits a pole-spot along our line of sight. Without the assistance of HPF, we would have struggled to unravel this unique puzzle. However, while the mystery surrounding the skewed shape is solved (with the help of HPF), ongoing monitoring of this system with HPF will yield new one-of-a-kind insights into the planet, the star, and the system as a whole.

You can read more about the TOI-3884 here.

Posted in Exoplanet Science, HPF Science | Tagged planet, starspot | Comments Off

TOI-5205 b: A Forbidden Planet?

Posted on February 20, 2023 by Shubham Kanodia

The Discovery

For the past four years alongside the HPF survey to discover new exoplanets, we have also been using HPF to follow-up on planet candidates discovered by NASA’s Transiting Exoplanet Survey Satellite (TESS) mission. The TESS survey is observing the entire sky every two years, and has already found thousands of planet candidates, many of which orbit M-dwarfs. By observing millions of M-dwarfs, it is able to find giant transiting exoplanets (six times Earth’s radius or larger) around M-dwarf stars, which are supposed to be quite rare. The impact of this mission on our sample of these planets can be seen below, where up until 2018 we knew of just four of these planets, while there are now around 15 of them.

Discovery of confirmed giant transiting exoplanets orbiting M-dwarfs. The huge impact of TESS and our follow-up efforts utilizing HPF and NEID can be seen with the increase in discoveries post 2019.

TOI-5205

As part of these follow-up efforts, we began observing one of these planet candidates–TOI-5205 b–with HPF and other ground-based observing resources. This candidate seemed particularly interesting since it orbits a mid-M dwarf that is just about 40% the size and mass of the Sun, and about 3400 K in temperature (compared to the 5800 K for the Sun). This is an especially small host star when compared to all the previous giant Jupiter-sized planets, which were found around early M-dwarfs (roughly 60% Solar mass and 4000 K in temperature).

We started the characterization of this planet candidate with ground-based transits from a multitude of telescopes, including the 3.5-meter ARC telescope (ARCTIC) at Apache-Point Observatory, USA, 0.6 m Red Buttes Observatory (RBO) in Wyoming, USA, and 0.3 m TMMT telescope in Las Campanas, Chile. This was important because the host-star (TOI-5205) had another star present about 0.1% of a degree away on the sky, which contaminates the low-resolution TESS images. These ground-based transits helped us estimate the level of this contamination and also refine the transit timing for the planet.

Different ground-based photometry obtained for TOI-5205 b. The transit models in red were not used for obtain the transit depth due to various systematics, but just to refine the transit timing.

The first of these transits on April 22, 2022 with APO confirmed that the initial contamination efforts were close, and we did indeed have a massive Jupiter-sized object orbiting a mid-M dwarf. This transit, at nearly 7% brightness reduction, is one of the deepest known transits for exoplanets orbiting main-sequence stars!

Artist rendition showing the relative size of the TOI-5205 b system compared to similar Jupiter-like planets around Solar-type stars (Credit: Katherine Cain, Carnegie)

Subsequently we obtained seven HPF radial velocities (RV), which helped us estimate the mass of this object to be planetary in nature, and in fact almost exactly that of Jupiter (~ 1.08 Jupiter masses). TOI-5205~b turned out to be the first confirmed Jupiter-type planet such a low-mass star. The ratio of the planet’s mass to that of the star is about 0.3%, which one of the largest for all M-dwarf planets.

TOI-5205 b with respect to the other massive planets orbiting M-dwarfs. Its red marker shows how its host star is much lower in mass compared to the other similar systems.

So why is TOI-5205 b forbidden?

Core Accretion

For context, a 0.3% mass ratio object around a Solar-type star would have a mass of about 3 Jupiter masses; at these masses, the conventional core-accretion theories of planet formation start to have a harder time to form these objects. Under this core-accretion paradigm, planets must first form a solid core that’s primarily made of heavy elements (not hydrogen/helium) and roughly 10 Earth masses. Once it reaches this threshold, the planet undergoes a runaway process of gas accretion where it quickly builds up a massive gaseous envelope reaching Jupiter in size and mass (see here for a review).

So far all the Jupiter-like planets have been around early M-dwarfs, where the primary challenge has been forming this initial heavy-element core before the protoplanetary disks in which these planets form evaporate away. This was first shown by the seminal papers from Gregory Laughlin, and also Ida & Lin, which suggested that these low-mass stars would struggle to form these massive-enough cores in a timely fashion before the disk vanishes because the host-star is lower in mass (flashback to Kepler’s third law, where the orbital period is inversely proportional to the host-star mass). One way around this was the idea of gravitational instabilities in turbulent disks forming planets.

Gravitational Instability

Under the gravitational instability scenario proposed by Alan Boss, massive disks (typically assumed to be > 10% the mass of the host-star) tend to form self-gravitating clumps of dust and gas far away from their host star where the disk is quite cool. These clumps start off in spiral arms (similar to the spiral arms of the Milky Way Galaxy), and eventually coalesce to form planets. The main reason this presents a viable alternative to core-accretion is that this is a very fast process, where planets can form quickly in a few thousand years (as opposed to the millions of years it takes through core-accretion).

However, the problem for TOI-5205 b orbiting a mid-M dwarf is not just about the timescale of formation in the protoplanetary disks, but more of mass budgets.

Mass Budgets

It is typically assumed that planet formation begins in Class II protoplanetary disks which are about 1 – 10 million years old. These disks can now be studied with facilities like the Atacama Large Millimeter/sub-millimeter Array (ALMA) in the Atacama Desert, Chile which consists of 66 individual radio telescopes, and enable estimates of the amount of dust and gas present in a star’s protoplanetary disk.

ALMA: Large array of radio telescopes in Chile (Credit: ESO)

Studies of disks suggest that they tend to be quite diverse in their properties, while also following an approximate relation where the disk masses are proportional to the host-star mass. For stars similar to TOI-5205, these relations suggest that they have about 10 Earth masses of dust (here and here), albeit with a large spread in the predictions from these measurements.

For comparison, models of planetary interiors utilize our understanding of how gases and metals react to high temperatures and pressures to estimate what the interiors of these planets look like (here). Put simply, based on the planet’s mass, radius, temperature, etc., they can predict how much metal (or non hydrogen/helium heavy elements) they have. These models also have a large scatter in their predictions, but on average suggest that a planet such as TOI-5205 b should have about 60 Earth masses of heavy elements inside it.

And herein lies the problem: even if the entire protoplanetary disk collapsed to form one big ball of heavy elements, it still falls short by about a factor of 5! That being said, we know that TOI-5205 b exists, therefore there is some gap in our understanding of these disks, or planetary interiors, or the process of planet formation (or the most likely scenario, all of the above!).

Possible Solutions

There is the possibility that the ALMA observations are underestimating the disk dust masses, and that these disks indeed have a lot more dust in them. A few studies have shown that this might be the case, especially if the dust is locked up in particles that are larger than the millimeter regions probed by ALMA (see here for a review). If this were the case, then there would be a lot more material available to form these massive planets, which would help resolve the discrepancy presented above.

Furthermore, studies have suggested that planet formation is already underway for Class II disks, especially when it comes to giant planets. This implies that the total mass budget available for forming planets is not what is observed by ALMA for such disks, but perhaps much earlier during their Class 0 or I phase, when the disks are more massive by almost an order of magnitude!

The evolution of protoplanetary disks. Typical observations with ALMA are during the Class II phase. However, it is now starting to be established that planet formation begins much earlier.

Lastly, some of the assumptions made by some of these planetary interior models have not stood the test of time. Recent observations of the Solar-system giant planets with the Juno and Cassini missions have shown that their interiors are quite complex, with diffused cores where the helium and metals are immiscible in metallic hydrogen (see here and here for a review). If this were true for giant extrasolar planets as well, then these models would be overestimating their heavy-element content.

Conclusion

The existence of TOI-5205 b can perhaps be attributed to a combination of factors suggested above, which suggest the need for revisions in our understanding of protoplanetary disks, planet formation, and also planet interiors. Follow-up efforts such as those conducted by HPF for Giant Exoplanets around M-dwarf Stars (GEMS) are already hinting at the presence of more such planets, which suggests that TOI-5205 b–while definitely an outlier–isn’t the only one. If so, how frequently do these forbidden planets form?

You can read more about this intriguing system in this manuscript from HPF team member Shubham Kanodia.

Posted in HPF Science | Tagged formation, planet | Comments Off

TOI-2136b: Habitable, but not to us?

Posted on September 15, 2022 by Paul M Robertson

Will Life on Other Planets Look Like Ours?

Finding life somewhere other than Earth would be a truly monumental discovery, but will we know it when we see it? For as long as astronomers have been studying the stars, they have imagined far off worlds, usually something like Earth. And that’s why exoplanet scientists covet the “Earth 2.0,” an exoplanet just like Earth, that may just house the life we have long sought. But that does not mean that scientists aren’t also thinking about and looking for life in unusual places…

TOI-2136b: Characterizing the Planet with HPF, TESS, and ARCTIC

Scientists working with HPF–led by Science Team member Corey Beard–have recently characterized the transiting exoplanet, TOI-2136b. First identified by the transiting exoplanet survey satellite (TESS), we then utilized the diffuser-assisted ARCTIC instrument to refine the size of TOI-2136b. We learned that it is a sub-Neptune, meaning that a huge part of the planet is a big, gaseous atmosphere, just like Neptune, but with a radius only twice as big as Earth. Using HPF, our group was able to constrain the planet mass by measuring the Doppler wobble of the system. As a result, we confirmed that this system is probably a gaseous, puffy planet, rather than a rocky planet like Earth.

Left: Transits of TOI-2136b observed with TESS. Right: Transit of TOI-2136b taken with the diffuser-assisted ARCTIC instrument. By measuring the brightness of a star every two minutes, we are able to identify periodic dips that suggest a planet is passing in front.

A Cold Haber World?

Even more fascinating, TOI-2136b just might have an environment amenable to an exotic kind of life we’ve never seen on Earth. Orbiting its star once every 7.8 days, TOI-2136b is a lot closer to its star than the Earth is to the Sun. Luckily, TOI-2136b’s host star is an M dwarf, a cooler, redder star than the Sun. As a result, the equilibrium temperature of TOI-2136b is “only” 129^o C (265^o F)… way too hot for humans to survive. However, because the planet is in a size regime where exoplanets most likely have large atmospheres of hydrogen and helium, the pressure beneath the atmosphere is likely to be very high: so high, in fact, that water–which would normally have boiled into steam–can be forced to stay in a liquid state. This is highly significant, as liquid water is essential for any currently known kind of life.

Scientists originally proposed the idea of a Cold Haber World in 2013, a unique environment where bacteria live inside of a planetary ocean surrounded by a thick gaseous atmosphere. On Earth, some industries use what we call the “Haber process” at extremely high temperatures to produce ammonia by combining H₂ and N₂. Bacteria on TOI-2136b could combine hydrogen and nitrogen from the atmosphere using this same Haber process, though the temperature would be lower in the planetary environment than in the lab on Earth (hence the use of the word “cold”). Much like photosynthesis on Earth, these bacteria would get energy from the process, making it an appealing way for bacteria to survive.

But so what? How would we ever know that this was happening? Well, it happens that TOI-2136b has the perfect properties to make it viable for biosignature searches. At only twice the size of Earth, TOI-2136b is large enough that it probably has a thick atmosphere of hydrogen and helium, but small enough (<3.75 Earth radii) that it can’t produce ammonia abiotically; that is, without the help of life. With a temperature of 129^o C, the planet can have liquid water due to the high pressure of a thick atmosphere. And the cherry on top? This system is close enough to Earth that we could look at it with the James Webb Space Telescope (JWST), and actually figure out what its atmosphere is made of! As a part of our analysis, our team simulated what a JWST observation with NIRSPEC–one of the spectrometers on JWST–would look like, and the results are exciting. Clear molecular features from water, carbon dioxide, and methane are visible for any mass of the planet allowed by our HPF observations. Our simulations don’t include any ammonia lines, but that is because this planet shouldn’t be producing ammonia without any bacteria. So, if we observe this system and detect it, that would be compelling preliminary evidence that there is life on the planet, using this Haber process to create energy for itself.

By using our knowledge of chemistry and physics, we are able to simulate what a JWST NIRSPEC spectrum of TOI-2136b might look like. While the planet transits the star, its atmosphere absorbs some of the light. By looking at what colors are absorbed, we are able to determine which chemicals exist in the planet’s atmosphere.

While we haven’t detected any biosignatures yet on TOI-2136b, keep an eye out. Other groups are already working on studying this system more, and it has a lot of potential!

Posted in Exoplanet Science, HPF Science | Comments Off