An artist's impression of the newly discovered planet orbiting Alpha Centauri B. The star's binary companion, Alpha Centauri A, is seen at lower left, while the Sun is the bright star at upper right. Image: ESO/L Calcada/N Risinger (skysurvey.org).

A rocky exoplanet the size of Earth has been unveiled in the closest star system to the Sun, the Alpha Centauri triple system, just 4.3 light years away. The planet, found by the HARPS instrument at the European Southern Observatory, is too close to its star for life but the discovery – hailed as the greatest since the first exoplanets were discovered 20 years ago – tantalisingly hints at the possibility of a planet existing in the habitable zone.

Alpha Centauri is composed of two Sun-like stars, A and B, with a third red dwarf star named Proxima a tenth of a light year closer to us. The new planet, which has a mass 1.13 times that of Earth, revolves around star B every 3.2 days at a distance of just 5.98 million kilometres, placing it so close to its star that its surface will surely melt in the 1,220 degree Celsius heat. Furthermore, gravitational interactions with Alpha Centauri B will mean that the planet is tidally locked, so that it always shows the same face to its star as the Moon does to Earth. If there is no atmosphere to transport heat around the planet, the star-facing hemisphere will be an ocean of magma whereas the far hemisphere will be frozen stiff in permanent night – a Jekyll and Hyde planet if ever there was one.

This alien world was discovered by European astronomers using the record-breaking HARPS (High Accuracy Radial velocity Planet Searcher) on the 3.6-metre telescope at the European Southern Observatory in Chile. Since 2008 they have searched for a planet around our nearest stellar neighbour, competing with other teams from the United States and New Zealand.

"This was one of the most intensive campaigns ever to find an exoplanet, with more than 450 observations, observing almost every night," says Xavier Dumusque of Geneva Observatory, who led the observing programme. Dumusque is lead author on a paper describing the results that appears in the 18 October issue of the journal Nature.

For over four years HARPS looked at Alpha Centauri B three times every possible night, with exposure times of ten minutes separated by two hours. As a spectrograph, HARPS is able to detect shifts in the spectrum of light from a star that are caused by attendant planets. Strictly speaking, planets don't orbit their stars, but rather the star and its planets orbit the centre of mass between them. Since most of the mass in a planetary system rests inside the star, this centre of mass is usually located within the star itself, but importantly it is offset from dead centre, causing the star to 'wobble' around this point. This causes blueshifts in the star's light as the star wobbles towards us, and redshifts as it wobbles away from us. The size and periodicity of this Doppler shift indicates the mass of an orbiting planet and how far away it is located from its star. The Doppler shift signal of the planet detected by HARPS is tiny – a wobble of just 51cm per second – and really pushes the envelope of what HARPS can do.

"The Geneva team has done something very difficult, but it's an approach that I've advocated for the past few years," says Professor Debra Fischer of Yale University, who leads one of the other teams searching for exoplanets in the Alpha Centauri system. "They are digging into their data to remove signals that arise from the surface of the star, leaving behind the whiff of a residual signal induced by a prospective planet."

Those signals from the surface of the star are what made this detection especially difficult. Alpha Centauri B, like our Sun, has its own cycle of magnetic activity lasting eight years rather than the Sun's eleven-year cycle. Heat rises to the surface via convection through the outer-third of a star, which causes the stellar photosphere (its surface) to rise, inducing a blueshift in the star's light. Ordinarily this can be corrected for but, during the period of peak magnetic activity, a star develops spots just like sunspots. These spots are areas of tightly knotted magnetic fields that inhibit convection from within the star, leading to the photosphere sinking and causing a redshift that can interfere with the measurement of the Doppler shift induced by the planet. If careful corrections are not made, then the tiny signal of an orbiting planet can be lost in the noise of the star's magnetic cycle. In light of the small signal, commentators in some quarters have cast doubt on the findings, but the Geneva team are relatively unconcerned.

"We are talking about very small signals, so we have to rely on statistics, but the probability of this being a true detection is high, greater than 99 percent," says team-member Stephane Udry, also of Geneva Observatory. "So yes, while we have to keep an open mind, I am very confident that more observations will confirm what we are seeing."

Fischer agrees. "I have no doubt that they have carried out every reasonable test of their data before making this announcement," she says. "Nevertheless, because these corrections essentially constitute a new approach, confirmation is critical."

Independent confirmation can come in two ways. One is the possibility of a transit – the planet crossing the face of its star, blocking just one ten-thousandth of the star's light, a measurement that may be achievable with the Hubble Space Telescope. The HARPS team are already formulating a proposal to Hubble's scientific governors, the Space Telescope Science Institute in Baltimore, USA, to do just that. The other means by which confirmation can be forthcoming is if one of the other groups searching for planets in the Alpha Centauri system also detect the planet. Fischer's team are armed with the CHIRON spectrograph, which was commissioned in 2011 following funding by the National Science Foundation and is placed on the 1.5-metre Cerro Tololo Inter-American Observatory (CTIO), also in Chile. "Our precision, following upgrades this summer, matches the HARPS precision, but yields a five-month string of data compared to the five-year baseline from HARPS," says Fischer.

Star trails over the 3.6-metre telescope, which houses the HARPS instrument that found a planet around Alpha Centauri B, at the European Southern Observatory. Image: ESA/A Santerne.

If Fischer's team want to confirm they will need many more measurements, but they are in a race against time. The angular distance in the sky between Alpha Centauri A and B as viewed from Earth is closing up as the two orbit one another, with their closest approach as we will see it coming in 2016 before they proceed to separate again. The closer the two stars are, the more difficult the measurements become.

However, long term the prospects for further study, exploration and characterisation of planets so close to us in the Alpha Centauri system look highly promising, but has coming second in the race deterred Fischer at all?

"Ah yes, the agony of defeat – I felt it sharply for the first few days, even with the vindication that a planet has been found orbiting Alpha Centauri B," she tells Astronomy Now via e-mail, before turning once again to optimism. "There is not a more exciting result for an individual star, even with the long line of spectacular results from the last two decades. The indication that our nearest neighbour has rocky planets is incredible. Furthermore, statistical results from NASA's Kepler mission suggest that where there is one, there are usually several rocky planets. This leaves open the possibility of a terrestrial planet – in fact, I think this strengthens the speculative possibility of a habitable world in the Alpha Centauri system."

And so, as one race ends, another begins. Already the Geneva team are gearing up for the quest to find a habitable planet around Alpha Centauri B, with the development of their successor to HARPS, named ESPRESSO (Echelle SPectrograph for Rocky Exoplanet and Stable Spectroscopic Observations) that will be installed on the much bigger Very Large Telescope, also in Chile to permit the detection of Earth mass planets in the habitable zone. Similar instruments are also being designed for the next generation of extremely large telescopes that will come online in the next decade. Meanwhile, observations also continue to look for planets around Alpha Centauri A and Proxima Centauri. HARPS won the marathon to discover the first planet in the Alpha Centauri system – who will win the sprint to discover the first planet there in the habitable zone?