Arxiv:1708.04241V3

Home , Milnesium, Superflare

arXiv:1708.04241v3 [astro-ph.EP] 16 Oct 2017 [email protected] orsodn uhr aav Lingam Manasvi author: Corresponding n supin soitdwt u nlss ercmedta t that recommend we analysis, our with associated assumptions and ysprae a o ensffiinl prcae,adta hum a that M- and of appreciated, zone sufficiently next habitable the been the positive in not in a has exoplanets play prove superflares and also by may Mars could present-day superflares Superflares on weaker pow compounds. life t hand, most organic correspond key the other could of that the formation periodicity propose On their We that record. and exoplanets. events, and extinction system, of Solar the in 2 1 onA alo colo niern n ple Sciences, Applied and Engineering of School Paulson St A. Garden John 60 Astrophysics, for Center Harvard-Smithsonian eepoesm ftermfiain rsn rmsprae nthe on superflares from arising ramifications the of some explore We yee sn L 2017 using 17, Typeset October version Draft IK O IEO AIAL LNT RMSPRLRSO THEI OF SUPERFLARES FROM PLANETS HABITABLE ON LIFE FOR RISKS ∼ 10 A T 3 E X er edn odvsaigeooi n ehooia oss In losses. technological and economic devastating to leading years twocolumn tl nAASTeX61 in style aav Lingam Manasvi ABSTRACT abig,M 23,USA 02138, MA Cambridge, , ,2 1, avr nvriy 9Ofr t abig,M 23,USA 02138, MA Cambridge, St, Oxford 29 University, Harvard n baa Loeb Abraham and ob ut ermna oteeouino complex of evolution the to detrimental quite be to eti atrsi h ersra osldiversity fossil terrestrial the in patterns certain o nt ih oetal ins ueflr event superflare a witness potentially might anity eerslssol evee ihdecaution. due with viewed be should results hese oei nbigteoii flf hog the through life of origin the enabling in role dKdaf.W ocueta h ikposed risk the that conclude We K-dwarfs. nd ru ueflrscnsrea luil drivers plausible as serve can superflares erful vltoayhsoyo at,ohrplanets other Earth, of history evolutionary 1 ih ftemn uncertainties many the of light OTSTARS HOST R 2

1. INTRODUCTION civilization. Finally, we summarize the salient results of Flares are eruptions of high-energy radiation from the paper in Sec. 5. stars, and phenomena associated with these events have been recorded, and studied, throughout human his- 2. CONNECTIONS BETWEEN THE TIMING OF tory (Eather 1980; Vaquero 2007). One of the power- SUPERFLARES AND SPECIES EXTINCTIONS ful solar flares on record, the Carrington event, dates In this Section, we shall explore the timescales associ- back to more than 150 years ago (Carrington 1859). ated with large superflares on the Sun and outline possi- Solar (and stellar) flares have been extensively stud- ble connections with the fossil diversity record. We also ied in recent times for a multitude of reasons. There delineate the caveats and assumptions in our model. has been a great deal of interest in understanding the physical mechanisms responsible for their origin, 2.1. Timescales for superflares and mass extinctions usually through magnetic reconnection (Priest 2014; Although large-scale extinction events have oc- Janvier 2017) resulting in the rapid release of magnetic curred multiple times on Earth, their exact num- energy (Shibata & Magara 2011; Janvier et al. 2015; ber remains uncertain. One of the important hy- Comisso et al. 2016, 2017). In addition, flares have been potheses put forward concerns the existence of puta- exhaustively studied in the context of space weather pre- tive periodic patterns in the fossil extinction record dictions (Schwenn 2006; Barnard et al. 2011), as they (Raup 1986; Hallam & Wignall 1997; Courtillot 1999). can indirectly cause damage to satellites and astronauts The estimates for the periodic timescale have typ- in orbit. As stellar flares are typically associated with ically ranged from 26 Myr (Raup & Sepkoski 1984, the emission of ultraviolet (UV) radiation and high- 1986) to 62 Myr (Rohde & Muller 2005). The ev- energy protons, several studies have been undertaken to idence in favor of and against this periodicity has gauge the robustness of life on Earth as well as other ex- been explored extensively over the past three decades oplanets to these events (Rind 2002; Buccino et al. 2006; (Patterson & Smith 1987; Benton 1995; Bambach 2006; Melott & Thomas 2011; Dartnell 2011; Atri & Melott Alroy 2008; Melott & Bambach 2014). To explain these 2014). extinction events, a wide range of astrophysical phe- A common theme in most of these papers is that the nomena such as gamma ray bursts (GRBs), supernovae, flares studied were not particularly extreme, as most the presence of a distant solar companion, and comet 32 of them were characterized by energies . 10 erg. impacts have been invoked; the reader may consult However, the launch of the Kepler mission to detect Bailer-Jones (2009) for further details. exoplanets greatly altered, and advanced, our under- If we choose a periodic timescale of τ = 26 Myr and standing of the statistics of flares (Walkowicz et al. hypothesize that the extinctions are caused by an astro- 2011; Maehara et al. 2012; Shibayama et al. 2013; physical phenomenon, the latter must repeat after this Candelaresi et al. 2014). The analysis of the Kepler interval of time. We shall posit that solar superflares data revealed that highly energetic flares, dubbed su- serve as a driver of these extinction events, and thereby perflares, occurr on M-, K- and G-type stars with a fairly examine whether they constitute a viable mechanism. high frequency. In turn, this discovery reignited interest We begin by noting that superflares are extremely rare in the possibility that superflares could occur on the Sun events, and solar observations have not been undertaken over the span of a few thousands of years (Shibata et al. for sufficiently long periods to directly document their 2013). In parallel, based on evidence from radionuclides existence (Usoskin 2017). Fortunately, the observations in tree rings (Miyake et al. 2012), it was suggested that of ∼ 105 solar-type stars by the Kepler mission have the deduced spike in cosmic rays could potentially be yielded a wealth of data (Maehara et al. 2012). For explained by a solar superflare that erupted in AD 775 slowly rotating, G-type stars like the Sun, the follow- (Melott & Thomas 2012; Usoskin et al. 2013). ing relation was empirically determined: In light of the mounting evidence concerning the importance of superflares, we carry out below an analysis dN ∝ E−α α & 2, (1) of their implications for life on Earth and exoplanets. dE The outline of the paper is as follows. In Sec. 2, we present connections between extinction events in the fos- where N was the occurrence rate of superflares as a sil record and the frequency of superflares and highlight function of the energy E (Maehara et al. 2012). A de- some of the assumptions inherent in our analysis. We tailed analysis of the Kepler data led to the conclu- 34 follow this by analyzing the effects of large superflares on sion that flares of energy ∼ 10 erg would occur ev- Earth in Sec. 3, and conclude that they can trigger mass ery ∼ 2000 yrs (Shibayama et al. 2013). In this con- extinctions. In Sec. 4, we explore the negative and pos- text, we observe that a superflare on the Sun with 34 itive consequences of superflares for life on Mars, Venus energy ∼ 10 erg that occurred in AD 775 (∼ 1250 and exoplanets orbiting low-mass stars. We also delin- years ago) has been posited (Melott & Thomas 2012; eate the economic risks posed by superflares to human Usoskin et al. 2013; Mekhaldi et al. 2015), although the proposed evidence and reasoning are open to other 3

2 interpretations (Miyake et al. 2012; Cliver et al. 2014; of the sunspot, and F = Aspot/ 2πR⊙ is the fraction Neuhäuser & Neuhäuser 2015). Another energetic event of the Sun’s surface covered by the sunspot with Aspot dating from AD 993 has been associated with a super- denoting its area; the normalization factor of 0.3 was flare (Miyake et al. 2013; Mekhaldi et al. 2015); how- selected based on the preceding facts. Here, we have ever, the corresponding astronomical evidence has been chosen a normalization of 10 kG for the magnetic field critiqued (Stephenson 2015). As the Sun has lower ac- as opposed to the standard value of 1 kG (Maehara et al. tivity levels with respect to most solar-type stars, it has 2015). Our choice is motivated by the fact that sunspots been suggested that the frequency of superflares on the with 6 kG are currently documented (Livingston et al. Sun with ∼ 1034 erg could be lower by an order of mag- 2006), while values . 30 kG have been conjectured for nitude (Karoff et al. 2016). solar-type starspots (Rubenstein & Schaefer 2000). We We are now in a position to answer the question: what have also normalized the efficiency by its characteristic is the energy E of a superflare that occurs with a fre- value of ∼ 0.1 (Shibata et al. 2013). quency of ∼ 20 Myr? Using the above information in Instead, if we suppose that α ∼ 2, the value of E cor- conjunction with α =2.3 (Maehara et al. 2012), we are responding to τ is correspondingly increased by about led to conclude that E ∼ 1037 erg. This raises the im- an order of magnitude, i.e. it must have a value of 1038 mediate question as to whether flares of this magnitude erg. A superflare with this energy would not be feasible are achievable on solar-type stars, since the Kepler sam- on the Sun, as the requirements for B, ǫ and F become ple only yielded values . 2 × 1036 erg (Shibayama et al. very stringent and unlikely. Thus, we propose that, for 2013; Maehara et al. 2015); on the other hand, it must the above choice of the parameters, the energy of the be recognized that the difference in the two maximal superflare that occurs once every 26 Myr ought to be values is less than one order of magnitude (a factor of E ∼ 1037 erg. We have also suggested that superflares 5). In order to answer this question, we shall rely upon a of this magnitude can, under a rare set of circumstances, combination of empirical and theoretical considerations. occur on the Sun. We can also invert this argument as From the observational standpoint, we note that flares follows: upon computing the maximum possible energy with energies ∼ 1038 erg have been documented in G- E of a solar superflare, we find that its frequency of oc- type stars (Schaefer et al. 2000). The result is pertinent currence corresponds to ∼ 20 Myr. This value is very since these stars are: (i) not rapid rotators, (ii) typically close to the periodic extinction timescale of 26 Myr pro- single, and (iii) not very young. Thus, in many respects posed by some authors (Raup & Sepkoski 1984). these stars are similar to the Sun, thereby suggesting At this stage, a few important points regarding so- that equally large flares may, in principle, also be man- lar superflares merit a mention: (i) they may occur ifested in the latter. We also wish to point out that at much longer intervals than ∼ 20 Myr (Gopalswamy superflares with energies ∼ 1037 erg have been docu- 2017), and (ii) they could be unevenly spaced. The mented for some G-type stars studied by the Kepler mis- former stems from the fact that the flare distribu- sion (Walkowicz et al. 2011; Basri et al. 2011). Based tion might decline rapidly due to an exponential falloff on the empirical evidence, it was suggested in Sec. V of at large values instead of the power law scaling (1). Schrijver et al. (2012) that the theoretical upper bound Such behavior has been documented for flux ropes that for superflares on “Sun-like” stars on the main sequence arise during the reconnection process (Janvier 2017; would be ∼ 1037 erg. However, we caution that some Lingam et al. 2017). We note that (ii) can be partly ex- of these stars possess higher ambient surface magnetic plained by invoking the fact that superflares in certain fields than the Sun; consequently, superflares on the Sun G-type stars are not strictly periodic since they have might have a very low probability of occurrence as dis- been documented to occur in ‘clusters’ (Maehara et al. cussed further in Sec. 2.2. 2012; Shibayama et al. 2013). Moreover, as stellar activ- Shibata et al. (2013) argued that superflares on the ity broadly declines with age (Mamajek & Hillenbrand Sun can arise provided that sufficiently large sunspots, 2008; Soderblom 2010), it is reasonable to expect that approximately 30% of the surface area, are formed. Fur- the frequency of superflares and the maximum energy thermore, some empirical evidence from other Sun-like released will reduce over time, implying that large su- stars also indicates that superflares could occur on the perflares would tend to become increasingly uncommon Sun (Nogami et al. 2014). If we consider active low- during later epochs. mass stars, a spot-coverage fraction of ∼ 0.4 appears to Hence, the above facts collectively indicate that the be fairly common (Jackson & Jeffries 2013). The energy likelihood of superflares on the Sun being rare and in- E of the flare can be expressed as termittent, as opposed to regular and periodic, is also quite high. In turn, any extinction events they po- ǫ B 2 F 3/2 tentially cause would also display the same proper- E = ǫE ≈ 1037 erg , mag 4 ties. Consequently, the hypothesis that some of the 0.1 10 G 0.3 (2) extinction events recorded since the Cambrian period where ǫ is the fraction of magnetic energy Emag convert- (Hallam & Wignall 1997) could have been triggered by a ible into flare energy, B is the magnetic field strength superflare merits further consideration. Before proceed- 4 ing further, we also wish to reiterate that the timescales available for explosive energy release; this necessitated discussed herein are subject to a fair degree of uncer- the inclusion of an efficiency factor ǫ in (2). tainty as the statistical properties are not robust; in- Lastly, we point out that the large value of the mag- stead, there is a paucity of data with respect to both the netic filling fraction (F ∼ 10%) employed in (2) is neces- fossil record and superflares on solar-type stars. Hence, sary for large superflares to occur (Shibata et al. 2013). these timescales should be interpreted as the character- The statistical analysis of the magnetic flux distribution istic values associated with the corresponding processes. on the Sun’s surface indicates that such large values of F Lastly, we observe that comparatively smaller super- are not feasible (Muñoz-Jaramillo et al. 2015). An im- flares, i.e. with energies much lower than 1037 erg, portant point that needs to be noted, however, is that may also play a role in regulating the biodiversity on detailed solar observations of sunspots date back to a few Earth. Although these events are not expected to cause centuries (Schrijver et al. 2012), which is clearly a very mass extinctions, their relative frequency is much higher short time span by geological and astronomical stan- compared to larger superflares (Maehara et al. 2012; dards; in particular, the plasma environment of Earth Shibata et al. 2013). Hence, it should be instructive to during the Hadean and Archean epochs could have been compare fossil biodiversity records against the predicted quite different (Airapetian et al. 2016). frequencies of superflares (with varying energies), and determine whether any significant correlations can be 3. EFFECTS CAUSED BY THE SUPERFLARE ON deduced. EARTH Next, we shall explore the environmental and biologi- 2.2. Caveats and assumptions for the model cal consequences arising from a superflare with E ∼ 1037 Here, we shall elucidate the assumptions and uncer- erg, and whether these effects are severe enough to trig- tainties associated with our preceding discussion. ger a mass extinction. We begin with the important observation that the We begin by estimating the energy E⊕ that is de- maximum energy as well as the constant of proportion- posited by the superflare on Earth. If one assumes that ality and the spectral index of solar superflares in (1) the flare energy is emitted isotropically, we find remains unknown. Hence, there is an inherent degen- 2 eracy that can be illustrated by the following example. R⊕ E⊕ = E , (3) Let us suppose that the timescale for a solar superflare 2a with 1034 erg is 6.5 × 104 yrs, which is higher than our previous choice by a factor of 30. Using α =2.3, we find where a = 1 AU. In contrast, it has been suggested that a superflare with E ∼ 1036 erg has a characteristic that the energy could be deposited in a non-isotropic ◦ timescale of τ = 26 Myr. In Sec. 3.1, we argue that manner with an opening angle of 24 (Melott & Thomas even a superflare with ∼ 1036 erg has the potential to 2012; Neuhäuser & Hambaryan 2014). If we consider ′ cause mass extinctions. A lower value of E would, in this scenario, the energy deposited will be E⊕ ∼ 100E⊕. turn, entail less stringent constraints on F and B in (2). Upon substituting the appropriate values in (3), we find 27 ′ 29 Next, it must be recognized that our analysis is based that E⊕ ∼ 4.5 × 10 erg and E⊕ ∼ 4.5 × 10 erg. on statistical considerations. Hence, in employing (1) we are implicitly relying on the assumption that the 3.1. Ozone depletion and its consequences sampling of a large number of G-type stars is roughly The role of ionizing radiation, produced by flares and equivalent to sampling the Sun over an extended pe- other catastrophic phenomena, on atmospheric chem- riod of time (around 4 × 105 yrs). However, it must istry and surface biology has been investigated quite be noted that not all G-type stars are “Sun-like”. This extensively (Dartnell 2011; Melott & Thomas 2011; intrinsic variability may imply that the corresponding Atri & Melott 2014). A number of factors, such as statistics for the Sun are not the same as (1). Hence, the Earth’s thick atmosphere, the presence of a mag- there is a distinct possibility that the Sun is incapable netic field, and the existence of ozone, serve to shield of giving rise to large superflares (Schrijver et al. 2012; the surface from the majority of biologically damaging Kitchatinov & Olemskoy 2016). On the other hand, the radiation - mostly Ultraviolet-B (UVB) and Ultraviolet- analysis of young solar-type stars based on Kepler data C (UVC). However, many of these studies concentrated (Shibayama et al. 2013) suggests that the Sun would on flares that were typically < 1033 erg. To the best have been more active when it was much younger. of our knowledge, the effects of a flare with E ∼ 1037 The estimate for the flare energy in (2) constitutes a erg do not appear to have been delineated in the litera- simple scaling analysis, and does not capture the time ture. Hence, our subsequent discussion will necessitate dependence of the flare energy released during the recon- a certain degree of extrapolation from known results. nection process (Shibata & Magara 2011; Priest 2014). It is important to recognize that, in discussing the Moreover, it is important to recognize that the mag- dangers arising from solar (or stellar) flares, there are netic energy ∝ B2 is not fully converted into the flare several distinct components associated with the lat- energy since the final magnetic field is not completely ter phenomena (Schwenn 2006; Shibata & Magara 2011; 5

Emslie et al. 2012; Benz 2017); for e.g., the electromag- We also note that the effects arising from strong flares netic radiation emitted, and the high-fluence outflow have been studied for Earth-analogs orbiting M-dwarfs. of solar energetic particles (SEPs). The second fac- Segura et al. (2010) considered a superflare on the active tor has been explored in detail (Miroshnichenko 2001) M-dwarf AD Leonis (AD Leo), and demonstrated that and identified as being particularly important, since it the UV radiation did not cause any significant ozone de- facilitates the formation of nitrogen oxides (NOx) by pletion. However, when the role of SEPs was taken into means of atmospheric ionization (Crutzen 1979). In account, the ozone depletion was shown to attain a max- turn, these compounds are responsible for the depletion imum of 94%. Here, two caveats must be recorded: the of ozone (Schumann & Huntrieser 2007; Jackman et al. Earth-analog was situated at a distance of 0.16 AU and 2008). As noted earlier, reduction in the ozone levels the maximum value was for an unmagnetized planet. enables higher doses of harmful UV radiation to reach Given that only a few data points are available, all re- the surface and leads to biological damage. We will now sults obtained from direct extrapolation must be inter- turn our attention to a few specific analyses of solar and preted with due caution. Furthermore, there are several stellar flares and their consequences. other factors involved in the extent of ozone depletion, The famous 1859 Carrington flare (Carrington 1859; for e.g. diurnal cycles (Verronen et al. 2005), which are Cliver & Svalgaard 2004) remains one of the most not considered in our analysis. Let us denote the ozone powerful solar storms ever documented, with a total depletion by DO3 and the SEP fluence by Fp. Assum- energy that is estimated to have been 5 × 1032 erg ing a power-law scaling, we suggest that the following (Cliver & Dietrich 2013). This event led to a globally expression serves as a reasonable fit for the SEP events averaged maximal decrease in the ozone levels of 5%, discussed earlier: although maximum depletion was 14% at higher lati- 2/5 tudes (Thomas et al. 2007). Other studies found that Fp D ∼ 5% . (4) ozone reduction of 20%-40% and 60% occurred in the O3 1010 protons cm−2 stratosphere and mesosphere respectively (Rodger et al. 2008a,b; Calisto et al. 2013); for comparison, the ozone The next step is to relate the flare energy E to the depletion in the stratosphere because of anthropogenic SEP fluence. To do this, we invoke the results from change is . 10% (Solomon 1999). We further point out Takahashi et al. (2016), where the scaling of the SEP that similar values for ozone reduction in the strato- flux Fp with E was obtained. Assuming an isotropic sphere and mesosphere were recorded for the 1972 angular distribution of the SEPs, and using Fp ∝ 4.35 1/6 1/6 (Heath et al. 1977), 1989 (Jackman et al. 2000), 2000- (VCME ) , VCME ∝ E , tCME ∝ E and Fp ∝ 2003 (López-Puertas et al. 2005; Jackman et al. 2005) tCME × Fp, the scaling relation is and 2005 (Seppäläet al. 2006) SEP events. In addition, significant changes in the surface air temperature were F E 9/10 p = , (5) identified for the Carrington flare, with Europe and 1010 protons cm−2 5 × 1032 erg Russia experiencing warming of . 7 ◦C (Calisto et al. 2013); the 2003 SEP event was responsible for temper- where we have normalized the fluence and flare energy in ature variations of up to ±3 ◦C (Jackman et al. 2007). terms of the Carrington event (Cliver & Dietrich 2013). The terrestrial effects of the putative superflare in AD Upon combining (4) and (5), we arrive at 775 were explored in Thomas et al. (2013), and it was concluded that an SEP fluence of ∼ 1012 protons cm−2 E 9/25 D ∼ 5% . (6) of particles with energies > 30 MeV would lead to se- O3 5 × 1032 erg vere damage of the biosphere. However, owing to the paucity of available data, a wide range of outcomes were If we substitute E ∼ 1037 erg in the above expression, predicted. The averaged ozone depletion ranged from a we find that DO3 ∼ 177%. As noted earlier, the same lower bound of 5% to an upper bound of 32% depending power-law behavior may not be valid for higher flare on the SEP fluence, with a fairly plausible intermediate energies and SEP fluences. It is also possible to com- value of 22%. In comparison, ozone depletion due to a pute the critical energy Ec that leads to 100% ozone 36 GRB at a distance of a few kpc is 38% (Thomas et al. depletion; we find it to be Ec ∼ 2 × 10 erg. If the 2005), and a supernova at 8 pc leads to a depletion of flare on AD Leo were scaled upwards to account for the 47% (Gehrels et al. 2003). The biotic effects due to the larger Earth-Sun distance, its equivalent energy would intermediate and upper cases were manifested as the in- be ∼ 1035 − 1036 erg. As this flare caused a maximum crease in UVB-induced damage of plants by 14% and of 94% ozone reduction (Segura et al. 2010), the flare 25% respectively. The SEP event in AD 775 was there- energy is roughly in agreement with the value of Ec cal- fore associated with moderate damage of the biosphere culated from our model. due to reduced photosynthesis in the oceans and land Let us recall that a GRB from a few kpc leads to (Thomas et al. 2013). ozone depletion of ∼ 40% (Thomas et al. 2005) and has been posited as the trigger for the Ordovician 6 mass extinction (Melott et al. 2004). A supernova at observe that some, albeit not all, of these factors have 8 pc has also been predicted to engender comparable been documented for the Permian-Triassic mass extinc- depletion (Crutzen & Bruhl 1996; Gehrels et al. 2003). tion (Knoll et al. 1996; Erwin 2006; Knoll et al. 2007) In contrast, as per our scaling relations, a flare en- although this does not necessarily imply that the P-T ergy upwards of Ec would cause complete destruction extinction was triggered (or exacerbated) by a super- of the ozone layer and correspond to a fluence of ∼ flare. 1013 protons cm−2. If such a superflare were to occur on the Sun (regardless of its periodicity), it seems rea- 3.2. Other ramifications from the superflare sonable enough to argue that the damage to the bio- Apart from the manifold consequences of sudden sphere would be great enough to trigger a mass extinc- ozone depletion and enhanced UVB radiation, super- tion, especially since severe ozone depletion engenders flares of this magnitude could also give rise to other widespread and major biological damage (Thomas et al. effects, some of which have an interesting mix of neg- 2015). Hence, if a large flare (even one with E ≪ Ec) ative and positive consequences. As described earlier, subsequently erupted before the ozone layer had been Calisto et al. (2013) concluded that the the Carrington replenished, virtually all organisms on the surface, in- event raised surface air temperatures by 7 ◦C. Here, it cluding extremophiles, would be critically endangered is worth recalling that the superflares we consider are (Estrela & Valio 2017). approximately five orders of magnitude larger. Hence, We will now briefly summarize some of the effects it seems reasonable to conclude that the air tempera- that arise due to ozone depletion. Ozone depletion has ture would be subjected to a much higher increase (or been linked with the increased penetration of biologi- decrease). Although this rise (or fall) in temperature cally harmful UVB radiation (Kerr & McElroy 1993), would be transient, we hypothesize that this could have an environmental stressor that leads to mutagenesis, re- a highly detrimental effect on most complex organisms duced fertility, suppression of physiological processes, for several reasons. and even death (Vincent & Roy 1993; Dahms & Lee First, we observe that most organisms have an opti- 2010). Recent research suggests that the primary in- mal body temperature at which they function. If the fluences of UVB radiation on life are likely to be mani- temperature exceeds this value by a non-trivial amount, fested at trophic levels - moving the focus away from in- there is a sharp decline in biochemical and physiologi- dividual organisms and species - through alterations of cal processes, ultimately leading to protein denaturation biogeochemical and climate cycles (Charlson et al. 1987; (Schulte 2015). Another factor that is even more impor- Häder et al. 2007; Zepp et al. 2011). For instance, UVB tant than the rise in temperature is the timescale over radiation may indirectly cause a reduction in carbon which it occurs. If the spike in temperature is sharp, the dioxide absorption, or a decline in the quantity/quality organism’s thermal adaption breaks down (Angilletta of nutrient cycling in marine food webs (Häder et al. 2009). Hence, it appears reasonable to conclude that 2015). the metabolic functioning of most organisms would be As a specific example, we point out that enhancement impaired, perhaps irreversibly, when subjected to a su- of UV radiation leads to a reduction in phytoplank- perflare. As temperature regulates a wide array of eco- ton photosynthesis (Cullen & Neale 1994; Day & Neale logical and evolutionary properties (Brown et al. 2004; 2002) and causes DNA damage (Malloy et al. 1997; Lingam & Loeb 2017a), we anticipate that an abrupt Castenholz & Garcia-Pichel 2012). The ozone deple- increase in temperature would severely impact the sta- tion in the Antarctic has been linked with a & 10% de- bility and functioning of ecosystems. Rapid fluctuations cline in the productivity of phytoplankton (Smith et al. along these lines have been posited as major causes be- 1992). Any such decline would have crucial effects on hind the ongoing Holocene extinction (Barnosky et al. marine ecosystems since phytoplankton are responsible 2012). It is therefore reasonable to surmise that past for 50% of the planet’s primary production (Field et al. mass extinctions could also have featured elevated tem- 1998). In addition, phytoplankton play a critical role peratures (Kiehl & Shields 2005). in regulating biogeochemical cycles, climate variations, Nitric acid rain is generated through the reaction of ni- biomass production, and the diversity, abundance and trogen dioxide (NO2) with the hydroxyl group (Crutzen functioning of marine ecosystems (Charlson et al. 1987; 1979; Toon et al. 1986). The ensuing nitrogen pollu- Sabine et al. 2004; Boyce et al. 2010). tion of aquatic ecosystems leads to a multitude of issues Any changes in plankton productivity will cause rip- including acidification, increased toxicity, and eutroph- ple effects that extend to different trophic levels, and ication (Camargo & Alonso 2006). On the other hand, thus alter the overall ecosystem response to UVB ra- since it can lead to a proliferation of primary producers, diation (Bothwell et al. 1994) as discussed in the pre- it may enable ecosystems to rebound after the initial de- ceding paragraph. A decline in phytoplankton could, structive phase. Similarly, it has been argued that NO2 in principle, disrupt the biological pump and lead to can reduce solar irradiance and cause large-scale glacia- increased CO2 levels in the atmosphere and an associ- tion (Reid et al. 1978). However, recent studies indicate ated rise in temperature via the greenhouse effect. We that this phase is transient and accompanied by a subse- 7 quent increase in solar irradiance (Thomas et al. 2015). distinguish between natural (for e.g. flares) and anthro- As some of the mass extinction events appear to have pogenic deposition of nitrates, since the latter has be- been followed by an upsurge in species diversification come increasingly important (Mayewski et al. 1986). (Benton 2009; Knoll 2015), factors with dual character- There has been some controversy as to whether SEP istics, like the ones identified above, might have played events are large enough to account for the observed an important role. peaks. For instance, even the highly energetic Car- Before proceeding further, we wish to highlight a cou- rington event has not left widespread traces in polar ple of self-evident, but nonetheless highly important, ice (Wolff et al. 2012). Several authors have there- conceptual points: ecosystems, as well as the biosphere, fore concluded that solar proton events in the Holocene are intrinsically nonlinear. The study of nonlinear dy- ∼ 104 yrs are not detectable, implying that nitrate namical systems has revealed the significance of “tip- spikes are not accurate proxies for these phenomena ping points”, i.e. states wherein infinitesimal pertur- (Wolff et al. 2012; Duderstadt et al. 2016). In contrast, bations can give rise to critical transitions leading to it has recently been suggested that hard-spectrum SEP qualitative changes (Lenton et al. 2008). Hence, even events can be unambiguously identified (Smart et al. when considering cases where superflares give rise to 2014; Melott et al. 2016) in the ice core record. As the only “minimal” changes in the environment, Earth’s cli- superflares considered in our work are much stronger mate and biosphere may respond in a nonlinear manner than those recorded in modern history, the level of (Scheffer et al. 2001), thereby possibly leading to the on- nitrate deposition due to the associated solar proton set of a mass extinction event. events should be much higher; on the other hand, their In the same spirit, we advocate that astrophysical age (∼ 10 Myr) and the accompanying erosion may ren- causes should not be viewed in isolation, as they are der this method non-viable. more effective when acting in tandem with geological From a long-term standpoint, it seems probable that phenomena, for e.g. geomagnetic field reversals, volcan- the use of isotope-based estimates represents a more ism and ocean circulation patterns. The Earth’s mag- promising endeavour. In particular, it has been sug- netic field is significantly reduced during the reversal gested that SEPs can produce 10Be and 14C in the at- process (Merrill et al. 1998), and an SEP event of lower mosphere, and that high-resolution isotope data may magnitude occurring during this period will therefore yield signatures of such events (Usoskin et al. 2006; be capable of causing the same degree of devastation Beer et al. 2013). A rapid increase in the 14C content (Reid et al. 1976; Raup 1985). Thus, superflares may of tree rings in Japan (Miyake et al. 2012) has been in- constitute one half (the impulse) of the proposed “press- voked as evidence in favour of a high-fluence solar proton pulse” mechanism for mass extinctions (Arens & West event in AD 775 (Melott & Thomas 2012; Usoskin et al. 2008).1 In turn, this could lead to mass extinction events 2013; Mekhaldi et al. 2015). Thus, if abrupt features are that display a superposition of stochasticity and peri- present in measurements of cosmogenic radionuclides at odicity; such patterns have been predicted to be duly the same period as one of the mass extinction events, manifested in the fossil record (Feulner 2011). they would lend credence to the hypothesis that superflares played a role in triggering species extinction; note 3.3. Signatures of solar superflares that these spikes must also be consistent with a solar proton event. In order to evaluate this conjecture, high- Having outlined the consequences arising from a mas- resolution data pertaining to these radionuclides should sive superflare, it is now instructive to ask whether such be collected from both terrestrial and lunar rocks. flares can be deduced from the geological record. We conclude by observing that evidence from radionu- As noted previously, one of the consequences of su- clides and ice cores in favor of superflares must be in- perflares is that they can give rise to large-scale SEP terpreted with due caution. One must identify potential events that promote the production of nitrogen oxide “false positives” that are capable of producing the same compounds. The most widely proposed method entails signatures as solar superflares, and may therefore be the use of ice cores in Greenland or Antarctica as a proxy mistaken for the latter. It is safe, however, to argue that for solar activity (Stothers 1980). The basic idea is that the aforementioned radiochemical evidence does not un- there exists a correspondence between nitrate concen- equivocally eliminate the possibility of solar flares.2 trations in the ice cores and flares; short-term nitrate features (spikes) can, in principle, reflect solar proton events (Legrand & Kirchner 1990; Dreschhoff & Zeller 3.4. Imprints in the fossil diversity record 1998; McCracken et al. 2001). However, when consid- We begin by observing that the putative causes for the ering studies reliant on this method, it is important to ‘Big Five’ mass extinctions have been quite thoroughly

1 In many cases, however, the distinction between “press” and 2 The identification of false positives also constitutes an impor- “pulse” is not readily apparent, and both give rise to a wide range tant component in the analogous field of detecting biosignatures of macroevolutionary responses (Grant et al. 2017). on exoplanets (Kaltenegger 2017). 8 documented (Hallam & Wignall 1997; Courtillot 1999; We will now explore some of the implications that Bambach 2006). Hence, the chances of superflares caus- superflares would have for life-as-we-know-it on other ing these particular extinction events are most likely planets (and moons). minimal. However, as noted in the previous sections, some of the observed extinction events with a periodic- 4.1. Implications for Mars and Venus ity of 26 Myr may have been caused by superflares, often Here, we shall consider only present-day Mars and acting in conjunction with other natural causes. We will Venus, and return to ancient Mars and Venus at a later therefore outline certain distinctive features that might stage. If seen purely from the viewpoint of energetics, characterize extinctions where superflares played a role. it may appear as though the UV and particle energies 1. Ozone depletion is predicted to increase (up to deposited on Mars (Venus) are only a factor of 2 lower a factor of 2-3) as one moves to higher latitudes (higher) than those deposited on Earth, provided that (Thomas et al. 2007). In light of the deleterious the emission is isotropic with an inverse-square law. consequences of ozone depletion outlined in Sec. In reality, the scenario is more complicated on ac- 3.1, we may expect the extinction probability to count of the fact that Mars has a very tenuous atmo- increase with latitude. We also note that the tem- sphere - the surface pressure and column density are perature spike described in Sec. 3.2 is likely to about two orders of magnitude lower (Owen 1992)- be more pronounced at higher latitudes. In con- and weak (crustal) magnetic fields (Acuna et al. 1998). trast, the opposite (extinction probability) trend Both of these factors have been identified as major ob- has been predicted to occur for extinction driven stacles in protecting the surface from Galactic Cosmic by current climate change (Thomas et al. 2004). Rays (GCRs). For instance, while considering exoplanets around low-mass stars, Grießmeier et al. (2015) con- 2. Organisms that are subterranean or those that cluded that the presence of a weak (or zero) magnetic dwell below the euphotic zone should be rela- field leads to an energetic particle flux that is more than tively protected from UVB radiation. Hence, the three orders of magnitude higher than on Earth. Hence, preferential extinction of surface-dwelling (land or strong magnetospheric shielding is necessary, especially aquatic) organisms could be a consequence of su- for planets with rarefied atmospheres, to prevent ele- perflares, unless they are equipped with screening vated surface radiation levels (Grießmeier et al. 2016). compounds (Cockell & Knowland 1999). As Mars possesses these characteristics, it seems reasonable to conclude extremely large superflares of the 3. The photosynthetic productivity of phytoplank- kind discussed herein would prove to be highly detri- ton has been argued to be more sensitive to UVB mental, and possibly fatal, to any life on the planet. levels when compared to terrestrial plants due to Several studies have attempted to trace the evolution its less effective screening (Day & Neale 2002). of Martian habitability over time (Cockell et al. 2000; Given the importance of the former in oceanic Fairén et al. 2010; Cockell 2014), and identify regions ecosystems (Field et al. 1998), we conjecture that where life could have persisted (Boston et al. 1992; aquatic species would be rendered more vulnerable Davila & Schulze-Makuch 2016). After the discovery to extinction when compared to terrestrial organ- of superflares in G-type stars (Maehara et al. 2012), de- isms. tailed studies of Martian habitability when subjected 4. Nitric acid rain and the production of nitrogen to such an event do not seem to have been undertaken. oxides are some of the outcomes that may result Although the radiation doses are unlikely to drive all from superflares, as discussed in Sec. 3.2. Their Martian lifeforms to extinction, any survivors would ensuing consequences have already been investi- have evolved a high radiation tolerance, akin to organ- gated in the context of the P-Tr and K-Pg extinc- isms like Thermococcus gammatolerans (Jolivet et al. tions (Prinn & Fegley 1987; Zahnle 1990), and in- 2003), Deinococcus radiodurans (Cox & Battista 2005) clude photosynthesis inhibition, toxicosis, foliage and Milnesium tardigradum (Horikawa et al. 2006). and respiratory damage. Invertebrates in freshwa- The situation for Venus is quite different. In this case, ter ecosystems are particularly vulnerable to acid- studying the role of surface ionizing radiation is quite ification (Schindler 1988), and might therefore be irrelevant since the temperature (at 740 K) is far too preferentially subject to extinction compared to extreme to host life-as-we-know-it. Instead, proposals their saltwater counterparts. for putative Venusian life have focused on sulphur-based chemoautotrophs situated in the clouds (Morowitz 1967; The above list is not meant to be definitive, but it can Cockell 1999; Schulze-Makuch et al. 2004). The effects potentially serve as a preliminary guide for locating ex- of the AD 775 SEP event on the Venusian atmosphere tinction events mediated by superflares. were studied by Dartnell et al. (2015), and it was concluded that the radiation dose was insufficient to cause 4. IMPLICATIONS OF SUPERFLARES FOR LIFE damage, although the atmospheric chemistry was af- ELSEWHERE IN THE UNIVERSE fected by the strong ionization. By using the scaling re- 9 lation (5), we conclude that SEP fluence would be nearly flares have hitherto assumed a maximum flare energy of four orders of magnitude higher for the superflares dis- . 1036 erg. However, as per our discussion in this paper, cussed in this paper when compared to the AD 775 the existence of more energetic superflares ought not be event. Hence, it seems likely that airborne ecosystems ruled out. M-dwarfs have a smaller surface area com- on Venus would be subject to high extinction risks when pared to G-type stars, and thus smaller (maximal) spot these superflares occur due to a much higher degree of sizes, but they also have higher surface magnetic fields ionization and greatly enhanced radiation dosage. (Morin et al. 2010), and may therefore still be capable In addition, a wide range of objects in our Solar sys- of generating large superflares. tem have been proposed as sites where life could exist Apart from the greatly enhanced radiation doses re- (Schulze-Makuch & Irwin 2006; Shapiro & Schulze-Makuch ceived on these planets, we note that superflares with 2009); the list includes well-known candidates like Ti- ∼ 1035 erg have an occurrence rate that is 20 and 5 tan, Europa and Enceladus, but also more exotic op- times higher for M- and K-dwarfs respectively compared tions such as Jupiter (Sagan & Salpeter 1976), asteroids to G-type stars (Maehara et al. 2012; Candelaresi et al. and comets (Clark et al. 1999). For subsurface environ- 2014). Hence, complex life on exoplanets around M- ments, we do not anticipate that superflares would play dwarfs could be subject to repeated extinction events on a major role provided that the crust is sufficiently thick; the timescale of O(1) Myr. In between these events, we see, however, Dartnell (2011). Atmospheric ecosystems, observe that smaller superflares occur at regular inter- on the other hand, are likely to be significantly per- vals, potentially lowering the chances for the biosphere turbed by major superflares along the lines described to repair itself. However, in light of the enhanced mu- earlier. tations and selection pressure induced by flares (Sagan 1961, 1973; Cockell 1998; Dartnell 2011), the periods 4.2. Implications for life on exoplanets around M- and in between these extinctions may witness rapid specia- K-dwarfs tion. Superflares could therefore be responsible for pe- We begin by observing that planets situated in the riodically varying diversification and extinction rates. habitable zone (HZ), the region theoretically capable of Thus, it seems plausible that short bursts of extinc- supporting liquid water, around M-dwarfs are charac- tion and speciation (Smith et al. 2004) might be inter- terized by two distinct and highly important properties: spersed with long periods of stasis; the suggested pat- (i) they are situated very close to the host star, and (ii) tern is somewhat reminiscent of punctuated equilibrium the host stars are very active (Scalo et al. 2007). (Eldredge & Gould 1972; Gould & Eldredge 1993). A combination of these two factors is responsible Let us now suppose that we consider the idealized sce- for ensuring that the atmospheres of M-dwarfs are nario where all of the energy from the superflare impacts rapidly stripped away through a combination of thermal the surface of an Earth-sized planet orbiting a low-mass and non-thermal escape processes (Dong et al. 2017; M-dwarf. The energy deposited Ep is estimated by uti- Lingam & Loeb 2017a; Garcia-Sage et al. 2017). Simu- lizing (3) and we will choose a ∼ 0.01 AU for the sake lations have illustrated that these mechanisms lead to of convenience; this value is somewhat close to the or- atmospheric depletion over < 1 Gyr timescales in the bital radii of Proxima b (Anglada-Escudéet al. 2016) absence of outgassing. Most of the exoplanets in the and the TRAPPIST-1 planets (Gillon et al. 2017). We 31 HZ of M-dwarfs are tidally locked, and have weak mag- find that Ep ∼ 4.5 × 10 erg for isotropic emission, and 33 netic moments (Khodachenko et al. 2007; Zuluaga et al. Ep ∼ 4.5 × 10 erg for non-isotropic emission with an ◦ 2013). On account of these two reasons, the shielding opening angle of 24 . We ask the question: what is the against coronal mass ejections, SEPs and GCRs is ex- mass M that will be raised to the boiling point of water? pected to be much lower (Grießmeier et al. 2005, 2009; It is computed via Vidotto et al. 2013; Grießmeier et al. 2015; Kay et al. E M = p , (7) 2016). Consequently, the biological hazards will be C heightened for planets in the HZ of M-dwarfs. For ex- ∆T oplanets orbiting active M-dwarfs, it has been shown where ∆T ∼ 100 K and C is the specific heat capacity of recently that the levels of surface UVB and UVC radia- water. In reality, note that all of the energy impacting tion (due to flares) would be lethal to most lifeforms on the planet will not be delivered to the surface, and the Earth (O’Malley-James & Kaltenegger 2017). value of ∆T < 100 K. With these values, we find that Atri (2017) studied the effects of solar proton events M ∼ 1019 kg for isotropic emission, and M ∼ 1021 kg on a wide range of Earth-analogs with varying orbital for the non-isotropic case. This leads us to the remark- distances, magnetic moments, atmospheric column den- able conclusion that, for the latter situation, a superflare sities and flare energies. For the most extreme case(s), of 1037 erg is capable of evaporating the oceans on this the radiation doses on the surface were demonstrated to planet provided that their total mass is comparable to be ∼ 104 Sv. In comparison, a dose of ∼ 100 Sv is lethal that of Earth’s oceans. Thus, in terms of an existential to most mammals and birds, and certain insects. Most threat, it should be placed in the same category as aster- studies concerning the biological ramifications of super- oids, GRBs and supernovae (Sloan et al. 2017) although 10 its frequency of occurrence is much higher. Even though K-dwarf? We suggest that superflares might represent a single superflare will not suffice to wholly evaporate a missing piece of the puzzle: their impact on exoplan- the oceans, a few of them, spanning a total of O(100) ets in the HZ of K-dwarfs is more profound, and these Myr, should be enough to dessicate a planet in the HZ events occur more frequently (by a factor of 5). Thus, of a low-mass M-dwarf. when all of these factors are taken into consideration, If we consider the isotropic emission case, the ramifi- our position around a G-type star may not be a fortu- cations are still severe, albeit not so dramatic. The eu- itous accident, but a fairly probable event instead. photic zone, the region where photosynthesis occurs and If a large fraction of M- and K-dwarfs are unsuited most of the marine life is situated, would be completely to host complex life on planets orbiting them, this evaporated, and the same fate would befall the rest of still leaves G-type stars. However, even in this cat- the pelagic zone. Even on Earth, the non-isotropic sce- egory, we note that a small, but non-trivial, fraction nario is capable of raising the temperature of the photic of them display evidence of regular superflare activity zone by a few degrees and could disrupt biogeochemical (Maehara et al. 2012; Shibayama et al. 2013). One may mechanisms and giving rise to outcomes like euxinia; thus be tempted to conclude that complex life is rare the latter is believed to have played an important role in the Universe, although simple microbial life could be in regulating ocean diversity over time (Meyer & Kump quite common. This line of reasoning has been advo- 2008), and in the Permian-Triassic mass extinction event cated by several authors in the past and is referred to as (Grice et al. 2005). the “Rare Earth” hypothesis (Ward & Brownlee 2000). Thus, to summarize, the prospects for complex life on However, we wish to caution that our study does not exoplanets in the HZ around M-dwarfs are severely ham- necessarily direct us to this conclusion since the fraction pered due to a multitude of reasons. The degree of ozone of G-type stars exhibiting unusual superflare activity is depletion and the radiation dosage received are likely to known to be small, and there exist considerable statisti- be much higher than those on Earth. Superflares occur cal uncertainties regarding the frequency and magnitude with a higher frequency on M-dwarfs and are thus more of superflare events on M-, K- and G-type stars. likely to give rise to frequent extinction events. Lastly, they could deposit enough energy into the oceans to boil 4.3. Risks to human civilization from superflares them completely or partially, and thereby severely im- Ever since the discovery of superflares in G-type stars, pact the growth and development of marine life. Al- several studies have briefly alluded to the risk to hu- though our discussion was oriented towards exoplanets, man civilization from such an event (Shibata et al. 2013; many of these considerations would be applicable to ex- Karoff et al. 2016). However, detailed analyses of the omoons in the HZ (Heller et al. 2014) as well. Some threats posed by a large superflare to technological civi- of the general conclusions regarding planets orbiting M- lizations (such as ours) do not appear to have been un- dwarfs are applicable to K-dwarfs to a lesser degree, as dertaken thus far (Lingam & Loeb 2017c). the latter fall between M- and G-type stars in terms of In Sec. 2, we presented data favoring the recurrence most of their properties.3 of a ∼ 1034 erg superflare every ∼ 2000 yrs. More- Collectively, these facts pave the way towards answer- over, superflares with energies of approximately 1035 ing a fundamental question delineated in Loeb et al. erg, 1036 erg and 1037 erg would occur with frequen- (2016): why is it that we orbit a G-type star in the cies of ∼ 40 Kyr, ∼ 800 Kyr and ∼ 20 Myr respec- present epoch and not an M-dwarf in the cosmic fu- tively. Even though superflares with relatively lower en- ture? This question was further studied through the ergies will cause negligible biological damage, they are use of Bayesian inference methods (Haqq-Misra et al. capable of causing tremendous destruction to human 2017). One approach to resolve this apparent para- civilization. Hence, it is imperative to constrain (and dox is by identifying reasons why life around M-dwarfs eventually predict) the frequencies with which these su- is selectively suppressed. Through considerations of perflares can occur on the Sun. The first step entails biodiversity, Lingam & Loeb (2017b) recently argued undertaking a thorough scrutiny of historical records that low-mass M-dwarfs are unsuitable for life-as-we- for evidence of large-scale aurorae and sunspots that know-it, implying that K- and G-type stars repre- could be indirectly associated with superflares. Al- sent the best chances for hosting complex biospheres though some studies along these lines have been un- (Heller & Armstrong 2014; Cuntz & Guinan 2016). Al- dertaken recently (Vaquero 2007; Hayakawa et al. 2015, though this conclusion ameliorates the problem, it does 2017; Tamazawa et al. 2017), a much higher degree of not fully solve it since we are left with the equivalent attention to this topic appears to be warranted. question: why do we orbit a G-type star and not a We also note that several studies have attempted to forecast the course of space weather over the next

3 few centuries (Barnard et al. 2011; Lockwood 2012; Since superflares are known to exist even on L-dwarfs Steinhilber & Beer 2013; Ineson et al. 2015) but most (Schmidt et al. 2016), we anticipate that our findings would also be valid to some degree to planets orbiting such stars. of them have focused on making predictions over short timescales, i.e. for the next 10 − 100 years. For in- 11 stance, models indicate that a Carrington-like event and modelling of SEP events was identified as one of the has a relatively high (. 10%) chance of occurring in highest priorities. We also note that solar proton events the next decade (Showstack 2011; Riley 2012; Kataoka damage the atmosphere by inducing chemical changes, 2013). The emphasis on short timescales is motivated disrupting climate feedback mechanisms, triggering elec- primarily by pragmatic considerations since the inher- trical discharges and altering the formation of clouds ent solar variability does not enable accurate forecasting (Gray et al. 2010; Solanki et al. 2013; Mironova et al. over longer epochs. However, as we shall argue below, 2015). Each of these environmental changes will, in there is a pressing need to take longer timescales into turn, also lead to concomitant ecological, social and eco- account. nomic losses that are likely to be quite significant. We begin by observing that the manifold impacts Although the scaling between economic losses and the of relatively moderate (in comparison to superflares) magnitude of catastrophes will not be linear, it is still space weather events have been thoroughly documented instructive to evaluate the energy of a superflare that (Schwenn 2006; Space Studies Board 2009; Hapgood would lead to damage equal to that of the world’s Gross 2011; Schrijver et al. 2015; Eastwood et al. 2017). Coro- Domestic Product (GDP). Using the values for the Car- nal mass ejections, typically associated with flares, give rington event described above, and the world’s current rise to powerful geomagnetic storms capable of signif- GDP,5 we find that the resultant value is ∼ 1034 erg. A icantly disrupting the planet’s magnetosphere (Kahler superflare with this energy could occur on the Sun once 1992; Webb & Howard 2012). Geomagnetic storms in- every ∼ 2000 years. If we further assume that the AD duce large electric fields and currents, which can severely 775 event was a superflare of this magnitude, we are led disrupt a wide range of electrical systems (Boteler et al. to the conclusion that the next such event might take 1998; Pirjola 2000; Pulkkinen 2007). A superflare place ∼ 750 years in the future. However, as noted in may also generate an electromagnetic pulses (EMP) Secs. 2.1 and 2.2, our understanding of solar superflares due to the abrupt ionization of the planet’s dayside is both rudimentary and based on statistical evidence. atmosphere, somewhat akin to the effects of a nu- Hence, the wait time of ∼ 750 years proposed above clear weapon (Glasstone & Dolan 1977; Longmire 1978; must be viewed with caveats. Volland 1984). Detailed calculations pertaining to these Based on the frequency of superflares outlined ear- processes, and the ensuing consequences for technologi- lier, we surmise that a event with energy ∼ 1036 erg cal civilizations, are beyond the scope of this paper. has a ∼ 10−4 chance of occurring in the next cen- The Carrington 1859 flare has garnered much atten- tury. As noted in Sec. 3.1, a superflare of this mag- tion since it represents a valuable benchmark against nitude may be sufficient to cause total ozone deple- which extreme space weather events can be measured. tion and lead to major ecological damage. In compar- In 1859, the Carrington flare caused the disruption of ison, the likelihood of a 2 km asteroid or comet hit- telegraph services (Boteler 2006), but the same event ting the Earth in the same period has been estimated to would lead to far more destructive effects in the current be 10−4 (Chapman & Morrison 1994) and would result era. For starters, we note that the worldwide disruption in widespread destruction (Toon et al. 1997). Hence, of power grids would lead to considerable economic dam- both of these events represent genuine hazards, and have age. The losses for the US alone have been documented a similar likelihood of occurring in the next century. to be ∼ 2 trillion dollars (Space Studies Board 2009). In However, despite the similar (or greater) dangers posed addition, breakdowns in satellite communications, nav- by superflares, asteroid and comet impacts have been igation and surveillance are anticipated. The total eco- subjected to detailed risk analyses (Posner 2004; Smith nomic losses have been estimated to be ∼ 70 billion dol- 2013). NASA has also put together extensive plans en- lars, and about 10% of the existing satellites orbiting the tailing the close monitoring of near-Earth objects, and Earth would be destroyed (Odenwald et al. 2006). More deflecting them if necessary; the Panoramic Survey Tele- devastatingly, Schulte in den Bäumen et al. (2014) con- scope and Rapid Response System (Pan-STARRS) mis- cluded that the long-term disruptions of global supply sion merits a mention in this regard.6. The total cost of chains due to extreme space weather events would lead a system to detect and deflect near-Earth objects is be- to losses worth 3.4 trillion dollars. Assuming these esti- tween 1 − 10 billion dollars.7 Although this discrepancy mates are correct, the resulting impact would be equiv- is partly explained by the recent discovery of superflares alent to the cumulative effects of anthropogenic climate in G-type stars, it can also be attributed to “anthropic change over a period of several decades.4 shadow” or cognitive biases that lead to underestima- In addition, we note that the SEPs produced during tion of risks posed by certain catastrophes (Yudkowsky extreme space weather events constitute a major hazard 2008; Cirkovićet´ al. 2010). to any space-based operations. Hence, in the roadmap of Schrijver et al. (2015), the need for further observations 5 http://data.worldbank.org/indicator/NY.GDP.MKTP.CD?year_high_desc=true 6 https://panstarrs.stsci.edu/ 4 http://www.ghf-ge.org/human-impact-report.pdf 7 https://www.nasa.gov/pdf/171331main_NEO_report_march07.pdf 12

We end our analysis with a brief comment on the It is, however, the latter compound that has at- well-known “Doomsday argument” which relies on prob- tracted a great deal of attention in recent times. In abilistic considerations to arrive at the total number of studies concerning the origin of life, the ‘RNA world’ human beings who will exist in the future. Gott (1993) hypothesis has been extensively investigated (Joyce undertook a famous analysis that led to an estimate of 2002; Ruiz-Mirazo et al. 2014). The formation of ac- the future lifetime of humans; the value ranged from tivated ribonucleotides, which undergo polymerization ∼ 5 × 103 years to ∼ 8 × 106 years. Although Gott’s to yield RNA, is difficult for a number of reasons. How- analysis has been critiqued by several authors (Bostrom ever, it was shown by Powner et al. (2009) that a mix- 2002), we can use these numbers to estimate the cor- ture of chemical compounds, including cyanamide and responding magnitude of the solar superflares by using cyanoacetylene, led to the synthesis of pyrimidine ri- (1). We find that the superflare energies must lie be- bonucleotides under conditions resembling the early tween 2 × 1034 erg and 6 × 1036 erg; based on the argu- Earth. Subsequently, Patel et al. (2015) demonstrated ments provided in this paper, the latter value is capable that the forerunners of the building blocks for protocells of causing a mass-extinction event. - nucleic acids, proteins, and lipids - could have arisen through the homologation of hydrogen cyanide and its derivatives. Hydrogen cyanide has therefore been iden- 4.4. Some positive implications of superflares tified as a putative ‘feedstock’ molecule which played Hitherto, we have restricted ourselves to exploring the a pivotal role in the origin of life (Saladino et al. 2012; negative consequences arising from superflares. How- Sutherland 2016). ever, as briefly noted in Sec. 4.2, these flares stimulate We note that the synthesis of these compounds mutations and thereby lead to bursts of rapid species di- need not have occurred on Earth since early Mars versification - a factor that may have been particularly (Wordsworth 2016) and Venus (Way et al. 2016) were important during the Archean era. also potential sites of prebiotic synthesis (Cockell Even if one supposes that solar superflares gave rise 2000). Asteroids and comets could have facilitated to mass extinctions, polyextremophiles like Deinococ- the exogenous delivery of prebiotic compounds to cus radiodurans would be easily able to survive such Earth by means of quasi-panspermia (Chyba & Sagan episodes. Moreover, the remarkable discovery of Desul- 1992; Thomas et al. 2006). Looking further afield, forudis audaxviator (Chivian et al. 2008), a sulfur- we anticipate that planets orbiting M-dwarfs, and K- reducing chemoautotroph, has revealed that species on dwarfs to a lesser extent, would be more conducive Earth and elsewhere can derive energy from radioactive to exogenous delivery mechanisms primarily on ac- sources for sustenance. Hence, even in high-radiation count of the shorter interplanetary distances involved environments that would result from superflares, a fair (Lingam & Loeb 2017d). number of species may possess UV radiation screen- Lastly, flares lead to elevated levels of UV radi- ing and DNA repair mechanisms (Cockell & Knowland ation, and have thus been invoked as a means of 1999), and prove to be adaptable enough to survive. ameliorating the UV deficiency (Buccino et al. 2007; In addition, certain habitats, associated with reduced Rugheimer et al. 2015) on planets around M-dwarfs. It levels of UV radiation, ought to be conducive to the sus- was noted in Ranjan et al. (2017) that UV-sensitive pre- tained existence of photoautotrophs (Cleaves & Miller biotic chemistry pathways could be functional over the 1998; Cockell & Raven 2004). duration of the flare, and become inactive during the Superflares could have played a beneficial and impor- quiescent phase. These findings may be valid to some tant role during the Hadean era through a number of degree for superflares, although they have a much lower channels. SEPs, as well as GCRs, have the capacity to frequency of occurrence. Additionally, it seems quite catalyze cloud formation (Kirkby 2007), the generation plausible that these extreme phenomena can adversely of strong electric fields, significant bursts of energetic impact the synthesis of prebiotic compounds once a radiation, and lightning (Dwyer 2003; Dwyer & Uman certain threshold value of the energy (and UV flux) is 2014); the last has important biological consequences exceeded. (Erlykin & Wolfendale 2010), given its relevance in prebiotic chemistry (Chyba & Sagan 1992). Further- 5. CONCLUSIONS more, SEPs impacting the Earth during such events Ever since the discovery of superflares on solar-type would have enabled a network of chemical reactions stars, there has been much interest in exploring the en- (Ehrenfreund et al. 2002), ultimately culminating in suing consequences of such events on Earth and other the formation of nitrous oxide and hydrogen cyanide exoplanets. We began our analysis by proposing that (Airapetian et al. 2016). The former’s importance stems superflares with energies . 1037 erg are potentially ca- from the fact that it is a highly efficient greenhouse gas pable of occurring on the Sun. The associated timescale that could have warmed Earth’s atmosphere, thereby of recurrence was found to be ∼ 20 Myr, a value that providing a potential resolution for the long-standing coincided with the periodic extinction timescale of 26 faint young Sun paradox (Sagan & Mullen 1972). Myr deduced by some authors from the fossil record 13

(Raup & Sepkoski 1984). This fact motivated us to ex- comet impacts. We reviewed the literature on the eco- plore the environmental and biological ramifications of nomic damage wrought by superflares due to the dis- ∼ 1036 − 1037 erg superflares on Earth. In addition, we ruption of power grids, breakdown in communications also specified the assumptions, caveats and uncertainties and supply chains (Pulkkinen 2007; Hapgood 2011). We associated with our analysis in Sec. 2.2. hypothesized that the overall losses could exceed the We concluded that a superflare of this magnitude world’s current GDP for certain superflares, and that could cause destruction of the ozone layer, thereby lead- an event of this magnitude has a very high chance of ing to widespread damage to ecosystems, and possibly transpiring during this millennium. triggering a mass extinction. In addition, the air sur- We completed our analysis of superflares by observing face temperature could rise abruptly by a considerable that, in certain instances, they can also lead to beneficial amount, damaging the metabolic functioning of biota outcomes. During the Hadean and Eoarchean eons on because of a breakdown in thermal adaptation. We Earth, when the Sun was much more active, superflares also raised the important point that small environmen- may have been an important factor in catalyzing the tal perturbations could lead to far-reaching implications origin of life (Ehrenfreund et al. 2002) and warming the for ecosystems due to nonlinear processes (Lenton et al. planet by inducing a greenhouse effect (Airapetian et al. 2008). We also suggested that superflares may have 2016). With regards to the former phenomenon, super- acted in concert with geological mechanisms giving rise flares could have played a critical role in the synthesis to extinction events that were neither wholly stochastic of hydrogen cyanide, a vital chemical compound that is nor periodic. Evidence for extreme superflares may ex- capable of giving rise to the precursors of proteins, lipids ist in the form of nitrate spikes in ice cores, anomalously and nucleic acids under prebiotic conditions. high concentrations of certain cosmogenic isotopes on Superflares ought to have therefore played a major Earth (Miyake et al. 2013), and perhaps directly in the role in shaping the evolutionary history of the Earth fossil extinction record. and other habitable exoplanets. They may have con- We followed our discussion by examining some of stituted an essential energy source in the synthesis of the implications for extraterrestrial life. We inferred prebiotic compounds, and thereby enabling abiogenesis. that present-day Mars and Venus are more suscepti- On the other hand, they could also have triggered quasi- ble to damage from superflares as they lack an intrin- periodic extinction events, although, in all probability, sic magnetic field or a thick atmosphere. We also con- not the ‘Big Five’ mass extinctions. Intriguingly, super- sidered exoplanets orbiting M-dwarfs and outlined why flares might serve as putative mechanisms by which the the prospects for complex life on these planets are typ- likelihood of life on planets around M- and K-dwarfs ically lowered compared to G-type stars. A combina- is selectively lowered compared to G-type stars like the tion of factors including weak magnetic moments, close Sun. Thus, they provide a potential explanation as to distances to the host star, extensive atmospheric strip- why we, Homo sapiens, have found ourselves dwelling ping, and enhanced frequency of superflare events are on a planet orbiting the Sun instead of one that is situ- all responsible for making environments around these ated in the habitable zone of an M-dwarf. stars hostile to life-as-we-know-it (Maehara et al. 2012; Lingam & Loeb 2017a). We also showed that, especially for planets orbiting low-mass M-dwarfs like Prox- We thank Andrew Knoll, John Raymond, Luca ima Centauri and TRAPPIST-1, a significant fraction of Comisso, Chuanfei Dong and Russell Kulsrud for their the oceans can be evaporated over sub-Gyr timescales valuable comments and suggestions concerning the pa- due to highly energetic superflares. per. This work was partly supported by grants from Although superflares are likely to pose a genuine the Breakthrough Prize Foundation for the Starshot threat to human civilization, their importance has not Initiative and Harvard University’s Faculty of Arts and been taken seriously in comparison to the likelihood of Sciences, and by the Institute for Theory and Compu- other astronomical catastrophes, for e.g. asteroid and tation (ITC) at Harvard University.

REFERENCES Acuna, M. H., Connerney, J. E. P., Wasilewski, P., et al. Anglada-Escudé, G., Amado, P. J., Barnes, J., et al. 2016, 1998, Science, 279, 1676 Nature, 536, 437 Airapetian, V. S., Glocer, A., Gronoff, G., Hébrard, E., & Arens, N. C., & West, I. D. 2008, Paleobiology, 34, 456 Danchi, W. 2016, Nat. Geosci., 9, 452 Atri, D. 2017, Mon. Not. R. Astron. Soc. Lett., 465, L34 Alroy, J. 2008, Proc. Natl. Acad. Sci. USA, 105, 11536 Angilletta, M. J. 2009, Thermal Adaptation: A Theoretical Atri, D., & Melott, A. L. 2014, Astropart. Phys., 53, 186 and Empirical Synthesis (Oxford Univ. Press) Bailer-Jones, C. A. L. 2009, Int. J. Astrobiol., 8, 213 14

Bambach, R. K. 2006, Annu. Rev. Earth Planet. Sci., 34, Cliver, E. W., & Dietrich, W. F. 2013, J. Space Weather 127 Space Clim., 3, A31 Barnard, L., Lockwood, M., Hapgood, M. A., et al. 2011, Cliver, E. W., & Svalgaard, L. 2004, Sol. Phys., 224, 407 Geophys. Res. Lett., 38, L16103 Cliver, E. W., Tylka, A. J., Dietrich, W. F., & Ling, A. G. Barnosky, A. D., Hadly, E. A., Bascompte, J., et al. 2012, 2014, Astrophys. J., 781, 32 Nature, 486, 52 Cockell, C. S. 1998, J. Theor. Biol., 193, 717 Basri, G., Walkowicz, L. M., Batalha, N., et al. 2011, —. 1999, Planet. Space Sci., 47, 1487 Astron. J., 141, 20 —. 2000, Orig. Life Evol. Biosph., 30, 467 Beer, J., McCracken, K. G., Abreu, J., Heikkilä, U., & —. 2014, Astrobiology, 14, 182 Steinhilber, F. 2013, Space Sci. Rev., 176, 89 Cockell, C. S., Catling, D. C., Davis, W. L., et al. 2000, Benton, M. J. 1995, Science, 268, 52 Icarus, 146, 343 —. 2009, Science, 323, 728 Cockell, C. S., & Knowland, J. 1999, Biol. Rev., 74, 311 Benz, A. O. 2017, Living Rev. Sol. Phys., 14, 2 Cockell, C. S., & Raven, J. A. 2004, Icarus, 169, 300 Boston, P. J., Ivanov, M. V., & McKay, C. P. 1992, Icarus, Comisso, L., Lingam, M., Huang, Y.-M., & Bhattacharjee, 95, 300 A. 2016, Phys. Plasmas, 23, 100702 Bostrom, N. 2002, Anthropic Bias: Observation Selection —. 2017, submitted to Astrophys. J., arXiv:1707.01862 Effects in Science and Philosophy (Routledge) Courtillot, V. 1999, Evolutionary Catastrophes: The Science of Mass Extinction (Cambridge Univ. Press) Boteler, D. H. 2006, Adv. Space Res., 38, 159 Cox, M. M., & Battista, J. R. 2005, Nature Rev. Boteler, D. H., Pirjola, R. J., & Nevanlinna, H. 1998, Adv. Microbiol., 3, 882 Space Res., 22, 17 Crutzen, P. J. 1979, Annu. Rev. Earth Planet. Sci., 7, 443 Bothwell, M. L., Sherbot, D. M. J., & Pollock, C. M. 1994, Crutzen, P. J., & Bruhl, C. 1996, Proc. Natl. Acad. Sci. Science, 265, 97 USA, 93, 1582 Boyce, D. G., Lewis, M. R., & Worm, B. 2010, Nature, 466, Cullen, J. J., & Neale, P. J. 1994, Photosynth. Res., 39, 303 591 Cuntz, M., & Guinan, E. F. 2016, Astrophys. J., 827, 79 Brown, J. H., Gillooly, J. F., Allen, A. P., Savage, V. M., & Dahms, H.-U., & Lee, J.-S. 2010, Aquatic Toxicology, 97, 3 West, G. B. 2004, Ecology, 85, 1771 Dartnell, L. R. 2011, Astrobiology, 11, 551 Buccino, A. P., Lemarchand, G. A., & Mauas, P. J. D. Dartnell, L. R., Nordheim, T. A., Patel, M. R., et al. 2015, 2006, Icarus, 183, 491 Icarus, 257, 396 —. 2007, Icarus, 192, 582 Davila, A. F., & Schulze-Makuch, D. 2016, Astrobiology, Calisto, M., Usoskin, I., & Rozanov, E. 2013, Environ. Res. 16, 159 Lett., 8, 045010 Day, T. A., & Neale, P. J. 2002, Annu. Rev. Ecol. Evol. Camargo, J. A., & Alonso, A. 2006, Environ. Int., 32, 831 Syst., 33, 371 Candelaresi, S., Hillier, A., Maehara, H., Brandenburg, A., Dong, C., Lingam, M., Ma, Y., & Cohen, O. 2017, & Shibata, K. 2014, Astrophys. J., 792, 67 Astrophys. J. Lett., 837, L26 Carrington, R. C. 1859, Mon. Not. R. Astron. Soc., 20, 13 Dreschhoff, G., & Zeller, E. J. 1998, Sol. Phys., 177, 365 Castenholz, R. W., & Garcia-Pichel, F. 2012, in Ecology of Duderstadt, K. A., Dibb, J. E., Schwadron, N. A., et al. Cyanobacteria II, ed. B. A. Whitton (Springer), 481–499 2016, J. Geophys. Res. D, 121, 2994 Chapman, C. R., & Morrison, D. 1994, Nature, 367, 33 Dwyer, J. R. 2003, Geophys. Res. Lett., 30, 2055 Charlson, R. J., Warren, S. G., Lovelock, J. E., & Andreae, Dwyer, J. R., & Uman, M. A. 2014, Phys. Rep., 534, 147 M. O. 1987, Nature, 326, 655 Eastwood, J. P., Biffis, E., Hapgood, M. A., et al. 2017, Chivian, D., Brodie, E. L., Alm, E. J., et al. 2008, Science, Risk Anal., 37, 206 322, 275 Eather, R. H. 1980, Majestic Lights: The Aurora in Chyba, C., & Sagan, C. 1992, Nature, 355, 125 Science, History, and the Arts (American Geophysical Cirković,´ M. M., Sandberg, A., & Bostrom, N. 2010, Risk Union), doi:10.1029/SP018 Anal., 30, 1495 Ehrenfreund, P., Irvine, W., Becker, L., et al. 2002, Rep. Clark, B. C., Baker, A. L., Cheng, A. F., et al. 1999, Orig. Prog. Phys, 65, 1427 Life Evol. Biosph., 29, 521 Eldredge, N., & Gould, S. J. 1972, in Models in Cleaves, H. J., & Miller, S. L. 1998, Proc. Natl. Acad. Sci. Paleobiology, ed. T. J. M. Schopf (Freeman, Cooper & USA, 95, 7260 Co, San Francisco), 82–115 15

Emslie, A. G., Dennis, B. R., Shih, A. Y., et al. 2012, Hayakawa, H., Tamazawa, H., Kawamura, A. D., & Isobe, Astrophys. J., 759, 71 H. 2015, Earth, Planets, and Space, 67, 82 Erlykin, A. D., & Wolfendale, A. W. 2010, Surv. Geophys., Hayakawa, H., Tamazawa, H., Uchiyama, Y., et al. 2017, 31, 383 Sol. Phys., 292, 12 Erwin, D. H. 2006, Extinction: How Life on Earth Nearly Heath, D. F., Krueger, A. J., & Crutzen, P. J. 1977, Ended 250 Million Years Ago (Princeton Univ. Press) Science, 197, 886 Estrela, R., & Valio, A. 2017, submitted to Astrobiology, Heller, R., & Armstrong, J. 2014, Astrobiology, 14, 50 arXiv:1708.05400 Heller, R., Williams, D., Kipping, D., et al. 2014, Fairén, A. G., Davila, A. F., Lim, D., et al. 2010, Astrobiology, 14, 798 Astrobiology, 10, 821 Horikawa, D. D., Sakashita, T., Katagiri, C., et al. 2006, Feulner, G. 2011, Int. J. Astrobiol., 10, 123 Int. J. Rad. Biol., 82, 843 Field, C. B., Behrenfeld, M. J., Randerson, J. T., & Ineson, S., Maycock, A. C., Gray, L. J., et al. 2015, Nat. Falkowski, P. 1998, Science, 281, 237 Commun., 6, 7535 Garcia-Sage, K., Glocer, A., Drake, J. J., Gronoff, G., & Jackman, C. H., Fleming, E. L., & Vitt, F. M. 2000, J. Cohen, O. 2017, Astrophys. J. Lett., 844, L13 Geophys. Res., 105, 11659 Gehrels, N., Laird, C. M., Jackman, C. H., et al. 2003, Jackman, C. H., Roble, R. G., & Fleming, E. L. 2007, Astrophys. J., 585, 1169 Geophys. Res. Lett., 34, L04812 Gillon, M., Triaud, A. H. M. J., Demory, B.-O., et al. 2017, Jackman, C. H., Deland, M. T., Labow, G. J., et al. 2005, Nature, 542, 456 Adv. Space Res., 35, 445 Glasstone, S., & Dolan, P. J. 1977, The Effects of Nuclear Jackman, C. H., Marsh, D. R., Vitt, F. M., et al. 2008, Weapons, Tech. rep., Department of Defense, Atmos. Chem. Phys., 8, 765 Washington DC Jackson, R. J., & Jeffries, R. D. 2013, Mon. Not. R. Astron. Gopalswamy, N. 2017, in Extreme Events in the Geospace: Soc., 431, 1883 Origins, Predictability and Consequences, ed. Janvier, M. 2017, J. Plasma Phys., 83, 535830101 N. Buzulukova (Elsevier), 91–119 Janvier, M., Aulanier, G., & Démoulin, P. 2015, Sol. Phys., Gott, III, J. R. 1993, Nature, 363, 315 290, 3425 Gould, S. J., & Eldredge, N. 1993, Nature, 366, 223 Jolivet, E., L’Haridon, S., Corre, E., Forterre, P., & Prieur, Grant, P. R., Grant, B. R., Huey, R. B., et al. 2017, Phil. D. 2003, Int. J. Syst. Evol. Microbiol., 53, 847 Trans. R. Soc. B, 372, 20160146 Joyce, G. F. 2002, Nature, 418, 214 Gray, L. J., Beer, J., Geller, M., et al. 2010, Rev. Geophys., Kahler, S. W. 1992, Annu. Rev. Astron. Astrophys., 30, 113 48, RG4001 Kaltenegger, L. 2017, Annu. Rev. Astron. Astrophys., 55, Grice, K., Cao, C., Love, G. D., et al. 2005, Science, 307, 433 706 Karoff, C., Knudsen, M. F., De Cat, P., et al. 2016, Nat. Grießmeier, J.-M., Stadelmann, A., Grenfell, J. L., Lammer, H., & Motschmann, U. 2009, Icarus, 199, 526 Commun., 7, 11058 Grießmeier, J.-M., Stadelmann, A., Motschmann, U., et al. Kataoka, R. 2013, Space Weather, 11, 214 2005, Astrobiology, 5, 587 Kay, C., Opher, M., & Kornbleuth, M. 2016, Astrophys. J., Grießmeier, J.-M., Tabataba-Vakili, F., Stadelmann, A., 826, 195 Grenfell, J. L., & Atri, D. 2015, Astron. Astrophys., 581, Kerr, J. B., & McElroy, C. T. 1993, Science, 262, 1032 A44 Khodachenko, M. L., Ribas, I., Lammer, H., et al. 2007, —. 2016, Astron. Astrophys., 587, A159 Astrobiology, 7, 167 Häder, D.-P., Kumar, H. D., Smith, R. C., & Worrest, Kiehl, J. T., & Shields, C. A. 2005, Geology, 33, 757 R. C. 2007, Photochem. Photobiol. Sci., 6, 267 Kirkby, J. 2007, Surv. Geophys., 28, 333 Häder, D.-P., Williamson, C. E., Wängberg, S.-A., et al. Kitchatinov, L. L., & Olemskoy, S. V. 2016, Mon. Not. R. 2015, Photochem. Photobiol. Sci., 14, 108 Astron. Soc., 459, 4353 Hallam, A., & Wignall, P. B. 1997, Mass Extinctions and Knoll, A. H. 2015, Life on a Young Planet: The First Three Their Aftermath (Oxford Univ. Press) Billion Years of Evolution on Earth, Princeton Science Hapgood, M. A. 2011, Adv. Space Res., 47, 2059 Library (Princeton Univ. Press) Haqq-Misra, J., Kopparpu, R. K., & Wolf, E. T. 2017, Int. Knoll, A. H., Bambach, R. K., Canfield, D. E., & J. Astrobiol., arXiv:1705.07813 Grotzinger, J. P. 1996, Science, 273, 452 16

Knoll, A. H., Bambach, R. K., Payne, J. L., Pruss, S., & Meyer, K. M., & Kump, L. R. 2008, Annu. Rev. Earth Fischer, W. W. 2007, Earth Planet. Sci. Lett., 256, 295 Planet. Sci., 36, 251 Legrand, M. R., & Kirchner, S. 1990, J. Geophys. Res., 95, Mironova, I. A., Aplin, K. L., Arnold, F., et al. 2015, Space 3493 Sci. Rev., 194, 1 Lenton, T. M., Held, H., Kriegler, E., et al. 2008, Proc. Miroshnichenko, L. I. 2001, Astrophysics and Space Science Natl. Acad. Sci. USA, 105, 1786 Library, Vol. 260, Solar Cosmic Rays (Springer), Lingam, M., Comisso, L., & Bhattacharjee, A. 2017, Phys. doi:10.1007/978-94-015-9646-6 Plasmas, arXiv:1702.05782 Miyake, F., Masuda, K., & Nakamura, T. 2013, Nat. Lingam, M., & Loeb, A. 2017a, Int. J. Astrobiol., Commun., 4, 1748 arXiv:1707.02996 Miyake, F., Nagaya, K., Masuda, K., & Nakamura, T. 2012, —. 2017b, Astrophys. J. Lett., 846, L21 Nature, 486, 240 —. 2017c, submitted to Astrophys. J. Lett., Morin, J., Donati, J.-F., Petit, P., et al. 2010, Mon. Not. R. arXiv:1709.05348 Astron. Soc., 407, 2269 —. 2017d, Proc. Natl. Acad. Sci. USA, 114, 6689 Morowitz, H. 1967, Nature, 215, 1259 Livingston, W., Harvey, J. W., Malanushenko, O. V., & Muñoz-Jaramillo, A., Senkpeil, R. R., Windmueller, J. C., Webster, L. 2006, Sol. Phys., 239, 41 et al. 2015, Astrophys. J., 800, 48 Lockwood, M. 2012, Surv. Geophys., 33, 503 Neuhäuser, R., & Hambaryan, V. V. 2014, Astron. Nachr., Loeb, A., Batista, R. A., & Sloan, D. 2016, J. Cosmol. 335, 949 Astropart. Phys., 8, 040 Neuhäuser, R., & Neuhäuser, D. L. 2015, Astron. Nachr., Longmire, C. 1978, IEEE Trans. Antennas. Propag., 26, 3 336, 225 López-Puertas, M., Funke, B., Gil-López, S., et al. 2005, J. Nogami, D., Notsu, Y., Honda, S., et al. 2014, Publ. Geophys. Res. A, 110, A09S43 Astron. Soc. Jpn, 66, L4 Maehara, H., Shibayama, T., Notsu, Y., et al. 2015, Earth, Odenwald, S., Green, J., & Taylor, W. 2006, Adv. Space Planets, and Space, 67, 59 Res., 38, 280 Maehara, H., Shibayama, T., Notsu, S., et al. 2012, Nature, O’Malley-James, J. T., & Kaltenegger, L. 2017, Mon. Not. 485, 478 R. Astron. Soc. Lett., 469, L26 Malloy, K. D., Holman, M. A., Mitchell, D., & Detrich, III, Owen, T. 1992, in Mars, ed. H. H. Kieffer, B. M. Jakosky, H. W. 1997, Proc. Natl. Acad. Sci. USA, 94, 1258 C. W. Snyder, & M. S. Matthews (Univ. of Arizona Mamajek, E. E., & Hillenbrand, L. A. 2008, Astrophys. J., Press), 818–834 687, 1264 Patel, B. H., Percivalle, C., Ritson, D. J., Duffy, C. D., & Mayewski, P. A., Lyons, W. B., Spencer, M. J., et al. 1986, Sutherland, J. D. 2015, Nat. Chem., 7, 301 Science, 232, 975 Patterson, C., & Smith, A. B. 1987, Nature, 330, 248 McCracken, K. G., Dreschhoff, G. A. M., Zeller, E. J., Pirjola, R. 2000, IEEE Trans. Plasma Sci., 28, 1867 Smart, D. F., & Shea, M. A. 2001, J. Geophys. Res., 106, Posner, R. A. 2004, Catastrophe: Risk and Response 21585 (Oxford Univ. Press) Mekhaldi, F., Muscheler, R., Adolphi, F., et al. 2015, Nat. Powner, M. W., Gerland, B., & Sutherland, J. D. 2009, Commun., 6, 8611 Nature, 459, 239 Melott, A. L., & Bambach, R. K. 2014, Paleobiology, 40, Priest, E. 2014, Magnetohydrodynamics of the Sun 177 (Cambridge Univ. Press) Melott, A. L., & Thomas, B. C. 2011, Astrobiology, 11, 343 Prinn, R. G., & Fegley, Jr., B. 1987, Earth Planet. Sci. —. 2012, Nature, 491, E1 Lett., 83, 1 Melott, A. L., Thomas, B. C., Laird, C. M., Neuenswander, Pulkkinen, T. 2007, Living Rev. Sol. Phys., 4, 1 B., & Atri, D. 2016, J. Geophys. Res. D, 121, 3017 Ranjan, S., Wordsworth, R., & Sasselov, D. D. 2017, Melott, A. L., Lieberman, B. S., Laird, C. M., et al. 2004, Astrophys. J., 843, 110 Int. J. Astrobiol., 3, 55 Raup, D. M. 1985, Nature, 314, 341 Merrill, R. T., McElhinny, M. W., & McFadden, P. L. 1998, —. 1986, Science, 231, 1528 International Geophysics Series, Vol. 63, The Magnetic Raup, D. M., & Sepkoski, J. J. 1984, Proc. Natl. Acad. Sci. Field of the Earth: Paleomagnetism, the Core, and the USA, 81, 801 Deep Mantle (Academic Press) —. 1986, Science, 231, 833 17

Reid, G. C., Isaksen, I. S. A., Holzer, T. E., & Crutzen, Segura, A., Walkowicz, L. M., Meadows, V., Kasting, J., & P. J. 1976, Nature, 259, 177 Hawley, S. 2010, Astrobiology, 10, 751 Reid, G. C., McAfee, J. R., & Crutzen, P. J. 1978, Nature, Seppälä, A., Verronen, P. T., Sofieva, V. F., et al. 2006, 275, 489 Geophys. Res. Lett., 33, L07804 Riley, P. 2012, Space Weather, 10, 02012 Shapiro, R., & Schulze-Makuch, D. 2009, Astrobiology, 9, Rind, D. 2002, Science, 296, 673 335 Rodger, C. J., Verronen, P. T., Clilverd, M. A., SeppäLä, Shibata, K., & Magara, T. 2011, Living Rev. Sol. Phys., 8, 6 A., & Turunen, E. 2008a, J. Geophys. Res. D, 113, Shibata, K., Isobe, H., Hillier, A., et al. 2013, Publ. Astron. D23302 Soc. Jpn, 65, 49 —. 2008b, J. Geophys. Res. D, 113, D23302 Shibayama, T., Maehara, H., Notsu, S., et al. 2013, Rohde, R. A., & Muller, R. A. 2005, Nature, 434, 208 Astrophys. J. Suppl., 209, 5 Rubenstein, E. P., & Schaefer, B. E. 2000, Astrophys. J., Showstack, R. 2011, EOS Transactions, 92, 374 529, 1031 Sloan, D., Batista, R. A., & Loeb, A. 2017, Sci. Rep., 7, Rugheimer, S., Segura, A., Kaltenegger, L., & Sasselov, D. 5419 2015, Astrophys. J., 806, 137 Smart, D. F., Shea, M. A., Melott, A. L., & Laird, C. M. Ruiz-Mirazo, K., Briones, C., & de la Escosura, A. 2014, 2014, J. Geophys. Res. A, 119, 9430 Chem. Rev., 114, 285 Smith, D. S., Scalo, J., & Wheeler, J. C. 2004, Orig. Life Sabine, C. L., Feely, R. A., Gruber, N., et al. 2004, Science, Evol. Biosph., 34, 513 305, 367 Smith, K. 2013, Environmental Hazards: Assessing Risk Sagan, C. 1961, Radiat. Res., 15, 174 and Reducing Disaster, 6th edn. (Routledge) —. 1973, J. Theor. Biol., 39, 195 Smith, R. C., Prezelin, B. B., Baker, K. S., et al. 1992, Sagan, C., & Mullen, G. 1972, Science, 177, 52 Science, 255, 952 Sagan, C., & Salpeter, E. E. 1976, Astrophys. J. Suppl., 32, Soderblom, D. R. 2010, Annu. Rev. Astron. Astrophys., 48, 737 581 Saladino, R., Crestini, C., Pino, S., Costanzo, G., & Di Solanki, S. K., Krivova, N. A., & Haigh, J. D. 2013, Annu. Mauro, E. 2012, Phys. Life Rev., 9, 84 Rev. Astron. Astrophys., 51, 311 Scalo, J., Kaltenegger, L., Segura, A. G., et al. 2007, Solomon, S. 1999, Rev. Geophys., 37, 275 Astrobiology, 7, 85 Space Studies Board. 2009, Severe Space Weather Schaefer, B. E., King, J. R., & Deliyannis, C. P. 2000, Events–Understanding Societal and Economic Impacts: Astrophys. J., 529, 1026 A Workshop Report (Natl. Acad. Press) Scheffer, M., Carpenter, S., Foley, J. A., Folke, C., & Steinhilber, F., & Beer, J. 2013, J. Geophys. Res. A, 118, Walker, B. 2001, Nature, 413, 591 1861 Schindler, D. W. 1988, Science, 239, 149 Stephenson, F. R. 2015, Adv. Space Res., 55, 1537 Schmidt, S. J., Shappee, B. J., Gagné, J., et al. 2016, Astrophys. J. Lett., 828, L22 Stothers, R. 1980, Nature, 287, 365 Schrijver, C. J., Beer, J., Baltensperger, U., et al. 2012, J. Sutherland, J. D. 2016, Angew. Chem. Int. Ed., 55, 104 Geophys. Res. A, 117, A08103 Takahashi, T., Mizuno, Y., & Shibata, K. 2016, Astrophys. Schrijver, C. J., Kauristie, K., Aylward, A. D., et al. 2015, J. Lett., 833, L8 Adv. Space Res., 55, 2745 Tamazawa, H., Kawamura, A. D., Hayakawa, H., et al. Schulte, P. M. 2015, J. Exp. Biol., 218, 1856 2017, Publ. Astron. Soc. Jpn, 69, 22 Schulte in den Bäumen, H., Moran, D., Lenzen, M., Cairns, Thomas, B. C., Jackman, C. H., & Melott, A. L. 2007, I., & Steenge, A. 2014, Nat. Hazard. Earth Syst. Sci., 14, Geophys. Res. Lett., 34, L06810 2749 Thomas, B. C., Melott, A. L., Arkenberg, K. R., & Snyder, Schulze-Makuch, D., Grinspoon, D. H., Abbas, O., Irwin, B. R. 2013, Geophys. Res. Lett., 40, 1237 L. N., & Bullock, M. A. 2004, Astrobiology, 4, 11 Thomas, B. C., Neale, P. J., & Snyder, II, B. R. 2015, Schulze-Makuch, D., & Irwin, L. N. 2006, Astrobiology, 15, 207 Naturwissenschaften, 93, 155 Thomas, B. C., Melott, A. L., Jackman, C. H., et al. 2005, Schumann, U., & Huntrieser, H. 2007, Atmos. Chem. Phys., Astrophys. J., 634, 509 7, 3823 Thomas, C. D., Cameron, A., Green, R. E., et al. 2004, Schwenn, R. 2006, Living Rev. Sol. Phys., 3, 2 Nature, 427, 145 18

Thomas, P. J., Hicks, R. D., Chyba, C. F., & McKay, C. P. Ward, P., & Brownlee, D. 2000, Rare Earth: Why Complex 2006, Comets and the Origin and Evolution of Life, Life Is Uncommon in the Universe (Copernicus) Advances in Astrobiology and Biogeophysics (Springer), Way, M. J., Del Genio, A. D., Kiang, N. Y., et al. 2016, doi:10.1007/10903490 Geophys. Res. Lett., 43, 8376 Toon, O. B., Pinto, J., Hamill, P., & Turco, R. P. 1986, Geophys. Res. Lett., 13, 1284 Webb, D. F., & Howard, T. A. 2012, Living Rev. Sol. Toon, O. B., Zahnle, K., Morrison, D., Turco, R. P., & Phys., 9, 3 Covey, C. 1997, Rev. Geophys., 35, 41 Wolff, E. W., Bigler, M., Curran, M. A. J., et al. 2012, Usoskin, I. G. 2017, Living Rev. Sol. Phys., 14, 3 Geophys. Res. Lett., 39, L08503 Usoskin, I. G., Kromer, B., Ludlow, F., et al. 2013, Astron. Wordsworth, R. D. 2016, Annu. Rev. Earth Planet. Sci., 44, Astrophys., 552, L3 381 Usoskin, I. G., Solanki, S. K., Kovaltsov, G. A., Beer, J., & Kromer, B. 2006, Geophys. Res. Lett., 33, L08107 Yudkowsky, E. 2008, in Global Catastrophic Risks, ed. Vaquero, J. M. 2007, Adv. Space Res., 40, 929 N. Bostrom & M. M. Cirković(Oxford´ Univ. Press), Verronen, P. T., SeppäLä, A., Clilverd, M. A., et al. 2005, 91–119 J. Geophys. Res. A, 110, A09S32 Zahnle, K. J. 1990, in Global Catastrophes in Earth Vidotto, A. A., Jardine, M., Morin, J., et al. 2013, Astron. History, ed. V. L. Sharpton & P. D. Ward, Vol. 247 Astrophys., 557, A67 (Geological Society of America), 271–288 Vincent, W. F., & Roy, S. 1993, Environ. Rev., 1, 1 Volland, H. 1984, Physics and Chemistry in Space, Vol. 11, Zepp, R. G., Erickson III, D. J., Paul, N. D., & Sulzberger, Atmospheric Electrodynamics (Springer-Verlag), B. 2011, Photochem. Photobiol. Sci., 10, 261 doi:10.1007/978-3-642-69813-2 Zuluaga, J. I., Bustamante, S., Cuartas, P. A., & Hoyos, Walkowicz, L. M., Basri, G., Batalha, N., et al. 2011, J. H. 2013, Astrophys. J., 770, 23 Astron. J., 141, 50