There is something almost comic about the name GJ 887 d.

It sounds like a storage code, not a place. The star it circles is better known to catalogues than to the public, and the planet itself has never been photographed. Yet this anonymous-looking system, 10.7 light-years away, now contains the second-closest confirmed planet in a star’s habitable zone.

Only Proxima Centauri b is nearer.

The confirmation appeared in March 2026 in Astronomy & Astrophysics. Christian Hartogh and his colleagues reported that a faint wobble detected years earlier really is caused by a planet orbiting the red dwarf GJ 887 every 50.77 days.

I went into the paper expecting a fairly simple nearby-planet story. What I found was more interesting. GJ 887 d spent six years balanced between two explanations: it might have been a planet, or it might have been the star creating an impressively planet-like false signal.

The new work makes the planetary explanation much stronger. It does not tell us that the world is habitable, let alone inhabited. Those are very different claims.

The planet was hiding inside the star’s own noise

GJ 887 is a red dwarf, also known as Lacaille 9352 and HD 217987. It has roughly half the Sun’s mass, less than four percent of its luminosity and a surface temperature of about 3,688 kelvin. At visible wavelengths, it is one of the brightest red dwarfs in our sky, although at magnitude 7.39 it remains just beyond normal naked-eye visibility.

In 2020, a team led by Sandra Jeffers reported two planets around it. GJ 887 b completes an orbit in about 9.3 days, while GJ 887 c takes about 21.8 days. Both are much closer to their star than Mercury is to the Sun.

The same measurements contained a third rhythm near 50.7 days. That was the intriguing one because it would place a planet in GJ 887’s habitable zone. But the researchers did not claim it. Their 2020 analysis treated the signal as dubious and probably related to stellar activity.

This is a recurring problem in exoplanet astronomy. A planet can tug its star towards and away from us, shifting the star’s spectral lines slightly through the Doppler effect. Astronomers call this the radial-velocity method. Unfortunately, starspots, bright magnetic regions and rotation can deform the same spectral lines and produce their own repeating shifts.

For GJ 887 d, the star’s rotation was especially awkward. The 2026 team measured it at 38.7 days, close enough to the suspected planet’s 50.77-day period that a loose observing schedule could blur the two together.

The distinction matters because the planet makes GJ 887 move by only about 1.7 metres per second. That is walking speed. Astronomers were trying to identify that repeated motion in light arriving from more than 100 trillion kilometres away while the surface of the star produced variations of its own.

Why the 2026 confirmation is more convincing

The RedDots collaboration returned with 101 new radial-velocity measurements from HARPS, a precision spectrograph on the European Southern Observatory’s 3.6-metre telescope in Chile, plus 12 measurements from the ESPRESSO instrument on the Very Large Telescope.

Combined with archival observations, the analysis used 277 nightly binned HARPS radial velocities and 12 from ESPRESSO. The useful change was not only the amount of data. The new observations were taken with a daily cadence, allowing the team to follow two full rotations of the star rather than catching scattered snapshots.

Hartogh’s team also tracked several indicators of magnetic activity and used a Gaussian-process model to represent the correlated noise produced by the star. In plain language, they gave the analysis a flexible description of how stellar activity changes through time, then asked whether a separate 50-day planetary rhythm was still needed.

It was.

The team reported strong Bayesian evidence for GJ 887 d and a radial-velocity detection of about 4.6 sigma. The 50.77-day signal remained coherent while the activity measurements pointed to a stellar rotation closer to 39 days.

The peer-reviewed paper, “RedDots: Multiplanet system around M dwarf GJ 887 in the solar neighborhood”, now supports a system with at least four planets. NASA’s Exoplanet Archive lists GJ 887 d as confirmed.

This is one 2026 analysis, not the final word on every property of the system. The planet’s status is now much firmer than it was in 2020, but nearly everything that would determine its actual environment remains unmeasured.

What “second closest” actually means

GJ 887 lies 3.2877 parsecs away, or about 10.7 light-years. Proxima Centauri is only 4.24 light-years from us, and its planet Proxima b also occupies the habitable zone.

After that, GJ 887 d is next. The ranking comes directly from the discovery paper: it is the second-closest known planet in a habitable zone after Proxima Centauri b.

Nearby is relative. Light leaving GJ 887 when I was a university student would only now be reaching Earth. A spacecraft travelling at Voyager 1’s present speed would need well over 100,000 years to cover the distance.

But for astronomy, 10.7 light-years is genuinely local. The closeness makes the star bright enough for precise measurements and gives future telescopes a better chance of separating planetary light from starlight. Distance does not make GJ 887 d reachable, but it makes it observable in ways that more distant habitable-zone planets may never be.

The lack of a famous name is partly a cultural accident. Proxima Centauri belongs to the Alpha Centauri system, the nearest star system to the Sun. TRAPPIST-1 acquired a memorable name and seven compact planets that cross in front of their star. GJ 887 is a southern red dwarf filed under several catalogue labels, and its planets do not transit from our point of view.

It may be one of the most useful nearby planetary systems while remaining almost invisible in popular culture.

The habitable-zone label is narrower than it sounds

GJ 887 d orbits at about 0.212 astronomical units, roughly 31.7 million kilometres from its star. That would be intolerably close to the Sun. Around a faint red dwarf, it receives an estimated 81 percent of the energy Earth receives from sunlight.

The team calculated an equilibrium temperature of roughly 241 kelvin, or minus 32 degrees Celsius. That figure does not include the warming effect of an atmosphere. Earth’s own equilibrium temperature is below freezing, while its actual global surface temperature is warmer because of the greenhouse effect.

According to the climate model used in the paper, GJ 887’s habitable zone covers orbital periods from about 43 to 122 days. The planet’s 50.77-day year puts it inside that range, closer to the warm inner side than the cold outer edge.

I wrote recently about how icy moons complicate the usual picture of a habitable zone. The same caution works in the other direction here. Being inside the zone is not a certificate of habitability. It is a first-pass calculation identifying where an approximately Earth-like world, with a suitable atmosphere, might sustain liquid water on its surface.

Venus is near the Sun’s habitable-zone boundary and has a surface hot enough to melt lead. Mars lies within some definitions of the zone and has a cold, dry, irradiated surface. Orbit provides an energy budget. Atmosphere, geology, magnetic protection, water inventory and history decide what happens with it.

We do not yet know what kind of planet this is

Radial velocity gives astronomers a minimum mass, not a radius. For GJ 887 d, that minimum is about six times Earth’s mass. The real mass depends on the tilt of the orbit, which has not been measured.

The label “super-Earth” is easy to misread. It describes a mass range, not an enlarged Earth. The 2026 paper is explicit that a planet in this range could be predominantly rocky, rich in water or wrapped in enough gas to resemble a small Neptune.

NASA’s public catalogue currently classifies GJ 887 d as Neptune-like. The discovery authors use super-Earth because its minimum mass falls between two and ten Earth masses. Those descriptions are not evidence of a disagreement about an observed surface. They expose how little is known when a planet does not transit and its size cannot be measured.

We have no detection of an atmosphere. We do not know whether the planet has oceans, continents, clouds or even a solid surface accessible beneath its atmosphere. No oxygen, methane, water vapour or other possible biosignature has been measured.

This is where my earlier question about whether Earth may be unusually lucky keeps returning. A favourable orbit is only one filter. A planet must also acquire and retain the right materials, survive its star, regulate its climate and remain stable for long enough. We do not know how frequently that whole chain holds.

A quiet red dwarf can still flare

GJ 887 is less magnetically active than many stars of the same type. Between 1998 and 2018 it showed low starspot coverage, little variation in visible brightness and weak activity in several spectral indicators. That relative calm helped astronomers find its planets.

It may also make the system kinder to atmospheres than Proxima Centauri. The 2026 paper cites modelling suggesting Earth-like stellar-wind conditions and an Earth-like level of galactic cosmic rays in GJ 887’s habitable zone.

There is a qualification. A 2020 examination of archival Hubble observations found ultraviolet flaring that ordinary visible-light monitoring had missed. Arizona State University’s account of that work described large ultraviolet brightness spikes during the limited Hubble observations.

The 2026 authors therefore call GJ 887 magnetically quiet while acknowledging that it can flare. Whether those events would strip or chemically alter GJ 887 d’s atmosphere depends on the flare history, the stellar wind, the atmosphere itself and the planet’s unknown magnetic field.

Quiet is comparative, not absolute.

Why this obscure system may become a priority

GJ 887 d does not pass in front of its star, so the most productive technique used by the James Webb Space Telescope for studying exoplanet atmospheres is unavailable. There is no thin ring of planetary air for Webb to examine during a transit.

Direct imaging is the longer-term possibility. The planet should reach a maximum apparent separation of about 65 milliarcseconds from its star, approximately the proposed inner working angle for one design tier of NASA’s future Habitable Worlds Observatory. Its predicted contrast may be bright enough, but separating it from the star would sit near the instrument’s limit.

The authors are appropriately cautious and call its detectability unclear. They also note that GJ 887 appears on target lists for proposed concepts including the Habitable Worlds Observatory and the space-based LIFE interferometer.

Before then, astronomers can keep refining the orbit, search for the system’s inclination and test a further 2.2-day signal that may belong to a fifth planet. That candidate would have a minimum mass around half Earth’s, but its signal is below the stability excursions of HARPS and is not yet secure.

What has changed in 2026 is not that a second Earth has been found nearby. It is that a persistent 50-day wobble, once reasonably dismissed as stellar noise, now belongs to a confirmed planet in the right orbital zone for liquid surface water under some conditions.

At 10.7 light-years, that is close enough to deserve years of difficult follow-up. The next task is to discover whether GJ 887 d is a rocky world, a water world or a small Neptune that happened to land in the most suggestive part of its system.