TOI-1243 b, TOI-4529 b, TOI-5388 b: small planets around M dwarfs
TESS and CARMENES confirm two sub-Neptunes and an Earth-sized planet around red dwarfs. We check the radii, the star densities and the wobbles with fibub's transit and mass tools.
Red dwarfs, or M dwarfs, are the most common stars in our galaxy. They are small and cool, so a small planet blocks a larger share of their light and tugs them harder than it would a Sun-like star. That makes them the best place to find and weigh planets not much bigger than Earth. A team has now confirmed three planets found by NASA’s TESS satellite around three nearby red dwarfs, using the CARMENES spectrograph in Spain to measure the stars’ wobbles. This article shows how the transit fitter and the exoplanet mass calculator let anyone check the sizes, the star densities and the masses.
The paper in brief
The three stars are 18 to 43 parsecs away (60 to 140 light years), with about a third to half the Sun’s mass and radius. Each has one transiting planet, a planet that passes in front of its star once an orbit and dims it slightly.
- TOI-1243 b orbits every 4.65 days. It is 2.33 times Earth’s radius and 7.7 ± 1.5 Earth masses, so its mean density is only 0.61 times Earth’s: a sub-Neptune, a planet with a thick envelope of gas or water over a rocky core.
- TOI-4529 b orbits every 5.88 days and is 1.77 Earth radii. Its wobble was too small to see; its mass is below 4.9 Earths.
- TOI-5388 b orbits every 2.59 days and is 0.99 Earth radii, the size of Earth. Its mass is below 2.2 Earths.
All three radii are known to better than 7%. The team also used ground-based telescopes to rule out other causes of the dips, such as a background eclipsing binary star.
Checked with the tools
We used the paper’s periods, radii, masses and star properties, and fibub’s maths. We did not have the TESS data, so the transit fit in Step 2 uses a simulated light curve and says so.
Step 1: how deep is each dip? The share of starlight a planet blocks is the square of its radius divided by the star’s. TOI-1243 b covers 0.0418 of the star’s width, so it dims the star by 1,747 parts per million, about 0.17%. TOI-4529 b dims its star by 1,143 ppm and TOI-5388 b, an Earth-sized planet, by 915 ppm. Around the Sun, an Earth would make a dip of only 84 ppm: this is why red dwarfs are good hunting grounds.
Step 2: fitting a transit. We made one 27-day TESS-like sector for TOI-1243 b (2-minute points with 1.5 parts per thousand of noise, an impact parameter of 0.4 and a typical red dwarf limb darkening, all our assumptions) and fitted it with the Transit light curve fitter’s maths, period fixed (Figure 1). It found a planet-to-star size ratio of 0.0402 ± 0.0041, which with the star’s 0.511 solar radii gives 2.24 ± 0.23 Earth radii, against the paper’s 2.33 ± 0.12. The depth is well measured, but the shape is not: the fit could not pin down the impact parameter or the orbit size (a/R* = 19 ± 11) from one noisy sector. This is the trade-off the tool warns about, and why papers like this one add sharper ground-based light curves.
Step 3: the star’s density from the orbit. Kepler’s law with the star’s mass gives TOI-1243 b an orbit of 0.0437 au, which is 18.4 star radii. Fed back into the fitter’s density formula, that gives a mean star density of 5.44 g/cm³, the same as 0.515 solar masses packed into 0.511 solar radii. That is about four times the Sun’s 1.41 g/cm³, as expected for a red dwarf. With a cleaner light curve, this is how a transit alone can check a star’s size.
Step 4: the wobble. A 7.7 Earth-mass planet on a 4.65-day orbit pulls its 0.515 solar-mass star back and forth at 4.59 m/s, a brisk walking pace, which CARMENES could measure. Masses at the paper’s limits would give 2.82 m/s for TOI-4529 b and 2.33 m/s for TOI-5388 b. If TOI-5388 b has Earth’s density, its pull would be only about 1.0 m/s, which explains why only an upper limit was possible.
Step 5: what the planets are made of. Mass divided by radius cubed gives the density in Earth units (Figure 2). TOI-1243 b comes out at 0.61 times Earth’s, about 3.3 g/cm³: too light for bare rock and iron. The limits for the other two are 0.88 and 2.27 times Earth’s, the second matching the paper’s 2.2 after rounding. As a rough guide, if each planet absorbed all its starlight and spread the heat evenly, they would be near 550 to 580 K, far too hot for liquid water.
| Check | Paper | Our tool result |
|---|---|---|
| TOI-1243 b radius | 2.33 ± 0.12 Earths | 2.24 ± 0.23 Earths from one simulated sector |
| Transit depths | (in the fits) | 1,747, 1,143 and 915 ppm |
| Star density from the orbit (TOI-1243) | 0.515 Suns, 0.511 solar radii | 5.44 g/cm³ either way |
| TOI-1243 b wobble | mass 7.7 ± 1.5 Earths | 4.59 m/s |
| TOI-1243 b density | 0.61 ± 0.15 Earth’s | 0.61 Earth’s (3.3 g/cm³) |
| Density limits | < 0.88 and < 2.2 Earth’s | 0.88 and 2.27 Earth’s |
How we made the graphs
The charts and every number above come from scripts/blog-figures/toi-1243-sub-neptune-m-dwarf-carmenes.mjs in fibub’s public code. It calls the tools’ own functions:
fluxAt,fitTransit,fold,durationsandstellarDensityfromsrc/lib/sci/transit.ts, the engine of the transit fitter.semiMajor,semiAmplitudeanddensityfromsrc/lib/sci/rvmass.ts, the maths behind the exoplanet mass calculator.
Where the inputs come from: periods, radii, masses, limits and the star masses, radii and temperatures are from the paper (abstract and Table 1). The simulated light curve’s noise (1.5 ppt per 2-minute point), impact parameter (0.4), limb darkening (0.3 and 0.3) and timing are our own choices, labelled in the script. The 550 to 580 K temperatures use a simple textbook formula (zero reflection, heat spread all round), also ours.
How we test the maths
On every change to fibub, tests/sci.mjs checks these functions against known answers:
- The transit maths must give HD 209458 b a transit lasting 3.07 hours and a star density of 1.02 g/cm³ from its published orbit, and the Sun a density of 1.41 g/cm³ from Earth’s orbit.
- A small planet crossing the centre of a limb-darkened star must block exactly the share the textbook formula gives, to eight decimal places.
- The orbit maths must give Jupiter’s pull on the Sun as 12.47 m/s, and 51 Pegasi b’s orbit as 0.053 au.
Check it yourself
The wobble, on the exoplanet mass calculator:
- Choose Mass from the wobble.
- Type 0.515 in Star mass (Suns), 4.65 in Orbital period (days) and 4.59 in Velocity semi-amplitude K (m/s). Leave Eccentricity at 0.
- It reads about 7.70 Earth masses, the paper’s mass (strictly the minimum mass; a transiting planet is seen almost edge-on).
- Type 0.1747 in Transit depth (%) and 0.511 in Star radius (Suns): Planet radius reads about 2.33 Earth radii.
The transit, on the transit fitter: paste a TESS light curve of TOI-1243 as time (days), brightness and error. Type 4.65 in Orbital period (days) and 0.511 in Star’s radius, if known (R☉). The depth should come out near 1.7 parts per thousand and the planet near 2.3 Earth radii; check Star’s mean density against 5.4 g/cm³. With one sector, expect loose values for the impact parameter, as in Step 2. We did not run this step on the real TESS data.
What researchers can use it for
- Predict the wobble a transiting planet should cause, to decide whether a spectrograph can weigh it.
- Turn a transit depth into a planet size for any star radius, and see how errors in the star carry into the planet.
- Check a star’s size by comparing the density from the transit with the density from its mass and radius.
- Sort planets by make-up: place them on a mass-radius diagram against lines of equal density.
- Teach why red dwarfs are the best places to find and weigh small planets.
The fine print
- The light curve in Figure 1 is simulated. It shows what one TESS sector can and cannot pin down, not the real data.
- The wobbles assume circular, edge-on orbits; the paper also assumed circular orbits.
- The depths ignore limb darkening, the dimmer edge of the star, which makes a real dip a little deeper at its centre.
- The temperatures are a rough guide that ignores clouds, reflection and how each atmosphere moves heat.
Sources
- Poultourtzidis, E., Lacedelli, G., Pallé, E. et al. (2026), Characterization of two new transiting sub-Neptunes and a terrestrial planet around M-dwarf hosts, Astronomy & Astrophysics 707, A106, doi:10.1051/0004-6361/202557155. Licence CC BY 4.0, checked on the journal page, 2026-11-16. Values used here, from the abstract: periods 4.65, 5.88 and 2.59 days; radii 2.33 ± 0.12, 1.77 (+0.09/−0.08) and 0.99 (+0.07/−0.06) Earths; TOI-1243 b mass 7.7 ± 1.5 Earths and density 0.61 ± 0.15 Earth's; 3-sigma mass limits 4.9 and 2.2 Earths and density limits 0.88 and 2.2 Earth's. From Table 1: star masses 0.515, 0.482, 0.292 solar; radii 0.511, 0.480, 0.300 solar; temperatures 3515, 3697, 3440 K.
If you are an author or publisher and want a credit or a passage changed or removed, tell us and we will act quickly.