Before NASA’s TESS (Transiting Exoplanet Survey Satellite) mission launched in 2018, astronomers tried to understand what it would find in advance. One study calculated that TESS would find between 4430 and 4660 new exoplanets during its primary two-year-long mission.
The primary mission (PM) is over, and TESS is in its extended mission (EM) now. The extended mission is 1.5 years old, and TESS has discovered 176 confirmed exoplanets and 5164 candidates. Scientists are still going through data from the primary mission, so the data might be hiding many more exoplanets. And TESS isn’t finished yet.
When TESS finds a new exoplanet candidate, the candidate becomes a TESS Object of Interest (TOI.) The mission recently announced another batch of exoplanet candidates, boosting the TOI count to more than 5,000. The new batch doubles the number of TOIs to date. This batch is mainly from the Minnesota Institute of Technology’s (MIT) Faint Star Search. MIT post-doc Michelle Kunimoto leads the Faint Star Search.
“This time last year, TESS had found just over 2,400 TOIs. Today, TESS has reached more than twice that number — a huge testament to the mission and all the teams scouring the data for new planets,” Kunimoto said in a press release. ”I’m excited to see thousands more in the years to come!”
TOIs require confirmation, and when TESS identifies one, it’s fed into the TESS Follow-up Observing Program (TFOP.) Over 40 groups of astronomers at facilities worldwide make up the TFOP. Their observations not only confirm the candidates as actual exoplanets; they also refine the exoplanets’ radii and masses. Once the TFOP does its work, the planet can then be called a confirmed exoplanet.
TESS is a complex mission that generates enormous quantities of data. Exoplanet candidates hide in that data, and much of the data requires manual inspection. But that’s an arduous, time-consuming task. The Faint Star Search uses a modified method to scour all data for exoplanet candidates.
The Faint Star Search FSS is different from the PM. TESS’s primary mission focused on nearby bright stars, but the FSS puts the spotlight on exoplanets around faint stars. The Faint Star Search is “… an independent vetting pipeline…incorporating both automated vetting tests and manual inspection to identify promising planet candidates around these fainter stars.”
TESS’s two-year primary mission covered 85% of the sky. The satellite observed the southern hemisphere sky in the first year and the northern hemisphere sky in the second. This new batch of exoplanets candidates comes from the first year of TESS’s extended mission, the mission’s third year overall. During that year, the spacecraft re-observed the Earth’s Southern Hemisphere sky, an area the mission previously studied in its first year. The re-observation led to dozens of new TOIs.
Katharine Hesse is part of the team that vets and releases TOIs at MIT’s Kavli Institute. “With data from the first year of the extended mission, we have found dozens of additional candidates to TOIs found during the prime mission, “ Hesse said in a press release. “I am excited to see how many multi-planet systems we can find during the rest of the extended mission and in upcoming years with TESS.” There are plans to extend the TESS mission to 2025 and beyond, and the satellite should unveil many more new planet candidates during that time.
TESS’s extended mission is different from the primary mission. It increased its imaging cadence from 30 minutes in the PM to 10 minutes in the EM. So TESS will observe a star in its EM three times longer than in the PM.
Michelle Kunimoto told Universe Today that the EM’s faster cadence benefits the candidate vetting process and results in fewer false positives. “I would expect the EM to have a lower false-positive rate than the PM. Part of this should be because most targets in the PM were re-observed in the EM, so stars have more data to work with, and more data means more transits and better constraints on planet parameters.” Kunimoto said.
“Another part of this should be because the EM full-frame image observations (where most TOIs have been found) were observed at an improved 10-minute cadence, faster than the 30-minute cadence of the PM observations. This means that a star observed in the EM will have three times as many data points as one observed in the PM over the same time period. This helps to distinguish transit shapes and helps TOI vetters spot tell-tale signs of astrophysical false positives in the lightcurves before they alert them as TOIs.”
False positives play a significant role in TESS’s data. Identifying them takes diligence, and they occur at different rates in the PM, the EM, and the Faint Star Search. TESS generates an enormous amount of data, and the data gathered in the PM, EM, and FSS is all different.
Kunimoto provided some numbers for false positives and an explanation for how it all fits together.
Of the 2241 Prime Mission TOIs (those in the official Prime Mission Catalog from Guerrero et al. 2021), there are:
- 606 False Positives (FPs)
- 192 Confirmed Planets (CPs)
- 269 Known Planets (KPs)
- 1174 Planet Candidates (PCs)
The high number of FPs in the primary mission might look shocking, but it’s an expected part of the process. The PM has the highest number of FPs and the highest number of Confirmed Planets. Why?
“Keep in mind that these are the oldest TOIs, which means they have had the community’s attention for the longest, and will have received the most follow-up work,” Kunimoto said. But most TOIs in the Prime Mission haven’t been dispositioned as false positives or confirmed as planets yet.
“It looks like 27.0% of the Prime Mission TOIs are false positives, 20.6% are bonafide planets, and the remaining 52.4% do not yet have final dispositions,” Kunimoto explained.
The detection data from the Extended Mission looks different. But that data hasn’t been scrutinized as much as the PM yet.
The EM identified 936 TOIs. That doesn’t include the Faint Star Search.
- 61 FPs
- 10 CPs
- 120 KPs (many of these were found originally by the K2 mission, and re-detected by TESS when it observed the ecliptic for the first time over the past ~5 months)
- 747 PCs
“6.5% are FPs, 13.9% are bonafide planets, and the leftover 77.5% are still candidates,” Kunimoto explained. A greater percentage of TOIs in the EM are still candidates compared to the PM because the data hasn’t been around as long.
Next is the Faint Star Search. The FSS isn’t a separate mission, just a different way of combing through the data. Since the target stars are fainter, there isn’t as much light. That means the data isn’t as strong, and there will likely be more False Positives. It also means the vetting process is more challenging and rigorous.
“The Faint Star TOIs so far account for 2033 TOIs, where 1617 were initially found using data from the Prime Mission (a year after the Prime Mission ended) and the latest batch of 416 (which pushed TESS over 5000 TOIs) have been from the Extended Mission,” Kunimoto said.
Here are the current results from the FSS:
- 133 FPs
- 1 CPs
- 15 KPs
- 1884 PCs
So far, the FSS data contains a greater percentage of false positives than the other missions. That’s not unexpected.
Kunimoto explains why:
“I would also expect the Faint Star TOIs to have a higher false-positive rate than other kinds of TOIs,” Kunimoto said. “These are TOIs around stars as faint as TESS magnitude = 13.5 magnitude, and almost no bright stars (< 10.5 magnitude). The light curves of fainter stars have lower precision, which makes it more challenging to find planets, distinguish transit shapes, and spot tell-tale signs of false positives.”
The lower magnitude of the faint stars leads to particular kinds of false positives. Most of them are Hot Jupiters. Why? They block more light which creates a stronger signal.
“That also means that most Faint Star TOIs should be giants, which will have very deep transit depths that can still stand out against noisy lightcurves,” Kunimoto explained. “In fact, over half of all Faint Star TOIs are hot Jupiters – giant planets with very short orbital periods!”
“Giants tend to be more easily confused with eclipsing binary false positives. Finally, fainter stars tend to have less stellar information available, so in some cases, we do not actually know the mass/radius/temperature/etc. of a host star,” Kunimoto said.
With less light to observe, astronomers are in a tough spot. They need light from the star to measure the star’s properties. And since all they know about the potential planet orbiting the star is based on the star itself, it leads to uncertainty.
“?We have to assume certain stellar parameters in order to estimate, for example, how large the planet is – so it may be the case that some Faint Star TOIs would be too large to be planetary if we had the correct stellar radius,” Kunimoto said.
There’s nothing wrong with false positives. They’re an expected part of the process and don’t represent a failure in any way. “False positives are an expected and unavoidable part of planet candidate identification,” Kunimoto emphasized. And they can be easier to identify than actual planets.
“Finally – I’d like to emphasize that it’s much easier to identify a false positive than it is to fully confirm a planet,” Kunimoto said.
The data from TESS so far represents a sort of snapshot in time. TOIs listed as planet candidates might have already been confirmed but not reported in any journals yet. “A TOI marked as simply a “Planet Candidate” above may have already received follow-up that shows it is highly like to be a real planet – but no one in the community has written up an official validation/confirmation paper for that planet yet!” Kunimoto said.
It’s impossible to know what TESS’s final numbers will be. But we’ve come a long way from the two exoplanets discovered around a pulsar in 1992. Thanks especially to Kepler and TESS, we know of 4903 confirmed exoplanets as of January 10th, 2022. T
TESS is still working, and there’s a backlog of data. So far, there are 176 TESS confirmed planets and 5164 TESS candidates.
Whatever numbers we end up with, TESS is a triumph so far.