The afternoon session of the workshop moved beyond the search for life in the solar system and focused instead on the search for life on distant exoplanets. This session was co-chaired by David Des Marais, NASA Ames Research Center, and Dimitar Sasselov, Harvard-Smithsonian Center for Astrophysics.
Victoria Meadows of the University of Washington prefaced her talk on extrasolar biosignatures by explaining that she would essentially summarize the 2016 workshop hosted by the Nexus for Exoplanet System Science (NExSS) and the NASA Astrobiology Institute, called the “Exoplanet Biosignatures Workshop Without Walls.” This workshop combined the expertise of NExSS, the NASA Astrobiology Institute, and the science and technology definition teams for exoplanet observation mission concepts to focus on the following three main questions:
- What are the known remotely observable biosignatures, the processes that produce them, and their known non-biological sources?
- How can we develop a more comprehensive framework for identifying additional biosignatures and their possible abiotic mimics?
- What standards can we agree to use for assessing biosignature observations, both known biosignatures and those we have yet to identify?
That workshop produced five coordinated papers on topics covered by the workshop: a biosignatures review, work on using O2 as a biosignature, developing a more general framework for observing and interpreting biosignatures, novel types of biosignatures, and a synthesis paper to guide future research on topics such as modeling and mission development.
Exoplanet Biosignature Review
The major question in exoplanet biosignatures, said Meadows, is how to detect life at great distances. In this case, life must have a global impact to be observable. Identifying biosignatures requires three things: reliability
that the signature is indeed biological, survivability of the potential biosignature, and the detectability of the possible signature. Meadows said that an alternative way to search for life would be to look for a disequilibrium or some sort of unexpected planetary process that cannot be explained by abiotic processes.
Meadows stated that typical biosignatures are atmospheric gases, such as oxygen in the presence of methane.1,2 However, she wanted to push the boundaries of what we know by exploring other types of gaseous biosignatures in different contexts and environments. There are also surface biosignatures, such as the “red edge,” which is due to the phenomenon that Earth’s plants are highly reflective in the near-infrared (see Figure 3.1).3 Other types of “edges” may be possible with different pigments, which may or may not be related to photosynthesis (e.g., UV protection).4,5,6 Temporal biosignatures are also possible, such as daily or seasonal changes.7 An example is the seasonal change in abundance of CO2 in Earth’s atmosphere.8 A large disequilibrium could also indicate signs of life. The classic example is Earth’s high abundance of both O2 and CH4. Since methane’s lifetime in the atmosphere is just 10 years,9 methane’s high abundance in the presence of O2 indicates an active source of the gas, and in the case of Earth, that is due to life (see Table 3.1). Meadows then cited some recent work that showed that the largest Gibbs energy disequilibrium on Earth is the fact that Earth has both N2 and O2 with an ocean. Without life, this would end up as nitrate dissolved in the ocean.10
She then defined three terms useful in thinking about biosignatures. First, an “antibiosignature” is an aspect of the planetary environment that suggests that life is not present, such as abundant CO on Mars, which would be an attractive energy source for life if it were there.11 A “false positive” is an abiotic source for a potential biosignature, such as O2 being produced by photolysis of H2O or CO2.12-17 A “false negative” is when processes on the planet work to reduce the detectability of a biosignature, such as oxidation on a planet’s surface.18,19
1 D.R. Hitchcock and J.E. Lovelock, 1967, Life detection by atmospheric analysis, Icarus 7:149.
2 V. Meadows, 2017, Reflections on O2 as a biosignature in exoplanetary atmospheres, Astrobiology, accepted.
3 D.M. Gates, H.J. Keegan, J.C. Schleter, and V.R. Weidner, 1965, Spectral properties of plants, Applied Optics 4:11.
4 E.W. Schwieterman, C.S. Cockell, and V.S. Meadows, 2015, Nonphotosynthetic pigments as potential biosignatures, Astrobiology 15:341.
5 S. Hegde, I.G. Paulino-Lima, R. Kent, L. Kaltenegger, and L. Rothschild, 2015, Surface biosignatures of exo-Earths: Remote detection of extraterrestrial life, Proceedings of the National Academy of Sciences of the U.S.A. 112:3886.
6 N.Y. Kiang, A. Segura, G. Tinetti, Govindjee, R.E. Blankenship, M. Cohen, J. Siefert, D. Crisp, and V.S. Meadows, 2007, Spectral signatures of photosynthesis. II. Coevolution with other stars and the atmosphere on extrasolar worlds, Astrobiology 7:252.
7 V.S Meadows, 2005, Modelling the diversity of extrasolar terrestrial planets, Proceedings of the International Astronomical Union 1:25.
8 C.D. Keeling, R.B. Bacastow, A.E. Bain-Bridge, C.A. Ekdahl, Jr., P.R. Guenther, LS. Waterman, and J.F.S. Chin, 1976, Atmospheric carbon dioxide variations at Mauna Loa Observatory, Hawaii, Tellus 28:538.
9 J.T. Houghton, L.G. Meira Filho, J. Bruce, H. Lee, B.A. Callander, E. Haites, N. Harris, and K. Maskell, eds., 1994, Climate Change 1994: Radiative Forcing of Climate Change and An Evaluation of the IPCC IS92 Emission Scenarios, Cambridge University Press, Cambridge, U.K.
10 J. Krissansen-Totton, D.S. Bergsman, and D.C. Catling, 2015, On detecting biospheres from chemical thermodynamic disequilibrium in planetary atmospheres, Astrobiology 16:39.
11 K. Zahnle, R.S. Freedman, and D.C Catling, 2011, Is there methane on Mars?, Icarus 212:493.
12 R. Luger and R. Barnes, 2015, Extreme water loss and abiotic O2 buildup on planets throughout the habitable zones of M dwarfs, Astrobiology 15:119.
13 F. Tian, 2015, History of water loss and atmospheric O2 buildup on rocky exoplanets near M dwarfs, Earth and Planetary Science Letters 432:126.
14 R. Wordsworth and R. Pierrehumbert, 2014, Abiotic oxygen-dominated atmospheres on terrestrial habitable zone planets, The Astrophysical Journal Letters 785:20.
15 P. Gao, R. Hu, T.D. Robinson, C. Li, and Y.L. Yung, 2015, Stabilization of CO2 atmospheres on desiccated M dwarf exoplanets, The Astrophysical Journal 806:249.
16 F. Tian, K. France, J.L. Linsky, P.J.D. Mauas, and M.C. Vieytes, 2014, High stellar FUV/NUV ratio and oxygen contents in the atmospheres of potentially habitable planets, Earth and Planetary Science Letters 385:22.
17 C.E. Harman, E.W. Schwieterman, J.C. Schottelkotte, and J.F. Kasting, 2015, Abiotic O2 levels on planets around F, G, K, and M stars: Possible false positives for life?, The Astrophysical Journal 812:137.
18 A.D. Anbar, Y. Duan, T.W. Lyons, G.L. Arnold, B. Kendall, R.A. Creaser, A.J. Kaufman, G.W. Gordon, C. Scott, J. Garvin, and R. Buick, 2007, A whiff of oxygen before the great oxidation event?, Science 317:1903.
19 C.T. Reinhard, S.L. Olson, E.W. Schwieterman, and T.W. Lyons, 2017, False negatives for remote life detection on ocean-bearing planets: Lessons from the early Earth, arXiv:1702.01137.
TABLE 3.1 Constituents of the Earth’s Atmosphere (Volume Mixing Ratios)
|Molecule||Standard Abundance (Ground-Truth Earth)||Galileo Valuea||Thermodynamic Equilibrum Value Estimate 1b||Thermodynamic Equilibrum Value Estimate 2c|
|H2O||0.001 to 0.03||0.001 to 0.01||0.001 to 0.03|
|Ar||9 × 10−3||9 × 10−3|
|CO2||3.5 × 10−4||5±2.5×10-4||3.5 × 10−4|
|CH4||1.6 × 10−6||3±1.5×10-6||<10−35||10−145|
|N2O||3 × 10−7||~10-6||2 × 10−20||2 × 10−19|
|O3||10−8 to 10−7||>10-8||6 × 10−32||3 × 10−30|
a Galileo values for O2, CH4, and N2O from Near-Infrared Mapping Spectrometer (NIMS) data; O3 estimate from Ultraviolet Spectrometer (UVS) data.
b At P = 1 bar, T = 280 K (see E.R. Lippincot, R.V. Eck, M.O. Dayhoff, and C. Sagan, 1967, Thermodynamic equilibria in planetary atmospheres, The Astrophysical Journal 147:753).
c At P = 1 bar, T = 290 K (see W.L. Chameides and D.D. Davis, 1992, Chemistry in the troposphere, Chemical and Engineering News 60:38).
d The observed value; it is in thermodynamic equilibrium only if the under-oxidized state of the Earth’s crust is neglected.
SOURCE: Reprinted by permission from Macmillan Publishers Ltd.: C. Sagan, W.R. Thompson, R. Carlson, D. Gurnett, and C. Hord, 1993, A search for life on Earth from the Galileo spacecraft, Nature 365:715-721, copyright 1993.
Meadows then put the idea of a biosignature in some historical context. Previously, O2 alone was considered a robust biosignature, as there was no known abiotic source that would produce it in high abundance on Earth. However, she said that it now isn’t as simple as that. She said that oxygen is still considered an excellent biosignature because it is produced via photosynthesis; its substrates (water and carbon dioxide) are likely abundant on habitable exoplanets; and O2 is potentially very detectable because it is evenly mixed throughout the atmosphere and is spectrally active at visible and near-infrared wavelengths. However, recent work has found several abiotic methods that can create a large O2 atmosphere, often involving photolysis of H2O or CO2.20, 21,22 Several of these methods occur on M-dwarf planets, which some consider to be particularly attractive to searches for life.
However, she said, false positives have signatures themselves (see Figure 3.2).23 For a planet that obtained an atmosphere rich in oxygen by boiling off its oceans while orbiting a young M-dwarf star,24 the oxygen atmosphere would become so pressurized and dense that a detectable amount of O4 would form in significant quantities.25 In a separate example, Meadows said that, if photolysis of CO2 were the source of O2 in the atmosphere, then large amounts of CO would also be apparent in the planet’s spectrum.26 In fact, these false positive indicators are often more observable than the biosignature itself, such as the O4 in the former scenario.27
Framework for Biosignature Assessment
Using the lessons learned from oxygen, Meadows said that the workshop aimed to develop a general framework for assessing biosignatures. The first step in this process is to characterize the important parameters in the planet’s host star and its entire planetary system. Afterwards, the planet’s properties must be characterized, and a search for biosignatures can be conducted. If any are found, potential false-positive scenarios must be further scrutinized.
In finding good biosignatures to choose from, Meadows listed three potential starting points. The easiest method would be to simply identify Earth’s current biosignatures.28 A disadvantage of this is that this limits you to the biosignatures of just one planet. Another method would be to explore Earth’s past.29 This expands the types of biosignatures one can search for, but knowledge of Earth’s past environments and biosignatures is not fully developed. The most general method would be to explore a large number of potential volatiles that may be biosignatures.30 However, without an example of a planet with life to analyze in context, this makes the risk of finding a false positive higher.
Giving a preview of the material in the workshop report, she showed a figure demonstrating that having liquid water on the surface is a function of the star, the properties of the planetary system, and the properties of the planet itself. She then listed four processes that could mimic false positives: geological/chemical (e.g., volcanism and serpentinization), mineralogical (e.g., surface reflectivity), photochemical (e.g., photolytic O2 and seasonal changes in gases), and atmospheric evolution (O2 produced from water loss). Ruling out these false positives could require additional observations beyond just the detection of a biosignature.
20 Luger and Barnes, 2015.
21 Gao et al., 2015.
22 Harman et al., 2015.
23 V. Meadows, 2017, Reflections on O2 as a biosignature in exoplanetary atmospheres, Astrobiology, accepted.
24 Luger and Barnes, 2015.
25 E.W. Schwieterman, V.S. Meadows, S.D. Domagal-Goldman, D. Deming, G.N. Arney, R. Luger, C.E. Harman, A. Misra, and R. Barnes, 2016, Identifying planetary biosignature impostors: Spectral features of CO and O4 resulting from abiotic O2/O3 production, The Astrophysical Journal Letters 819:13.
26 Gao et al., 2015.
27 Schwieterman et al., 2016.
28 J.E. Lovelock, 1975, Thermodynamics and the recognition of alien biospheres, Proceedings of the Royal Society of London. Series B, Biological Sciences 189:167.
29 G. Arney, S.D. Domagal-Goldman, V.S. Meadows, E.T. Wolf, E. Schwieterman, B. Charnay, M. Claire, E. Hébrard, and M.G. Trainer, 2016, The pale orange dot: The spectrum and habitability of hazy Archean Earth, Astrobiology 16:873.
30 S. Seager and W. Bains, 2015, The search for signs of life on exoplanets at the interface of chemistry and planetary science, Science Advances 1:e1500047.
Meadows then briefly mentioned the difficulty of determining the confidence of any detected, potential biosignature and the possibility of novel biosignatures. She then finished with the following list of questions to guide the field forward:
- How do we discover biosignatures with high detection significance?
- How do we know that we’re properly interpreting these as biosignatures in the right environmental context?
- Do we have the instrumental capability today or in planned missions to detect and identify biosignatures and their environments in order to put the results in context?
An audience member asked Meadows whether Earth’s biosphere would be detectable before the rise of oxygen. Meadows said that any biosignatures present there would be harder to observe and to properly interpret. However, there could have been a disequilibrium present in the Archean Earth that might have been readily detectable. Pigments might also have existed that did not require photosynthesis to occur, such as for UV protection. However, she then emphasized again that it would be harder to detect life on an Archean Earth.
Responding to that, another participant then cautioned against the idea of just looking for disequilibrium gases. He gave an example of a gas like CO that could build up abiotically in some exoplanet atmospheres. He then stated that, on an Archean Earth, you might have methanogens that would actually drive the system towards equilibrium by metabolizing carbon dioxide and hydrogen into methane. Meadows agreed, but stressed the importance of the abundances of such potential substances. She said that serpentinization would produce maybe 5 ppm of methane naturally,31 while biotic sources of life could produce enough methane to make up multiple percent of the atmosphere’s composition.
Another audience member then said that Earth’s history could have had multiple enrichment periods as new types of metabolism evolved into being, particularly in the first couple billions of years before animal predation. For example, methanogenesis could have led to hydrogen-based photosynthesis, then H2S-based photosynthesis, and then iron-based photosynthesis, each of which could have greatly enriched the atmosphere with certain biosignatures. Meadows agreed that this was an interesting point and emphasized again the abundance argument, the idea that there would be so much of a substance that it would immediately trigger our interest.
The idea of long-term trajectories was then raised by one member of the audience. For example, planets that develop plate tectonics might all evolve in similar ways. Deviations from this evolution could then be a potential biosignature. Meadows answered that Norm Sleep of Stanford University has looked at the way that life can affect the geology of a planet.32 Noting the possibility that she might not be remembering it correctly, she said it was about how life could affect mineral assemblages and the behavior of rock in certain situations. However, she thought that geological evolution might be so variable that general statements about a planet’s geological state may not be possible. Touching on a couple of asides, she then brought up the great difficulty of dating most stars, which often have uncertainty ranges on the order of billions of years. This then makes it difficult to date the age of the planets in the system. Then she said that habitability is not just a function of location, but of time. A planet that is habitable now may not have been in the past and/or may not be habitable in the future and vice versa.
Another workshop participant then added that Dr. Sleep also talked about a biological carbon pump that could send reductants to the sea floor and therefore affect the redox state of mineral assemblages. He then continued along the path of long-term, geological trajectories, saying that, to him, the habitable zone is the region where feedbacks push the planet toward having liquid water as opposed to away from having liquid water. However, he said, once life begins, it then becomes part of the feedback loops. Additionally, he stated that we need to build up a complete model of a lifeless planet and all its processes and understand how it works in order to obtain a null model. Figuring out how to do this using Earth is difficult, as life is nearly everywhere. Meadows agreed with that, stating
31 A. Guzmán-Marmolejo, A Segura, and E. Escobar-Briones, 2013, Abiotic production of methane in terrestrial planets, Astrobiology 13:550.
32 N.H. Sleep, D.K. Bird, and E. Pope, 2012, Paleontology of Earth’s mantle, Annual Review of Earth and Planetary Sciences 40:277.
that “life is a planetary process.” If life exists, you need to take it into account to get an explanation of the global system. She then stressed that looking at biosignatures must include understanding its environment and context.
An audience member then cautioned against the idea of an antibiosignature as it pertains to carbon monoxide. Abiotic sources, in principle, could create more CO than life could use, and therefore, a detection of CO would not necessarily mean that there was no life. He then cautioned against thinking we know what an abiotic planet might look like, since we cannot explain the amount of water on Earth or know how much water a typical habitable zone rocky planet might contain.
The final question directed toward Dr. Meadows regarded the technological capabilities of future missions being able to resolve the surface of an exoplanet and measure the composition of its optically thin atmosphere and whether it would ever be possible in the near or more distant future. Meadows answered that the Habitable Exoplanet Imaging Mission (HabEx) or the Large UV/Optical/Infrared Surveyor (LUVOIR) should be able to do it, presumably in about 20 years. She said that the Wide Field Infrared Survey Telescope (WFIRST) (expected launch date in the mid-2020s) might be able to do it too, if it is incredibly lucky. Transmission spectroscopy, where the atmosphere is viewed as the planet transits its star, would not be able to see a planetary surface, but direct imaging could see down to the surface, even if the atmosphere is partially cloudy. In fact, she said, if one could take images of a planet as it rotates, one could create a longitudinal map of the surface and potentially measure surface composition. However, disentangling the atmospheric signature from the surface signature would still be necessary.
William Bains of the Massachusetts Institute of Technology continued on the subject of searching for biosignatures in exoplanets, but thought about it in a broader context of what life could produce. In “thinking outside the box,” he said that what one really needs to do is to think in a much larger box. He chose this box to be “chemistry.” Biosignatures visible over large distances would likely be volatiles in the atmosphere or colors on the surface,33 so there needs to be an understanding on why life could make volatiles or colors using only the laws of chemistry as starting assumptions. He then gave the following list of three types of biosignatures gases:34
- Type I: A byproduct of energy capture (e.g., H2S + 3/2O2 → SO2 +H2O))
- Type II: A byproduct of biomass capture (e.g., CO2 + H2O + hν → [CH2O] + O2)
- Type III: No chemical “reason” at all (e.g., C6H14N4O2 + 2O2 + 3[H] → NO + C6H13N3O3 + 2H2O)
Bains said that there is a common expectation that life on other planets would use redox disequilibria to capture energy, the waste product(s) of which is (are) termed Type I biosignatures. How this plays out depends on the atmospheric composition of the planet. One can make predictions on what life might do in an atmosphere dissimilar to Earth’s atmosphere of N2 and O2, such as an atmosphere dominated by CO2, H2, or N2. On such a planet, life would presumably react crustal rocks with the atmosphere, reducing them in the case of an H2-rich atmosphere. Some of these chemical reactions could produce energy that life could extract. An observable biosignature needs to be one of these energy-extracting chemical reactions that produces a volatile.
In an H2 atmosphere, the only such volatiles are likely to be CH4, H2S, H2O, and NH3.35 The first three would be expected in an H2 atmosphere anyway, but Bains said that ammonia could be a good biosignature in an H2 atmosphere. Ammonia would need to be produced at a rate that, at the very minimum, would maintain a detectable amount of it in the atmosphere despite ammonia removal through atmospheric photochemistry. This requires a certain level of biomass. A calculation of the minimum biomass (under the most favorable conditions) needed to
33 S. Seager and W. Bains, 2015, The search for signs of life on exoplanets at the interface of chemistry and planetary science, Science Advances 1:e1500047.
34 S. Seager, M. Schrenk, and W. Bains, 2012, An astrophysical view of Earth-based metabolic biosignature gases, Astrobiology 12:61.
35 S. Seager, W. Bains, and R. Hu, 2013, Biosignature gases in H2-dominated atmospheres on rocky exoplanets, The Astrophysical Journal 777:95.
TABLE 3.2 Biomass Needed to Maintain Detectable Levels of Biosignature Gases in an Atmosphere of a Habitable Zone Planet with an H2-Dominated Atmosphere at P = 1 bar
|Compound||Biosignature Gas Type||Thermal Emission (gm/cm2)||Transmission (gm/cm2)|
|Sun-like||Active M Dwarf||Quiet M Dwarf||Active M Dwarf||Quiet M Dwarf|
|NH3||Type I||4.0 × 10−4||8.0 × 10−6||9.5 × 10−6||1.1||1.8 × 10−9|
|CS2||Type III||5.5 × 107||2.3 × 107||37||1.5 × 107||24|
|OCS||Type III||1.3 × 105||5,500||0.67||9.9 × 104||12|
maintain a detectable level of ammonia shows that is possible (see Table 3.2). Bains noted that the apparently barren Sechura Desert in Peru has 10 times the biomass necessary for thermal detection on a planet with an H2 atmosphere orbiting a Sun-like star and orders of magnitude more for an M-dwarf star, if that life used ammonia production as a primary source of energy. However, careful consideration of false positives and false negatives is necessary.
Type II gases are produced by biomass capture. On Earth, life needs to grab carbon from CO2 and throw away the oxygen. On a planet with an H2-dominated atmosphere, it would need to take carbon from CH4, the most likely dominant carbonaceous gas, and throw out the hydrogen.36 Bains ran through the likely Type II chemical reactions and came to the conclusion that the most plausible path to get a biosignature is a planet using methane, water, and energy to produce biomass and H2 gas, a result he called “incredibly disappointing” since the planet already has a hydrogen-dominated atmosphere from abiotic sources. This reaction can use near-infrared photons to power it, meaning that there likely would be not a red edge on this planet either. On the other hand, Bains said, false positives are less of a problem because they are thermodynamically implausible. A false negative could occur if life had not yet evolved to that stage though.
Going more in-depth on the issue of color, Bains noted Earth’s red edge, the fact that Earth’s plants are very reflective in the near-infrared. However, this is not inevitable. He pointed out that begonia leaves (not petals) are blue. He then showed a figure of a spectral analysis of many different types of things, in which the rocks and living materials were distinct enough that it could be used to classify a spectrum as coming from something living or non-living. A problem with this is that it does not take into account that, in real environments, rocks would be mixed in with plants, and deconvoluting them could be difficult.
Bains then asked what color alien life would be. Photosynthesis merely requires plants to absorb somewhere in the star’s spectral range, but that is a weak constraint. Beyond that, he thought that we don’t really have an idea what color life would be. He then went through a few examples. UV protection might be generally favorable, but on Earth, melanin is black and looks like rock. Pigments to capture photons evolved at least four independent times (chlorophyll, bacteriorhodopsins, aphid carotenes, and melanized fungi), each pigment having its own absorption characteristics. He then noted that, out of these four cyclic catalytic pathways to capture CO2, only one has been exhibited in biomass, and Bains didn’t think that we knew why.
Moving on to Type III, Bains performed the same type of biomass calculation as before (see Table 3.2).37 Many of them would require huge amounts of biomass to maintain an atmosphere with a detectable level of a Type III biosignature (especially for Sun-like stars), up to an equivalent of “a column of cabbages hundreds of meters high”
36 W. Bains, S. Seager, and A. Zsom, 2014, Photosynthesis in hydrogen-dominated atmospheres, Life 4:716.
37 Seager et al., 2013.
for thermally detected CS2 around a Sun-like star. All of this was based on gas production rates found in Earth life, however, where there is little understanding of why life chose particular secondary metabolite products. For example, there are 34 halomethanes possible, but life is only known to produce 22 of them, and nobody knows why. Another example is that life does not use fluorine well. He said that, while it is very electronegative, so is oxygen, which life uses aplenty.
In order to approach all of this systematically, Bains noted that Sara Seager suggested building up a catalog of all small molecules possible and then working backwards to filter for stability, volatility, and so on.38,39 This even includes things not produced by life directly, but molecules produced by industry too. Two of the top-level filters they are applying to these molecules are whether they are detectable and whether they can be produced geochemically. Entropy of formation is also important to look at because bigger molecules are less likely to form spontaneously. This catalog is intended to build up a “geological plausibility index” to determine how likely it is that a molecule might be produced by geology and, inversely, how likely it might be produced by life.
An audience member asked about whether kinetic arguments, rather than purely thermodynamic arguments, can guide interpretation. He cited N2 as an example. Treating an N2 atmosphere only as an energetic sink means that life would be unlikely to exist on such a planet. However, since N2 is a kinetic sink, this suggests that a detection of N2 does not preclude life. Referring back to the previous talk by Meadows, Bains said that nitrogen is very robust. While nitrogen and oxygen will eventually get turned into nitrate in the ocean, he said that it will then be broken back down into nitrogen in hydrothermal systems. Therefore, he said, finding a thermodynamic disequilibrium is not really useful without a better understanding of the system’s kinetics.
Agreeing with Bains, another member of the audience disputed the idea that, if there is a disequilibrium, it means that there is no life because otherwise life would have used it to make more life. He said that this is a visible fact just by looking at O2 and N2 in the atmosphere or seeing a forest outside in an O2 atmosphere. This is not a failure of life, he said, but a failure of Darwinism. However, after life becomes intelligent, it tries to remove Darwinism in favor of Lamarckism. He then asked what the consequences would be of an intelligent species exploiting Lamarckism. Bains agreed to the first part, saying that a thermodynamic disequilibrium is a red herring. Bains commented that evolution is inefficient at finding the optimal solution, and as the audience member’s comment said, it may leave thermodynamic disequilibria unexploited. He also commented that evolution was a poor biomarker, one reason being that you cannot observe evolution of life on other planets.
A workshop participant then brought up the topic of geochemical false positives, saying that one must take into account time. He gave hydrogen as an example. An average-sized planet, he said, would not be able to produce H2 effectively early on, but after about a billion years, it could then do so. He said that one must look at the planet’s age and environment. Bains agreed.
Going back to Bains’ point about ammonia being a good biosignature in an H2-dominated atmosphere, an audience member suggested that ammonia in the atmosphere could be produced by a comet impact before the observation and asked how to get around these kinds of special events. Bains answered that it requires knowing about the temporal context: how old the world is and how it’s changing. This could show that an observation is not a one-off event.
On a similar note, another audience member said that an old paper discovered the reduction of nitrogen to ammonia on desert sands using titanium dioxide as a catalyst, which showed that there is an abiotic way to create ammonia.40 Bains noted that the audience member had brought this point to Bains’ attention before. He then stressed the abundance point from the previous talk, noting that a whole lot of titanium dioxide would be necessary for that scenario.
38 Seager and Bains, 2015.
39 S. Seager, W. Bains, and J.J. Petkowski, 2016, Toward a list of molecules as potential biosignature gases for the search for life on exoplanets and applications to terrestrial biochemistry, Astrobiology 16:465.
40 G.N. Schrauzer and T.D. Guth, 1977, Photolysis of water and photoreduction of nitrogen on titanium dioxide. Journal of the American Chemical Society 99:7189.
A workshop participant then asked Bains what his practical suggestion was for moving forward. Bains answered that they are trying to categorize all the small molecules to try to rule out things that are likely to have a geological origin. He went on to say that there are major gaps in our understanding in reaction chemistry, such as how stable molecules are in water. The same goes for atmospheric photochemistry. Bains wants a huge database of how different molecules react in different environments under different conditions.
Then an audience member asked for Bains’s opinion on the limits of life using different energy sources. He said that, for example, no life extracts mechanical energy. The audience member then said that life is lazy and doesn’t want to do anything if there is an available gradient to use instead. Bains replied that there are some energy sources that are just too diffuse to be usable, such as Earth’s magnetic field. Bains then agreed that life is lazy (and intelligent life lazier). Difficult steps could take a long time to accomplish. For example, making oxygen from water is chemically difficult to do, and it seems like it took a long time for life to be able to do it. A planet that could support methanogenesis or oxygenesis might have life that has not yet evolved to do it.
A question was then posed to Bains about whether there could be biosignatures in the UV region that could complement the more commonly suggested biosignatures. Bains said that there are biological molecules that absorb in the UV, but he was unsure of geological molecules. Another audience member then answered that ozone photolytically produced by O2 is the best example. In the Proterozoic era when there was less O2, however, the ozone signal may nonetheless be visible in the UV. Methane, on the other hand, absorbs at UV wavelengths that telescopes are unlikely to be able to observe. Other molecules could work too, but they all have better lines in the visible and the infrared. Another audience member then chimed in to say that pigments could absorb in the UV, but that these are not apparent on Earth because little UV radiation makes it to the surface. On other planets, however, these pigments could potentially create a strong surface signature.
Nick Siegler of the NASA Jet Propulsion Laboratory (JPL) began his talk by stating that the main goal of the Exoplanet Exploration Program technology effort is to enable future space missions to observe a planetary spectrum of a rocky planet in the habitable zone of its star and understand it in the context of potential life. He went on to say that the main exoplanet discovery tools—the radial velocity and transit techniques, which have discovered more than 95 percent of the more than 3,400 known exoplanets—will not be the techniques to directly image exoplanets, which is needed to get a reflected light spectrum. Spectroscopy will be hard because there simply aren’t many photons available to use, but it will not be the biggest problem. The biggest problem will be suppressing the light from the stars, which can be 10 billion times brighter than a rocky planet in the habitable zone of a Sun-like star. Starlight suppression could be done in one of the following three ways: internal occulters (i.e., coronagraphs), external occulters (i.e., starshades), and nulling interferometers. The latter option is the least technologically mature of the options and one that NASA is not currently pursuing.
While the Hubble Space Telescope (HST) and the James Webb Space Telescope (JWST) both have coronagraphs, Siegler explained, WFIRST will be the first space telescope with a coronagraph (or possibly a starshade) specifically designed for directly imaging exoplanets. WFIRST’s Wide-Field Instrument (WFI) will arguably help answer questions in three of the biggest astrophysical areas—dark matter, dark energy, and exoplanets (via microlensing and coronagraphy). The telescope’s coronagraph instrument (CGI) will be used for the direct imaging and spectroscopy of exoplanets. WFIRST is in its formulation phase (Phase A) at this time. The project, telescope, and WFI are managed by NASA Goddard Space Flight Center, while the CGI is managed by JPL. The project has now also been directed to study the compatibility of a starshade with WFIRST. The current state of the art for coronagraphs is the Gemini Planet Imager (GPI) and the Very Large Telescope Spectro-Polarimetric High-contrast Exoplanet Research instrument (VLT SPHERE). WFIRST would improve upon their contrast ratio capability by 2 to 3 orders of magnitude and also improve upon the ability to probe smaller planet-star separations (see Figure
3.3). Further technological advancement would be required to observe rocky planets in the habitable zone of stars at a distance of 10 parsec (pc) and further.
Siegler then showed a video from JPL about how a classical coronagraph works.41 As a star’s light, depicted in the form of a wavefront, passes through the telescope, it becomes distorted by the slight imperfections inherent in any telescope’s optics. Diffraction adds concentric rings to the images. To see the planets, a mask is inserted to block most of the star’s light and redirect the rest of the light to the outer edge. A washer-shaped object then blocks most of the redirected light. Because the planet’s light comes in at an angle, it misses the first mask and goes through the center hole of the washer-shaped object. At this point, the planet’s light is still obscured by the residual starlight leaking through. To reduce the amount of leaking starlight, a deformable mirror is used to correct the distortions in the incoming light beam. This can then reveal the existence of a planet in the image up to a billion times fainter than the star. The video finished by saying that the planet’s light can then be directed into a spectrograph for spectral analysis.
Siegler then continued, elaborating a list of what a future telescope with a coronagraph would need in order to study Earth-like planets in Earth-like orbits around Sun-like stars. It would need to improve its contrast ratio sensitivity relative to that of WFIRST’s coronagraph by about two orders of magnitude. Deformable mirrors and image post-processing are fairly well advanced but need to go farther. Integration times would be days to weeks typically, so the system needs to be extremely stable. Otherwise, telescope vibrations and thermal distortions can cause blurriness. Siegler said that wavefront sensors would need to be able to measure wavefront distortions up to 10 picometers (pm), a couple of orders of magnitude better than HST (the current best), and correct for them. The technological capability to build large, segmented mirrors in a way that the optics are phase coherent—which may be required to build telescopes with primary mirrors exceeding 4 m—to within at least nanometers is not yet developed. Because of the long integration times, photon rates will be measured in photons per minute, so detectors with ultralow read noise are necessary, especially in the infrared. The size of the telescope is another question, especially with regard to a large, monolithic mirror versus a segmented mirror. Siegler then showed an image of potential telescope architectures for 12-m segmented mirrors of various segment sizes and shapes (hexagonal to a more radial, pie-like structure). The main problem with segmented mirrors is that all the small gaps add additional layers of diffraction, and the primary purpose of a coronagraph is to remove diffraction.
Siegler then showed an animation of a telescope with a starshade.42 They were two separate spacecraft with separate propulsion systems. When aligned, the starshade blocked the star’s light, revealing the reflected light of the planets. The starshade possesses a petal-like shape which serves to reorient the diffraction, creating a dark shadow for the telescope. He claimed that, in many ways, a starshade is a simpler method than the coronagraph because the starshade is doing all the work. It drastically reduces wavefront-control requirements on sensitivity, segment phasing, and other corrections. It has a higher tolerance for error as long as the starshade performs as designed. The starshade would be tens of meters across and tens of thousands of kilometers away. The starshade needs to be able to deploy and position its petals and maintain its physical stability, suppress the starlight, and fly in formation with a telescope separated from it by tens of thousands of kilometers and maintain the telescope’s lateral offset within acceptable limits. He then showed a starshade optical demonstration performed by Northrop Grumman in the Nevada desert, which was able to detect a simulated planet 100 million times fainter. Another experiment used a baseline of 2.4 km with a solar telescope to block out Arcturus and observe background stars. Another test, currently ongoing at Princeton University, has exceeded a contrast ratio of 10−8 at a single wavelength of 632 nanometers.
The starshade will be challenging to manufacture. The petals, he said, will need to be about 6 to 8 meters in length and fabricated to a tolerance of about 100 microns. The petals will need to be deployed to millimeter-level
41 NASA, “The Search for Alien Earths—How Coronagraphs Find Hidden Planets,” video, https://exoplanets.nasa.gov/exep/coronagraphvideo/, accessed December 5, 2016.
precision. JPL tested a deployment method for the petals, showing proof of concept. Another challenge is how to store an opaque starshade for launch and then deploy it without snagging or damaging it, since the starshade relies on its ability to remain opaque. A small, origami-like folding technique worked, so Siegler said that “JPL held back no expense” and performed a larger version (about half the size WFIRST would need) using corrugated cardboard and three interns. A recent prototype demonstrated a smaller starshade, but with more flight-like materials such as Mylar and high-density polyurethane. He then added that they think they have figured out formation flying to meter-level precision using current equipment on WFIRST.
Siegler said that everything in his talk could be found in the Exoplanet Exploration Program Technology Plan Appendix from 2016 (the 2017 update is now released and can be found at their website).43 He then brought up a slide showing past and future NASA and European Space Agency exoplanet missions, such as CHEOPS and PLATO. He requested that future planning think favorably of exoplanets, since we won’t be able to analyze biosignatures and false positives or negatives unless we can directly image these exoplanets. Siegler finished by mentioning two NASA-chartered mission concept studies that will be considered for possible future missions that could dramatically advance the field of exoplanets: the Habitable Exoplanet Imaging Mission (HabEx), a 4-m monolithic mirror or 6.5-m segmented mirror, or the Large UltraViolet/Optical/Infrared Surveyor (LUVOIR), a 9- to 16-m segmented mirror.
An audience member asked what the expected lifetime of a starshade would be and how many targets it could reach before running out of fuel. Siegler pointed out that this is a valid question due to the fact that micrometeoroids in space would likely pierce the starshade, limiting its lifetime. He said that, with multiple plies in the starshade, a micrometeoroid is unlikely to pierce perfectly orthogonal to the starshade where leaked light could do the most damage. He estimated a lifetime on the order of years. Pressed on the topic of fuel, Siegler explained a scenario that uses chemical propulsion to keep the starshade aligned with micro-thrusts and uses solar-electric propulsion for slewing to different targets. Another option is having two starshades so that one could be in operation while the other one was slewing.
Staying on the topic of the starshade, another participant pointed out that the tips of the petals have to be precise and sharp and then asked how they would clean dust off of them. Siegler said that they don’t know yet, but agreed that the tips need to be razor thin, about one micron thick. Dust, typically on the order of a wavelength, could be a problem.
A biologist then asked why astronomers were so focused on Earth-like planets and so pessimistic about hot Jupiters. He thought that only about 30 percent of the NASA exoplanet program portfolio should be about Earth-like planets, not 100 percent. Siegler said that he embraced that view but explained that by focusing the technology development on detecting Earth-like planets, you get the other planets for free. WFIRST, for example, would be able to detect hundreds of cold Jupiters, Saturns, and Neptunes too. Another audience member then commented that WFIRST will get about 10 times more total planets than rocky planets in the habitable zone by doing an observational sweep in direct imaging.
That same commenter then raised a new question about whether the trick allowing a potential starshade to work with WFIRST would also allow one to work with JWST. Siegler answered that NASA did study whether JWST could be designed to be compatible with a starshade, but they decided against it for technical and programmatic reasons. He then moved back to WFIRST and the collaboration between the JPL starshade and coronagraph teams and the Goddard spacecraft team. He said that the teams had found a relatively simple approach that addressed telescope-starshade alignment requiring minimal modifications to existing instrumentation. The WFIRST project has been asked to continue carrying starshade compatibility in their designs subject to review. A final decision would likely be made by NASA no later than fiscal year 2018.
An audience member then asked about the precision of the stability between the starshade and the telescope. Siegler noted that this is the formation flying issue. He said that the dark shadow of the starshade is about 2 m in
43 NASA, “Technology Needs and Gap Lists,” https://exoplanets.nasa.gov/exep/technology/gap-lists/, accessed December 5, 2016.
diameter and is cylindrically shaped. The lateral precision needs to be within 1 m, but the on-axis precision can have tolerances of hundreds of kilometers. The 1-m control precision has been done before on other spacecraft, including those docking with the space shuttle, but the angular alignment required with WFIRST is on order of milli-arcseconds, which is in a whole new regime. The audience member then asked him how problematic he felt this was. Siegler answered that recent testbed demonstrations were relieving him of his concern. He no longer thinks that the two spacecraft sensing their relative positions is a problem and said that the necessary control has never been a problem.
The final questioner then asked what WFIRST could do for exomoons. Prompted by the audience, Siegler responded by saying that WFIRST’s microlensing capabilities could potentially detect an exomoon, which would have a very unique lensing signature. However, spectral characterization would be impossible.
Matteo Brogi of the University of Colorado, Boulder, began his talk by saying that he has been using ground-based telescopes to look at hot Jupiters with high-resolution spectroscopy. He said that this could be possible for smaller, fainter planets in the future, such as Proxima Centauri b, which orbits the nearest star to the Sun (1.3 pc away).44 Proxima Centauri b has a minimum mass (msini) of 1.27 M⊕, a semi-major axis of 0.05 AU, a period of 11 days, a radial velocity semi-amplitude of 1.4 m/s, and does not appear to transit.
M-Dwarf Habitable-Zone Planets
Brogi then stressed that, not only is there a planet around the closest star, but that the size of the planet represents the most common type of planet around FGK dwarfs45 and especially M dwarfs,46 according to Kepler. Having a small planet orbiting a small star gives a higher planet-to-star contrast ratio than if the same planet orbited a larger star. This is true for measurements of the planet’s transit depth, reflected light, and thermal emission. The transit depth depends on the relative size of the planet and the star. The thermal emission also depends on the square of the relative planet-to-star radius, and additionally, it has a strong dependence on the temperature of the two bodies. Smaller stars are also cooler, hence they appear fainter when compared to the planet. Finally, habitable-zone planets orbiting M dwarfs need to be very close to the star. With the quadratic dependence on semi-major axis, the reflected light contrast ratio will drastically improve. Brogi then pointed out that M dwarfs are the most common type of stars nearby (about 70 to 80 percent are M dwarfs). Putting all the information together, Dressing and Charbonneau47 in 2015 predicted that the nearest transiting and non-transiting exoplanets in the habitable zone is about 10.6 pc and 2.6 pc away, respectively. Proxima Centauri b is even closer than that.
From basic energy arguments, Brogi said that the habitable zone will be much closer to the star than for a Sun-like star. Proxima Centauri b is in the middle of the classical habitable zone for an M-dwarf star (like Proxima Centauri).48 He stressed again that this means that the reflected light signal will be enhanced. The transit probability (radius of the star divided by the semi-major axis, assuming a circular orbit) of an M-dwarf, habitable-zone planet will also be enhanced, and because of the shorter period, transits are more frequent and can be stacked to increase the signal-to-noise ratio.
44 G. Anglada-Escudé, P.J. Amado, J. Barnes, Z.M. Berdiñas, R.P. Butler, G.A.L. Coleman, I. de la Cueva, et al., 2016, A terrestrial planet candidate in a temperate orbit around Proxima Centauri, Nature 536:437.
45 F. Fressin, G. Torres, D. Charbonneau, S.T. Bryson, J. Christiansen, C.D. Dressing, J.M. Jenkins, L.M. Walkowicz, and N. Batalha, 2013, The false positive rate of Kepler and the occurrence of planets, The Astrophysical Journal 766:81.
46 C.D. Dressing and D. Charbonneau, 2015, The occurrence of potentially habitable planets orbiting M dwarfs estimated from the full Kepler dataset and an empirical measurement of the detection sensitivity, The Astrophysical Journal 807:45.
48 R.K. Kopparapu, R. Ramirez, J.F. Kasting, V. Eymet, T.D. Robinson, S. Mahadevan, R.C. Terrien, S. Domagal-Goldman, V. Meadows, and R. Deshpande, 2013, Habitable zones around main-sequence stars: New estimates, The Astrophysical Journal 765:131.
Characterizing Exoplanet Atmospheres
Up until a few years ago, transiting planets offered the only opportunity to characterize atmospheres, Brogi said. While one cannot separate a planet’s light from the starlight solely based on the system’s geometry and differential measurements in time, it is possible to discriminate between the two. When the planet is in front of the star, the star’s flux appears to drop, and some of the starlight gets filtered through the planet’s atmosphere. The atmosphere’s spectral fingerprint is then imprinted onto the observed spectrum, which is called a transmission spectrum. When the planet moves behind the star, the flux from the planet (thermal and reflected) disappears, so one can compare the spectrum before and after to characterize the planet’s atmosphere. Stable, sensitive instrumentation can also measure the total light of system as a function of the planet’s orbital phase, which is called a phase curve. All of these can be used to help determine the atmosphere’s composition, thermal structure, and energy balance.
However, Brogi reminded us that Proxima Centauri b does not transit. For nontransiting planets, he proposes to use high-resolution Doppler spectroscopy to separate the planet’s light from the starlight in the spectral domain in addition to the temporal domain. An advantage of high-resolution spectroscopy, he said, is that each molecular species is resolved into the individual lines, resulting in unique and very specific fingerprints. Matching techniques, such as cross-correlation, can be used to detect these species unambiguously. Planets orbiting close to their parent stars also acquire a very distinct Doppler signature due to their orbital motion. Brogi showed a figure demonstrating, with a toy model, the visibility of the planet signature with respect to the telluric lines from Earth’s atmosphere (see Figure 3.4). While the planet is moving along the orbit, its radial velocity changes by tens of kilometers per second. In contrast, Earth’s atmospheric absorption (telluric absorption) lines remain stationary in velocity (i.e., in wavelength). This duality allows us to effectively disentangle the contaminating telluric signal from the exoplanet signal and to remove the former very effectively without altering the latter. The residual data is then cross-correlated with model spectra for exoplanet atmospheres to combine the signal of all molecular lines. In this way, detections of molecular species also deliver the planet’s radial velocity. When compared to the previously known stellar radial velocity, the planet and the star are treated as a spectroscopic binary. This technique allows for a measurement of the planet’s mass and inclination without needing the planet to ever transit. A caveat to this, Brogi said, is that the result is not a real planet spectrum, but a likelihood function that is subject to uncertainties in the theoretical models. Brogi and collaborators have used this method successfully, mostly on the Very Large Telescope’s Cryogenic High-Resolution InfraRed Echelle Spectrograph (VLT CRIRES) around 2.3 and 3.2 microns. They have been able to measure the mass and orbital inclinations of three hot Jupiters (τ Boo b,49 HD 179949 b,50 and 51 Peg b51). Both CO and H2O have been confidently measured in the atmospheres of transiting and nontransiting planets.52,53,54 (They did not, however, find CH4, which is not surprising for these high-temperature planets.) While no thermal inversions have been detected, exoplanet rotation and winds have been measured based on the broadening of the cross-correlation function.55,56
49 M. Brogi, I.A.G. Snellen, R.J. de Kok, S. Albrecht, J. Birkby, and E.J.W. de Mooji, 2012, The signature of orbital motion from the dayside of the planet τ Boötis b, Nature 486:502.
50 M. Brogi, I.A.G. Snellen, R.J. de Kok, S. Albrecht, J.L. Birkby, and E.J.W. de Mooji, 2013, Detection of molecular absorption in the dayside of exoplanet 51 Pegasi b?, The Astrophysical Journal 767:27.
51 M. Brogi, R.J. de Kok, J.L. Birkby, H. Schwarz, and I.A.G. Snellen, 2014, Carbon monoxide and water vapor in the atmosphere of the non-transiting exoplanet HD 179949 b, Astronomy and Astrophysics 565:124.
52 I.A.G. Snellen, R.J. de Kok, E.J.W. de Mooij, and S. Albrecht, 2010, The orbital motion, absolute mass and high-altitude winds of exoplanet HD209458b, Nature 465:1049.
53 J.L. Birkby, R.J. de Kok, M. Brogi, E.J.W. de Mooij, H. Schwarz, S. Albrecht, and I.A.G. Snellen, 2013, Detection of water absorption in the day side atmosphere of HD 189733 b using ground-based high-resolution spectroscopy at 3.2 μm, Monthly Notices of the Royal Astronomical Society 436:35.
54 R.J. de Kok, M. Brogi, I.A.G. Snellen, J. Birkby, S. Albrecht, and E.J.W. de Mooij, 2013, Detection of carbon monoxide in the high-resolution day-side spectrum of the exoplanet HD 189733b, Astronomy and Astrophysics 554:82.
55 I.A.G. Snellen, B.R. Brandl, R.J. de Kok, M. Brogi, J. Birkby, and H. Schwarz, 2014, Fast spin of the young extrasolar planet β Pictoris b, Nature 509:63.
56 M. Brogi, R.J. de Kok, S. Albrecht, I.A.G. Snellen, J.L. Birkby, and H. Schwarz, 2016, Rotation and winds of exoplanet HD 189733 b measured with high-dispersion transmission spectroscopy, The Astrophysical Journal 817:106.
In order to use this same method on rocky, habitable-zone planets, Brogi illustrated how to combine it with high-spatial resolution via future integral-field, high-resolution spectrographs (IFSs).57,58 This allows for the same analysis as on the hot Jupiters mentioned previously, but for each individual pixel of the IFS. He said that spreading out the photons spectrally does not cause a loss of signal, because spectra are recombined at a later stage during cross-correlation. Brogi’s group tested the method already with VLT CRIRES on known directly imaged planets.59,60 He said that it worked because of two factors that help increase the signal-to-noise ratio. First, the signal-to-noise ratio increases with the square root of the number of lines analyzed. Second, the contaminating light from the star is suppressed. They ran a simulation of this method using the parameters of the future European Extremely Large Telescope (E-ELT, 39 m diameter) with the Mid-Infrared E-ELT Imager and Spectrograph (METIS), which is a high-resolution, near-infrared spectrograph with an integral field unit. With just classical direct
57 I. Snellen, R. de Kok, J. L. Birkby, B. Brandl, M. Brogi, C. Keller, M. Kenworthy, H. Schwarz, and R. Stuik, 2015, Combining high-dispersion spectroscopy with high contrast imaging: Probing rocky planets around our nearest neighbors, Astronomy and Astrophysics 576:59.
58 C. Lovis, I. Snellen, D. Mouillet, F. Pepe, F. Wildi, N. Astudillo-Defru, J.-L. Beuzit, et al. 2017, Atmospheric characterization of Proxima b by coupling the SPHERE high-contrast imager to the ESPRESSO spectrograph, Astronomy and Astrophysics 599:16.
59 I.A.G. Snellen, B.R. Brandl, R.J. de Kok, M. Brogi, J. Birkby, and H. Schwarz, 2014, Fast spin of the young extrasolar planet β Pictoris b, Nature 509:63.
60 H. Schwarz, C. Ginski, R.J. de Kok, I.A.G. Snellen, M. Brogi, and J.L. Birkby, 2016, The slow spin of the young substellar companion GQ Lupi b and its orbital configuration, Astronomy and Astrophysics 593:74.
imaging, a planet slightly larger and warmer than Earth orbiting Alpha Centauri B would barely be detectable. However, when using cross-correlation filtering, a 5σ detection of the planet would be made after 10 hours (8σ after 30 hours). An exact Earth copy orbiting Alpha Centauri B would be just barely detectable, but a confident detection of Proxima Centauri b would be possible. For optical reflected light, a detection would be more challenging. However, one advantage is that, with an M dwarf’s many spectral lines, cross-correlation can result in a gain of 65 to 80 in signal-to-noise ratio. Even so, Proxima Centauri b’s reflected light would only be detectable with starlight suppression via extreme adaptive optics. In that case, a 10-hour observation would suffice. Brogi then gave a caveat that this is based on a scaled version of Earth and does not take into account trying to retrieve the planet’s properties. However, he has a plan to make this feasible for the next generation of extremely large telescopes.
Brogi then emphasized that Proxima Centauri b is just a prototype. While the habitability of planets orbiting M dwarfs is in question, he said that Proxima Centauri b is the best near-term chance to allow us to test our observational skills and characterize a potentially habitable world. He then concluded by saying that this ground-based, high-resolution technique is capable of getting the masses, inclinations, rotations, and wind speeds for nontransiting hot Jupiters already, but if combined with high spatial resolution, it could allow the same thing for potentially habitable planets in the future.
An audience member asked how many M dwarfs the next generation of extremely large telescopes will be able to survey. Brogi answered at least 10. The Giant Magellan Telescope—the smallest of the three planned, extremely large, ground-based telescopes at 25 m—is his favorite because it will have high-resolution spectroscopic capabilities from the start. One issue, however, is how much telescope time will be available for these observations. If the noise is not Gaussian, then it becomes more difficult. He said that 50 to 100 hours of observational time would be a reasonable time investment.
Another audience member then asked about the difficulty of using ground-based telescopes to look for biosignatures due to contamination by Earth’s atmosphere. Brogi replied that the Doppler shift of the planet will help distinguish it from Earth’s atmosphere, both in terms of where the lines appear and how the lines change due to the planet’s orbit. He then said that, at a resolution of 100,000, Earth’s atmosphere will not prevent you from detecting these features.
GENERAL DISCUSSION: PRACTICAL BIOSIGNATURES THAT CAN BE EXPLOITED TO SEARCH FOR LIFE IN SITU IN THE SOLAR SYSTEM AND FROM AFAR ON EXTRASOLAR WORLDS
The moderator for the general discussion on practical biosignatures both in the solar system and for exoplanets was Gary Blackwood from JPL and manager of NASA’s Exoplanet Exploration Program. The second question in the workshop’s statement of task (see Appendix C) was the focus of this discussion: Are we today positioned to design, build and conduct experiments or observations capable of life detection remotely or in situ in our own solar system and from afar on extrasolar planets? Blackwood then informally polled the audience on their opinion on the answer to that question. There was a mix of yes, no, and unsure. Blackwood then listed five topics to guide the discussion.61
What’s Changed Since 2000?
In 2000, the National Research Council released a workshop report called Signs of Life: A Report Based on the April 2000 Workshop on Life Detection Techniques.62 Blackwood asked what is new since then and what has changed in technology, scientific discoveries, and understanding. He then opened the floor for discussion.
61 The text in this section is not necessarily in chronological order. Comments have been moved out of chronological order to improve flow and preserve continuity of thought.
62 National Research Council, Signs of Life: A Report Based on the April 2000 Workshop on Life Detection Techniques, The National Academies Press, Washington, D.C., 2002.
Starting with Earth biology, an audience member said that Robert Hazen has articulately advocated that our understanding of mineral complexity on Earth is largely attributable primarily to life (but also to liquid water). She asked how we should use this to interpret results from other bodies in our solar system and even exoplanets. She brought up the idea of geobiology and emphasized that life is a planetary process.
Another workshop participant then said that more than one example of oxygenic photosynthesis is now known. She said that at longer wavelengths, life co-evolved with its environment. Another person then said that new developments have been made in studying the terrestrial biosphere, such as the extent of the deep biosphere and the ability to detect microbes there and differentiate between active and dormant biomass.
A new discovery since 2000, an audience member said, is the Lost City, a field of hydrothermal vents in the middle of the Atlantic Ocean. This has led to the development of a robust model for life emerging at one of these alkaline, hydrothermal systems.
A member of the audience then said that we now understand what chemical features a universal genetic molecule would have: a one-dimensional biopolymer with a backbone of repeating charges. He then said that work done in the laboratory has shown that RNA was likely the first molecule on Earth to gain access to Darwinism. Changing topics to Mars, he said that we now know that the surface of Mars is not self-sterilizing. Earlier notions to the contrary were attributable to misinterpretations of the 1976 Viking result.
Sticking with the subject of Mars, an audience member said that the Mars rover missions, particularly the Curiosity rover in Gale Crater, have discovered long-lived, aqueous environments. He said that this demonstrated not a biosignature, but the ability of sedimentary rock to preserve organic matter over a long period of time. Another participant followed up saying that there is now lots of evidence for liquid water processes on Mars that were unknown in 2000, such as recurring slope lineae from the Phoenix mission and the High Resolution Imaging Science Experiment (HiRISE) on the Mars Reconnaissance Orbiter (see Figure 3.5). She said that she thinks that the case for modern life on Mars is growing but has been ignored.
A new technological improvement, another participant said, was the ability to perform high-precision measurements of isotopes in molecules on Mars using both Curiosity and the Mars Atmosphere and Volatile EvolutioN (MAVEN) orbiter. This has given us a window into Mars’ past in terms of its atmospheric pressure and how it evolved over time. Another innovation, he said, has been the tunable laser spectrometer. The ExoMars Trace Gas Orbiter should be able to do even better for trace gases. Another member of the audience agreed. She said that two classes of measurements have been miniaturized for Mars and can now be used for other bodies in the solar system: the ability to precisely measure stable isotopes and the ability to do precision chemical and mineralogical analysis at the micro scale. She noted that the rovers and orbiters have discovered an amazing diversity on the surface of Mars.
A member of the audience then gave a plug for sample return since 2000. He said that we as a community have returned material from comets and are on our way to an asteroid. There are plans to do the same with Mars. The possibility also exists for a sample return from a plume of Enceladus.
Changing the topic to Venus, another participant said that life could exist in the clouds of Venus. He lamented the focus on the habitable zone as referring to only the surface temperature and not considering habitable temperatures elsewhere in the atmosphere. Since 2000, there has been discussion of exploring Venus’s clouds using an unmanned aerial vehicle, which might be possible in the next decade. He said that, although there haven’t been any new scientific discoveries, life in the Venusian clouds is reasonable considering that the properties of bacteria on Earth (chemical composition, spectral properties, and size distribution) are similar to the cloud particles on Venus. He suggested we might actually be observing bacteria. He then brought up the UV absorber in Venus’s atmosphere, whose origin remains unknown after 50 years. Bacteria, he thought, could be its origin too, and he asked the community to consider this idea.
Moving to the outer solar system, another workshop participant brought up Cassini’s discoveries on Enceladus. We now know that it has a liquid water ocean, and there is compelling evidence for hydrothermal vents as well. She then brought up the discoveries on Titan and the possibility of weird life. More generally, she said that community interest in ocean worlds as interesting targets has grown. Another audience member then mentioned the possibility of plate tectonics on the Europan ice shell and progress on researching different types of ice phases.
Going now beyond the solar system, a member of the audience talked about the discoveries from the Kepler space telescope. She said that it shows that terrestrial exoplanets are probably common, mentioning Proxima Centauri b as an example. Another major thing, she said, was the evolution of our understanding of false positives when it comes to biosignatures. This has removed the idea of O2 being an easy, straightforward biosignature. Therefore, we now know that we must have an understanding of the entire environmental context and of life as a planetary process. Continuing on this topic, another audience member referred to a debate on martian magnetofossils, which helped us become more skeptical and focus more on potential false positives.
In the final comment about this topic, an audience member talked about what has not yet happened. There is still no magical, Star Trek tricorder. All of our progress has been based on a priori knowledge of our biochemistry that isn’t going to be known for life beyond Earth.
In Situ and Remote Sensing
Blackwood then moved to the topic of doing in situ and remote sensing of life. He asked what we should search for and why, what processes we should use, and how we could improve the robustness of detection and subsequent interpretation.
An audience member began the discussion by talking about how multiple methods should be used. For example, studies in Antarctica, the Atacama Desert, and the Mojave Desert have revealed hidden microbes, sometimes under just a millimeter of rock. She thought that we should try to identify spectral signatures from not just orbiting satellites, but maybe aircraft and rovers as well.
A member of the audience then mentioned the topic of clumped isotopes, thinking that it would be interesting to do an analysis to learn the sources of the hydrogen and methane on Mars. Going back to the previous discussion, he said that a big discovery since 2000 is life in deep sediments that could have generation times of millions of years. He then asked how we would account for and measure it if it were on Mars. Another person agreed, saying that our frame of reference is one of high energy, but our targets could have very low energy. This should cause us to shift the way we think about biosignatures.
Another participant in the workshop said that, for Mars, the mantra of “follow the water” has been very successful. Expanding this idea to exoplanets, he said that a planet the size of Earth in the habitable zone of a G-dwarf star might not be adequate because it might not be able to get enough water (or, at least, enough water to be detectable). Coming back to Mars, he said that we still do not have evidence of an aquifer on Mars. He thought that we should extend our search for water on Mars deeper below the surface.
A conference participant then said that Mars isn’t just telling us to “follow the water,” but to “follow the salty water.” He thinks that the same is probably true for Europa. Therefore, we should focus on biosignatures that can be preserved in salty places. As a brief aside, he also said that a search for extraterrestrial intelligence would be a great place to start doing remote sensing.
Following up on the “follow the water” mantra, another audience member said to “follow the carbon.” He said we have been doing this, but not enough has been found to get excited about it. Therefore, we should target the sediments mostly likely to reveal organics with Curiosity’s Sample Analysis at Mars (SAM) instrument. In 2020, he said, we will have better remote detection capabilities of organics. He said that if we keep discovering water over and over again, people will eventually catch on.
Again emphasizing the need to look below the surface of Mars, another person commented that we should develop better subsurface exploration techniques using either direct or indirect observations. He said that this was also applicable to Europa. He then moved to exoplanets. He said that, with the Transiting Exoplanet Survey Satellite (TESS), we could have 10 to 50 exoplanets that we could follow up on, but only the time for maybe one or two of them. Because choosing which one(s) could be tricky, he said, we should develop models to predict a planet’s environment from only the planet’s mass, radius, and host star.
Then a member of the audience said that, while he was excited about directly imaging exoplanets, we should start thinking about instruments and observational techniques to complement direct spectroscopy. Both the planets and the stars, he said, will need to be better characterized in terms of mass and diameter. He gave Kepler as an example, the proposal for which included follow-up observations of ground-based radiovelocity detection for any discoveries made by Kepler itself.
Another member of the audience emphasized the idea of understanding the context of any discoveries. Multiple lines of evidence are important. Another workshop participant then said that the same thing applies to exoplanets, which means having a well-characterized host star.
Referencing Europa, Enceladus, and other outer bodies, another participant then said that we might not have the luxury of multiple mission campaigns to search for life. For these objects, the audience member said that we need to speed up the mission cadence and make careful predictions, so that follow-up missions in the works can handle whatever new discoveries were made in the meantime.
A member of the audience then said that we need more instruments that can do liquid-based analysis, especially if we are going to places looking for water-soluble organic molecules.
Solar System and Extrasolar Worlds
Blackwood then moved on to the third topic and asked, What is common to both scenarios in regards to the remote or in situ detection of biosignatures?
A workshop participant then said that the questions of what is practical for remote detection and in situ detection are very different. He was confident that any form of life could be detected in situ, even if it did not use DNA as its genetic material.
A WebEx viewer then posed a question about whether Juno could fly through Europa’s plumes to try to detect signs of life. An audience member then confirmed that Juno would not be capable of that. It does have a UV spectrometer, but it cannot do the observations in the way that would be needed.
Sticking to the topic of gas giants, a member of the audience then said that he was excited about the elemental abundances of giant planets, and not just hot Jupiters, because we are getting to the point where we can find colder gas giants. Comparison between the elemental abundances between Jupiter and Saturn versus extrasolar giant planets, effectively comparative planetology, could be interesting.
Another member of the audience was struck about the idea of fingerprinting a world. Looking at Venus, Earth, and Mars, she wasn’t sure what the lessons were. She wondered what properties of planets really control the composition of the atmosphere and enable a stable redox state through time. The level of coupling between the planet’s interior evolution and the nature of degassing, volcanism, and magnetic fields are not understood, she thought.
Continuing on that subject, a workshop participant said that Venus, Earth, and Mars all had liquid water in the past. Venus could have had liquid water for 2 billion years. He said that Venus could still have life in the clouds. Many exoplanets could also be Venus-like. For an in situ detection, he said that an airplane filled with hydrogen or helium could survive up to a year in Venus’s clouds. For remote detection, practical techniques could include doing isotopic measurements and looking for disequilibrium using Raman LIDAR. Since no planetary protection precautions for Venus are required per the Committee on Space Research (COSPAR), he said that it would be easy.
Referring back to earlier discussions, an audience member said that “follow the water” is not enough on Mars anymore. He said that we now need to determine which of the sites with water are the best. Similarly, he asked which exoplanets make the best targets. He then wondered what properties of a planetary system could give a sense of the composition of the planets.
A member of the audience wanted to tie “follow the water” together with Venus, Earth, and Mars using observations to see what they all have in common. One such thing is patterns in circulation, and we have general (or global) circulation models (GCMs). 3D GCMs can explore the habitability of a wide range of planets.
Another participant at the workshop emphasized the importance of the planetary interior. He said that processes like plate tectonics and volcanism are not linear or predictable but can be addressed in a probabilistic way depending on things like the planet’s mass, composition, evolution, volcanism, and magnetic fields. In this way, we could potentially see what sort of features make certain processes more likely, such as hydrogen outgassing, oxidizing outgassing, dynamo generation, or plate tectonics. In this framework, Venus would be an important data point. A mission to Venus to learn its interior properties, like its rheology, water distribution, seismology, or dynamo generation, would be useful.
Are We Ready?
Blackwood then asked if we are ready today to engineer and observe life detection remotely or in situ for either the solar system or exoplanets.
Starting with in situ detection, a member of the audience said that we were technologically ready. However, we were not ready to deal with the environmental context of the detection. Life might have a low signal-to-noise ratio, and the environment could produce a lot of measurement noise.
Another member of the audience then took exception to the phrasing of the question. He said that we are capable of designing a mission to go to an aqueous environment to detect life. However, in situ detection of extinct life is more difficult. It is uncertain whether or not that is possible today. Continuing on, he said that designing a mission to do remote detection of life using just spectroscopy is impossible right now. Another audience member also criticized the phrasing of the question. She said that many instruments could detect something that we interpret as life. However, if we go to a place beyond Earth, which may or may not have Earth-like life, we might not know whether or not we have found life. Many assumptions go into any interpretation.
After Blackwood asked for someone on the side of “no,” a workshop participant came out firmly on the negative side. She touched on previous themes and said that detecting life, especially very-low-biomass life, is non-trivial to do in real environments (whereas in a laboratory, it could be easy). Another aspect is the time dimension. Organisms in an environment that don’t do much could have a chemical signal. However, we might not be sure if it indicates life, since our assumptions are based on fast microbes. She said that the planet, the crust, and, to some extent, even the oceans and the ocean’s sediments are dominated by slow microbes.
Another conference participant said that many people believe that the best places to find life beyond Earth are in completely inaccessible areas, such as deep under the martian surface or below the crust of an icy world. Even
sampling a plume is nontrivial. She then said that people never really agree on an unambiguous biosignature. The interpretation will be very difficult.
A participant at the workshop then said that interpretation might not be a black and white issue. That being the case, an instrument relevant to detecting a biosignature should at least give a better characterization of the environment. This would then lead to approaching the issue better for the next mission.
Moving to the exoplanet side, a member of the audience said that we were definitely not ready for in situ detection of life on exoplanets. For remote detection, obtaining a secondary biosignature to constrain the fluxes of the primary biosignature is very difficult, such as methane for oxygen. However, he said, it is completely inadequate compared to Mars, which has an unexplained, variable amount of methane in its atmosphere. In response to this, he stressed the idea of using population statistics in terms of atmospheric detections in order to rule out certain scenarios. This could lead to results showing either life on several worlds or highlighting a profound ignorance of geology and geochemistry.
What Can We Learn from One Another?
Blackwood then moved on to the fifth and final topic of the discussion. He asked what we could learn about how to perform in situ and remote sensing of life or potential biosignatures both in the solar system and beyond. He also wanted to know what we could learn from people in other disciplines, such as planetary science, astronomy, biology, geology, oceanography, geochemistry, and others.
One member of the audience said that we have no choice but to learn from people in other disciplines. Otherwise, nobody in the scientific community will believe any claimed detection. Another workshop participant agreed, saying that, no matter how hard one tries, it is very difficult to be an expert in more than one field. Therefore, she thought that we needed to help each other out, particularly in avoiding false positives and false negatives.
Then a workshop participant said that one discipline that is really needed is statistics. The geological context, he said, is all statistical priors, not estimates on the probability that we have detected life. These priors must be understood. He also said that there is an entire industry devoted to designing experiments to optimize the outcome when your knowledge and your priors are uncertain: the clinical trials industry. These people might be useful when designing missions.
Two more disciplines were then added to the list by another audience member. The first was glaciology, since these are clearly visible from space. Glaciers could help concentrate organics or create drainage patterns. The other discipline she suggested would be useful is climatology. Climate change is one example of how life can alter the climate. For example, in addition to just global warming, oceans are rising and becoming warmer and more acidic.
A member of the audience then brought up work he had done that concluded that we should search for evidence of evolution. One thing he said we could learn from one another is how an evolving community of organisms changes the environment. Earth has probably had many previous versions of itself, but we do not understand them. He would like people to understand life as a planetary process and how life has been involved in planetary evolution.
A team member for both the Habitable Exoplanet Imaging Mission (HabEx) and the Large UV/Optical/ Infrared Surveyor (LUVOIR) then extended an invitation for people to communicate with them on what kind of observations they would like to make.
Another audience member brought up the idea of comparative planetology again. There is Venus, Earth, and Mars, but also Mercury and the Moon. He thought we needed to understand how their evolution was different and why they are so different today. Without knowing at least how Venus, Earth, and Mars are different, understanding exoplanets will not be possible.
To finish out the discussion, a member of the audience then addressed how in situ and remote detection could play off one another. For example, the detection of methane is a potential biosignature. In situ detection by the Curiosity rover, remote detection by Mars orbiter missions, and remote detection from Earth could answer the question of whether it truly is a biosignature. Something she thought was critical was using a combination of both modeling and laboratory work. The biggest thing that irked her was that we don’t know the effect that radiation has on the organic molecules on the martian surface, which she thought needs to be simulated.