Volcanism on Mercury
- David A. RotheryDavid A. RotheryFaculty of Science, Technology, Engineering & Mathematics, The Open University
The history of volcanism on Mercury is almost the entire history of the formation of its crust. There are no recognized tracts of intact primary crust analogous to the Moon’s highland crust, probably because the density of Mercury’s iron-poor magma ocean was insufficient to enable crystalized silicate phases to float. Mercury’s surface consists of multiple generations of lavas. These were emplaced, rather like terrestrial “large igneous provinces” or LIPs, in their greatest volumes prior to about 3.5 Ga. Subsequently, erupted volumes decreased, and sites of effusive eruption became largely confined to crater floors. Plains lava surfaces younger than about 3.7 Ga have become scarred by sufficiently few impact craters that they are mapped as “smooth plains.” The older equivalents, which experienced the inner solar system’s “late heavy bombardment,” are mapped as intercrater plains. There is no consensus over whether plains with superimposed-crater characteristics that are intermediate between the smooth plains and intercrater plains end members can be consistently mapped as “intermediate plains.” However, any subdivision of the volcanic plains for mapping purposes arbitrarily splits apart a continuum.
The volcanic nature of Mercury’s smooth plains was ambiguous on the basis of the imagery returned by the first mission to Mercury, Mariner 10, which made three fly-bys in 1974–1975. Better and more complete imaging by MESSENGER (in orbit 2011–2015) removed any doubt by documenting innumerable ghost craters and wrinkle ridges. No source vents for the plains are apparent, but this is normal in LIPs where effusion rate and style characteristically flood the vent beneath its own products. However, there are good examples of broad, flat-bottomed valleys containing streamlined islands suggesting passage of fast-flowing low viscosity lava.
Although the causes of the mantle partial melting events supplying surface eruptions on Mercury are unclear, secular cooling of a small, one-plate planet such as Mercury would be expected to lead to the sort of temporal decrease in volcanic activity that is observed. Factors include loss of primordial heat and declining rate of radiogenic heat production (both of which would make mantle partial melting progressively harder), and thermal contraction of the planet (closing off ascent pathways).
Lava compositions, so far as can be judged from the limited X-ray spectroscopic and other geochemical measurements, appear to be akin to terrestrial komatiites but with very low iron content. Variations within this general theme may reflect heterogeneities in the mantle, or different degrees of partial melting.
The cessation of flood volcanism on Mercury is hard to date, because the sizes of the youngest flows, most of which are inside <200-km craters, are too small for reliable statistics to be derived from the density of superposed craters. However, it probably continued until approximately 1 Ga ago.
That was not the end of volcanism. MESSENGER images have enabled the identification of over a hundred “pits,” which are noncircular holes up to tens of km in size and up to about 4 km deep. Many pits are surrounded by spectrally red deposits, with faint outer edges tens of km from the pit, interpreted as ejecta from explosive eruptions within the pit. Many pits have complex floors, suggesting vent migration over time. Pits usually occur within impact craters, and it has been suggested that crustal fractures below these craters facilitated the ascent of magma despite the compressive regime imposed by the secular thermal contraction. These explosive eruptions must have been driven by the violent expansion of a gas. This could be either a magmatic volatile expanding near the top of a magma conduit, or result from heating of a near-surface volatile by rising magma. MESSENGER showed that Mercury’s crust is surprisingly rich in volatiles (S, Cl, Na, K, C), of which the one likely to be of most importance in driving the explosive eruptions is S.
We do not know when explosive volcanism began on Mercury. Cross-cutting relationships suggest that some explosion pits are considerably less than 1 Ga old, though most could easily be more than 3 Ga. They characteristically occur on top of smooth plains (or less extensive smooth fill of impact craters), and while some pits have no discernible “red spot” around them (perhaps because over time, it has faded into the background), there is no known example of part of a red spot peeping out from beneath the edge of a smooth plains unit. There seems to have been a change in eruptive style over time, with (small volume) explosions supplanting (large volume) effusive events.
This article summarizes what is known of volcanism on Mercury, in the light of mature (but still ongoing) analysis of data returned by the MESSENGER spacecraft that orbited the planet 2011–2015. It appears that almost the whole of Mercury’s surface is volcanic, so that a survey of Mercury’s volcanic history is pretty much a survey of upper crustal history, as well as offering windows into magmagenesis, intrusive processes, and the availability and mobility of volatiles.
Mercury’s surface consists of multiple generations of lavas emplaced in a manner similar to terrestrial “large igneous provinces” or LIPs. The greatest volumes were erupted prior to about 3.5 Ga. Subsequently erupted volumes decreased, and sites of eruption became largely confined to crater floors. Lava effusion appears to have ceased sometime within the past 1 to 2 Ga. The youngest known volcanism on Mercury (within the past Ga) is explosive, via vents punched up through lava plains and impact craters. This supplanting of large-scale effusive volcanism by smaller-scale explosive volcanism may have much to tell about magmagenesis, magma migration, and the activity of volatiles over time.
Mercury is an airless heavily-cratered body, and at first sight looks deceptively like the Earth’s Moon. However, there are important contrasts. Mercury lacks the Moon’s prominent surface dichotomy between higher-albedo highlands and lower-albedo maria, which is a consequence of contrasting conditions in their magma oceans soon after formation. Mercury has a very large core whose outer part is fluid and is the seat of generation of a magnetic field (Anderson et al., 2011) that is sufficiently strong to deflect the solar wind most of the time, whereas the Moon’s core is tiny and does not generate a magnetic field (Weber et al., 2011). Mercury’s crust is rich in volatile elements, whereas the Moon (including its crust) is depleted in volatiles, a contrast that remains true despite recent detection of water, chlorine, and fluorine by sophisticated analysis of lunar samples (e.g., Barnes et al., 2016). (Cold-trapped volatiles of presumed late cometary origin found in permanently shadowed polar craters, which occur on both bodies, are unrelated to each body’s intrinsic volatile inventory.)
Mercury is an end-member planet in the solar system. It is the smallest terrestrial planet (2440-km radius), has the greatest uncompressed density, and is closest to the Sun. In studying it, we learn about processes likely to affect the most easily observed exoplanets, and about possible inward migration of planetary embryos in the early stages of our own solar system. Its proximity to the Sun makes it hard to study from Earth, and thus far it has been visited by only two spacecraft, Mariner 10 and MESSENGER. The joint ESA-JAXA (European Space Agency, Japan Aerospace Exploration Agency) mission BepiColombo will deliver two orbiters to the planet in 2025, from which we can expect to learn much more.
The Pre-MESSENGER View
Before the space age, little could be learned of Mercury that was of relevance to its volcanology. Mercury has no moon from which its mass could be determined, but orbital perturbations to passing comets showed its mass to be surprisingly high for its size. Johann Encke obtained a mass within 20 percent of the correct value as long ago as 1841. Mercury’s density is 5.43 × 103 kg m−3, only slightly less than the Earth’s 5.51 × 103 kg m−3. A more meaningful comparison between the two planets can be made by allowing for the internal self-compression that results from each planet’s own gravity. On this basis, Mercury’s uncompressed density is 5.3 × 103 kg m−3, whereas Earth’s is only 4.0 × 103 kg m−3 (e.g., Siegfried & Solomon, 1974). From these data alone, it was apparent that Mercury’s core (the densest part of any planet) must be relatively much larger than the Earth’s. In fact, the total thickness of silicates (crust plus mantle) overlying Mercury’s core can be no more than about 400 km.
It was also known, mostly from radar observations in the 1960s (Pettengill & Dyce, 1965), which proved to be a better guide than visual telescopic observations, that Mercury rotates on its axis exactly three times per every two orbits of the Sun. This discredited an earlier assumption that it must be tidally locked into synchronous rotation (one rotation per orbit). It also fitted neatly with new understanding of the tidal torque resulting from Mercury’s eccentric solar orbit, which showed that such a 1:1 spin-to-orbit coupling would be unstable, in contrast to the stability of the observed 3-to-2 coupling (Peale & Gold, 1965).
Mariner 10 and the Smooth Plains Controversy
Mariner 10, the first spacecraft to visit Mercury, made three fly-bys in 1974–1975. In addition to its unexpected discovery of the planetary magnetic field (Ness et al., 1975), these encounters, limited by orbital dynamics to be made under identical illumination conditions each time, resulted in imaging adequate to permit mapping of 40 to 45 percent of the planet (Murray, 1975). The images were essentially black and white (color imaging was of poor quality) and mostly with a spatial resolution of 2 to 4 km, though small areas could be imaged at 100-m resolution during the hour of closest approach (Murray, 1975).
These images showed many similarities to the Moon, and the density of superposed impact craters demonstrated Mercury’s surface to be ancient. Differing crater densities in adjacent terrains clearly demonstrated age differences. Figure 1 shows a typical region as seen by Mariner 10. It is immediately obvious that Mercury lacks the albedo contrast that makes the lunar maria so distinct from brighter lunar highlands. In fact, the albedo everywhere on Mercury is more similar to lunar maria than to the lunar highlands. Probably the most remarkable features revealed on the images were numerous contractional tectonic features, most notably lobate scarps up to 2 km high and up to several hundred km long, interpreted to be surface manifestations of thrust faults (Dzurisin, 1978; Strom et al., 1975). These were attributed to global thermal contraction, with an uncertain degree of influence by tidal de-spinning (Melosh & McKinnon, 1988). They are two orders of magnitude bigger than equivalent features on the Moon (Binder & Gunga, 1985; Watters et al., 2012).
Study and interpretation of images such as those used for Figure 1 allowed those working with Mariner 10 images to define a chronostratigraphic timescale for Mercury (Spudis, 1985; Spudis & Guest, 1988) analogous to the one that had been established for the Moon during the Apollo era (Shoemaker & Hackman, 1962; Wilhelms, 1970; Wilhelms & McCauley, 1971). Table 1 names Mercury’s five chronostratigraphic systems, and shows the originally proposed ages of their bases plus revised ages for the starts of the two youngest systems based on more recent modeling of impactor flux.
Table 1. The Chronostratigraphic Timescale for Mercury, as Defined Using Mariner 10 Data, Plus Revisions to the Younger End by Banks et al. (2017). The bases of the four youngest systems are defined by the formation of the basins or craters Tolstoj, Caloris, Mansur, and Kuiper
Major Units and Characteristics
Age of Base/Ga
Age of Base/Ga Revised
Fresh craters with rays
0.28 ± 0.06
Fresh craters without surviving rays
1.7 ± 0.2
Caloris basin and its ejecta. Plains inside and outside Caloris. Partially degraded craters
Tolstoj basin, its ejecta and associated plains. More heavily degraded craters
Intercrater plains and basins older than Tolstoj. Most heavily degraded craters
It is clear on Figure 1 that there are large tracts of “smooth plains” with relatively few superimposed craters. These include Caloris Planitia (the floor of the Caloris basin); the plain extending west, northwest, and north of the crater Mansur; and the shallow, flat-floored (flooded) 270-km crater Van Eyck immediately southeast of Mansur. Mariner 10 imaging was sufficient for it to be recognized (e.g., Spudis & Guest, 1988) that smooth plains of Calorian age were the last resurfacing event of regional extent, and that Mansurian-age plains are confined to crater floors.
There are essentially three plausible origins of smooth plains on an airless body such as Mercury. Plains associated with basins could be impact melt that was generated by the basin-forming impact (this can be ruled out when the basin is too deeply flooded to be accounted for by a plausible melt-volume, as is usually the case on Mercury); plains at any location could be lavas supplied by effusive volcanism; or plains could represent sheets of fluidized impact ejecta. Most Mariner-era workers favored effusive volcanism, but were rightly cautious (e.g., Strom et al., 1975) in view of a recent debacle when Apollo 16 had shown that the Moon’s Cayley plains are not in fact a high-albedo mare lava or tuff facies but impact ejecta from the Imbrium basin (Muehlberger et al., 1972).
The lack of albedo contrast between Mercury’s plains and older terrain seemed to support an ejecta origin for the plains rather than a volcanic origin. Doubts at the time were nicely summarized by Don Wilhelms (1976) in a cautionary analysis directed at the proponents of plains volcanism on Mercury: “I cannot disprove Mercurian volcanism but wish to point out that many of their arguments are very similar to those advanced for the volcanic origin of light plains material on the Moon during the stage of telescopic and orbital study, but which were discredited by the Apollo missions. Therefore the question of Mercurian volcanism, with its important implications for the thermal and impact history of Mercury and the other inner planets, should be kept open.”
The nature and origin of plains material remained controversial until MESSENGER imaging settled the argument firmly in favor of effusive volcanism (see the section “Effusive volcanism conclusively recognized”), for the widespread plains units at least, although the smooth fill of some craters could still be impact melt.
Mariner 10 was not equipped to determine Mercury’s elemental abundances or its surface mineralogy. This is clearly of relevance to the nature of the surface rocks, or rather of the regolith that is believed to cover the bedrock from which it is derived even more effectively than is the case on the Moon (Kreslavsky et al., 2014; Langevin, 1997; Shevchenko, 2002, 2004). Color imaging was adequate to demonstrate heterogeneity, but also a lack of relationship between color boundaries and the smooth plains (Rava & Hapke, 1987). A few dark blue diffuse-edged deposits were suggested to have a pyroclastic origin (Robinson & Lucey, 1997), but MESSENGER images subsequently showed these to be these to be fairly unremarkable ejecta patches of “low-reflectance material” that should not be confused with the bright red deposits subsequently recognized as pyroclastic in MESSENGER data (Head et al., 2008).
Telescopic study of Mercury is hampered by the planet’s proximity to the Sun, preventing it from being observed in a dark sky except when very close to the horizon. Prior to MESSENGER, reflected light spectroscopy and emitted infrared spectroscopy using optical telescopes were among the few techniques capable of giving information on Mercury’s surface mineralogy. The lack of a strong absorption at 900–1100 nm, expected from electron transitions in the Fe-O bond, was used to place an upper limit of about 3 percent by weight of Fe-O in silicates (Warell & Blewett, 2004), possibly occurring mostly in clinopyroxenes as suggested by mid-IR spectroscopy from Hawaii (Warell et al., 2006). Even more stringently, ground-based microwave imaging showed that Mercury’s regolith is remarkably transparent to microwaves (low dielectic loss tangent), putting an upper limit on the FeO+TiO2 content of about 2 wt percent. Such low FeO was surprising (for comparison, units on the Moon range from 4 to 27 percent FeO) and it was suggested that this could be explained by space weathering in which the Fe-O bonds had been broken to form nanophase iron, which would lower the albedo while being spectrally featureless (e.g., Noble & Pieters, 2003).
The issue of whether the expected iron at Mercury’s surface is hidden in an undetectable mineral phase or is genuinely absent was settled thanks to measurements by MESSENGER, as described next.
The MESSENGER View
MESSENGER, a name ponderously contrived from MErcury Surface Space ENvironment, GEochemistry, and Ranging, was a spacecraft with a 1,107 kg launch mass, of which over half was propellant. It was launched on August 3, 2004, flew back past the Earth a year later, then made two fly-bys of Venus and three of Mercury (January 14, 2008; October 6, 2008; and September 29, 2009) before achieving orbit around Mercury on March 18, 2011. The successive planetary fly-bys were “gravity assist” maneuvers that enabled the probe to match velocity with Mercury on its final approach so it could be captured into orbit without having to carry an unfeasible amount of fuel (Solomon et al., 2007).
The thermal environment that a Mercury orbiter has to endure is severe, especially over the dayside of the planet when the spacecraft must contend with overhead solar flux ten times stronger than at Earth, and heat radiated from the 700-K noontime surface below. The severity of this was mitigated by placing MESSENGER in an eccentric 12-hour polar orbit about Mercury, passing about 200 km from the surface at high northern latitudes but 15,200 km over high southern latitudes. Because orbital speed decreases with increasing range, MESSENGER swooped quickly through the closest and hottest part of its orbit and had longer to radiate excess heat to space while in the distant part of its orbit. MESSENGER’s planned nominal mission was one year in orbit, but it remained healthy and in April 2012 the far point (“apoherm”) of the orbit was lowered to 10,314 km, so that the orbital period became about eight hours. From September 2014 the near point of the orbit (“periherm”) was allowed to drift below 30-km altitude on four occasions until, its propellant exhausted, the probe impacted the surface on April 30, 2015.
The latitude of periherm was between about 80° and 60° N throughout the entire mission, so that the nature and quality of the data obtained differs between hemispheres. For example, beyond about 10° S the laser altimeter could detect no usable return signal, and the fields of view of the X-ray spectrometer and gamma-ray and neutron spectrometer rendered them unable to spatially resolve detail in the southern hemisphere. On the other hand, the high northern latitudes have gaps between swaths imaged by MESSENGER’s narrow-angle camera at highest resolution (e.g., Thomas et al., 2014a), because of the trade-off between swath width and spatial resolution depending on altitude above the surface.
The X-ray and gamma ray spectrometers on MESSENGER proved that iron in any form is scarce at Mercury’s surface. Nittler et al. (2011) found an upper limit of 4 wt percent Fe based on X-ray spectrometry, whereas Evans et al. (2012) estimated a northern hemisphere average of 1.9±0.3 wt percent Fe (Evans et al., 2012). This crustal depletion in iron is now regarded as a result of extreme partitioning of iron into the core under highly reducing conditions (McCubbin et al., 2012; Namur et al., 2016a).
Effusive Volcanism Conclusively Recognized
Even the first MESSENGER fly-by convinced most people that Mercury’s plains are volcanic. Evidence included numerous cases of impact craters flooded by lavas (“ghost craters”), sometimes so deeply that the only surface manifestation is a ring-like wrinkle ridge that traces the location the flooded crater rim (Figure 2; Head et al., 2008). The Caloris basin was now seen in full for the first time (Figure 3), and crater-counting showed that the exterior plains are younger than the interior plains, and both are younger than the rim materials. This does not allow the exterior plains to be ejecta deposits or the interior plains to be impact melt (Fassett et al., 2009; Head et al., 2009), leaving a volcanism as the only viable option. However, the fact that the interior and exterior plains are spectrally distinct (Head et al., 2009; Watters et al., 2009) raises issues about flow length and spatially heterogeneous magmagenesis that are not yet resolved (e.g., Rothery et al., 2017). The first fly-by also revealed a number of irregularly shaped “rimless” depressions on the Caloris floor that were recognized as explosive volcanic vents. That class of feature is discussed after first examining Mercury’s history of effusive volcanism.
Mercury’s Smooth Plains
Imaging from orbit further strengthened the evidence that Mercury’s smooth plains were made by effusive volcanism. However, it also revealed some mistaken interpretations of the fly-by imaging. For example, the ejecta of some craters of up to 100 km in size inside the Caloris basin had been thought to be embayed by the floor plains (Murchie et al., 2008; Watters et al., 2009). This would have been a significant finding because, as in the case of the “ghost craters” elsewhere on the planet, it would demonstrate an interval of time to allow these craters to have been formed on the basin floor before partial fill by plains volcanism. However, this particular interpretation was disproven by detail from orbit (Figure 4). For example, while agreeing that the Caloris plains are volcanic, Ernst et al. (2015) wrote “higher-resolution orbital images reveal that the craters previously classified as embayed actually postdate the flooding: each crater has a well-defined ejecta deposit and a secondary crater field superposed on the surrounding interior plains” (p. 416). Instead, what convinced people that the Caloris fill is volcanic was its evident thickness; the numerous wrinkle ridges (typical, though not diagnostic, of lava surfaces); and its crater-retention age which is younger than the exposed Caloris rim (Denevi et al., 2013; Fassett et al., 2009; Spudis & Guest, 1988), showing that the surface of the basin-fill must postdate the basin excavation by a barely resolvable time interval possibly as short as about 0.1 Ga.
Denevi et al. (2013) found that about 27 percent of the planet’s surface is occupied by smooth plains (Figure 5), on a count that deliberately excluded smooth fill-in craters of less than 100 km diameter on the grounds that this is more likely to be impact melt. Most examples of smooth plains occur as infill of impact basins, the most extensive exceptions being the circum-Caloris plains (otherwise known as the Caloris exterior plains) and the “northern plains” (Figure 6). Those were first recognized as smooth plains from orbit, and alone cover about 6 percent of the planet (Head et al., 2011; Ostrach et al., 2015).
The smooth plains are mostly Calorian in age. Denevi et al. (2013) used crater-counting to date all extensive smooth plains at ~3.7–3.9 Ga, while recognizing that some small patches were likely to be Mansurian, with ages as young as 1 Ga. Ostrach et al. (2015) cautioned that the age of the northern plains could be anything from 3.7 to 2.5 Ga, depending on which crater production function is assumed to apply. Byrne et al. (2016) were able to resolve slightly younger crater ages for nine areas of smooth plains smaller than those previously dated, from which they concluded that widespread plains volcanism had ceased by about 3.5 Ga. However, they recognized the apparently younger (~1 Ga, late Mansurian) ages of the plains in the 290-km Rachmaninoff basin (Figure 7; Prockter et al., 2010) and the 265-km Raditladi basin (Marchi et al., 2011), and the possibility of further young ages in areas smaller than 6 × 104 km2. Fegan et al. (2017) found ages as young as possibly 0.9 Ga in a survey of lava-flooded basins in the 150–400-km diameter range.
Figure 8 shows an example of the complexity of relationships that can be observed between basins on Mercury and smooth plains. The view is centered on a 470-km diameter basin named Aneirin, of probable Calorian age. The floor of Aneirin is covered by low-reflectance blue plains, whose thickness appears to increase towards the southeast, to judge from the flooded craters in the east of the basin and the continuity between the interior plains and the exterior plains beyond the flooded eastern rim of the basin. These plains and Aneirin’s rim have been superposed by several later craters, the largest of which is Dario, a 150-km crater that overlaps the western rim. Dario itself appears to have been flooded by lava to a depth sufficient to bury all trace of the central peak complex likely to occur in a crater of this size. The floor of Dario, and that of a shallow 30-km crater within it (which also looks to be flooded by lava), is cut by a lobate scarp that traces the location of the part of Aneirin’s western rim that was obliterated by the Dario-forming impact. This is an example of a compressional tectonic feature reaching the surface along the interface between basin-floor and basin-fill in a manner that appears widespread on Mercury (Fegan et al., 2017), and shows that plains emplacement here ceased before the thrust motion.
Based on crater-counting, Fegan et al. (2017) suggest a 2.2 Ga age for the smooth plains inside Aneirin (although Byrne et al., 2016, suggested 3.7 Ga), and a last resolvable scarp movement at 0.6 Ga. The floor of Dario is too small to yield a statistically meaningful cratering-age, because craters smaller than about 4 km are likely to include a substantial, and unevenly distributed, proportion of secondary craters (Strom et al., 2008). However, it must fall between those two events, whatever their correct ages. Dario and other sub-200 km craters with smooth plains confined to their floors are examples of plains volcanism postdating the 3.5 Ga end of “widespread effusive volcanism” documented by Byrne et al. (2016).
Occurrence of young plains volcanism largely within impact features has been interpreted as collocation with areas of crustal weakness (impact features), allowing eruption despite the overall global compressive tectonic regime that would be expected to inhibit magma ascent (Byrne et al., 2016). Another example is areas of less than about 104 km2 within plains that have visually low crater density. These are impossible to distinguish from flukes resulting from the clustered distribution of secondary craters, except when one edge is sharply bounded by a tectonically controlled topographic feature such in Figure 9. Here, there is a persuasive case for it being a young lava surface, possibly fed by eruptions taking advantage of the bounding fault as an ascent pathway (Malliband et al., 2018).
Smooth Plains Emplacement Mechanism
The extent of the majority of smooth plains on Mercury makes them fit into the category of “large igneous provinces” recognized originally on Earth but occurring also on the Moon, Mars and Venus (Bryan & Ernst, 2008; Ernst, 2014). The extent and ability of Mercury’s plains-forming volcanism to bury impact craters suggests emplacement of low-viscosity flood basalts (Denevi et al., 2013; Head et al., 2011). On Earth, flood basalts typically bury their own source fissures, none of which have been positively identified on Mercury. Byrne et al. (2013) noted various “coalesced depressions” in a study of part of Mercury’s northern plains but found no resolvable lava flows emanating from them, and so these are likely to be later explosive or collapse features. However, Byrne et al. (2013) did describe broad channels with streamlined islands consistent with erosive flow of voluminous high-temperature low-viscosity lavas (Figure 10).
Some narrow lava channels with similarities to lunar sinuous rilles have been suggested (Hurwitz et al., 2013), but they are at the limits of resolution and inconclusive as regards origin via thermal erosion or collapse of lava-tube roofs. The margins of individual flows and boundaries within flow units have not been identified. However, after presenting evidence for flow of Caloris exterior plains into the basin, weaker evidence of outward flow of interior plains, and of continuity of plains across a flooded gap in the western rim of Caloris, Rothery et al. (2017) suggested that flow lengths were more likely to be hundreds rather than thousands of km.
Smooth Plains Composition
There are many reasons to be cautious or even skeptical when using Mercury surface compositional data for petrologic modeling. Elemental abundances estimated from gamma and neutron spectroscopy sample about the top meter of the regolith whereas X-ray spectroscopy senses only the uppermost few micrometers, and optical spectroscopy from which mineralogic inferences are made senses approximately the uppermost millimeter (Rothery et al., 2010). Although the regolith is expected to be well-mixed, the mixing rate must be balanced against the rates of various kinds of space weathering, so we cannot be sure of the extent to which different measures are representative of the regolith. When we add to this the fact that the regolith is a mixture of material redistributed on a variety of scales by impact gardening (Langevin, 1997), and that the best spatial resolution of the MESSENGER X-ray Spectrometer (XRS) was of the order of 100 km and considerably worse for some elements (Nittler et al., 2011; Weider et al., 2015), then even the best measurement carries a high likelihood of being of a mixture of rock types rather than an effectively uncontaminated single rock type.
MESSENGER provided data adequate to show compositional heterogeneity on Mercury, including variations between different smooth plains. Denevi et al. (2009) and Ernst et al. (2010) used color imaging to show that smooth plains range from high-reflectance red plains (notably the Caloris interior plains and the northern smooth plains) to low-reflectance blue plains (such as the Caloris exterior plains and the plains in and around Aneirin). Data from the Mercury Atmospheric and Surface Composition Spectrometer (MASCS) confirmed the weakness of any 1-μm absorption attributable to the Fe-O bond, and showed geographic spectral variations that fitted with other data while not being diagnostic (Izenberg et al., 2014).
The XRS showed that the northern plains and the Caloris interior plains have a distinctly low Mg/Si ratio, but the Caloris interior plains have a uniformly high Al/Si ratio while the northern plains have a spatially heterogeneous low-to-medium Al/Si ratio (Weider et al., 2012, 2015). While they saw some compositional (element ratio) variations that do not appear to coincide with independently recognizable terrain boundaries, Weider et al. (2015) were able to distinguish four plains compositions: Caloris interior plains (fairly MORB-like on Mg/Si v Al/Si), the northern part of the northern smooth plains (lower in Al/Si), the southern part of the northern smooth plains (slightly higher than either in Mg/Si), and the smooth plains inside Rachmaninoff (even higher in Mg/Si). They concluded that the best terrestrial analogs for Mercury’s surface materials are low-Fe basaltic komatiites or basalts.
The available evidence does not enable the cause of magmagenesis to be identified, but is consistent with both mantle heterogeneity, and with time-variable and depth-variable partial melting events. Namur et al. (2016b) suggested that Mercury’s older volcanic crust (see Older Plains and the Burial of Primary Crust) required partial melting at a depth of about 360 km and a mantle potential temperature of 1650°C, whereas by the time of Northern Plains eruption the depth of melting had decreased to 160 km at a mantle potential temperature of 1410°C. The degree of mantle partial melting would also have decreased over time (Namur & Charlier, 2017). Vander Kaaden and McCubbin (2016) model the Northern Plains as an alkali-rich boninite produced by partial melting of an olivine-dominant, pyroxene- and plagioclase-bearing mantle at very low oxygen fugacity, and Namur and Charlier (2017) suggested that the local mineralogy would be Na-rich plagioclase, diopside, and forsterite. Vetere et al. (2017) measured low viscosities of about 10 Pa s in an experimental analog of Northern Plains lavas. They calculated that high effusion rates (>104 m3 s−1) would be necessary for their emplacement, but determined that spreading as an inflating flow-field beneath a flexible cooling crust would be insensitive to Mercury’s day-night temperature difference of over 600 K.
Older Plains and the Burial of Primary Crust
Mariner 10–era mappers distinguished two other spatially extensive plains units that are older than the smooth plains. These were “intercrater plains” (Figure 11) consisting of rolling, densely cratered surfaces between >30 km craters (Trask & Guest, 1975) and a less-cratered equivalent that became known as “intermediate plains” (Schaber & McCauley, 1980).
These units resemble older and hence more cratered versions of the smooth plains. With the MESSENGER confirmation that Mercury lacks a low-density anorthositic primary crust like the Moon (Nittler et al., 2011), and experiments and modeling by Vander Kaaden and McCubbin (2015) demonstrating that Mercury’s iron-poor magma ocean was not dense enough to allow an anorthosite primary crust to form by flotation, it became accepted that almost the whole of Mercury’s surface is secondary crust (in the sense defined by Taylor, 1989), consisting of multiple generations of flood lavas (Figure 12). These seem to have erupted as a succession of large igneous provinces supplied by mantle partial melting. Mercury’s primary crust, formed by flotation of buoyant differentiated mineral phases from the initial magma ocean, is likely to have been no more than a few hundred meters of graphite-rich material (Vander Kaaden & McCubbin, 2015) manifested today as “low-reflectance material” uplifted or excavated and ejected by large craters and found across about 15 percent of the surface (Denevi et al., 2009; Peplowski et al., 2016).
Whitten et al. (2014) used better MESSENGER images to reassess areas that had been mapped as intermediate plains on Mariner 10 images. They concurred that these are volcanic in origin, being more cratered, and hence older, equivalents of smooth plains. However, using N(20) and N(10) statistics (the number of craters ≥20 and ≥10 km diameter per 106 km2) they found such an overlap in ages, with both units ranging through the pre-Tolstojian and Tolstojian, that they recommended that intermediate plains should be abandoned as a mapping unit. Part of their argument rested on their recognition of previously unnoticed small patches of smooth plains (>750 km2, enough to occupy over a third of a 50-km diameter crater) material within intercrater plains, which had lowered the apparent N20) and N(10) values of the count areas. They recommended that such occurrences should be remapped as outliers of smooth plains surrounded by intercrater plains. This observation ties in with the more recently recognized small sizes of the youngest areas of smooth plains (e.g., Figure 9) and suggests that similarly small effusive volumes may have been produced at earlier times too.
The Whitten et al. (2014) recommendation would also require abandoning any attempt to distinguish genuinely Tolstojian plains from the (probably more extensive) pre-Tolstojian plains. Some mappers in the MESSENGER era continue to find intermediate plains to be a useful unit (Galluzzi et al., 2016; Guzzetta et al., 2017; Figure 13), distinguishable from the older intercrater plains at the 1:400k mapping scale used in the preparation of 1:3M quadrangle maps.
One of the greatest geological revelations by MESSENGER was the abundant evidence of explosive volcanism in the form of irregularly shaped pit-like craters. Some of them are considerably deeper than similarly sized impact craters. They were at first often described as “rimless” (e.g., Head et al., 2008; Murchie et al., 2008), but this is an unhelpful term because the pits usually have well-defined edges that many would call a “rim”; what they actually lack is a rampart marking the rim in the same manner as impact craters. Most pits are surrounded by bright spectral anomalies with diffuse outer edges and an increasingly strong “red” color towards their middles (Figure 14). The pits were quickly accepted as explosive volcanic vents and the “red spots” (awarded the IAU-approved descriptor term: “faculae” in 2018) as pyroclastic ejecta (Head et al., 2008, 2009; Kerber et al., 2009, 2011; Murchie et al., 2008). These bear no relationship to the pyroclastic deposits suggested by Robinson and Lucey (1997) on the basis of recalibrated Mariner 10 color images, which can be seen on MESSENGER images to be low-reflectance material excavated by impacts.
Global surveys (Goudge et al., 2014; Jozwiak et al., 2018; Thomas et al., 2014a) have revealed about a hundred candidate volcanic vents. These are widespread across the globe but very few occur within extensive smooth plains, with the notable exception of those close inside the Caloris rim (Figures 3 and 14) and there are no examples clearly within the northern volcanic plains. Most occur on the floors of impact craters (a category that includes those inside Caloris, because that is an impact basin), where they can usually be seen to postdate any smooth floor material.
The vents associated with Caloris are all inside the basin, and occur close to the edge (less than about 150 km from its rim). The plains fill is likely to be thinnest there, so maybe a really thick lava pile was a barrier to explosive eruptive ascent. A few are on smooth plains floors of rim reentrants (e.g., Figure 14 near 160° E), which led to an initial misinterpretation (Watters et al., 2009) that these are outside the basin.
On an airless body such as Mercury, explosively ejected material must be dispersed on ballistic trajectories, like the eruption plumes seen on Jupiter’s moon Io (e.g., McEwen et al., 1998; Strom & Schneider, 1982). A requirement for explosive eruptions to occur is the presence of volatiles, whose violent expansion as a gas phase is what fragments the ejecta and drives it upwards. Such a gas could come either from exsolution of volatiles in response to decreasing confining pressure within a volatile-rich magma as it rises, or from volatiles encountered by magma in the shallow subsurface environment. In either case there is a requirement for sufficiently abundant volatiles capable of becoming a gas. Prior to MESSENGER, Mercury was expected to be poor in volatiles because of its proximity to the Sun and the giant impact event assumed to have stripped away so much of Mercury’s original silicate inventory that it was left with a comparatively large core. However MESSENGER provided several independent lines of evidence that Mercury’s surface is rich in volatiles, including high K/Th ratio (from gamma-ray spectroscopy; Peplowski et al., 2011, 2012); sulfur (presumably as sulfides) ranging from 2 percent to 4 percent across the whole surface (from XRS; Nittler et al., 2011); and “hollows” where km-sized, 10 m-deep patches of surface are being lost to space, presumably via loss of a volatile constituent (Blewett et al., 2011; Thomas et al., 2014a). Crustal richness in sulfur can be explained by the highly reducing conditions postulated for Mercury’s formation, because sulfur solubility in silicate melts increases as oxygen fugacity decreases (Namur et al., 2016a; Zolotov et al., 2013), and a similar argument applies to potassium (McCubbin et al., 2012), but it is not clear that the same explanation can account for crustal richness in other volatiles. Studies using MASCS confirmed the redness of the pyroclastic deposits throughout the visible and near infrared, but contained no compositionally diagnostic information (Besse et al., 2015).
Neither the identity of the volatile that drives Mercury’s explosive eruptions nor its source is known, but it could involve either sulfur or carbon in the case of the largest red spot (variously known as NE Rachmaninoff or Rachmaninoff-Copeland, but now officially named Nathair Facula, 130-km radius—see Figure 15). This is the only example with a spatially resolved XRS measurement, and here sulfur is seen to be depleted (Weider et al., 2016). Carbon also appears to be locally low in abundance according to gamma-ray spectrometry (Peplowski et al., 2016; Weider et al., 2016). Weider et al. (2016) consider CO, COS, and CS2 as not entirely satisfactory candidates, but the loss of C and S in such gas species would account for the low C and S in the solid residue of the pyroclastic ejecta.
The radius of each red spot can be used to estimate the velocity at which particles were ejected, and from this a volatile abundance can be modeled (Kerber et al., 2009; Thomas et al., 2014b). This yields volatile abundances of thousands of p.p.m. (depending on species), something like double the abundance necessary to account for the extent of deposits from lunar pyroclastic eruptions (Kerber et al., 2009).
Study of detailed vent morphology is hindered by the limited numbers that were imaged at sufficiently high resolution. Rothery et al. (2014) used targeted high-resolution narrow angle camera images from orbit to study a prominent 27 × 13 km vent in the southwest of the Caloris basin (22.3° N, 146.2° E) that had been previously described, on the basis only of images from the first fly-by (Head et al., 2008), as a “kidney-shaped depression superimposed on a broad, smooth domelike feature.” Rothery et al. (2014) found that images with the Sun low in the west revealed an extension to the structure, destroying the illusion of a simple kidney shape (Figure 16), and used laser altimeter profiles to show that the vent does not lie at the summit of a topographic rise of any significance (the “broad dome” having been inferred largely on the misleading basis of an albedo feature). In this case the floor of the vent is over 1 km below its rim, whereas the rim is only about 100 m above the general level of the terrain and the flanks slope at only 0.2°. In a subsequent study of several other examples, Thomas et al. (2014b) found outer flank slopes usually of only about 3°, leveling off considerably more proximally to the vent than the visible edge of the red spot deposit. Vents therefore appear to be essentially explosive, excavational features, with little sign of a surrounding constructional edifice.
By employing both low-Sun and higher-Sun images, Rothery et al. (2014) were able to identify nine individual vents as separate depressions surrounded by the overall rim of the “kidney-shaped vent.” They likened this to a compound volcano on Earth, where the locus of magmatic activity has migrated to and fro over time, and suggested that the most recent activity inside this particular compound vent was at the smaller and more central depressions, on whose floor the shadow-free higher-Sun images revealed fine scale texture that is absent from the older vent floors (those that are partially cross-cut by the younger vents). There, the original texture has likely become muted by proximal fallout from later eruptions and general regolith-forming processes.
Many of the other pits on Mercury can also be described as compound vents. There are various others in the Caloris basin (Rothery et al., 2014); the NE Rachmaninoff vent (Figure 15) is a coalescence of two main vents (Thomas et al., 2014b) that may each contain subsidiary vents on its floor; the “north Rachmaninoff vent” (35.9° N, 57.3° E, Figure 15) has the form of an equidimensional cluster of vents enclosed by a common rim; and there are numerous examples of vents excavating a moat around the site of a crater’s central peak (Figure 17; Jozwiak et al., 2018; Thomas et al., 2015a). In most examples the resolution of the images is inadequate to resolve internal textural differences of the kind apparent in Figure 16, but it seems more reasonable to assume a migration of the locus of activity over time, rather than an array of vents or a complete ring-moat erupting simultaneously.
Explosive volcanism on Mercury seems to have outlasted plains-forming effusive volcanism. The known vents all punch through smooth (or intermediate) plains, with their “red spot” deposit overlying the plains unit. There is no reported example of a lava-flooded vent or an incomplete “red spot” peeking out beyond the edge of a plains lava flow that has buried the other part (though it is also the case that some red spots have no identified vent). Rothery et al. (2014) pointed out that the fine-scale texture within parts of the Caloris vent in Figure 16 is consistent with an age billions of years younger than smoother parts of the vent floor, though there are other possible explanations. Thomas et al. (2014c) obtained Calorian to early Mansurian crater-retention ages for the pyroclastic deposits surrounding NE and N Rachmaninoff and the Caloris vent in Figure 16, but found late Mansurian or even early Kuiperian ages for some sub-30 km craters that host vents, providing a maximum possible age for the vents themselves.
Thomas et al. (2014b) suggested that the common association of explosive vents with craters and tectonic features is an indication that crater-related fractures and some tectonic faults were able to act as magma or volatile pathways or both despite Mercury’s general compressive tectonic regime. They went on to argue (2015b) that the lack of floor fracturing in craters hosting explosive vents on Mercury demonstrates deeper magma storage than on the Moon, and that this was a response to global compressive tectonics, permitting exsolution of volatiles and explosive eruption while inhibiting effusive volcanism.
Mercury has turned out to be a fascinating and perplexing planet, and although there surely remains much to discover among the trove of MESSENGER data, it is fortunate that Mercury will soon be studied by an even more capable mission. This is BepiColombo, due for launch late in 2018 and likely to commence science operations in Mercury orbit in 2026. It will consist of two orbiters, one European and one Japanese, which between them will study the planet and its environment with a more sophisticated and wider suite of instruments than MESSENGER (Benkhoff et al., 2010; Milillo et al., 2010; Rothery et al., 2010). Europe’s BepiColombo Mercury Planetary Orbiter will have a less eccentric orbit than MESSENGER, with a periherm lying closer to the equator, which should enable it to obtain altimetric and compositional data equally well over the north and south hemispheres. Among its payload is a thermal infrared imaging system (MERTIS) that will provide new mineralogical information by spectroscopy in a region of the spectrum not previously imaged at Mercury (Hiesinger et al., 2010).
More geographically complete compositional measurements at higher spatial resolution and greater sensitivity will deliver opportunities to improve our petrologic modelling of the plains-forming volcanism. Improved and globally complete geological mapping, beginning with MESSENGER images and other data (Galluzzi et al., 2016; Guzzetta et al., 2017; Mancinelli et al., 2016) and later incorporating BepiColombo images and other data will put observations into a firmer context and clarify the spatiotemporal history of plains volcanism.
Outstanding questions include whether there are any exposed tracts of intact primary crust, and if so what their composition is. During MESSENGER’s end-of-mission low-altitude campaign, its neutron spectrometer was able to demonstrate an increase in thermal neutron count rate when passing over low-reflectance material that is consistent with this being the remains of graphite-bearing primary crust (Peplowski et al., 2016). It is unlikely that BepiColombo will be able to reproduce those data because it is not planned for the spacecraft to ever be that close to the surface, but on the other hand it may be able to detect and map carbon with its more capable X-ray spectrometer, MIXS (Fraser et al., 2010). It is also important to pin down whether Mercury is enriched in volatiles in general, or only in those whose concentration into the silicate fraction of the planet can be explained as a consequence of extreme reducing conditions during differentiation. It is uncertain which volatile species was/were responsible for explosive volcanism. MESSENGER X-ray spectroscopy was able to spatially resolve only the largest “red spot,” but BepiColombo’s MIXS instrument should be able to measure sulfur and possibly carbon in and around several smaller red spots.
Further work on the ages, eruption mechanisms, and eruption history of pyroclastic vents is needed. For example, did any explosive volcanism occur simultaneously with, or even precede, effusive volcanism, or did it always come afterwards, as some kind of last gasp? Targeted high resolution and stereoscopic imaging by BepiColombo’s camera system, SIMBIO-SYS (Flamini et al., 2010), will surely help in this respect.
Identifying and dating the youngest examples of both explosive and effusive volcanism are important goals, but absolute dating may prove intractable because of the small sizes of likely examples, the unreliability of cratering statistics over such small areas, and the thinness of pyroclastic deposits that makes it hard to be sure whether an impact crater penetrates or is mantled by such a deposit.
Many questions about Mercury’s formation, differentiation and magmagenesis would be best answered by geochemical and isotopic analysis of samples. Sadly, neither in situ analysis nor sample return missions are likely for well over a decade, and in fact our first chance to study a sample may come from finding and recognizing a meteorite from Mercury, within which a reasonable fraction of crater-ejecta can find itself in an Earth-crossing orbit (Gladman & Coffey, 2009).
- Balogh, A., Ksanfomality, L., & Steiger, R. (Eds.). (2008). Mercury. Space Sciences Series of ISSI. Dordrecht, The Netherlands: Springer.
- Clark, P. E. (2007). Dynamic planet: Mercury in the context of its environment. New York: Springer Science & Business Media.
- Hartmann, W. K. (Ed.). (1975). The planet Mercury [Special section]. Journal of Geophysical Research, 80, 2341–2344.
- Kerr, R. (2011). Mercury looking less exotic, more a member of the family. Science, 333(6051), 1847–1868.
- Rothery, D. A. (2016). Planet Mercury. New York: Springer.
- Solomon, S., Anderson, B., & Nittler, L. (Eds). (Forthcoming). Mercury: The view after MESSENGER. New York: Cambridge University Press.
- Solomon, S. C., Prockter, L. M., & Blewett, D. T. (Eds.). (2009). MESSENGER at Mercury: An introduction to the Special Issue. Earth and Planetary Science Letters, 285(3–4). Introduction to a special issue of Planetary and Space Science, 59(15), 1827–1828.
- Strom, R. G., & Sprague, A. L. (2003). Exploring Mercury: The iron planet. New York: Springer Science & Business Media.
- Vilas, F., Chapman, C. R., & Matthews, M. S. (Eds.). (1988). Mercury: Papers presented at the Mercury Conference held 6–9 Aug. 1986 in Tucson. Tucson: University of Arizona Press.
- Anderson, B. J., Johnson, C. L., Korth, H., Purucker, M. E., Winslow, R. M., Slavin, J. A., . . . Zurbuchen, T. H. (2011). The global magnetic field of Mercury from MESSENGER orbital observations. Science, 333(6051), 1859–1862.
- Banks, M. E., Xiao, Z., Braden, S. E., Barlow, N. G., Chapman, C. R., Fassett, C. I., & Marchi, S. S. (2017). Revised constraints on absolute age limits for Mercury’s Kuiperian and Mansurian stratigraphic systems. Journal of Geophysical Research: Planets, 122(5), 1010–1020.
- Barnes, J. J., Anand, M., & Franchi, I. A. (2016). Investigating the history of magmatic volatiles in the moon using NanoSIMS. Microscopy and Microanalysis, 22(S3), 1804–1805
- Benkhoff, J., van Casteren, J., Hayakawa, H., Fujimoto, M., Laakso, H., Novara, M., . . . Ziethe, R. (2010). BepiColombo—Comprehensive exploration of Mercury: Mission overview and science goals. Planetary and Space Science, 58(1), 2–20.
- Besse, S., Doressoundiram, A., & Benkhoff, J. (2015). Spectroscopic properties of explosive volcanism within the Caloris basin with MESSENGER observations. Journal of Geophysical Research: Planets, 120(12), 2102–2117.
- Binder, A. B., & Gunga, H. C. (1985). Young thrust-fault scarps in the highlands: Evidence for an initially totally molten Moon. Icarus, 63(3), 421–441.
- Blewett, D. T., Chabot, N. L., Denevi, B. W., Ernst, C. M., Head, J. W., Izenberg, N. R., . . . Xiao, Z. (2011). Hollows on Mercury: MESSENGER evidence for geologically recent volatile-related activity. Science, 333(6051), 1856–1859.
- Bryan, S. E., & Ernst, R. E. (2008). Revised definition of large igneous provinces (LIPs). Earth-Science Reviews, 86(1), 175–202.
- Byrne, P. K., Klimczak, C., Williams, D. A., Hurwitz, D. M., Solomon, S. C., Head, J. W., . . . Oberst, J. (2013). An assemblage of lava flow features on Mercury. Journal of Geophysical Research: Planets, 118(6), 1303–1322.
- Byrne, P. K., Ostrach, L. R., Fassett, C. I., Chapman, C. R., Denevi, B. W., Evans, A. J., . . . Solomon, S. C. (2016). Widespread effusive volcanism on Mercury likely ended by about 3.5 Ga. Geophysical Research Letters, 43(14), 7408–7416.
- Denevi, B. W., Ernst, C. M., Meyer, H. M., Robinson, M. S., Murchie, S. L., Whitten, J. L., . . . Chapman, C. R. (2013). The distribution and origin of smooth plains on Mercury. Journal of Geophysical Research: Planets, 118(5), 891–907.
- Denevi, B. W., Robinson, M. S., Solomon, S. C., Murchie, S. L., Blewett, D. T., Domingue, D. L., . . . Chabot, N. L. (2009). The evolution of Mercury’s crust: A global perspective from MESSENGER. Science, 324(5927), 613–618.
- Dzurisin, D. (1978). The tectonic and volcanic history of Mercury as inferred from studies of scarps, ridges, troughs, and other lineaments. Journal of Geophysical Research: Solid Earth, 83(B10), 4883–4906.
- Ernst, R. E. (2014). Large igneous provinces. Cambridge, UK: Cambridge University Press.
- Ernst, C. M., Denevi, B. W., Barnouin, O. S., Klimczak, C., Chabot, N. L., Head, J. W., . . . Solomon, S. C. (2015). Stratigraphy of the Caloris basin, Mercury: Implications for volcanic history and basin impact melt. Icarus, 250, 413–429.
- Ernst, C. M., Murchie, S. L., Barnouin, O. S., Robinson, M. S., Denevi, B. W., Blewett, D. T., . . . Roberts, J. H. (2010). Exposure of spectrally distinct material by impact craters on Mercury: Implications for global stratigraphy. Icarus, 209(1), 210–223.
- Evans, L. G., Peplowski, P. N., Rhodes, E. A., Lawrence, D. J., McCoy, T. J., Nittler, L. R., . . . Weider, S. Z. (2012). Major‐element abundances on the surface of Mercury: Results from the MESSENGER Gamma‐Ray Spectrometer. Journal of Geophysical Research: Planets, 117(E12).
- Fassett, C. I., Head, J. W., Blewett, D. T., Chapman, C. R., Dickson, J. L., Murchie, S. L., . . . Watters, T. R. (2009). Caloris impact basin: Exterior geomorphology, stratigraphy, morphometry, radial sculpture, and smooth plains deposits. Earth and Planetary Science Letters, 285(3), 297–308.
- Fegan, E. R., Rothery, D. A., Marchi, S., Massironi, M., Conway, S. J., & Anand, M. (2017). Late movement of basin-edge lobate scarps on Mercury. Icarus, 288, 226–234.
- Flamini, E., Capaccioni, F., Colangeli, L., Cremonese, G., Doressoundiram, A., Josset, J. L., . . . Marinangeli, L. (2010). SIMBIO-SYS: The spectrometer and imagers integrated observatory system for the BepiColombo planetary orbiter. Planetary and Space Science, 58(1), 125–143.
- Fraser, G. W., Carpenter, J. D., Rothery, D. A., Pearson, J. F., Martindale, A., Huovelin, J., . . . Benkoff, J. (2010). The mercury imaging X-ray spectrometer (MIXS) on bepicolombo. Planetary and Space Science, 58(1), 79–95.
- Galluzi, V., Carli, C., Zambon, F., Giacomini, L., Guzzetta, L., Ferranti, L., & Palumbo, P. (2017). The Intermediate Plains of Mercury: Considerations on a debated unit. In European Planetary Science Congress 2017, held 17–22 September, 2017 in Riga, Latvia.
- Galluzzi, V., Guzzetta, L., Ferranti, L., Di Achille, G., Rothery, D. A., & Palumbo, P. (2016). Geology of the Victoria quadrangle (H02), Mercury. Journal of Maps, 12(Supp1.), 227–238.
- Gladman, B., & Coffey, J. (2009). Mercurian impact ejecta: Meteorites and mantle. Meteoritics & Planetary Science, 44(2), 285–291.
- Goudge, T. A., Head, J. W., Kerber, L., Blewett, D. T., Denevi, B. W., Domingue, D. L., . . . Izenberg, N. R. (2014). Global inventory and characterization of pyroclastic deposits on Mercury: New insights into pyroclastic activity from MESSENGER orbital data. Journal of Geophysical Research: Planets, 119(3), 635–658.
- Guzzetta, L., Galluzzi, V., Ferranti, L., & Palumbo, P. (2017). Geology of the Shakespeare quadrangle (H03), Mercury. Journal of Maps, 13(2), 227–238.
- Head, J. W., Chapman, C. R., Strom, R. G., Fassett, C. I., Denevi, B. W., Blewett, D. T., . . . Prockter, L. M. (2011). Flood volcanism in the northern high latitudes of Mercury revealed by MESSENGER. Science, 333(6051), 1853–1856.
- Head, J. W., Murchie, S. L., Prockter, L. M., Robinson, M. S., Solomon, S. C., Strom, R. G., . . . Gillis-Davis, J. J. (2008). Volcanism on Mercury: Evidence from the first MESSENGER fly-by. Science, 321(5885), 69–72.
- Head, J. W., Murchie, S. L., Prockter, L. M., Solomon, S. C., Chapman, C. R., Strom, R. G., . . . Dickson, J. L. (2009). Volcanism on Mercury: Evidence from the first MESSENGER fly-by for extrusive and explosive activity and the volcanic origin of plains. Earth and Planetary Science Letters, 285(3), 227–242.
- Hiesinger, H., Helbert, J., & Team, M. C. I. (2010). The Mercury radiometer and thermal infrared spectrometer (MERTIS) for the BepiColombo mission. Planetary and Space Science, 58(1–2), 144–165.
- Hurwitz, D. M., Head, J. W., Byrne, P. K., Xiao, Z., Solomon, S. C., Zuber, M. T., . . . Neumann, G. A. (2013). Investigating the origin of candidate lava channels on Mercury with MESSENGER data: Theory and observations. Journal of Geophysical Research: Planets, 118(3), 471–486.
- Izenberg, N. R., Klima, R. L., Murchie, S. L., Blewett, D. T., Holsclaw, G. M., McClintock, W. E., . . . Helbert, J. (2014). The low-iron, reduced surface of Mercury as seen in spectral reflectance by MESSENGER. Icarus, 228, 364–374.
- Jozwiak, L. M., Head, J. W., & Wilson, L. (2018). Explosive volcanism on Mercury: Analysis of vent and deposit morphology and modes of eruption. Icarus, 302(1), 191–212.
- Kreslavsky, M. A., Head, J. W., Neumann, G. A., Zuber, M. T., & Smith, D. E. (2014). Kilometer‐scale topographic roughness of Mercury: Correlation with geologic features and units. Geophysical Research Letters, 41(23), 8245–8251.
- Jeanloz, R., Mitchell, D. L., Sprague, A. L., & de Pater, I. (1995). Evidence for a basalt-free surface on Mercury and implications for internal heat. Science, 268(5216), 1455–1457.
- Langevin, Y. (1997). The regolith of Mercury: Present knowledge and implications for the Mercury Orbiter mission. Planetary and Space Science, 45(1), 31–37.
- Kerber, L., Head, J. W., Blewett, D. T., Solomon, S. C., Wilson, L., Murchie, S. L., . . . Domingue, D. L. (2011). The global distribution of pyroclastic deposits on Mercury: The view from MESSENGER fly-bys 1–3. Planetary and Space Science, 59(15), 1895–1909.
- Kerber, L., Head, J. W., Solomon, S. C., Murchie, S. L., Blewett, D. T., & Wilson, L. (2009). Explosive volcanic eruptions on Mercury: Eruption conditions, magma volatile content, and implications for interior volatile abundances. Earth and Planetary Science Letters, 285(3), 263–271.
- Malliband, C., Rothery, D. A., Balme, M. R., & Conway, S. J. (2018). Smooth patches—a new small scale magmatic process on Mercury. Mercury: Current and Future Science 2018. Lunar & Planetary Science Institute Contrib. No. 2047.
- Mancinelli, P., Minelli, F., Pauselli, C., & Federico, C. (2016). Geology of the Raditladi quadrangle, Mercury (H04). Journal of Maps, 12(Supp1.), 190–202.
- Marchi, S., Massironi, M., Cremonese, G., Martellato, E., Giacomini, L., & Prockter, L. (2011). The effects of the target material properties and layering on the crater chronology: The case of Raditladi and Rachmaninoff basins on Mercury. Planetary and Space Science, 59(15), 1968–1980.
- McCubbin, F. M., Riner, M. A., Vander Kaaden, K. E., & Burkemper, L. K. (2012). Is Mercury a volatile‐rich planet? Geophysical Research Letters, 39(9).
- McEwen, A. S., Keszthelyi, L., Geissler, P., Simonelli, D. P., Carr, M. H., Johnson, T. V., . . . Magee, K. P. (1998). Active volcanism on Io as seen by Galileo SSI. Icarus, 135(1), 181–219.
- Melosh, H. J., & McKinnon, W. B. (1988). The tectonics of Mercury. Mercury, 374–400.
- Milillo, A., Fujimoto, M., Kallio, E., Kameda, S., Leblanc, F., Narita, Y., . . . McKenna-Lawlor, S. (2010). The BepiColombo mission: an outstanding tool for investigating the Hermean environment. Planetary and Space Science, 58(1), 40–60.
- Muehlberger, W. R., Batson, R. M., Boudette, E. L., Duke, C. M., Eggleton, R. E., Elston, D. P., . . . Hall, T. A. (1972). Preliminary geologic investigation of the Apollo 16 landing site.
- Murchie, S. L., Watters, T. R., Robinson, M. S., Head, J. W., Strom, R. G., Chapman, C. R., . . . Blewett, D. T. (2008). Geology of the Caloris basin, Mercury: A view from MESSENGER. Science, 321(5885), 73–76.
- Murray, B. C. (1975). The Mariner 10 pictures of Mercury: An overview. Journal of Geophysical Research, 80(17), 2342–2344.
- Murray, B. C., Strom, R. G., Trask, N. J., & Gault, D. E. (1975). Surface history of Mercury: Implications for terrestrial planets. Journal of Geophysical Research, 80(17), 2508–2514.
- Namur, O., & Charlier, B. (2017). Silicate mineralogy at the surface of Mercury. Nature Geoscience, 10(1), 9–13.
- Namur, O., Charlier, B., Holtz, F., Cartier, C., & McCammon, C. (2016a). Sulfur solubility in reduced mafic silicate melts: Implications for the speciation and distribution of sulfur on Mercury. Earth and Planetary Science Letters, 448, 102–114.
- Namur, O., Collinet, M., Charlier, B., Grove, T. L., Holtz, F., & McCammon, C. (2016b). Melting processes and mantle sources of lavas on Mercury. Earth and Planetary Science Letters, 439, 117–128.
- Ness, N. F., Behannon, K. W., Lepping, R. P., & Whang, Y. C. (1975). The magnetic field of Mercury, 1. Journal of Geophysical Research, 80(19), 2708–2716.
- Nittler, L. R., Starr, R. D., Weider, S. Z., McCoy, T. J., Boynton, W. V., Ebel, D. S., . . . Lawrence, D. J. (2011). The major-element composition of Mercury’s surface from MESSENGER X-ray spectrometry. Science, 333(6051), 1847–1850.
- Ostrach, L. R., Robinson, M. S., Whitten, J. L., Fassett, C. I., Strom, R. G., Head, J. W., & Solomon, S. C. (2015). Extent, age, and resurfacing history of the northern smooth plains on Mercury from MESSENGER observations. Icarus, 250, 602–622.
- Nittler, L. R., Starr, R. D., Weider, S. Z., McCoy, T. J., Boynton, W. V., Ebel, D. S., . . . Lawrence, D. J. (2011). The major-element composition of Mercury’s surface from MESSENGER X-ray spectrometry. Science, 333(6051), 1847–1850.
- Noble, S. K., & Pieters, C. M. (2003). Space weathering on Mercury: Implications for remote sensing. Solar System Research, 37(1), 31–35.
- Peale, S. J., & Gold, T. (1965). Rotation of the planet Mercury. Nature, 206(4990), 1240–1241.
- Peplowski, P. N., Evans, L. G., Hauck, S. A., McCoy, T. J., Boynton, W. V., Gillis-Davis, J. J., . . . McNutt, R. L. (2011). Radioactive elements on Mercury’s surface from MESSENGER: Implications for the planet’s formation and evolution. Science, 333(6051), 1850–1852.
- Peplowski, P. N., Lawrence, D. J., Rhodes, E. A., Sprague, A. L., McCoy, T. J., Denevi, B. W., . . . Stockstill‐Cahill, K. R. (2012). Variations in the abundances of potassium and thorium on the surface of Mercury: Results from the MESSENGER Gamma‐Ray Spectrometer. Journal of Geophysical Research: Planets, 117(E12).
- Peplowski, P. N., Klima, R. L., Lawrence, D. J., Ernst, C. M., Denevi, B. W., Frank, E. A., . . . Solomon, S. C. (2016). Remote sensing evidence for an ancient carbon-bearing crust on Mercury. Nature Geoscience, 9(4), 273–276.
- Pettengill, G. H., & Dyce, R. B. (1965). A Radar Determination of the Rotation of the Planet Mercury. Nature, 206(4990), 1240.
- Prockter, L. M., Ernst, C. M., Denevi, B. W., Chapman, C. R., Head, J. W., Fassett, C. I., . . . Cremonese, G. (2010). Evidence for young volcanism on Mercury from the third MESSENGER fly-by. Science, 329(5992), 668–671.
- Rava, B., & Hapke, B. (1987). An analysis of the Mariner 10 color ratio map of Mercury. Icarus, 71(3), 397–429.
- Robinson, M. S., & Lucey, P. G. (1997). Recalibrated Mariner 10 color mosaics: Implications for mercurian volcanism. Science, 275(5297), 197–200.
- Rothery, D. A., Mancinelli, P., Guzzetta, L., & Wright, J. (2017). Mercury’s Caloris basin: Continuity between the interior and exterior plains. Journal of Geophysical Research: Planets, 122(3), 560–576.
- Rothery, D., Marinangeli, L., Anand, M., Carpenter, J., Christensen, U., Crawford, I. A., . . . Fraser, G. (2010). Mercury’s surface and composition to be studied by BepiColombo. Planetary and Space Science, 58(1), 21–39.
- Rothery, D. A., Thomas, R. J., & Kerber, L. (2014). Prolonged eruptive history of a compound volcano on Mercury: Volcanic and tectonic implications. Earth and Planetary Science Letters, 385, 59–67.
- Schaber, G. G., & McCauley, J. F. (1980). Geologic map of the Tolstoj (H-8) quadrangle of Mercury. United States Geological Survey, Miscellaneous Investigations Series, Map I-1199.
- Siegfried, R. W., & Solomon, S. C. (1974). Mercury: Internal structure and thermal evolution. Icarus, 23(2), 192–205.
- Shevchenko, V. V. (2002). The structure of the surface of Mercury’s regolith from remote sensing data. Solar System Research, 36(5), 359–366.
- Shevchenko, V. V. (2004). Remote estimation of the structure of the surface layer of Mercury. Advances in Space Research, 33(12), 2147–2151.
- Shoemaker, E. M., & Hackman, R. J. (1962). Stratigraphic basis for a lunar time scale. In The Moon (Vol. 14, pp. 289–300). New York: Academic Press.
- Solomon, S. C., McNutt, R. L., Gold, R. E., & Domingue, D. L. (2007). MESSENGER mission overview. Space Science Reviews, 131(1–4), 3–39.
- Sprague, A., Warell, J., Cremonese, G., Langevin, Y., Helbert, J., Wurz, P., . . . Milillo, A. (2008). Mercury’s surface composition and character as measured by ground-based observations. In A. Balogh, L. Ksanfomality, & R. von Steiger (Eds.), Mercury (pp. 217–249). New York: Springer.
- Spudis, P. D. (1985). A Mercurian chronostratigraphic classification. Reports of the Planetary Geology and Geophysics Program, NASA Technical Memorandum, 87563, 595–597.
- Spudis, P. D., & Guest, J. E. (1988). Stratigraphy and geologic history of Mercury. Mercury, 118–164.
- Strom, R. G., Chapman, C. R., Merline, W. J., Solomon, S. C., & Head, J. W. (2008). Mercury Cratering Record Viewed from MESSENGER’s First Fly-by. Science, 321(5885), 79–81.
- Strom, R. G., & Schneider, N. M. (1982). Volcanic eruption plumes on Io. In Satellites of Jupiter (pp. 598–633).
- Strom, R. G., Trask, N. J., & Guest, J. E. (1975). Tectonism and volcanism on Mercury. Journal of Geophysical Research, 80(17), 2478–2507.
- Taylor, S. R. (1989). Growth of planetary crusts. Tectonophysics, 161(3–4), 147–156.
- Thomas, R. J., Lucchetti, A., Cremonese, G., Rothery, D. A., Massironi, M., Re, C., . . . Anand, M. (2015a). A cone on Mercury: Analysis of a residual central peak encircled by an explosive volcanic vent. Planetary and Space Science, 108, 108–116.
- Thomas, R. J., Rothery, D. A., Conway, S. J., & Anand, M. (2014a). Hollows on Mercury: Materials and mechanisms involved in their formation. Icarus, 229, 221–235.
- Thomas, R. J., Rothery, D. A., Conway, S. J., & Anand, M. (2014b). Mechanisms of explosive volcanism on Mercury: Implications from its global distribution and morphology. Journal of Geophysical Research: Planets, 119(10), 2239–2254.
- Thomas, R. J., Rothery, D. A., Conway, S. J., & Anand, M. (2014c). Long-lived explosive volcanism on Mercury. Geophysical Research Letters, 41(17), 6084–6092.
- Thomas, R. J., Rothery, D. A., Conway, S. J., & Anand, M. (2015b). Explosive volcanism in complex impact craters on Mercury and the Moon: Influence of tectonic regime on depth of magmatic intrusion. Earth and Planetary Science Letters, 431, 164–172.
- Trask, N. J., & Guest, J. E. (1975). Preliminary geologic terrain map of Mercury. Journal of Geophysical Research, 80(17), 2461–2477.
- Vander Kaaden, K. E., & McCubbin, F. M. (2015). Exotic crust formation on Mercury: Consequences of a shallow, FeO‐poor mantle. Journal of Geophysical Research: Planets, 120(2), 195–209.
- Vander Kaaden, K. E., & McCubbin, F. M. (2016). The origin of boninites on Mercury: An experimental study of the northern volcanic plains lavas. Geochimica Et Cosmochimica Acta, 173, 246–263.
- Vetere, F., Rossi, S., Namur, O., Morgavi, D., Misiti, V., Mancinelli, P., . . . Perugini, D. (2017). Experimental constraints on the rheology, eruption, and emplacement dynamics of analog lavas comparable to Mercury’s northern volcanic plains. Journal of Geophysical Research: Planets, 122(7), 1522–1538.
- Warell, J., & Blewett, D. T. (2004). Properties of the Hermean regolith: V. New optical reflectance spectra, comparison with lunar anorthosites, and mineralogical modelling. Icarus, 168(2), 257–276.
- Warell, J., Sprague, A. L., Emery, J. P., Kozlowski, R. W. H., & Long, A. (2006). The 0.7–5.3 μm IR spectra of Mercury and the Moon: Evidence for high-Ca clinopyroxene on Mercury. Icarus, 180(2), 281–291.
- Watters, T. R., Murchie, S. L., Robinson, M. S., Solomon, S. C., Denevi, B. W., André, S. L., & Head, J. W. (2009). Emplacement and tectonic deformation of smooth plains in the Caloris basin, Mercury. Earth and Planetary Science Letters, 285(3), 309–319.
- Watters, T. R., Robinson, M. S., Banks, M. E., Tran, T., & Denevi, B. W. (2012). Recent extensional tectonics on the Moon revealed by the Lunar Reconnaissance Orbiter Camera. Nature Geoscience, 5(3), 181.
- Weber, R. C., Lin, P. Y., Garnero, E. J., Williams, Q., & Lognonne, P. (2011). Seismic detection of the lunar core. Science, 331(6015), 309–312.
- Weider, S. Z., Nittler, L. R., Murchie, S. L., Peplowski, P. N., McCoy, T. J., Kerber, L., . . . Izenberg, N. R. (2016). Evidence from MESSENGER for sulfur‐and carbon‐driven explosive volcanism on Mercury. Geophysical Research Letters, 43(8), 3653–3661.
- Weider, S. Z., Nittler, L. R., Starr, R. D., Crapster-Pregont, E. J., Peplowski, P. N., Denevi, B. W., . . . Solomon, S. C. (2015). Evidence for geochemical terranes on Mercury: Global mapping of major elements with MESSENGER’s X-Ray Spectrometer. Earth and Planetary Science Letters, 416, 109–120.
- Weider, S. Z., Nittler, L. R., Starr, R. D., McCoy, T. J., Stockstill‐Cahill, K. R., Byrne, P. K., . . . Solomon, S. C. (2012). Chemical heterogeneity on Mercury’s surface revealed by the MESSENGER X‐Ray Spectrometer. Journal of Geophysical Research: Planets, 117(E12).
- Whitten, J. L., Head, J. W., Denevi, B. W., & Solomon, S. C. (2014). Intercrater plains on Mercury: Insights into unit definition, characterization, and origin from MESSENGER datasets. Icarus, 241, 97–113.
- Wilhelms, D. E. (1970). Summary of lunar stratigraphy-telescopic observations. Professional Paper 599-F. United States Geological Survey. Washington, DC: USG Printing Office.
- Wilhelms, D. E. (1976). Mercurian volcanism questioned. Icarus, 28(4), 551–558.
- Wilhelms, D. E., & McCauley, J. F. (1971). Geologic map of the near side of the Moon. IMAP 703. Washington, DC: US Geological Survey.
- Zolotov, M. Y., Sprague, A. L., Hauck, S. A., Nittler, L. R., Solomon, S. C., & Weider, S. Z. (2013). The redox state, FeO content, and origin of sulfur‐rich magmas on Mercury. Journal of Geophysical Research: Planets, 118(1), 138–146.