A Re-appraisal of the Habitability of Planets Around M Dwarf Stars

Jill C. Tarter,1 Peter R. Backus,1 Rocco L. Mancinelli,1 Jonathan M. Aurnou,2 Dana E. Backman,1 Gibor S. Basri,3 Alan P. Boss,4 Andrew Clarke,5 Drake Deming,6 Laurance R. Doyle,1 Eric D. Feigelson,7 Friedmann Freund,1 David H. Grinspoon,8 Robert M. Haberle,9 Steven A. Hauck, II,10 Martin J. Heath,11 Todd J. Henry,12 Jeffery L. Hollingsworth,9 Manoj M. Joshi,13 Steven Kilston,14 Michael C. Liu,15 16 Eric Meikle,17 I. Neill Reid,18 Lynn J. Rothschild,9 John M. Scalo,19 Antigona Segura,20 Carol M. Tang,21 James M. Tiedje,22 Margaret C. Turnbull,4 Lucianne M. Walkowicz,23 Arthur L. Weber,1 and Richard E. Young9

Abstract

Stable, hydrogen-burning, M dwarf stars comprise about 75% of all stars in the Galaxy. They are extremely long-lived and because they are much smaller in mass than the Sun (between 0.5 and 0.08 MSun), their temperature and stellar luminosity are low and peaked in the red. We have re-examined what is known at present about the potential for a terrestrial planet forming within, or migrating into, the classic liquid-surface-water habitable zone close to an M dwarf star. Observations of protoplanetary disks suggest that planet-building materials are common around M dwarfs, but N-body simulations differ in their estimations of the likelihood of potentially-habitable, wet planets residing within their habitable zones, which are only ~ 1/5 to 1/50 of the width of that for a G star. Particularly in light of the claimed detection of the planets with masses as small as 5.5 and 7.5 MEarth orbiting M stars, there seems no reason to exclude the possibility of terrestrial planets. Tidally locked synchronous rotation within the narrow habitable zone doesn’t necessarily lead to atmospheric collapse, and active stellar flaring may not be as much of an evolutionarily disadvantageous factor as has previously been supposed. We conclude that M dwarf stars may indeed be viable hosts for planets on which the origin and evolution of life can occur. A number of planetary processes such as cessation of geothermal activity, or thermal and non-thermal atmospheric loss processes may limit the duration of planetary habitability to periods far shorter than the extreme lifetime of the M dwarf star. Nevertheless, it makes sense to include M dwarf stars in programs that seek to find habitable worlds and evidence of life. This paper presents the summary conclusions of an interdisciplinary workshop (http://mstars.seti.org ) sponsored by the NASA Astrobiology Institute and convened at the SETI Institute.

Keywords: Planets, Habitability, M Dwarfs, Stars

1 SETI Institute 2 University of California Los Angeles 3 University of California Berkeley 4 Carnegie Institution of Washington 5 British Antarctic Survey 6 NASA Goddard Space Flight Center 7 Pennsylvania State University 8 Southwest Research Institute 9 NASA Ames Research Center 10 Case Western Reserve University 11 Ecospheres Project 12 Georgia State University 13 U.K. Meteorological Office 14 Ball Aerospace & Technologies Corp. 15 University of Hawaii 16 Alfred P. Sloan research Fellow 17 National Center for Science Education 18 Space Telescope Science Institute 19 University of Texas at Austin 20 California Institute of Technology 21 California Academy of Sciences 22 Michigan State University 23 University of Washington

1

I. Introduction The discovery of the first extrasolar planet (Mayor and Queloz, 1995), and the realization that Starting with the work of Huang (1959, 1960) type I and II migration plus n-body dynamics can and continuing with Dole’s “Habitable Planets for dramatically rearrange the geometry of planetary Man” (1964), stars whose masses greatly exceed or systems subsequent to their formation (numerous fall below the mass of the Sun were considered authors in Deming and Seager, 2003), and thus inhospitable to biology, particularly the complex, deliver and remove potentially habitable planets from intelligent variety. This conclusion stemmed from a habitable zone, expanded astronomers’ attention simple arguments. Massive stars live their lives too beyond planetary systems that were exact analogs of quickly – measured in millions of years – exhausting the solar system. This broadening of astronomical their nuclear fuel too rapidly (and self-destructing in perspective combined with the discovery of spectacular fashion) to permit the multi-billion year hydrothermal habitats that are apparently independent evolutionary timescales required on Earth to convert of the Sun (Corliss et al, 1979), an increasing stardust into beings capable of contemplating the appreciation for the incredible tenacity of stars. While low-mass stars have much longer extremophiles on Earth (Rothschild and Mancinelli, lifetimes, their luminosity is so feeble that any planet 2001), and the realization that M dwarfs comprise would need to be nestled very close to the star in perhaps 75% or more of the stars (excluding Brown order to permit the possibility of having a surface Dwarfs) and half of the total stellar mass in the temperature conducive to liquid water, which seems Galaxy (Henry, 2004), has once again raised the issue to be the sine qua non of life as we know it. In such of the suitability of M dwarfs as hosts for habitable small orbits (.1 to .35 AU for an M0 star, and closer worlds. Several authors (for example: Basri et al., still for smaller stars) any planet would become 2004) have even challenged their colleagues with tidally locked to the star, that is, it would keep the conference posters arguing that M dwarfs may be the same side continuously facing the star (an analog is favored locations for the origin and evolution of life. the Earth’s moon). Early, and perhaps overly We have every reason to believe that M stars are simplistic, arguments suggested that such a planet’s favorable hosts for planets. Protoplanetary accretion atmosphere would boil off on the star-facing side and disks are as common around low-mass stars as they freeze out on the dark side – not the most clement are around solar-type stars, with around half of all conditions for life. As a result, for decades, objects at ages of a few Myr possessing such disks exobiologists and SETI researchers excluded M (e.g. Haisch et al. 2001, Liu et al. 2003, Lada et al. dwarf stars from their consideration. However, in 2006, Sicilia-Aguilar et al. 2005). Older debris disks 1994, an international conference reconsidered the around M stars are thus far rare, but their paucity may question of circumstellar habitable zones and be due to the limited sensitivity of previous surveys. concluded that late K and early M dwarf stars One of the very nearest young M stars, the 12 Myr provided the best opportunity for environments that old M0.5 star AU Mic, at distance of only 10 pc, were continuously habitable over the 4.6 Gyr possess a prominent cold, dusty disk easily detected required by evolutionary timescales on Earth (Doyle by thermal continuum emission and scattered light 1996). Based on a simple energy-balance model, it (Liu et al. 2004; Kalas, Liu and Matthews 2004). was argued that atmospheric heat transport could The AU Mic disk likely represents a late stage of prevent freeze-out on the dark side (Haberle et al. planetary accretion in which planetary-mass objects 1996). More sophisticated 3D climate modeling by have stirred up remnant planetesimals to collide and Joshi et al. (1997) suggested that a surface pressure fragment in a Kuiper Belt-like structure. Thus, the of as little as 0.1 bar could prevent atmospheric potential for planet formation around M stars appears collapse on the dark side, while a surface pressure of to be robust. 1-2 bars could allow liquid water on most parts of the Ground-based radial velocity searches have been surface. However, the magnetic activity and looking for extrasolar planets around several hundred consequent flaring of M dwarfs were considered M dwarfs, with several planets found to date problematic because of the large flux of UV-B that (http://exoplanets.org , would be delivered to the surface of the planet. Later http://obswww.unige.ch/~udry/planet/planet.html). studies (Segura et al., 2005) indicated that abiotic The lowest mass exoplanets orbiting main sequence production of atmospheric ozone could mitigate any stars, with claimed masses of 5.5 and ≥7.5 MEarth, flare activity. However, planets in the habitable orbit M dwarfs (Beaulieu et al., 2006 and Rivera et zones of dwarf M stars would likely be so very al., 2005). Regardless of whether or not small unearthly, that they have not attracted a great deal of planets are preferentially formed around M dwarf attention until now. stars, the current detection technologies preferentially

2 detect them there. For a given planet mass and Although we may soon know whether radius, both the radial velocity reflexes and the appropriate planets lie within the habitable zones of photometric transit depths are larger for lower mass M dwarf stars, we are a long way from knowing stellar hosts. However, no rocky planets have yet whether they are likely to be inhabited. This paper been discovered within an M dwarf habitable zone, presents the summary conclusions of an and N-body simulations (Ida and Lin 2005; interdisciplinary workshop (http://mstars.seti.org) Raymond, Meadows and Scalo 2006; Montgomery sponsored by the NASA Astrobiology Institute and and Laughlin 2006) provide conflicting answers convened at the SETI Institute to consider what is regarding the likelihood of such planets. Ground- reasonably known about the topic, and what based observations using radial velocity techniques theoretical and observational projects might be as well as transit photometry may soon resolve the undertaken in the near term to constrain the issue. possibilities. Section II assesses what is known Plans for increasingly sophisticated space about M dwarf stars, the formation of planets around missions to search for terrestrial planets around M dwarfs, the probable atmospheric characteristics of nearby stars and search for biosignatures in their planets orbiting within the habitable zone of such atmospheres (e.g. Darwin http://sci.esa.int/science- stars, and the basic requirements for the origin and e/www/area/index.cfm?fareaid=28, CoRoT evolution of biology on any such planets. Section III http://sci.esa.int/science- makes a differential comparison between a G dwarf e/www/area/index.cfm?fareaid=39, and the and M dwarf stellar host, enumerates the most Terrestrial Planet Finders C and I favorable and unfavorable environmental conditions http://planetquest.jpl.nasa.gov/TPF/tpf_index.html) for biology (in a relative way), and examines those demand high fidelity finding lists of nearby stars processes that might terminate the habitability of a most likely to harbor habitable planets. If these favored planet long before the host M dwarf star should turn out to be M dwarf stars, then constraints ceases fusing hydrogen in its core. In conclusion we imposed by the planet/star contrast ratio will be suggest a framework for research to be conducted in eased, but the challenges of spatially resolving a the next two years that should better delimit the planet in the habitable zone will be significantly constraints on the spectral type and ages of M dwarf enhanced (Turnbull, 2005). The Kepler mission stars that would be appropriate targets for missions (Borucki et al. 2003) is scheduled to launch in 2008 searching for life beyond Earth, including and should provide, within a few years thereafter, the technological civilizations. A second workshop in demographics of terrestrial-size planets within the 2007 will provide the opportunity for re-examination habitable zones of stars as a function of spectral type, and further refinements to our approaches. from F to at least M5 dwarfs. The Allen Telescope Array is currently under II. Requirements for Habitability construction in Northern California (http://astron.berkeley.edu/ral/ata/) and efficient use II.a Nomenclature of its wide field of view for commensal SETI observations requires a large list of target stars To deal with large numbers and phenomenology around which habitable planets and their life forms they can measure, but perhaps not understand, might exist. In creating a list of ‘Habstars’ to be used astronomers have a long history of using to guide SETI observations with the Allen Telescope classification schemes that are not always intuitively Array, Turnbull and Tarter (2003a,b) did not understood by scientists in other fields or the general explicitly exclude M dwarfs, except for those public. At the risk of offending some well-schooled expressing high levels of chromospheric activity. readers, this section presents a mini-guide to those Even so, the magnitude limit of the parent Hipparcos classification schemes that are used in this paper to sample meant that proportionately few M dwarfs discuss the habitability of planets orbiting M dwarf ended up in the final target list (~600 M stars of stars. ~19,000 total habstars). A SETI target list of ‘M’ is a stellar classification based on the Habstars incorporates an additional selection criterion characteristic features found in relatively low related to the stellar age, requiring at least 3 Gyr for resolution spectra, initially obtained at optical evolution to have potentially led to technological wavelengths nearly a century ago. It was later civilizations. If M dwarfs are viable Habstars, then understood that the observable spectral features were additional efforts will be required to inventory the actually providing information about stellar surface local neighborhood, overcoming the biases of temperatures and mass, but by then the classification magnitude limited sampling. alphabet was well established running OBAFGKML from the hottest most massive star, to the coolest

3 objects, some of which may not be massive enough expended and continues to be invested in calibrating to be true stars and stably fuse H to He at their cores. different measurement results against one another. These alphabetical classifications also have fractional One approach is to attach a letter designation to divisions from 0 to 9, running from hottest to coolest magnitude measurements, where the letter indicates within the class. Colors are also used as an the spectral band and response of the receiver e.g. MV abbreviation for temperatures; as one might expect, denotes absolute magnitude as measured in the visual blue stars are hot and red stars are cool. And or V band. The difficulty comes in practice when not astronomers use the terms ‘earlier’ and ‘later’ to refer all ‘V’ bands are the same, but that level of detail is to the spectral sequence from hotter to cooler. beyond the scope of this paper. Common ‘Dwarf’ (as opposed to Supergiant, luminous and spectroscopic bands are: U (ultraviolet), B (blue), V normal Giant, Subgiant) is a luminosity classification (visual), R (red), and I (infrared), plus J, H, K (other and tells us that a star resides on the ‘main sequence’ near-infrared). The difference between the brightness (a locus of points in a plot of stellar luminosity vs. of an object in two spectral bands e.g. MB-MV is an temperature) where it will spend the majority of its astronomical color and designated (B-V). In contrast lifetime in a stable configuration with nuclear fusion to these wavelength-specific measurements, as its power source. These same named luminosity astronomers use the term bolometric to connote a classes are also numbered for abbreviation; I, II, III, property, such as luminosity, that includes IV, V runs from Supergiant to dwarf. Our own Sun contributions from all wavelengths. is spectral type G2 and luminosity class V (or dwarf). Finally, because of the ungainliness of Less massive dwarf stars are cool, such as M dwarfs. astronomical quantities when expressed in cgs or The length of time a particular star remains on the other scientific systems, astronomers tend to use main sequence, and the rate at which the end phases relative measures e.g. the mass of an object expressed 33 of its evolution cause it to expand, contract, change in terms of the mass of the Sun (Msun = 1.99 x 10 g), its surface gravity and therefore to appear within or the distance light travels in one year (ly = 9.46 x other luminosity, and spectral classifications is 1017 cm), or the parsec which is the distance of an determined by its mass. Massive stars consume their object whose measured angular parallax is 1 arc sec nuclear fuel rapidly, while M dwarfs remain stable (pc = 3.08 x 1018 cm). for a very long time. In the history of the Milky Way Most of the stellar members of the solar Galaxy, no M dwarf star has yet had sufficient time neighborhood, and presumably the Universe, are red to evolve away from the main sequence and end its dwarfs (low mass, cool, main sequence stars). They life. comprise at least 75% of all stars (excluding Brown Since the human eye is a logarithmic sensor, Dwarfs) and continue to be found at distances less astronomers have historically used a logarithmic or than 5 pc (Henry et al. 1997; Jao et al. 2005; Henry magnitude scale for recording and relating stellar et al. 2006). These objects span a huge range in brightness. On this scale, a difference of 5 in brightness (9 < Mv < 20), have masses between 0.5- magnitude implies a factor of 100 difference in the 0.6 and 0.08 Msun (Henry et al. 2004, Delfosse et al. brightness, and the sense of the comparison is such 2000), and have been assigned spectral type M. that a 6th magnitude star is 100 times fainter than a 1st Because of their overwhelming numbers, they magnitude star. The zero point of this scale is set by contribute more to the total stellar mass budget of the observations of A0 V stars, which are assigned a Galaxy than any other spectral type star, even at their magnitude of 0. To distinguish between the relatively low masses. Their intrinsic faintness brightness of a star as it appears on the sky, which makes them elusive, evidenced by the fact that not a depends on its distance, and the intrinsic luminosity single one can be seen with the naked eye. of the star, there are two magnitude scales; apparent Table 1 outlines the basic characteristics for stars magnitudes are designated as lower case m, and representing the broad range of M dwarf absolute magnitudes designated as upper case M, and characteristics, with our Sun as a benchmark. defined as the apparent magnitude that a star would Absolute brightnesses at optical (MV, 0.55 microns) have if it were at a distance of 10 parsecs (~33 light and infrared (MK, 2.2 microns) wavelengths are years) from the observer. Observations were provided, as well as a broad baseline color (V-K). historically made with the naked eye, then eyes with Guidelines for effective surface temperatures, and telescopic assistance, then recorded on photographic masses and luminosities relative to our Sun are also plates, and now recorded with instrumentation listed. It is worth noting that for the later M types, spanning a much larger range of frequencies than can the mass inferred from color temperature can depend be perceived by the human eye. Each of these on age (younger stars are hotter, then cool off). sensors has different sensitivities over different spectral ranges. Significant energy has been

4

Table 1 SpType MV MK V-K Temp Mass Lum Sun(G2V) 4.8 3.3 1.5 5800 1.00 100.% M0 V 9.0 5.3 3.8 3900 0.50 6.% M3 V 11.7 6.8 4.9 3600 0.29 3.% M6 V 16.6 9.4 7.2 3000 0.10 0.5% M9 V 19.4 10.5 8.9 2400 0.08 0.02%

Given that one magnitude corresponds to a factor other hand, the surface pressures are actually higher of ~2.5 in brightness, one notes that an M0 V star than for solar-type stars despite the low mass produces only 1.9% of the light in the visual band as (because gravity is an inverse square force). Because the Sun, but 16% of the light in the infrared K band. of the low outer temperatures and high pressures, the Corresponding values for an M9 V star are 1.4 X 10-4 atmosphere can support molecules as well as atoms, % at V and 0.13% at K. Thus, the balance of and the optical and infrared spectra are dominated by radiation relative to our Sun is very different for M molecular bands. In the optical these arise from dwarfs, which produce relatively more infrared heavy metal oxides like TiO and VO, not because radiation than visible. those molecules are abundant but because they The range of different stellar environments is happen to have high optical opacities. In the infrared larger for dwarf stars defined as spectral type M than the spectrum is dominated by steam and carbon for any other stellar type. The full gamut of stellar monoxide spectral features. spectral types runs OBAFGKML. To match the The high outer opacities and generally lower range in visual brightness of the M dwarfs temperatures throughout mean that the interior (magnitudes 9.0 to 19.4) more than four full spectral opacities are also higher, and the stars find it easier to types of stars are required: A0.0 V stars have MV~0, move luminosity outward by convection rather than whereas K7.0 V stars have MV~9. When considering radiation throughout their bulk. This convective stellar mass, arguably the single most important mixing has the effect of making the whole star characteristic of a star because mass determines available as nuclear fuel. Given that the Sun only nearly every other characteristic, the factor of six in burns 0.1 of its mass (due to its radiative core), and mass from M0.0 V to M9.0 V is again only matched coupled with the fact that M stars have much lower by spanning types AFGK. luminosities, this makes their hydrogen burning In summary, M dwarfs are low mass stars that lifetimes (the main sequence) much longer. These span a substantial mass range (~0.1 to ~0.5 solar can range from 50 Gyr for the most massive M stars masses) and exhibit a wide range of fundamental to several trillion years for the least massive ones. properties. During an extended youth (ages < ~ 1 The mixing also means that the core composition has Gyr), they are significantly more magnetically active an extremely slow gain in mean molecular weight, at UV and X-ray wavelengths than solar-type stars and a consequently very slow change in emitted flux. and exhibit powerful flares whose radiation could This is in contrast to stars like the Sun, in which inhibit the emergence of life. They dominate the local hydrogen is converted to helium in an isolated (non- stellar population by number, contributing convective) core, and the relatively faster increase in approximately half of the stellar mass in the Galaxy. mean molecular weight results in a steady brightening of solar-type stars even while on the II.b The Stellar Properties of M Stars main sequence. As a result, the habitable zone (discussed next) for an M star stays in place for much As described above, M stars have relatively low longer as well; and the continuously habitable zone is stellar temperatures, which makes their light red. a much larger percentage of it. Thus, if a planet exists They also have smaller radii (the size scales in the habitable zone of a typical M star, it may approximately with the mass), so that a 0.1 solar remain in that zone for 100 Gyr. In essence, mass object is roughly 0.1 solar radii as well. This therefore, M stars last so long that the length of accounts for their low luminosities, although one habitability becomes more of a planetary than stellar could also say that these low luminosities are caused issue. An important exception to this could be the by the low central pressure and temperature required effect of magnetic activity on the star; we defer to support the lower mass against gravity, which discussion of this until Section IIIa. leads to much smaller nuclear fusion rates. On the

5 II.c The Habitable Zone of magnitude drier than the surface (as it is on Earth), to being almost as wet. The rate of hydrogen The habitable zone (HZ) is a concept that that is production from water photolysis in the middle used extensively in the context of searching for atmosphere increases accordingly, as does escape of planets that might be suitable abodes for life. There hydrogen to space. Over time, the planet's surface are a number of different definitions of the HZ (e.g.: dries (Kasting and Pollack 1983). A number of Dole 1964; Heath et al. 1999), but here we define it caveats on this mechanism, such as climate as that region around a star in which a planet with an feedbacks, are present, and are reviewed in Kasting et atmosphere can sustain liquid water on the surface. al. (1993). This is because liquid water is basic requirement for An M dwarf presents specific issues for the life on Earth, so its presence would appear to be the habitability of planets that lie in the HZ. For decades, most important criterion for including a given the main problem was thought to be tidal locking; a location in any search for life, if indeed any selection planet in the habitable zone of an M dwarf would lie of locations is desirable in the first place. An so close to its parent star that it was very likely to additional reason for the choice of this definition is become tidally locked. In other words, the planet that extrasolar planets on which liquid water and life always presented the same face to its star, much as are present at the surface should be observable Earth's moon does to the Earth (Dole 1964). It was spectroscopically in a search for evidence of life assumed that a consequence of locking was that (Leger et al., 1993; Angel and Woolf, 1996; Tinetti et atmospheric volatiles would freeze out on the dark al. 2005), whereas subsurface biospheres may not be side on the planet, leaving an atmosphere where the detectable. surface pressure is controlled by a balance between The nature of the HZ around F, G, and M stars the latent heat of condensation and radiative cooling has been dealt with elsewhere (Kasting et al. 1993; on the dark side. Such an atmosphere would have a Heath et al. 1999), and the reader is referred to those surface pressure orders of magnitude below that of papers for detailed overviews of the subject. We only present-day Earth, and far below the triple point of summarize some of their results here. The outer and water - the minimum pressure at which water can inner edges of the HZ are defined as those values of stably exist as a liquid. It would therefore be the stellar insolation (SI) at which liquid water ceases considered not habitable. to be stable at a planetary surface. In terms of actual The above view was first challenged by Haberle size, the HZ around an M-dwarf will be smaller in et al. (1996), who used an energy balance model that size than the HZ around a G star, because of the parameterized atmospheric heat transport to show inverse square law governing radiation from the that an atmosphere containing only 100 mb of CO2 parent star, and because of the range of M star masses transported enough heat to the dark side to prevent the width of the habitable zone will vary by an order freezing of volatiles. The result of Haberle et al. was of magnitude from 0.2 AU to 0.02 AU. The given significantly more weight by Joshi et al. seriousness of this constraint on habitability, given (1997), who explored the sensitivity of surface the likelihood of a planet forming at a given distance temperature gradient to factors such as surface from its star, is discussed in section III.b. pressure and atmospheric IR optical depth using a 3D For the outer edge, Kasting et al. (1993) global circulation model which actually simulated the considered that on geological timescales, the amount climate and weather of the atmosphere. They, like of CO2 in the Earth's atmosphere is controlled by Haberle et al. (1996), found that approximately 100 processes such as weathering. The dependence of mb of CO2 was enough to prevent atmospheric weathering rate on temperature ensures that if a collapse. Joshi (2003) showed that inclusion of the planet's surface freezes, weathering ceases, CO2 hydrological cycle (including advection of water builds up in the atmosphere, and thus warms the vapor) on a tidally-locked Earth did not destabilize planet up again. This negative feedback acts to the climate, and achieved qualitatively similar results extend the width of the habitable zone outwards in to Joshi et al. (1997). terms of distance from the parent star, and to lower Note that the above studies are "worst-case" values of SI. The limit to the negative feedback, i.e.: scenarios. For instance, not all close-in planets the outer edge of the HZ, occurs at that value of SI at become tidally locked; another possibility is that they which CO2 condenses in the atmosphere, rendering have a spin/orbit resonance like Mercury, which the weathering feedback irrelevant. rotates 3 times every two orbits around the Sun. If a The inner edge of the habitable zone can be planet is not tidally locked, or the effect of heat defined as that point at which a planet's stratosphere transport by oceans is included, the temperature (which on Earth lies between 15 and 40 km in difference between day and night will be significantly altitude), changes in composition from being orders lower than the model predictions. It is perhaps

6 counterintuitive to think that atmospheric motions destabilize climate. A tidally locked planet in the HZ can dramatically change the surface environment of a of an M dwarf would have an orbital period of days planet. Nevertheless, the models do show that while to weeks (Heath et al. 1999), and so the climatic tidal locking might alter the expected edge of the HZ consequences of an eccentric orbit would be heavily (see below), it can no longer be considered a serious damped by a planet's atmosphere and oceans. We barrier to habitability. further discuss the effects of stellar variability, and For a given SI, the substellar point (the location indeed orbital variations, in section IV.b.vi below. directly beneath the star) is warmer than if the stellar The effects of flares, which might present an obstacle radiation was distributed evenly in longitude (Joshi to habitability (Kasting et al. 1993; Heath et al. 1999) 2003). Thus one might expect the inner edge of the is also considered later in III.d. The potential effects HZ (as defined by a wet stratosphere) to happen at a of Coronal Mass Ejections are discussed in papers by lower value of SI than is the case for a G star. On the Khodachenko et al. (2006) and Lammer et al. (2006) other hand, climate model studies show that the in this issue. planetary albedo of an ocean covered synchronously To summarize, the limits of the HZ around an rotating Earth is 0.35, which is 20% higher than if M-star might be expected to be defined by similar tidal locking did not happen (Joshi 2003). A higher processes to those that define the limits around a G albedo would move the inner edge of the HZ nearer star. However, the atmospheres of planets in the HZ its parent star. The rise is due to the presence of ice of M stars would certainly be very unfamiliar to us in and clouds, and works in the opposite direction to the terms of their circulation, radiation and chemistry. relatively low planetary albedo expected of a planet However, if planets can keep their atmospheres and orbiting an M dwarf (Kasting et al. 1993). water inventories, none of these differences present The distribution over wavelength of the large obstacles to their potential habitability. ultraviolet radiation from M dwarfs could lead to a Therefore, from the point of view of atmospheric and different photochemistry in their planets, moving the climate science, planets in the HZ of M dwarfs outer edge of the HZ away from the star, because should have almost as high a chance of being gases such as CH4 or N2O might exist in much higher habitable as planets in the HZ of G stars. concentrations than on Earth (Segura et al. 2005). The CO2 condensation limit would no longer apply, II.d. Planets Within the Habitable Zone as even without any CO2, significant greenhouse gases would exist in the atmosphere. This effect The formation of planets around M stars requires would make the habitable zone wider. that the protostar be surrounded by a protostellar disk Kasting et al. (1993) used the concept of a with sufficient material. Primordial (optically thick) continuous habitable zone (CHZ) to take account of circumstellar disks are readily detected by their changes in SI over billions of years. Since M dwarfs strong thermal IR emission, far above that produced are very long lived, the CHZ might exist for an order by stellar photospheres. Both ground-based and of magnitude longer in time around an M dwarf than space-based IR surveys have identified many disks around a G dwarf. The time that life would have to around low-mass stars and brown dwarfs in the take hold on a planet orbiting an M dwarf would nearest young star-forming regions (e.g. Muench et therefore be much longer, and life might therefore be al. 2001; Liu et al. 2003; Jayawardhana et al. 2003; more likely on such a planet. Such considerations are Luhman et al. 2005), with about half of the low mass of course based on the premise that a planet orbiting population (spectral type of M0 to M9) having disks. an M dwarf could keep a substantial atmosphere In at least some cases, the masses of these primordial despite the various loss processes that would disks appear to be quite substantial, about a few inevitably occur. This is discussed in section III.c percent of the stellar host mass (e.g. Andrews and below. Williams 2005). Thus, it appears that M stars are as While the amplitude of long-term variability of likely a venue for planet formation as more massive SI on M dwarf is small, short-term variability might stars. present a problem. For instance, stellar radiation can Less theoretical work has been done on planet oscillate by several percent on timescales of days due formation around low mass stars, because of the to rotationally modulated starspots (Rodono 1986). emphasis on explaining the origin of our own However, Joshi et al. (1997) showed that an planetary system. Our terrestrial planets are atmosphere having a surface pressure of 1 bar would universally thought to have formed through the not freeze out even if subjected to a starspot that collisional accumulation of successively larger solid reduced SI by 40%, and lasted a month. In addition, a bodies -- starting with sub-micron-sized dust grains, tidally locked planet would be prevented from through kilometer-sized planetesimals, Moon-sized undergoing large swings in obliquity that might planetary embryos, and culminating after about 30

7 Myr in the formation of the Earth and the other consists largely of silicate minerals, mostly olivine terrestrial planets (Wetherill 1990). Wetherill (1996) and pyroxenes (Draine and Lee 1984; Adamson, extended his Monte Carlo models of collisional Whittet, et al. 1990; Greenberg and Li, 1996) or of accumulation to include the case of lower-mass stars. amorphous condensates in the multicomponent He found that Earth-like planets were just as likely to system MgO-FeO-SiO2 (Rietmeijer, Nuth, et al. form from the collisional accumulation of solids 1999; Rietmeijer, Hallenbeck, et al. 2002; Rietmeijer, around M dwarfs with half the mass of the Sun as Nuth, et al. 2002). Judging from the relative they were to form around solar-mass stars. More intensities of the silicate and “organic” bands in the recent modeling by Ida and Lin (2005) investigated infrared spectra a surprisingly large fraction of the the effects of varying stellar (as presumably mineral dust grains in the diffuse interstellar medium protoplanetary disk) metallicities and mass. The across our galaxy and even in neighboring galaxies results are consistent with the detection statistics on consists of organics, about 10% (Sandford, exoplanets and predict a peak in formation of Allamandola, et al. 1991; Pendleton, Sandford, et al. Neptune-mass planets in short orbits around M stars. 1994). Most of these organics are expected to survive Simulations by Montgomery and Laughlin (2006) intact the accretionary processes leading to suggest that terrestrial mass planets form consistently planetesimals and to comet-sized bodies. A smaller in the habitable zone of late type M stars, whereas but still significant fraction may survive collisions Raymond, Meadows, and Scalo (2006) find that it between planetesimals and even cometary impacts on may only be the early type M stars that form larger planets such as the Earth (Chyba 1989). In this terrestrial planets within the habitable zone having way organics formed in the outflow of distant dying sufficient mass to retain an atmosphere and a stars (Dey, Van Breugel, et al. 1997; Matsuura, favorable water content. Zijlstra, et al. 2004) that persisted for hundreds of Because the collisional accumulation process is millions of years in the interstellar medium may have inherently stochastic, it is not possible to make contributed to the pool of complex organic molecules definitive predictions regarding the outcome of a on the early Earth from which life arose. given set of initial conditions for the protoplanetary The organics associated with the interstellar dust disk, but basic trends can be discerned, of which have the spectroscopic signature of fully saturated perhaps the most important is the effect of Jupiter- (aliphatic) hydrocarbons, in which C atoms are linked mass planets on terrestrial planet formation to other C atoms by single bonds and each is bonded (Wetherill 1996). If a Jupiter-mass planet forms prior to either two or three H atoms. They are essentially to the final phases of terrestrial planet formation, as straight or branched Cn chains, where n can be a large seems to be necessary to explain the gaseous bulk number, such as in carboxylic (fatty) acids or composition of a Jupiter-like planet, then the location parafinic sections linking more sturdy polyaromatic of such a massive gravitational perturber can have a hydrocarbons (Sandford, Allamandola, et al. 1991; controlling influence on the formation of Earth-mass Pendleton, Sandford, et al. 1994; Pendleton and planets in the habitable zone of solar-mass stars: a Allamandola 2002). How the delicate saturated Jupiter closer than about 4 AU to its star could hydrocarbons survive the intense UV radiation field prevent the growth of Earth-mass planets at 1 AU. It that permeates the diffuse interstellar medium and the may well be necessary for life to have a long-period incessant bombardment by high-energy particles has Jupiter to scatter water-bearing planetary embryos long been a question of great interest and discussion. toward the early Earth (Morbidelli et al. 2000) and Various models have been proposed, of which then much later on to intercept and reduce the the “core-mantle” concept (Greenberg 1968) has cometary flux onto the inner planets that might received the widest attention (Sandford, Allamandola otherwise frustrate the evolution of advanced life et al. 1991; Allamandola, Bernstein et al. 1999). It is though periodic catastrophic impacts (Wetherill based on the assumption that dust grains forming in 1994). N-body simulations with many more particles the outflow of stars acquire a layer of ice composed by Raymond, Quinn and Lunine (2006) suggest that not only of H2O but also other gaseous components the delivery of water to terrestrial planets near the such as CO, NH3 etc. Through UV photolysis in the habitable one is statistically more robust than ice matrix organic molecules would form which, previously thought, particularly if the water-bearing upon sublimation of the ice, will remain on the grain embryos form slowly. A long-period Jupiter might surface as a thin veneer (Greenberg, Li, et al. 1995). then be a prerequisite for a habitable Earth-like Such a thin veneer of organic matter on the surface of planet. grains, however, is not expected to survive for long Interstellar dust is the main contributor to the the intense radiation field and constant sputtering in formation of accretionary disks and, thence, to the the diffuse interstellar medium. Another model formation of Earth-like planets. The interstellar dust (Mathis, Rumpl, et al. 1977) postulates separate grain

8 population of silicate minerals and hydrogenated dwarf stars, and perhaps at somewhat smaller orbital amorphous carbon to account for the observed distances. However, the longer orbital periods at a infrared spectral features and the dielectric properties given distance from a lower mass star mean that core of the dusty interstellar space (Draine and Lee 1984; accretion will generally be too slow to produce Draine 2003; Zubko, Dwek, et al. 2004). It is in Jupiter-mass planets in orbit around M dwarfs before disagreement with the observational fact that the the disk gas disappears (Laughlin et al. 2004). In the silicate-to-organics ratio throughout our galaxy and competing disk instability model for gas giant planet even other galaxies is surprisingly constant, at about formation, however, calculations for M dwarf 10:1, in spite of very different conditions of dust protostars (Boss 2006a) show that Jupiter-mass formation in different astrophysical environments. clumps can form in less than 1000 years in Recently a dust grain model was presented, marginally gravitationally unstable disks with masses based on the laboratory observation that C-H bonds of 0.02 to 0.07 solar masses orbiting around stars can form inside the mineral matrix (Freund and with masses of 0.5 and 0.1 solar masses. These Freund 2006). It treats the organics associated with ongoing models suggest that there is no reason why the dust as Cn–Hm entities embedded into the mineral M dwarf stars should not be able to form gas giant matrix of the dust grains, forming a single phase solid protoplanets rapidly, if disk instability occurs. Hence solution. This model appears to be most consistent it may well be that mini-solar systems, similar to our with the astronomical observations (Sandford, own except reduced in spatial scale, are able to form Allamandola, et al. 1991; Pendleton, Sandford, et al. in orbit about M dwarf stars. Zhou, Asrseth, Lin and 1994). It explains both the complexity and hardiness Nagasawa (2005) have recently proposed an of the organics in interstellar space and their observational test to discriminate between these two survivability through time and space. models of planet formation. Numerous short period While a general consensus exists regarding the terrestrial planets can be expected to be found in basic mechanism for terrestrial planet formation, the systems with Jovian mass planets, as the result of situation is quite different for the formation of giant sequential accretion, but not as the result of planets. The conventional wisdom is that gas giant gravitational instability. However, models of planets form by the process of core accretion, where terrestrial planet formation by Kortenkamp, a roughly 10 Earth-mass solid core forms first by Wetherill, and Inaba (2001) have shown that rapid collisional accumulation, and then accretes disk gas formation of gas giant planets by disk instability can to form a massive gaseous envelope (Pollack et al. actually facilitate the growth of inner rocky planets, 1996). However, even when the surface density of contrary to the assumptions made by Zhou et al. solids is high enough for runaway accretion (Lissauer (2005). Ground-based spectroscopic surveys have 1987) to assemble a seed mass for the solid core in begun to detect planets in orbit around M dwarfs, about 0.5 Myr, the timescale for subsequent growth with masses ranging from about 7.5 Earth-masses to and accretion of the gaseous envelope can be Jupiter-mass and above (Rivera et al. 2005). Boss sufficiently long (typically a few to 10 Myr) to raise (2006b) has shown that disk instability in the outer the danger that the disk gas may have been dissipated regions coupled with collisional accumulation in the well before the solid cores could accrete enough gas inner regions is able to explain the super-Earths and to grow to Jupiter-mass. Instead, “failed cores” gas giants found in orbit around M dwarfs by both similar to Uranus and Neptune might result. An microlensing and spectroscopic planet searches. alternative mechanism has been proposed and investigated, where gas giant planets form rapidly II.e Atmospheres of Habitable Planets through a gravitational instability of the gaseous portion of the disk (Boss 1997, 2001). Disk Since the circulations of terrestrial planetary instability can occur within thousands of years, well atmospheres are primarily driven by the spatial before the disk dissipates, and if this mechanism can distribution of short wave radiation, one can expect occur, it would presumably outrace the core accretion that the atmospheres of planets orbiting M-dwarfs mechanism. will be significantly different in character to the Boss (1995) studied the thermodynamics of terrestrial paradigm, both in terms of circulation and protoplanetary disks around stars with masses from composition. Again, for a fuller description of the 0.1 to 1.0 solar mass, finding that the location of the atmospheric circulation on tidally locked planets, the ice condensation point only moved inward by a few reader is referred to Joshi et al. (1997), which used a AU at most when the stellar mass was decreased to simple two-stream gray radiation scheme to model IR that of late M dwarfs. In the core accretion model of emission and assumed that all solar radiation hit the giant planet formation, this implies that gas giant surface, and Joshi (2003), which explicitly modeled planets should be able to form equally well around M the effects of the hydrological cycle, and had higher

9 horizontal and vertical resolution. The effect of a the same way as the Western Pacific region, which water cycle was to have more uplift on the dayside has the hottest sea surface temperature, appears very and downwelling on the nightside than in the dry cold when seen in the IR. Such observations might model, resulting in different flow at low levels across actually provide evidence for a planet with an active the terminator. The strengths of the east-west, or hydrological cycle. Significant efforts have been zonal, jet streams in the upper layers of each model expended on predicting and interpreting the spectra to were also different as a result of the different be observed by the TPF-C and TPF-I missions configurations. seeking to directly image any terrestrial planets in the It appears to be quite commonly thought that habitable zones of nearby stars. The moisture content even if atmospheric collapse is prevented on tidally and average planetary albedo are critical to these locked planets, the large thermal gradients and models, as are considerations of phase and smearing associated high winds across the terminator inhibit over orbital cycles (Segura et al. 2003). A great deal habitability, especially forest habitability. However, more research needs to be done on the topic. atmospheric heat transport tends to reduce thermal Photolysis processes on a planetary atmosphere gradients, both by reducing their amplitude, and by are controlled by the UV radiation of its parent star. smearing them out to larger horizontal scales. Figure Because of the peculiar flux distribution over 8 of Joshi et al. (1997) shows the so-called radiative- wavelength of the active M dwarfs and the low convective equilibrium temperature field (obtained radiation in the ultraviolet region of the spectrum by running the atmospheric model without the effect from quiescent M stars, the atmospheric chemistry of of large-scale horizontal motions) and the actual trace species may be greatly affected. Simulations of time-averaged temperature field. The former case has planets that lie in the habitable zone of an M star, and large gradients on the terminator, which are have the same atmospheric composition and input of completely smeared out in the latter case. Thermal biogenic compounds as present-day Earth, show that gradients, and any potential high wind speeds the atmospheric lifetimes of compounds like associated with them, should therefore not be methane, nitrous oxide and methyl chloride are larger considered a barrier to habitability. than on Earth (Segura et al. 2005). As a result, the The terminator is thought to present the most abundance of these compounds could be potentially likely location for life, due to the relatively small three orders of magnitude more than their terrestrial effect of flares at that location (potentially damaging abundance (see Table 2). Planets around quiescent M radiation travels through a longer atmospheric stars (referred as model on Table 2) show a “methane pathlength suffering greater degradation, and the runway”, i.e.: methane builds up in the atmosphere absolute flux is diminished by the slant angle). because there is not enough UV radiation to destroy However, the terminator would also receive far less this compound in the stratosphere and the main sink sunlight than the substellar point or SP (the location of methane, OH, is highly reduced in the troposphere where the parent star is directly overhead) and so (see discussion on Segura et al. 2005). To avoid this forest habitability might be impeded here (Heath et problem a methane mixing ratio of 500 ppm is set for al. 1999). The presence of thick clouds associated the simulated planets around the quiescent M stars. with convection would have an effect on this The methane flux needed to maintain this mixing conclusion, by changing both the amount of direct ratio is 21% of the present methane production on and diffuse solar radiation at the SP. Plants at this Earth for a planet around the coolest star considered location might be less susceptible to flares for in Segura et al. (2005). Since methane is a instance. greenhouse gas, this result points to the possibility of The remotely-observed IR properties of a tidally extending the outer limit of the habitable zone for locked planet will be greatly affected by the presence planets that produce methane either biologically or of an atmosphere with an active hydrological cycle. not. On a dry world, or a planet whose atmosphere has For life detection purposes these biogenic gases collapsed, IR emission mostly comes from the may be more detectable on planets around quiescent surface, so that the dark side will appear far colder or active M dwarfs than on Earth (Figures 5 to 8 in than the dayside when viewed at these frequencies. Segura et al. 2005). Finally, a planet with the present However, with a hydrological cycle comes the concentration of O2 in the habitable zone of an active presence of water vapor and clouds – the SP will M dwarf may be able to develop an ozone layer as have clouds present for most of the time, and so IR large as the terrestrial one which will protect the emission here will come mostly from the cold cloud planetary surface from UV radiation (Tables 2 and 3 tops. Indeed, the dayside of such a planet might from Segura et al. 2005). actually appear as cold in the IR as the darkside, in

10

Table 2. Mixing ratios of biogenic gases calculated by a 1-D photochemical model for Earth-like planets around different stars (Segura et al. 2005). Methyl Parent star Methane Nitrous oxide chloride Sun 1.6×10-6 3.0×10-7 5.0×10-10 AD Leo (M4.5V)a 4.6×10-4 1.3×10-6 1.4×10-6 GJ 643 (M3.5V)a 3.8×10-4 1.1×10-6 8.8×10-7 b -4 -5 -7 M1 model (Teff=3650 K) 5.0×10 3.5×10 1.2×10 b -4 -4 -7 M3 model (Teff=3400 K) 5.0×10 1.0×10 2.0×10 b -4 -3 -7 M5 model (Teff=3100 K) 5.0×10 1.3×10 5.4×10

a For these planets the surface flux of these compounds was consider to be the same as present Earth: 14 13 12 9.54×10 g CH4/yr, 1.32×10 g N2O/yr, and 7.29×10 g CH3Cl/yr. b -4 On these planets the boundary conditions were: fixed mixing ratio for CH4 (5.0×10 ), fixed deposition -4 -4 velocities for H2 (2.4×10 cm/s) and CO2 (1.2×10 cm/s), and fixed surface fluxes for N2O and CH3Cl.

Another aspect of synchronously rotating planet happened to approximately match the gray optical habitability that needs attention in future work is how depth of the present terrestrial atmosphere with ~ 350 the carbonate-silicate cycle would control the partial ppm CO2 (τ~0.9), as well as an atmosphere having pressure of the atmospheric greenhouse gas CO2 1,000 mb pure CO2 (τ~1). It is therefore not (pCO2) and thereby climate on timescales of > 0.5 necessary to invoke high pCO2s to prevent freeze-out Myr. Changes in outgassing rates from the interior of on synchronously rotating planets – a conclusion that the planet and in continental weathering regimes, was repeated in Joshi (2003), which used present-day caused by changes in the distribution of continents atmospheric composition. across climatic zones would be expected to be These results meant that it was not necessary to associated with adjustments in pCO2. As mentioned limit discussions of habitable synchronously rotating earlier, Kasting et al. (1993) used the classic work by planets to planets on which pCO2 was constrained to Walker et al. (1981) that demonstrated how the extreme values, or to invoke and explain special carbonate-silicate cycle could act as a natural circumstances in which this was possible with Earth- thermostat to dampen climatic change, and to level insolation. Once again, then, the discussion of establish the limits of the habitable zone. The forest-habitability was found to follow on naturally greenhouse gas CO2 enters the atmosphere through from the climatic models without unduly contrived outgassing from the Earth’s interior, and is removed conditions. from the atmosphere as it dissolves in rainwater to There has been no investigation of the possible form dilute carbonic acid. In the event that insolation long-term evolution of climate in relation to the increased, the effect on weathering processes would carbonate-silicate cycle on synchronously rotating be to adjust pCO2 downwards. This would reduce the planets; values of pCO2 for different continental atmospheric greenhouse effect and so compensate for distributions have not been estimated. This is a major increased insolation. Reduced insolation would result area for future work, and the results will be of in an increased pCO2 so that an accentuated fundamental importance to taking discussion of greenhouse effect would compensate and prevent a synchronously rotating planet habitability forward. geologically active planet from freezing over. The geophysical regime of a synchronously In the early climatic models for synchronously rotating planet will play an important part in rotating planets, rather high pCO2s of up to 1,000 mb determining whether or not it is habitable. On a and 1,500 mb were assumed, but it is now clear that planet with plate tectonics, the drift of plates through extreme values are not needed. The model of Joshi et the colder zones near the terminator would extinguish al. (1997) actually employed a gray radiation scheme Earth-type arboreal forms, whilst drift onto the dark whose optical depth τ could be specified side would spell extinction for a wide range of independently of the surface pressure. The only organisms. Alternatively, one-plate planets such as aspect of their model atmosphere that was actually Mars and Venus with histories of long-term geologic “CO2” was the gas constant. A value of τ of 1.0 was activity (e.g., Basilevsky et al. 1997; Hauck et al. chosen for their control scenario because this 1998; Head et al. 2001) may also have regions of

11 persistent habitability. In particular, volcanically ~380K (>100oC) to below 233K (-40oC), the absolute active localities can act as a source of juvenile or limit for under-cooled water at a pressure of 1 bar). recycled volatiles and can generate surface heat The discovery of so-called extremophiles that have fluxes commensurate with plate tectonics settings. expanded the boundaries of habitable real estate has It was pointed out in Heath et al. (1999) that dramatically changed perceptions about the even if a layer of ice were to form on the dark side of possibility of life beyond Earth, a modern review of a synchronously rotating planet, Earth-like levels of currently understood limits on habitation can be geothermal heat from the interior of a geologically found in Rothschild and Mancinelli (2001). Life as active planet should ensure that oceans should not we know it does, however, depend on the presence of freeze to their floors. Bada et al. (1994) had advanced liquid water, and the existence of the liquid state this mechanism to demonstrate that liquid water ultimately sets the boundary conditions for life on environments would have been possible on an early Earth. Not only does water provide the solvent for Earth receiving reduced insolation from the early biochemistry, but there are very few physiological Sun, even in the absence of a massive atmospheric processes in which water does not play a part as greenhouse effect. What it means for a synchronously reactant or product. rotating planet is that supply of snow to the top of a Life on Earth depends on solution chemistry in thick layer of sea ice should be balanced by melting water, a solvent having unique physical and chemical at its base, and so, if deep ocean basins communicate characteristics, a set of chemicals primarily between the dark and lit hemispheres, a vigorous consisting of a few simple and common atoms (C, H, hydrological cycle is possible. This mechanism N, O, P and S), and relatively small differences of demonstrates one way the hydrological cycle and free energy (Ball 2004; Benner, et al. 2004,). Within geological activity are linked, and future research organisms, energy tends to be transduced using should investigate how as a planet’s internal heat flux electrical gradients across membranes, or by the declines with time, and the thickness of the oceanic oxidation of organic molecules to generate energy- crust declines, the volume of liquid water beneath the rich molecules (typically containing phosphate) and ice responds. organic electron pair carriers (often involving sulfur) However, it should be noted that the connection that are used to drive the synthesis of essential bio- between climate and geological activity is not one- molecules. Such energy transduction processes directional. Rather, geologic activity, especially frequently involve the use of transition metals, through volcanism, may be necessary to provide making their presence in trace amounts necessary. appropriate greenhouse gasses to the atmosphere. In Because of the complexity and intimate the case of some planets, such as Venus, there may involvement of water in all life processes on Earth, be a closed feedback loop where increasing or because its four valence electrons endow carbon with decreasing atmospheric temperatures temper the extraordinary versatility to create multiple chemical amount of geologic activity (e.g., Phillips et al. bonds, and because both water and carbon are 2001). Water also strongly affects the melting and common in the universe, it is often assumed that life deformational behavior of the silicate minerals that elsewhere, if it exists, will be based on liquid water make up the crust and mantle of terrestrial planets and carbon. Alternative physiologies may exist, (e.g., Mackwell et al. 1998) and hence affects the however, that may widen the range of physical type and vigor of geological activity such as has been conditions over which life could exist. Extrapolation suggested for Venus (Kaula, 1995). However, water from the one example of life known to us would, alone likely does not explain even the diversity of however, suggest that the two solvents most likely to tectonic regimes within our own Solar System. For act as a basis for life are water and ammonia. In the example, like Venus, Mars is a one-plate planet, absence of evidence, however, it seems pragmatic to however it has clearly has had water throughout its confine our search to life based upon water and history. Geodynamical regimes of geologically carbon (Ball 2004, Benner et al. 2004). active Earth-sized planets subject to substantial tidal Given the basic requirements outlined above, we torque remain to be investigated. are therefore looking for a rocky planet, with subsurface or surface liquid water (the latter requiring II.f. Requirements for Life an atmosphere), that is geologically active to provide a continual recycling of elements, and capable of The envelope of physical conditions for life on supporting liquid water for a sufficient period of time Earth is in many ways remarkably wide. for life to arise and evolve. The question of life on Environmental pH can range from <1 to 12, pressure planets around M dwarf stars thus centers on whether from <1 to >500 bar, salinity from zero (freshwater) planets matching the requirements for life and the to saturation in saline lakes, and temperature from essential elements could exist within the continuously

12 habitable zone of their host star, and how these In summary, if a rocky planet orbiting within the conditions might be altered by the effects of tidal habitable zone of an M dwarf star possesses the locking on the climate, and of radiation on life. chemical constituents necessary for life, liquid water, Since we are also interested in the opportunities and these conditions persist for a sufficient time to afforded by M dwarfs planets to enable evolution of allow life to originate (a timescale that is not yet technological civilizations (of interest to SETI constrained) and evolve (again this is relatively searches), it is worthwhile to describe the unconstrained, but the bias is certainly toward requirements of ecosystems with multicellular billions of years) then there is no reason to conclude organisms, such as plants, that has been termed that the planet is not habitable. A habitable planet is ‘forest-habitability’ (Heath, 1996). On Earth, forests more likely to produce a technological civilization if dominate biomass at the crust-atmosphere interface. evolutionary histories are convergent rather than It is estimated that in the absence of human contingent. Recently Vermeij (2006) has provided interference, forests might cover around 40% of our new data that favor convergence of evolutionary planet’s land area, and account for 60% of net innovations, citing information loss over geologic primary productivity in land environments. This time as the reason for the appearance of singular, massive photosynthetically-produced biomass, contingent innovations. It is therefore reasonable to implies that photosynthesis is a highly favorable speculate that the end point of convergent solution to the problem of harvesting and exploiting evolutionary processes adapting to the very un- natural energy sources, not merely an accident of Earthly environments on a planet around an M dwarf history. Further support for this is that fact that there might not look anything like us, but they might exist alternative forms of autotrophy amongst the nonetheless be sufficiently technologically complex prokaryotes, and yet none have dominated any but as to warrant investigation by SETI researchers. small, specialized niches. Forests also play an important role in biogeochemical cycles, in the III. A Differential Analysis Between G Dwarfs hydrological cycle, and in modification of climate and M Dwarfs (notably through control of transpiration, and by changing ground albedo). In addition they have III.a Magnetic Activity and Stellar Evolution provided an important environment for life, including that in which our own primate ancestors developed The first flaring M star was discovered in 1947 key adaptations concerned with hand-eye-brain co- by Edwin Carpenter, who noted a 3 magnitude operation that would later prove valuable in the brightening of UV Ceti (M5.5e, 2.7 pc) in 12 evolution of tool-making hominids. minutes. By 1970, 50 flare stars were known and it The temperature tolerance of higher plants on was recognized that they were nearly all mid- to late- Earth is narrower than the -40°C to >100°C range of type M stars lying within 15 pc of the Sun microorganisms. For example, the sclerophyll trees (Gurzadyan 1970). By the 1990s, the number had and shrubs include forms that suffer cold injury at doubled. A much greater number of flare stars was 5oC - 2oC and heat injury at 50oC - 60oC. The cold found in young open clusters; the Pleiades alone has o o and hot limits of CO2 uptake were 5C - 0 C and over 500 observed with another 500 estimated to be 45oC - 50oC respectively (Larcher, 1995). The present (Ambartsumian et al. 1970). A considerable temperature ranges at which CO2 uptake is 50 % that phenomenology of M star flares has been garnered in at the temperature optimum are 15oC - 20oC to 40oC - the optical and ultraviolet bands (Houdebine 2003 45oC. Also in the Earth’s seasonal regime, full-sized and references therein, West et al. 2004, Walkowicz, trees are found where mean daily temperatures Hawley and West 2005). The U band flux is exceed 10oC for a minimum of one month a year. correlated with Balmer line emission for M dwarf Therefore on synchronously rotating M dwarf stars whose levels of activity span 5 orders of planets, where there are no seasons, we might take magnitude. Chromospheric surface fluxes in the 10oC isotherm as the cold limit for trees. hydrogen Balmer lines can reach 108 erg/s/cm2, as Although another planet would have its own unique seen in the young system AU Mic. Flare intensity biosphere, these Earth-biased considerations suggest and duration are correlated from 1 to 100 minute that the temperature range of M-star ‘forest- decay phases. The chromospheric cooling budget is habitability’ would probably be near 10°C - 50°C. often dominated by ultraviolet continuum, rather than Furthermore, since the Joshi et al. (1997) model emission line. Stars earlier than M7 maintain a predicts wind speeds near the surface of typically 5 to persistent quiescent chromosphere and transition 10 m s-1 (even at the terminator) no special region between flares as indicated by Hα activity, biomechanics would be required for tree growth. with later type stars remaining active longer (Hawley and Johns-Krull 2003, and Silvestri, Hawley and

13 Oswalt 2005). Significant fractions of the stellar M stars, flare emissions in the different wavebands photosphere can be covered with long-lived, cool, are largely decoupled from each other (Smith et al. strongly magnetized starspots. 2005). The activity of M stars follows a rotation-activity The activity in electromagnetic radiation is relation, as seen in hotter solar-type stars, but only undoubtedly accompanied by high energy particle when the rotational velocities are below 4 (10) km/s fluxes as well. Since the X-rays in flare stars are for mid- (late-) M dwarfs (Mohanty and Basri 2003). harder (have relatively more high energy photons) For the more rapidly rotating stars, the dynamos or than from solar flares, it is reasonable to assume that surfaces of most M stars appear to be fully saturated. the particles may be more energetic as well. They The fact that M stars are convective, conductive, and also sometimes expel substantial quantities of mass at spin means that they can generate magnetic fields by high speed (mostly in the form of energetic protons). dynamo action. When they are fully convective Such ejections may be a substantial contributor to the (those M3 and later) the cyclical solar-type magnetic early mass loss rates, which could be as high as 104 dynamos (which operate at the radiative-convective greater than the current solar rate (Mullan et al. interface) cannot work, and the magnetic fields must 1992). (These mass ejections can also affect the come from partially or fully turbulent dynamos. This atmosphere of a planet in the habitable zone likely means that the typical field structures tend to (Khodachenko et al. 2006 and Lammer et al. 2006). be smaller, the sensitivity to rotation is less, and this We also know, however, that by the age of Proxima may account for the generally longer spindown time Centauri (several Gyr) and likely well before that, the for M stars (perhaps through the field geometry). mass loss rate has fallen below the solar rate (Wood There is no evident transition in activity properties et al. 2001). Thus, it is unlikely that an M star loses a around M3-M4 when the radiative core disappears significant amount of its mass either in coronal mass and the interior becomes fully convective. This is not ejections or a steady stellar wind. Perhaps the best understood and suggests either that a saturated evidence for this is that we do observe spindown in convective dynamo dominates the field generation or most M stars. The spindown is a feedback that it alters the interior structure of the cooler stars mechanism which has the seeds of its own (Mullan and MacDonald 2001). destruction; the slower the star spins the less activity We do know that M stars can generate strong and mass loss it has. fields when they are young, the fields are stronger The M flare stars identified prior to 1970 than solar due to the higher gas pressure, and can constituted roughly one-tenth of the dM population in cover nearly the whole star (Saar and Linsky, 1985; the solar neighborhood (Gurzadyan 1970). Johns-Krull and Valenti, 1996; Valenti and Johns- Comparison of flare star populations in clusters of Krull, 2001). These lead to flares which also different ages showed a decay of activity over several sometimes reach stellar dimensions (Orsten et al. hundred million years, and several groups found that 2005; Güdel et al. 2004) and create bursts of the lowest mass M stars exhibit a longer flare star luminosity which are a greater fraction of the total phase than higher mass M stars (Mirzoyan 1990; stellar luminosity than the strongest solar flares. The Stauffer et al. 1991; Hawley et al. 2000). However, a range of flare intensities is very wide (up to 105 or kinematic study of 93 UV Ceti stars within 25 pc of more), with young flare stars at the maximum values. the Sun showed a fairly high velocity dispersion σ = Flare luminosities decrease with stellar mass, roughly 30 km/s indicating a mean age around 3 Gyr and no maintaining a maximum ratio of 10-4 with the dependence on mass (Poveda et al. 1996). A small bolometric luminosity (Pettersen 1991). Of course, fraction of these nearby flare stars are either similar energies are also emitted at shorter identifiable members young disk groups or members wavelengths, into the X-rays. A number of recent X- of an old thick disk population. ray studies with simultaneous multiwavelength Two recent studies give new insight into the coverage provide a close-up view of individual flares. long-term decay of M star activity. Silvestri et al. On Proxima Centauri, low level X-ray emission (2005) examine a sample of 139 older M stars whose formerly labeled "quiescent" is now clearly shown to ages can be estimated from their white dwarf be the superposition of many small flares and companions. Using V-I color as a chromospheric modeling a powerful X-ray flare implies a loop size activity indicator, they find that many dM and dMe comparable to the stellar radius (Güdel et al. 2004). stars with ages 1-10 Gyr are more magnetically A continuous study of EV Lac (M4.5e, 5.1 pc) over active than predicted by the decline of chromospheric two days revealed a powerful radio flare activity seen in younger clusters with ages <1 Gyr. accompanied by an impulsive U-band event, Feigelson et al. (2004) obtained a small sample of 11 numerous optical white light flares, and many X-ray X-ray selected stars from a very deep Chandra X-ray flares are seen (Osten et al. 2005). Here, and in other Observatory field at high Galactic latitude. Seven of

14 these are dM stars at distances 50-500 pc and latter indicative of mass loss at the rate of up to10-10 represent the high-activity tail of older stars in the solar masses per year. Following this procedure, upper disk. Most of these stars exhibit X-ray flares Doyle et al. (1996) put an upper limit on mass loss with peak luminosities around 1027 – 1028 erg/s, from two young solar-type stars, the smaller upper similar to young flare stars. Comparison with limit being about 9 x10-11 solar masses per year, evolution models suggests a steep decay law Lx ~ t-2 indicating also what the effect of such mass loss in flare activity. This is consistent with X-ray studies would have been on the early solar system (the of dM stars in younger cluster populations and the circumstellar habitable zone migrating inward while solar neighborhood which indicate a flare decay law the planets migrated outward). The migration of the of Lx ~ t-1 for ages <1 Gyr and steepening at later circumstellar habitable zone due to mass loss, ages (Preibisch and Feigelson 2005). migration due to tidal friction (from the hydrological The fraction of M stars with Hα emission cycle on a synchronously rotating planet), and the denoting strong magnetic activity was known to be possible loss of a tidally-locked planet’s atmosphere low among early-M stars, increases to nearly 100% due to a reduced magnetic field, have been pointed around M5.5-7 stars, and declines rapidly among out in Doyle (2006) and Lammer et al. (2006). early-L stars (Joy and Abt 1974; Gizis et al. 2000). We calculate what level of stellar mass loss The largest samples studied to date consists of 499 M would significantly affect a planet’s habitability. As dwarfs from the volume-complete Palomar/MSU the star loses mass its luminosity decreases with the survey of nearby stars, and nearly 8000 M- and early- result that the Habitable Zone (HZ) will contract. L-type stars from a flux-limited sample obtained But also, stellar mass loss will cause the planet’s from the Sloan Digital Sky Survey (Gizis et al. 2002; orbit to spiral outward by conservation of angular West et al. 2004). The Palomar/MSU sample momentum. Life would be made more difficult to suggests that the magnetic activity of M0-M3 stars is sustain on such a planet by the combined effects of bimodal with most exhibiting no Hα emission, some stellar mass loss on the position of the habitable zone exhibiting strong emission but few at intermediate and the planet’s orbital position. A planet initially levels. The Sloan sample shows the fraction of dMe within the habitable zone could find itself outside stars peaks at 75% of M8 stars with >50% between after sufficient mass loss by the parent star. M5 and L0; the somewhat lower fraction of dM3 The calculation proceeds as follows: stars is probably due to the higher fraction of older Letting m = stellar mass, mp = planet mass, L = stars in this high Galactic latitude survey. However, stellar luminosity, r = planet orbit radius (orbit using a more quantitative measure, the average assumed to be always circular for slow mass loss), LHα/Lbol=2 x 10-4 from M0 to M5 and declines to 5 then in order to keep constant insolation: r2 ∝ L x 10-5 for M7-L0. The Palomar/MSU sample shows (applies to center or boundaries of HZ). an average around LHα/Lbol =1 x 10-4 from M0 to Main sequence luminosity evolution M3, followed by an increased range of activity (with (theoretically understood to be driven by stellar a small decline in average LHα/Lbol) from M3 to structure and nuclear fusion rates) depends on stellar M5. mass, with L ∝ mq. For high-mass stars, q ~ 4.5, but for M dwarfs the luminosity relationship has flattened III.b Truncation of Planetary Habitability to q~2.5. Therefore since r ∝ Ll/2, constant insolation requires that the radius decrease with mass as r ∝ Although the stable lifetimes of M dwarf stars m1.25. But in order to conserve orbital angular can be counted in the hundreds of gigayears, momentum of the planet as the parent star loses mass, habitable planets that form initially may not remain mpvr must remain constant, and therefore the radius habitable for the lifetime of the star. must increase in time as r1/ro = mo/m1. A factor of two characterizes the inner and outer III.b.i Truncation of Planetary Habitability: radii of habitable zone for the solar system (about Mass Loss and the Evolution of Planetary Position 0.75 to 1.5 AU in our solar system). So for stellar Relative to the Habitable Zone mass change to cause a planet starting out at the inner edge of the habitable zone to drift to the outer edge of Mass loss from young solar-type stars was first the (shrinking) HZ requires that r1/ro = 2 and suggested by Whitmire and Doyle (1995) as an -1/(1+1.25) therefore, m1/m0 = 2 = .73. Therefore the star explanation for the faint Sun paradox (Sagan and can lose at most about 25% of its mass before a Mullen 1972). Using the –2/3 spectral index excess planet once resident at the inner edge of the HZ indicative of a stellar winds at radio wavelengths, moves out beyond the outer edge. Mullen et al. (1989; 1992) placed limits on the stellar -14 At the Sun’s mass loss rate of 2 x 10 Msun per wind from a K2 dwarf and two M-dwarf stars, the year, a main sequence M star would take several

15 hundred billion years to lose 25% of its mass. Wood fluid motions within the liquid portion of the planet’s et al. 2005, find that the sustained mass loss rate from metallic core (Olson et al. 1999; Stevenson, 2003). M dwarfs is far below solar, so even an initial burst Such fields can create a magnetosphere that of mass loss cannot produce any substantial effects surrounds the planet, buffering it from the on orbital change. Migration out of the habitable surrounding space environment and limiting the zone due to sustained mass loss of the M dwarf star is atmospheric loss rate from cosmic ray sputtering unlikely to be a contributing factor to early truncation (Johnson, 1994; Grießmeier et al. 2005). The critical of habitability, a terrestrial planet that starts in the ingredients for planetary dynamo action are a habitable zone will remain there. sufficient volume of electrically conducting fluid, an energy source to drive motions in the fluid and net III.b.ii. Truncation of Planetary Habitability: organization of the flow field (Roberts and Planetary Interiors Glatzmaier, 2000; Aurnou, 2004). Typically, Coriolis forces produced by planetary rotation act to For planets of both M and G dwarfs, truncation organize the planetary core flow. Planetary rotation of planetary habitability is likely to occur with the periods within the M dwarf habitability zone are loss of large-scale endogenic geologic activity. Loss estimated to be between 10 – 100 days. These of heat via convective heat transfer in a terrestrial rotation rates, although slower than that of Earth, will planet’s silicate mantle is the primary driver of still produce strong Coriolis forces in ~ 100 – 1000 endogenic geologic activity over solar system km deep core fluid layers. Thus, in terms of dynamo timescales. Mantle convection, in turn, can prove theory (Stevenson, 2003), tidally locked planets important for chemical recycling, for mantle orbiting M dwarfs are viable candidates for core degassing of atmospheric components and for dynamos. sustaining sufficient internal heat flow to power a Recent numerical studies of planetary dynamo magnetic dynamo in the planet’s metallic core. The action propose the following scaling law for lifetime of mantle convective motions will tend to planetary magnetic dipole moment (Olson and increase with i) the thickness of the mantle (which Christensen, 2006; Christensen and Aubert, 2006): will tend to increase with planetary radius) and ii) the radioactive element content that provides heat during 1/3 3 M ~(QB D) RC (1) the decay process. The time scale over which radioactive decay is relevant is roughly 3 – 10 Gyrs where M is the magnetic dipole moment, QB is the timescales for Earth-like planets (e.g., Sleep, 2000). buoyancy flux, D is the thickness of the dynamo However, the dynamics of mantle flow may be rather generation region, and RC is the radius of the planet’s different on planets considerably more massive than core. This study has been carried out using only Earth due to the effects of extreme pressures and dipole-dominant dynamo solutions. Multipolar temperatures on mantle materials (e.g., Van den Berg numerical dynamo solutions also exist, although no et al. 2005). In addition to planetary radius, the other clear scaling behavior has been identified for these primary control on the depth of the mantle is the cases (Christensen and Aubert, 2006). In (1), note relative amount of metallic iron in the planet. For that the scaling predicts, quite surprisingly, that the example, the Earth’s metallic core comprises ~55% planetary dipole moment does not depend on the of the planet’s radius compared with ~75% inferred rotation rate of the planet. This does not mean that for Mercury (e.g., Lodders and Fegley, 1998). rotation rate is irrelevant in the dynamo process. It Therefore, bulk composition and internal structure of means, instead, that in the regime where the core 1-10 Earth-mass planets (e.g., Valencia et al. 2006) convection is dominated by the effects of planetary will also affect the truncation of terrestrial planet rotation, the typical strength of the dynamo field is habitability. Assuming typical enrichment in controlled by the amount of available buoyancy radioactive elements, planetary geological activity on power (Christensen and Aubert, 2006). Thus, in planets orbiting M dwarfs is likely to follow a terms of planetary dynamo action, a fast or slow similarly broad spectrum as observed in our solar rotator is correctly defined by the ratio of buoyancy system. In addition, tidal dissipation may be an forces versus Coriolis forces (Aurnou et al. 2006), important additional mantle heat source on bodies in not by the ratio of a planet’s rotation rate versus that eccentric orbits, as occurs on the Galilean satellites of the Earth. A planet that is a fast rotator must have (Showman and Malhotra, 1999; Moore, 2003). a buoyancy flux that is large enough to generate core Internally generated planetary magnetic fields convection, but not so large that the buoyancy forces are just as likely to occur on planets orbiting M swamp out the Coriolis forces that lead to the dwarfs as on planets around G dwarfs. Planetary magnetic fields are generated on terrestrial planets by

16 production of a well-organized planetary-scale process is most efficient for small planets with weak magnetic field. gravity fields, and is believed to drive the escape of O, C, and N at Mars (Mc Elroy, 1972; Hunten, 1993; III.b.iii. Truncation of Planetary Habitability: Fox and Hac, 1997). However, the only flux of Atmospheric Escape escaping material measured at Mars is due to ion escape wherein the electric field associated with the The ability of planets orbiting M stars to retain solar wind directly accelerates ions to the escape their atmosphere will depend critically on the nature velocity (Lundin et al. 1989). This process differs of atmospheric escape processes that such planets from sputtering in that the latter occurs when the ions experience. There are a variety of ways an reimpact the atmosphere with sufficient energy to atmosphere can escape to space. Perhaps the most eject all molecules at the exobase (mainly C, O, CO, well known are the thermal escape mechanisms of N, N2, and CO2). Both ion escape and sputtering are Jeans escape and hydrodynamic escape (see Walker, facilitated by the lack of a magnetic field (which 1977 for details). Jeans escape occurs above the allows direct interaction of the solar wind with the exobase where molecular collisions are infrequent atmosphere) and are sensitive to the UV flux. Thus, and some of the upward traveling molecules have M star planets lacking a strong enough magnetic field high enough velocities to escape to space. This type are susceptible to loss by each of these non-thermal of escape will occur in any planetary atmosphere escape mechanisms. with the escape flux being dependent on planet mass. This is far from a simple problem. Although For the terrestrial planets, only the lightest elements some effort has been spent on the evolution of the (hydrogen and helium) can escape by this processes atmospheres of close-in hot-Jupiters (Grießmeier et and the total loss over the lifetime of the solar system al. 2004), very little research on atmospheric escape has not significantly depleted their atmospheres. from M-star planets appears in the published This type of escape is also likely to be literature. Clearly, however, this is a major issue and inconsequential for M star planets. needs to be investigated. Hydrodynamic escape, however, may be more of a factor. This kind of escape occurs when III.c. Origin vs. Sustainability of Biology thermospheric temperatures are high enough such that the thermal energy of molecular motions is Heath et al. (1999) explored the opportunities for comparable to the kinetic energy at escape velocity. known types of higher plants in terms of three main In this situation the escaping molecules, generally factors, the distribution of climatic zones across the hydrogen, expand into the vacuum of space with such surfaces of synchronously rotating planets, the intensity that they can drag the heavier elements with amount of Photosynthetically Active Radiation them (e.g., C, O, N, etc.). Though still hypothetical, arriving in M star insolation, and stellar variability. hydrodynamic escape has been invoked to explain the Photosynthetically Active Radiation (PAR), fractionation of the xenon isotopes in the from about 4,000 Ǻ to 7,000 Ǻ, is not simply a atmospheres of Mars and Earth (Bogard et al. 2001; random part of the electromagnetic spectrum that just Hunten et al. 1987). It is believed to have occurred happens to have been picked by plants during the very early in solar system history when the solar course of evolution. It is rather of special biological EUV flux was much higher than it is today because it interest because it is neither so energetic that it is the absorption of solar EUV that powers damages cells, nor so weak that it cannot power hydrodynamic escape (Chassefiere, 1996). Since M water-splitting photosynthesis. As regards a perfect stars are very active in the EUV for much longer black body, the wavelength of peak energy output is periods of time than G stars (Allard et al. 1997), it inversely proportional to its temperature (Wien’s raises the possibility that hydrodynamic escape could displacement law). However, stars are more complex significantly deplete the atmospheres of the planets than this, because light emitted from any level of the that orbit them (Scalo et al. 2006). photosphere (a layer hundreds of kilometers deep) Non-thermal escape mechanisms include can be absorbed and re-emitted by the higher layers dissociative recombination, ion escape, and of the star, in a manner determined by stellar sputtering (Chassefiere et al. 2004 and Lammer et al. composition. The proportion of IR in M star 2006). Unlike thermal escape, which mainly affects insolation is substantially greater than for our Sun, the lightest elements, non-thermal escape can remove and as effective photospheric temperatures drop heavier elements and molecules. Dissociative through the range for spectral class M, the amount of recombination occurs when UV photons ionize PAR falls. Molecular bands of TiO2 are particularly molecules, which then recombine with electrons important in reducing emitted stellar radiation in the producing energetic neutrals that can escape. This PAR range, and red dwarf stars achieve energy

17 balance by back-warming of their continua at other not used efficiently by plants on Earth, because when wavelengths, notably around 10,000 Ǻ. Stars that chlorophyll absorbs a photon of blue light, this will formed out of material with low contents of “metals” promote an electron to an unstable upper excited -12 (astronomical parlance for elements heavier than singlet state that decays within a mere 10 second. hydrogen and helium) should put out the most PAR This means that both blue light photons and less but will have had smaller amounts of material with energetic red light photons produce the first excited which to form terrestrial planets. These would be singlet, as the starting point for energy transfer stars of the ancient Galactic halo and the older stars (Nobel, 1974). Moreover (for example, Salisbury and of the disk, whilst stars with the highest metallicities, Ross 1978), action spectra of certain crop plants and those of the Galactic bar/bulge and the intermediate trees are very much biased towards red light (blue and young disk populations will have more material spruce does not respond at all to radiation of to form terrestrial planets, but suffer the greatest wavelength < 5,000 Ǻ). There is also an effect known reduction in PAR output. as the red drop, whereby photosynthesis is impaired Rough estimates were made of the amount of when plants are illuminated only with radiation of > energy in and adjacent to the PAR region by 6800 Ǻ. However, they can benefit from the Emerson reference to published (Allen, 1973) stellar Effect (Emerson et al. 1957) whereby the efficiency luminosities in the U, B, V, R and I bandpasses and of photosynthesis at around 7,000 Ǻ is boosted if a to synthetic spectra for stars and substellar brown shorter red wavelength, such as < 6500 Ǻ, is present, dwarfs in the Teff range 4,000 to 2,000 K (Allard and and the combined level of photosynthetic efficiency Hauschildt, 1995). It was estimated that that the is greater than that expected from the sum for both amount of PAR which would be received at the top wavelengths taken separately. Heath et al. (1999) of a planet’s atmosphere (assuming identical also outlined the role of accessory pigments in insolation to that on Earth) from a star of Teff 4,000 K increasing light-harvesting in environments with (slightly hotter than a M0 star) would be roughly a reduced PAR (such as under water), and of known third, and the Photic Zone Window Radiation (PAR types of non-water splitting near-IR photosynthesis, in the range that penetrates clear ocean water most noting the possibility of linked photosystems effectively; 4,500 Ǻ to 5,500 Ǻ) would be a quarter harvesting in the near-IR and transferring energy to that incident on the Earth. The respective figures for water-splitting photosynthesis. a star of Teff 2,800 K are less than a twelfth and a There will be both advantages and disadvantages twentieth. These values are, nevertheless, sufficient from having the sun stationary in the sky. When the for photosynthesis to be metabolically profitable. sun casts illumination from a changing direction, its Most of the Earth’s forest trees are C3 plants, and light can penetrate beneath the forest canopy from they are often found to reach light saturation at a different angles at different times, benefiting foliage fraction of full sunlight (Black 1973). in different layers, including that of the understory. In Heath et al. (1999) also considered the way in the case of the Earth, seasonal changes allow PAR which light penetrates Earth-like atmospheres, having and summer warmth to be distributed to higher to pass through greater air masses with distance from latitudes, whilst rains in many areas are strongly the sub-stellar point. The efficiency with which light seasonal. On a synchronously rotating planet, passing through the atmosphere is subject to Rayleigh photosynthetic surfaces could be permanently angled scattering is inversely proportional to the fourth perpendicular to incoming light beams, and for the power of the wavelength, which means that red light hottest M dwarfs, continuous daylight and absence of (which is much more important than blue light in red a yearly migration of the sun between the tropics dwarf insolation), will suffer significantly less would mean that reduced PAR could be compensated scattering than blue light, and transmission effectively at the sub-stellar point. Unfortunately, the coefficients for direct irradiation of light of 7,000 Ǻ o model of Joshi (2003) indicates that this part of the out to about 60 from the sub-stellar point would be globe will be subject to possible climatic extremes, so about 80 % or greater than that for the sub-stellar other parts of the daylight hemisphere will need to be point. Some blue light will be down-scattered from considered in terms of the optimum location for the sky as a diffuse component, but the blueness of photosynthesis. This is a problem presently under the sky would fade for stars of progressively later investigation. In his model with a global ocean, there spectral type. Detailed studies of the direct and was intense cloudiness, not to mention massive diffuse illumination available on planets of M dwarfs precipitation that might have implications for erosion remain to be undertaken, and would benefit from of both soil and bedrock. In his simulation with the ongoing studies of M star spectra. northern hemisphere covered with land, cloud cover It is important to note that photosynthesis need was reduced, but temperatures could reach 80oC in not be impaired with red-biased light. Blue light is some regions, which is decidedly unhealthy for

18 Earth-type higher plants. Other land-sea damage caused by UV were discussed, and it was combinations, or changing the distance of the planet concluded that flares need not be a fatal problem for from its star need to be explored in future work. either micro-organisms or higher plants. Howbeit, the existence of such problems regarding These optimistic conclusions were based, the PAR-rich sub-stellar point, where light has to however, on limited data about flaring on dwarf M penetrate minimal thickness of atmosphere, cannot stars and it is essential that they be reconsidered in but detract from the global potential for the light of future observation programs that photosynthetic productivity of a synchronously investigate flare output across the spectrum. rotating planet. Recently Grießmeier et al. (2005) have Starspots and flares pose potential problems for considered the biological implications of cosmic ray life, but in the analysis of Heath et al. (1999) they impact on synchronously rotating terrestrial planets would not be insurmountable. The climatic in habitable zones around M dwarfs. They conclude implications of starspots had been modeled in Joshi that because of tidal locking, the magnetic moment, et al. (1997), who had noted that spots on red dwarf and therefore protection against primary cosmic rays, stars can reduce insolation by 10-40% for a few Earth will be reduced, allowing ~50-100% of the planetary months. A 40% decrease for 4 months would reduce surface to be reached by secondary cosmic rays that surface temperature by 27 K in some areas, and may can have biological consequences. allow regional temperatures to fall below the freezing It would appear from preliminary considerations point of water. However, many Earth tree species that red dwarf stars may well boast planets that have evolved to cope with seasonal cycles where support not only super-robust prokaryotic-grade temperatures fall far below freezing during the organisms, but organisms comparable to our higher winter, whilst within the Earth’s boreal forest zone, plants. Notwithstanding, many problems are evident annual temperature maxima and minima range from with the environments available on planets, and this about -40oC to 70oC. conclusion must be treated with caution. There is Heath et al. (1999) discussed the danger posed certainly a case to be answered. As regards to the by stellar flares, which can be commonplace from relative ease of initiating life on Earth, or on planets dwarf M stars. The key issue here is the amount of orbiting red dwarf stars (which may not, in any case, flare UV that could actually penetrate a planetary have had time to enter tidal lock before the atmosphere to ground level. Atmosphere-penetrating appearance of life), it is hard to offer a comparison, UV has been subdivided (Koller, 1965) into UV-A since (despite no shortage of hypotheses) we know so (3150 Ǻ to 3900 Ǻ), UV-B (2900 Ǻ to 3150 Ǻ), and little about the origin of life on Earth. UV-C (<2900 Ǻ). UV-B is the most dangerous, since it is not only biologically damaging, but can reach the IV. Lifting the Veil ground in significant amounts. Although peak germicidal efficiency occurs at ~ 2600 Ǻ, not much IV.a Input Data Required for Accurate Modeling UV-C arrives at the Earth’s surface. The U-bandpass used by astronomers (3310 Ǻ to 3990 Ǻ in Allen As has been stated through this paper there are 1973; 3260 Ǻ to 3940 Ǻ in Hawley and Pettersen, several issues concerning the interaction between a 1991) corresponds quite closely to UV-A, so apart planet and its parent M star that have not been from certain data obtained from above the Earth’s addressed yet. In part this is related to the intrinsic atmosphere by satellite (the International Ultraviolet characteristics of the M dwarfs, for example, they are Explorer satellite had provided observations in the fainter than other stars making difficult to measure ranges 1,150 Ǻ to 2,000 Ǻ and 1,900 Ǻ to 3,100 Ǻ.) their spectra, especially on the short wavelengths there was limited information about UV-B and UV-C (ultraviolet, X rays) unless the star is very active. in M dwarf flares. Other problems are related to planetary processes that The biological implications of an increased need to be better understood for any planet regardless proportion of UV during M star flares were put into the spectral type of its parent star, like atmospheric perspective with reference to the very low quiescent escape and geologic activity. Habitability analysis UV outputs of M dwarfs, and flare data, for specific must also consider stellar age, as the flaring activity stars. Data, both ground-based and from the appears to decay over the first 1-3 Gyr International Ultraviolet Explorer satellite, were used To study the ways in which an M star affects the to show that in many cases UV output reaching the chemistry and climate of a planet in the habitable surface of a planet would be comparable to that zone it is necessary to have spectra ranging from the provided continuously on Earth from the Sun, or ultraviolet to the infrared. Such spectra from the blue exceed terrestrial norms in transient events. part of the spectrum to the infrared are available for a Biological means of avoiding, tolerating or repairing number of nearby M stars with large apparent

19 brightness. However much of the photochemistry in more subtle than the cloud patterns, but may also be a planetary atmosphere is driven by ultraviolet detected (Ford et al.2001; Tinetti et al. 2005). photons, so spectra below 300nm are needed. In this However, time-resolved spectroscopy of different wavelength region, spectra exist only for a few of the views of the planet, combined with 3-D planetary most active M stars, not only because of the interest spectral models will be required to determine the in understanding the processes that drive their fractional coverage of different surface types (Tinetti activity, but because they are bright enough in the et al. 2005). To determine planetary atmospheric UV and shorter wavelengths to allow spectra to be composition, radiative transfer and climate models recorded above the instrumental noise. will be required to simulate the observed spectrum In order to obtain a better sense of the average and infer the presence and abundance of atmospheric radiation environment that a planet in the habitable gases. If some observational data is missing or zone would be exposed to, it will be necessary to unobtainable, modeling will also allow us to obtain UV spectra of M stars with a range of activity constrain possible planetary characteristics (see e.g. levels. Flares affect the visible spectrum and because Kasting et al. 1993; Joshi et al. 1997; Segura et al. of that it is important to have simultaneous 2005). observations of the UV and visible, in particular the blue part of the spectrum. Such observations will IV.b Planned Research for the Near Term require new space instrumentation, given that the Hubble Space Telescope was the only means for IV.b.i. HST ACS Prism Snapshot Survey of M obtaining UV spectra of astronomical objects, and Stars in the Near UV that previous instruments (e.g. IUE) only detected the few most active M stars. The situation is improving The surface habitability of a planet is a as the recently launched GALEX satellite is already complicated balance between having enough UV to providing new insights into the near- and far-UV provide the oxygen atoms needed to form ozone, and emission of M star flares (Robinson et al. 2005; having enough ozone to shield the planet's surface Welsh et al. 2006a and 2006b). from the UV. Preliminary studies of M dwarf In the future, observations of terrestrial planets habitability zones indicate that typical chromospheric orbiting M-dwarfs may be obtained by ground-based UV flux will produce observable ozone layers, radial velocity measurements, the Kepler transit comparable to that of Earth, on terrestrial planets mission, and NASA’s Space Interferometry Mission, around M dwarfs (see Figure 3b in Segura, et al. and Terrestrial Planet Finder-Coronagraph and - 2005). These results also indicate that the 9.6 micron Interferometer missions. Using data obtained from ozone band will be more prominent than for Earth these missions as input to models, we will be able to due to colder stratospheres on these planets (the derive information on the actual climate of these ozone band is seen in absorption-- a cold stratosphere planets. above a warm surface will produce a deep absorption For 1-D radiative-convective climate modeling, feature; Segura et al. 2003). Thus, it may actually be the planetary mass, radius, orbital parameters, albedo, easier to pick out ozone absorption in these planetary and atmospheric composition provide a atmospheres than for an Earth-like planet around a comprehensive set of model input data. These star like our Sun. In addition, vertical mixing ratio parameters will allow us to model the planetary calculations for terrestrial planets around M dwarfs surface temperature, which is the most crucial point toward an increased atmospheric concentration planetary characteristic to determine habitability. For of biomarker gases N2O and CH4 relative to Earth. 3-D climate modeling, other parameters such as This result is due to the differing UV energy planetary and atmospheric rotation rate, and the distribution from the parent M dwarf, and it is distribution of ocean and land are important to notable that these increased atmospheric determine the day-night and equator to pole variation concentrations are found for planets around both of surface temperature. active and inactive stars (Segura et al. 2005). Planetary radius and mass (and hence surface However, the stability of planetary atmospheres gravity) could be determined from the combination of around inactive M-stars is based on stellar model radial velocity and transit detection observations photospheres in the FUV, rather than actual (Henry et al. 2000; Charbonneau et al. 2000). Time- observations of such stars. resolved photometry will most likely reveal Despite these intriguing results and those variability in planetary cloud cover, if present, and discussed earlier in this paper, modeling of the may assist in the determination of atmospheric habitable zones around M dwarfs is severely limited rotation rate. Signals from changes in viewed surface by the lack of spectral data in the near ultraviolet composition on a planet with clouds, are likely to be (1700-3000 Å). Because of the delicate balance of

20 UV flux necessary, it is essential to model individual Edition of the Catalogue of Trigonometric Parallaxes stars using empirical UV spectra as input. (van Altena et al. 1995), which includes ~16,000 Composites of measured UV spectra of parallaxes measured from the ground; and the representative dwarfs have adequate signal/noise for Hipparcos catalogue (ESA 1997), which includes F, G and active K stars, but are insufficient for M ~118,000 parallaxes measured by the Hipparcos dwarfs. To remedy this lack of data, a Hubble Space satellite. The latter catalogue is effectively complete Telescope Snapshot survey is currently underway led for A, F and G dwarfs to ~40 parsecs (~2,000 by PI: S. L. Hawley (UW) in collaboration with Co- systems), but includes only a few hundred M dwarfs, Is: L. Walkowicz (UW), A. Segura (VPL), C. Johns- since the Hipparcos survey is complete to V~8.5, Krull (Rice), I. N. Reid (STScI), and M. Cohen with the faintest stars at only V~12.5. (UCB). This survey will use the Advanced Camera At present, there are currently only two robust for Surveys prism to capture near-ultraviolet spectral parallax programs operating. In the northern snapshots of a sample of 107 M dwarfs in the hemisphere, astronomers at the United States Naval immediate solar neighborhood. The sample has been Observatory in Flagstaff, AZ have both a CCD-based extensively studied in the Palomar/Michigan State effort led by Conard Dahn (Dahn et al. 2002) and an University (PMSU) survey of nearby stars (Reid et infrared array-based effort led by Fred Vrba (Vrba et al. 1995; Hawley et al. 1996) and is composed of al. 2004). In the southern hemisphere, the RECONS stars bright enough to be observed with short group at Georgia State University has a CCD-based snapshot exposures. We have homogeneous effort carried out at the Cerro Tololo Inter-American spectroscopic data of the Hα region for all of these Observatory in Chile, led by Todd Henry (Jao et al. stars. Transition region spectra obtained with STIS 2005). Both programs, however, involve star-by-star (Space Telescope Imaging Spectrograph) in the far targeted observations, rather than wide-field surveys, UV are also available for several of the targets, from and, combined, will produce only ~500 parallaxes our own observing programs and others in the over the next few years. Future projects will measure archive. In addition, two of the targets in our sample, parallaxes for large numbers of stars (see Section Gl 876 and Gl 581, have recently been found to IV.c), but at the present juncture trigonometric harbor Neptune-mass planets. parallax programs are not the solution to finding the These observations will provide data in the missing nearby stars. important near-UV wavelength region for M dwarfs Instead, the task of completing and expanding with a wide range of spectral types and activity the sample of nearby cool stars rests on photometric strengths. As of fall 2005, observations have been and spectroscopic techniques. Thanks to the taken for roughly a tenth of the sample, with new correlation between surface temperature and observations being scheduled regularly. We luminosity (or mass and energy generation) for core- anticipate that these new data will be ready for hydrogen burning stars, absolute magnitudes for inclusion in the Virtual Planetary Laboratory main-sequence dwarfs can be estimated from either photochemical models by early 2006, and will vastly color (Reid et al. 2002; Henry et al. 2004) or spectral improving our understanding of conditions on M type (Cruz et al. 2003). The resulting photometric or dwarf planets. spectroscopic parallaxes are both statistical (i.e., they refer to the average luminosity of single stars at the IV.b.ii An Improved Census of Nearby M dwarfs given color/spectral type) and less accurate than trigonometric parallaxes. At best, the uncertainty in

There is no question that current nearby-star distance, σd, is ±15-20%, but in some regions of the catalogues become incomplete at relatively small HR diagram, the uncertainties more than double, to distances from the Sun; although there is some debate ±60% (e.g., spectral types M3/M4, Reid and Cruz, over the extent of incompleteness (Henry et al. 1997; 2002; Williams et al. 2002). Reid et al. 2003), between 30% and 50% of M Nonetheless, with the completion of the DENIS dwarfs within 20 parsecs remain uncatalogued. The and 2MASS near-infrared sky surveys, photometric main complication in finding the missing stars is selection is a viable means of finding nearby M dwarf contrast: sifting out the small numbers of nearby candidates, particularly when coupled with proper dwarfs against the much more numerous distant, motion data. Several major efforts along these lines background stars at faint magnitudes. are currently underway, feeding targets to the Trigonometric parallax is the most reliable trigonometric parallax programs: method of distance determination, achieving • The RECONS survey (Henry et al. 1997) is accuracies better than 1% for the nearest few hundred focused primarily on identifying stars within systems. There are two fundamental compendia of 10 parsecs of the Sun. A crucial part of that trigonometric parallax measurements: the Fourth program is using photometric parallaxes to

21 select nearby candidates from the and M dwarfs is by measuring the incidence of debris SuperCOSMOS southern proper motion disks for stars covering a wide range of masses. survey (Hambly et al. 2004). As of After dissipation of their primordial planet-forming September, 2006 there were 348 objects in disks of gas and dust, many stars possess debris disks 249 systems with d<10 pc and trigonometric (e.g. Backman and Paresce 1993). Such disks are

parallax accuracy, σd<10%. This is a ~20% composed solely of regenerated dust, produced by increase since 2000, although many of the collisions from larger parent bodies which are new stars had prior photometric or otherwise undetectable. These systems represent the spectroscopic distance estimates; nearly all extrasolar analogs of the asteroid belt and Kuiper of the additions are M dwarfs. Belt in our own solar system. Therefore, debris disks • The New Neighbours program (Crifo et al. provide signposts that large solid bodies have 2005) uses DENIS IJK photometry to formed, and hence can shed light on the likelihood of identify nearby candidates, employing planets around M dwarfs compared to more massive follow-up optical spectroscopy to refine the stars. classification and distance estimates. The Ground- and space-based photometric studies program is mainly targeted toward late-type have identified over a hundred debris disks around A, M dwarfs within 30 parsecs. F, and G-type dwarfs from the infrared thermal • The 2MASS NStars survey targets M (Reid emission of the circumstellar dust grains (e.g., and Cruz, 2002) and L (Cruz et al. 2005) Greaves and Wyatt 2003; Bryden et al. 2006). dwarfs within 20 parsecs of the Sun. L However, only a handful of confirmed examples of dwarf candidates are identified directly from M dwarf debris disks are known (Song et al. 2002; the 2MASS JHK photometry, with more Liu et al. 2004; Low et al. 2005). One of them, the refined distances derived from optical and nearby (10 pc) M dwarf AU Mic, has proven to be a near-infrared spectroscopy; the current rich opportunity for studying disk and planet survey places 80 systems within 20 parsecs, formation up close, given its proximity to Earth although only ~30% have trigonometric (Kalas, Liu and Matthews 2004; Liu 2004; Krist et al. parallaxes. Most of the M dwarf candidates 2005; Metchev et al. 2005; Roberge et al. 2005). are selected from Luyten’s Two-tenths Myriad past searches for debris disks have neglected (NLTT) proper motion survey, with refined and/or overlooked M dwarfs, largely due to distances based on photometric and (optical) sensitivity limitations. Since the dust emission spectroscopic parallaxes. With follow-up originates from heating by incident starlight, the observations for approximately half the sky, much lower (10-1000X) luminosity of M dwarfs over 300 systems have been added to the 20- compared to G dwarfs means that debris disks around pc census (Reid et al. 2004). M dwarfs are much harder to detect (e.g. Wray, Liu • The LSPM catalog (Lépine and Shara, 2005) and Reid 2005). is a proper motion survey of (at present) the New ground-based (mid-infrared and sub- northern sky. Follow-up spectroscopy has so millimeter) and space-based (infrared) astronomical far concentrated on candidate cool instruments at last have the sensitivity to detect and subdwarfs (local members of the metal-poor characterize debris disks around large numbers of M Galactic halo), but, combining 2MASS data dwarfs, and thereby probe the frequency of planet with optical photographic photometry, this formation as a function of stellar mass. Spitzer Space survey will prove a happy hunting ground Telescope has an unprecedented level of sensitivity at for M dwarfs within 30-40 parsecs of the far-IR wavelengths (24-160 microns) and could Sun. readily detect debris disks around M dwarfs. Spitzer observations at 70 microns of 37 M stars closer than The combined results from these surveys are likely to 5 pc revealed no excesses (3 sigma limit) from lead to great improvements in the local M dwarf planetary debris at temperatures around 50-100 K (N. census over the next few years, with the potential of Gautier, priv. comm.). However, only debris disks as identifying almost all systems within 20-25 parsecs. exceptionally dense as the ones around beta Pictoris and AU Mic could be detected at 5 pc around a mid- IV.b.iii. The Frequency of Planets Around M M star. Such disks are rare (<1%; Jura et al. 1995), Dwarfs Compared To G Dwarfs From Debris and thus, the null result from the Spitzer work to date Disks Surveys is consistent with detection statistics for debris disks around more luminous stars. A much larger sample One promising means to study the relative stars needs to be surveyed to determine how common frequency of planet formation around Sun-like stars debris disks are around low-mass stars. Future

22 surveys should focus on the youngest M dwarfs in the have increased its luminosity by more than a percent solar neighborhood: AFG-type dwarfs show a higher or so, and M star planets have an exceedingly stable incidence of disks and larger dust masses at younger radiation environment over long timescales. An ages (e.g. Zuckerman and Becklin 1993; Habing et exception that is interesting for astrobiology is that al. 2001; Decin et al. 2003; Liu et al. 2004; Rieke et the time to reach the main sequence, during which al 2005; Najita and Williams 2005; Moor et al. 2006) the stellar luminosity decreases significantly, is about and thus young stars are the most appealing targets 0.3 Gyr for 0.15 Msun and approaches 1 Gyr for 0.1 for detecting disks. Msun (Laughlin and Bodenheimer 1997, Burrows et al. 2001). Thus retention of an early outgassed IV.b.iv. Response of Planetary Atmospheric atmosphere or an early origin of life may be Structure, Chemistry, and Escape Processes Due challenging for (eventually) habitable planets orbiting To Stellar Winds, Flares and Other Variability the least massive of main sequence stars. A change in S caused by changes in a planet’s As explained on section II.e the chemistry of a orbit can also provide a constraint to habitability. planet around an M star is different because of the Such changes will be minimal if a planet within the ultraviolet distribution the stellar spectrum. Those habitable zone becomes synchronously rotating, but models obtained steady-state solutions for may be important during the time required for tidal atmospheres that are submitted to the same stellar locking (Heath and Doyle 2004), circularization, or if flux on time. As we know, active M stars exhibit the initial eccentricity and semi-major axis are so strong and variable emissions on the ultraviolet and large that synchronous rotation is never achieved, or shorter wavelengths. It is not clear how the if resonant pumping increased the eccentricity, or atmospheric chemistry and the climate of the planet after the orbit has been dynamically perturbed by a would respond to these variable fluxes. This issue third body. Variations of this sort can be studied will be studied by the Virtual Planetary Laboratory quantitatively using so-called radiative convective project (http://vpl.ipac.caltech.edu/) that is currently models, which model the mean atmospheric radiative developing a coupled photochemical/radiative- properties in a 1D vertical column, and make simple convective model to study the effect of the M stars assumptions for the effect of vertical heat transport variability on Earth-like planets on their habitable by dynamics (e.g.: Kasting et al. 1993). zone. The short-term variability of S can also be due to changes in stellar luminosity or changes in planetary IV.b.v Evolutionary Changes in Stellar Flux orbits. M-dwarfs display starspots whose magnitude compared to S is large when compared to spots on G- Since the incoming stellar radiation S drives the stars. While the effect of a starspot would be climate of a planetary atmosphere, significant significant on climate, it is unlikely that an Earthlike variability in this quantity will have consequences for atmosphere on a planet in the HZ of an M-dwarf the potential habitability of a planet orbiting an M- would collapse as a result (Joshi et al. 1997). Further dwarf. The variability of S can be divided into two research is needed however on quantifying the effect components: long-term (periods >> orbital period) of starspots on specific aspects of climate such as and short-term (periods ≤ orbital period). forest habitability. The long-term variability of S can be due to Tidally locked planets should have circular either long-term changes in stellar luminosity, mass orbits, since the tidal removal of spin angular loss (see section IIIe), and changes in planetary orbits momentum increases the orbital angular momentum, (see section IIId). Although it has long been which reduces the eccentricity. It is nevertheless recognized that very low-mass stars can potentially possible that non-tidally locked planets around M- live for many tens of Gyr, most estimates of M star stars might have eccentric orbits and hence main sequence lifetimes have been based on experience cyclical changes in S. However, even if extrapolations of evolutionary calculations for this were the case, the effect of orbital eccentricity higher-mass stars. Laughlin and Bodenheimer (1993) should be far less severe on a planet orbiting an M- and Laughlin et al. (1997) produced evolutionary dwarf than one orbiting a G-star. This is because the calculations for stars in the mass range 0.08 to 0.50 orbital period of a planet in the HZ of an M-dwarf is Msun. The main sequence lifetimes at 0.50, 0.20, and much smaller than the orbital period of a planet in the 0.08 Msun turn out to be about 100, 1000, and 10,000 HZ of a G-star (Heath et al. 1999), and so the effects Gyr, and stars less massive than 0.2 Msun surprisingly of changes of S are lowered in the former case than never evolve through the red giant phase (Laughlin et the latter case by the thermal damping effect of a al. 1997). These calculations indicate that in the time planet’s atmosphere. since the beginning of the big bang, no M star should

23 More research is needed to quantify the effects understanding what countermeasures such organisms of large orbital eccentricity on habitability. In order might have to evolve. to do this, one has to employ a combination of the 1D model described above, more complex energy- IV.c Opportunities For Research That Would Be balance models that parameterize the effect of Particularly Enlightening horizontal transport on climate (e.g.: Williams and Pollard 2002), and 3D climate models (e.g.: Joshi IV.c.i. In M Star Disks, Are There 2003), which, while computationally expensive, do Proportionately More Volatiles At a Given explicitly represent the important large-scale physical Radius? processes. Given that temperatures at a fixed orbital IV.b.vi Radiation Damage To Life distance will be somewhat lower around an M dwarf than around a G dwarf (Boss 1995), one would Radiation damage to a potential life form on a expect that volatile compounds (e.g., water) will be planet orbiting an M dwarf star would have to be found in greater abundances closer to an M than a G estimated from what is known about radiation dwarf. Volatile abundances in the Solar System's damage to terrestrial life. This is complicated by four objects are generally thought to be primarily a result unknowns: what life would be like on a planet of the global temperature gradient in the nebula orbiting an M dwarf star, what the radiation regime (Cassen 1996, 2001; Boss 1998; Woolum and Cassen would be like (discussed elsewhere), what radiation 1999), though mixing and transport processes in the regime would the organisms be exposed to, and what gaseous disk (e.g., Boss 2004) and that produced by countermeasures would such organisms take. In the the chaotic orbital evolution of planetesimals (e.g., next decade, all of these estimates will improve. Morbidelli et al. 2000) can have a significant effect While we will not know what life around even on the radial distribution of volatiles and their one M-dwarf is like until we find it, it is most likely delivery to planetary surfaces. to be based on organic carbon (Rothschild and Mancinelli, 2001). Thus, estimates of the stability of IV.c.ii Prediction of Duration of Geothermal organic compounds, particularly long-chain Activity Within Planets in the Habitable Zone of polymers, in the presence of radiation should give us M Dwarf Stars an idea of the radiation tolerance of any potential life forms. Similarly, experiments with terrestrial We are not aware of any quantitative or organisms on radiation damage, a field which has qualitative studies that characterize the duration of recently been re-invigorated by NASA's geological activity on terrestrial planets as a function announcement in September 2006 of the selection of of radii and composition. However, because of the 12 new proposals to study space radiation damage. similar metallicities of G and M dwarfs we broadly The radiation regime that an organism is actually expect the sizes and compositions and hence the exposed to can vary enormously as a result of range of potential thermal histories to be similar on physical changes in the planetary environment such rocky planets orbiting either type of star. Mantle as the rotation of a planet and cloud cover, and the convection provides a mechanism for cycling and ecology, behavior and physiology of the organism. release of materials between the surface and For example, organisms can live in or under atmosphere. Thus, a first-order measure of the protective layers, migrate in and out of exposed duration of endogenic geologic activity is the time areas, and produce protective pigmentation (e.g., period over which convective motions are generated Rothschild and Giver, 2003; Rothschild and in the planet’s interior. Planetary thermal evolution Mancinelli, 2001). could be studied via parameterized convection Countermeasures to deal with radiation exposure models (e.g., Schubert et al. 1979; Schubert et al. range from avoidance, such as outlined above, to 2001; Hauck and Phillips 2002). In such a study one repair of damage. There is enormous interest in the could vary the radius from 0.1 – 10 times that of the study of radiation damage repair because of its Earth and vary the radioactive element content from medical importance on earth as well as its critical role 0.1 – 10 times that of the chondritic values that are in determining the ability of humans to survive in typical of our solar system (Stacey, 1992). The space. Thus, while direct research is not currently results of such calculations would give ballpark planned for mimicking the radiation exposure on a estimates of the geological lifetimes of extrasolar planet orbiting an M-dwarf, terrestrial-based research rocky bodies, as well as the longevity of conditions will certainly provide an increasing base for conducive to planetary dynamo action. Further studies might extend these calculations to include the

24 effects of mantle tidal dissipation (e.g., Showman and d = the distance to the star in parsecs. Malhotra, 1999; Moore, 2003). If we assume a single wind velocity (see Doyle et al. IV.c.iii Mass Loss Measurements From M Dwarf Stars 1995 for a five-segment model based on the solar wind), with the simplifying assumptions for the wind 6 One can ask, for example, if it is reasonable be that: Z = γ = µ = 1, T ≈ 10 , υ∞ ≅ 200 km/sec and able to detect relevant mass loss rates in the radio d ≈ 15 parsecs, then the above equation simplifies to: region (1 to 10 GHz) for nearby M-dwarf star systems (i.e., those within the range of Terrestrial • 4 4 / 3 2 / 3 2 / 3 Planet Finder as currently designed ~ 15 parsecs). For Sν = 8.8 ×10 M ν (21.8 −1.26 logν ) (3) a typical mid-M-dwarf stellar mass of about 0.3 solar masses, losing 10% of its mass over 5 billion years -12 where the units are as stated above. The maximum would constitute a mass-loss rate of 6x10 solar opaque frequency (wind detectable above the stellar masses per year. Such a mass loss would cause the photosphere) would then be about 15 GHz. For the circumstellar habitable zone (CHZ) to migrate inward mass-loss rate of about 6x10-12 solar masses per year, by about 7% of its distance using the mass- the flux expected at a distance of 15 parsecs at 10 luminosity relationship for solar-type stars of GHz would then be about 0.003 milliJansky. This 2.5 L ∝ M , while the planet (if angular momentum may be within the capability of, for example, the as conserved) would be expected to migrate outward Allen Array Telescope, but days-long integrations by about 4%. (The impact of stellar mass loss on may be necessary to ensure detection. Measurements planetary habitability is discussed in section III.b.ii.) of mass loss from stellar wind and Coronal Mass Thus such a mass loss rate would have Ejections are important for a better understanding of significant repercussion on the habitability of any the evolution of the radiation (X-ray, FUV, etc.) and planets initially formed within the narrow CHZ of M- particle/plasma environment of planets in the dwarf systems (although see Joshi et al. 1997 habitable zone as a function of spectral type and indicating a proportionately wider CHZ for M-stars stellar age. compared to other dwarf systems due to their peak flux being in the infrared where the moist runaway IV.c.iv. Construction of a Large Catalog of M greenhouse mechanism is not directly applicable for Dwarf Stars the inner boundary and the outer boundary albedo in the infrared is much smaller.) The combination of M dwarfs' overwhelming The radio flux, Sν (in milliJanskys) expected for numbers and relative obscurity mirrors a situation an optically thick free-free electron wind may be familiar to biologists --- it is the numerous, but tiny, formulated (Wright and Barlow 1975; Leitherer and microbes for which comprehensive catalogs are least Robert 1991; Doyle et al. 1995; see also Lamers and complete. As outlined above, M dwarfs remain Cassinelli 1999) as: poorly catalogued at distances beyond 10-20 parsecs of the Sun. SETI on the ATA really needs on the • order of one million stars, more than 700,000 of 10 4 / 3 2 / 3 −4 / 3 2 (2) Sν = 2.32x (M Z) (γ g ν ν ) (v∞ µ) d which will be M dwarfs if a volume-limited sample is desired. Here we consider how such a catalogue where might be constructed. As an estimate of scale, there are ~110 M dwarf • systems within 8 parsecs of the Sun and ~2,000 G M = the stellar mass-loss rate in solar masses per dwarfs within 40 parsecs. Extrapolating those number year, densities, with due allowance for the density gradient Z = the rms ionic charge of the wind particles, in the disk perpendicular to the Galactic Plane, a γ = the number of electrons per ion, Megastar survey limited to G dwarfs would need to gν = the free-free Gaunt factor = reach ~450 parsecs; a Megastar survey combining 10.6 + 1.90 logT –1.26 logν –1.26 logZ FGK dwarfs (300,000 stars) and M dwarfs (700,000 (at radio frequencies) where T is the stellar wind stars) would need to extend to ~250 and ~150 temperature, parsecs, respectively. Compiling reliable catalogues ν = the frequency in Hertz, at those distances is a non-trivial prospect. v∞ = the stellar wind velocity in kilometers per Photometric parallax surveys offer one option, second, namely estimating the distances for low-mass stars µ = the mean molecular weight of the wind particles, from their colors and magnitudes. The Sloan Digital and Sky Survey covers π steradians at moderate and high

25 Galactic latitude in the regions of the celestial sphere without the need for additional spectroscopic accessible from northern terrestrial latitudes; confirmation. Simulations of PS-1 performance (E. combining those data with the infrared 2MASS data Magnier, priv. comm.) find that PS-1 parallaxes will could produce substantial numbers of low-mass stars be obtained for all M dwarfs out to around 100 pc. (e.g. Hawley et al. 2002; Bochanski et al. 2005). In In the more distant future, even larger volume- the near future, both the Pan-STARRS project in the limited samples will come. The full resolution of this northern hemisphere (Kaiser et al. 2005) and the issue may have to await results from the Gaia satellite Large Synoptic Survey Telescope in the southern (Perryman, 2005). Slated for launch in 2011, Gaia hemisphere (LSST; Tyson, 2002) will provide aims to observe 109 stars brighter than 20th multiple photometric scans at optical wavelengths of magnitude, obtaining trigonometric parallaxes the northern sky, greatly surpassing the SDSS data in accurate to <11 µarcsec at V<15, degrading to 160 both depth and areal coverage. µarcsec at V=20. These observations will provide a There are, however, limitations to reliable catalogue of G dwarfs to well beyond 500 photometrically selected samples. First, the parsecs, and early- and mid-type M dwarfs to 100- uncertainties in photometric parallax vary with 200 parsecs. Unfortunately, the catalogue will not be spectral type, leading to non-uniform selection available until ~2018, at the earliest. effects, notably Malmquist bias (the tendency to include stars that are brighter than average at a given V. Conclusion spectral type). Second, unresolved binaries will also populate the sample in large numbers: ~30% of M We have tried to summarize what is known dwarfs are binary, with approximately half being about the potential for habitable planets to orbit nearly equal-mass systems; since such binaries are around M dwarf stars. Given the large percentage of twice as bright, they are sampled over a larger all stars of this type, two scientific efforts that would volume, and may contribute almost half of the stars be strongly impacted by the conclusion; 1) attempts in the photometric sample. These binaries are not to image terrestrial planets in orbit around nearby necessarily the best candidates for habitable planetary stars, and conduct a spectroscopic assay of their systems. These two biases effectively degrade the atmospheres for evidence of biosignatures, and 2) efficiency of SETI’s survey by contaminating the commensal radio searches for technosignatures using target sample. Follow-up spectroscopic observations the Allen Telescope Array now under construction. can address these issues to some extent, but obtaining Much is unknown, but we conclude that M dwarf such data for ~106 stars would be a very challenging stars are back on the table – for now. task. The aforementioned Pan-STARRS project will VI. Acknowledgment be an unprecedented resource for constructing the first volume-limited catalog of low-mass stars. Pan- This paper is based upon a workshop organized STARRS comprises a unique optical survey by the SETI Institute's NASA Astrobiology Institute instrument of four co-aligned 1.8-meter telescopes, Lead Team and supported by the National each equipped with a wide-field (7 sq. degree) CCD Aeronautics and Space Administration through the camera, with an estimated completion date in 2010. NASA Astrobiology Institute under Cooperative As a prototype and testbed for the complete 4- Agreement “Planetary Biology, Evolution, and telescope system, the project is now finishing the first Intelligence” Number NNA04CC05A. We thank 1.8-meter telescope ("PS-1"), and science operations SETI Institute for hosting the workshop and the are scheduled to begin in 2007. The large survey collaborative web site that supported the workshop power (etendue) of PS-1 will enable a deep, multi- and subsequent research and editorial activities. band, multi-epoch optical survey of the entire sky observable from Hawai`i (3π steradians). While the VII. References depth and area of the PS-1 survey will greatly exceed all previous photographic and digital optical surveys, Adamson, A.J., Whittet, D.C.B., and Duley, W.W. it is the high astrometric precision of the multi-epoch (1990) The 3.4-micron interstellar absorption dataset (1-d RMS error of 10 milliarcseconds per feature in CYG OB2 no. 12. Monthly Notices observing epoch) that will enable a transformational Royal Astronomical Society 243(1), 400-404. parallax census: in contrast to all previous parallax Allamandola, L.J., Bernstein, M.P., Sandford, S.A., programs which obtained measurements on a star-by- and Walker, R.L. (1999) Evolution of interstellar star basis, PS-1 will determine parallaxes over most ices. Space Science Reviews 90, 219-232. of the sky. Therefore, low-mass stars can be directly selected as objects with faint absolute magnitudes,

26 Allard, F. and Hauschildt, P.H. (1995) Model G., and Gizis, John E (2005) Spectroscopic atmospheres for M (sub)dwarf stars. 1. The base survey of M Dwarfs within 100 Parsecs of the grid model. Astrophys. J. 445, 433-450. Sun. AJ, 130, 1871. rd Allen, C.W. (1973) Astrophysical Quantities, 3 Bogard, D.D., Clayton, R.N., Marti, K., Owen, T., Edition, The Athlone Press, London, UK. and Turner, G. (2001) Martian volatiles: isotopic Ambartsumian, V.A., Mirzoyan, L., Parsamian, E.S., composition, origin, and evolution. Space Sci. Chavushian, O.S., and Erastova, L.K. (1970) Rev. 96, 425-458. Flare stars in the Pleiades. Astrofizika 6, 7-30. Bolin, B.T., Degens, E.T., Duvigneaud, P. and Andreeshchev, A., Scalo, J., and Smith, D. (2002) Kempe, S. (1979) The global biogeochemical Very low mass stars as optimum sites of carbon cycle, (Table 1.2, P. Duvigneaud). In The habitable planets and their detection. Poster Global Carbon Cycle, SCOPE 13, edited by B. presented at Abstrobiology Science Conference, Bolin, E.T. Degens, S. Kempe, and P. Ketner, NASA Ames Research Center, Moffett Field, John Wiley and Sons, Chichester, UK, pp. 1-56. CA. Borucki, W.J., Koch, D., Basri, G., Brown, T., Andrews, S. and Williams, J. (2005) Circumstellar Caldwell, D., DeVore, E., Dunham, E., Gautier, dust disks in Taurus-Auriga: the submillimeter T., Geary, J., Gilliland, R., Gould, A., Howell, perspective. ApJ, 631, 1134. S., and Jenkins, J. (2003) Kepler mission: a Angel, J.R.P. and Woolf, N.J. (1996) Searching for mission to find Earth-size planets in the life on other planets. Sci. Am. 274, 60-66. habitable zone. In Proceedings of the Conference Aurnou, J.M., (2004) Secrets of the deep. Nature on Towards Other Earths: Darwin/TPF and the 428, 134-135. Search for Extrasolar Terrestrial Planets, edited Aurnou, J.M., Heimpel, M.H., and Wicht, J. (2006) A by M. Fridlund and T. Henning, ESA SP-539, deep convection model for zonal flow on the Ice pp. 69-81. Giants, Icarus, in review. Boss, A.P. (1995) Proximity of Jupiter-like planets to Backman, D.E. and Paresce, F. (1993) In Protostars low mass stars. Science 267, 360-362. and Planets III, edited by E.H. Levy and J.I. Boss, A.P. (1997) Giant planet formation by Lunine, University of Arizona Press, Tucson, p. gravitational instability. Science 276, 1836-1839. 1253. Boss, A.P. (1998) Temperatures in protoplanetary Bada, J.L., Bigham, C., and Miller, S.L. (1994) disks. Annu. Rev. Earth and Plan. Sci. 26, 53-80. Impact melting of frozen oceans on the early Boss, A.P. (2001) Gas giant protoplanet formation: Earth: Implications for the origin of life. Proc. disk instability models with thermodynamics and Natl. Acad. Sci. USA. radiative transfer. Astrophys. J. 563, 367-373. Ball, P. (2004) Water, water, everywhere. Nature Boss, A.P. (2004) Evolution of the solar nebula. VI. 427, 19-20. Mixing and transport of isotopic heterogeneity. Basilevsky, A.T., Head, J.W., Schaber, G.G., and Astrophys. J. 616, 1265-1277. Strom, R.G. (1997) The resurfacing history of Boss, A.P. (2006a) Rapid formation of gas giant Venus. In Venus II-Geology, Geophysics, planets around M Dwarf stars. ApJ, 643, 501. Atmosphere, and Solar Wind Environment, Boss, A.P. (2006b) Rapid formation of super-Earths edited by S.W. Bougher, et al., University of around M Dwarf stars. ApJ (Letters), 644, L79- Arizona Press, Tucson, pp. 1047-1086. L82. Basri, G., Stassun, K.G., Ardila, D.R., Barsony, M., Bryden G., Beichman, C. A., Trilling, D. E., Rieke, and Mathieu, R.D. (2004) The origin of X-rays G. H., Holmes, E. K., Lawler, S. M., Stapelfeldt, in pre-main-sequence stars. Poster 105.09, AAS K. R., Werner, M. W., Gautier, T. N., Blaylock, 205th Meeting, San Diego, CA. M., Gordon, K. D., Stansberry, J. A., and Su, Beaulieu, J-P. and 72 other authors (2006) Discovery K.Y.L. (2006) Frequency of debris disks around of a cool planet of 5.5 Earth masses through solar-type stars: first results from a Spitzer MIPS gravitational microlensing. Nature 439, 437-440. survey. ApJ, 636, 1098. Benner, S.A., Ricardo, A., and Carrigan, M.A. (2004) Burrows, A., Hubbard, W.B., Lunine, J.I., and Is there a common chemical model for life in the Liebert, J. (2001) The theory of brown dwarfs universe? Current Opinions in Chemical Biology and extrasolar planets. Rev. Mod. Phys. 73, 719- 8, 672-689. 765. Black, C.C. (1973) Photosynthetic carbon fixation in Cassen, P. (1996) Models for the fractionation of relation to net CO2 uptake. Ann. Rev. Plant. moderately volatile elements in the solar nebula. Physiol. 24, 253-286. Meteoritics Planet. Sci. 31, 793-806. Bochanski, J., Hawley, Suzanne L., Reid, I. Neill, Covey, Kevin R., West, Andrew A., Tinney, C.

27 Cassen, P. (2001) Nebula thermal evolution and the Improved mass-luminosity relations. Astron. properties of primitive planetary materials. and Astrophys. 364, 217-224. Meteoritics Planet. Sci. 36, 671-700. Deming, D. and Seagar, S., editors (2003) Scientific Charbonneau, D., Brown, T.M., Latham, D.W., and Frontiers in Research on Extrasolar Planets, Mayor, M. (2000) Detection of planetary transits ASP Conference Series 294, San Francisco. across a Sun-like star. Astrophys. J. 529, L45- Dey, A., Van Breugel, W. Vacca, W., and Antonucci, L48. R. (1997) Triggered star formation in a massive Chassefiere, E. (1996) Hydrodynamic escape of galaxy at Z = 3.8: 4C 41.17. Astrphys. J. 490, oxygen from primitive atmospheres: applications 698-. to the cases of Venus and Mars. Icarus, 124, Dole, S.H. (1964) Habitable planets for man. 537-552 Blaisdell, New York. Chassefiere, E. and Leblanc, F. (2004) Mars Doyle, L.R., Vikramsingh, R., Whitmire, D.P., and atmospheric escape and evolution; interaction Heather, N.C. (1995) Circumstellar habitable with the solar wind. Planet. and Space Sci. 52, zones and mass loss from young solar-type stars 1039-1058. II. Observational considerations. In Progress in Christensen, U.R. and Aubert, J. (2006) Scaling the Search for Extraterrestrial Intelligence, properties of convection-driven dynamos in edited by G.S. Sostak, ASP Conf. Ser. 74, pp. rotating spherical shells and application to 195-204. planetary magnetic fields. Geophys. J. Int. 140, Doyle, L.R., editor (1996) Circumstellar Habitable 97-114. Zones: Proceedings of the First International Chyba, C.F. (1989) Impact delivery and erosion of Conference, Travis House Publications, Menlo planetary oceans. Nature 343, 317-332. Park, CA. Corliss, J.B., Dymond, J., Gordon, L.I., Edmond, Doyle, L.R., Heather, N.C., Vikramsingh, R., and J.M., von Herzen, R.P., Ballard, R.D., Green, K., Whitmire, D.P. (1996) Stellar mass loss and the Williams, D., Bainbridge, A. Crane, K., and van young solar system habitable zone. In Andel, T.H. (1979) Submarine thermal springs Circumstellar Habitable Zones, edited by L.R. on the galapagos rift. Science 203, 1073-1083. Doyle, Travis House Publications, Menlo Park, Costa, E., Mendez, R.A., Jao, W.C., Henry, T.J., CA, pp. 157-179. Subasavage, J.P., Brown, M.A., Ianna, P.A., and Doyle, L.R., (2006) Overview of extrasolar planet Bartlett, J. (2005) The solar neighborhood XIV. detection methods and M-stars as near-term Parallaxes from the Cerro Tololo inter-American targets for exobiological searches. In observatory parallax investigation - first results Proceedings of the 2004 Canary Islands Winter from the 1.5m telescope program. Astron. J. 130, School XVI on Extrasolar Planets, edited by H.- 337-349. J. Deeg and J.-A. Belmonte, Cambridge Crifo, F., Phan-Bao, N., Delfosse, X., Forveille, T., University Press. Guibert, J., Martín, E.L., and Reylé, C. (2005) Draine, B.T. and Lee, H.M. (1984) Optical New neighbours. VI. Spectroscopy of DENIS properties of interstellar graphite and silicate nearby stars candidates. Astron. Astrophys. 441, grains. Astrophys. J. 285, 89-108. 653. Draine, B.T. (2003) Interstellar dust grains. Ann. Rev. Cruz, K.L. and Reid, I.N. (2002) Meeting the cool Astron. Astrophys. 41, 241-289. neighbors. III. Spectroscopy of northern NLTT Emerson, R., Chalmers, R., and Cederstrand, C. stars. Astron. J. 123, 2828. (1957) Some factors influencing the long-wave Cruz, K.L., Reid, I.N., Liebert, J., Kirkpatrick, J.D., limit of photosynthesis. Proc. Natl. Acad. Sci. and Lowrance, P.J. (2003) Meeting the cool 43, 133-143. neighbors. V. A 2MASS-selected sample of ESA (1997) The Hipparcos and Tycho Catalogues, ultracool dwarfs. Astrophys. J. 126, 2412. ESA SP-1200, Noordwijk. Dahn, C.C., et al. (2002) Astrometry and photometry Feigelson, E.D., Hornschemeier, A.E., Micela, G., for cool dwarfs and brown dwarfs. Astron. J. Bauer, F.E., Alexander, D.M., Brandt, W.N., 124, 1170. Favata, F., Sciortino, S., and Garmire, G.P. Decin, G., Dominik, C., Malfait, K., Mayor, M., and (2004) The Chandra deep field-north survey. Waelkens, C. (2000) The Vega phenomenon XVII. Evolution of magnetic activity in old late- around G dwarfs. Astron. Astrophys. 357, 533. type stars. Astrophys. J. 611, 1107-1120. Delfosse, X., Foreveille, T., Segransan, D., Beuzit, J- Ford, E.B., Seager, S., and Turner, E.L. (2001) L., Udry, S., Perrier, C., and Mayor, M. (2000) Characterization of extrasolar terrestrial planets Accurate masses of very low mass stars IV. from diurnal photometric variability. Nature 412, 885-887.

28 Fox, J.L. and Hac, A. (1997) Spectrum of hot O at Trams, N., and Bouchet, P. (2001) Incidence and the terrestrial planets. J. Geophys. Res., 102, survival of remnant disks around main-sequence 24005-24011. stars. Astron. Astrophys. 365, 545. Freund, M.M. and Freund, F.T. (2006) Solid solution Haberle, R.M., McKay, C.P., Tyler, D., and model for interstellar dust grains and their Reynolds, R.T. (1996) Can synchronously organics. Astrophys. J. 639, 210-226. rotating planets support an atmosphere? In Gautier, N. (2005) private communication. Circumstellar Habitable Zones: Proceedings of Gizis, J.E., Monet, D.G., Reid, I.N., Kirkpatrick, J.D., the First International Conference, edited by Liebert, J., and Williams, R.J. (2000) New L.R. Doyle, Travis House Publications, Menlo neighbors from 2MASS: Activity and kinematics Park, Calif., pp. 29-33. at the bottom of the main sequence. Astron. J. Haisch, K., Lada, E. and Lada, C. (2001) 120, 1085-1099. Circumstellar disks in the IC 348 cluster. AJ Gizis, J.E., Reid, I.N., and Hawley, S.L. (2002) The 2001, 121. Palomar/MSU nearby star spectroscopic survey. Hambly, N.C., Henry, T.J., Subasavage, J.P., Brown, III. Chromospheric activity, M dwarf ages, and M.A., and Jao, W.C. (2004) The solar the local star formation history. Astron. J. 123, neighborhood VIII. Discovery of new high 3356-3369. proper motion nearby stars using the Greaves, J.S. and Wyatt, M.C. (2003) MNRAS 345, SuperCOSMOS sky survey. The Astron. J. 128, 1212. 437. Greenberg, J.M. (1968) Interstellar grains. In Hauck, S.A. II, Phillips, R.J., and Price, M. (1998) Nebulae and Interstellar Matter, Vol. 7, edited Venus: Crater distribution and plains resurfacing by B.M. Middlehurst and L.H. Aller, Chicago models. J. Geophys. Res. 103, 13635-13642. University Press, Chicago, p. 221. Hauck, S.A. II and Phillips, R.J. (2002) Thermal and Greenberg, J.M. and Li, A. (1996) What are the true crustal evolution of Mars. J. Geophys. Res. 107, astronomical silicates? Astron. Astrophys. 309, 5052, doi:10.1029/2001JE001801. 258-266. Hawley, S., et al. (2002) Characterization of M, L, Greenberg, J.M., Li, A., Mendoza-Gomez, C.X., and T dwarfs in the Sloan Digital Sky Survey. AJ Schutte, W.A., Gerakines, P.A., et al. (1995) 123, 3409. Approaching the interstellar grain organic Hawley, S.L. and Johns-Krull, C.M. (2003) refractory component. ApJ Letters 455, L177. Transition region emission from very low mass Grießmeier, J-M., Stadelmann, A., Motschmann, U., stars 3003. Astrophys. J. Suppl. Lett. 588, L109- Belisheva, N.K., Lammer, H., and Biernat, H.K. L112. (2005). Cosmic ray impact on extrasolar Earth- Hawley, S.L. and Pettersen, B.R. (1991) The great like planets in close-in habitable zones, flare of 1985 April 12 on AD Leonis. Astrophys. Astrobiology, 5, 587-603. J. 378, 725-741. Grießmeier, J.-M., Stadelmann, A., Penz, T., Hawley, S.L., Gizis, J.E., and Reid, I.N. (2000) The Lammer, H., Selsis, F., Ribas, I., Guinan, E.F., Palomar/MSU nearby star spectroscopic survey. Motschmann, U., Biernat, H.-K., and Weiss II. The southern M dwarfs and investigation of W.W. (2004) The effect of tidal locking on the magnetic activity. Astron. J. 112, 2799-2827. magnetospheric and atmospheric evolution of Head, J.W., Greeley, R., Golombek, M.P., Hartmann, ‘Hot Jupiters’. Astron. and Astrophys., 425,753- W.K., Hauber, E., Jaumann, R., Masson, P., 762. Neukum, G., Nyquist, L.E., and Carr, M.H. Güdel, M. and Benz, A.O. (1993) X-ray/microwave (2001) Geological processes and evolution. relation of different types of active stars. Space Sci. Rev. 96, 263-292. Astrophys. J. Lett. 405, L63-L66. Heath, M.J. (1996) The forest-habitability of Earth- Güdel, M., Audard, M., Reale, F., Skinner, S. L., and like planets. In Circumstellar Habitable Zones: Linsky, J. L. (2004) Flares from small to large: Proceedings of the First International X-ray spectroscopy from Proxima Centauri with Conference, edited by L.R. Doyle, Travis House XMM-Newton. Astron. Astrophys. 416, 713-732. Publications, Menlo Park, Calif., pp. 445-457. Gurzadyan, G.A. (1970) Flare Stars, Moscow Nauka Heath, M.J., Doyle, L.R., Joshi, M.M., and Haberle, Haro, G., and Morgan, W.W. 1953, Rapid R.M. (1999) Habitability of planets around red variables in the Orion Nebula, Astrophys. J. 118, dwarf stars. Origin of Life and Evolution of the 16-19. Biosphere 29, 405-424. Habing, H.J., Dominik, C., Jourdain de Muizon, M., Heath, M. and L.R. Doyle (2004) From near- Laureijs, R.J., Kessler, M.F., Leech, K., synchronously rotating planets to tidal lock: a Metcalfe, L., Salama, A., Siebenmorgen, R., new class of habitable planets examined for

29 forest habitability. In Bioastronomy 2002: Life Jayawardhana, R., Ardila, D., Stelzer, B. and Haisch, Among the Stars, IAU Symposium 213, edited by K. (2003) A disk census for young Brown R.P Norris and F.H. Stootman, Astronomical Dwarfs. AJ 126, 1515. Society of the Pacific, San Francisco, Calif., pp. Johnson, R.E. (1994) Plasma-induced sputtering of 225-229. an atmosphere. Space Sci. Rev. 69, 215-253. Henry, G.W., Marcy, G.W., Butler, R.P., Vogt, S.S. Johns-Krull, C.M., Valenti, J.A. (1996) Detection of (2000) A transiting “51 Peg-like” planet. strong magnetic fields on M Dwarfs. ApJL 495, Astrophys. J. 529, L41-L44. L95-L98. Henry, T.J., Ianna, P.A., Kirkpatrick, J.D., and Joshi M.M., Haberle, R.M., and Reynolds, R.T. Jahreiss, H. (1997) The solar neighborhood IV: (1997) Simulations of the atmospheres of discovery of the twentieth nearest star. A.J. 114, synchronously rotating terrestrial planets orbiting 388-395. M dwarfs: conditions for atmospheric collapse Henry, T.J. (2004) The mass-luminosity relation from and implications for habitability. Icarus 129, end to end. In Spectroscopically and Spatially 450-465. Resolving the Components of the Close Binary Joshi, M.M. (2003) Climate model studies of Stars, ASP Conference Series 318, edited by synchronously rotating planets. Astrobiology 3, R.W. Hilditch, H. Hensberge, and K. 415-427. Pavlovskipp, San Francisco, Calif., pp. 159-165. Joy, A.H. and Abt, H.A. (1974) Spectral types of M Henry, T.J., Subasavage, J.P., Brown, M.A., dwarf stars. Astrophys. J. Suppl. S. 28, 1-18. Beaulieu, T.D., Jao, W.C., and Hambly, N.C. Kaiser, N., et al. (2005) DPS, 37.0301: The Pan- (2004) The solar neighborhood X. New nearby STARRS Large Survey Telescope Project. stars in the southern sky and accurate Kalas, P., Liu, M.C., and Mathews, B.C. (2004) photometric distance estimates for red dwarfs. Discovery of a large dust disk around the nearby The Astron. J. 128, 2460. star AU microscopii. Science 303, 1990. Henry, T.J., Jao, W.C., Subasavage, J.P., Beaulieu, Kasting, J.F., Whitmire D.P., and Reynolds R.T. T.D., Ianna, P.A., Costa, E., and Mendez, R.A. (1993) Habitable zones around main sequence (2006) The solar neighborhood. XVII. parallax stars. Icarus 101, 108-128. results from the CTIOPI 0.9m program: 20 new Kasting, J.F. and Pollack, J.B. (1983) Loss of water members of the RECONS 10 pc sample. Astron. from Venus. I: Hydrodynamic escape of J. in press. hydrogen. Icarus 53, 479-508. Holland, H.D. (1978) The Chemistry of the Kaula, W.M. (1995) Venus reconsidered. Science Atmosphere and Oceans. Princeton University 270, 1460-1464. Press, Princeton. Khodachenko, M.L., Ribas, I., Lammer, H, Houdebine, E.R. (2003) Dynamics of flares on late Grießmeier, J.-M., Leitner, M., Selsis, F., Eirora, type dMe stars. IV. Constraints from C., Hanslmeier, A., Biernat, H.K., Farrugia, C.J. spectrophotometry in the visible. Astron. and Rucker, H.O. (2006) CME activity of low Astrophys. 397, 1019-1034. mass M stars as an important factor for the Huang, S-S. (1959) The problem of life in the habitability of terrestrial exoplanets I: CME universe and the mode of star formation. PASP impact on expected magnetospheres of Earth-like 71, 421-424. exoplanets in close-in habitable zones. Huang, S-S. (1960) Life outside the solar system. Sci. Astrobiology, this issue. Am. 202, 55-63. Koller, L.R. (1965) Ultraviolet Radiation. John Hunten, D.M., Pepin, R.O., and Walker, J.C.G. Wiley and Sons, London, U.K. (1987) Mass fractionation in hydrodynamic Kortenkamp, S.J., Wetherill, G.W., Inaba, S. (2001) escape. Icarus 69, 532-549. Runaway growth of planetary embryos Hunten, D.M (1993) Atmospheric evolution of the facilitated by massive bodies in a protoplanetary terrestrial planets. Science, 259, 915-920. disk. Science 293, 1127-1129. Ida, S., Lin, D.N.C. (2005) Dependence of exoplanets Krist, J.E., Ardila, D.R., Golimowski, D.A., Clampin, on host stars’ metallicity and mass, Prog. M., Ford, H.C., Illingworth, G.D., Hartig, G.F., Theoretical Physics Suppl. 158, 68-85. Bartko, F., Benítez, N., Blakeslee, J.P., Jao, W-C., Henry, T.J., Subasavage, J.P., Brown, Bouwens, R.J., Bradley, L.D., Broadhurst, T.J., M.A., Ianna, P.A., Bartlett, J.L., Costa, E., and Brown, R.A., Burrows, C.J., Cheng, E.S., Cross, Méndez, R.A. (2005) The solar neighborhood. N.J.G., Demarco, R., Feldman, P.D., Franx, M., XIII. Parallax results from the CTIOPI 0.9 meter Goto, T., Gronwall, C., Holden, B., Homeier, N., program: Stars with µ >= 1.0" yr-1 (MOTION Infante, L., Kimble, R.A., Lesser, M.P., Martel, Sample). Astron. J. 129, 1954-1967. A.R., Mei, S., Menanteau, F., Meurer, G.R.,

30 Miley, G.K., Motta, V., Postman, M., Rosati, P., Lodders, K. and Fegley, B. (1998) The Planetary Sirianni, M., Sparks, W.B., Tran, H.D., Scientists’s Companion, Oxford University Tsvetanov, Z.I., White, R.L., and Zheng, W. Press, New York. (2005) Hubble Space Telescope advanced Low, F.J., Smith, P.S., Werner, M. Chen, C., Krause, camera for surveys coronagraphic imaging of the V., Jura, M., and Hines, D.C. (2005) Exploring AU Microscopii Debris Disk. Astron. J. 129, terrestrial planet formation in the TW Hydrae 1008. association. Astrophys. J., 631, 1170-1179. Krumholz, L.R., McKinley, J.P., Ulrich, G.A., and Luhman, K.L., Lada, C. J., Hartmann, L., Muench, A. Suflita, J.M. (1997) Confined subsurface A., Megeath, S. T., Allen, L. E., Myers, P. C., microbial communities in Cretaceous rock. Muzerolle, J., Young, E., Fazio, G.G. 2005, The Nature 386, 64-66. disk fractions of Brown Dwarfs in IC 348 and Lada C.J., Muench, A.A., Luhman, K. L., Allen, L., Chamaeleon I. ApJL 631, 69. Hartmann, L., Megeath, T., Myers, P., Fazio, G., Lunding, R., Zakharov, A., Pellinen, R., Borg, H., Wood, K., Muzerolle, J., Rieke, G., Siegler, N., Hultqvist, B., Pissarenko, B.N., Dubinin, E.M., and Young, E. (2006) Spitzer observations of IC Barabash, S.W., Liede, I., and Koskinen, H. 348: The disk population at 2-3 million years. AJ (1989) First measurement of the ionospheric 131, 1574. plasma escape from Mars. Nature 341, 609. Lamers, H.J.G.L.M. and Cassinelli, J.P. (1999) Mackwell, S.J., Zimmerman, M.E., and Kohlstedt, Introduction to Stellar Winds, Cambridge D.L. (1998) High-temperature deformation of University Press, Cambridge, UK. dry diabase with application to tectonics on Lammer, H., Lichtenegger, H.I.M., Kulikov, Y.N., Venus. J. Geophys. Res. 103, 975-984. Grießmeier, J.-M., Terada, N., Erkaev, N.V., Mathis, J.S., Rumpl, W., and Nordsieck, K.H. (1977) Biernat, H.K., Khodachenko, M.L., Ribas, I., The size distribution of interstellar grains. Penz, T., and Selsis, F. (2006) CME activity of Astrophys. J. 217, 425-433. low mass M stars as an important factor for the Matsuura, M., Zijlstra, A.A., Molster, F.J., Hony, S., habitability of terrestrial exoplanets II: CME Waters, L. B. F. M., Kemper, F., Bowey, J. E., induced ion pick-up of Earth-like exoplanets in Chihara, H., Koike, C., and Keller, L. P. (2004) close-in habitable zones. Astrobiology, this issue. Polycyclic aromatic hydrocarbons and crystalline Larcher, W. (1995) Photosynthesis as a tool for silicates in the bipolar post-asymptotic Giant indicating temperature stress events. In Branch Star IRAS 16279-4757. Astrophys. J. Ecophysiology of Photosynthesis, edited by E.-D. 604(2), 791-799. Schulze and M. M. Caldwell, Springer-Verlag, Mayor, M. and Queloz, D. (1995) A Jupiter-mass Berlin, pp. 263-277. companion to a solar-type star. Nature 378, 355. Laughlin, G. and Bodenheimer, P. (1993) Luminosity Metchev, S.A. and Hillenbrand, L.A. (2004) functions for very low mass stars and brown Astrophys. J. 617, 1330. dwarfs. Astrophys. J. 403, 303-314. Mirzoyan, L.V. (1990) Flare stars in star clusters, Laughlin, G., Bodenheimer, P., and Adams, F.C. associations and the solar vicinity. In Flare Stars (1997) The end of the main sequence. Astrophys. in Star Clusters, Associations and the Solar J. 482, 420-432. Vicinity, IAU Symposium 137, edited by L.V. Laughlin, G., Bodenheimer, P., and Adams, F.C. Mirzoyan, et al., Kluwer Academic Press, (2004) The core accretion model predicts few Dordrecht, pp. 1-13. Jovian-mass planets orbiting red dwarfs. Mohanty, S. and Basri, G. (2003) Rotation and Astrophys. J. 612, L73-L76. activity in mid-M to L field dwarfs. Astrophys. J. Leger, A., Pirre, M., and Marceau, F.J. (1993) Search 583, 451-472. for primitive life on a distant planet: relevance of Montgomery, R. and Laughlin, G. (2006) AbSciCon O2 and O3 detections. Astron. Astrophys. 277, 2006 abstract, Washington, DC. 309-313. Moór, A., Ábrahám, P., Derekas, A., Kiss, Cs., Kiss, Leitherer, C., and Robert, C. (1991) Observations of L.L., Apai, D., Grady, C., and Henning, Th. stellar winds from hot stars at 1.3 millimeters. (2006) Nearby debris disk systems with high Astrophys. J. 377, 629-638. fractional luminosity reconsidered. ApJ 644, 525. Lissauer, J.J. (1987) Timescales for planetary Moore, W. (2003) Tidal heating and convection in Io. accretion and the structure of the protoplanetary J. Geophys. Res. 108, disk. Icarus 69, 249-265. doi:10.1029/2002JE001943. Liu, M.C. (2004) Science 305, 1442. Morbidelli, A., Chambers, J., Lunine, J.I., Petit, J.M., Liu, M.C., Matthews, B.C., Williams, J.P., and Kalas, Robert, F., Valsecchi, G.B., and Cyr, K.E. P.G. (2004) Astrophys. J. 608, 526. (2000) Source regions and timescales for the

31 delivery of water to the Earth. Meteoritics Poveda, A., Allen, C., Herrera, M.A., Cordero, G., Planet. Sci. 35, 1309-1320. and Lavalley, C. (1996) Kinematics, ages, and Muench, A., Alves, J., Lada, C., and Lada, E. (2001) evolutionary status of UV Ceti stars. Astron. Evidence for circumstellar disks around young Astrophys. 308, 55-65. Brown Dwarfs in the Trapezium Cluster. ApJL Preibisch, T. and Feigelson, E.D. (2005) The 558, 51. evolution of X-ray emission in young stars. Mullen, D.J., Sion, E.M., Bruhweiler, F.C., and Astrophys. J. Suppl., in press. Carpenter, K.G. (1989) Evidence for a cool wind Raymond, S., Meadows, V., and Scalo, J. (2006) from the K2 dwarf in the detached binary V471 AbSciCon 2006 abstract, Washington, DC. Tauri. Astrophys. J. Lett. 339, L33-L36. Raymond, S., Quinn, T., and Lunine J.I. (2006) High Mullen, D.J., Doyle, J.G., Redman, R.O., and resolution simulations of the final assembly of Mathioudakis, M. (1992) Limits on detectability Earth-like planets 2: Water delivery, in press. of mass loss from cool dwarfs. Astrophys. J. 397, Rieke, G. H., Su, K. Y. L., Stansberry, J. A., Trilling, 225-231. D., Bryden, G., Muzerolle, J., White, B., Mullan, D.J. and MacDonald, J. (2001) Are Gorlova, N., Young, E. T., Beichman, C. A., magnetically active low-mass M dwarfs Stapelfeldt, K. R., and Hines, D. C. (2005) completely convective? Astrophys. J. 559, 353- Decay of planetary debris disks. Astrophys. J. 371. 620, 1010. Najita and Williams (2005) Astrophys. J. in press Rietmeijer, F.J.M., Hallenbeck, S.L., Nuth, J.A. III, (astro-ph/0508165). and Karner, J.M. (2002). Amorphous Nobel, P.S. (1974) Introduction to Biophysical Plant magnesiosilicate smokes annealed in vacuum: Physiology, W.H. Freeman and Company, San the evolution of magnesium silicates in Francisco. circumstellar and cometary dust. Icarus 156(1), Olson, P.L. and Christensen, U.R. (2006) Dipole 269-286. moment scaling for convection-driven planetary Rietmeijer, F.J.M., Nuth, J.A. III, and Karner, J.M. dynamos. Earth Planet. Sci. Lett., in press. (1999) Metastable eutectic, gas to solid, Osten, R.A., Hawley, S L., Allred, J.C., Johns-Krull, condensation in the FeO-Fe2O3-SiO3 system. C.M., and Roark, C. (2005) From radio to X-ray: Phys. Chem. Chem. Phys. 1, 1511-1516. flares on the dMe flare star EV Lacertae. Rietmeijer, F.J.M., Nuth, J.A. III, Karner, J.M., and Astrophys. J. 621, 398-416. Hallenbeck, S.L. (2002) Gas-to-solid Pendleton, Y.J., Sandford, S.A., Allamandola, L.J., condensation in a Mg-SiO-H2-O2 vapor: Tielens, A.G.G.M., and Sellgren, K. (1994) Near metastable eutectics in the MgO-SiO2 phase infrared spectroscopy of hydrocarbon absorption diagram. Phys. Chem. Chem. Phys. 4, 546-551. in the interstellar medium. Astrophys. J. 437, Rivera, E., and 8 other authors (2005) A ~7.5 Earth- 683-696. mass planet orbiting the nearby star, GJ 876. Pendleton, Y.J. and Allamandola, L.J. (2002) The Astrophys. J. 634, 625-640. organic refractory material in the diffuse Roberge, A., Weinberger, A.J., Redfield, S., and interstellar medium: Mid-infrared spectroscopic Feldman, P.D. (2005) Rapid dissipation of constraints. ApJ. Suppl. Series 138(1), 75-98. primordial gas from the AU Microscopii debris Perryman, M.A.C. (2005) Overview of the Gaia disk. Astrophys. J. 626, 105. mission. In Astrometry in the Age of the Next Roberts, P.H. and Glatzmaier, G.A. (2000) Generation of Large Telescopes, edited by P. Geodynamo theory and simulations. Rev. Mod. Kenneth Seidelmann and A.K.B. Monet, San Phys. 72, 1081-1123. Francisco, p. 3. Robinson, R. D., et al. (2005) GALEX observations Pettersen, B.R. (1991) The nearby flare stars. Mem. of an energetic ultraviolet flare on the dM4e star Soc. Astr. Ital. 62, 217-242. GJ 3685A. ApJ 633, 447-451. Phillips, R. J., Bullock, M., Hauck, S.A. II (2001) Rodono M. (1986) The atmospheres of M dwarfs: Climate and interior coupled evolution on observations. In The M-type Stars, NASA SP- Venus. Geophys. Res. Lett. 28, 1779-1782, doi: 492, edited by H.R. Johnson and F.R. Querci, US 10.1029/2000GL011821 Govt. Printing Office, Washington DC, pp. 409- Pollack, J.B., Hubickyj, O., Bodenheimer, P. 453. Lissauer, J.J., Podolak, M., and Greenzweig, Y. Rothschild, L.J. and Mancinelli, R.L. (2001) Life in (1996) Formation of the giant planets by extreme environments. Nature 409, 1092. concurrent accretion of solids and gas. Icarus Rothschild, L.J. and Giver, L.J. (2003) 124, 62-85. Photosynthesis below the surface in a cryptic microbial mat. Intl. J. Astrobiology 1, 295-304.

32 Saar, S.H. and Linsky, J.L. (1985) The photospheric not migrate? Global Ecology and Biogeography magnetic field of the dM3.5e flare star AD Letters 6, 139-148. Leonis. ApJL 299, L47-L50. Song, I., Weinberger, A. J., Becklin, E. E., Sagan, C. and Mullen, G. (1972) Earth and Mars: Zuckerman, B., and Chen, C. (2002) M-type evolution of atmospheres and surface Vega-like stars. Astron. J. 124, 514. temperatures. Science 177, 52-56. Stacey, F.D. (1992) Physics of the Earth. Brookfield Salisbury, F.B. and Ross, C.W. (1978) Plant Press, Brisbane. Physiology, California, Wadsworth, Belmont. Stevenson, D.J. (2003) Planetary magnetic fields. Sandford, S.A., Allamandola, L.J., Tielens, A. G. G. Earth Planet. Sci. Lett. 208, 1-11. M., Sellgren, K., Tama, M., et al. (1991) The Tinneti, G., Meadows, V. Crisp, D., Kiang, N., Kahn, interstellar C-H stretching band near 3.4 B.H., Bosc, E., Fishbein, E., Velusamy, T., and microns: constraints on the composition of Turnbull, M. (2005) Detectability of planetary organic material in the diffuse interstellar characteristics in disk-average spectra II: medium. Astrophys. J. 371, 607-620. Synthetic spectra and light-curves of Earth. Sass, J.H. (1972) The Earth’s heat and internal Astrobiology, submitted. temperatures. In Understanding the Earth, 2nd Turnbull, M.C. and Tarter, J.C. (2003a) Target Edition, edited by I.G. Gass, P.J. Smith, and selection for SETI I: a catalog of nearby R.C.L. Wilson, Open University Press, Sussex, habitable stellar systems. Astrophys. J. Suppl. pp. 81-87. 145, 181-198. Segura, A., Krelove, K., Kasting, J.F., Sommerlatt, Turnbull, M.C. and Tarter, J.C. (2003b) Target D., Meadows, V., Crisp, D., Cohen, M., and selection for SETI II: Tycho-2 dwarfs, old open Mlawer, E. (2003) Ozone concentrations and clusters, and the nearest 100 stars. Astrophys. J. ultraviolet fluxes on Earth-like planets around Suppl. 149, 423-436. other stars. Astrobiology 3, 689-708. Tyson, J.A. (2002) Large synoptic survey telescope: Segura A., Kasting, J.F., Meadows, V., Cohen, M., overview. SPIE 4836, 10. Scalo, J., Crisp, D., Butler, R.A.H., and Tinetti, Valencia, D., O’Connell, R. J., Sasselov, D. (2006) G. (2005) Biosignatures from Earth-like planets Internal structure of massive terrestrial planets. around M-dwarfs. Astrobiology 5, 706-725. Icarus 181, 545-554. Scalo, J., Kaltenegger, L., Segura, A., Fridlund, M, Valenti, J.A., Johns-Krull, C. (2001) Magnetic field Ribas, I., Kulikov, Y.N., Grenfell, J.L., Rauer, measurements for cool stars. In Magnetic Fields H., Odert, P., Leitzinger, M., Selsis, F., Across the Hertzsprung-Russell Diagram, ASP Khodachenko, M.L., Eiroa, C., Kasting, J. and Conf. Ser. 248, p. 179. Lammer, H. (2006) M stars as targets for van Altena, W.F., Lee, J.T., and Hoffleit, E.D. (1995) terrestrial exoplanet searches and biosignature The General Catalogue of Trigonometric Stellar detection. Astrobiology, this issue. Parallaxes, 4th edition, Yale Univ. Obs., New Schubert, G., Cassen, P., and Young, R.E. (1979) Haven. Subsolidus convective cooling histories of Van den Berg, A.P., Beebe, G., and Yuen, D.A. terrestrial planets. Icarus 38, 192-211. (2005) Mantle dynamics of super-Earth Schubert, G., Turcotte, D., and Olson, P. (2001) extrasolar planet under extreme temperature and Mantle Convection in the Earth and Planets, pressure conditions. Eos Trans. AGU 86, Fall Cambridge University Press, Cambridge. Meet. Suppl., Abstract MR23B-0061. Showman, A.P. and Malhotra, R. (1999) The van den Besselaar, E.J.M., Raassen, A.J.J., Mewe, R., Galilean satellites. Science 286, 77-84. van der Meer, R.L.J., Gudel, M., and Audard, M. Sicilia-Aguilar, A., Hartmann, L.W., Hernandez, J. (2004) AD Leonis: Flares observed by XMM- Briceno, C., and Calvet, N. (2005) Astron. J. Newton and Chandra. Astron. Astrophys 411, 130, 188. 587-593. Silvestri, N.M., Hawley, S.L., and Oswalt, T.D. Vermeij, G. (2006) Historical contingency and the (2005) Astron. J. 129, 2428-2450. purported uniqueness of evolutionary Sleep, N.H., (2000) Evolution of the mode of innovations. PNAS 103, 1804-1809. convection within terrestrial planets. J. Geophys. Vincente, S., and Alves, J. (2005) Size distribution of Res. 105, 17563-17578. circumstellar disks in the Trapezium cluster. Smith, K., Gudel, M., and Audard, M. (2005) Flares Astron. and Astrophys. 441, 195-205. observed with XMM-Newton and the VLA. Vrba, F.J., Henden, A. A., Luginbuhl, C. B., Guetter, Astron. Astrophys. 436, 241-251. H. H., Munn, J. A., Canzian, B., Burgasser, A. J., Solomon, A.M. and Kirilenko, A.P. (1997) Climate Kirkpatrick, J.D., Fan, X., Geballe, T. R., change and terrestrial biomass: what if trees do Golimowski, D. A., Knapp, G. R., Leggett, S. K.,

33 Schneider, D. P., and Brinkmann, J. (2004) Wetherill, G.W. (1996) The formation and Preliminary parallaxes of 40 L and T dwarfs habitability of extra-solar planets. Icarus 119, from the US Naval Observatory Infrared 219-238. Astrometry Program. Astron. J. 127, 2948. Williams, D.M. and Pollard, D. (2002) Earth-like Walker, J.C.G., Hays, P.B., and Kasting, J.F. (1981) worlds on eccentric orbits: excursions beyond A negative feedback mechanism for the long- the habitable zone. International Journal of term stabilization of the Earth’s surface Astrobiology 1(1), 61-69. temperature. J. Geophys. Res. 8, 9,77-79,782. Whitmire, D.P. and Doyle, L.R. (1995) A slightly Walker, J.C.G. (1977) Evolution of the Atmosphere, more massive sun as the explanation of warm Macmillan Publishing Co., Inc. ISBN 0-02- temperatures on early Mars. Journal of 854390-4. Geophysical Research: Planets 100, 5457-5464. Walkowicz, L.M., Hawley, S.L., and West, A.A. Wood, B.E., Linsky, J.L., Muller, H.-R., and Zank, (2004) The χ factor: determining the strength of G.P. (2001) Astrophys. J. 547, L49. activity in low mass dwarfs. PASP 116, 1105- Woolum, D.S., and Cassen, P. (1999) Astronomical 1110. constraints on nebular temperatures: implications Welsh, B.Y.; Wheatley, J.M., Seibert, M., Browne, for planetesimal formation. Meteoritics Planet. S.E., West, A.A., Siegmund, O.H.W., Barlow, Sci. 34, 879-907. T.A., Forster, K., Friedman, P.G., Martin, D.C., Wray, J., Liu, M., and Reid, I. (2005) New debris Morrissey, P., Small, T., Wyder, T., disks around low-mass stars. BAAS 2007, 10.07. Schiminovich, D., Neff, S., and Rich, R.M. Wright, A.E. and Barlow, M.F. (1975) The radio and (2006a) The detection of M-dwarf UV flare infrared spectrum of early-type stars undergoing events in the GALEX data archives. ApJS, in mass loss. Monthly Notices of the Royal press (http://arxiv.org/abs/astro-ph/0605328). Astronomical Society 170, 41-51. Welsh, B.Y.; Wheatley, J.M.; Browne, S.E.; Zhou, J.-L., Aarseth, S.J.; Lin, D.N.C.; Nagasawa, M. Siegmund, O.H.W.; Doyle, J.G.; O'Shea, E.; (2005) Origin and ubiquity of short-period Earth- Antonova, A.; Forster, K.; Seibert, M.; like planets: evidence for the sequential accretion Morrissey, P.; Taroyan, Y. (2006b) GALEX high theory of planet formation. Astrophys. J. Lett. time-resolution ultraviolet observations of dMe 631, L85-L88. flare events. A and A, in press Zubko, V., Dwek, E., and Arendt, R.G. (2004) (http://arxiv.org/abs/astro-ph/0608254). Interstellar dust models consistent with West, A.A., Hawley, S.L., Walkowicz, L.M., Covey, extinction, emission, and abundance constraints. K.R., Silvestri, N.M., Raymond, S.N., Harris, Astrophys. J. Suppl. Series 151(V151/58303), 1- H.C., Munn, J.A., McGehee, P.M., Ivezić, 40. Željko, and Brinkmann, J. (2004) Spectroscopic Zuckerman, B. and Becklin, E.E. (1993) properties of cool stars in the Sloan Digital Sky Submillimeter studies of main-sequence stars. Survey: an analysis of magnetic activity and a Astrophys. J. 414, 793. search for subdwarfs. Astron. J. 128, 426-436. Wetherill, G.W. (1990) Formation of the Earth. Ann. Rev. Earth Planet. Sci. 18, 205-256. Wetherill, G.W. (1994) Possible consequences of absence of “Jupiters” in planetary systems. Astrophys. Space Sci. 212, 23-32.

34