Epsilon Eridani, also known as HD 22049, is the nearest non-binary Sun-like star, as well as the nearest star to host a candidate exoplanet. Located at a distance of only 10.5 light years, this young orange dwarf has the further distinction of harboring a huge debris disk, with a far greater mass in ice, rocks, and dust than our own Kuiper Belt (see also Debris Disk Systems).

The star belongs to spectral class K2, with a luminosity only 34% Solar (Kaler), and a mass and radius of 0.83 and 0.73 Solar, respectively (Di Folco et al. 2004, Benedict et al. 2006). These values correspond to a primordial ice line at about 2.25 AU and a current habitable zone centered around 0.6 AU (Menou & Tabachnik 2003, Raymond et al. 2007). Epsilon Eridani’s metallicity is typical for stars in the Solar neighborhood, but lower than the average exoplanetary host, with [Fe/H] estimated between -0.06 and -0.09 (Di Folco et al. 2004).

All sources agree that Epsilon Eridani is younger than 1 billion years, but beyond that, no consensus has emerged. Estimates of its age range from 200 to 850 million years. A careful review by Markus Janson and colleagues found no conclusive evidence to narrow this range, but on the basis of stellar rotation they favored a value of 440 million years (Janson et al. 2008). At an equivalent stage in its youth, our Solar System was a chaotic field of asteroids and comets that ricocheted among the planets and created impact craters on every available surface. Similar circumstances may explain the extensive debris observed around Epsilon Eridani.

gas giant planet

Epsilon Eridani has been the object of radial velocity observations for more than 20 years (Hatzes et al. 2000). Various teams have presented data indicating the presence of one gas giant planet, conventionally named Epsilon Eridani b, in an orbit wider than 3 AU. Despite the system's proximity, however, the star presents a difficult target on account of its extensive sunspot activity, which introduces high levels of background noise into the data (Gray & Baliunas 1995, Moran et al. 2004). Active chromospheres are typical of young stars, compromising the analytic precision necessary to characterize an exoplanet (Setiawan et al. 2008). The table below summarizes results published over the last decade.

No Consensus Yet on Epsilon Eridani b

Column 1: Source of parameter; for full bibliographic information, see below. Column 2: Orbital period in years. Column 3: Semimajor axis in astronomical units (AU). Column 4: Minimum mass as a fraction of Jupiter's mass. Column 5: Orbital eccentricity, with zero = a perfect circle.

Investigators generally agree that the radial velocity data contain a periodic signal that repeats approximately every 7 years. Nevertheless, the broad scatter in data points encourages doubt as to the signal's shape and source. Sunspot cycles or other chromospheric activity has not been conclusively ruled out. Thus two recent studies by two different astronomical teams regard the proposed planet as "controversial," "tentative," and "suspected but still unconfirmed" (Moran et al. 2004, Backman et al. 2009).

Most consistent among published parameters are the values for orbital period (7 years, give or take a few months) and semimajor axis (about 3.4 AU, which in the Solar System would fall midway between the orbits of Mars and Jupiter). Thus Epsilon Eridani is one of the few announced exoplanetary systems that resembles our own, insofar as it hosts a gas giant orbiting largely or entirely outside the system's ice line.

Values for minimum mass vary widely, and estimates of eccentricity range from about 0.0 to more than 0.7, within error margins (not shown in the table). If we accept the eccentricity presented by Benedict's group, Epsilon Eridani b has a periastron of 1 AU and an apastron of 5.8 AU, equivalent to a journey from Earth to Jupiter and back every seven years. On the other hand, if we accept the value presented by Butler's group, its periastron increases to 2.5 AU and its apastron shrinks to 4.2 AU. Even the low end of these eccentricity estimates may still be too high.

debris belts

We view Epsilon Eridani’s debris disk in a roughly face-on orientation, at an inclination of about 30 degrees against the plane of the sky (Hatzes et al. 2000, Greaves et al. 2005, Benedict et al. 2006). Dana Backman and colleagues have recently presented a comprehensive examination, augmenting previous results from Greaves and colleagues (Backman et al. 2009).

They find three concentric regions of dusty debris:

• A broad outer belt extending from 35 AU to 90 AU, with an exterior “halo” stretching outward as far as 110 AU. The outer belt's composition is primarily ice particles, with a small proportion of silicates. This region is a more massive and extensive analog of the Solar System’s Kuiper Belt. Backman and colleagues speculate that no major collisions have occurred in this area over the past 100,000 years.
  
  
• A narrow, sharply defined ring at about 20 AU, probably composed of silicate particles and hinting at a belt of rocky asteroids about 2 AU in width.
  
  
• Another narrow ring at 3 AU, definitely composed of silicates and about 1 AU in width, suggesting a near-twin to the Solar System’s Asteroid Belt.

Like analogous structures around stars of all spectral types, Epsilon Eridani’s outer disk must include a substantial population of ice dwarf planets resembling Pluto and Eris (Kenyon and Bromley 2008), whose perturbations stir up the surrounding debris and cause dust-producing collisions. The inner belts probably contain objects similar to our own asteroids, which range from cratered spheres like Ceres to irregular rocks like Gaspra.

system architecture

Many investigators have concluded that the morphology of Epsilon Eridani’s dust disk, with its extensive inner clearing and large outer radius, cannot be explained by the presence of the proposed planet at 3.39 AU (Greaves et al. 2005, Janson et al. 2008, Backman et al. 2009). Backman et al. consider that the most likely explanation for this complex structure is the presence of two or more additional planets orbiting in the gaps separating the three rings (Backman et al. 2009).

They propose that the outer belt is sculpted by a sub-Saturn mass planet orbiting just inside 30 AU, while the middle belt seems to require a planetary shepherd just beyond 20 AU, with a mass similar to Jupiter’s. Radial velocity data for Epsilon Eridani do indeed hint at the presence of at least one outer planet with an orbital period well in excess of 30 years (Janson et al. 2008).

The narrow belt at 3 AU needs a shepherd of its own to maintain its edges. Epsilon Eridani b could fulfill this role, but only if its orbit is less eccentric than most orbital solutions proposed to date (see table, above). Backman notes that the observed structure of the inner belt is inconsistent with the existence of the planet proposed by Hatzes, Benedict, and colleagues. A massive planet on such an eccentric orbit would rapidly clear out the observed ring. Therefore, either the original detection by Hatzes’ group is spurious or the planet’s eccentricity must be less than 0.25. A planet whose semimajor axis is 3.4 AU must have an eccentricity no greater than 0.12 to avoid crossing an astrocentric radius of 3 AU at periastron.

In short, Epsilon Eridani appears to harbor a complex planetary system, even if its components cannot yet be precisely constrained. Both radial velocity and infrared observations are consistent with an outer system of two or three giant planets in the same range of masses as the Solar System's four giants. None will substantially exceed the diameter of Jupiter, and all are likely to host families of moons and at least simple ring systems.

The asteroid belt at 3 AU hints at an inner system of rocky bodies analogous to the Solar System's terrestrial planets. Given Epsilon Eridani's youth, these hypothetical objects may not have developed stable crusts yet, and they must be subject to frequent cometary and asteroid impacts. Even if the system is destined one day to harbor an Earth twin, such a world would still be in its formative stages.

Last updated January 2009