- Email: [email protected]
Heat-pipe planets
Earth and Planetary Science Letters 474 (2017) 13–19
Contents lists available at ScienceDirect
Earth and Planetary Science Letters www.elsevier.com/locate/epsl
Heat-pipe planets William B. Moore a,b,∗ , Justin I. Simon c , A. Alexander G. Webb d a
Department of Atmospheric and Planetary Sciences, Hampton University, Hampton, VA 23668, USA National Institute of Aerospace, Hampton, VA 23666, USA c Center for Isotope Cosmochemistry and Geochronology, Astromaterials Research and Exploration Science, NASA Johnson Space Center, 2101 NASA Parkway, Houston, TX 77058-3696, USA d Department of Earth Sciences and Laboratory for Space Research, University of Hong Kong, Hong Kong, China b
a r t i c l e
i n f o
Article history: Received 30 July 2015 Received in revised form 5 June 2017 Accepted 7 June 2017 Available online xxxx Editor: A. Yin Keywords: planetary thermal evolution volcanism terrestrial planet lithosphere terrestrial planet surface
a b s t r a c t Observations of the surfaces of all terrestrial bodies other than Earth reveal remarkable but unexplained similarities: endogenic resurfacing is dominated by plains-forming volcanism with few identifiable centers, magma compositions are highly magnesian (mafic to ultra-mafic), tectonic structures are dominantly contractional, and ancient topographic and gravity anomalies are preserved to the present. Here we show that cooling via volcanic heat pipes may explain these observations and provide a universal model of the way terrestrial bodies transition from a magma-ocean state into subsequent single-plate, stagnant-lid convection or plate tectonic phases. In the heat-pipe cooling mode, magma moves from a high melt-fraction asthenosphere through the lithosphere to erupt and cool at the surface via narrow channels. Despite high surface heat flow, the rapid volcanic resurfacing produces a thick, cold, and strong lithosphere which undergoes contractional strain forced by downward advection of the surface toward smaller radii. We hypothesize that heat-pipe cooling is the last significant endogenic resurfacing process experienced by most terrestrial bodies in the solar system, because subsequent stagnant-lid convection produces only weak tectonic deformation. Terrestrial exoplanets appreciably larger than Earth may remain in heat-pipe mode for much of the lifespan of a Sun-like star. © 2017 Elsevier B.V. All rights reserved.
1. Introduction There is currently great interest in exploring other terrestrial planets, for a broad range of reasons – e.g., to learn more about the nature of solar systems; to seek other possible environments in which life may be sustainable, or have arisen; to provide perspective and insight for our understanding of our own planet. Our exploration of terrestrial bodies in our own solar system has largely emphasized the different geologic histories experienced by the different bodies: Mercury’s early contraction, Venus’s desiccated, relatively young surface, Earth’s plate tectonics, the Moon’s bimodal crustal composition, Mars’ hemispheric dichotomy and isolated volcanic edifices, Io’s exceptionally rapid volcanic resurfacing. In the face of such diversity and the likely discovery of still different planetary surfaces in exoplanetary solar systems, it is necessary to explore what aspects of geological history may be held in common by all of these terrestrial bodies, in order to address the larger question of what features may be shared by terrestrial planets across the galaxy.
*
Corresponding author. E-mail address: [email protected] (W.B. Moore).
http://dx.doi.org/10.1016/j.epsl.2017.06.015 0012-821X/© 2017 Elsevier B.V. All rights reserved.
It is generally agreed that the terrestrial planets have initial evolutionary phases in common: accretion of planetesimals led to heating, differentiation into metallic cores and silicate mantles. The immense early heat input – potentially augmented by short-lived radionuclides or tidal heating – required that significant fractions of their mantles were molten (magma oceans), even for objects as small as Vesta. In contrast, the subsequent path from magma oceans to the generation of relatively rigid outer lithospheres and convecting mantles is largely thought to be distinct for each of our solar systems’ terrestrial planets: At Mercury, a rapidly developed lithosphere is generally thought to have experienced global contraction due to bulk cooling of the interior (Solomon, 1977; Byrne et al., 2014). Venus is widely thought to have experienced catastrophic global resurfacing, possibly multiple times, due to formation of a stagnant lid that inhibits heat flow. Earth developed plate tectonics, but the pathway is unclear: some interpretations posit plate tectonics from nearly the initial development of lithosphere, or at least as far back as 4.3–4.1 Ga (e.g. Harrison, 2009), whereas others suggest formation of a single-plate lithosphere akin to Venus’ that gradually or catastrophically transitioned to a plate tectonic lid system (e.g. Debaille et al., 2013). The Moon’s evolution remains puzzling. Many workers argue that its dominantly plagio-
14
W.B. Moore et al. / Earth and Planetary Science Letters 474 (2017) 13–19
clase crust must have floated up and crystallized at the top of the magma ocean (Warren, 1985), but generating nearly pure plagioclase compositions by this mechanism requires extremely efficient melt drainage (Piskorz and Stevenson, 2014). Mars’ early evolution may have involved whole mantle overturn (Elkins-Tanton et al., 2005) or a proto-plate tectonic regime, potentially generated by early impacts (Yin, 2012), that later seized up isolating longlived chemically distinct lithospheric reservoirs (Borg and Draper, 2003). These five terrestrial bodies may have experienced divergent evolutions, but they likely have all been continually cooling since initial accretion and differentiation. Jupiter’s moon Io, however, diverges from that pattern: it has been kept above its solidus temperature for the majority of its evolution because its dominant heat source is frictional heating caused by its gravitational (tidal) interactions with Jupiter and the other Galilean moons (O’Reilly and Davies, 1981). With a surface heat flux forty times that of Earth, Io is dominated by volcanic heat advection. The rapid cooling and subsequent burial of volcanic rocks at the surface leads to a downward advection of cold surface material, chilling the lithosphere. Perhaps counter-intuitively, the terrestrial body with the highest known heat flux exhibits a thick, cold single-plate mechanical lithosphere which is the consequence of rapid crustal/lithospheric formation from the top by addition of new flows. Recently, based on a survey of geological and geochemical observations, heat-pipe cooling has been suggested as a possible mechanism for the early cooling and lithospheric generation of Earth, for as much as the first third of Earth history (Moore and Webb, 2013). In this work we explore the possibility that the end of the magma ocean is not the point at which the solar system’s rocky bodies diverge in their evolution. Specifically, we review the records of the solar system’s stagnant-lid terrestrial bodies to explore whether they too may have experienced an early phase of heat-pipe cooling, as hypothesized for Earth and observed at present on Io. This overview highlights that ancient features common to these terrestrial bodies, e.g., the dominance of lowviscosity, mafic volcanic rocks, commonality of contractional features and scarcity of extensional features, and preservation of ancient topographic signatures, are consistent with the early heatpipe hypothesis. The tantalizing possibility motivating this exploration is that if heat-pipe cooling may have been experienced by all of our solar system’s terrestrial planets, it may be a phase experienced by terrestrial planets universally (Ricard et al., 2014). We begin with a detailed description of the heat-pipe cooling process before discussing the relevant observations of each stagnant-lid planet. 2. Heat pipe cooling The geological evolution of terrestrial (rocky) planets is driven by the transport of heat from their interiors. Although impacts may obscure endogenic resurfacing processes, everywhere that evidence for such processes is preserved, it is dominated by volcanic landforms (except on Earth, where liquid water, life, and plate tectonics dominate surface evolution). A terrestrial body cooling via heat pipes experiences persistent global volcanic resurfacing by which older layers are progressively buried and advected downward to form a thick, cold, and strong mechanical lithosphere (O’Reilly and Davies, 1981) (Fig. 1, Heat-Pipe inset). At this point we should note that the mechanical lithosphere of a heat pipe planet is composed almost entirely of the products of melting, that is, it is also the crust. Only when the mantle drops below the solidus can a portion of it adhere to the base of the crust forming what’s known as the mantle-lithosphere. While on present-day Earth, the crust is mechanically weaker than the mantle by a significant factor, the products of high-degree, heat-pipe
Fig. 1. Modeled evolution of lithospheric thickness (measured as a fraction of the mantle thickness) over time as internal heat production decreases by a factor of four (dashed line shows equilibrium lithospheric thickness increasing). Planets evolving through a heat-pipe phase (red) develop a thick lithosphere early in their history, which thins as volcanism wanes and then thickens as stagnant-lid convection takes over. Planets without melt transport transition directly from the magma ocean to stagnant-lid convection (blue) and begin with thin weak lithospheres that monotonically thicken and strengthen over time.
melting are not as evolved or silica rich and hence tend to be more similar in strength to the mantle. On Earth, this may be responsible for the longevity of cratonic mantle (e.g. the “tectosphere” of Jordan, 1978). The implications of heat pipes for the tectonic history of terrestrial bodies are illustrated in Fig. 1 by contrasting the modeled evolution of the lithospheric thickness for a planet with a heatpipe phase of evolution vs. one with a stagnant lid and no melting as heat generation decreases by a factor of four. The models were computed using STAGYY in a 4 × 1, two-dimensional domain with purely internal heating (decaying exponentially with time), strongly temperature-dependent viscosity (viscosity contrast 107 ), and a linearly pressure-dependent solidus with instantaneous melt extraction (see Kankanamge and Moore, 2016 for model details). Although greatly simplified, our treatment of the melting process captures the first order effects of melting and melt extraction on the heat transport which are that heat and material are advected to the surface on timescales that are rapid with respect to advective or diffusive processes. Unlike the classic stagnant-lid planet (blue) with a monotonically thickening lithosphere that tracks the equilibrium thickness (dashed), the heat-pipe planet (red) develops an early, thick lithosphere that is capable of recording and preserving early deformation events. Heat-pipe operation leads to: 1) Thick, cold, and strong lithospheres even though heat flow is high, 2) Dominance of compressive stresses as buried layers are forced to smaller radii, 3) Continuous replacement of lithospheric material, 4) High melt-fraction (mafic to ultra-mafic), low-viscosity eruptions and efficient degassing of the interior, and 5) A rapid transition to stagnant-lid or plate tectonic behavior. Heat pipe cooling ends when the internal heat production drops below the rate that can be accommodated by primarily sub-solidus convection (with or without plate tectonics). In the following sections, we briefly review the observations relevant to the formation of the surfaces of each of the terrestrial planets and current models that have been proposed to explain them. We then discuss any outstanding problems and show how the heat-pipe hypothesis can resolve these in a consistent way across all planets.
W.B. Moore et al. / Earth and Planetary Science Letters 474 (2017) 13–19
3. Mercury Mercury’s surface has now been extensively imaged by the MESSENGER spacecraft, and two important observations have emerged. First, Mercury was globally resurfaced until about 3.7 Ga by volcanic eruptions emplacing smooth plains with few identifiable eruption centers. The presence of lava channels exhibiting erosional features and low slopes (<1 deg) is also evidence for low-viscosity flows (Hurwitz et al., 2013). This volcanic resurfacing essentially ceased at 3.5 Ga (Byrne et al., 2016) with little activity other than several kilometers of global contraction since (Byrne et al., 2014). Surface modification prior to this time was able to degrade even very large (>500 km diameter) impact structures, reducing their density relative to the Moon (Fassett et al., 2012). The second important discovery of the Messenger mission is that Mercury’s surface composition is dominated by highly magnesian mafic to ultra-mafic basalts (Charlier et al., 2013) and that the geochemically distinct terranes identified by remote sensing are not strictly associated with mapped geologic units (Weider et al., 2015). Analysis of the melting conditions based on the observed surface compositions (Namur et al., 2016) reveals shallow source depths and large degrees of melting. Mercurian basalts show evidence for a gradual decrease in the extent of melting of their source rocks, from nearly 50% melting in the older Intra-Crater Plains (ICP) and Heavily Cratered Terrain (HCP) to ∼25% melting in the Northern Volcanic Plains province (Namur et al., 2016). Estimates of the crustal thickness reveal that the total amount of crustal production as a fraction of mantle volume (∼10%) is the highest of all terrestrial bodies for which we have observational constraints. Existing models of Mercury’s thermal and tectonic history are constrained by the observations of about 1 Ga of volcanic activity, the high melt fractions of the erupted melts, the total melt production (though any amount of crustal recycling is unknown), the total contraction recorded by the lithosphere, and the presence of a convecting liquid outer core. In fitting many of these constraints, recent models identify an early and possibly episodic period of magmatic activity ending after less than about 1 Ga (Tosi et al., 2013). The nature of these episodes and the implications for lithospheric evolution have not, however, been investigated in detail. In particular, Mercury’s response to the global compressive regime is represented by the network of lobate scarps, indicative of a strong elastic lithosphere that existed from early times, and not of a weak lithosphere in equilibrium with high conductive heat flow. The interplay of volcanic and impact processes may have led to a layered crust mixed to various degrees by large impacts and thus to geochemical terrains that cannot be easily correlated with currently visible geologic units (Weider et al., 2015). The geological observations summarized above and the thermochemical model results point to an episode of heat pipes operating for somewhat less than the first billion years of Mercury’s evolution. As heat-pipe volcanism continually replaces the lithosphere, the new layers take on the shape supported by the scaffolding of the old layers, accommodating any previous deformation while producing compressive stresses due to downward advection of the surface. Thus, the history of crustal growth and the history of lithospheric stress are coupled and the contraction recorded by the observable scarps is a mixture of thermal contraction and heatpipe-related shortening. The rapid transition of volcanism from a global resurfacing process up to about 3.7 Ga to a regional process to complete cessation at about 3.5 Ga (Byrne et al., 2016) is a result of the high efficiency of melt extraction as a heat transport process that essentially shuts itself off as soon as heat production is no longer able to sustain a global melt layer. The coincidence of this decrease with the rapid decline in impactor flux resulted in
15
a surface displaying a mixture of impact-dominated and volcanicdominated surfaces. 4. Venus The surface of Venus, like that of Mercury, is dominated by mafic lavas, with broad plains made up of numerous flows spanning hundreds of kilometers at low slope and with few identifiable source structures. Other volcanic features include fields of small shields on the volcanic plains, hundreds of roughly circular coronae, and a few dozen large shields associated with topographic highlands known as volcanic rises (e.g. Tanaka et al., 1997). Geoid and topography data require a generally thick lithosphere (200–400 km) which thins by up to 80% beneath volcanic highlands (Moore and Schubert, 1997; Orth and Solomatov, 2012). Plateau-shaped highlands, on the other hand, are supported by thickened crust (Moore and Schubert, 1997), show abundant evidence for compressional tectonics, and are among the earliest surface features. Observations of deformation patterns and craters indicate that Venus’s surface records a decrease in tectonic activity (Ivanov and Head, 2015). Early contractional structures include fold-thrust belts with mountains as high as 12 km, whereas deformation of the youngest ∼10% of the surface is largely limited to low-strain (<5%) rift belts extending between major volcanic centers (Grimm, 1994). The random distribution of Venusian craters has long been taken as evidence of catastrophic resurfacing around 0.5 Ga, although it is also consistent with a wider range of volcanic resurfacing models (e.g. Bjonnes et al., 2012). Recent analyses of crater morphology indicate that the majority (∼80%) of craters have experienced volcanic infilling, suggesting a young (∼150 Ma average age) surface that has experienced hundreds of meters of volcanic deposition (Herrick and Rumpf, 2011). Nonetheless, nearly all craters display pristine circular shapes (Herrick and Rumpf, 2011), so cessation of significant deformation must occur prior to 0.5 Ga. Embayment relationships indicate that the present high-standing topographic features are among the earliest components in Venus’s surface record (Gilmore et al., 1998), so the thick lithosphere constituting their support must also have been present early. Early speculation that Venus might host plate tectonics has been dismissed in light of findings that the materials and deformation of the surface are dominated by a single narrow age range and thus lack the time-progressive signature of plate-like crustal formation and recycling (Kaula and Phillips, 1981). Turcotte (1989) explored whether active heat-pipe cooling might explain lithospheric development at Venus, noting that this is consistent with the dominance of volcanics, thick lithosphere, and support of high topography, but he concluded that such activity would require a much higher volcanic flux than observed. Catastrophic volcanic resurfacing models involving global subduction and wholesale lithospheric replacement have been proposed (e.g. Parmentier and Hess, 1992). Because these models involve progressive lithospheric thickening subsequent to resurfacing events, they are difficult to reconcile with observations showing that the contractional highlands and their supporting thick lithosphere are among the earliest features seen on Venus. Similarly, models of contractional terrane development across thin lithosphere which progressively thickened (Hansen, 2006) are difficult to reconcile with the lack of deformation across the embayed margins of deformed highlands. Although Venus does not display sufficient volcanic flux to currently experience active heat-pipe cooling, the features which led (Turcotte, 1989) to explore this possibility may yet be evidence of a fossil heat-pipe lithosphere that subsequently transitioned into stagnant-lid cooling mode. The dominance of volcanics, thick lithosphere, and support of high topography could all be largely relicts of heat-pipe cooling, and the evident waning of both volcanism
16
W.B. Moore et al. / Earth and Planetary Science Letters 474 (2017) 13–19
and deformation could reflect the transition to the stagnant-lid regime. In this case, the single narrow age range of most of the Venusian surface would not reflect complete lithospheric replacement, but rather the last areally extensive new layers atop a lithosphere created via heat-pipe cooling. Finally, we acknowledge that the effective erasure of all available Venusian surface records prior to 0.5 Ga limit our ability to evaluate Venus’ tectonic evolution across most of its history. Early plate tectonic and/or episodic subduction periods could have preceded the proposed heat-pipe cooling period. Indeed, coupled mantle–atmosphere modeling suggests that low surface temperatures in the first billion years of Venus’ history would have encouraged mobile lid behavior (Gillmann and Tackley, 2014). Nonetheless, the surface record permits that a subsequent single-plate lithosphere could have been dominated by heat-pipe cooling and later transitioned into a stagnant-lid phase with limited ongoing volcanism. 5. Mars Among the most important surface features on Mars are its large volcanos (e.g., Olympus Mons), ancient cratered terrains, and its north–south crustal dichotomy. Volcanic flows have extensively modified the northern lowlands, burying a surface similar in age to the southern highlands (Frey, 2006), but there remains a significant and abrupt change in crustal thickness (Neumann et al., 2004) between the cratered uplands of the south and the smoother lowlands of the north. This dichotomy formed less than 50 Ma after accretion (Frey, 2006) and is associated with only minor contractional structures (Nimmo and Tanaka, 2005). Another key observation for the early evolution of Mars is the extreme range of trace element abundances and isotopic compositions among the main group of Martian meteorites, basaltic rocks called Shergottites. Evidence for early global differentiation of Mars comes from the ancient Rb–Sr and Pb–Pb isotopic model ages of their mantle source materials (Chen and Wasserburg, 1986). These ancient ages are further supported by age constraints obtained by the short-lived 146 Sm–142 Nd decay system that suggest global differentiation of Mars occurred a few 10’s to 100 Ma after solar system formation (Borg et al., 2003). Martian meteorites also show large variation in 142 Nd/144 Nd and 143 Nd/144 Nd (the product of long lived decay of 147 Sm). These extreme variations are ∼4 times greater than those of comparable mantle sources on Earth and the Moon and likely exist because portions of the Martian mantle were isolated into geochemically distinct reservoirs more than ∼4 Ga ago. Volatile studies (Usui et al., 2012) indicate that Shergottites support an early depleted dry (15–50 ppm water) interior. It remains unclear whether magmatic/tectonic or impact processes were responsible for the formation of the crustal dichotomy between the elevated southern hemisphere and the depressed northern hemisphere. Endogenic models for the formation of the dichotomy invoke either horizontal (plate-tectonic) or vertical (instability or plume) motions. Horizontal models lack evidence of the necessary plate boundary structures (the equivalent of spreading centers or convergent margins on Earth) that should coincide with the dichotomy boundary, while the internal instability models involving mantle convection with a single plume (Zhong and Zuber, 2001) are too slow to form the dichotomy given the ancient age of the northern basement. Models that explain the early formation of the dichotomy by magma ocean (Elkins-Tanton et al., 2005) or impact processes (Reese et al., 2011) fail to account for the preservation of the dichotomy through the early period when Martian heat flow was high. Preservation of the dichotomy boundary requires either a very dry or very cold crust able to resist deformation at high heat flow. Differences among the reported meteorite ages have been used to suggest that Mars had a magma ocean with a protracted
(∼100 Ma) crystallization history (Debaille et al., 2009). Such a protracted period of time is difficult to reconcile with standard magma ocean cooling models (Elkins-Tanton et al., 2005). Additionally, although the current surface of Mars is dry the conditions at the surface of early Mars and for much of its history were likely quite wet. The budget of volatile evolution in terrestrial planets is debated. One endmember view is that volatiles were introduced during accretion and lost from the interior by volcanic degassing (Drake and Righter, 2002). An alternative view is that planets accreted “dry” and are later hydrated by addition of volatile-rich materials from the outer solar system (Albarède, 2009). If Mars accreted wet it must have lost volatiles very early as they are found in very low abundances in the source region(s) of mantle-derived meteorites (Usui et al., 2012). Heat-pipe operation is capable of producing a thick, cold, and strong lithosphere very early in Mars’ history (Fig. 1) through rapid downward lithosphere advection, thereby preserving the crustal dichotomy and its sharp boundary. Heat pipe driven convergence also explains the observed dominance of contractional features. A heat-pipe planet that transitions to stagnant-lid convection (Fig. 1) would exhibit early and significant extraction of incompatible material to form a primordial crust with little to no subsequent remixing. It is reasonable to suggest that the level of geochemical depletion indicated by Martian meteorites at such an early stage is not seen among mantle melts on Earth because it no longer exists due to later subduction and return of significant amounts of enriched crustal materials to Earth’s interior. Much less crustal material has been recycled into the interior of Mars (Usui et al., 2012). A another prediction of the heat-pipe hypothesis would be that the isotopic systematics used to suggest an ancient period of mantle overturn and protracted magma ocean solidification (Debaille et al., 2009) actually reflects the onset and punctuated magmatic activity related to heat-pipe cooling. Finally, samplebased volatile studies (Usui et al., 2012) supporting an early dry mantle are consistent with the expectation of efficient degassing by heat-pipe volcanism, which would have aided in the preservation of an early crustal dichotomy. 6. Moon The Moon’s primordial history has been overprinted by early global chemical segregation, meteor bombardment, and volcanism. The most visible evidence of lunar chemical segregation is the existence of two very different terrains on the Moon: the highlands and the maria. The lunar highlands represent an ancient terrain dominated by plagioclase rich anorthositic rocks whose origins likely hold the key to understanding the early evolution of the Moon and potentially terrestrial planet formation in general. The prevalence of anorthosite in the lunar highlands and the existence of a rock type called ferroan anorthosite or “FAN” returned by the Apollo missions are generally attributed to crystallization and flotation of less dense plagioclase at the top of an ancient lunar magma ocean (LMO model) (Warren, 1985). The processes and timescales of how differentiation of the highlands occurred remain unresolved (Borg et al., 2011). This uncertainty is at least partly due to the fact that meteor impact gardening has reworked much of the ancient lunar surface, possibly to a depth of 3 km (Yamamoto et al., 2012). Nevertheless, remote sensing observations focused on outcrops within large impact craters (e.g., Orientale Basin) demonstrate that the upper lunar lithosphere contains thick, widespread, extremely plagioclase-rich anorthosite to depths of tens of km (Ohtake et al., 2009), supporting the importance of widespread differentiation on the Moon. Novel volatile element measurements (Hui et al., 2013) and some new age determinations of FANs (Borg et al., 2011) along with the recently reported geochemical characteristics of a diver-
W.B. Moore et al. / Earth and Planetary Science Letters 474 (2017) 13–19
17
Fig. 2. Stages of early Lunar crustal differentiation: I. Magma ocean, II. Lithospheric-scale magma crystal mush with stagnant crustal lid, III. Transition to Heat pipe, where melts cycle through crystal mush leading to buoyancy-driven crystal-melt separation, IV. Heat pipe, rapid lithospheric thickening in which relatively less mafic, Mg-suite melts intrude the bottom of the plagioclase-bearing gabbroic crust providing a mechanism that admixes a substantial portion of the lunar lithosphere. Heating at the bottom of the thickening lithosphere produces buoyant anorthositic melts that rise towards the surface. It is at this time that the observed highly refined plagio-crust (FAN) is finally produced.
sity of anorthosite clasts contained in lunar meteorite breccias (i.e., Takeda et al., 2006) have introduced new complications to the standard LMO hypothesis. These materials now appear to imply: (1) a wetter Moon than conventionally believed (McCubbin et al., 2010; Hui et al., 2013), (2) younger (4.3 Ga) FAN ages that no longer support the crystallization sequence originally used to develop the LMO model (Borg et al., 2015), and (3) igneous breccia clasts that exhibit more heterogeneity than previously seen in the Apollo collection expanding the compositional range to bridge the apparent gap between the FAN and the more alkali-enriched “Mg suite” rocks on classic anorthite content (Ca/[Ca + Na + K]) versus Mg-number (100Mg/[Mg + Fe]) plots (Takeda et al., 2006). Along with evidence for greater heterogeneity in their trace element abundances (e.g., rare-earth elements) these geochemical data imply that these sample did not all crystallize from a common magma source (Russell et al., 2014). Moreover, it is unclear whether a simple flotation crust is capable of producing the extremely high plagioclase contents (>98%) observed in both Apollo samples (Haskin et al., 1981) and remote sensing data (Ohtake et al., 2009), since a mostly solid lithosphere forms before plagioclase feldspar reaches saturation at 70% solidification (Snyder and Taylor, 1992; Lin et al., 2017). It is possible to model with reasonable physical parameters that a significant amount of felsic melt can escape from a crystal-mush network to form an early-formed plagioclase-rich floatation lid (Piskorz and Stevenson, 2014). However, the simplest formation scenarios for a floating cumulate are complicated by the near-uniformity of the alkali composition of the plagioclase, even as the mafic phases record significant variations in Mg content (Warren, 1985). In addition to these recent insights into its unusual crustal chemistry, the Moon stands out as having a shape that is dramatically out of hydrostatic equilibrium (even at the longest wavelengths spherical harmonic degree 2). The Moon’s shape is not a fossil of a synchronous rotator at any distance from the Earth (the ratio of the gravity coefficients J2 :C22 is in excess by a factor of 3: Goudas, 1964), but instead must record some other orbital/rotational state or combination of states. Collectively, these observations either support other models of crust formation on the Moon besides crystal flotation or suggest that there are complexities in the LMO scenario that allow for multiple generations of anorthosite formation (Norman et al., 1995; Gross et al., 2014) and hydrostatic disequilibrium. The only means by which the Moon’s disequilibrium shape can be preserved over geologic time is substantial lithospheric strength, but all present explanations for the observed shape rely on processes that occur very early in the Moon’s evolution when it is much hotter. What is required is a way to rapidly produce a strong lithosphere even when the Moon is young and hot, which is precisely the expected behavior of a body experiencing heat-pipe cooling (Fig. 1). The lithosphere is created rapidly and continuously, causing the shape to be recorded around the time the heat-
pipe mechanism shuts off, and leaving behind a strong, distorted lithosphere. Likewise, the aforementioned geochemical complications can be resolved for the Moon if the plagioclase-rich crust is produced and refined through a widespread episode of heatpipe magmatism (Fig. 2) rather than a process solely produced by density-driven plagioclase flotation. In the heat-pipe scenario, the base of the thickened lithosphere is remelted by incompatible element rich (i.e., alkalis and volatiles) mafic magmas from the deep interior (possibly equivalent to the Mg suite rocks), producing melts that are higher in plagioclase while at the same time introducing water and a range of mafic components. This refinement process occurs through volatile-assisted remelting, where the lowered solidus produces buoyant, Al-rich melts (Fig. 2). Repeated operation of this refining mechanism can result in the nearly pure plagioclase melts and textures observed in Apollo samples (Haskin et al., 1981), while at the same time explaining the variations in Mg contents, the diversity of materials contained in lunar meteorites, and the relatively high water abundance measured by Hui et al. (2013) in FAN rocks. If volatiles in FANs are introduced by the heat pipe process (analogous to the formation of Tonalite–Trondhjemite–Granodiroties on Earth) and not due to the canonical LMO fractional crystallization differentiation model, this evidence for a “wet” Moon is no longer valid. A wide-spread phase of heat-pipe volcanism that exchanges deep mantle-derived material with crustal rocks is also consistent with the latest geochronological observations that indicate that there is significant overlap between the ages determined for Mg suite rocks, FANs, and the ∼4.3 Ga model calculated for the source of younger maria basalts (Borg et al., 2015). A global period of serial volcanism (Longhi, 2003) eliminates the outstanding age problem with the standard LMO model (Shearer et al., 2006), because the re-melted FANs would be related and thus coeval with intrusion of the mafic Mg suite rocks as opposed to being older as required by the standard magma ocean model (Shearer et al., 2006). 7. Discussion and conclusions Heat pipes are an important feature of terrestrial planets at high heat flow, as illustrated by Io’s present activity. Evidence for their operation early in Earth’s history (Moore and Webb, 2013) suggests that all terrestrial bodies should experience an early episode of heat-pipe cooling. We have shown that the geological, geochemical, and geochronological evidence from the terrestrial bodies in our solar system is consistent with heat-pipe operation providing the main mechanism of crustal formation and endogenic resurfacing. Io is in the heat-pipe mode to the present day due to intense tidal heating, while the other planets (except Earth) have all transitioned to stagnant-lid convection with subsequent levels of geologic activity proportional to their heat content (Fig. 3). Venus is the most active, while Mercury and the Moon are long
18
W.B. Moore et al. / Earth and Planetary Science Letters 474 (2017) 13–19
tience and numerous reviewers for their assistance in improving this paper. References
Fig. 3. Illustration of terrestrial planet heat flow versus internal temperature. The sense of evolution as heat sources and internal heat content decline is shown by arrows. The terrestrial bodies are labeled, LTP stands for Large Terrestrial exo-Planet (several Earth-masses), and the initial magma ocean stage is indicated. The current temperature coordinates are notional (Venus may be warmer than Earth, for example). Heat loss is high and thermal evolution is rapid in the upper right of the diagram, where the magma ocean gives way to heat-pipes (dotted), and heat flow decreases as heat-pipes transition to either plate tectonics (dashed) or stagnant lid sub-solidus convection.
dead. Note that the main planetary (as opposed to material) properties contributing to the efficiency of heat transport by subsolidus convection are gravity, surface to interior temperature difference, and, to a lesser degree, mantle depth, therefore smaller planets have less efficient convection as do planets with higher surface temperatures (e.g. Venus). Only on Earth have buoyancy forces in the mantle overcome the strength of the lithosphere resulting in a transition to plate tectonics as indicated by the dashed line. The heat-pipe hypothesis therefore provides a uniform explanation for several common features of the known terrestrial planets that have not undergone plate tectonics and should be considered an important aspect of their evolution. Finally, we note that the conditions for the operation of plate tectonics have been the subject of considerable debate, particularly when applied to the large terrestrial bodies (LTP in Fig. 3) that have recently been found orbiting other stars (e.g. Stamenkovic´ et al., 2011). It has been argued that such bodies are both more and less likely to have plate tectonics. Earth appears to have transitioned to plate tectonics about 1 Ga after formation (Shirey and Richardson, 2011; Moore and Webb, 2013). Since the equilibrium heat flux of a planet scales as mass/area (for most plausible heat sources), terrestrial planets more massive than the Earth should experience longer heat-pipe episodes prior to the initiation of plate tectonics. A planet twice as massive as Earth should take more than twice as long to cool because the area increases less rapidly than a simple mass scaling would suggest due to the compressibility of terrestrial materials (Stamenkovic´ et al., 2011). For the large “super-Earths” over five Earth masses, the lifetime of the heat-pipe phase may exceed the lifetime of Sun-like parent stars and thus any subsequent plate-tectonic phase may never be observed. Such planets might better be called “super-Ios,” driving us to reconsider what types of surfaces and atmospheres to expect as we expand our exploration of other solar systems. Acknowledgements W.B.M. acknowledges funding from NSF EAR-1246983 and NExSS NNX15AE05G. J.I.S. acknowledges NASA funding from the Lunar Advanced Science & Exploration Research Program 10-LASER10-0077, the Mars Fundamental Research Program 10-MFRP10-0022, and the Solar System Workings Program 15-SSW15-2-0411. We would like to thank the editor for his pa-
Albarède, F., 2009. Volatile accretion history of the terrestrial planets and dynamic implications. Nature 461, 1227–1233. Bjonnes, E.E., Hansen, V.L., James, B., Swenson, J.B., 2012. Equilibrium resurfacing of Venus: results from new Monte Carlo modeling and implications for Venus surface histories. Icarus 217, 451–461. Borg, L.E., Connelly, J.N., Boyet, M., Carlson, R.W., 2011. Chronological evidence that the Moon is either young or did not have a global magma ocean. Nature 477, 70–72. Borg, L.E., Draper, D.S., 2003. A petrogenic model for the origin and compositional variation of the martian basaltic meteorites. Meteorit. Planet. Sci. 38, 1713–1731. Borg, L.E., Gaffney, A.M., Shearer, C.K., 2015. A review of lunar chronology revealing a preponderance of 4.34–4.37 Ga ages. Meteorit. Planet. Sci. 50, 715–732. Borg, L.E., Nyquist, L.E., Wiesmann, H., Shih, C.-Y., Reese, Y., 2003. The age of Dar al Gani 476 and the differentiation history of the Martian meteorites inferred from their radiogenic isotopic systematics. Geochim. Cosmochim. Acta 67, 3519–3536. Byrne, P.K., Klimczak, C., Sengör, ¸ A.M.C., Solomon, S.C., Watters, T.R., Hauck, S.A.I., 2014. Mercury’s global contraction much greater than earlier estimates. Nat. Geosci. 7, 301–307. Byrne, P.K., Ostrach, L.R., Fassett, C.I., Chapman, C.R., Denevi, B.W., Evans, A.J., Klimczak, C., Banks, M.E., Head, J.W., Solomon, S.C., 2016. Widespread effusive volcanism on Mercury likely ended by about 3.5 Ga. Geophys. Res. Lett. 43, 7408–7416. Charlier, B., Grove, T.L., Zuber, M.T., 2013. Phase equilibria of ultramafic compositions on Mercury and the origin of the compositional dichotomy. Earth Planet. Sci. Lett. 363, 50–60. Chen, J.H., Wasserburg, G.J., 1986. Formation ages and evolution of Shergotty and its parent planet from U–Th–Pb systematics. Geochim. Cosmochim. Acta 50, 955–968. Debaille, V., Brandon, A.D., O’Neill, C., Yin, Q.-Z., Jacobsen, B., 2009. Early Martian mantle overturn inferred from isotopic composition of nakhlite meteorites. Nat. Geosci. 2, 548–552. Debaille, V., O’Neill, C., Brandon, A.D., Haenecour, P., Yin, Q.-Z., Mattielli, N., Treiman, A.H., 2013. Stagnant-lid tectonics in early Earth revealed by 142 Nd variations in late Archean rocks. Earth Planet. Sci. Lett. 373, 83–92. Drake, M.J., Righter, K., 2002. Determining the composition of the Earth. Nature 416, 39–44. Elkins-Tanton, L.T., Hess, P.C., Parmentier, E.M., 2005. Possible formation of ancient crust on Mars through magma ocean processes. J. Geophys. Res., Planets 110 (E9), 12. Fassett, C.I., Head, J.W., Baker, D.M.H., Zuber, M.T., Smith, D.E., Neumann, G.A., Solomon, S.C., Klimczak, C., Strom, R.G., Chapman, C.R., Prockter, L.M., Phillips, R.J., Oberst, J., Preusker, F., 2012. Large impact basins on Mercury: global distribution, characteristics, and modification history from MESSENGER orbital data. J. Geophys. Res., Planets 117, E00L08. Frey, H.V., 2006. Impact constraints on, and a chronology for, major events in early Mars history. J. Geophys. Res. 111, 8. Gillmann, C., Tackley, P., 2014. Atmosphere/mantle coupling and feedbacks on Venus. J. Geophys. Res., Planets 119, 1189–1217. Gilmore, M.S., Collins, G.C., Ivanov, M.A., Marinangeli, L., Head, J.W., 1998. Style and sequence of extensional structures in tessera terrain, Venus. J. Geophys. Res. 103, 16813–16840. Goudas, C.L., 1964. Moments of inertia and gravity field of the Moon. Icarus 3, 375–409. Grimm, R.E., 1994. The deep structure of Venusian plateau highlands. Icarus 112, 89–103. Gross, J., Treiman, A.H., Mercer, C.N., 2014. Lunar feldspathic meteorites: constraints on the geology of the lunar highlands, and the origin of the lunar crust. Earth Planet. Sci. Lett. 388, 318–328. Hansen, V.L., 2006. Geologic constraints on crustal plateau surface histories, Venus: the lava pond and bolide impact hypotheses. J. Geophys. Res., Planets 111 (E10), E11010. Harrison, T.M., 2009. The Hadean Crust: evidence from >4 Ga zircons. Annu. Rev. Earth Planet. Sci. 37, 479–505. Haskin, L.A., Lindstrom, M.M., Salpas, P.A., 1981. Some observations on compositional characteristics of lunar anorthosites. In: Lunar and Planetary Science Conference, vol. 12, pp. 406–408. Technical Report. Herrick, R.R., Rumpf, M.E., 2011. Postimpact modification by volcanic or tectonic processes as the rule, not the exception, for Venusian craters. J. Geophys. Res., Planets 116, 2004. Hui, H., Peslier, A.H., Zhang, Y., Neal, C.R., 2013. Water in lunar anorthosites and evidence for a wet early Moon. Nat. Geosci. 6, 177–180. Hurwitz, D.M., Head, J.W., Byrne, P.K., Xiao, Z., Solomon, S.C., Zuber, M.T., Smith, D.E., Neumann, G.A., 2013. Investigating the origin of candidate lava channels
W.B. Moore et al. / Earth and Planetary Science Letters 474 (2017) 13–19
on Mercury with MESSENGER data: theory and observations. J. Geophys. Res., Planets 118, 471–486. Ivanov, M.A., Head, J.W., 2015. Volcanically embayed craters on Venus: testing the catastrophic and equilibrium resurfacing models. Planet. Space Sci. 106, 116–121. Jordan, T.H., 1978. Composition and development of the continental tectosphere. Nature 274, 544–548. Kankanamge, D.G.J., Moore, W.B., 2016. Heat transport in the hadean mantle: from heat pipes to plates. Geophys. Res. Lett. 43, 3208–3214. Kaula, W.M., Phillips, R.J., 1981. Quantitative tests for plate tectonics on Venus. Geophys. Res. Lett. 8, 1187–1190. Lin, Y., Tronche, E.J., Steenstra, E.S., van Westrenen, W., 2017. Experimental constraints on the solidification of a nominally dry lunar magma ocean. Earth Planet. Sci. Lett. 471, 104–116. URL http://www.sciencedirect.com/science/ article/pii/S0012821X1730239X. Longhi, J., 2003. A new view of lunar ferroan anorthosites: postmagma ocean petrogenesis. J. Geophys. Res., Planets 108, 5083. McCubbin, F.M., Steele, A., Hauri, E.H., Nekvasil, H., Yamashita, S., Hemley, R.J., 2010. Nominally hydrous magmatism on the Moon. Proc. Natl. Acad. Sci. 107, 11223–11228. Moore, W.B., Schubert, G., 1997. Venusian crustal and lithospheric properties from nonlinear regressions of highland geoid and topography. Icarus 128, 415–428. Moore, W.B., Webb, A.A.G., 2013. Heat-pipe earth. Nature 501, 501–505. Namur, O., Collinet, M., Charlier, B., Grove, T.L., Holtz, F., McCammon, C., 2016. Melting processes and mantle sources of lavas on Mercury. Earth Planet. Sci. Lett. 439, 117–128. Neumann, G.A., Zuber, M.T., Wieczorek, M.A., McGovern, P.J., Lemoine, F.G., Smith, D.E., 2004. Crustal structure of Mars from gravity and topography. J. Geophys. Res., Planets 109, 8002. Nimmo, F., Tanaka, K., 2005. Early Crustal Evolution of Mars. Annu. Rev. Earth Planet. Sci. 33, 133–161. Norman, M.D., Keil, K., Griffin, W.L., Ryan, C.G., 1995. Fragments of ancient lunar crust: petrology and geochemistry of ferroan noritic anorthosites from the Descartes region of the Moon. Geochim. Cosmochim. Acta 59, 831–847. Ohtake, M., Matsunaga, T., Haruyama, J., Yokota, Y., Morota, T., Honda, C., Ogawa, Y., Torii, M., Miyamoto, H., Arai, T., Hirata, N., Iwasaki, A., Nakamura, R., Hiroi, T., Sugihara, T., Takeda, H., Otake, H., Pieters, C.M., Saiki, K., Kitazato, K., Abe, M., Asada, N., Demura, H., Yamaguchi, Y., Sasaki, S., Kodama, S., Terazono, J., Shirao, M., Yamaji, A., Minami, S., Akiyama, H., Josset, J.-L., 2009. The global distribution of pure anorthosite on the Moon. Nature 461, 236–240. O’Reilly, T.C., Davies, G.F., 1981. Magma transport of heat on Io: a mechanism allowing a thick lithosphere. Geophys. Res. Lett. 8, 313–316. Orth, C.P., Solomatov, V.S., 2012. Constraints on the Venusian crustal thickness variations in the isostatic stagnant lid approximation. Geochem. Geophys. Geosyst. 13, Q11012. Parmentier, E.M., Hess, P.C., 1992. Chemical differentiation of a convecting planetary interior: consequences for a one plate planet such as Venus. Geophys. Res. Lett. 19, 2015–2018. Piskorz, D., Stevenson, D.J., 2014. The formation of pure anorthosite on the Moon. Icarus 239, 238–243. Reese, C.C., Orth, C.P., Solomatov, V.S., 2011. Impact megadomes and the origin of the Martian crustal dichotomy. Icarus 213, 433–442.
19
Ricard, Y., Labrosse, S., Dubuffet, F., 2014. Lifting the cover of the cauldron: convection in hot planets. Geochem. Geophys. Geosyst. 15 (11). Russell, S.S., Joy, K.H., Jeffries, T.E., Consolmagno, G.J., Kearsley, A., 2014. Heterogeneity in lunar anorthosite meteorites: implications for the lunar magma ocean model. Philos. Trans. R. Soc. Lond. Ser. A 372, 20130241. Shearer, C.K., Hess, P.C., Wieczorek, M.A., Pritchard, M.E., Parmentier, E.M., Borg, L.E., Longhi, J., Elkins-Tanton, L.T., Neal, C.R., Antonenko, I., Canup, R.M., Halliday, A.N., Grove, T.L., Hager, B.H., Lee, D.-C., Wiechert, U., 2006. Thermal and magmatic evolution of the Moon. In: Joliff, B.L., Wieczorek, M.A., Shearer, C.K., Neal, C.R. (Eds.), New Views of the Moon. Min. Soc. Am., Chantilly, VA, pp. 365–518. Shirey, S.B., Richardson, S.H., 2011. Start of the Wilson cycle at 3 Ga shown by diamonds from subcontinental mantle. Science 333, 434–436. Snyder, G.A., Taylor, L.A., 1992. Imperfect fractional crystallization of the lunar magma ocean and formation of the lunar mantle: a realistic chemical approach. In: Agee, C.B., Longhi, J. (Eds.), Physics and Chemistry of Magma Oceans from 1 Bar to 4 Mbar. Solomon, S.C., 1977. The relationship between crustal tectonics and internal evolution in the moon and Mercury. Phys. Earth Planet. Inter. 15, 135–145. ´ V., Breuer, D., Spohn, T., 2011. Thermal and transport properties of Stamenkovic, mantle rock at high pressure: applications to super-Earths. Icarus 216, 572–596. Takeda, H., Yamaguchi, A., Bogard, D.D., Karouji, Y., Ebihara, M., Ohtake, M., Saiki, K., Arai, T., 2006. Magnesian anorthosites and a deep crustal rock from the farside crust of the moon. Earth Planet. Sci. Lett. 247, 171–184. Tanaka, K.L., Senske, D.A., Price, M., Kirk, R.L., 1997. Physiography, geomorphic/geologic mapping and stratigraphy of Venus. In: Bougher, S.W., Hunten, D.M., Phillips, R.J. (Eds.), Venus II: Geology, Geophysics, Atmosphere, and Solar Wind Environment, p. 667. Tosi, N., Grott, M., Plesa, A.-C., Breuer, D., 2013. Thermochemical evolution of Mercury’s interior. J. Geophys. Res., Planets 118, 2474–2487. Turcotte, D.L., 1989. A heat pipe mechanism for volcanism and tectonics on Venus. J. Geophys. Res. 94, 2779–2785. Usui, T., Alexander, C.M.O., Wang, J., Simon, J.I., Jones, J.H., 2012. Origin of water and mantle-crust interactions on Mars inferred from hydrogen isotopes and volatile element abundances of olivine-hosted melt inclusions of primitive shergottites. Earth Planet. Sci. Lett. 357, 119–129. Warren, P.H., 1985. The magma ocean concept and lunar evolution. Annu. Rev. Earth Planet. Sci. 13, 201–240. Weider, S.Z., Nittler, L.R., Starr, R.D., Crapster-Pregont, E.J., Peplowski, P.N., Denevi, B.W., Head, J.W., Byrne, P.K., Hauck, S.A., Ebel, D.S., Solomon, S.C., 2015. Evidence for geochemical terranes on Mercury: global mapping of major elements with MESSENGER’s X-ray spectrometer. Earth Planet. Sci. Lett. 416, 109–120. Yamamoto, S., Nakamura, R., Matsunaga, T., Ogawa, Y., Ishihara, Y., Morota, T., Hirata, N., Ohtake, M., Hiroi, T., Yokota, Y., Haruyama, J., 2012. Olivine-rich exposures in the South Pole-Aitken Basin. Icarus 218, 331–344. Yin, A., 2012. An episodic slab-rollback model for the origin of the Tharsis rise on Mars: implications for initiation of local plate subduction and final unification of a kinematically linked global plate-tectonic network on Earth. Lithosphere 4, 553–593. Zhong, S., Zuber, M.T., 2001. Degree-1 mantle convection and the crustal dichotomy on Mars. Earth Planet. Sci. Lett. 189, 75–84.
Contents lists available at ScienceDirect
Earth and Planetary Science Letters www.elsevier.com/locate/epsl
Heat-pipe planets William B. Moore a,b,∗ , Justin I. Simon c , A. Alexander G. Webb d a
Department of Atmospheric and Planetary Sciences, Hampton University, Hampton, VA 23668, USA National Institute of Aerospace, Hampton, VA 23666, USA c Center for Isotope Cosmochemistry and Geochronology, Astromaterials Research and Exploration Science, NASA Johnson Space Center, 2101 NASA Parkway, Houston, TX 77058-3696, USA d Department of Earth Sciences and Laboratory for Space Research, University of Hong Kong, Hong Kong, China b
a r t i c l e
i n f o
Article history: Received 30 July 2015 Received in revised form 5 June 2017 Accepted 7 June 2017 Available online xxxx Editor: A. Yin Keywords: planetary thermal evolution volcanism terrestrial planet lithosphere terrestrial planet surface
a b s t r a c t Observations of the surfaces of all terrestrial bodies other than Earth reveal remarkable but unexplained similarities: endogenic resurfacing is dominated by plains-forming volcanism with few identifiable centers, magma compositions are highly magnesian (mafic to ultra-mafic), tectonic structures are dominantly contractional, and ancient topographic and gravity anomalies are preserved to the present. Here we show that cooling via volcanic heat pipes may explain these observations and provide a universal model of the way terrestrial bodies transition from a magma-ocean state into subsequent single-plate, stagnant-lid convection or plate tectonic phases. In the heat-pipe cooling mode, magma moves from a high melt-fraction asthenosphere through the lithosphere to erupt and cool at the surface via narrow channels. Despite high surface heat flow, the rapid volcanic resurfacing produces a thick, cold, and strong lithosphere which undergoes contractional strain forced by downward advection of the surface toward smaller radii. We hypothesize that heat-pipe cooling is the last significant endogenic resurfacing process experienced by most terrestrial bodies in the solar system, because subsequent stagnant-lid convection produces only weak tectonic deformation. Terrestrial exoplanets appreciably larger than Earth may remain in heat-pipe mode for much of the lifespan of a Sun-like star. © 2017 Elsevier B.V. All rights reserved.
1. Introduction There is currently great interest in exploring other terrestrial planets, for a broad range of reasons – e.g., to learn more about the nature of solar systems; to seek other possible environments in which life may be sustainable, or have arisen; to provide perspective and insight for our understanding of our own planet. Our exploration of terrestrial bodies in our own solar system has largely emphasized the different geologic histories experienced by the different bodies: Mercury’s early contraction, Venus’s desiccated, relatively young surface, Earth’s plate tectonics, the Moon’s bimodal crustal composition, Mars’ hemispheric dichotomy and isolated volcanic edifices, Io’s exceptionally rapid volcanic resurfacing. In the face of such diversity and the likely discovery of still different planetary surfaces in exoplanetary solar systems, it is necessary to explore what aspects of geological history may be held in common by all of these terrestrial bodies, in order to address the larger question of what features may be shared by terrestrial planets across the galaxy.
*
Corresponding author. E-mail address: [email protected] (W.B. Moore).
http://dx.doi.org/10.1016/j.epsl.2017.06.015 0012-821X/© 2017 Elsevier B.V. All rights reserved.
It is generally agreed that the terrestrial planets have initial evolutionary phases in common: accretion of planetesimals led to heating, differentiation into metallic cores and silicate mantles. The immense early heat input – potentially augmented by short-lived radionuclides or tidal heating – required that significant fractions of their mantles were molten (magma oceans), even for objects as small as Vesta. In contrast, the subsequent path from magma oceans to the generation of relatively rigid outer lithospheres and convecting mantles is largely thought to be distinct for each of our solar systems’ terrestrial planets: At Mercury, a rapidly developed lithosphere is generally thought to have experienced global contraction due to bulk cooling of the interior (Solomon, 1977; Byrne et al., 2014). Venus is widely thought to have experienced catastrophic global resurfacing, possibly multiple times, due to formation of a stagnant lid that inhibits heat flow. Earth developed plate tectonics, but the pathway is unclear: some interpretations posit plate tectonics from nearly the initial development of lithosphere, or at least as far back as 4.3–4.1 Ga (e.g. Harrison, 2009), whereas others suggest formation of a single-plate lithosphere akin to Venus’ that gradually or catastrophically transitioned to a plate tectonic lid system (e.g. Debaille et al., 2013). The Moon’s evolution remains puzzling. Many workers argue that its dominantly plagio-
14
W.B. Moore et al. / Earth and Planetary Science Letters 474 (2017) 13–19
clase crust must have floated up and crystallized at the top of the magma ocean (Warren, 1985), but generating nearly pure plagioclase compositions by this mechanism requires extremely efficient melt drainage (Piskorz and Stevenson, 2014). Mars’ early evolution may have involved whole mantle overturn (Elkins-Tanton et al., 2005) or a proto-plate tectonic regime, potentially generated by early impacts (Yin, 2012), that later seized up isolating longlived chemically distinct lithospheric reservoirs (Borg and Draper, 2003). These five terrestrial bodies may have experienced divergent evolutions, but they likely have all been continually cooling since initial accretion and differentiation. Jupiter’s moon Io, however, diverges from that pattern: it has been kept above its solidus temperature for the majority of its evolution because its dominant heat source is frictional heating caused by its gravitational (tidal) interactions with Jupiter and the other Galilean moons (O’Reilly and Davies, 1981). With a surface heat flux forty times that of Earth, Io is dominated by volcanic heat advection. The rapid cooling and subsequent burial of volcanic rocks at the surface leads to a downward advection of cold surface material, chilling the lithosphere. Perhaps counter-intuitively, the terrestrial body with the highest known heat flux exhibits a thick, cold single-plate mechanical lithosphere which is the consequence of rapid crustal/lithospheric formation from the top by addition of new flows. Recently, based on a survey of geological and geochemical observations, heat-pipe cooling has been suggested as a possible mechanism for the early cooling and lithospheric generation of Earth, for as much as the first third of Earth history (Moore and Webb, 2013). In this work we explore the possibility that the end of the magma ocean is not the point at which the solar system’s rocky bodies diverge in their evolution. Specifically, we review the records of the solar system’s stagnant-lid terrestrial bodies to explore whether they too may have experienced an early phase of heat-pipe cooling, as hypothesized for Earth and observed at present on Io. This overview highlights that ancient features common to these terrestrial bodies, e.g., the dominance of lowviscosity, mafic volcanic rocks, commonality of contractional features and scarcity of extensional features, and preservation of ancient topographic signatures, are consistent with the early heatpipe hypothesis. The tantalizing possibility motivating this exploration is that if heat-pipe cooling may have been experienced by all of our solar system’s terrestrial planets, it may be a phase experienced by terrestrial planets universally (Ricard et al., 2014). We begin with a detailed description of the heat-pipe cooling process before discussing the relevant observations of each stagnant-lid planet. 2. Heat pipe cooling The geological evolution of terrestrial (rocky) planets is driven by the transport of heat from their interiors. Although impacts may obscure endogenic resurfacing processes, everywhere that evidence for such processes is preserved, it is dominated by volcanic landforms (except on Earth, where liquid water, life, and plate tectonics dominate surface evolution). A terrestrial body cooling via heat pipes experiences persistent global volcanic resurfacing by which older layers are progressively buried and advected downward to form a thick, cold, and strong mechanical lithosphere (O’Reilly and Davies, 1981) (Fig. 1, Heat-Pipe inset). At this point we should note that the mechanical lithosphere of a heat pipe planet is composed almost entirely of the products of melting, that is, it is also the crust. Only when the mantle drops below the solidus can a portion of it adhere to the base of the crust forming what’s known as the mantle-lithosphere. While on present-day Earth, the crust is mechanically weaker than the mantle by a significant factor, the products of high-degree, heat-pipe
Fig. 1. Modeled evolution of lithospheric thickness (measured as a fraction of the mantle thickness) over time as internal heat production decreases by a factor of four (dashed line shows equilibrium lithospheric thickness increasing). Planets evolving through a heat-pipe phase (red) develop a thick lithosphere early in their history, which thins as volcanism wanes and then thickens as stagnant-lid convection takes over. Planets without melt transport transition directly from the magma ocean to stagnant-lid convection (blue) and begin with thin weak lithospheres that monotonically thicken and strengthen over time.
melting are not as evolved or silica rich and hence tend to be more similar in strength to the mantle. On Earth, this may be responsible for the longevity of cratonic mantle (e.g. the “tectosphere” of Jordan, 1978). The implications of heat pipes for the tectonic history of terrestrial bodies are illustrated in Fig. 1 by contrasting the modeled evolution of the lithospheric thickness for a planet with a heatpipe phase of evolution vs. one with a stagnant lid and no melting as heat generation decreases by a factor of four. The models were computed using STAGYY in a 4 × 1, two-dimensional domain with purely internal heating (decaying exponentially with time), strongly temperature-dependent viscosity (viscosity contrast 107 ), and a linearly pressure-dependent solidus with instantaneous melt extraction (see Kankanamge and Moore, 2016 for model details). Although greatly simplified, our treatment of the melting process captures the first order effects of melting and melt extraction on the heat transport which are that heat and material are advected to the surface on timescales that are rapid with respect to advective or diffusive processes. Unlike the classic stagnant-lid planet (blue) with a monotonically thickening lithosphere that tracks the equilibrium thickness (dashed), the heat-pipe planet (red) develops an early, thick lithosphere that is capable of recording and preserving early deformation events. Heat-pipe operation leads to: 1) Thick, cold, and strong lithospheres even though heat flow is high, 2) Dominance of compressive stresses as buried layers are forced to smaller radii, 3) Continuous replacement of lithospheric material, 4) High melt-fraction (mafic to ultra-mafic), low-viscosity eruptions and efficient degassing of the interior, and 5) A rapid transition to stagnant-lid or plate tectonic behavior. Heat pipe cooling ends when the internal heat production drops below the rate that can be accommodated by primarily sub-solidus convection (with or without plate tectonics). In the following sections, we briefly review the observations relevant to the formation of the surfaces of each of the terrestrial planets and current models that have been proposed to explain them. We then discuss any outstanding problems and show how the heat-pipe hypothesis can resolve these in a consistent way across all planets.
W.B. Moore et al. / Earth and Planetary Science Letters 474 (2017) 13–19
3. Mercury Mercury’s surface has now been extensively imaged by the MESSENGER spacecraft, and two important observations have emerged. First, Mercury was globally resurfaced until about 3.7 Ga by volcanic eruptions emplacing smooth plains with few identifiable eruption centers. The presence of lava channels exhibiting erosional features and low slopes (<1 deg) is also evidence for low-viscosity flows (Hurwitz et al., 2013). This volcanic resurfacing essentially ceased at 3.5 Ga (Byrne et al., 2016) with little activity other than several kilometers of global contraction since (Byrne et al., 2014). Surface modification prior to this time was able to degrade even very large (>500 km diameter) impact structures, reducing their density relative to the Moon (Fassett et al., 2012). The second important discovery of the Messenger mission is that Mercury’s surface composition is dominated by highly magnesian mafic to ultra-mafic basalts (Charlier et al., 2013) and that the geochemically distinct terranes identified by remote sensing are not strictly associated with mapped geologic units (Weider et al., 2015). Analysis of the melting conditions based on the observed surface compositions (Namur et al., 2016) reveals shallow source depths and large degrees of melting. Mercurian basalts show evidence for a gradual decrease in the extent of melting of their source rocks, from nearly 50% melting in the older Intra-Crater Plains (ICP) and Heavily Cratered Terrain (HCP) to ∼25% melting in the Northern Volcanic Plains province (Namur et al., 2016). Estimates of the crustal thickness reveal that the total amount of crustal production as a fraction of mantle volume (∼10%) is the highest of all terrestrial bodies for which we have observational constraints. Existing models of Mercury’s thermal and tectonic history are constrained by the observations of about 1 Ga of volcanic activity, the high melt fractions of the erupted melts, the total melt production (though any amount of crustal recycling is unknown), the total contraction recorded by the lithosphere, and the presence of a convecting liquid outer core. In fitting many of these constraints, recent models identify an early and possibly episodic period of magmatic activity ending after less than about 1 Ga (Tosi et al., 2013). The nature of these episodes and the implications for lithospheric evolution have not, however, been investigated in detail. In particular, Mercury’s response to the global compressive regime is represented by the network of lobate scarps, indicative of a strong elastic lithosphere that existed from early times, and not of a weak lithosphere in equilibrium with high conductive heat flow. The interplay of volcanic and impact processes may have led to a layered crust mixed to various degrees by large impacts and thus to geochemical terrains that cannot be easily correlated with currently visible geologic units (Weider et al., 2015). The geological observations summarized above and the thermochemical model results point to an episode of heat pipes operating for somewhat less than the first billion years of Mercury’s evolution. As heat-pipe volcanism continually replaces the lithosphere, the new layers take on the shape supported by the scaffolding of the old layers, accommodating any previous deformation while producing compressive stresses due to downward advection of the surface. Thus, the history of crustal growth and the history of lithospheric stress are coupled and the contraction recorded by the observable scarps is a mixture of thermal contraction and heatpipe-related shortening. The rapid transition of volcanism from a global resurfacing process up to about 3.7 Ga to a regional process to complete cessation at about 3.5 Ga (Byrne et al., 2016) is a result of the high efficiency of melt extraction as a heat transport process that essentially shuts itself off as soon as heat production is no longer able to sustain a global melt layer. The coincidence of this decrease with the rapid decline in impactor flux resulted in
15
a surface displaying a mixture of impact-dominated and volcanicdominated surfaces. 4. Venus The surface of Venus, like that of Mercury, is dominated by mafic lavas, with broad plains made up of numerous flows spanning hundreds of kilometers at low slope and with few identifiable source structures. Other volcanic features include fields of small shields on the volcanic plains, hundreds of roughly circular coronae, and a few dozen large shields associated with topographic highlands known as volcanic rises (e.g. Tanaka et al., 1997). Geoid and topography data require a generally thick lithosphere (200–400 km) which thins by up to 80% beneath volcanic highlands (Moore and Schubert, 1997; Orth and Solomatov, 2012). Plateau-shaped highlands, on the other hand, are supported by thickened crust (Moore and Schubert, 1997), show abundant evidence for compressional tectonics, and are among the earliest surface features. Observations of deformation patterns and craters indicate that Venus’s surface records a decrease in tectonic activity (Ivanov and Head, 2015). Early contractional structures include fold-thrust belts with mountains as high as 12 km, whereas deformation of the youngest ∼10% of the surface is largely limited to low-strain (<5%) rift belts extending between major volcanic centers (Grimm, 1994). The random distribution of Venusian craters has long been taken as evidence of catastrophic resurfacing around 0.5 Ga, although it is also consistent with a wider range of volcanic resurfacing models (e.g. Bjonnes et al., 2012). Recent analyses of crater morphology indicate that the majority (∼80%) of craters have experienced volcanic infilling, suggesting a young (∼150 Ma average age) surface that has experienced hundreds of meters of volcanic deposition (Herrick and Rumpf, 2011). Nonetheless, nearly all craters display pristine circular shapes (Herrick and Rumpf, 2011), so cessation of significant deformation must occur prior to 0.5 Ga. Embayment relationships indicate that the present high-standing topographic features are among the earliest components in Venus’s surface record (Gilmore et al., 1998), so the thick lithosphere constituting their support must also have been present early. Early speculation that Venus might host plate tectonics has been dismissed in light of findings that the materials and deformation of the surface are dominated by a single narrow age range and thus lack the time-progressive signature of plate-like crustal formation and recycling (Kaula and Phillips, 1981). Turcotte (1989) explored whether active heat-pipe cooling might explain lithospheric development at Venus, noting that this is consistent with the dominance of volcanics, thick lithosphere, and support of high topography, but he concluded that such activity would require a much higher volcanic flux than observed. Catastrophic volcanic resurfacing models involving global subduction and wholesale lithospheric replacement have been proposed (e.g. Parmentier and Hess, 1992). Because these models involve progressive lithospheric thickening subsequent to resurfacing events, they are difficult to reconcile with observations showing that the contractional highlands and their supporting thick lithosphere are among the earliest features seen on Venus. Similarly, models of contractional terrane development across thin lithosphere which progressively thickened (Hansen, 2006) are difficult to reconcile with the lack of deformation across the embayed margins of deformed highlands. Although Venus does not display sufficient volcanic flux to currently experience active heat-pipe cooling, the features which led (Turcotte, 1989) to explore this possibility may yet be evidence of a fossil heat-pipe lithosphere that subsequently transitioned into stagnant-lid cooling mode. The dominance of volcanics, thick lithosphere, and support of high topography could all be largely relicts of heat-pipe cooling, and the evident waning of both volcanism
16
W.B. Moore et al. / Earth and Planetary Science Letters 474 (2017) 13–19
and deformation could reflect the transition to the stagnant-lid regime. In this case, the single narrow age range of most of the Venusian surface would not reflect complete lithospheric replacement, but rather the last areally extensive new layers atop a lithosphere created via heat-pipe cooling. Finally, we acknowledge that the effective erasure of all available Venusian surface records prior to 0.5 Ga limit our ability to evaluate Venus’ tectonic evolution across most of its history. Early plate tectonic and/or episodic subduction periods could have preceded the proposed heat-pipe cooling period. Indeed, coupled mantle–atmosphere modeling suggests that low surface temperatures in the first billion years of Venus’ history would have encouraged mobile lid behavior (Gillmann and Tackley, 2014). Nonetheless, the surface record permits that a subsequent single-plate lithosphere could have been dominated by heat-pipe cooling and later transitioned into a stagnant-lid phase with limited ongoing volcanism. 5. Mars Among the most important surface features on Mars are its large volcanos (e.g., Olympus Mons), ancient cratered terrains, and its north–south crustal dichotomy. Volcanic flows have extensively modified the northern lowlands, burying a surface similar in age to the southern highlands (Frey, 2006), but there remains a significant and abrupt change in crustal thickness (Neumann et al., 2004) between the cratered uplands of the south and the smoother lowlands of the north. This dichotomy formed less than 50 Ma after accretion (Frey, 2006) and is associated with only minor contractional structures (Nimmo and Tanaka, 2005). Another key observation for the early evolution of Mars is the extreme range of trace element abundances and isotopic compositions among the main group of Martian meteorites, basaltic rocks called Shergottites. Evidence for early global differentiation of Mars comes from the ancient Rb–Sr and Pb–Pb isotopic model ages of their mantle source materials (Chen and Wasserburg, 1986). These ancient ages are further supported by age constraints obtained by the short-lived 146 Sm–142 Nd decay system that suggest global differentiation of Mars occurred a few 10’s to 100 Ma after solar system formation (Borg et al., 2003). Martian meteorites also show large variation in 142 Nd/144 Nd and 143 Nd/144 Nd (the product of long lived decay of 147 Sm). These extreme variations are ∼4 times greater than those of comparable mantle sources on Earth and the Moon and likely exist because portions of the Martian mantle were isolated into geochemically distinct reservoirs more than ∼4 Ga ago. Volatile studies (Usui et al., 2012) indicate that Shergottites support an early depleted dry (15–50 ppm water) interior. It remains unclear whether magmatic/tectonic or impact processes were responsible for the formation of the crustal dichotomy between the elevated southern hemisphere and the depressed northern hemisphere. Endogenic models for the formation of the dichotomy invoke either horizontal (plate-tectonic) or vertical (instability or plume) motions. Horizontal models lack evidence of the necessary plate boundary structures (the equivalent of spreading centers or convergent margins on Earth) that should coincide with the dichotomy boundary, while the internal instability models involving mantle convection with a single plume (Zhong and Zuber, 2001) are too slow to form the dichotomy given the ancient age of the northern basement. Models that explain the early formation of the dichotomy by magma ocean (Elkins-Tanton et al., 2005) or impact processes (Reese et al., 2011) fail to account for the preservation of the dichotomy through the early period when Martian heat flow was high. Preservation of the dichotomy boundary requires either a very dry or very cold crust able to resist deformation at high heat flow. Differences among the reported meteorite ages have been used to suggest that Mars had a magma ocean with a protracted
(∼100 Ma) crystallization history (Debaille et al., 2009). Such a protracted period of time is difficult to reconcile with standard magma ocean cooling models (Elkins-Tanton et al., 2005). Additionally, although the current surface of Mars is dry the conditions at the surface of early Mars and for much of its history were likely quite wet. The budget of volatile evolution in terrestrial planets is debated. One endmember view is that volatiles were introduced during accretion and lost from the interior by volcanic degassing (Drake and Righter, 2002). An alternative view is that planets accreted “dry” and are later hydrated by addition of volatile-rich materials from the outer solar system (Albarède, 2009). If Mars accreted wet it must have lost volatiles very early as they are found in very low abundances in the source region(s) of mantle-derived meteorites (Usui et al., 2012). Heat-pipe operation is capable of producing a thick, cold, and strong lithosphere very early in Mars’ history (Fig. 1) through rapid downward lithosphere advection, thereby preserving the crustal dichotomy and its sharp boundary. Heat pipe driven convergence also explains the observed dominance of contractional features. A heat-pipe planet that transitions to stagnant-lid convection (Fig. 1) would exhibit early and significant extraction of incompatible material to form a primordial crust with little to no subsequent remixing. It is reasonable to suggest that the level of geochemical depletion indicated by Martian meteorites at such an early stage is not seen among mantle melts on Earth because it no longer exists due to later subduction and return of significant amounts of enriched crustal materials to Earth’s interior. Much less crustal material has been recycled into the interior of Mars (Usui et al., 2012). A another prediction of the heat-pipe hypothesis would be that the isotopic systematics used to suggest an ancient period of mantle overturn and protracted magma ocean solidification (Debaille et al., 2009) actually reflects the onset and punctuated magmatic activity related to heat-pipe cooling. Finally, samplebased volatile studies (Usui et al., 2012) supporting an early dry mantle are consistent with the expectation of efficient degassing by heat-pipe volcanism, which would have aided in the preservation of an early crustal dichotomy. 6. Moon The Moon’s primordial history has been overprinted by early global chemical segregation, meteor bombardment, and volcanism. The most visible evidence of lunar chemical segregation is the existence of two very different terrains on the Moon: the highlands and the maria. The lunar highlands represent an ancient terrain dominated by plagioclase rich anorthositic rocks whose origins likely hold the key to understanding the early evolution of the Moon and potentially terrestrial planet formation in general. The prevalence of anorthosite in the lunar highlands and the existence of a rock type called ferroan anorthosite or “FAN” returned by the Apollo missions are generally attributed to crystallization and flotation of less dense plagioclase at the top of an ancient lunar magma ocean (LMO model) (Warren, 1985). The processes and timescales of how differentiation of the highlands occurred remain unresolved (Borg et al., 2011). This uncertainty is at least partly due to the fact that meteor impact gardening has reworked much of the ancient lunar surface, possibly to a depth of 3 km (Yamamoto et al., 2012). Nevertheless, remote sensing observations focused on outcrops within large impact craters (e.g., Orientale Basin) demonstrate that the upper lunar lithosphere contains thick, widespread, extremely plagioclase-rich anorthosite to depths of tens of km (Ohtake et al., 2009), supporting the importance of widespread differentiation on the Moon. Novel volatile element measurements (Hui et al., 2013) and some new age determinations of FANs (Borg et al., 2011) along with the recently reported geochemical characteristics of a diver-
W.B. Moore et al. / Earth and Planetary Science Letters 474 (2017) 13–19
17
Fig. 2. Stages of early Lunar crustal differentiation: I. Magma ocean, II. Lithospheric-scale magma crystal mush with stagnant crustal lid, III. Transition to Heat pipe, where melts cycle through crystal mush leading to buoyancy-driven crystal-melt separation, IV. Heat pipe, rapid lithospheric thickening in which relatively less mafic, Mg-suite melts intrude the bottom of the plagioclase-bearing gabbroic crust providing a mechanism that admixes a substantial portion of the lunar lithosphere. Heating at the bottom of the thickening lithosphere produces buoyant anorthositic melts that rise towards the surface. It is at this time that the observed highly refined plagio-crust (FAN) is finally produced.
sity of anorthosite clasts contained in lunar meteorite breccias (i.e., Takeda et al., 2006) have introduced new complications to the standard LMO hypothesis. These materials now appear to imply: (1) a wetter Moon than conventionally believed (McCubbin et al., 2010; Hui et al., 2013), (2) younger (4.3 Ga) FAN ages that no longer support the crystallization sequence originally used to develop the LMO model (Borg et al., 2015), and (3) igneous breccia clasts that exhibit more heterogeneity than previously seen in the Apollo collection expanding the compositional range to bridge the apparent gap between the FAN and the more alkali-enriched “Mg suite” rocks on classic anorthite content (Ca/[Ca + Na + K]) versus Mg-number (100Mg/[Mg + Fe]) plots (Takeda et al., 2006). Along with evidence for greater heterogeneity in their trace element abundances (e.g., rare-earth elements) these geochemical data imply that these sample did not all crystallize from a common magma source (Russell et al., 2014). Moreover, it is unclear whether a simple flotation crust is capable of producing the extremely high plagioclase contents (>98%) observed in both Apollo samples (Haskin et al., 1981) and remote sensing data (Ohtake et al., 2009), since a mostly solid lithosphere forms before plagioclase feldspar reaches saturation at 70% solidification (Snyder and Taylor, 1992; Lin et al., 2017). It is possible to model with reasonable physical parameters that a significant amount of felsic melt can escape from a crystal-mush network to form an early-formed plagioclase-rich floatation lid (Piskorz and Stevenson, 2014). However, the simplest formation scenarios for a floating cumulate are complicated by the near-uniformity of the alkali composition of the plagioclase, even as the mafic phases record significant variations in Mg content (Warren, 1985). In addition to these recent insights into its unusual crustal chemistry, the Moon stands out as having a shape that is dramatically out of hydrostatic equilibrium (even at the longest wavelengths spherical harmonic degree 2). The Moon’s shape is not a fossil of a synchronous rotator at any distance from the Earth (the ratio of the gravity coefficients J2 :C22 is in excess by a factor of 3: Goudas, 1964), but instead must record some other orbital/rotational state or combination of states. Collectively, these observations either support other models of crust formation on the Moon besides crystal flotation or suggest that there are complexities in the LMO scenario that allow for multiple generations of anorthosite formation (Norman et al., 1995; Gross et al., 2014) and hydrostatic disequilibrium. The only means by which the Moon’s disequilibrium shape can be preserved over geologic time is substantial lithospheric strength, but all present explanations for the observed shape rely on processes that occur very early in the Moon’s evolution when it is much hotter. What is required is a way to rapidly produce a strong lithosphere even when the Moon is young and hot, which is precisely the expected behavior of a body experiencing heat-pipe cooling (Fig. 1). The lithosphere is created rapidly and continuously, causing the shape to be recorded around the time the heat-
pipe mechanism shuts off, and leaving behind a strong, distorted lithosphere. Likewise, the aforementioned geochemical complications can be resolved for the Moon if the plagioclase-rich crust is produced and refined through a widespread episode of heatpipe magmatism (Fig. 2) rather than a process solely produced by density-driven plagioclase flotation. In the heat-pipe scenario, the base of the thickened lithosphere is remelted by incompatible element rich (i.e., alkalis and volatiles) mafic magmas from the deep interior (possibly equivalent to the Mg suite rocks), producing melts that are higher in plagioclase while at the same time introducing water and a range of mafic components. This refinement process occurs through volatile-assisted remelting, where the lowered solidus produces buoyant, Al-rich melts (Fig. 2). Repeated operation of this refining mechanism can result in the nearly pure plagioclase melts and textures observed in Apollo samples (Haskin et al., 1981), while at the same time explaining the variations in Mg contents, the diversity of materials contained in lunar meteorites, and the relatively high water abundance measured by Hui et al. (2013) in FAN rocks. If volatiles in FANs are introduced by the heat pipe process (analogous to the formation of Tonalite–Trondhjemite–Granodiroties on Earth) and not due to the canonical LMO fractional crystallization differentiation model, this evidence for a “wet” Moon is no longer valid. A wide-spread phase of heat-pipe volcanism that exchanges deep mantle-derived material with crustal rocks is also consistent with the latest geochronological observations that indicate that there is significant overlap between the ages determined for Mg suite rocks, FANs, and the ∼4.3 Ga model calculated for the source of younger maria basalts (Borg et al., 2015). A global period of serial volcanism (Longhi, 2003) eliminates the outstanding age problem with the standard LMO model (Shearer et al., 2006), because the re-melted FANs would be related and thus coeval with intrusion of the mafic Mg suite rocks as opposed to being older as required by the standard magma ocean model (Shearer et al., 2006). 7. Discussion and conclusions Heat pipes are an important feature of terrestrial planets at high heat flow, as illustrated by Io’s present activity. Evidence for their operation early in Earth’s history (Moore and Webb, 2013) suggests that all terrestrial bodies should experience an early episode of heat-pipe cooling. We have shown that the geological, geochemical, and geochronological evidence from the terrestrial bodies in our solar system is consistent with heat-pipe operation providing the main mechanism of crustal formation and endogenic resurfacing. Io is in the heat-pipe mode to the present day due to intense tidal heating, while the other planets (except Earth) have all transitioned to stagnant-lid convection with subsequent levels of geologic activity proportional to their heat content (Fig. 3). Venus is the most active, while Mercury and the Moon are long
18
W.B. Moore et al. / Earth and Planetary Science Letters 474 (2017) 13–19
tience and numerous reviewers for their assistance in improving this paper. References
Fig. 3. Illustration of terrestrial planet heat flow versus internal temperature. The sense of evolution as heat sources and internal heat content decline is shown by arrows. The terrestrial bodies are labeled, LTP stands for Large Terrestrial exo-Planet (several Earth-masses), and the initial magma ocean stage is indicated. The current temperature coordinates are notional (Venus may be warmer than Earth, for example). Heat loss is high and thermal evolution is rapid in the upper right of the diagram, where the magma ocean gives way to heat-pipes (dotted), and heat flow decreases as heat-pipes transition to either plate tectonics (dashed) or stagnant lid sub-solidus convection.
dead. Note that the main planetary (as opposed to material) properties contributing to the efficiency of heat transport by subsolidus convection are gravity, surface to interior temperature difference, and, to a lesser degree, mantle depth, therefore smaller planets have less efficient convection as do planets with higher surface temperatures (e.g. Venus). Only on Earth have buoyancy forces in the mantle overcome the strength of the lithosphere resulting in a transition to plate tectonics as indicated by the dashed line. The heat-pipe hypothesis therefore provides a uniform explanation for several common features of the known terrestrial planets that have not undergone plate tectonics and should be considered an important aspect of their evolution. Finally, we note that the conditions for the operation of plate tectonics have been the subject of considerable debate, particularly when applied to the large terrestrial bodies (LTP in Fig. 3) that have recently been found orbiting other stars (e.g. Stamenkovic´ et al., 2011). It has been argued that such bodies are both more and less likely to have plate tectonics. Earth appears to have transitioned to plate tectonics about 1 Ga after formation (Shirey and Richardson, 2011; Moore and Webb, 2013). Since the equilibrium heat flux of a planet scales as mass/area (for most plausible heat sources), terrestrial planets more massive than the Earth should experience longer heat-pipe episodes prior to the initiation of plate tectonics. A planet twice as massive as Earth should take more than twice as long to cool because the area increases less rapidly than a simple mass scaling would suggest due to the compressibility of terrestrial materials (Stamenkovic´ et al., 2011). For the large “super-Earths” over five Earth masses, the lifetime of the heat-pipe phase may exceed the lifetime of Sun-like parent stars and thus any subsequent plate-tectonic phase may never be observed. Such planets might better be called “super-Ios,” driving us to reconsider what types of surfaces and atmospheres to expect as we expand our exploration of other solar systems. Acknowledgements W.B.M. acknowledges funding from NSF EAR-1246983 and NExSS NNX15AE05G. J.I.S. acknowledges NASA funding from the Lunar Advanced Science & Exploration Research Program 10-LASER10-0077, the Mars Fundamental Research Program 10-MFRP10-0022, and the Solar System Workings Program 15-SSW15-2-0411. We would like to thank the editor for his pa-
Albarède, F., 2009. Volatile accretion history of the terrestrial planets and dynamic implications. Nature 461, 1227–1233. Bjonnes, E.E., Hansen, V.L., James, B., Swenson, J.B., 2012. Equilibrium resurfacing of Venus: results from new Monte Carlo modeling and implications for Venus surface histories. Icarus 217, 451–461. Borg, L.E., Connelly, J.N., Boyet, M., Carlson, R.W., 2011. Chronological evidence that the Moon is either young or did not have a global magma ocean. Nature 477, 70–72. Borg, L.E., Draper, D.S., 2003. A petrogenic model for the origin and compositional variation of the martian basaltic meteorites. Meteorit. Planet. Sci. 38, 1713–1731. Borg, L.E., Gaffney, A.M., Shearer, C.K., 2015. A review of lunar chronology revealing a preponderance of 4.34–4.37 Ga ages. Meteorit. Planet. Sci. 50, 715–732. Borg, L.E., Nyquist, L.E., Wiesmann, H., Shih, C.-Y., Reese, Y., 2003. The age of Dar al Gani 476 and the differentiation history of the Martian meteorites inferred from their radiogenic isotopic systematics. Geochim. Cosmochim. Acta 67, 3519–3536. Byrne, P.K., Klimczak, C., Sengör, ¸ A.M.C., Solomon, S.C., Watters, T.R., Hauck, S.A.I., 2014. Mercury’s global contraction much greater than earlier estimates. Nat. Geosci. 7, 301–307. Byrne, P.K., Ostrach, L.R., Fassett, C.I., Chapman, C.R., Denevi, B.W., Evans, A.J., Klimczak, C., Banks, M.E., Head, J.W., Solomon, S.C., 2016. Widespread effusive volcanism on Mercury likely ended by about 3.5 Ga. Geophys. Res. Lett. 43, 7408–7416. Charlier, B., Grove, T.L., Zuber, M.T., 2013. Phase equilibria of ultramafic compositions on Mercury and the origin of the compositional dichotomy. Earth Planet. Sci. Lett. 363, 50–60. Chen, J.H., Wasserburg, G.J., 1986. Formation ages and evolution of Shergotty and its parent planet from U–Th–Pb systematics. Geochim. Cosmochim. Acta 50, 955–968. Debaille, V., Brandon, A.D., O’Neill, C., Yin, Q.-Z., Jacobsen, B., 2009. Early Martian mantle overturn inferred from isotopic composition of nakhlite meteorites. Nat. Geosci. 2, 548–552. Debaille, V., O’Neill, C., Brandon, A.D., Haenecour, P., Yin, Q.-Z., Mattielli, N., Treiman, A.H., 2013. Stagnant-lid tectonics in early Earth revealed by 142 Nd variations in late Archean rocks. Earth Planet. Sci. Lett. 373, 83–92. Drake, M.J., Righter, K., 2002. Determining the composition of the Earth. Nature 416, 39–44. Elkins-Tanton, L.T., Hess, P.C., Parmentier, E.M., 2005. Possible formation of ancient crust on Mars through magma ocean processes. J. Geophys. Res., Planets 110 (E9), 12. Fassett, C.I., Head, J.W., Baker, D.M.H., Zuber, M.T., Smith, D.E., Neumann, G.A., Solomon, S.C., Klimczak, C., Strom, R.G., Chapman, C.R., Prockter, L.M., Phillips, R.J., Oberst, J., Preusker, F., 2012. Large impact basins on Mercury: global distribution, characteristics, and modification history from MESSENGER orbital data. J. Geophys. Res., Planets 117, E00L08. Frey, H.V., 2006. Impact constraints on, and a chronology for, major events in early Mars history. J. Geophys. Res. 111, 8. Gillmann, C., Tackley, P., 2014. Atmosphere/mantle coupling and feedbacks on Venus. J. Geophys. Res., Planets 119, 1189–1217. Gilmore, M.S., Collins, G.C., Ivanov, M.A., Marinangeli, L., Head, J.W., 1998. Style and sequence of extensional structures in tessera terrain, Venus. J. Geophys. Res. 103, 16813–16840. Goudas, C.L., 1964. Moments of inertia and gravity field of the Moon. Icarus 3, 375–409. Grimm, R.E., 1994. The deep structure of Venusian plateau highlands. Icarus 112, 89–103. Gross, J., Treiman, A.H., Mercer, C.N., 2014. Lunar feldspathic meteorites: constraints on the geology of the lunar highlands, and the origin of the lunar crust. Earth Planet. Sci. Lett. 388, 318–328. Hansen, V.L., 2006. Geologic constraints on crustal plateau surface histories, Venus: the lava pond and bolide impact hypotheses. J. Geophys. Res., Planets 111 (E10), E11010. Harrison, T.M., 2009. The Hadean Crust: evidence from >4 Ga zircons. Annu. Rev. Earth Planet. Sci. 37, 479–505. Haskin, L.A., Lindstrom, M.M., Salpas, P.A., 1981. Some observations on compositional characteristics of lunar anorthosites. In: Lunar and Planetary Science Conference, vol. 12, pp. 406–408. Technical Report. Herrick, R.R., Rumpf, M.E., 2011. Postimpact modification by volcanic or tectonic processes as the rule, not the exception, for Venusian craters. J. Geophys. Res., Planets 116, 2004. Hui, H., Peslier, A.H., Zhang, Y., Neal, C.R., 2013. Water in lunar anorthosites and evidence for a wet early Moon. Nat. Geosci. 6, 177–180. Hurwitz, D.M., Head, J.W., Byrne, P.K., Xiao, Z., Solomon, S.C., Zuber, M.T., Smith, D.E., Neumann, G.A., 2013. Investigating the origin of candidate lava channels
W.B. Moore et al. / Earth and Planetary Science Letters 474 (2017) 13–19
on Mercury with MESSENGER data: theory and observations. J. Geophys. Res., Planets 118, 471–486. Ivanov, M.A., Head, J.W., 2015. Volcanically embayed craters on Venus: testing the catastrophic and equilibrium resurfacing models. Planet. Space Sci. 106, 116–121. Jordan, T.H., 1978. Composition and development of the continental tectosphere. Nature 274, 544–548. Kankanamge, D.G.J., Moore, W.B., 2016. Heat transport in the hadean mantle: from heat pipes to plates. Geophys. Res. Lett. 43, 3208–3214. Kaula, W.M., Phillips, R.J., 1981. Quantitative tests for plate tectonics on Venus. Geophys. Res. Lett. 8, 1187–1190. Lin, Y., Tronche, E.J., Steenstra, E.S., van Westrenen, W., 2017. Experimental constraints on the solidification of a nominally dry lunar magma ocean. Earth Planet. Sci. Lett. 471, 104–116. URL http://www.sciencedirect.com/science/ article/pii/S0012821X1730239X. Longhi, J., 2003. A new view of lunar ferroan anorthosites: postmagma ocean petrogenesis. J. Geophys. Res., Planets 108, 5083. McCubbin, F.M., Steele, A., Hauri, E.H., Nekvasil, H., Yamashita, S., Hemley, R.J., 2010. Nominally hydrous magmatism on the Moon. Proc. Natl. Acad. Sci. 107, 11223–11228. Moore, W.B., Schubert, G., 1997. Venusian crustal and lithospheric properties from nonlinear regressions of highland geoid and topography. Icarus 128, 415–428. Moore, W.B., Webb, A.A.G., 2013. Heat-pipe earth. Nature 501, 501–505. Namur, O., Collinet, M., Charlier, B., Grove, T.L., Holtz, F., McCammon, C., 2016. Melting processes and mantle sources of lavas on Mercury. Earth Planet. Sci. Lett. 439, 117–128. Neumann, G.A., Zuber, M.T., Wieczorek, M.A., McGovern, P.J., Lemoine, F.G., Smith, D.E., 2004. Crustal structure of Mars from gravity and topography. J. Geophys. Res., Planets 109, 8002. Nimmo, F., Tanaka, K., 2005. Early Crustal Evolution of Mars. Annu. Rev. Earth Planet. Sci. 33, 133–161. Norman, M.D., Keil, K., Griffin, W.L., Ryan, C.G., 1995. Fragments of ancient lunar crust: petrology and geochemistry of ferroan noritic anorthosites from the Descartes region of the Moon. Geochim. Cosmochim. Acta 59, 831–847. Ohtake, M., Matsunaga, T., Haruyama, J., Yokota, Y., Morota, T., Honda, C., Ogawa, Y., Torii, M., Miyamoto, H., Arai, T., Hirata, N., Iwasaki, A., Nakamura, R., Hiroi, T., Sugihara, T., Takeda, H., Otake, H., Pieters, C.M., Saiki, K., Kitazato, K., Abe, M., Asada, N., Demura, H., Yamaguchi, Y., Sasaki, S., Kodama, S., Terazono, J., Shirao, M., Yamaji, A., Minami, S., Akiyama, H., Josset, J.-L., 2009. The global distribution of pure anorthosite on the Moon. Nature 461, 236–240. O’Reilly, T.C., Davies, G.F., 1981. Magma transport of heat on Io: a mechanism allowing a thick lithosphere. Geophys. Res. Lett. 8, 313–316. Orth, C.P., Solomatov, V.S., 2012. Constraints on the Venusian crustal thickness variations in the isostatic stagnant lid approximation. Geochem. Geophys. Geosyst. 13, Q11012. Parmentier, E.M., Hess, P.C., 1992. Chemical differentiation of a convecting planetary interior: consequences for a one plate planet such as Venus. Geophys. Res. Lett. 19, 2015–2018. Piskorz, D., Stevenson, D.J., 2014. The formation of pure anorthosite on the Moon. Icarus 239, 238–243. Reese, C.C., Orth, C.P., Solomatov, V.S., 2011. Impact megadomes and the origin of the Martian crustal dichotomy. Icarus 213, 433–442.
19
Ricard, Y., Labrosse, S., Dubuffet, F., 2014. Lifting the cover of the cauldron: convection in hot planets. Geochem. Geophys. Geosyst. 15 (11). Russell, S.S., Joy, K.H., Jeffries, T.E., Consolmagno, G.J., Kearsley, A., 2014. Heterogeneity in lunar anorthosite meteorites: implications for the lunar magma ocean model. Philos. Trans. R. Soc. Lond. Ser. A 372, 20130241. Shearer, C.K., Hess, P.C., Wieczorek, M.A., Pritchard, M.E., Parmentier, E.M., Borg, L.E., Longhi, J., Elkins-Tanton, L.T., Neal, C.R., Antonenko, I., Canup, R.M., Halliday, A.N., Grove, T.L., Hager, B.H., Lee, D.-C., Wiechert, U., 2006. Thermal and magmatic evolution of the Moon. In: Joliff, B.L., Wieczorek, M.A., Shearer, C.K., Neal, C.R. (Eds.), New Views of the Moon. Min. Soc. Am., Chantilly, VA, pp. 365–518. Shirey, S.B., Richardson, S.H., 2011. Start of the Wilson cycle at 3 Ga shown by diamonds from subcontinental mantle. Science 333, 434–436. Snyder, G.A., Taylor, L.A., 1992. Imperfect fractional crystallization of the lunar magma ocean and formation of the lunar mantle: a realistic chemical approach. In: Agee, C.B., Longhi, J. (Eds.), Physics and Chemistry of Magma Oceans from 1 Bar to 4 Mbar. Solomon, S.C., 1977. The relationship between crustal tectonics and internal evolution in the moon and Mercury. Phys. Earth Planet. Inter. 15, 135–145. ´ V., Breuer, D., Spohn, T., 2011. Thermal and transport properties of Stamenkovic, mantle rock at high pressure: applications to super-Earths. Icarus 216, 572–596. Takeda, H., Yamaguchi, A., Bogard, D.D., Karouji, Y., Ebihara, M., Ohtake, M., Saiki, K., Arai, T., 2006. Magnesian anorthosites and a deep crustal rock from the farside crust of the moon. Earth Planet. Sci. Lett. 247, 171–184. Tanaka, K.L., Senske, D.A., Price, M., Kirk, R.L., 1997. Physiography, geomorphic/geologic mapping and stratigraphy of Venus. In: Bougher, S.W., Hunten, D.M., Phillips, R.J. (Eds.), Venus II: Geology, Geophysics, Atmosphere, and Solar Wind Environment, p. 667. Tosi, N., Grott, M., Plesa, A.-C., Breuer, D., 2013. Thermochemical evolution of Mercury’s interior. J. Geophys. Res., Planets 118, 2474–2487. Turcotte, D.L., 1989. A heat pipe mechanism for volcanism and tectonics on Venus. J. Geophys. Res. 94, 2779–2785. Usui, T., Alexander, C.M.O., Wang, J., Simon, J.I., Jones, J.H., 2012. Origin of water and mantle-crust interactions on Mars inferred from hydrogen isotopes and volatile element abundances of olivine-hosted melt inclusions of primitive shergottites. Earth Planet. Sci. Lett. 357, 119–129. Warren, P.H., 1985. The magma ocean concept and lunar evolution. Annu. Rev. Earth Planet. Sci. 13, 201–240. Weider, S.Z., Nittler, L.R., Starr, R.D., Crapster-Pregont, E.J., Peplowski, P.N., Denevi, B.W., Head, J.W., Byrne, P.K., Hauck, S.A., Ebel, D.S., Solomon, S.C., 2015. Evidence for geochemical terranes on Mercury: global mapping of major elements with MESSENGER’s X-ray spectrometer. Earth Planet. Sci. Lett. 416, 109–120. Yamamoto, S., Nakamura, R., Matsunaga, T., Ogawa, Y., Ishihara, Y., Morota, T., Hirata, N., Ohtake, M., Hiroi, T., Yokota, Y., Haruyama, J., 2012. Olivine-rich exposures in the South Pole-Aitken Basin. Icarus 218, 331–344. Yin, A., 2012. An episodic slab-rollback model for the origin of the Tharsis rise on Mars: implications for initiation of local plate subduction and final unification of a kinematically linked global plate-tectonic network on Earth. Lithosphere 4, 553–593. Zhong, S., Zuber, M.T., 2001. Degree-1 mantle convection and the crustal dichotomy on Mars. Earth Planet. Sci. Lett. 189, 75–84.