All volcanoes are the same. You start with liquid rock some distance below the surface. It tries to rise because molten rock is less dense than the solid rock that surrounds it. Once it reaches the surface it is called a volcano. There are many variations, of course. The liquid may pour out and form lava ponds or flows, or it may cause a big explosion. The composition of the rock that melted makes a difference. The temperature is also not always the same. Some lavas form beautiful stratovolcanoes, other form messy cones; some flows are smooth, others are so rough and sharp they tear your shoes to pieces. But those are details. Earth’s 1500 volcanoes may look diverse, but under the skin they are all the same. Just like us, really.
But that is on Earth. How about space volcanoes? Do you get little green volcanoes on Mars, and multi-eyed monster volcanoes on planet Zorg? How much of the uniformity of Earth’s volcanoes is caused by Earth? Do we need Spock to explain to us ‘It is volcanic life – but not as we know it, Jim’?
In our tour of the Solar System, we have previously looked at Mars, Venus and Pluto, and found volcanoes very different from ours. Let’s do another one, to add to the volcanic diversity: our least known planet, mysterious Mercury.
Mercury: the fleeting messenger
It is not the easiest planet to observe, being so close to the Sun. After sunset, it will never be far above the horizon, as it chases the Sun down. It is brighter than you may expect, but few people in our modern world consciously see it, even if they could get off their phones long enough to recognize a planet. (Last month, on my travels, I unexpectedly saw the space station come over one evening. I pointed it out to a student. He had never heard of it.)
Mercury is the messenger, traveling faster on the sky than any other planet. It is a small planet, our smallest since the demotion of Pluto, but heavier than one might expect. From the excess weight it is clear that there is more iron than rock, the only one of the terrestrial planets to have this. Images showed a densely cratered surface. Our current understanding of Mercury’s innards are depicted in the figure, as published by David Smith and colleagues (Science, 2012). Scaled to the Earth, Mercury’s iron core is huge, and its mantle is tiny. Like Earth, the core is cooling and the inner part has already solidified. As it cools, solid iron sulphide floats upward and it has formed a solid layer on top of the liquid outer core. The thin, dense mantle rests on this.
The model is not undisputed, as it is not clear why Mercury should have so much sulphur. There was very little sulphur in the inner Solar System at the time the planets formed, and in fact Earth doesn’t have that much. But the large amount of iron compared to rock is well established. You would expect Mercurian volcanoes to be ultra-mafic: iron makes this world go round. You’d be wrong, though.
The surface is airless and cratered, inhospitable, baked by the menacing Sun. And the Sun takes forever to set. Uniquely among the planets, Mercury’s day lasts longer than its year. A Mercury year takes 88 of our days, but a Mercury day lasts 116, exactly two of its years. It gives the term ‘birthday’ a whole new meaning. At mid-day, every other year, the temperature gets up to 450 C, and at night it cools down to -120C, the largest range in the Solar System. Still, Venus gets hotter and the mid-day heat is not hot enough to melt the rock. In any case, surface heat can’t form volcanoes, only internal heat can. The interior has cooled a lot (a small planet looses heat faster than a big one). So the strong expectation was that Mercury probably would have had volcanoes, but would do so no longer. The Messenger space craft, which orbited Mercury from 2011 to 2015, confirmed this, but it found that ancient Mercury had been very active indeed.
Getting to Mercury is hellishly difficult. Venus and Mars have received many spacecraft, but Mercury has only been visited twice. The journey takes too much fuel; it is at the limit of what rockets can do. Mariner 10 was the first one to come, just before the focus of the US space program decisively shifted to the outer Solar System (after it, Mariner 11 was renamed Voyager 1). It managed three fly-by’s of Mercury, in accordance with the Coleridge poem The Ancient Mariner, with which it shared its name: He is an ancient Mariner, And he stoppeth one of three. Except that Mariner 10 never could stop.
To get to Mercury, Mariner 10 used a gravity assist from Venus, the first Solar system probe to steal a bit of extra velocity that way. Every little helps: every kilogram of fuel you don’t have to carry means 10 kilogram of extra useful weight, and Mariner 10 had only about 500 kilogram of load to begin with. After this Venus fly-by, it arrived at Mercury on 16 March 1974 but it lacked the fuel needed to stop and flew past at high speed. A second flyby was rather distant, and finally the close encounter of the third kind brought it to within 700 kilometer of the surface. A week after the third fly-by, NASA turned off the radio. Presumably it is still there, forever chasing Mercury and wondering what on Earth has happened to us and to its albatross, Mercury. Again in the words of the Ancient Mariner: Alone, alone, all, all alone, Alone on a wide wide sea! For many years, its grainy images of half the surface were the only ones we had.
35 years (!) later, the Messenger probe arrived. This was a far more ambitious probe. Launched in 2004, it took 7 years to get into orbit around Mercury, and at the end of its mission, in 2015 , after 5 years or orbiting, it was allowed to crash into the planet. The probe weighted 1100 kilograms. To cope with the heat, out where the Sun is ten times brighter than on Earth, it took its clue from Mary Poppins and carried a parasol: all of the spacecraft was shielded by a sun shade. Only (obviously) the solar panels peeked out from underneath the shade. The sun shade could not protect Messenger from the heat reflected of the surface of Mercury. To cope with this, it followed a highly elliptical orbit, taking it within 200 kilometer of the surface, followed by a spell as far as 15,000 kilometer from the planet where it could shed the incinerating heat reflected off the planet. The two integrated electronics modules (one as a backup) ran off a 25-MHz processor. This was, after all, 20 year old technology. The solar panels provided 450 Watt of power, a positive luxury for a deep space mission.
In order to keep the fuel demand within reasonable limits, Messenger made generous use of other people’s gravity. It was launched in August 2004. On its post-launch orbit, it didn’t even get as far as Venus. After one year, it was back at Earth with nothing to show for it. But now the tricks began. It flew low over Earth, and let Earth’s gravity bend its orbit and so picked up extra speed. (To be precise, you need to lose speed to go the inner Solar System, not gain, but the principle is the same.) Now it could get to Venus, without having had to spend a drop of extra fuel. In October 2006 and again in June 2007, it flew close to Venus, picking up (negative) speed both times. Now it finally could get to Mercury. But it still wasn’t enough – on arrival at Mercury in January 2008, it whizzed past the planet. No problem, it used this encounter to decelerate further, did the same at the next encounter in October 2008, and the next one in October 2009, and finally was caught by Mercury on the fourth encounter, in March 2011. Space travel has become complex and slow. Mercury is only a little further than Mars, but whilst going to Mars takes 7 months, getting to Mercury took 7 years.
To finally get into orbit required 200 kilogram of fuel. After 7 years of free loading, it burned 20% of its weight in just 15 minutes.
Five years of scientific studies followed. Because of the highly elliptical orbit, one side of the planet (the north) cold be studied in much more detail than the other. The X-ray instrument didn’t work so well at first, because the Sun (which provided the X-rays) had an unusually deep minimum. But the mission was a flying success. It ended in 2015 when Messenger ran out of fuel and crashed into the surface. (This was a planned crash, no need to worry.)
There is one further mission to Mercury scheduled: a joint European/Japanese mission to Mercury, Bepi-Colombo, will be launched next year. A surface landing has been planned for this mission, but sadly it was sacrificed to the budget gods. Both partners have a history of unplanned crash-landings on non-terrestrial surfaces, so possibly the insurance premiums just became astronomical.
The topography of Mercury shows the usual aspects of planetary surfaces: high regions, low regions, craters of all sizes, and ridges. The range of heights isn’t large, by planetary standards. From the lowest to the highest point is less than 10 kilometer in altitude, considerable less than the Moon (20 kilometer), Mars (30 kilometer) or even Earth (20 kilometer). On Mars, which has similar surface gravity, the range of altitude is dominated by Tharsis, a huge volcanic area undoubtedly pushed up by the mantle. This suggests that Mercury never formed tall volcanoes, and either lacked strong mantle plumes or has a lithosphere too weak to carry a big bulge.
The polar regions are a bit lower, whilst most of the highlands are around the equatorial regions. That is probably not accidental. It can be caused by something called ‘true polar wander’ where the crust slowly reorients itself so that the highest points are now at the equator. Mars has done this: the Tharsis bulge is exactly on the equator, even though the surface structures suggests that when it formed it was 20 degrees off the equator. It requires a strong solid crust, and a liquid lithosphere or upper mantle which allows the entire crust to shift.
394 craters in Mercury have been named. If they are larger than 250 kilometer, they are called a ‘basin’ rather than a crater. There are many more smaller craters, of course: counting craters in one area and scaling to the surface area of Mercury, I estimate upward of half a million craters in total. This includes quite a lot of secondary craters: debris kicked up by one impact forms more craters as it rains down. These form the rays surrounding some of craters. Mercurian craters tend to be a bit deeper than similar-sized craters on Mars, possibly because the average impact velocity is higher.
The largest crater by far is the Caloris basin, a staggering 1550 kilometer across. It is one of the largest craters in the Solar System, the scar of an impactor around 100 kilometer diameter. Mariner 10 only imaged half of it: it was located close to the edge of its coverage. Messenger completed the job 35 years later. But in spite of its fame and glory, it is not that easy to see Caloris. On the topographical map above it is not visible, and there are few images available that clearly show it. The crater walls are easier to pick out than the interior. It is better seen in (enhanced) colour images, because the inside of the basin is smoother and redder than the outside. But it is not deep. Part of the inside is even higher than the crater wall: clearly there has been infill and perhaps uplift, an attempt to wipe it off the map. Outside of the main crater wall, which is 3 kilometer high in places, there is a second wall, about 100 kilometer further out. Outside of that there is a vast debris field where the ejecta ended up.
The next largest craters are Rembrandt (720 km), Beethoven (62 km) and Tolstoj (510 km), also sizeable wounds. They are better visible.
The presence of craters allow one to identify which are younger and which are older terrains: the former will have fewer craters. The area around, and inside the Caloris Basin stands out as having few craters. Either Caloris was the last of the major impacts, or the surface here was reworked later on. ‘Younger’ is relative: it would still be 3.5-4 billion years old. Rembrandt has a similar age but Beethoven and Tolstoj are older. (The order of the names may not be entirely historically correct.)
Scarps and rupes
The surface of Mercury is more than just craters and basins. The cliff running through Rembrandt is an example of something different. There a number of other curved scarps. Discovery Rupes (Latin for cliff) is 500 km long and over 1 km high. These are thrust faults, and as they cut across craters they must be ore recent that these impacts. How can a planet without plate tectonics have faults? It is caused by its small size. Mercury has cooled over time and much of the core has solidified. The interior has shrunk because of this, and the crust was left oversized. As the planet shrank, by perhaps 5 kilometer, the crust wrinkled. The scarps are these wrinkles, the signs of old-age.
But there may be a bit more to them. Take a hard look and there are impressions of buried impact craters in places. Some of the rupes and scarps seem to run as arcs around these vaguely visible features. Perhaps they are signs of an underlying older surface, rather than the deeper interior. But many scarps are not related to such buried features, and remain attributed to planetary shrinking.
Around a quarter of the surface of Mercury are the ‘lowlands’, and about a quarter of these are ‘smooth’ regions. The main regions are in the north (the North Volcanic Province) and surrounding the Caloris basin. Both are regions of thinner crust, 20-40 kilometers, as compared to the 50-80 kilometer-thick crust around the equatorial regions.
The smooth plains are young volcanics. The lava flows appear to be up to 1-2 kilometer thick and they cover an older terrain, flowing over impact craters and debris fields. The covered impact sites are known as the ghost craters. There are fewer craters on the surface of the smooth plains, and this shows that they formed relatively late, meaning 3.5-4 billion years ago. The ones around Caloris are clearly associated in some way with the impact basin. There may also be a large impact basin underneath the northern volcanic plains but this is disputed.
The smooth lava flows created low shield volcanoes or lava plains. The lava was very fluid (low viscosity), and was able travel very large distances. No stratovolcanoes or cinder cones here! The volcanic plains show that Mercury had a phase of massive flood lavas, putting Iceland to shame.
nnels. In the image, the partly buried mounts inside the channel showed how the lava flowed, the kipukas of Mercury. Strange pits in the surface nearby suggests that some of the lava flowed underground.
Not all the flows flowed. In some places there is a bright, reddish material surrounding a small volcanic flow, believed to be debris from pyroclastic flows. These volcanoes were clearly a combination of effusive and explosive eruptions, and the explosions were very powerful, with the pyroclastics reaching 50 kilometer from the vents. There is only one case where there is a pyroclastic deposit but no effusive lava. The explosions were unexpected: as Mercury formed so close to the Sun, all the volatiles (water, CO2, etc) were believed to have evaporated before Mercury ever formed. Without such volatiles, you can’t get explosions. Clearly that expectation had been wrong.
Some of the smaller lava fields show more than one point of origin. One field inside the Caloris basin formed from nine different, overlapping vents. These are compound volcanoes, perhaps caused by an eruption site migrating along a dyke. Most of them trace explosive eruptions.
Pits and hollows
The image shows a field of ‘hollows’ where the surface appears to have been eaten away, or more likely collapsed. The field is 10-15 kilometer long. There are quite a few such fields on Mercury.
Mercury’s acne comes in two different types. Pit craters form inside impact craters, are deeper and have irregular floors. They are surrounded by ejecta and appear to be volcanic in nature. Hollows are smaller and shallow and have flat floors. Curiously, hollows are more common on slopes that face towards the equator, and are also more common on the two hottest areas of Mercury (the two regions where ‘noon’ occurs when Mercury is at its closest to the Sun). This suggest they form as something sublimates, but it is not clear what this ‘something’ is. It is the second indication, after the pyroclastics, that Mercury has unexpected volatiles.
Hollows tend to form on darker, less reflective surfaces areas but the flat bottoms are bright. We don’t quite understand the formation, but perhaps they form where a thin layer (of lava?) overlays a different surface, and whatever sublimates is found only in the top layer. We don’t know whether the volatiles are unique to the dark material, or that these are more susceptible to sublimation because dark material gets hotter in the Sun.
The surface of Mercury showed some surprises. For such an iron-rich planet, there was rather little iron on the surface. There was also a lack of aluminium and calcium, but a lot of magnesium. The composition was intermediate between that of Komatiite (very hot lava) on Earth and basaltic mare on the Moon. Sulphur was much more abundant than had been expected, ten times higher than in the crust of the other terrestrial planets. It was found especially in places with higher calcium, and perhaps it came up as CaS.
Back to Caloris
The Caloris basin is much more notable from its colour than from its topography. It has clearly been filled in, while the region outside of the crater has also been flooded with lava but of a different colour. They are separated by the crater rim, up to 3 kilometer high but in places obliterated by the later lava. Where the crater rim is visible, it is much more densely cratered than the lava plains. This means that the plains formed considerably later than the crater itself. Although there are many places where the two lavas are adjacent, it is not clear whether one consistently covered another. Thus, we don’t know which one came first.
Where the redder material inside has been hit by a later impact, the deepest impacts show that the underlying material was blue (note: all colours are greatly exaggerated. This is not what your eye would see.) But similarly, a deep crater in the blue region revealed a red underlay. It is confusing.
The only way this seems to makes sense is that the original material was blue. This may have been a flood event sometime after the impact. It was embayed by 1-2 kilometer of red flood basalt coming from within the crater, which in many places overflowed the rim. Finally, it was covered by another batch blue flood basalt coming from outside the crater but in a few places flowing into the crater. The process was completed by 3.5 billion years ago, and there has been little or no volcanic activity since, apart from the formation of the hollows.
Composition measurements from Messenger show that the red material has high Si and Al, and low Fe, Mg, Ca and S compared to the rest of the planet. The blue material is more standard (for Mercury) in composition. Clearly, they were formed by different eruptions from different magma reservoirs. Of course, Mg, Fe and Ca are most affected by partial crystallization: if the magma sits for a while and cools a bit, these elements will form solids and drop to the bottom. The composition of the red material inside Caloris is consistent with this. But it is unique on Mercury: no other region appears to have this composition.
To explain these events, we need a large magma chamber which was sitting underneath Caloris. For some reason underneath Caloris it cooled faster or further than it did underneath the exterior. This may be because of the thinner crust left by the impact, which provided less thermal insulation. A hundred million years after the impact (speculation alert), the magma erupted on the surface. What triggered this? That is not known – perhaps the shrinking of the planet provided the pathways. The magma came up, and flooded the interior of the basin, and perhaps shortly after this the exterior region flooded from its underlying magma.
The lava almost filled the basin. It took a long time to cool. As it cooled, it compacted, and for unclear reasons an intricate spider web of cracks developed. The image below shows this web. It is centred on a crater (‘only’ 40 kilometer across) but whether the spider came from this secondary impact is now known. It is amazingly similar to the ‘astrums’ on Venus, and these are due to doming: a local uplift. The spider web, named pantheon fossae, remains a mystery.
Mercury has been dead for 3.5 billion years. The scars show a fascinating history, volcanic eruptions unlike Earth has ever seen, but it was a long time ago. Underneath the blazing Sun, the planet has cooled too far.
But it could still rule the Solar System. Mercury has a funny orbit, close enough to the Sun that it feels a different, non-Newtonian gravity. Its precessing elliptical orbit had at first been interpreted as due to perturbations from another planet. This new planet was even given a name: orbiting between Mercury and the Sun, the appropriate name was clearly Volcano. A few astronomers saw it crossing in front of the Sun. At least until Einstein decided that gravity itself was to blame. He came up with General Relativity, and this explained Mercury’s precession perfectly. Volcano was never seen again.
This precession, though, is one of the danger points of the Solar System. The rate of precession is, by a complete fluke, very close to that of Jupiter. In the future, the two may fall into sync, and Jupiter would distort the elliptical orbit of Mercury and pull it away from the Sun. Free-flying Mercury would play havoc with the inner Solar System. In some models, it collides with Venus – or with Earth. In one model, an interaction causes Venus and Earth to swap places. An unguided messenger is a terrible danger! The chance of this happening is quite small: it is estimated as only 1-2% over the next few billion years. But it is not zero, and it is the biggest danger the Solar System faces. Take care.