Introduction
The current status of exoplanet characterization shows a surprisingly diverse set of giant planets. For a subset of these, some properties have been measured or inferred using radial velocity (RV), micro-lensing, transits and astrometry. These observations have yielded measurements of planetary mass, orbital elements and the planetary radius and during the last few years, physical and chemical characteristics of the upper atmosphere of some of the transiting planets. Specifically, observations of transits, combined with RV information, have provided estimates of the mass, radius and density of a subset of these planets, ranging from giant planets to rocky planets such as Corot 7b (Léger et al. Reference Léger2009) and Kepler 10b (Batalha et al. Reference Batalha2011). Due to a detection bias for both transiting and the RV method, most detected planets orbit extremely close to their host star and therefore receive high amounts of stellar irradiation and have subsequent high surface temperature. The first directly imaged exoplanet candidates around young stars show the improvement in direct detection techniques that are designed to resolve the planet and collect its photons. This can currently be achieved for widely separated young objects and has already detected exoplanets candidates (see e.g. Kalas et al. Reference Kalas, Graham, Chiang, Fitzgerald, Clampin, Kite, Stapelfeldt, Marois and Krist2008; Marois et al. Reference Marois, Macintosh, Barman, Zuckerman, Song, Patience, Lafrenière and Doyon2008; Lagrange et al. Reference Lagrange2009).
Recent investigations of high precision RV data samples have shown that between 20 and 50% of all sample stars exhibit RV variations indicating the presence of super-Earths or ice giants (Howard et al. Reference Howard, Marcy, Bryson, Jenkins, Rowe, Batalha and Borucki2011). Among the hundreds of confirmed planets, already a few close by, low mass RV planets such as Gl 581 d (Udry et al. Reference Udry2007), with minimum masses below 10 Earth masses, MEarth, orbit in the habitable zone (HZ) of their parent star. These close by planets provide excellent targets for future atmospheric exploration. Several Kepler transit planetary candidates from the February 2011 data release (Borucki et al. Reference Borucki2011) that are consistent with rocky models, orbit their host stars also in the HZ, providing first statistics of the number of planets and Earth-like planets (etaEarth) (see e.g. Traub Reference Traub2011) and a more complete sample is expected in the second Kepler data release in 2012.
The discovery of transiting planets with masses below 10 MEarth and radii consistent with rocky planetary models, answered the important question if planets more massive than Earth could potentially be rocky. Ten MEarth are used from formation theories as the upper limit for rocky planet formation, for comparison, Uranus has about 15 MEarth. Above that mass the planet is thought to accumulate a substantial amount of gas that makes it akin to a gas giant, not a rocky planet with an outgassed atmosphere, what we use as the definition of Super-Earth here. Where exactly such a cut-off mass is – if it exists at all – an open question.
Recent discoveries by ground based, as well as the Corot and Kepler space-mission, found planets with radii below 2 Earth radii and masses below 10 MEarth and densities akin to Neptune as well as Earth, suggesting that there is not one cut-off mass above which a planet is like Neptune and below which it is rocky like Earth or Venus. The first planets below 10 MEarth, with both mass estimates and radius measurements, have provided a wide range of observed radii and densities (e.g., Seager et al. Reference Seager, Kuchner, Hier-Majumder and Militzer2007; Sotin et al. Reference Sotin, Grasset and Mocquet2007; Valencia et al. Reference Valencia, Sasselov and O'Connell2007; Grasset et al. Reference Grasset, Schneider and Sotin2009).
Especially in the mass range below 5MEarth, two planets in the multiple planet system, Kepler 11b and Kepler 11f (Lissauer et al. Reference Lissauer2011), with 4.3 and 2.3 MEarth have radii of 1.97 and 2.61 Earth radii and mean densities of 3.1 and 0.7 g/cm3, respectively. These derived densities allow substantial envelopes of light gases for this mass range. Recent atmosphere observations for a 6.55 ± 0.98 Earth mass planet, GJ 1214 b (Charbonneau et al. Reference Charbonneau2009Reference Christensen and Pearl9; Bean et al. Reference Bean, Miller-Ricci Kempton and Homeier2011; Désert et al. Reference Désert2011), with a mean density of 1.8 g/cm3, indicates either hazes or high cloud cover in an expanded atmosphere (Miller-Ricci et al. Reference Miller-Ricci Kempton, Zahnle and Fortney2012). For comparison Neptune has a mean density of 1.6 g/cm3, Earth a mean density of 5.5 g/cm3.
Observing mass and radius alone cannot break the degeneracy of a planet's nature due to the effect of an extended atmosphere that can block the stellar light and increase the observed planetary radius significantly. Even if a unique solution would exist, planets with similar density, such as Earth and Venus, present very different planetary environments in terms of habitable conditions. Therefore, the question refocuses on atmospheric features to characterize a planetary environment. Future space missions have the explicit purpose of detecting other Earth-like worlds, analysing their characteristics, determining the composition of their atmospheres, investigating their capability to sustain life as we know it and searching for signs of life. They also have the capacity to investigate the physical properties and composition of a broader diversity of planets, to understand the formation of planets and interpret potential biosignatures.
In this paper, we discuss how we can read a planet's spectral fingerprint and characterize if it is potentially habitable. In section 1, we discuss the hot, close in planetary sample, in section 2, the first steps to detect a habitable planet, set biomarker detection in context and focuses on low resolution biomarkers in the spectrum of an Earth-like planet are discussed. In section 3, we discuss spectral evolution of a habitable planet, abiotic sources of biomarkers and Earth's spectra around different host stars, and section 4 summarizes the article.
Hot rocky exoplanets
Exoplanet surveys have already discovered the first rocky, strongly irradiated exoplanets: Corot 7b (Léger et al. Reference Léger2009) and Kepler 10b (Batalha et al. Reference Batalha2011). A whole population of hot planets with minimum masses below 10 Earth masses, consistent with rocky planet models (super-Earths) has already been detected. In the Kepler February 2011 data release (Borucki et al. Reference Borucki2011), the Kepler team presented 997 stars with a total of 1235 planetary candidates. Six hundred and fifteen of these objects have radii consistent with rocky planetary models (smaller than 2 Earth radii within measurement errors) and a subsample of 193 might have extremely high temperatures on their surfaces (T P > 1000 K), due to stellar irradiation alone, assuming a rocky planet model with a Bond albedo of 0.01 (Miguel et al. Reference Miguel, Kaltenegger, Fegley and Schaefer2011).
Such extreme conditions are not seen on any planet in our Solar System and therefore the potential atmospheric composition of these hot super-Earths remains largely unknown. For temperatures above 1000 K, the surface of a rocky planet would be covered with a magma ocean which will vapourize, forming an outgassed atmosphere. Schaefer & Fegley (Reference Schaefer and Fegley2009) modelled this process for Corot 7b, showing a silicate atmosphere is possible for such hot objects. Miguel et al. (Reference Miguel, Kaltenegger, Fegley and Schaefer2011) developed a simple approach linking the observable data of a planet (radius, semi-major axis and stellar effective temperature) to the atmospheric composition for such hot planets. The results were applied to the hot, potentially rocky, planets in the Kepler sample computing the chemical equilibrium between the melt and vapour in a magma exposed at temperatures between 1000 and 3500 K using the MAGMA-code (Fegley & Cameron Reference Fegley and Cameron1987; Schaefer & Fegley Reference Schaefer and Fegley2004).
The resulting column densities of the atmospheric gases outgassed for a planet of 10 Earth masses as a function of the surface temperature are shown in Fig. 1(a) with the most abundant gases being Fe, Mg, Na, O, O2 and SiO. The abundance of each gas in the atmosphere depends on the surface temperature. The temperatures where the abundance of the most important gases change significantly are indicated in the figures, and define different atmospheric types (Miguel et al. Reference Miguel, Kaltenegger, Fegley and Schaefer2011). Fig. 1(b) shows the stellar effective temperature versus the planet's distance to its star. Kepler planetary candidates with T p > 1000 K are also shown in different point sizes according to their radius: 2–2.5 Earth radii (large points), 1.5–2 Earth radii (medium points) and candidates with R p < 1.5 Earth radii (small points). According to these results, there are five types of initial atmospheres for hot rocky planets (for details, see Miguel et al. Reference Miguel, Kaltenegger, Fegley and Schaefer2011).
Fig. 1. Column densities of atmospheric gases as a function of the planet surface temperature (left panel). Temperatures where the most abundant gases change are indicated. Stellar effective temperature versus semi-major axis (right panel) (adapted from Miguel et al. Reference Miguel, Kaltenegger, Fegley and Schaefer2011).
Characterize a habitable planet (learning from Earth)
In the next few years, ground- and space-based missions will give us statistics on the number, size, period and orbital distance of planets, extending to terrestrial planets on the lower mass range end as a first step, while future space missions are designed to characterize their atmospheres. A temperate planet is a very faint, small object close to a very bright and large object, its parent star. For directly imaging planets, in the visible part of the spectrum we observe the starlight, reflected off the planet, in the IR we detect the planets own emitted flux. The Earth–Sun intensity ratio is about 10−7 in the thermal infrared (IR) (∼10 μm) and about 10−10 in the visible (∼0.5 μm) (see Fig. 2). Nevertheless, the contrast ratio of hot extrasolar Giant planets (EGP) to their parent stars as well as the contrast ratio of a planet to a smaller parent star is much more favourable, making Earth-like planets around small stars very interesting targets.
Fig. 2. Model of our Solar System (left) (assumed here to be Black Bodies with Earth spectrum shown). Synthetic transmission spectra (right) of the Earth from UV to IR shown. The intensity is given as a fraction of solar intensity as well as the relative height in the atmosphere. The atmospheric features are indicated (Kaltenegger et al. Reference Kaltenegger, Segura and Mohanty2011).
Low mass Main Sequence M dwarfs are the most abundant stars in the galaxy, representing about 75% of the total stellar population. Many planets including potentially rocky planets within the HZ, such as Gl581d (see e.g. Kaltenegger et al. Reference Kaltenegger, Segura and Mohanty2011; von Paris et al. Reference von Paris, Gebauer, Godolt, Grenfell, Hedelt, Kitzmann, Patzer, Rauer and Stracke2010; Wordsworth et al. Reference Wordsworth, Forget, Selsis, Millour, Charnay and Madeleine2011), are thus likely to be found in the near future. These will provide excellent targets (see e.g. Scalo et al. Reference Scalo2007) that can be probed for atmospheric components, especially for hot planets with extended atmospheres.
Different strategies exist to characterize a planet's atmosphere: direct detection resolves the planet and star individually, and transmission as well as secondary eclipse measurements subtract the stellar light from a combined star–planet detection. Future telescopes will allow for direct detection in close orbits for close by stars.
A spectral fingerprint of an Earth-like atmosphere
The atmosphere of a planet contains the detectable information to explore the planetary environment remotely. On Earth some atmospheric species exhibiting noticeable spectral features in the planet's spectrum result directly or indirectly from biological activity: the main ones are O2, O3, CH4 and N2O (see Fig. 2). Figure 2 shows the detectable features in the planet's reflection, emission and transmission spectrum using the Earth itself as a proxy for observations and model fits to data of spectra of the Earth (Kaltenegger et al. Reference Kaltenegger, Traub and Jucks2007; Kaltenegger & Traub Reference Kaltenegger and Traub2009). CO2 and H2O are in addition important as greenhouse gases in a planet's atmosphere and potential sources for high O2 concentration from photosynthesis. Sagan et al. (Reference Sagan, Thompson, Carlson, Gurnett and Hord1993) analysed a spectrum of the Earth taken by the Galileo probe, searching for signatures of life and concluded that the large amount of O2 and the simultaneous presence of CH4 traces are strongly suggestive of biology for a planet around a Sun-like star.
Figure 3 shows observations and model fits to spectra of the Earth in three wavelength ranges (Kaltenegger et al. Reference Kaltenegger, Traub and Jucks2007). The data shown in Fig. 3 (left) are the visible Earthshine spectrum (Woolf et al. Reference Woolf, Smith, Traub and Jucks2002), (centre) the near-IR (NIR) Earthshine spectrum (Turnbull et al. Reference Turnbull, Traub, Jucks, Woolf, Meyer, Gorlova, Skrutskie and Wilson2006) and (right) the thermal IR spectrum of Earth as measured by a spectrometer en route to Mars (Christensen & Pearl Reference Christensen and Pearl1997). The data are shown in black and the SAO model in red. In each case, the constituent gas spectra in a clear atmosphere are shown in the bottom panel, for reference.
Fig. 3. Observed reflectivity spectrum in the visible (Woolf et al. Reference Woolf, Smith, Traub and Jucks2002) (top), NIR (Turnbull et al. Reference Turnbull, Traub, Jucks, Woolf, Meyer, Gorlova, Skrutskie and Wilson2006) (middle) and emission spectrum in the IR (Christensen & Pearl Reference Christensen and Pearl1997) (bottom panel) of the integrated Earth, as determined from Earthshine and space, respectively. The data are shown in black and the SAO model in red. The reflectivity scale is arbitrary (Kaltenegger et al. Reference Kaltenegger, Traub and Jucks2007).
Our search for signs of life in exoplanets is based on the assumption that extraterrestrial life shares fundamental characteristics with life on Earth, in that it requires liquid water as a solvent and has a carbon-based chemistry (see e.g. Brack Reference Brack1993; DesMarais et al. Reference Des Marais, Harwit, Jucks, Kasting, Lin, Lunine, Schneider, Seager, Traub and Woolf2002). Life based on a different chemistry is not considered here because these life forms would produce signatures in their atmosphere that are so far unknown. Therefore, we assume extraterrestrial life is similar to life on Earth in its use of the same input and output gases, and that it exists out of thermodynamic equilibrium (Lovelock Reference Lovelock1975).
The viewing geometry results in different flux contribution of the overall detected signal from the bright and dark side, for the reflected light, the terminator region for transmitted light and the planet's hot and cold regions for the emitted flux. Biosignatures are used here to mean detectable species, or set of species, whose presence at significant abundance strongly suggests a biological origin (e.g. couple CH4 + O2, or CH4 + O3). It is their quantities and detection along with other atmospheric species, in a certain context (for instance the properties of the star and the planet) that points toward a biological origin. O2, O3 with a reducing gas such as CH4 are good biosignatures that can be detected by a low-resolution (resolution < 80) spectrograph.
Visible and NIR as well as IR spectral regions contain the signature of atmospheric gases that can be observed with low resolution and can indicate habitable conditions and, possibly, the presence of a biosphere: CO2, H2O, O3, CH4 and N2O in the thermal IR, and H2O, O3, O2, CH4 and CO2 in the visible to NIR (see e.g. DesMarais et al. Reference Des Marais, Harwit, Jucks, Kasting, Lin, Lunine, Schneider, Seager, Traub and Woolf2002; Meadows & Seager Reference Meadows, Seager, Seager, Tucson and Tucson2010; Kaltenegger et al. Reference Kaltenegger, Segura and Mohanty2011, and references therein for detailed reviews). The presence or absence of these spectral features (detected collectively) will indicate similarities or differences with the atmospheres of terrestrial planets, and its astrobiological potential. Note that the presence of biogenic gases such as O2/O3 + CH4 may imply the presence of an active biosphere, but their absence does not imply the absence of life. Life existed on Earth before the interplay between oxygenic photosynthesis and carbon cycling produced an oxygen-rich atmosphere (see section 4).
Characterizing planetary environment
It is relatively straightforward to remotely ascertain Earth is a habitable planet, replete with oceans, a greenhouse atmosphere, global geochemical cycles, and life if one has data with arbitrarily high signal-to-noise and spatial and spectral resolution. The interpretation of observations of other planets with limited signal-to-noise ratio and spectral resolution as well as absolutely no spatial resolution, as envisioned for the first generation instruments, will be far more challenging and implies that we need to gather information on the planet environment to understand what we will see. After detection, we will focus on the main properties of the planetary system, its orbital elements as well as the presence of an atmosphere using the light curve of the planet (Selsis et al. Reference Selsis, Kasting, Levrard, Paillet, Ribas and Delfosse2003; Gaidos & Williams Reference Gaidos and Williams2004; Moskovitz et al. Reference Moskovitz, Gaidos and Williams2009; Wordsworth et al. Reference Wordsworth, Forget, Selsis, Millour, Charnay and Madeleine2011) or/and a crude estimate of the planetary nature using very low-resolution information (three or four channels) (see e.g. Traub et al. Reference Traub2003).
Knowing the temperature and planetary radius is crucial for the general understanding of the physical and chemical processes occurring on the planet (tectonics and hydrogen loss to space). Presently, radius measurements can only be performed when the planet transits in front of its parent star, by an accurate photometric technique. If the secondary eclipse of the transiting planet can be observed (when the planet passes behind the star) or the planet can be imaged, then the thermal emission of the planet can be measured, allowing the retrieval of mean brightness temperature T b, and the radius from the IR flux. One can also calculate the stellar energy of the star F star that is received at the measured orbital distance. The surface temperature of the planet at this distance depends on its albedo and on the greenhouse warming by atmospheric compounds. However, with a low-resolution spectrum of the thermal emission, the mean effective temperature and the radius of the planet can be obtained.
The accuracy of the radius and temperature determination will depend on the quality of the fit (and thus on the sensitivity and resolution of the spectrum), the precision of the Sun–star distance, the cloud coverage and also the distribution of brightness temperatures over the planetary surface. Assuming the effective temperature of our planet was radiated from the uppermost cloud deck at about 12 km would introduce about 2% error on the derived Earth radius. For transiting planets, for which the radius is known, within a certain measurement error, the measured IR flux can directly be converted into a brightness temperature that will provide information on the temperature of the atmospheric layers responsible for the emission.
If only the mass of non-transiting planets can be measured (by RV and/or astrometric observations), a first estimate of the radius can be made by assuming a bulk composition of the planet which can then be used to convert IR fluxes into temperatures. The ability to associate a surface temperature with the spectrum relies on the existence and identification of spectral windows probing the surface or the same atmospheric levels. Such identification is not trivial. For an Earth-like planet there are some atmospheric windows that can be used in most of the cases, especially between 8 and 11 μm as seen in Fig. 2. This window would, however, become opaque at high H2O partial pressure (e.g. the inner part of the HZ where a lot of water is vapourized) and at high CO2 pressure (e.g. a very young Earth or the outer part of the HZ).
The orbital flux variation in the IR can distinguish planets with and without an atmosphere in the detection phase (see also Selsis et al. Reference Selsis, Kasting, Levrard, Paillet, Ribas and Delfosse2003; Gaidos & Williams Reference Gaidos and Williams2004). Strong variation of the thermal flux with phase reveals a strong difference in temperature between the day and night hemisphere of the planet, and is a consequence of the absence of a dense atmosphere. In such a case, estimating the radius from the thermal emission is made difficult because most of the flux received comes from the small and hot substellar area. The ability to retrieve the radius would depend on the assumptions that can be made about the orbit geometry and the rotation rate of the planet. In most cases, degenerate solutions will exist. When the mean brightness temperature, T b, is stable along the orbit, the estimated radius is more reliable. The radius can be measured at different points of the orbit and thus for different values of T b, which should allow an estimate of the error.
The mean value of T b estimated over an orbit can be used to estimate the Bond albedo of the planet, its reflectivity, through the balance between the incoming stellar radiation and the outgoing IR emission. In the visible ranges, the reflected flux allows us to measure the product of Bond Alebdo times planetary area (a small but reflecting planet appears as bright as a big but dark planet).
The thermal light curve (i.e. the integrated IR emission measured at different position on the orbit) exhibits smaller variations due to the phase (whether the observer sees mainly the day side or the night side) and to the season on a planet with an atmosphere than the corresponding visible lightcurve variations. Therefore, planets with dense atmospheres can be distinguished from airless or Mars-like planets by the amplitude of the observed variations of T b (see Fig. 4).
Fig. 4. Orbital light curve for black body planets in a circular orbit with null obliquities, with and without an atmosphere in the visible (top) and thermal IR (bottom) (after Selsis et al. Reference Selsis, Kasting, Levrard, Paillet, Ribas and Delfosse2003).
As a second step a higher resolution spectrum will be used for interesting planetary targets to identify the compounds of the planetary atmosphere, constrain the temperature and radius of the observed exoplanet. In that context, we can test if we have an abiotic explanation of all compounds seen in the atmosphere of such a planet. If we do not, we can begin to consider the exciting biotic hypothesis.
The HZ
Different aspects of what determines the boundaries of the HZ have been discussed broadly in the literature. Habitability and the HZ are functions of the stellar flux at the planet's location as well as the planet's atmospheric composition. The latter determines the albedo and the greenhouse effect in the atmosphere. The inner and outer boundaries of the HZ differ for clear and cloudy conditions because the overall planetary albedo A, is a function of the chemical composition of the clear atmosphere as well as the fraction of clouds, A=A clear+A cloud.
The main differences among studies of the HZ are the imposed chemical composition and cloud fraction of the planet's atmosphere. Examples of atmospheres with different chemical compositions include the original CO2/H2O/N2 model with a water reservoir (e.g. Earth's), or model atmospheres with high H2/He concentrations (Pierrehumbert & Gaidos Reference Pierrehumbert and Gaidos2011) or limited water supply (Abe et al. Reference Abe, Abe-Ouchi, Sleep and Zahnle2011). Two concepts are commonly used throughout the literature for cloud-free (see Kasting et al. Reference Kasting, Whitmire and Reynolds1993) and cloudy atmospheres (Selsis et al. Reference Selsis, Kasting, Levrard, Paillet, Ribas and Delfosse2007). According to this model, the HZ is an annulus around a star where a rocky planet with a CO2/H2O/N2 atmosphere and sufficiently large water content (such as in Earth) can host liquid water permanently on a solid surface. This definition of the HZ assumes that the planet is rocky, water is present, the main atmospheric composition is CO2/H2O/N2, and the abundance of H2O and CO2 in the atmosphere is regulated by a geophysical cycle similar to that of Earth, resulting in an H2O- and CO2-dominated atmosphere on the inner and the outer edges of the HZ, respectively. Between those limits on a geological active planet, climate stability is provided by a feedback mechanism in which atmospheric CO2 concentration varies inversely with planetary surface temperature. In this definition, the locations of the two edges of the HZ are determined based on the equilibrium temperature of the planet (see e.g. Kasting et al. Reference Kasting, Whitmire and Reynolds1993; Selsis et al. Reference Selsis, Kasting, Levrard, Paillet, Ribas and Delfosse2003; Kaltenegger & Sasselov Reference Kaltenegger and Sasselov2011).
As shown by Selsis et al. (Reference Selsis, Kasting, Levrard, Paillet, Ribas and Delfosse2003) and from the cases studied in Kasting et al. (Reference Kasting, Whitmire and Reynolds1993), for a star with luminosity L and an effective temperature T Star between 3700 and 7200 K, the inner and outer boundaries of the HZ (l in, l out) can be obtained from
$$l_x (T_{{\rm Star}}, L_{{\rm Star}} ) = (l_{x{\rm Sum}} {\rm -} a_x T_S {\rm -} b_x T_S^2 )\left( {\displaystyle{{L_{{\rm Star}}} \over {L_{{\rm Sum}}}}} \right)^{1/2}. $$In this equation, l x (l in and l out) is in AU, a in = 2.7619 × 10−5, b in = 3.8095 × 10−9, a out = 1.3786 × 10−4, b out = 1.4286 × 10−9 and T S (K)=T Star(K) − 5700. As indicated by eq. (1), the extrapolated boundaries of the HZ are functions of the input parameter l x − Sun, the inner and outer limits of the Sun's HZ. As mentioned earlier, these limits are model dependent and different for different values of cloud fraction and atmosphere composition. In this study, we consider the values of l in–Sun and l out–Sun corresponding to the following scenarios.
For the inner boundary of the Sun's HZ (Fig. 1), we consider the water-loss limit (T surf = 373 K) of the HZ of our Solar System for 0 and 50% cloud fraction (Kasting et al. Reference Kasting, Whitmire and Reynolds1993; Selsis et al. Reference Selsis, Kasting, Levrard, Paillet, Ribas and Delfosse2003). In this scenario, the value of l in–Sun is equal to 0.95 and 0.76 AU for 0 and 50% cloud fractions (f), respectively. The outer boundary denotes the distance from the star where the maximum greenhouse effect fails to keep CO2 from condensing permanently, leading to runaway glaciation. For the outer boundary of the Sun's HZ, we consider the theoretical values of l out–Sun = 1.67 and 1.95 AU corresponding to a cloud-free and 50% cloud-fraction CO2 atmosphere. We choose those model values because they correspond to the empirical limits based on the initial Solar flux received at the position of Venus and Mars with l in–Sun = 0.75 AU and l out–Sun = 1.77 AU, respectively (see Kasting et al. Reference Kasting, Toon and Pollack1988, Reference Kasting, Whitmire and Reynolds1993 for details). Note that in the empirical Mars limit, the value of the outer boundary of the HZ is biased by Mars’ small size, which did not allow this planet to maintain a dense greenhouse atmosphere. At the limits of the HZ, the Bond albedo of a habitable planet is fully determined by its atmospheric composition and depends on the spectral distribution of the stellar irradiation.
In this definition, the two edges of the HZ (see Fig. 5), depend on the Bond albedo of the planet A, the luminosity of the star L star, the planet's semi-major axis D, as well as the eccentricity e, of the orbit, and in turn the average stellar irradiation at the planet's location. A more eccentric orbit increases the annually averaged irradiation proportional to (1−e 2)1/2 (Williams & Pollard Reference Williams and Pollard2002). However, the limits of the HZ are known qualitatively, more than quantitatively. This uncertainty is mainly due to the complex role of clouds and three-dimensional climatic effects not yet included in the modelling. Thus, planets slightly outside the computed HZ could still be habitable, while planets at habitable orbital distance may not be habitable because of their size or chemical composition. Subsurface life that could exist on planets with very different surface temperatures is not considered here, because of the lack of remotely detectable atmospheric to assert habitability.
Fig. 5. Extent of the HZ for water loss limit for 0 and 50% cloud coverage (inner limits) and 100% cloud coverage (outer limit dashed line) and position of potentially habitable Kepler planetary candidates in the HZ, individual HZ limits are indicated with crosses (Kaltenegger & Sasselov Reference Kaltenegger and Sasselov2011).
Applying this definition to the Kepler February 2011 data release, assuming circular orbits and albedo corresponding to 50% cloud coverage (consistent with the empirical ‘Venus’-limit of the HZ), leads to 27 Kepler planetary candidates in the HZ. Among those are three planetary candidates with radii smaller than two Earth radii (Kaltenegger & Sasselov Reference Kaltenegger and Sasselov2011). The potentially rocky Kepler planet candidates in multiple systems are especially interesting objects because their mass could be determined using transit time variations to calculate a mean density and potentially confirm high density and rocky characteristics.
Due to the large mean distance to the Kepler stars and planets, the characterization of the planets atmospheres will not be in the near future, although it is an excellent mission to derive statistics. Small planets detected around stars close to the Sun will provide the planetary targets that can be followed up with the next generation of ground- and space-based telescopes.
Characterization of planetary spectra
Influence of host-stars
The range of characteristics of planets is likely to exceed our experience with the planets and satellites in our own Solar System by far. Models of planets more massive than our Earth – rocky super-Earths – need to consider the changing atmosphere structure, as well as the interior structure of the planet (see e.g. Seager et al. Reference Seager, Kuchner, Hier-Majumder and Militzer2007; Sotin et al. Reference Sotin, Grasset and Mocquet2007; Valencia et al. Reference Valencia, Sasselov and O'Connell2007; Grasset et al. Reference Grasset, Schneider and Sotin2009). Also, Earth-like planets orbiting stars of different spectral type might evolve differently. Modelling these influences will help to optimize the design of the proposed instruments to search for Earth-like planets.
Using a numerical code that simulates the photochemistry of a wide range of planetary atmospheres several groups (Selsis Reference Selsis2000; Segura et al. Reference Segura, Krelove, Kasting, Sommerlatt, Meadows, Crisp, Cohen and Mlawer2003, Reference Segura, Kasting, Meadows, Cohen, Scalo, Crisp, Butler and Tinetti2005; Grenfell et al. Reference Grenfell, Stracke, von Paris, Patzer, Titz, Segura and Rauer2007; Pallé et al. Reference Pallé, Zapatero Osorio, Barrena, Montañés-Rodríguez and Martín2009) have simulated the atmospheric composition of a replica of our planet orbiting different types of star: F-type star (more massive and hotter than the Sun) and a K-type star (smaller and cooler than the Sun). The models assume similar background composition of the atmosphere as well as similar strength of biogenic sources.
For an Earth-analogue planet around a spectral grid of host stars ranging in 6250 K<T eff < 6500 K at a resolution λ/Δλ = 25 Fig. 6 shows the IR emergent modelled spectrum for F, G and K stars (Rugheimer et al. submitted) for a clear sky emergent spectra as well as a 60% global cloud coverage analogous to Earth (40% 1 km cloud layer, 40% 6 km cloud layer and 20% 12 km cloud layer). The spectra are presented relative to the planet's surface temperature black body. Clouds have high reflection in the visible. In the IR, the temperature difference between the emitting/absorbing layer and the surface determines the strength of the spectral feature. Since on Earth clouds emit at temperatures generally colder than the surface, they can increase as well as reduce the depth of features in the IR as can be seen in the cases of CO2 at 15 μm, O3 at 9.6 μm and CH4 feature at 7.7 μm.
Fig. 6. Clear sky model (left) and 60% cloud coverage (right) of the IR spectral region at a resolution of 25 (corresponding to proposed resolution of future space missions) for an Earth-analogue planet around FGK stars (Rugheimer et al. submitted).
A planet orbiting a K star has a thin O3 layer, compared with Earth's one, but still exhibits a deep O3 absorption: indeed, the low UV flux is absorbed at lower altitudes than on Earth which results in a less efficient warming (because of the higher heat capacity of the dense atmospheric layers). Therefore, the ozone layer is much colder than the surface and this temperature contrast produces a strong feature in the thermal emission. The process works the other way around in the case of an F-type host star. Here, the ozone layer is denser and warmer than the terrestrial one, exhibiting temperatures about as high as the surface temperature. Thus, the resulting low temperature contrast produces only a weak and barely detectable feature in the IR spectrum. This comparison shows that planets orbiting G (solar) and K-type stars may be better candidates for the search for the O3 signature than planets orbiting F-type stars (see Fig. 6). This result is promising since G- and K-type stars are much more numerous than F-type stars, the latter being rare and affected by a short lifetime (<1 Gyr). Due to the hot stratosphere for all F-type stars (T eff > 6000 K), the CO2 absorption feature at 15 μm has a prominent central emission peak visible. The central peak can be thought of as an indirect feature indicating a temperature inversion in the atmosphere (see also Selsis Reference Selsis2000). The O3 feature at 9.6 μm is increasingly difficult to resolve for hotter stellar types, despite increasing ozone abundance, due to the hotter stratosphere of the F-type stars (see also Selsis Reference Selsis2000). The CH4 feature at 7.7 μm, while visible for F- and G-type stars, is more prominent in the late K dwarfs (4500 and 4250 K) than in hotter stars due to its higher abundance from a lower UV environment.
Evolution of biomarkers over geological times on Earth
One crucial factor in interpreting planetary spectra is the point in the evolution of the atmosphere when its biomarkers and its habitability become detectable. The spectrum of the Earth has not been static throughout the past 4.5 Ga. This is due to the variations in the molecular abundances, the temperature structure and the surface morphology over time. At about 2.3 Ga oxygen and ozone became abundant, affecting the atmospheric absorption component of the spectrum. At about 0.44 Ga, an extensive land plant cover followed, generating the red chlorophyll edge in the reflection spectrum. The composition of the surface (especially in the visible), the atmospheric composition and temperature–pressure profile can all have a significant influence on the detectability of a signal. Figure 7 shows theoretical visible and mid-IR spectra of the Earth at six epochs during its geological evolution (Kaltenegger et al. Reference Kaltenegger, Traub and Jucks2007). The epochs are chosen to represent major developmental stages of the Earth, and life on Earth. If an extrasolar planet is found with a corresponding spectrum, we can use the stages of evolution of our planet to characterizing it, in terms of habitability and the degree to which it shows signs of life. Furthermore, we can learn about the evolution of our own planet's atmosphere and possibly the emergence of life by observing exoplanets in different stages of their evolution. Earth's atmosphere has experienced dramatic evolution over 4.5 billion years, and other planets may exhibit similar or greater evolution, and at different rates. Figure 7 shows epochs that reflect significant changes in the chemical composition of Earth's atmosphere. The oxygen and ozone absorption features could have been used to indicate the presence of biological activity on Earth anytime during the past 50% of the age of the solar system. Different signatures in the atmosphere are clearly visible over Earth's evolution and observable with low resolution. The spectral resolution required for optimal detection of habitability and biosignatures has to be able to detect those features on our own planet for the dataset we have over its evolution.
Fig. 7. The visible to NIR (top) and mid-IR (bottom) spectral features on an Earth-like planet change considerably over its evolution from a CO2-rich (epoch 0) to a CO2/CH4-rich atmosphere (epoch 3) to a present-day atmosphere (epoch 5). The bold lines show spectral resolution of 80 and 25 comparable with the proposed visible TPF and Darwin/proposed future space missions (TPF-I) mission concept, respectively (Kaltenegger et al. Reference Kaltenegger, Traub and Jucks2007).
Spectra of the Earth exploring temperature sensitivity (hot house and cold scenario) and different singled out stages of its evolution (e.g. Pavlov et al. Reference Pavlov, Kasting, Brown, Rages, Freedman and Greenhouse2000; Schindler & Kasting Reference Schindler and Kasting2000; Traub & Jucks Reference Traub, Jucks, Mendillo, Nagy and Waite2006) produce a variety of spectral fingerprints for our own planet. Strong volcanism (Kaltenegger et al. Reference Kaltenegger, Segura and Mohanty2011), as well as other geochemical cycles could also be detected in a planet's spectrum (Domagal-Goldman et al. Reference Domagal-Goldman, Meadows, Claire and Kasting2011; Kaltenegger & Sasselov Reference Kaltenegger and Sasselov2011).
Those spectra will be used as part of a grid to characterize any exoplanets found and influences the design requirements for a spectrometer to detect habitable planets.
Surface and red-edge features
Although they efficiently absorb the visible light, photosynthetic plants have developed strong IR reflection (possibly as a defence against overheating and chlorophyll degradation) resulting in a steep change in reflectivity around 700 nm, called the red-edge. The primary molecules that absorb the energy and convert it to drive photosynthesis (H2O and CO2 into sugars and O2) are chlorophyll A (0.450 μm) and B (0.680 μm). The exact wavelength and strength of the spectroscopic ‘vegetation red-edge’ (VRE) depends on the plant species and environment. On Earth around 440 million years ago (Schopf Reference Schopf1993; Pavlov et al. Reference Pavlov, Hurtgen, Kasting and Arthur2003), an extensive land plant cover developed, generating the red chlorophyll edge in the reflection spectrum between 700 and 750 nm. Averaged over a spatially unresolved hemisphere of Earth, the additional reflectivity of this spectral feature is typically a few percent (see also Arnold et al. Reference Arnold, Gillet, Lardiere, Riaud and Schneider2002; Montanes-Rodriguez et al. Reference Montañés-Rodriguez, Pallé, Goode, Hickey and Koonin2005; Cowan et al. Reference Cowan2009). Several groups (Woolf et al. Reference Woolf, Smith, Traub and Jucks2002; Turnbull et al. Reference Turnbull, Traub, Jucks, Woolf, Meyer, Gorlova, Skrutskie and Wilson2006; Montanes-Rodriguez et al. Reference Montanes-Rodriguez, Pallé and Goode2007) have measured the integrated Earth spectrum via the technique of Earthshine, using sunlight reflected from the non-illuminated, or ‘dark’, side of the moon. Earthshine measurements have shown that detection of Earth's VRE is feasible if the resolution is high and the cloud coverage is known, but is made difficult owing to its broad, essentially featureless spectrum and cloud coverage. Space-based EPOXI measurements have shown similar results (Cowan et al. Reference Cowan2009; Livengood et al. Reference Livengood2011).
Our knowledge of the reflectivity of different surface components on Earth – such as deserts, ocean and ice – helps in assigning the VRE of the Earthshine spectrum to terrestrial vegetation. Earth's hemispherical-integrated VRE signature is very weak, but planets with different rotation rates, obliquities, land–ocean fraction and continental arrangement may have lower cloud-cover and higher vegetated fraction (see e.g. Seager & Ford Reference Seager and Ford2002). Knowing that other pigments exist on Earth and that some minerals can exhibit a similar spectral shape around 750 nm (Seager et al. Reference Seager, Turner, Schafer and Ford2005), the detection of the red-edge of the chlorophyll on exoplanets, despite its interest, will not be unambiguous.
Picking the most different reflecting surfaces (snow with a high albedo and sea with an extremely low albedo) shows the maximum effect surface coverage could have on the amount of light reflected from an exoplanet – assuming the whole planet surface is covered with that one material, the surface area is the same, and also artificially assuming similar cloud coverage and atmospheres for comparison. If similar photosynthesis would evolve on a planet around other stellar types, the possible different types of a vegetation spectral signature have been modelled (Kiang et al. Reference Kiang, Siefert and Blankenship2007). Those signatures will be difficult to verify through remote observations as being of biological origin.
Clouds reduce the relative depths, full widths and equivalent widths of spectral features, weakening the spectral lines in the visible (Kaltenegger et al. Reference Kaltenegger, Traub and Jucks2007). In the thermal IR, clouds emit at temperatures that are generally colder than the surface and can both enhance or reduce the depth of the absorption features, while in the visible the clouds themselves have different spectrally dependent albedo that further influence the overall shape of the spectrum.
If one could record the planet's signal with a very high time resolution (a fraction of the rotation period of the planet) and SNR, one could determine the overall contribution of clouds to the signal (Pallé et al. Reference Pallé, Ford Eric, Seager, Montañés-Rodríguez and Vazquez2008). During each of these individual measurements, one has to collect enough photons for a high individual SNR per measurement to be able to correlate the measurements to the surface features, what precludes this method for first-generation missions that will observe a minimum of several hours to achieve an SNR of 5–10. For Earth (Cowan et al. Reference Cowan2009; Pallé et al. Reference Pallé, Ford Eric, Seager, Montañés-Rodríguez and Vazquez2008; Livengood et al. Reference Livengood2011) such high SNR measurements show a correlation to Earth's surface feature because the individual measurements are time resolved as well as have an individual high SNR, making it a very interesting concept for future generations of missions.
Summary
Spectroscopy of the atmosphere of extrasolar planets allows us to remotely explore a planets environment, to distinguish Mini-Neptunes from rocky Super-Earths, and to explore atmospheric compositions as well as search for indications for habitability. Any information we collect on habitability is only important in a context that allows us to interpret, what we find. To search for signs of life we need to understand how the observed atmosphere physically and chemically works. Knowledge of the temperature and planetary radius is crucial for the general understanding of the physical and chemical processes occurring on the planet. These parameters as well as an indication of habitability can be determined with low resolution spectroscopy and low photon flux, as assumed for first generation space missions. The combination of spectral information in the visible (starlight reflected of the planet) as well as in the mid-IR (planet's thermal emission) allows a confirmation of atmospheric species, a more detailed characterization of individual planets, but also to explore a wide domain of planet diversity. Being able to measure the outgoing shortwave and longwave radiation, as well as their variations along the orbit, and to determine the albedo and identify greenhouse gases would, in combination, allow us to explore the climate system at work on other worlds. Ultimately future missions will allow us probe planets similar to our own for habitable conditions.
Acknowledgements
The authors acknowledge support from DFG funding ENP Ka 3142/1-1 and NASA NAI.
