Selected Star Systems Within 10 Parsecs (32.6 Light Years)
DEEP FLY 2008

This diagram is a visual summary of the Solar neighborhood, seen from the perspective of an observer in the north celestial hemisphere. Within 32.6 light years of the Sun, it includes:

• Most of the brightest stars
• All known exoplanetary systems
• All systems containing debris fields resembling our Asteroid and Kuiper Belts
• A selection of Sun-like stars with the potential to host extrasolar planets

Contemporary astronomy tends to focus on phenomena that are extremely remote in space and time: the evolution of galactic nuclei, the physics of star-forming clouds, the geometry of dark matter. Meanwhile, basic information on the Sun’s immediate neighborhood remains fragmentary or disputed. A reasonably complete stellar census has been conducted only for stars located within 10 parsecs (32.6 light years) of our Solar System, and even within this limited volume many details are obscure. Nevertheless, the patchy data so far accumulated provide a glimpse into the Sun’s back yard.

recent history

Our neighborhood is relatively free of interstellar gas and dust, as established by studies of the local distribution of molecular hydrogen clouds. Hence the name Local Bubble, although our bubble is actually an extended chimney-shaped void (Lallement et al. 2003, Perrot & Grenier 2003). Current investigations suggest that this structure has been created over the past 50 million years by supernovae explosions associated with the expanding front of starburst activity known as the Gould Belt (Maiz-Apellaniz 2001).

The same investigations imply that the Solar System was located near the heart of this active starburst region in the astronomically recent past (within the past 10-50 million years). Narciso Benitez and colleagues have argued that, two million years ago, a nearby supernova associated with the Gould Belt damaged the Earth's ozone layer and caused a minor extinction event at the transition to the Pleistocene epoch (Benitez et al. 2002). With the subsequent expansion of the Gould Belt, our system is now quite distant from any star-forming clouds and their embedded stellar giants. The closest such region is probably the Rho Ophiuchi complex, about 145 parsecs (475 light years) away (Makarov 2007).

star populations

As a result, the immediate Solar neighborhood is devoid of the brightest, hottest stars, represented by spectral class O. The lifetimes of these dazzling objects are so brief that they are found only in the vicinity of their native clouds. Our neighborhood also lacks stars of spectral class B, which like O stars are extremely bright, short-lived, and rare. No B stars exist within 20 parsecs (65 light years), and only about a dozen are found between 20 and 40 parsecs (130 light years).

Within 10 parsecs of the Sun, the RECONS Survey counts a minimum of 344 stars and brown dwarfs, with the following distribution by spectral type:

D = white dwarfs; L + T = brown dwarfs. Data from the RECONS Survey as of 2008.

Bright white stars of spectral class A are the least numerous, and spectral class F is only slightly better represented. Populations increase as mass and luminosity decrease, such that lightweight M dwarfs outnumber Sun-like G stars by a ratio of 11 to 1.

RECONS data also indicate that most systems (70%) within 10 parsecs contain only one star, while 22% of systems contain two stars, 6% contain three stars, and 2% contain four or more. At least in our region of the Galaxy, stellar multiplicity correlates closely with spectral type, such that about 60% of G and K stars, 30% of M dwarfs, and 20% of brown dwarfs occur in binary or multiple systems (Lada 2005, Allen 2007). Accordingly, most of the single stars in our neighborhood are red stars of class M, while more than half of all nearby Sun-like stars (i.e., members of spectral classes F, G, and K) are found in binary or multiple systems.

These distributions by stellar multiplicity and spectral type may be typical of the entire Milky Way, although we can expect considerable local variation. In any case, the neighborhood population evidently represents stars born in many different parts of the Galaxy, whose Galactic orbits have migrated over hundreds of millions of years and fortuitously coincide at this epoch (Famaey et al. 2007, Ecuvillon et al. 2007).

exoplanetary systems

Outside our Solar System, only one Sun-like star within 10 parsecs is known to harbor a planetary system. This is Epsilon Eridani, a K2 dwarf without any stellar companions located 3.2 parsecs (10.5 light years) distant. Epsilon Eridani is evidently a very young star, less than a billion years old. It is surrounded by extensive debris belts with an outer radius of about 105 AU (Backman et al. 2008). A single gas giant planet (1.55 Mjup) has so far been detected, with a period of 6.85 years and a semimajor axis of 3.39 AU (Benedict et al. 2006). Other planetary companions are possible, but planet “b” is probably the most massive object in the system. It may harbor an extensive family of icy moons. Depending on the eccentricity of this planet's orbit, terrestrial planets may or may not have been able to assemble in the system’s habitable zone, which extends from about 0.4 AU to 0.7 AU (Raymond et al. 2007).

Given the total population of 71 Sun-like stars within 10 parsecs, it is unclear why Epsilon Eridani should be the only exoplanetary host. Although planets are unlikely to form or survive in the vicinity of close binaries (those with periastron separations less than 20 AU), relatively few of the nearby Sun-like stars occur in such configurations. Not even the star’s metal content can explain its distinction. Enhanced stellar metallicity (greater than 0.0 and especially over +0.2) is associated with the presence of massive planets within a few AU of the host star (Fischer & Valenti 2005). Yet the metallicity of Epsilon Eridani is slightly less than Solar, at –0.03 (CNE).

In fact, despite good theoretical arguments to the contrary (e.g., Laughlin 2004, Raymond et al. 2007), current observations suggest that local M dwarfs provide environments at least as favorable for planetary systems as local Sun-like stars. Seven M dwarfs within 10 parsecs have so far been identified as exoplanetary host stars, and an eighth (GJ 436) has been detected just outside this limit. The seven nearby M dwarf hosts range in mass from GJ 317 at 0.24 Msol to GJ 176 at 0.50 Msol, and in metallicity from GJ 832 at -0.31 to GJ 849 at +0.16 (Bailey et al. 2008). Four of them (GJ 876, GJ 832, GJ 849, GJ 317) host gas giants, and the other three (GJ 647, GJ 581, GJ 176) host Super Earth to Neptune-mass planets. At least two systems (GJ 876, GJ 581) contain additional Super Earth planets, bringing the local total of M dwarf exoplanets to 11. Notably, none of the nearby systems harbors a Hot Jupiter. The general observation that massive star-grazing planets are the products of highly metallic environments (Fischer & Valenti 2005, Greaves et al. 2007) remains unchallenged.

In addition to gas giants, ice giants, and terrestrial-mass planets, several nearby stars have also been shown to harbor debris disks analogous to the Asteroid and Kuiper Belts of the Solar System. These range from hot, massive stars of spectral class A (Vega, Fomalhaut), through Sun-like stars of spectral classes G and K (61 Virginis, Tau Ceti, Epsilon Eridani), to cool young M dwarfs (AU Microscopii). Leading researchers into the debris disk phenomenon have called these structures “signposts of planetary system formation” (Trilling et al. 2007), implying that where there is debris, there will also be planets. Out of the neighborhood population, however, only two systems -- Epsilon Eridani and Fomalhaut -- show persuasive evidence of planetary companions. Continuing observations with increasingly sensitive methods should someday clarify the status of the other nearby debris disk systems, with Vega an especially likely candidate for planets.

Counting our Sun as well as Epsilon Eridani, the planet-hosting rate among Sun-like stars within 10 parsecs stands at 2.8% (2/71). The rate for M dwarfs within the same volume is quite similar, at 2.9% (7/239). Even as additional M dwarf stars are discovered nearby, the detection rate for exoplanets orbiting such stars is likely to remain steady or even increase. This prediction is based on the enormous progress in the study of local M dwarf systems that has occurred just in the past decade. Between 2000 and 2007, 40 new M dwarfs were announced within 10 parsecs. In a still shorter period of time, between 2005 and 2008, 5 of the 7 currently known M dwarf hosts were identified in the same volume of space.

Intriguingly, one of the four nearby A stars has been shown to harbor an exoplanet. This is Fomalhaut, for which the Hubble Space Telescope has directly imaged a gas giant orbiting just inside the system's far-flung debris belt (Chiang et al. 2008). The Hubble images bring the planet-hosting rate among neighborhood A-type stars to a whopping 25%. While this discovery may be a statistical fluke, it may alternatively provide evidence of a large planetary population around the Milky Way's bright stars.

Whether any additional F, G, or K stars within 10 parsecs may also be exoplanetary hosts is unknown. One key reason for pessimism is the low metallicity of most Sun-like stars in our immediate neighborhood.

elsewhere in the orion arm . . .

We can get a clearer perspective on our local 10-parsec sphere by comparing it with similar volumes of space elsewhere in the Orion Arm. For example, the Orion Nebula Cluster is a young star-forming region located about 415 parsecs away (1350 light years; Menten 2007), centered on the four bright stars of the Trapezium Cluster. Within a radius of only 3 parsecs, this region contains at least 3500 visible stars, ranging in spectral type from O through M (Hillenbrand et al. 1997). Its full population is more than an order of magnitude larger than our local sphere, so that the stars at its core in the Trapezium are separated on average by only 0.05 parsecs. Despite an abundance of bright O and B stars, however, this population is still dominated by newborn M dwarfs, in proportions similar to those found in the Solar neighborhood (Hillenbrand et al. 1997). All are younger than a few million years.

Substantially closer, at a distance of only 170 parsecs (550 light years; Kraus & Hillenbrand 2007), is the Praesepe Cluster. At an approximate age of 600 million years, this open cluster has long since lost its primordial hydrogen clouds, as well as a substantial fraction of its least massive stars. Within a radius of about 10 parsecs, Praesepe currently harbors about 1000 members, covering a range of spectral types (A through M) similar to those found in the immediate Solar neighborhood (Kraus & Hillenbrand 2007). Praesepe probably has at least three times as many star systems as our own 10-parsec sphere, with a higher proportion of binary systems.

An important difference between these two dense regions and the Sun's more sparsely populated back yard is our local diversity. Stars in clusters have identical ages and very similar chemical compositions, so that all members of a given spectral class reach the same evolutionary stage at the same time. The stars in our immediate neighborhood span a range of ages from just a few million years to more than 10 billion years, and a range of metallicities from an extreme low of about -1.50 (Groombridge 34) to a high of about +0.30 (HD 32147).

Star clusters are also relatively rare. Thus, in the context of our entire Local Bubble, the nearest 10 parsecs are quite packed. Although the disk of the Milky Way measures more than 30,000 parsecs (100,000 light years) from edge to edge, its vertical thickness at the so-called Solar Circle is only about 600 parsecs (Bonatto et al. 2006). Stars crowd the midplane of the Galactic disk, with population densities falling off above and below. Our Sun travels in the thick of the local crowd, as various recent estimates place it only 15 to 30 parsecs north of the Galactic plane (Bonatto et al. 2006). Star populations are likely to thin out substantially within a few hundred parsecs in either direction, north or south.