Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A shorter Archean day-length biases interpretations of the early Earth's climate

A shorter Archean day-length biases interpretations of the early Earth's climate A shorter Archean day-length biases interpretations of the early Earth’s climate 1 2 1 Christopher Spalding , and Woodward W. Fischer Division of Geological and Planetary Sciences California Institute of Technology, Pasadena, CA 91125 Department of Astronomy Yale University, New Haven, CT 06511 March 12, 2019 Abstract to global glaciation. We show that within the faster- spinning regime, the greenhouse warming required to Earth’s earliest sedimentary record contains evidence generate an ice-free Earth can differ from that re- that surface temperatures were similar to, or per- quired to generate an Earth with permanent ice caps haps even warmer than modern. In contrast, stan- by the equivalent of 1–2 orders of magnitude of pCO . dard Solar models suggest the Sun was 25% less Accordingly, the resolution of the Faint Young Sun luminous at this ancient epoch, implying a cold, problem depends significantly on whether the early frozen planet—all else kept equal. This discrepancy, Earth was ever, or even at times, ice-free. known as the Faint Young Sun Paradox, remains un- resolved. Most proposed solutions invoke high con- centrations of greenhouse gases in the early atmo- 1 Introduction sphere to offset for the fainter Sun, though current The Earth hosts an extensive variety of habitable geological constraints are insufficient to verify or fal- environments, which approximately track the places sify these scenarios. In this work, we examined sev- where liquid water persists. Liquid water, therefore, eral simple mechanisms that involve the role played is considered the most critical component of habit- by Earth’s spin rate, which was significantly faster ability as applied to other planets, ranging from Mars during Archean time. This faster spin rate enhances to extrasolar worlds (Kasting, 2010). The Earth’s ge- the equator-to-pole temperature gradient, facilitat- ological record has preserved over 4 billion years of ing a warm equator, while maintaining cold poles. global climate history, revealing that large bodies of Results show that such an enhanced meridional gra- liquid water have persisted throughout the majority dient augments the meridional gradient in carbon- ate deposition, which biases the surviving geological of Earth’s past. Accordingly, the Earth is not just record away from the global mean, toward warmer habitable today, but has remained habitable for bil- waters. Moreover, using simple atmospheric mod- lions of years (Grotzinger and Kasting, 1993). els, we found that the faster-spinning Earth was less Despite the empirically-derived presence of tem- sensitive to ice-albedo feedbacks, facilitating larger perate surface environments during Earth’s history, meridional temperature gradients before succumbing models of the Sun’s evolution lead to an apparent arXiv:1903.03817v1 [astro-ph.EP] 9 Mar 2019 contradiction. Standard models suggest that the sedimentary basins are rarely known and hard to Sun’s luminosity has been steadily increasing, such measure, making it difficult to ascertain whether or that 3.8 billion years ago, its luminosity was only 75% not the geological indicators of a warm climate (e.g. of its modern value (Gough, 1981). For greenhouse absence of glacial deposits) reflects a truly global con- effects similar to today, the planet would be expected dition (Evans and Pisarevsky, 2008). to enter a global snowball state at Solar luminosities The early atmospheric composition facilitated the of only  10% below today’s value (Hoffman et al., development of life and thus, before settling upon 1998; Yang et al., 2012). In spite of observations the concentration of greenhouse gases required to ex- to the contrary, the Sun’s history predicts a glob- plain the geological record, it is important to take full ally glaciated climate during most of the Archean era account of climactic drivers aside form atmospheric ( 3:8 2:5 Gya)(Budyko, 1969; Sellers, 1969; Kast- composition. One such influence is the Earth’s ro- ing et al., 1993; Ikeda and Tajika, 1999; Abbot et al., tation rate, which is known to have been decreas- 2011). ing monotonically over geological time owing to tides The discrepancy between the presence of liquid wa- raised on Earth from the moon (Touma and Wisdom, ter, and a faint early Sun has been dubbed the “Faint 1994). The most apparent consequence of a faster Young Sun Paradox” and gained widespread aware- rotation rate is an enhancement of the equator-to- ness owing to the work of Sagan and Mullen (1972). pole temperature gradient (North, 1975; Kaspi and A resolution to the paradox requires that the ancient Showman, 2015; Liu et al., 2017), which suggests Earth retained a substantially augmented fraction of that the Archean Earth exhibited a hotter equator incoming Solar heat than today, and/or that the re- relative to its poles than today. Our results suggest constructions of Solar luminosity are in error (Feul- that the extra warmth trapped at the equator gener- ner, 2012). ated signals in the geological record that would have A majority of previous work has proposed that tended toward higher temperatures, thereby creating the Archean greenhouse atmospheric composition dif- a naturally-biased representation of the mean global fered radically from today in order to accommodate Archean climate. the fainter Sun (Kasting et al., 1993; Wordsworth and Pierrehumbert, 2013). Under certain scenarios, cli- mate models possessing heightened levels of green- 2 Length of day house gases, such as CH and CO , successfully re- 4 2 produce the required global temperatures under a The length of Earth’s day was undoubtedly shorter in faint early Sun. Indeed, a promising negative feed- the past, but an accurate reconstruction back in time back has long been known to exist for generating ad- is currently beyond model capabilities, largely owing ditional CO when the climate cools (Walker et al., to the uncertain tidal dissipation rates throughout 1981). Earth’s history (Touma and Wisdom, 1994; Egbert Despite the promise shown by the hypothesis of and Ray, 2000). Growth bands in the skeletons of elevated greenhouse warming, geological constraints marine organisms reveal a Cambrian day length of upon the Archean atmospheric composition are un-  21 hours (Williams, 2000). However, reconstruc- able to adequately distinguish between paleoclimate tions of greater antiquity, before abundant availabil- models (Feulner, 2012). The Archean geological ity of animal skeletons, are highly controversial. De- record is sufficiently sparse and fragmented such that spite such limitations, it has been suggested that a it is not clear whether the early Earth was ice-free, 21 hour resonance in atmospheric thermal tidal forc- or whether ice caps persisted (de Wit and Furnes, ing maintained a 21 hour day back as far as 2.5 Gya 2016). Below we show that distinguishing between (Zahnle and Walker, 1987; Bartlett and Stevenson, these two scenarios is critical to evaluating the like- 2016), and recent stratigraphic techniques are emerg- lihood that greenhouse gases can feasibly solve the ing that may soon improve the reliability of Precam- paradox. Moreover, the paleolatitudes of Archean brian day length reconstructions (Meyers and Malin- 2 Solar luminosity (relative to modern) verno, 2018). of 2 (Figure 1). The time spent between day-lengths The evolution of the Earth-Moon semi-major axis of 6 and 9 hours is geologically brief, such that for may be written as (Bills and Ray, 1999) most of the Archean, the Earth was likely rotating at a rate above 6 hours. da 11=2 = fa (1) dt 24 1 Archean era 21-hour resonance model where f is a dimensional quantity that encodes in- 0.95 formation related to Earth’s dissipative properties and a is the Earth-Moon semi-major axis (assuming 0.9 negligible orbital eccentricity). Today, f  6:29 45 13=2 1 10 m yr , measured using the recession of the 0.85 moon’s orbit. Owing to conservation of angular mo- ocean model mentum, the Earth-Moon separation is trivially con- 0.8 60% modern modern tidal verted to a length of day for the Earth (assuming dissipation dissipation constancy of structural parameters). A length of day 0.75 3.5 3 2.5 2 1.5 1 0.5 0 4.5 4 of 6 hours corresponds to an orbital distance of about time before present (Gyr) 9.26 Earth radii. Such a close orbit would likely leave observable features upon the Earth and so we con- Figure 1: The length of Earth’s day as a function of sider it as a lower limit for the Earth’s day-length. time given different assumptions about how tidal dis- The evolution of Earth’s length of day is shown sipation has changed over Earth history. Red, black in Figure 1 for four different possible scenarios, each and gray lines reflect the evolution under modern, differing in their assumptions regarding variations in 60% modern and 35% modern tidal dissipation rates. tidal dissipation over time. Assuming a modern dis- The blue line, drawn from Webb (1982), was derived sipation agrees well with geological proxies back to from a model of tidal dissipation with a more sophis- the Cambrian Period (Williams, 2000; Meyers and ticated model of the oceans. All results show that Malinverno, 2018), but predicts that the Earth and tidal dissipation must be less than 60% of today av- Moon came within 10 Earth radii about 1.5 Gyr into eraged over the history of the Earth-moon system, the past—a condition wholly inconsistent with ob- given the lack of evidence for a cataclysmic origin servations from the Proterozoic and Archean record. of the moon until prior to  4 Gya. By the end Even dissipation 60% of today’s value suggests an in- of Archean time, the day-length was likely above 12 ferred age of the Moon that is too young. Assuming hours, having changed substantially during that era. 35% of today’s dissipation produces a reasonable age The gold line represents the predicted solar luminos- of the moon, but conflicts with the Phanerozoic sed- ity following the computation within Gough (1981). imentary record. The third model (blue line) was derived by Webb (1982), and includes a careful computation of the dis- In order to take account of the range of possibili- sipation within an ocean model; it agrees better than ties, we investigated spin rates ranging from mod- a simple modulation of f in equation 1. The fourth ern to the equivalent of 6 hours, i.e., 1 4, hypothesis (shown by a grey line) fits best, which is and denotes the modern-day spin angular veloc- the 21-hour resonance model as presented in Bartlett ity. This choice of range is in part owing to recent and Stevenson (2016). General Circulation Model (GCM) simulations that Given that tides get stronger as the moon is closer, performed numerical experiments across this 4-fold the day length would increase at a faster rate dur- range of spin rates (Liu et al., 2017), allowing us to ing the Archean than during later time periods, such calibrate our simplified mathematical framework us- that the early and late Archean worlds would differ ing results of more complex models. However, the markedly in their rotation rates, perhaps by a factor 6 hour case should be seen as a lower limit to the day length of day (hours) day-length known length—most of Archean time was likely character- order to retain the same equator-to-pole heat flux, a ized by a slower spin rate, between 9 and 18 hours steeper temperature gradient is retained. The value (Figure 1). of K cannot be derived from first principles, and eff must either be measured directly, or extracted from General Circulation Models. In this work, we chose 3 Influence of Earth’s spin rate values of K as inferred from the simulations per- eff formed in Liu et al. (2017). We fitted the depen- Earth’s climate is driven by the distribution of Solar dence of diffusivity and spin rate from their simula- insolation, with more energy being received at the tions with the relationship equator than at the poles. Owing to the presence of the atmosphere, heat is transported from hotter eff = ; (3) to cooler latitudes via turbulent processes that may 0 0 be modelled as an effective diffusivity, which lowers the meridional temperature gradient. We write this 6 2 1 where n = 0:7 and K  1:7 10 m s . diffusivity as (see appendix) The precise magnitude of K varies between in- eff vestigations. The important feature here is the re- D  K ; (2) eff lationship between and K , which we fixed at eff gR 0:7 K / . However, it is interesting to compare eff 3 1 1 where c = 10 J kg K is the specific heat of air, this to dependences derived elsewhere. Kaspi and = 1 bar is the surface air pressure, g = 10 m s is Showman (2015) performed global climate simula- the acceleration due to gravity and R = 6400 km is tions at various spin rates within the context of extra- the Earth’s radius. K represents the efficiency of solar planets. They did not explicitly measure K , eff eff meridional heat transfer by way of turbulent eddies but the heat flux from eddies F was presented. In (North, 1975), and is the parameter most directly a diffusive parameterization, F / K =T , where E eff influenced by rotation rate, as we discuss below. T is the equator-to-pole temperature difference. Equation 2 indicates that meridional heat trans- Reading F and T from their figure 8 yields an 0:71 port depends not just upon spin rate, but also upon approximate relationship of K / , which eff atmospheric pressure and heat capacity (Williams is remarkably similar to that used here, as derived and Kasting, 1997). Some evidence has been put for- from Liu et al. (2017). However, the earlier work of ward that the atmospheric pressure differed in the Williams and Kasting (1997) utilized a relationship past (Goldblatt et al., 2009), though theory and geo- where n = 2: a much steeper relationship. Accord- logical proxies often disagree over whether the pres- ingly, the dependence upon spin rate may be even greater than that derived here, though the qualita- sure was more or less than today. If the atmosphere tive picture will not change. was thick with CO , its heat capacity may have been larger, additionally helping to warm and homoge- A rise in meridional temperature gradient is a ro- nize the global climate. However, preserved raindrop bust outcome of global climate models incorporat- imprints, and the size distribution of basaltic vesi- ing faster spin rates (Kaspi and Showman, 2015; Liu cles (Som et al., 2016), suggest atmospheric pressures et al., 2017). However, an alternative approach used similar to, or lower than, modern. Given these dis- in previous work is to utilize the principle of maxi- crepancies, we focus upon the influence of spin rate, mized entropy production to derive meridional tem- but note that a full understanding of the ancient cli- perature gradients (Lorenz et al., 2001). Though this mate requires an auxiliary treatment of the interplay approach was met with moderate success for Titan between, as yet uncertain, changes in atmospheric and Mars, the methodology has been criticized, and composition, alongside eddy-driven heat transport. its utility has been questioned (Goody, 2007). Thus, Generally, the magnitude of K is reduced at we continue with the assumption that rotation rate eff faster rotation rates (Liu et al., 2017), such that in plays an important role in heat transport via the at- 4 unstable branches mosphere, owing to the extensive and consistent lit- theoretical approach is that within a specific range erature surrounding these mechanics. of Solar luminosities, multiple stable solutions ex- ist. These include an ice-covered state, an ice-free More uncertain than the relationship between state, along with a solution possessing ice extend- meridional temperature gradient and spin rate, is the ing from the poles to an “ice-line” latitude denoted influence of spin upon global mean temperature. As . The value of  was computed using a pre- the planet spins faster, the Hadley cell is reduced s s defined critical temperature for permanent ice caps in meridional extent (Kaspi and Showman, 2015), (T = T  10 C). The simplicity of this model which potentially reduces cloud coverage and there- s allows a mechanistic exploration of the relationship fore albedo. Consequently, more Solar insolation is between spin rate and the ice-albedo feedback. absorbed and the planet warms (Liu et al., 2017). This argument is encouraging for the faint young Sun problem, however, it is important to note that differ- 2 x today’s ent global climate models have exhibited contradic- today’s spin rate spin rate tory results regarding the impact of spin rate upon 4 x today’s global cloud cover and global temperature (Vallis and spin rate Farneti, 2009). Thus we focus here predominantly upon the meridional temperature profile and the ice- faster spin, weaker feedback albedo feedback, rather than upon the global mean temperature. In what follows, we argue that the enhanced 15 meridional temperature gradient favors the deposi- tion of sedimentary rocks at low latitudes, skewing 0.9 0.95 1 1.05 1.1 1.15 1.2 the Archean record toward warmer temperatures. We Stellar luminosity (relative to present Sun) identified that rotation rate and resulting temper- ature distribution, have two primary influences on Figure 2: The equilibrium ice line (latitude at which interpretations of the ancient climate record. First, T = 10 C) as a function of Solar luminosity, for 3 the most basic interpretation of the Archean record spin rates. The solution is stable if the slope is posi- is that the planet was not globally glaciated. We dis- tive (solid lines) and unstable if the slope is negative cuss the role of a faster rotation rate in decreasing (dashed lines). The reduced slope within the stable sensitivity of the planet to the ice-albedo feedback. regions for higher spin rates amounts to a greater re- Second, we discuss the tendency of a faster-rotating silience to the ice-albedo feedback. temperature distribution to have biased the deposi- The zonally-averaged temperature at any given lat- tion pattern of Archean sedimentary rocks toward itude was computed as a balance between incom- warmer waters, and in turn, biasing interpretations ing, short-wave radiation (from the Sun), out-going of the ancient climate. longwave radiation (from the Earth and its atmo- sphere), together with the effective diffusion of heat between latitudes, with efficiency K (defined in sec- eff 4 Spin rate and the ice-albedo tion 2). The diffusive term represents turbulent phe- feedback nomena that tend to transport heat from the equator to the poles, such as baroclinic instabilities (Kaspi In order to analyze the influence of rotation rate upon and Showman, 2015). the planet’s susceptibility to the ice-albedo feedback, Table 1 contains the values for the various pa- we constructed a simple energy balance model based rameters used, which were largely taken from North upon the Budyko-Sellers framework (Budyko, 1969; (1975). In Figure 2 we plotted the solution for the ice Sellers, 1969; North, 1975). A characteristic of this line as a function of Solar luminosity for 3 values of multiple stable states Ice-line latitude (degrees) Parameter values the faster the spin rate, the weaker the feedback. 2 2 1 a a S A (Wm ) B (Wm K ) 0 1 2 The aforementioned influence upon the feedback 0.68 0.38 0:482 196:2 1:953 strengths lead to two critical and complementary findings. The first, is that ice may extend further Table 1: The parameter values chosen to solve the toward the equator without triggering a global snow- Budyko-Sellers type energy balance model. Val- ball, providing a greater chance that the equator may ues taken from (North, 1975), except A which was be kept ice-free despite low Solar luminosity. How- tuned to match today’s climate with with K = eff ever, a second, larger effect is that a lower mini- 6 2 1 1:7 10 m s mum luminosity is required to entirely deglaciate the planet from a glaciated starting point. This impacts 30 solutions to the faint young Sun paradox because there are substantial differences based on whether or not the early Earth was ice free (even at the poles). Figure 2 captures the result that an ice-free condition ⌦ = ⌦ 0 is significantly more difficult to achieve under faster ice-line -10 spin rates. -20 ⌦ =2⌦ -30 ⌦ =4⌦ 5 Depositional biases -40 -50 The inference of a warm Archean Earth is based 0 15 30 45 60 75 90 largely upon evidence from sedimentary rocks as- Latitude (degrees) sociated with early cratons and marine platforms (Grotzinger and Kasting, 1993). Though controver- Figure 3: Meridional temperature profile arising from sial, absolute temperature estimates have been at- the Budkyo-Sellers model for 3 spin rates. Upon com- tempted using these records, from which the overly- parison to Figure 1, the presence of ice is capable of ing marine waters appear to have been at least as decreasing polar temperatures while maintaining a warm as surface seawater today (Knauth and Lowe, warm equator. Faster spins enhance this equator-to- 2003). However, a key unknown with respect to these pole difference. early archives is the latitude, and often the environ- ment, of deposition. In this section, we examine how the lithotypes used to infer Archean climate are typi- spin rate, translating to 3 values of K . The modern eff cally biased toward sedimentation in warmer waters, climate possesses ice caps down to roughly 72 of lat- thereby skewing the implied Archean climate toward itude, and this was used to tune the parameters for warmer temperatures. In conjunction with the en- the blue line, representing today’s spin rate (North, hanced equator-to-pole temperature gradient associ- 1975). The other two lines represent spin rates of 2 ated with a faster rotation rate, Archean climate may and 4 have been significantly cooler than implied by a naive Although the earliest descriptions of the faint reading of the geochemical record. young Sun paradox were motivated by raising global mean temperature, more recent treatments outline the importance the Earth’s susceptibility to global 5.1 Meridional bias in carbonate de- glaciations (Kasting et al., 1993; Ikeda and Tajika, position 1999). A drop in Solar insolation leads to the spread of ice, which causes an increase in albedo. In reflect- Some of the most reliable evidence for a temperate ing extra sunlight, this additional albedo amplifies Archean climate stems from the existence of shallow the initial drop in insolation. Figure 2 indicates that marine carbonate platforms (Sumner and Grotzinger, Temperature ( C) 3.5 3.5 ⌦ = ⌦ and James, 2000; Ridgwell et al., 2003; Higgins et al., ⌦ =2⌦ 2009; Fischer and Knoll, 2009). ⌦ =4⌦ Key differences with modern times include a lack of 2.5 2.5 skeletonized metazoa and open-ocean planktic calci- fiers. In such a global regime, abiotic mechanisms of kinetic preference for equatorial carbonate production would have dominated the car- deposition of aragonite 1. 1.5 5 bon cycle; fluxes of dissolved inorganic carbon and al- kalinity into the oceans would be balanced almost en- tirely by abiotic precipitation (Ridgwell et al., 2003). 0. 0.5 5 Moreover, the predominance of alkalinity-favoring, anaerobic respiration in shallow sediments would am- 0 15 30 45 60 75 90 0 15 30 45 60 75 90 plify carbonate precipitation and preservation within Latitude (degrees) sediments (Higgins et al., 2009). In Archean basins, carbonate precipitation would Figure 4: The distribution of carbonate deposition have predominantly occurred in shallow-water shelf rates as a function of latitude for 3 spin rate scenar- environments, with a spatial distribution governed ios (modern; blue, twice modern; red, and four times largely by temperature. Inferences regarding global modern; black) for the Budyko-Sellers climate mod- climate by way of Archean carbonates must therefore els presented in the text. All curves are normalized acknowledge the existence of a strong chemical pref- to have an area under the curve of unity, such as to erence, both kinetically and thermodynamically, for highlight the meridional distribution of deposition, precipitation of these rocks within the warmest shal- as opposed to the absolute magnitude, which would low water environments on the planet. In what fol- depend more sensitively upon whole ocean carbonate lows, we estimate the expected bias associated with a chemistry. Notice that, particularly when ice occurs rotationally-enhanced equatorial temperature gradi- at the poles, the rate of aragonite deposition has de- ent, and argue that the widespread occurrence of liq- creased by an order of magnitude by  30 40 and uid water observed may simply be reflecting a warm has essentially dropped to zero by  60 . equator, kept stable by the fast-rotating planet. 1996). In addition, widespread measurements of sta- 5.2 Kinetic model of Carbonate depo- ble oxygen isotope ratio data from carbonates and sition cherts have have been interpreted as indicative of high temperatures (Knauth and Lowe, 2003). Ac- The saturation state of seawater with respect to arag- cordingly, inferences regarding the paleoenvironment onite decreases toward the poles in the modern of the Archean by way of sedimentary records are fil- oceans (Jiang et al., 2015). This trend generally tered through the lens of geochemical environments arises from two causes: 1) aragonite is more soluble favoring the deposition of carbonates and cherts. in colder waters, and 2) the carbonate system adjusts In modern oceans, calcium carbonate salts are pref- in the presence of an atmosphere containing CO to erentially deposited in warm, tropical waters. This raise the concentration of Dissolved Inorganic Car- pattern reflects the evolution and ecology of organ- bon (DIC) in colder waters. In today’s oceans, the isms with carbonate skeletons, but it is also directly saturation state of aragonite varies with temperature related to the inorganic chemistry of the carbonate approximately linearly as system (Zeebe and Wolf-Gladrow, 2001), whereby warmer waters are more saturated with respect to a(T= C) + b: (4) CaCO salts. The dominant mechanisms of carbon- ate precipitation during Precambrian time remain where a  0:09 and b  1:5 in the modern ocean uncertain (Grotzinger and Reed, 1983; Grotzinger (Jiang et al., 2015). g( ) Fraction of aragonite deposition Carbonate compensation is broadly considered To illustrate the effect of temperature, we defined to have been limited by over geological time the relative rate of aragonite deposition within a lat- (Grotzinger and James, 2000; Ridgwell et al., 2003), itudinal band d as g()d, where, 2+ as opposed to Ca -limitation. Accordingly, no mat- g() = r() cos(); (6) ter the precise dependence of upon temperature and latitude, the Precambrian carbon cycle is ex- which follows from the definition of r as a rate per pected to have exhibited a drop in from the equa- unit area on the globe. The absolute magnitude of tor toward the poles. k is unimportant for our investigation; what mat- To proceed, we adopted this relationship for the ters is the relative precipitation rate at each latitude. modern, but note that carbonate deposition in Pre- Accordingly, we treated k as a normalization factor cambrian oceans may not have exhibited a linear de- that satisfies pendence. The values of a (the meridional gradient) " # =2 and b (the saturation state at the freezing point) are g()d = 1: (7) also likely to change, but not independently. In par- ticular, if a is raised, in order to achieve an equivalent Figure 4 illustrates the relative degree of carbonate global precipitation of carbonate, b must drop. Phys- precipitation as a function of latitude. The temper- ically, if saturation state varies more markedly with ature T () was taken from the Budyko-Sellers tem- latitude, the saturation state at high latitudes may be lower while retaining an equivalent global deposi- perature model (Figure 3), using an equivalent so- tion. A stronger or weaker dependence will constitute lar forcing of the modern day. Note that no depo- an additional determinant upon potential biases that sition occurs at latitudes that are ice-covered year- we do not tackle here. Our goal is simply to compare round. Latitudes with annual mean temperatures the biases associated with different rotation rates, for below freezing, but above T = 10 C will allow which it suffices to approximate the dependence as deposition during the fraction of the year when ice is linear across all rotation rates. not present. Though strictly unphysical, the rate at We used the dependence given by equation 4 to these latitudes is reflected by allowing for sub-freezing construct an idealized kinetic model of carbonate de- temperatures in the computation of g(), rather than position. The precipitation rate per unit area of cal- fixing the temperature at the freezing point. The two cium carbonate is generally written (Zeebe and Wolf- choices do not significantly differ, given the slow pre- Gladrow, 2001) cipitation rates at such low temperatures relative to the equator. r() = k exp(E =R T ())( (T ()) 1) ; (5) 0 a g A where k is a reference rate, is the saturation 0 A 5.3 Preference for equatorial carbon- state of aragonite, n  1:7 (Romanek et al., 2011) ate deposition is the order of the reaction, E  71 kJ mol is the 1 1 activation energy and R = 8:31J mol K is the As is illustrated in Figure 4, a significant enhance- universal gas constant. ment of deposition rates near the equator occurs as The rate of the reaction depends upon temper- a result of the combination of enhanced tempera- ature in two fundamentally different aspects. The tures and surface area at low latitudes. An order of first is that the saturation state itself varies with lat- magnitude enhancement of carbonate deposition oc- itude, being larger in warmer tropical waters. How- curs at the equator relative to the lower latitudes of ever, the dominant effect lies in the Arrhenius factor  30 40 , and deposition drops to zero close to the exp(E =R T ), such that if the whole ocean is to ice-line. Crucially, the existence of systematic bias a g balance the fluxes of ions entering it, calcium car- is present for all 3 rotation rates, meaning that the bonate will precipitate faster in the warmer regions Precambrian climate record exhibits an ever-present owing to a kinetic enhancement of reaction rates. tendency towards an equatorial bias. 8 Despite the presence of bias irrespective of rota- tion rate, the magnitude and interpretation of the bias depends upon rotation rate. Upon comparison mean temperature of carbonates to Figures 2 and 3, at 1 Solar luminosity, ice extends rotationally-induced significantly further toward the equator at higher ro- bias toward high temperatures tation rates than modern. In other words, receiving information only about lower latitudes gives a signif- icantly more biased picture of global mean temper- mean global temperature atures when rotation rates are higher. In order to quantify this bias, we defined the true mean global temperature as 1 2 3 4 R Earth’s spin rate (relative to today) =2 T () cos()d =2 Figure 5: The true, global mean temperature T (up- cos()d per, red line) and the temperature that would be in- =2 = T () cos()d: (8) ferred from biased carbonate deposition T (lower, obs blue line) as a function of rotation rate. We set lu- minosity equal to Solar for each. Despite the spread In contrast, the averaged signal captured by paleocli- of ice sheets leading to a drop in T , the temperature mate records in the geological record will be skewed inferred from the geological record would increase ow- by the kinetic factor g(), such that the mean inferred ing to a strong preference toward equatorial deposi- from the record is given by tion. =2 T = g()T ()d: (9) obs Knoll, 2009). There appear also to have been pri- mary modes of chert deposition as sedimentary parti- Using the definitions above, we computed the de- cles generated from waters supersaturated with silica pendence of both T and T as a function of , illus- obs (Stefurak et al., 2014). If indeed much of Archean trated in Figure 5. Notably, faster rotation rates led silica deposition was driven by evaporation (Maliva to an enhancement of the mean temperature of pre- et al., 2005; Fischer and Knoll, 2009), the most likely cipitated carbonates, despite a reduced global mean bias in chert formation is at the descending limb temperature. A strong bias remained for all rotation of the Hadley cell, which moves equator-ward at rates, but the magnitude of this bias was amplified faster spin-rates (Kaspi and Showman, 2015). Given as rotation rate increased. Accordingly, care must be these uncertainties, we did not develop a quantita- taken when inferring a warm Archean climate, sim- tive model of chert deposition, but to the degree that ply from an apparently warm signal in sedimentary this is correct, the potential climate bias from obser- deposits. vations of chert will plausibly follow a qualitatively The biases affecting the chert record are less clear, similar trend to carbonate deposition. in large part because we know far less about the de- positional mechanisms of Precambrian cherts. How- ever, in many cases cherts are likely to have been 5.4 Paleolatitude of shelf environ- associated with diagenetic replacements of carbon- ments ate rocks (Maliva et al., 2005). The paleoenviron- ment of many cherts associated with carbonate plat- In the previous subsection, we demonstrated that a forms suggests that they have mechanisms of precip- relatively hotter equator, at higher rotation rates, itation wherein evaporative concentration and over tends to bias the carbonate record toward warmer saturation may have been important (Fischer and environments. However, a key assumption was that Temperature ( C) 1.2 1.2 all of the Earth’s ocean surface was available for car- bonate deposition (meaning that in the computation 1.15 1.15 of T , we simply averaged over all latitudes). In obs 1.1 1.1 ice-free reality, geological observations have shown that car- permanent polar ice caps bonate deposition during Precambrian time was pro- 1.05 are possible 1.05 nounced in shallow shelf environments (Grotzinger and Reed, 1983; Sumner and Grotzinger, 1996; Ridg- well et al., 2003). A test, in principle, of the sug- 0.95 0.95 gestion that the Archean record is biased toward the 0.9 0.9 equator, would be to reconstruct the paleolatitudes of snowball sedimentary strata from which the paleoclimate data 1 1 2 2 3 3 4 4 were generated. Earth’s spin rate (relative to today) Limited paleomagnetic evidence from the Bar- berton Greenstone belt—a critical archive of early Figure 6: The range of Solar luminosities facilitating Archean paleoclimate data currently—suggests a lo- a partially ice-covered state as a function of Earth’s cation close to the paleoequator at  3:5 Ga (Kröner spin rate (see Figure 4). At = 4 , the increase in and Layer, 1992). Unfortunately, the paleolatitudes Solar luminosity required to deglaciate the planet is of Archean cratons, and indeed the nature of plate many times larger than that required at modern spin motions at such early times, is debated (Evans and rates. Pisarevsky, 2008; Korenaga, 2013), reducing confi- dence in specific reconstructions. planet and amplified by a climate system that was We took a different approach to examine this using less efficient at moving heat from the equator. This the distribution of paleolatitudes for continental cra- raises the hypothesis that the poles may have retained tons for more recent times, and assume that the same cold ice caps while the planet preserved paleoclimate distribution holds for earlier, Archean, times. Our data from the warm equator. calculation above of T essentially assumed that obs over geological time, the distribution of continental cratons is isotropic in latitude. In Figure 7 we con- 6 Implications for the Faint structed a histogram of 45 robust, paleolatitude mea- surements from Evans and Pisarevsky (2008) all more Young Sun Paradox ancient than 800 Ma, and compared it to an isotropic (random) distribution. If no latitude was more likely, A fundamental aspect of the Archean Earth was that the distribution would approximate a cosine—more its day was significantly shorter—and it is valuable to deposition occurs at lower latitudes in proportion to frame problems regarding early climate in this con- the greater surface area of the Earth available there. text. The enhanced spin rate would have influenced The measured distribution closely approximates that the transport of heat and the distribution of temper- expected from no preferred latitude, validating our atures over the globe. In turn this would have im- computation of T . pacted the locus of deposition of chemical sediments. obs To conclude, the Archean oceans, and global mean Archean Earth, like the modern, deposited calcium temperature more broadly, are likely not as warm as carbonate salts in order to balance weathering fluxes a straightforward read of the archive suggests. The into the oceans. As opposed to the biogenic domi- faster-spinning Earth augmented the equatorial tem- nance of such deposition today, the Archean carbon- perature relative to the poles, substantially increas- ate cycle would largely have been driven by abiotic ing the deposition rate of carbonate at low latitudes influences, principally temperature. Combined with relative to the poles. The surviving record, thus, is the enhanced equatorial temperatures, this leads to intrinsically biased toward the warmer parts of the a significant, kinetic bias toward deposition of car- Solar luminosity (relative to modern) bonates at low latitudes (Figure 4). Given that a stable. substantial fraction of the surviving Archean sedi- If an ice-free Earth is required to match the ge- mentary record formed as carbonate, our picture of ological record, a resolution to the faint young Sun the Archean world is skewed towards these warmer problem may require revisiting the founding premise temperatures. behind the whole paradox. I.e., precisely how well We also demonstrated that a faster-rotating Earth do we understand the Solar luminosity over time? If, was generally less sensitive to the ice-albedo feedback. for example, the Sun’s mass had been a few percent This lower sensitivity makes it substantially more dif- higher in the Archean, its higher luminosity would ficult to generate an ice-free planet 3.5 billion years have valuable implications for how we think about ago. Simply put, the Faint Young Sun paradox is far the composition of Earth’s early atmosphere (Feul- easier to explain if Earth always maintained ice at ner, 2012). This idea is promising and remains to the poles than it would be if the planet did not (Wolf be thoroughly tested (Spalding et al., 2018), but is and Toon, 2013). Over the past 500 million years, we somewhat inconsistent with mass-loss rates observed have a decent understanding of when the planet was in Sun-like stars. glaciated and when it was not, from a combination The Archean sedimentary record presents us with of sedimentological and geochemical datasets. a valuable, yet limited set of robust windows into Going further back in time the geological record its climate, and drivers of habitability. It is criti- is sufficiently fragmented that it becomes extremely cal to thoroughly understand the filters applied at challenging to discern glacial and ice-free inter- the moment of sedimentation in order to reconstruct vals. The gold standard geological observations for the conditions that truly persisted soon after the ori- early Earth’s climate come from glacial deposits gin of life. We demonstrated the key role played by (diamictites, ice-rafted debris, striated clasts and rotation rate in sculpting the distribution of temper- pavements)—but these records are inherently sparse. ature, and therefore carbonate deposition, upon the The earliest observations of glacial rocks in the record early Earth. These biases yield an apparently warmer come from the  3 billion-year-old Pongola and Archean than may have existed in reality. Witswatersrand basins of South Africa (Young et al., 1998); yet earlier glacial deposits from  3:5 Ga have been described (de Wit and Furnes, 2016) though al- Acknowledgements ternative non-glaciogenic interpretations remain pos- sible. Results shown here, however, illustrate just C.S thanks the generous support of the NASA how valuable those sedimentological and geochemi- NESSF Graduate Fellowship in Earth and Planetary cal observations are. If further geological investiga- Science, and of the Heising-Simons Foundation’s 51 tion finds sufficiently strong evidence of ice-free states Pegasi b Postdoctoral Fellowship. W.F. acknowl- of the early Earth, then the faint young Sun paradox edges support from the Simons Foundation Collabo- becomes exceedingly more difficult to solve. ration on the Origins of Life. Additionally, we thank In a simple global model, Solar luminosity may be Konstantin Batygin, Greg Laughlin, Dorian Abbot considered as an equivalent to greenhouse gas forcing. and Seth Finnegan for illuminating conversations. Fi- Figure 2 in Ikeda and Tajika (1999) demonstrates a nally, we thank several referees for their important similar result to our Figure 2, but with pCO in place contributions to the work. of Solar luminosity. A comparison between our result and theirs shows that upper limit of the region facili- tating multiple stable states corresponds to 12 or- ders of magnitude greater pCO under Archean spin References rates than the modern equivalent. In turn, roughly 1 order of magnitude lower pCO levels are tolerated Abbot, D. S., Voigt, A., and Koll, D. (2011). The jor- at faster spins before only a single snowball state is mungand global climate state and implications for 11 14 evidence. When did plate tectonics begin on planet Evans & Pisarevsky (2008) Earth, 440:249–263. (paleomagnetic data) 10 Feulner, G. (2012). The faint young sun problem. Reviews of Geophysics, 50(2). 8 uniform global distribution Fischer, W. W. and Knoll, A. H. (2009). An iron shut- tle for deepwater silica in late archean and early pa- leoproterozoic iron formation. Geological Society of America Bulletin, 121(1-2):222–235. Goldblatt, C., Claire, M. W., Lenton, T. M., 0 15 30 45 60 75 90 Matthews, A. J., Watson, A. J., and Zahnle, K. J. paleo - latitude (degrees) (2009). Nitrogen-enhanced greenhouse warming on early earth. Nature Geoscience, 2(12):891. Figure 7: The distribution of paleolatitudes older than 800 Ma in a dataset complied by Evans and Goody, R. (2007). Maximum entropy production in Pisarevsky (2008) (bars). If the distribution was uni- climate theory. Journal of the atmospheric sci- formly distributed across the globe, the bars and red ences, 64(7):2735–2739. line should coincide. The data show that low latitude Gough, D. (1981). Solar interior structure and lumi- cratons—and their sedimentary basins—are slightly nosity variations. In Physics of Solar Variations, overrepresented. pages 21–34. Springer. Grotzinger, J. and Reed, J. (1983). Evidence for neoproterozoic glaciations. Journal of Geophysical primary aragonite precipitation, lower proterozoic Research: Atmospheres, 116(D18). (1.9 ga) rocknest dolomite, wopmay orogen, north- Bartlett, B. C. and Stevenson, D. J. (2016). Analysis west canada. Geology, 11(12):710–713. of a precambrian resonance-stabilized day length. Grotzinger, J. P. and James, N. P. (2000). Precam- Geophysical Research Letters, 43(11):5716–5724. brian carbonates: evolution of understanding. Bills, B. G. and Ray, R. D. (1999). Lunar orbital evo- Grotzinger, J. P. and Kasting, J. F. (1993). New con- lution: A synthesis of recent results. Geophysical straints on precambrian ocean composition. The Research Letters, 26(19):3045–3048. Journal of Geology, 101(2):235–243. Budyko, M. I. (1969). The effect of solar radia- Higgins, J., Fischer, W., and Schrag, D. (2009). Oxy- tion variations on the climate of the earth. tellus, genation of the ocean and sediments: consequences 21(5):611–619. for the seafloor carbonate factory. Earth and Plan- de Wit, M. J. and Furnes, H. (2016). 3.5-ga hy- etary Science Letters, 284(1-2):25–33. drothermal fields and diamictites in the barberton Hoffman, P. F., Kaufman, A. J., Halverson, G. P., greenstone belt paleoarchean crust in cold environ- and Schrag, D. P. (1998). A neoproterozoic snow- ments. Science advances, 2(2):e1500368. ball earth. science, 281(5381):1342–1346. Egbert, G. and Ray, R. (2000). Significant dissipa- Ikeda, T. and Tajika, E. (1999). A study of the tion of tidal energy in the deep ocean inferred from energy balance climate model with co2-dependent satellite altimeter data. Nature, 405(6788):775. outgoing radiation: Implication for the glaciation Evans, D. A. and Pisarevsky, S. A. (2008). Plate tec- during the cenozoic. Geophysical Research Letters, tonics on early earth? weighing the paleomagnetic 26(3):349–352. frequency Jiang, L.-Q., Feely, R. A., Carter, B. R., Greeley, Meyers, S. R. and Malinverno, A. (2018). Protero- D. J., Gledhill, D. K., and Arzayus, K. M. (2015). zoic milankovitch cycles and the history of the so- Climatological distribution of aragonite saturation lar system. Proceedings of the National Academy state in the global oceans. Global Biogeochemical of Sciences, 115(25):6363–6368. Cycles, 29(10):1656–1673. North, G. R. (1975). Analytical solution to a simple climate model with diffusive heat transport. Jour- Kaspi, Y. and Showman, A. P. (2015). Atmo- nal of the Atmospheric Sciences, 32(7):1301–1307. spheric dynamics of terrestrial exoplanets over a wide range of orbital and atmospheric parameters. Ridgwell, A. J., Kennedy, M. J., and Caldeira, The Astrophysical Journal, 804(1):60. K. (2003). Carbonate deposition, climate sta- bility, and neoproterozoic ice ages. Science, Kasting, J. (2010). How to find a habitable planet. 302(5646):859–862. Princeton University Press. Romanek, C. S., Morse, J. W., and Grossman, E. L. Kasting, J. F., Whitmire, D. P., and Reynolds, R. T. (2011). Aragonite kinetics in dilute solutions. (1993). Habitable zones around main sequence Aquatic geochemistry, 17(4-5):339. stars. Icarus, 101(1):108–128. Sagan, C. and Mullen, G. (1972). Earth and mars: evolution of atmospheres and surface tempera- Knauth, L. P. and Lowe, D. R. (2003). High archean tures. Science, 177(4043):52–56. climatic temperature inferred from oxygen isotope geochemistry of cherts in the 3.5 ga swaziland su- Sellers, W. D. (1969). A global climatic model based pergroup, south africa. Geological Society of Amer- on the energy balance of the earth-atmosphere sys- ica Bulletin, 115(5):566–580. tem. Journal of Applied Meteorology, 8(3):392–400. Korenaga, J. (2013). Initiation and evolution of Som, S. M., Buick, R., Hagadorn, J. W., Blake, T. S., plate tectonics on earth: theories and observations. Perreault, J. M., Harnmeijer, J. P., and Catling, Annual Review of Earth and Planetary Sciences, D. C. (2016). Earth’s air pressure 2.7 billion years 41:117–151. ago constrained to less than half of modern levels. Nature Geoscience, 9(6):448. Kröner, A. and Layer, P. (1992). Crust formation Spalding, C., Fischer, W. W., and Laughlin, G. and plate motion in the early archean. Science, (2018). An orbital window into the ancient 256(5062):1405–1411. sun’s mass. The Astrophysical Journal Letters, Liu, X., Battisti, D. S., and Roe, G. H. (2017). The 869(1):L19. effect of cloud cover on the meridional heat trans- Stefurak, E. J., Lowe, D. R., Zentner, D., and Fischer, port: Lessons from variable rotation experiments. W. W. (2014). Primary silica granules-a new mode Journal of Climate, 30(18):7465–7479. of paleoarchean sedimentation. Geology, 42(4):283– Lorenz, R. D., Lunine, J. I., Withers, P. G., and McKay, C. P. (2001). Titan, mars and earth: Sumner, D. Y. and Grotzinger, J. P. (1996). Were Entropy production by latitudinal heat transport. kinetics of archean calcium carbonate precipita- Geophysical Research Letters, 28(3):415–418. tion related to oxygen concentration? Geology, 24(2):119–122. Maliva, R. G., Knoll, A. H., and Simonson, B. M. (2005). Secular change in the precambrian silica Touma, J. and Wisdom, J. (1994). Evolution of the cycle: insights from chert petrology. Geological So- earth-moon system. The Astronomical Journal, ciety of America Bulletin, 117(7-8):835–845. 108:1943–1961. 13 Vallis, G. K. and Farneti, R. (2009). Meridional en- A Solution to the energy bal- ergy transport in the coupled atmosphere–ocean ance model system: Scaling and numerical experiments. Quar- terly Journal of the Royal Meteorological Society, In this appendix, we provided the derivation of the 135(644):1643–1660. solution to the Budyko-Sellers energy-balance cli- mate model. The equation describing the steady- Walker, J. C., Hays, P., and Kasting, J. F. (1981). A state meridional temperature distribution T , written negative feedback mechanism for the long-term sta- in terms of a new variable x  sin(), takes the form bilization of earth’s surface temperature. Journal (North, 1975) of Geophysical Research: Oceans, 86(C10):9776– d dT Webb, D. (1982). Tides and the evolution of the D (1 x ) = Qa(x; x )S(x) I (T; pCO ) ; s 2 dx dx | {z } | {z } earth-moon system. Geophysical Journal of the | {z } Solar insolation re-emitted heat diffusive transport Royal Astronomical Society, 70(1):261–271. (10) Williams, D. M. and Kasting, J. F. (1997). Habitable planets with high obliquities. Icarus, 129(1):254– where the coefficient Williams, G. E. (2000). Geological constraints on D  K ; (11) eff the precambrian history of earth’s rotation and the gR moon’s orbit. Reviews of Geophysics, 38(1):37–59. and Q is the Solar constant divided by 4. In the 3 1 1 Wolf, E. and Toon, O. (2013). Hospitable archean above, c = 10 J kg K is the specific heat of air, climates simulated by a general circulation model.  = 1 bar is the surface air pressure, g = 10 m s Astrobiology, 13(7):656–673. is the acceleration due to gravity and R = 6400 km is the Earth’s radius. The Solar insolation function Wordsworth, R. and Pierrehumbert, R. (2013). S(x) is defined as Hydrogen-nitrogen greenhouse warming in earth’s early atmosphere. science, 339(6115):64–67. S(x) = 1 + S P (x) 2 2 Yang, J., Peltier, W., and Hu, Y. (2012). The initi- = 1 + S (3x 1) (12) ation of modern soft and hard snowball earth cli- mates in ccsm4. Climate of the Past, 8(3):907–918. where P (x) is a degree 2 Legendre Polynomial. The function a(x; x ) yields the fraction of incom- Young, G. M., Brunn, V. V., Gold, D. J., and Minter, ing Solar radiation that is absorbed (one minus the W. (1998). Earth’s oldest reported glaciation: albedo) as a function of latitude. The ice-albedo feed- physical and chemical evidence from the archean back is built into a(x; x ); defined as mozaan group ( 2.9 ga) of south africa. The Jour- nal of geology, 106(5):523–538. a ; if x  x 0 s a(x; x ) (13) Zahnle, K. and Walker, J. C. (1987). A constant a ; otherwise daylength during the precambrian era? Precam- brian Research, 37(2):95–105. where a > a , as ice is more reflective than ocean. 0 1 The outgoing radiation is prescribed through the Zeebe, R. E. and Wolf-Gladrow, D. A. (2001). CO2 function in seawater: equilibrium, kinetics, isotopes. Num- ber 65. Gulf Professional Publishing. I (T ) = A + BT (x) (14) 14 where A and B are constants to be determined by tuning to the modern climate. Such linearity allows us to replace T (x) with I (T )=B as the dependent variable (North, 1975). Carrying out the substitution and defining D  (15) we may formulate the problem into an ordinary differential equation satisfying d dI (x) 1 Q (1 x ) I (x) = S(x)a(x; x ): 0 0 dx dx D D (16) In order to solve the differential equation above, we separated the solution into two regimes; one at low latitudes where the albedo is smaller and a second at high latitudes where the albedo is that of ice. For this part of the calculation, we ignored the effect of spin rate upon albedo, and included only the effect of rotation rate upon diffusivity. In each region (i = 0 for low latitudes and i = 1 for high latitudes), the general solution may be written (North, 1975) I (x) = A P (x) +B Q (x) i i l i l S P (x) 2 2 + Qa 1 + ; i = 0; 1 (17) 6D + 1 where A and B are constants to be determined by i i satisfying boundary conditions, and the parameter l is defined as D = : (18) l(1 + l) The boundary conditions require zero heat flux at the equator and pole. They also require continuity of gradient at x = x and at this point I = I = I , s 0 1 s which was computed using a predefined critical tem- perature for permanent ice caps (T = T  10 C). http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Astrophysics arXiv (Cornell University)

A shorter Archean day-length biases interpretations of the early Earth's climate

Astrophysics , Volume 2019 (1903) – Mar 9, 2019

Loading next page...
 
/lp/arxiv-cornell-university/a-shorter-archean-day-length-biases-interpretations-of-the-early-earth-sFBxFw5szG

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

ISSN
0012-821X
eISSN
ARCH-3330
DOI
10.1016/j.epsl.2019.02.032
Publisher site
See Article on Publisher Site

Abstract

A shorter Archean day-length biases interpretations of the early Earth’s climate 1 2 1 Christopher Spalding , and Woodward W. Fischer Division of Geological and Planetary Sciences California Institute of Technology, Pasadena, CA 91125 Department of Astronomy Yale University, New Haven, CT 06511 March 12, 2019 Abstract to global glaciation. We show that within the faster- spinning regime, the greenhouse warming required to Earth’s earliest sedimentary record contains evidence generate an ice-free Earth can differ from that re- that surface temperatures were similar to, or per- quired to generate an Earth with permanent ice caps haps even warmer than modern. In contrast, stan- by the equivalent of 1–2 orders of magnitude of pCO . dard Solar models suggest the Sun was 25% less Accordingly, the resolution of the Faint Young Sun luminous at this ancient epoch, implying a cold, problem depends significantly on whether the early frozen planet—all else kept equal. This discrepancy, Earth was ever, or even at times, ice-free. known as the Faint Young Sun Paradox, remains un- resolved. Most proposed solutions invoke high con- centrations of greenhouse gases in the early atmo- 1 Introduction sphere to offset for the fainter Sun, though current The Earth hosts an extensive variety of habitable geological constraints are insufficient to verify or fal- environments, which approximately track the places sify these scenarios. In this work, we examined sev- where liquid water persists. Liquid water, therefore, eral simple mechanisms that involve the role played is considered the most critical component of habit- by Earth’s spin rate, which was significantly faster ability as applied to other planets, ranging from Mars during Archean time. This faster spin rate enhances to extrasolar worlds (Kasting, 2010). The Earth’s ge- the equator-to-pole temperature gradient, facilitat- ological record has preserved over 4 billion years of ing a warm equator, while maintaining cold poles. global climate history, revealing that large bodies of Results show that such an enhanced meridional gra- liquid water have persisted throughout the majority dient augments the meridional gradient in carbon- ate deposition, which biases the surviving geological of Earth’s past. Accordingly, the Earth is not just record away from the global mean, toward warmer habitable today, but has remained habitable for bil- waters. Moreover, using simple atmospheric mod- lions of years (Grotzinger and Kasting, 1993). els, we found that the faster-spinning Earth was less Despite the empirically-derived presence of tem- sensitive to ice-albedo feedbacks, facilitating larger perate surface environments during Earth’s history, meridional temperature gradients before succumbing models of the Sun’s evolution lead to an apparent arXiv:1903.03817v1 [astro-ph.EP] 9 Mar 2019 contradiction. Standard models suggest that the sedimentary basins are rarely known and hard to Sun’s luminosity has been steadily increasing, such measure, making it difficult to ascertain whether or that 3.8 billion years ago, its luminosity was only 75% not the geological indicators of a warm climate (e.g. of its modern value (Gough, 1981). For greenhouse absence of glacial deposits) reflects a truly global con- effects similar to today, the planet would be expected dition (Evans and Pisarevsky, 2008). to enter a global snowball state at Solar luminosities The early atmospheric composition facilitated the of only  10% below today’s value (Hoffman et al., development of life and thus, before settling upon 1998; Yang et al., 2012). In spite of observations the concentration of greenhouse gases required to ex- to the contrary, the Sun’s history predicts a glob- plain the geological record, it is important to take full ally glaciated climate during most of the Archean era account of climactic drivers aside form atmospheric ( 3:8 2:5 Gya)(Budyko, 1969; Sellers, 1969; Kast- composition. One such influence is the Earth’s ro- ing et al., 1993; Ikeda and Tajika, 1999; Abbot et al., tation rate, which is known to have been decreas- 2011). ing monotonically over geological time owing to tides The discrepancy between the presence of liquid wa- raised on Earth from the moon (Touma and Wisdom, ter, and a faint early Sun has been dubbed the “Faint 1994). The most apparent consequence of a faster Young Sun Paradox” and gained widespread aware- rotation rate is an enhancement of the equator-to- ness owing to the work of Sagan and Mullen (1972). pole temperature gradient (North, 1975; Kaspi and A resolution to the paradox requires that the ancient Showman, 2015; Liu et al., 2017), which suggests Earth retained a substantially augmented fraction of that the Archean Earth exhibited a hotter equator incoming Solar heat than today, and/or that the re- relative to its poles than today. Our results suggest constructions of Solar luminosity are in error (Feul- that the extra warmth trapped at the equator gener- ner, 2012). ated signals in the geological record that would have A majority of previous work has proposed that tended toward higher temperatures, thereby creating the Archean greenhouse atmospheric composition dif- a naturally-biased representation of the mean global fered radically from today in order to accommodate Archean climate. the fainter Sun (Kasting et al., 1993; Wordsworth and Pierrehumbert, 2013). Under certain scenarios, cli- mate models possessing heightened levels of green- 2 Length of day house gases, such as CH and CO , successfully re- 4 2 produce the required global temperatures under a The length of Earth’s day was undoubtedly shorter in faint early Sun. Indeed, a promising negative feed- the past, but an accurate reconstruction back in time back has long been known to exist for generating ad- is currently beyond model capabilities, largely owing ditional CO when the climate cools (Walker et al., to the uncertain tidal dissipation rates throughout 1981). Earth’s history (Touma and Wisdom, 1994; Egbert Despite the promise shown by the hypothesis of and Ray, 2000). Growth bands in the skeletons of elevated greenhouse warming, geological constraints marine organisms reveal a Cambrian day length of upon the Archean atmospheric composition are un-  21 hours (Williams, 2000). However, reconstruc- able to adequately distinguish between paleoclimate tions of greater antiquity, before abundant availabil- models (Feulner, 2012). The Archean geological ity of animal skeletons, are highly controversial. De- record is sufficiently sparse and fragmented such that spite such limitations, it has been suggested that a it is not clear whether the early Earth was ice-free, 21 hour resonance in atmospheric thermal tidal forc- or whether ice caps persisted (de Wit and Furnes, ing maintained a 21 hour day back as far as 2.5 Gya 2016). Below we show that distinguishing between (Zahnle and Walker, 1987; Bartlett and Stevenson, these two scenarios is critical to evaluating the like- 2016), and recent stratigraphic techniques are emerg- lihood that greenhouse gases can feasibly solve the ing that may soon improve the reliability of Precam- paradox. Moreover, the paleolatitudes of Archean brian day length reconstructions (Meyers and Malin- 2 Solar luminosity (relative to modern) verno, 2018). of 2 (Figure 1). The time spent between day-lengths The evolution of the Earth-Moon semi-major axis of 6 and 9 hours is geologically brief, such that for may be written as (Bills and Ray, 1999) most of the Archean, the Earth was likely rotating at a rate above 6 hours. da 11=2 = fa (1) dt 24 1 Archean era 21-hour resonance model where f is a dimensional quantity that encodes in- 0.95 formation related to Earth’s dissipative properties and a is the Earth-Moon semi-major axis (assuming 0.9 negligible orbital eccentricity). Today, f  6:29 45 13=2 1 10 m yr , measured using the recession of the 0.85 moon’s orbit. Owing to conservation of angular mo- ocean model mentum, the Earth-Moon separation is trivially con- 0.8 60% modern modern tidal verted to a length of day for the Earth (assuming dissipation dissipation constancy of structural parameters). A length of day 0.75 3.5 3 2.5 2 1.5 1 0.5 0 4.5 4 of 6 hours corresponds to an orbital distance of about time before present (Gyr) 9.26 Earth radii. Such a close orbit would likely leave observable features upon the Earth and so we con- Figure 1: The length of Earth’s day as a function of sider it as a lower limit for the Earth’s day-length. time given different assumptions about how tidal dis- The evolution of Earth’s length of day is shown sipation has changed over Earth history. Red, black in Figure 1 for four different possible scenarios, each and gray lines reflect the evolution under modern, differing in their assumptions regarding variations in 60% modern and 35% modern tidal dissipation rates. tidal dissipation over time. Assuming a modern dis- The blue line, drawn from Webb (1982), was derived sipation agrees well with geological proxies back to from a model of tidal dissipation with a more sophis- the Cambrian Period (Williams, 2000; Meyers and ticated model of the oceans. All results show that Malinverno, 2018), but predicts that the Earth and tidal dissipation must be less than 60% of today av- Moon came within 10 Earth radii about 1.5 Gyr into eraged over the history of the Earth-moon system, the past—a condition wholly inconsistent with ob- given the lack of evidence for a cataclysmic origin servations from the Proterozoic and Archean record. of the moon until prior to  4 Gya. By the end Even dissipation 60% of today’s value suggests an in- of Archean time, the day-length was likely above 12 ferred age of the Moon that is too young. Assuming hours, having changed substantially during that era. 35% of today’s dissipation produces a reasonable age The gold line represents the predicted solar luminos- of the moon, but conflicts with the Phanerozoic sed- ity following the computation within Gough (1981). imentary record. The third model (blue line) was derived by Webb (1982), and includes a careful computation of the dis- In order to take account of the range of possibili- sipation within an ocean model; it agrees better than ties, we investigated spin rates ranging from mod- a simple modulation of f in equation 1. The fourth ern to the equivalent of 6 hours, i.e., 1 4, hypothesis (shown by a grey line) fits best, which is and denotes the modern-day spin angular veloc- the 21-hour resonance model as presented in Bartlett ity. This choice of range is in part owing to recent and Stevenson (2016). General Circulation Model (GCM) simulations that Given that tides get stronger as the moon is closer, performed numerical experiments across this 4-fold the day length would increase at a faster rate dur- range of spin rates (Liu et al., 2017), allowing us to ing the Archean than during later time periods, such calibrate our simplified mathematical framework us- that the early and late Archean worlds would differ ing results of more complex models. However, the markedly in their rotation rates, perhaps by a factor 6 hour case should be seen as a lower limit to the day length of day (hours) day-length known length—most of Archean time was likely character- order to retain the same equator-to-pole heat flux, a ized by a slower spin rate, between 9 and 18 hours steeper temperature gradient is retained. The value (Figure 1). of K cannot be derived from first principles, and eff must either be measured directly, or extracted from General Circulation Models. In this work, we chose 3 Influence of Earth’s spin rate values of K as inferred from the simulations per- eff formed in Liu et al. (2017). We fitted the depen- Earth’s climate is driven by the distribution of Solar dence of diffusivity and spin rate from their simula- insolation, with more energy being received at the tions with the relationship equator than at the poles. Owing to the presence of the atmosphere, heat is transported from hotter eff = ; (3) to cooler latitudes via turbulent processes that may 0 0 be modelled as an effective diffusivity, which lowers the meridional temperature gradient. We write this 6 2 1 where n = 0:7 and K  1:7 10 m s . diffusivity as (see appendix) The precise magnitude of K varies between in- eff vestigations. The important feature here is the re- D  K ; (2) eff lationship between and K , which we fixed at eff gR 0:7 K / . However, it is interesting to compare eff 3 1 1 where c = 10 J kg K is the specific heat of air, this to dependences derived elsewhere. Kaspi and = 1 bar is the surface air pressure, g = 10 m s is Showman (2015) performed global climate simula- the acceleration due to gravity and R = 6400 km is tions at various spin rates within the context of extra- the Earth’s radius. K represents the efficiency of solar planets. They did not explicitly measure K , eff eff meridional heat transfer by way of turbulent eddies but the heat flux from eddies F was presented. In (North, 1975), and is the parameter most directly a diffusive parameterization, F / K =T , where E eff influenced by rotation rate, as we discuss below. T is the equator-to-pole temperature difference. Equation 2 indicates that meridional heat trans- Reading F and T from their figure 8 yields an 0:71 port depends not just upon spin rate, but also upon approximate relationship of K / , which eff atmospheric pressure and heat capacity (Williams is remarkably similar to that used here, as derived and Kasting, 1997). Some evidence has been put for- from Liu et al. (2017). However, the earlier work of ward that the atmospheric pressure differed in the Williams and Kasting (1997) utilized a relationship past (Goldblatt et al., 2009), though theory and geo- where n = 2: a much steeper relationship. Accord- logical proxies often disagree over whether the pres- ingly, the dependence upon spin rate may be even greater than that derived here, though the qualita- sure was more or less than today. If the atmosphere tive picture will not change. was thick with CO , its heat capacity may have been larger, additionally helping to warm and homoge- A rise in meridional temperature gradient is a ro- nize the global climate. However, preserved raindrop bust outcome of global climate models incorporat- imprints, and the size distribution of basaltic vesi- ing faster spin rates (Kaspi and Showman, 2015; Liu cles (Som et al., 2016), suggest atmospheric pressures et al., 2017). However, an alternative approach used similar to, or lower than, modern. Given these dis- in previous work is to utilize the principle of maxi- crepancies, we focus upon the influence of spin rate, mized entropy production to derive meridional tem- but note that a full understanding of the ancient cli- perature gradients (Lorenz et al., 2001). Though this mate requires an auxiliary treatment of the interplay approach was met with moderate success for Titan between, as yet uncertain, changes in atmospheric and Mars, the methodology has been criticized, and composition, alongside eddy-driven heat transport. its utility has been questioned (Goody, 2007). Thus, Generally, the magnitude of K is reduced at we continue with the assumption that rotation rate eff faster rotation rates (Liu et al., 2017), such that in plays an important role in heat transport via the at- 4 unstable branches mosphere, owing to the extensive and consistent lit- theoretical approach is that within a specific range erature surrounding these mechanics. of Solar luminosities, multiple stable solutions ex- ist. These include an ice-covered state, an ice-free More uncertain than the relationship between state, along with a solution possessing ice extend- meridional temperature gradient and spin rate, is the ing from the poles to an “ice-line” latitude denoted influence of spin upon global mean temperature. As . The value of  was computed using a pre- the planet spins faster, the Hadley cell is reduced s s defined critical temperature for permanent ice caps in meridional extent (Kaspi and Showman, 2015), (T = T  10 C). The simplicity of this model which potentially reduces cloud coverage and there- s allows a mechanistic exploration of the relationship fore albedo. Consequently, more Solar insolation is between spin rate and the ice-albedo feedback. absorbed and the planet warms (Liu et al., 2017). This argument is encouraging for the faint young Sun problem, however, it is important to note that differ- 2 x today’s ent global climate models have exhibited contradic- today’s spin rate spin rate tory results regarding the impact of spin rate upon 4 x today’s global cloud cover and global temperature (Vallis and spin rate Farneti, 2009). Thus we focus here predominantly upon the meridional temperature profile and the ice- faster spin, weaker feedback albedo feedback, rather than upon the global mean temperature. In what follows, we argue that the enhanced 15 meridional temperature gradient favors the deposi- tion of sedimentary rocks at low latitudes, skewing 0.9 0.95 1 1.05 1.1 1.15 1.2 the Archean record toward warmer temperatures. We Stellar luminosity (relative to present Sun) identified that rotation rate and resulting temper- ature distribution, have two primary influences on Figure 2: The equilibrium ice line (latitude at which interpretations of the ancient climate record. First, T = 10 C) as a function of Solar luminosity, for 3 the most basic interpretation of the Archean record spin rates. The solution is stable if the slope is posi- is that the planet was not globally glaciated. We dis- tive (solid lines) and unstable if the slope is negative cuss the role of a faster rotation rate in decreasing (dashed lines). The reduced slope within the stable sensitivity of the planet to the ice-albedo feedback. regions for higher spin rates amounts to a greater re- Second, we discuss the tendency of a faster-rotating silience to the ice-albedo feedback. temperature distribution to have biased the deposi- The zonally-averaged temperature at any given lat- tion pattern of Archean sedimentary rocks toward itude was computed as a balance between incom- warmer waters, and in turn, biasing interpretations ing, short-wave radiation (from the Sun), out-going of the ancient climate. longwave radiation (from the Earth and its atmo- sphere), together with the effective diffusion of heat between latitudes, with efficiency K (defined in sec- eff 4 Spin rate and the ice-albedo tion 2). The diffusive term represents turbulent phe- feedback nomena that tend to transport heat from the equator to the poles, such as baroclinic instabilities (Kaspi In order to analyze the influence of rotation rate upon and Showman, 2015). the planet’s susceptibility to the ice-albedo feedback, Table 1 contains the values for the various pa- we constructed a simple energy balance model based rameters used, which were largely taken from North upon the Budyko-Sellers framework (Budyko, 1969; (1975). In Figure 2 we plotted the solution for the ice Sellers, 1969; North, 1975). A characteristic of this line as a function of Solar luminosity for 3 values of multiple stable states Ice-line latitude (degrees) Parameter values the faster the spin rate, the weaker the feedback. 2 2 1 a a S A (Wm ) B (Wm K ) 0 1 2 The aforementioned influence upon the feedback 0.68 0.38 0:482 196:2 1:953 strengths lead to two critical and complementary findings. The first, is that ice may extend further Table 1: The parameter values chosen to solve the toward the equator without triggering a global snow- Budyko-Sellers type energy balance model. Val- ball, providing a greater chance that the equator may ues taken from (North, 1975), except A which was be kept ice-free despite low Solar luminosity. How- tuned to match today’s climate with with K = eff ever, a second, larger effect is that a lower mini- 6 2 1 1:7 10 m s mum luminosity is required to entirely deglaciate the planet from a glaciated starting point. This impacts 30 solutions to the faint young Sun paradox because there are substantial differences based on whether or not the early Earth was ice free (even at the poles). Figure 2 captures the result that an ice-free condition ⌦ = ⌦ 0 is significantly more difficult to achieve under faster ice-line -10 spin rates. -20 ⌦ =2⌦ -30 ⌦ =4⌦ 5 Depositional biases -40 -50 The inference of a warm Archean Earth is based 0 15 30 45 60 75 90 largely upon evidence from sedimentary rocks as- Latitude (degrees) sociated with early cratons and marine platforms (Grotzinger and Kasting, 1993). Though controver- Figure 3: Meridional temperature profile arising from sial, absolute temperature estimates have been at- the Budkyo-Sellers model for 3 spin rates. Upon com- tempted using these records, from which the overly- parison to Figure 1, the presence of ice is capable of ing marine waters appear to have been at least as decreasing polar temperatures while maintaining a warm as surface seawater today (Knauth and Lowe, warm equator. Faster spins enhance this equator-to- 2003). However, a key unknown with respect to these pole difference. early archives is the latitude, and often the environ- ment, of deposition. In this section, we examine how the lithotypes used to infer Archean climate are typi- spin rate, translating to 3 values of K . The modern eff cally biased toward sedimentation in warmer waters, climate possesses ice caps down to roughly 72 of lat- thereby skewing the implied Archean climate toward itude, and this was used to tune the parameters for warmer temperatures. In conjunction with the en- the blue line, representing today’s spin rate (North, hanced equator-to-pole temperature gradient associ- 1975). The other two lines represent spin rates of 2 ated with a faster rotation rate, Archean climate may and 4 have been significantly cooler than implied by a naive Although the earliest descriptions of the faint reading of the geochemical record. young Sun paradox were motivated by raising global mean temperature, more recent treatments outline the importance the Earth’s susceptibility to global 5.1 Meridional bias in carbonate de- glaciations (Kasting et al., 1993; Ikeda and Tajika, position 1999). A drop in Solar insolation leads to the spread of ice, which causes an increase in albedo. In reflect- Some of the most reliable evidence for a temperate ing extra sunlight, this additional albedo amplifies Archean climate stems from the existence of shallow the initial drop in insolation. Figure 2 indicates that marine carbonate platforms (Sumner and Grotzinger, Temperature ( C) 3.5 3.5 ⌦ = ⌦ and James, 2000; Ridgwell et al., 2003; Higgins et al., ⌦ =2⌦ 2009; Fischer and Knoll, 2009). ⌦ =4⌦ Key differences with modern times include a lack of 2.5 2.5 skeletonized metazoa and open-ocean planktic calci- fiers. In such a global regime, abiotic mechanisms of kinetic preference for equatorial carbonate production would have dominated the car- deposition of aragonite 1. 1.5 5 bon cycle; fluxes of dissolved inorganic carbon and al- kalinity into the oceans would be balanced almost en- tirely by abiotic precipitation (Ridgwell et al., 2003). 0. 0.5 5 Moreover, the predominance of alkalinity-favoring, anaerobic respiration in shallow sediments would am- 0 15 30 45 60 75 90 0 15 30 45 60 75 90 plify carbonate precipitation and preservation within Latitude (degrees) sediments (Higgins et al., 2009). In Archean basins, carbonate precipitation would Figure 4: The distribution of carbonate deposition have predominantly occurred in shallow-water shelf rates as a function of latitude for 3 spin rate scenar- environments, with a spatial distribution governed ios (modern; blue, twice modern; red, and four times largely by temperature. Inferences regarding global modern; black) for the Budyko-Sellers climate mod- climate by way of Archean carbonates must therefore els presented in the text. All curves are normalized acknowledge the existence of a strong chemical pref- to have an area under the curve of unity, such as to erence, both kinetically and thermodynamically, for highlight the meridional distribution of deposition, precipitation of these rocks within the warmest shal- as opposed to the absolute magnitude, which would low water environments on the planet. In what fol- depend more sensitively upon whole ocean carbonate lows, we estimate the expected bias associated with a chemistry. Notice that, particularly when ice occurs rotationally-enhanced equatorial temperature gradi- at the poles, the rate of aragonite deposition has de- ent, and argue that the widespread occurrence of liq- creased by an order of magnitude by  30 40 and uid water observed may simply be reflecting a warm has essentially dropped to zero by  60 . equator, kept stable by the fast-rotating planet. 1996). In addition, widespread measurements of sta- 5.2 Kinetic model of Carbonate depo- ble oxygen isotope ratio data from carbonates and sition cherts have have been interpreted as indicative of high temperatures (Knauth and Lowe, 2003). Ac- The saturation state of seawater with respect to arag- cordingly, inferences regarding the paleoenvironment onite decreases toward the poles in the modern of the Archean by way of sedimentary records are fil- oceans (Jiang et al., 2015). This trend generally tered through the lens of geochemical environments arises from two causes: 1) aragonite is more soluble favoring the deposition of carbonates and cherts. in colder waters, and 2) the carbonate system adjusts In modern oceans, calcium carbonate salts are pref- in the presence of an atmosphere containing CO to erentially deposited in warm, tropical waters. This raise the concentration of Dissolved Inorganic Car- pattern reflects the evolution and ecology of organ- bon (DIC) in colder waters. In today’s oceans, the isms with carbonate skeletons, but it is also directly saturation state of aragonite varies with temperature related to the inorganic chemistry of the carbonate approximately linearly as system (Zeebe and Wolf-Gladrow, 2001), whereby warmer waters are more saturated with respect to a(T= C) + b: (4) CaCO salts. The dominant mechanisms of carbon- ate precipitation during Precambrian time remain where a  0:09 and b  1:5 in the modern ocean uncertain (Grotzinger and Reed, 1983; Grotzinger (Jiang et al., 2015). g( ) Fraction of aragonite deposition Carbonate compensation is broadly considered To illustrate the effect of temperature, we defined to have been limited by over geological time the relative rate of aragonite deposition within a lat- (Grotzinger and James, 2000; Ridgwell et al., 2003), itudinal band d as g()d, where, 2+ as opposed to Ca -limitation. Accordingly, no mat- g() = r() cos(); (6) ter the precise dependence of upon temperature and latitude, the Precambrian carbon cycle is ex- which follows from the definition of r as a rate per pected to have exhibited a drop in from the equa- unit area on the globe. The absolute magnitude of tor toward the poles. k is unimportant for our investigation; what mat- To proceed, we adopted this relationship for the ters is the relative precipitation rate at each latitude. modern, but note that carbonate deposition in Pre- Accordingly, we treated k as a normalization factor cambrian oceans may not have exhibited a linear de- that satisfies pendence. The values of a (the meridional gradient) " # =2 and b (the saturation state at the freezing point) are g()d = 1: (7) also likely to change, but not independently. In par- ticular, if a is raised, in order to achieve an equivalent Figure 4 illustrates the relative degree of carbonate global precipitation of carbonate, b must drop. Phys- precipitation as a function of latitude. The temper- ically, if saturation state varies more markedly with ature T () was taken from the Budyko-Sellers tem- latitude, the saturation state at high latitudes may be lower while retaining an equivalent global deposi- perature model (Figure 3), using an equivalent so- tion. A stronger or weaker dependence will constitute lar forcing of the modern day. Note that no depo- an additional determinant upon potential biases that sition occurs at latitudes that are ice-covered year- we do not tackle here. Our goal is simply to compare round. Latitudes with annual mean temperatures the biases associated with different rotation rates, for below freezing, but above T = 10 C will allow which it suffices to approximate the dependence as deposition during the fraction of the year when ice is linear across all rotation rates. not present. Though strictly unphysical, the rate at We used the dependence given by equation 4 to these latitudes is reflected by allowing for sub-freezing construct an idealized kinetic model of carbonate de- temperatures in the computation of g(), rather than position. The precipitation rate per unit area of cal- fixing the temperature at the freezing point. The two cium carbonate is generally written (Zeebe and Wolf- choices do not significantly differ, given the slow pre- Gladrow, 2001) cipitation rates at such low temperatures relative to the equator. r() = k exp(E =R T ())( (T ()) 1) ; (5) 0 a g A where k is a reference rate, is the saturation 0 A 5.3 Preference for equatorial carbon- state of aragonite, n  1:7 (Romanek et al., 2011) ate deposition is the order of the reaction, E  71 kJ mol is the 1 1 activation energy and R = 8:31J mol K is the As is illustrated in Figure 4, a significant enhance- universal gas constant. ment of deposition rates near the equator occurs as The rate of the reaction depends upon temper- a result of the combination of enhanced tempera- ature in two fundamentally different aspects. The tures and surface area at low latitudes. An order of first is that the saturation state itself varies with lat- magnitude enhancement of carbonate deposition oc- itude, being larger in warmer tropical waters. How- curs at the equator relative to the lower latitudes of ever, the dominant effect lies in the Arrhenius factor  30 40 , and deposition drops to zero close to the exp(E =R T ), such that if the whole ocean is to ice-line. Crucially, the existence of systematic bias a g balance the fluxes of ions entering it, calcium car- is present for all 3 rotation rates, meaning that the bonate will precipitate faster in the warmer regions Precambrian climate record exhibits an ever-present owing to a kinetic enhancement of reaction rates. tendency towards an equatorial bias. 8 Despite the presence of bias irrespective of rota- tion rate, the magnitude and interpretation of the bias depends upon rotation rate. Upon comparison mean temperature of carbonates to Figures 2 and 3, at 1 Solar luminosity, ice extends rotationally-induced significantly further toward the equator at higher ro- bias toward high temperatures tation rates than modern. In other words, receiving information only about lower latitudes gives a signif- icantly more biased picture of global mean temper- mean global temperature atures when rotation rates are higher. In order to quantify this bias, we defined the true mean global temperature as 1 2 3 4 R Earth’s spin rate (relative to today) =2 T () cos()d =2 Figure 5: The true, global mean temperature T (up- cos()d per, red line) and the temperature that would be in- =2 = T () cos()d: (8) ferred from biased carbonate deposition T (lower, obs blue line) as a function of rotation rate. We set lu- minosity equal to Solar for each. Despite the spread In contrast, the averaged signal captured by paleocli- of ice sheets leading to a drop in T , the temperature mate records in the geological record will be skewed inferred from the geological record would increase ow- by the kinetic factor g(), such that the mean inferred ing to a strong preference toward equatorial deposi- from the record is given by tion. =2 T = g()T ()d: (9) obs Knoll, 2009). There appear also to have been pri- mary modes of chert deposition as sedimentary parti- Using the definitions above, we computed the de- cles generated from waters supersaturated with silica pendence of both T and T as a function of , illus- obs (Stefurak et al., 2014). If indeed much of Archean trated in Figure 5. Notably, faster rotation rates led silica deposition was driven by evaporation (Maliva to an enhancement of the mean temperature of pre- et al., 2005; Fischer and Knoll, 2009), the most likely cipitated carbonates, despite a reduced global mean bias in chert formation is at the descending limb temperature. A strong bias remained for all rotation of the Hadley cell, which moves equator-ward at rates, but the magnitude of this bias was amplified faster spin-rates (Kaspi and Showman, 2015). Given as rotation rate increased. Accordingly, care must be these uncertainties, we did not develop a quantita- taken when inferring a warm Archean climate, sim- tive model of chert deposition, but to the degree that ply from an apparently warm signal in sedimentary this is correct, the potential climate bias from obser- deposits. vations of chert will plausibly follow a qualitatively The biases affecting the chert record are less clear, similar trend to carbonate deposition. in large part because we know far less about the de- positional mechanisms of Precambrian cherts. How- ever, in many cases cherts are likely to have been 5.4 Paleolatitude of shelf environ- associated with diagenetic replacements of carbon- ments ate rocks (Maliva et al., 2005). The paleoenviron- ment of many cherts associated with carbonate plat- In the previous subsection, we demonstrated that a forms suggests that they have mechanisms of precip- relatively hotter equator, at higher rotation rates, itation wherein evaporative concentration and over tends to bias the carbonate record toward warmer saturation may have been important (Fischer and environments. However, a key assumption was that Temperature ( C) 1.2 1.2 all of the Earth’s ocean surface was available for car- bonate deposition (meaning that in the computation 1.15 1.15 of T , we simply averaged over all latitudes). In obs 1.1 1.1 ice-free reality, geological observations have shown that car- permanent polar ice caps bonate deposition during Precambrian time was pro- 1.05 are possible 1.05 nounced in shallow shelf environments (Grotzinger and Reed, 1983; Sumner and Grotzinger, 1996; Ridg- well et al., 2003). A test, in principle, of the sug- 0.95 0.95 gestion that the Archean record is biased toward the 0.9 0.9 equator, would be to reconstruct the paleolatitudes of snowball sedimentary strata from which the paleoclimate data 1 1 2 2 3 3 4 4 were generated. Earth’s spin rate (relative to today) Limited paleomagnetic evidence from the Bar- berton Greenstone belt—a critical archive of early Figure 6: The range of Solar luminosities facilitating Archean paleoclimate data currently—suggests a lo- a partially ice-covered state as a function of Earth’s cation close to the paleoequator at  3:5 Ga (Kröner spin rate (see Figure 4). At = 4 , the increase in and Layer, 1992). Unfortunately, the paleolatitudes Solar luminosity required to deglaciate the planet is of Archean cratons, and indeed the nature of plate many times larger than that required at modern spin motions at such early times, is debated (Evans and rates. Pisarevsky, 2008; Korenaga, 2013), reducing confi- dence in specific reconstructions. planet and amplified by a climate system that was We took a different approach to examine this using less efficient at moving heat from the equator. This the distribution of paleolatitudes for continental cra- raises the hypothesis that the poles may have retained tons for more recent times, and assume that the same cold ice caps while the planet preserved paleoclimate distribution holds for earlier, Archean, times. Our data from the warm equator. calculation above of T essentially assumed that obs over geological time, the distribution of continental cratons is isotropic in latitude. In Figure 7 we con- 6 Implications for the Faint structed a histogram of 45 robust, paleolatitude mea- surements from Evans and Pisarevsky (2008) all more Young Sun Paradox ancient than 800 Ma, and compared it to an isotropic (random) distribution. If no latitude was more likely, A fundamental aspect of the Archean Earth was that the distribution would approximate a cosine—more its day was significantly shorter—and it is valuable to deposition occurs at lower latitudes in proportion to frame problems regarding early climate in this con- the greater surface area of the Earth available there. text. The enhanced spin rate would have influenced The measured distribution closely approximates that the transport of heat and the distribution of temper- expected from no preferred latitude, validating our atures over the globe. In turn this would have im- computation of T . pacted the locus of deposition of chemical sediments. obs To conclude, the Archean oceans, and global mean Archean Earth, like the modern, deposited calcium temperature more broadly, are likely not as warm as carbonate salts in order to balance weathering fluxes a straightforward read of the archive suggests. The into the oceans. As opposed to the biogenic domi- faster-spinning Earth augmented the equatorial tem- nance of such deposition today, the Archean carbon- perature relative to the poles, substantially increas- ate cycle would largely have been driven by abiotic ing the deposition rate of carbonate at low latitudes influences, principally temperature. Combined with relative to the poles. The surviving record, thus, is the enhanced equatorial temperatures, this leads to intrinsically biased toward the warmer parts of the a significant, kinetic bias toward deposition of car- Solar luminosity (relative to modern) bonates at low latitudes (Figure 4). Given that a stable. substantial fraction of the surviving Archean sedi- If an ice-free Earth is required to match the ge- mentary record formed as carbonate, our picture of ological record, a resolution to the faint young Sun the Archean world is skewed towards these warmer problem may require revisiting the founding premise temperatures. behind the whole paradox. I.e., precisely how well We also demonstrated that a faster-rotating Earth do we understand the Solar luminosity over time? If, was generally less sensitive to the ice-albedo feedback. for example, the Sun’s mass had been a few percent This lower sensitivity makes it substantially more dif- higher in the Archean, its higher luminosity would ficult to generate an ice-free planet 3.5 billion years have valuable implications for how we think about ago. Simply put, the Faint Young Sun paradox is far the composition of Earth’s early atmosphere (Feul- easier to explain if Earth always maintained ice at ner, 2012). This idea is promising and remains to the poles than it would be if the planet did not (Wolf be thoroughly tested (Spalding et al., 2018), but is and Toon, 2013). Over the past 500 million years, we somewhat inconsistent with mass-loss rates observed have a decent understanding of when the planet was in Sun-like stars. glaciated and when it was not, from a combination The Archean sedimentary record presents us with of sedimentological and geochemical datasets. a valuable, yet limited set of robust windows into Going further back in time the geological record its climate, and drivers of habitability. It is criti- is sufficiently fragmented that it becomes extremely cal to thoroughly understand the filters applied at challenging to discern glacial and ice-free inter- the moment of sedimentation in order to reconstruct vals. The gold standard geological observations for the conditions that truly persisted soon after the ori- early Earth’s climate come from glacial deposits gin of life. We demonstrated the key role played by (diamictites, ice-rafted debris, striated clasts and rotation rate in sculpting the distribution of temper- pavements)—but these records are inherently sparse. ature, and therefore carbonate deposition, upon the The earliest observations of glacial rocks in the record early Earth. These biases yield an apparently warmer come from the  3 billion-year-old Pongola and Archean than may have existed in reality. Witswatersrand basins of South Africa (Young et al., 1998); yet earlier glacial deposits from  3:5 Ga have been described (de Wit and Furnes, 2016) though al- Acknowledgements ternative non-glaciogenic interpretations remain pos- sible. Results shown here, however, illustrate just C.S thanks the generous support of the NASA how valuable those sedimentological and geochemi- NESSF Graduate Fellowship in Earth and Planetary cal observations are. If further geological investiga- Science, and of the Heising-Simons Foundation’s 51 tion finds sufficiently strong evidence of ice-free states Pegasi b Postdoctoral Fellowship. W.F. acknowl- of the early Earth, then the faint young Sun paradox edges support from the Simons Foundation Collabo- becomes exceedingly more difficult to solve. ration on the Origins of Life. Additionally, we thank In a simple global model, Solar luminosity may be Konstantin Batygin, Greg Laughlin, Dorian Abbot considered as an equivalent to greenhouse gas forcing. and Seth Finnegan for illuminating conversations. Fi- Figure 2 in Ikeda and Tajika (1999) demonstrates a nally, we thank several referees for their important similar result to our Figure 2, but with pCO in place contributions to the work. of Solar luminosity. A comparison between our result and theirs shows that upper limit of the region facili- tating multiple stable states corresponds to 12 or- ders of magnitude greater pCO under Archean spin References rates than the modern equivalent. In turn, roughly 1 order of magnitude lower pCO levels are tolerated Abbot, D. S., Voigt, A., and Koll, D. (2011). The jor- at faster spins before only a single snowball state is mungand global climate state and implications for 11 14 evidence. When did plate tectonics begin on planet Evans & Pisarevsky (2008) Earth, 440:249–263. (paleomagnetic data) 10 Feulner, G. (2012). The faint young sun problem. Reviews of Geophysics, 50(2). 8 uniform global distribution Fischer, W. W. and Knoll, A. H. (2009). An iron shut- tle for deepwater silica in late archean and early pa- leoproterozoic iron formation. Geological Society of America Bulletin, 121(1-2):222–235. Goldblatt, C., Claire, M. W., Lenton, T. M., 0 15 30 45 60 75 90 Matthews, A. J., Watson, A. J., and Zahnle, K. J. paleo - latitude (degrees) (2009). Nitrogen-enhanced greenhouse warming on early earth. Nature Geoscience, 2(12):891. Figure 7: The distribution of paleolatitudes older than 800 Ma in a dataset complied by Evans and Goody, R. (2007). Maximum entropy production in Pisarevsky (2008) (bars). If the distribution was uni- climate theory. Journal of the atmospheric sci- formly distributed across the globe, the bars and red ences, 64(7):2735–2739. line should coincide. The data show that low latitude Gough, D. (1981). Solar interior structure and lumi- cratons—and their sedimentary basins—are slightly nosity variations. In Physics of Solar Variations, overrepresented. pages 21–34. Springer. Grotzinger, J. and Reed, J. (1983). Evidence for neoproterozoic glaciations. Journal of Geophysical primary aragonite precipitation, lower proterozoic Research: Atmospheres, 116(D18). (1.9 ga) rocknest dolomite, wopmay orogen, north- Bartlett, B. C. and Stevenson, D. J. (2016). Analysis west canada. Geology, 11(12):710–713. of a precambrian resonance-stabilized day length. Grotzinger, J. P. and James, N. P. (2000). Precam- Geophysical Research Letters, 43(11):5716–5724. brian carbonates: evolution of understanding. Bills, B. G. and Ray, R. D. (1999). Lunar orbital evo- Grotzinger, J. P. and Kasting, J. F. (1993). New con- lution: A synthesis of recent results. Geophysical straints on precambrian ocean composition. The Research Letters, 26(19):3045–3048. Journal of Geology, 101(2):235–243. Budyko, M. I. (1969). The effect of solar radia- Higgins, J., Fischer, W., and Schrag, D. (2009). Oxy- tion variations on the climate of the earth. tellus, genation of the ocean and sediments: consequences 21(5):611–619. for the seafloor carbonate factory. Earth and Plan- de Wit, M. J. and Furnes, H. (2016). 3.5-ga hy- etary Science Letters, 284(1-2):25–33. drothermal fields and diamictites in the barberton Hoffman, P. F., Kaufman, A. J., Halverson, G. P., greenstone belt paleoarchean crust in cold environ- and Schrag, D. P. (1998). A neoproterozoic snow- ments. Science advances, 2(2):e1500368. ball earth. science, 281(5381):1342–1346. Egbert, G. and Ray, R. (2000). Significant dissipa- Ikeda, T. and Tajika, E. (1999). A study of the tion of tidal energy in the deep ocean inferred from energy balance climate model with co2-dependent satellite altimeter data. Nature, 405(6788):775. outgoing radiation: Implication for the glaciation Evans, D. A. and Pisarevsky, S. A. (2008). Plate tec- during the cenozoic. Geophysical Research Letters, tonics on early earth? weighing the paleomagnetic 26(3):349–352. frequency Jiang, L.-Q., Feely, R. A., Carter, B. R., Greeley, Meyers, S. R. and Malinverno, A. (2018). Protero- D. J., Gledhill, D. K., and Arzayus, K. M. (2015). zoic milankovitch cycles and the history of the so- Climatological distribution of aragonite saturation lar system. Proceedings of the National Academy state in the global oceans. Global Biogeochemical of Sciences, 115(25):6363–6368. Cycles, 29(10):1656–1673. North, G. R. (1975). Analytical solution to a simple climate model with diffusive heat transport. Jour- Kaspi, Y. and Showman, A. P. (2015). Atmo- nal of the Atmospheric Sciences, 32(7):1301–1307. spheric dynamics of terrestrial exoplanets over a wide range of orbital and atmospheric parameters. Ridgwell, A. J., Kennedy, M. J., and Caldeira, The Astrophysical Journal, 804(1):60. K. (2003). Carbonate deposition, climate sta- bility, and neoproterozoic ice ages. Science, Kasting, J. (2010). How to find a habitable planet. 302(5646):859–862. Princeton University Press. Romanek, C. S., Morse, J. W., and Grossman, E. L. Kasting, J. F., Whitmire, D. P., and Reynolds, R. T. (2011). Aragonite kinetics in dilute solutions. (1993). Habitable zones around main sequence Aquatic geochemistry, 17(4-5):339. stars. Icarus, 101(1):108–128. Sagan, C. and Mullen, G. (1972). Earth and mars: evolution of atmospheres and surface tempera- Knauth, L. P. and Lowe, D. R. (2003). High archean tures. Science, 177(4043):52–56. climatic temperature inferred from oxygen isotope geochemistry of cherts in the 3.5 ga swaziland su- Sellers, W. D. (1969). A global climatic model based pergroup, south africa. Geological Society of Amer- on the energy balance of the earth-atmosphere sys- ica Bulletin, 115(5):566–580. tem. Journal of Applied Meteorology, 8(3):392–400. Korenaga, J. (2013). Initiation and evolution of Som, S. M., Buick, R., Hagadorn, J. W., Blake, T. S., plate tectonics on earth: theories and observations. Perreault, J. M., Harnmeijer, J. P., and Catling, Annual Review of Earth and Planetary Sciences, D. C. (2016). Earth’s air pressure 2.7 billion years 41:117–151. ago constrained to less than half of modern levels. Nature Geoscience, 9(6):448. Kröner, A. and Layer, P. (1992). Crust formation Spalding, C., Fischer, W. W., and Laughlin, G. and plate motion in the early archean. Science, (2018). An orbital window into the ancient 256(5062):1405–1411. sun’s mass. The Astrophysical Journal Letters, Liu, X., Battisti, D. S., and Roe, G. H. (2017). The 869(1):L19. effect of cloud cover on the meridional heat trans- Stefurak, E. J., Lowe, D. R., Zentner, D., and Fischer, port: Lessons from variable rotation experiments. W. W. (2014). Primary silica granules-a new mode Journal of Climate, 30(18):7465–7479. of paleoarchean sedimentation. Geology, 42(4):283– Lorenz, R. D., Lunine, J. I., Withers, P. G., and McKay, C. P. (2001). Titan, mars and earth: Sumner, D. Y. and Grotzinger, J. P. (1996). Were Entropy production by latitudinal heat transport. kinetics of archean calcium carbonate precipita- Geophysical Research Letters, 28(3):415–418. tion related to oxygen concentration? Geology, 24(2):119–122. Maliva, R. G., Knoll, A. H., and Simonson, B. M. (2005). Secular change in the precambrian silica Touma, J. and Wisdom, J. (1994). Evolution of the cycle: insights from chert petrology. Geological So- earth-moon system. The Astronomical Journal, ciety of America Bulletin, 117(7-8):835–845. 108:1943–1961. 13 Vallis, G. K. and Farneti, R. (2009). Meridional en- A Solution to the energy bal- ergy transport in the coupled atmosphere–ocean ance model system: Scaling and numerical experiments. Quar- terly Journal of the Royal Meteorological Society, In this appendix, we provided the derivation of the 135(644):1643–1660. solution to the Budyko-Sellers energy-balance cli- mate model. The equation describing the steady- Walker, J. C., Hays, P., and Kasting, J. F. (1981). A state meridional temperature distribution T , written negative feedback mechanism for the long-term sta- in terms of a new variable x  sin(), takes the form bilization of earth’s surface temperature. Journal (North, 1975) of Geophysical Research: Oceans, 86(C10):9776– d dT Webb, D. (1982). Tides and the evolution of the D (1 x ) = Qa(x; x )S(x) I (T; pCO ) ; s 2 dx dx | {z } | {z } earth-moon system. Geophysical Journal of the | {z } Solar insolation re-emitted heat diffusive transport Royal Astronomical Society, 70(1):261–271. (10) Williams, D. M. and Kasting, J. F. (1997). Habitable planets with high obliquities. Icarus, 129(1):254– where the coefficient Williams, G. E. (2000). Geological constraints on D  K ; (11) eff the precambrian history of earth’s rotation and the gR moon’s orbit. Reviews of Geophysics, 38(1):37–59. and Q is the Solar constant divided by 4. In the 3 1 1 Wolf, E. and Toon, O. (2013). Hospitable archean above, c = 10 J kg K is the specific heat of air, climates simulated by a general circulation model.  = 1 bar is the surface air pressure, g = 10 m s Astrobiology, 13(7):656–673. is the acceleration due to gravity and R = 6400 km is the Earth’s radius. The Solar insolation function Wordsworth, R. and Pierrehumbert, R. (2013). S(x) is defined as Hydrogen-nitrogen greenhouse warming in earth’s early atmosphere. science, 339(6115):64–67. S(x) = 1 + S P (x) 2 2 Yang, J., Peltier, W., and Hu, Y. (2012). The initi- = 1 + S (3x 1) (12) ation of modern soft and hard snowball earth cli- mates in ccsm4. Climate of the Past, 8(3):907–918. where P (x) is a degree 2 Legendre Polynomial. The function a(x; x ) yields the fraction of incom- Young, G. M., Brunn, V. V., Gold, D. J., and Minter, ing Solar radiation that is absorbed (one minus the W. (1998). Earth’s oldest reported glaciation: albedo) as a function of latitude. The ice-albedo feed- physical and chemical evidence from the archean back is built into a(x; x ); defined as mozaan group ( 2.9 ga) of south africa. The Jour- nal of geology, 106(5):523–538. a ; if x  x 0 s a(x; x ) (13) Zahnle, K. and Walker, J. C. (1987). A constant a ; otherwise daylength during the precambrian era? Precam- brian Research, 37(2):95–105. where a > a , as ice is more reflective than ocean. 0 1 The outgoing radiation is prescribed through the Zeebe, R. E. and Wolf-Gladrow, D. A. (2001). CO2 function in seawater: equilibrium, kinetics, isotopes. Num- ber 65. Gulf Professional Publishing. I (T ) = A + BT (x) (14) 14 where A and B are constants to be determined by tuning to the modern climate. Such linearity allows us to replace T (x) with I (T )=B as the dependent variable (North, 1975). Carrying out the substitution and defining D  (15) we may formulate the problem into an ordinary differential equation satisfying d dI (x) 1 Q (1 x ) I (x) = S(x)a(x; x ): 0 0 dx dx D D (16) In order to solve the differential equation above, we separated the solution into two regimes; one at low latitudes where the albedo is smaller and a second at high latitudes where the albedo is that of ice. For this part of the calculation, we ignored the effect of spin rate upon albedo, and included only the effect of rotation rate upon diffusivity. In each region (i = 0 for low latitudes and i = 1 for high latitudes), the general solution may be written (North, 1975) I (x) = A P (x) +B Q (x) i i l i l S P (x) 2 2 + Qa 1 + ; i = 0; 1 (17) 6D + 1 where A and B are constants to be determined by i i satisfying boundary conditions, and the parameter l is defined as D = : (18) l(1 + l) The boundary conditions require zero heat flux at the equator and pole. They also require continuity of gradient at x = x and at this point I = I = I , s 0 1 s which was computed using a predefined critical tem- perature for permanent ice caps (T = T  10 C).

Journal

AstrophysicsarXiv (Cornell University)

Published: Mar 9, 2019

References