Dawn over Ceres: the lonely volcano

Ceres is different. It was the first asteroid to be discovered and is by some distance the largest. Ceres contains a quarter of all the mass on the entire asteroid belt. (That sounds more impressive than it is: the mass is just over 1% of that of the Moon.) But it does not look like the others. It is the only asteroid that is round, pulled into that shape by its own gravity. That qualifies it as a dwarf planet: Ceres is the only one in the inner solar system. Vesta, which is a little smaller, is nowhere near round. That already tells you that Ceres is made from weaker materials. Vesta is rocky, a proto-planet in all but name. Ceres is different, like no other planet in the inner solar system. It is midway between the rocky objects of the inner solar system, and the icy moons, comets and dwarfs of the outer solar system. It is unique – and being unique can be a lonely experience. Ceres could have empathized with lonely George, the last of its species for a century of its life. But for science, uniqueness is an opportunity. Scientists love studying something that is different from anything else. There is so much to learn. And in its uniqueness, Ceres may hold the key to one of the mysteries of the solar system. Perhaps it can tell us where Earth got its water from.

This is the third instalment of the Dawn trilogy. Part 1 was about the spacecraft and its unique propulsion system. Part 2 discussed Vesta. They can be read independently.

Ice and rock

The problem of Earth’s water is a simple one. Planets grow from smaller things which collide and stick together. These smaller things must be solid: baby planets have too little gravity to hold on to any gas. For something to be solid, the temperature should be below its sublimation temperature. Rock is fine up to temperatures exceeding1000 Kelvin. This allows space rocks to exist as close as 20 million kilometers from the Sun. Indeed, Mercury is some 60 million kilometers from the Sun, end even though a bit hot, the planet is neither melting nor evaporating. The asteroid Icarus gets even closer to the Sun than Mercury, befitting its name, but it too is in no danger, apart from getting a bit toasty and thoroughly sterilized. The Parker Solar Probe which was recently launched is playing a much more dangerous game. It will get so close to the Sun that its temperature will reach over 2000 Kelvin. That is too much for most materials. Most of the spacecraft will be hiding behind a thick and innovative sun shield, made from carbon which is stable even at these temperatures. The solar panels will have to peek around the sun shield (for obvious reasons) but they contain an active cooling system, and will limit their exposure to the sun to stop them from getting too hot.

Ice is more fragile: it evaporates at much lower temperatures than rock or metals do. Under vacuum, ice becomes vapour at 200 K (snow in vacuum requires much lower temperatures than snow on Earth). Those kind of temperatures are only found further our, at 500 million kilometers from the Sun or more, three times the distance from the Earth to the Sun. The limit for frozen water is somewhere within the asteroid belt. Planets, moons, or other objects which formed further out could incorporate ice, and because both oxygen and hydrogen are much more common in the Universe than silicates and other rocky elements, they quickly become mainly water. This allowed comets to form, as well icy objects such as the moons of the gas giants, with their thick oceans – and even Pluto. A comet could not have formed in or inside the asteroid belt, because of the lack of frozen water. This limit is called the snow line. Sometimes we see a comet, unwisely, getting very close to the Sun, the proverbial snowball in hell; they become spectacularly bright (albeit difficult to see because they are so close to the Sun), but afterwards they often disappear, completely vaporized by the heat. They are called the kamikaze comets.

This also explains how rocky planets formed. They are made almost exclusively from elements that are rare in space, such as silicon and iron. These are the only elements which could form solids close to the Sun: all other elements evaporated and disappeared before the growing planet could capture them. Vesta belongs to this group, and it is dry and dusty, without water. Ceres formed closer to the snow line, and it managed to attract quite a lot of water. Effectively, it is made of wet rock.

But the surface of the Earth, the ultimate rocky planet, also has quite a lot of water. How does that fit in? It turns out all the Earth’s water is near or at the surface. Down in the mantle, the planet becomes bone dry. The total amount of water, adding the atmosphere, the surface and the crust, is about three times the amount in the oceans. Compared to the total mass of the Earth, that is a very small fraction. It just so happens that most of it congregates around Manchester, giving the impression that Earth is much wetter than it really is.

Water: in spite of the occasional problems, we can’t live without it. But where did the Earth get its water from? And how come we have just enough to fill the oceans but not flood the continents?

From the fact that all the water is near the surface, you can already deduce it was a late addition. The water came when the planet was already fully formed and formed a wet veneer. But where did it come from?

The rocks of Ceres

We already knew quite a bit about the rocks on the surface of Ceres before Dawn arrived. No meteorite has ever been traced to Ceres, so we lacked the direct evidence that we had for Vesta. But it is amazing how much can be done with a large telescope. The observations showed a dark surface: the dark colours were interpreted as a mix between carbon and clay. Infrared spectra showed several characteristic absorptions in the infrared which indicated the presence of a form of phyllosilicates, containing water or ammonia.

Phyllosilicates, or sheet silicates, contain connected rings of tetrahydrons, where each tetrahydron consists of a silicon atom surround by four oxygen atoms. Three of the oxygen atoms are shared with other tetrahydrons. Often, captive OH molecules are present within the ring. Phyllosilicates form from weathering of rock, involving water. Clay is an example of a phyllosilicate, and this was a good candidate for part of the surface of Ceres. Water is itself difficult to detect from Earth, because the water in our atmosphere gets in the way, and you don’t know what is space water and what is local.

The density of Ceres was found to be around 2100 kg/m3. This is intermediate between silicate rock and ice, which suggested that Ceres was made of rock with a significant fraction (20-30%) of water.

Where would you expect the water on Ceres to be? After all, on Earth it ended up on or near the outside. Models suggested that the radioactive elements presents in the young solar system would have heated up the interior of Ceres quite effectively. Even if Ceres formed a little later, when the strongest heat sources would have decayed, the water in the interior would still have comprehensively melted. The silicate rock would have ‘weathered’ at this time. The heavier elements would have sunk through the melt, and the water and altered silicates would have migrated upward. These models predict that the core will be dry rock, and the mantle of Ceres will contain most of its water. The early radioactive heat has long gone, and the interior will have cooled down. Even in the mantle, the water became ice.

Dawn confirmed this basic structure. The core of Ceres has a radius of around 250 km, and the ice-rich mantle is some 100 km thick. The outer crust has a thickness of 30 kilometer on average, with a thicker region close to the equator and a few near-holes where an impact did real damage. The density of the mantle was found to be 2400 kg/m3. The crust has much lower density, 1200 kg/m3, almost low enough to float. The mantle is rock with ice, but the density of the crust suggests it is more like ice with some rock. But the surface itself is pure rock: any ice would quickly sublimate under the faint sun.

Thickness of the crust, as measured by Dawn. Ermakoff et al. 2017, Journal of Geophysical Research: Planets, 122, 2267–2293.

The surface

Of course, pictures and other observations mainly show the surface, and this is what gets most news coverage. Dawn arrived at Ceres in March 2015. A few images were obtained early on, and these caught the attention when two bright white spots were seen, of unknown origin. The first phase was spend in a high altitude orbit. Afterwards Dawn moved closer, finally getting close enough for images which could see details as small as 50 meters. The images became more detailed as Ceres moved closer, and slowly we deciphered the surface.

Let’s first gives some basics. Ceres is chilly, being 3 times further from the Sun than we are. The day-time surface temperatures hover around -40C (which for the convenience of our US readers is the same as -40F). The poles are colder and they even contain a few areas that are in permanent darkness. A day on Ceres lasts ten hours, and we are still close enough to the Sun that it makes a difference. So almost any ice ice that makes it to the surface, will over time experience unsuitably hot weather and slowly disappear in the heat of the day.

Craters of Ceres. The linear features come from debris thrown out by a larger impact. Note that many craters have a ghost-like appearance, looking as if they have been eroded or erased.

The images showed the usual crater-covered surface of airless bodies. But a closer look revealed that there was something missing: there were no large craters. The largest impact crater was 280 km across. Craters twice that size had been expected. There were 16 craters larger than 100 kilometer, but there should have been more than 40. Something was removing the largest impact scars. There were in fact a few basins, shallow depressions around 500 kilometer across. If these were the missing craters, the scars had been mostly healed. Volcanic eruptions could have covered them, as happened on Mars. But ice can do the same thing: a crater formed on ice becomes less distinct as over time the ice rebounds and flows back into the gap. The basins are densely covered with smaller craters, so the healing happened a long time ago, perhaps when Ceres still had radioactive heat which kept the ice ductile.

From Ermakoff et al. 2017, Journal of Geophysical Research: Planets, 122, 2267–2293. the colour shows the height above the standard surface level. The ancient largest basin can just be recognized around 120-150 degree longitude, and latitude from -20 to +60 degrees.

But ice turned out to be too flexible. Craters in ice would disappear too quickly. Somehow the material was stronger than ice, but weaker than rock. The missing ingredient appears to be clathrate hydrates: material where gas molecules are caught inside the ice. This strengthens the ice, and now the properties fitted with what is observed on Ceres. Indeed, this could have slowly erased the oldest and largest craters on Ceres.

Oxo and Haulani

The impact crater Oxo turned out to be particularly interesting. It is relatively small, at 9 km diameter, but stands out because it is bright. This image shows the crater. The sharp structures suggest it is young: the age is estimated as less than 200,000 years. One side shows bright streaks. Spectra have shown that these are water: this is the only direct evidence for water on Ceres, and the fact that it was revealed by this impact shows that it is found just below the surface. Note the big slump on the south east (bottom right) which has pushed part of the crater rim a kilometer into the crater. On the north and west, the bright stripes suggest that the ejecta from the crater also contained water. But interestingly, the water is only seen on one side.

This image is of a crater called Haulani, 34-kilometers in diameter. It shows a blueish material surrounding and inside the crater. It is part of the ejecta blanket, but differs from the reddish surface material (the colours are of course enhanced and not as the human eye would see it). The blue material comes from below the surface and was excavated by the impact. It likely came from the water-rich material hiding below. Some other craters show the same effect, but it is only seen in young craters. The lack of impact craters inside of Haulani shows that this crater is also young. Over time, the blueish material degrades back to the normal surface composition.

White lava

The high plateau near the equator is called the Haunami planum. This region caught the attention very quickly after Dawn’s arrival. The first images showed a bright spot in this area. A closer look showed that the spot was double, with one half spot in the very center of a crater. The crater is now called Occator, and is 92 kilometer wide and 4 kilometer deep. The crater itself is unexceptional, apart from some clear landslides along the rim, with debris lobes. One of the lobes covers a quarter of the crater floor. The spots are the outstanding feature. There are some other spots on Ceres, but none are as bright as these. What are they, how did they form and why are they in the centre of an impact crater?

The spots were given their own name: the one in the center is named Cerealia Facula and the off-center one (in fact a cluster of spots) is known as Vinalia Faculae. The names do not roll off the tongue. The center one is located on and around a central dome 400 meter tall and 3 kilometer wide. It is surrounded by a depression, 11 kilometer wide and 600 meter deep, within the main crater. The surface around the dome appears to contain carbonates, forming a dark material.

Occator crater, with its bright spots

Fractures in Occator

The central dome is surrounded by a pattern of fractures, some concentric, some more complex. It is as if something tried to push its way out, and managed to bulge and break the surface, without actually escaping. When did this happen? The crater is estimated to be 30 million years old. But the dome appears to be much younger, perhaps only 4 million years. Whatever happened, it was long after the crater formed. There are some remnants of a central peak, partly obliterated by the dome.

From Nathues et al, 2017, Astronomical Journal, 153, 112


The topography gives another indication that the white material came from below: it is found only where the crater floor is deepest.

Dawn measurements told us what the white material is: it is sodium carbonate, a kind of salt. Cracks and fractures showed that it came from below. They allowed water to percolate up and reach the surface. There, the water evaporated and disappeared (it is near the equator, after all). But the water was briny, and the salts in the water stayed behind. The sodium carbonate is white. Carbonates can contain other elements than sodium, and this can change the colour. Pure carbonates are very dark. It gives a stark contrast.

There are stlll many questions. Did the dome form from below or by deposition of salt? Did the water come up from a shallow reservoir or from deeper? Was heat involved? Why did such spots form here? This remains unclear. But the spots of Occator have given us a view of the ice and brine hiding below the surface. And can we call this kind of activity volcanic? It involves molten material coming to the surface through cracks. But perhaps Ceres is a little too wet for liquid water to count as magma, and the dome is a bit small. It is a borderline case.

Ahuna Mons

But there is a real volcano on Ceres. It is the only volcano on a unique planet: perhaps this is the loneliest volcano in the Solar System. Dawn kept it company.

Ahuna Mons. The vertical scale is exaggerated by a factor of 2. Source: NASA

Source: Ruesch et al., 2016, Science, 353

Ahuna Mons is a pyramid-shaped mountain, 17 kilometer wide and rising over 4 kilometer above the surroundings. It really stands alone: the region is undulating but not hilly, and the mountain seems to come from nowhere. And although there are higher regions on Ceres, there is no other mountain like it on the entire dwarf planet. It is unique.

The top is partly flat, giving the pyramid a mesa-like appearance. The mountain stands next to a crater so that the side abuts the crater rim, but the crater and volcano appear unrelated. There are fractures visible on the top. The sides are uniformly steep, 30 to 40 degrees and have streaks from rock falls.

The crater is estimated to be 200 million years old. The surface of Ahuna Mons appears to be considerably younger than this. The streaks along the side show that the slopes are still unstable. The stripes will slowly erode over time, and the fact that they are clearly visible shows they are young. (‘Young’ should be read in a relative way: perhaps up to a few tens of millions of years, which to an astronomer is toddler age.) The slopes still have a very sharp separation from the surrounding material: again this will smooth over time. In space there is still erosion, due to a continuous bombardment by micrometeorites. The meteoric rain mixes up the upper centimeter of the surface within about a million years in a process that (somewhat optimistically) is called gardening.

There are several ways to form such a mountain. There is compression, a tectonic process which forms mountain chains on Earth. But this creates faults and those are not seen around Ahuna Mons. The second way is emplacement, where a magma chamber inflates the surface. But this requires a thin, elastic crust, more so than Ceres appears to have. And there is extrusion, where you push stuff out through a hole to the surface until you have a mountain. This is similar to how a volcanic cryptodome (or lava dome) works, and it is the most likely explanation for Ahuna Mons. The structure is not uniform: there are multiple small hills and short separate fractures which suggest the extrusion was not a single event but happened over a long time. The lack of lobes and the steep slopes shows that the material had a high viscosity, and cooled by conduction within a hundred thousand years.

This model suggests that Ahuna Mons formed from a cryomagma creating a massive cryptodome. The cryomagma is thought to be a combination of chloride-rich brines, carbonates and phyllo-silicates, and water ice. In this mixture, the brines would have melted, with a melt fraction of a few per cent, allowing the magma the travel up. The fractures that allowed the magma to move up may have come from the nearby impacts.

This leaves one question unanswered. What provided the heat to create the melt? Ceres is a small world, which cooled down a long time ago. The answer may lie in the past. Minerals can dissolve into liquid water, and at one time Ceres had liquid water throughout much of its interior. One of those elements was potassium. It has a radioactive isotope, 40K (K stand for the germanic name, kalium). Early on in the evolution of Ceres, this became dissolved into the icy mantle, enriched it, and migrated with the briny water. Now there may be pockets of enhanced potassium below the surface of Ceres, ready and waiting. 40K has a half time of a billion years, and so it is still active.

Ceres is rather warm for an icy body because it is relatively close to the sun. Temperatures below the crust are not far off values where a bit of melt could occur. Add a bit of excess heat from radioactive potassium, and Ceres is ready for action. Whenever conditions are right, a little melt happens and the viscous cryomagma begins to travel towards the surface, if it can find the right fracture. On top, sometimes the sticky fluid may create salty puddles, leading to white spots. At other times, perhaps once in a 100 million years, a 4-kilometer tall cryptodome may form.

So Ceres isn’t a lost cause. It is like a teenager in the early morning – to all accounts dead to the world, but with hidden energy inside. It holds hope and promise.

Part of the dome of Ahuna Mons


What does this tell us about the mystery of Earth? In the traditional model, Earth obtained its water through a hail of comets. They formed in the outer solar system where ice was the main solid. But it takes a lot of comets to carry an ocean. Dawn showed us that there were water-rich asteroids much closer to us. Jupiter would have disturbed the orbits of many of them, sending them into the inner solar system where Earth, the largest of the terrestrial planets, was ready and waiting. All it needed was the right asteroid to come too close and get caught in the spider web of gravity.

In spite of its tiny mass (0.01% of Earth), Ceres contains a lot of water: roughly 10% of the amount on Earth. Earth needed a couple of large icy asteroids, not too many, not too few, but just right to fill the oceanic basins. Too few asteroids would give Earth a wet crust but without surface water; too many would cover all the continents underneath kilometers of water. We were looking for the Goldilocks asteroids. Perhaps one of those icy asteroids was just right for us.

An object twice the size of Ceres could have provided Earth with all the water we would ever need. Was Earth lucky, and did it win a space ice lottery? Only Ceres knows.

Ceres and Vesta are the two bosses of the asteroid belt. How different they are! One is a protoplanet, the kind of object from which the Earth formed. The other is a water container, and example of the kind of object that may seeded the Earth with water and changed its future. Dawn really did find the morning glory of the Solar System.

Albert, August 2018

Originally this post stated that Ceres contained more water than Earth does. This was a numerical error (wrong by a factor of 10), as pointed out by a commenter below. It has been updated to fix this mistake.

210 thoughts on “Dawn over Ceres: the lonely volcano

  1. This might settle the debate about whether kilauea has a lake.

    Halemaumau is about 2 km wide, and it is close enough to a circle that it has probably got a surface area of 3 km2. The nearest gauge to kilauea shows about half a meter of water (actually less than half a meter but it makes it easier). Half a meter of water falling on an area of 3 km2 would give a volume of 1,500,000 m3. The bottom of the crater is sort of a cone shape, and a cone with a volume of 1.5 million m3 and a slope of 45 degrees (probably close enough to the actual slope) would be about 60 meters deep and 100 meters wide. However it could be about twice the volume if it was 40 inches/1 meter ( 3 million m3 – 90 meters deep) and a lot of the rain fell on mauna loa directly upslope of kilauea, so the groundwater will flow down towards it and so it is quite likely that a pretty big lake (likely the biggest in the state of Hawaii) either already exists and even if it doesn’t yet it will in the near future.

    Water has about 5 times the heat capacity of basalt (4.18 for water and 0.82 for basalt) so 1.5 million m3 of water has the same cooling capacity as 7.5 million m3 of basalt. However basalt is also 3 times denser than water (7.5 million/3 = 2.5 million) and basalt lava is about 4 times hotter than supercritical steam (2.5 million/4 = 624,000) so it would only take about 620,000 m3 of lava to evaporate the entire lake and I think I read somewhere that on mt etna a lava fountain with a height of 1000 meters would have to have an eruption rate of about 1000 m3/s so it would only take about 10 minutes for a high fountain on kilauea to erupt that much lava. It could very well be that I am majorly underestimating the eruption rate too, the fountains during the skaftar fires were only 50% taller than that but closer to 10 times the eruption rate, and etna erupts a more gas rich magma so it could be less than ideal for a volcano like kilauea and maybe skaftar is a better comparison given the similar magma.

    1.5 million m3 of water could become 2.2 km3 if it got flashed to steam by a sudden eruption under it. The volume of the new crater is somewhere between 0.8 and 1 km3. It might take only a minute or less to evaporate the new lake and a lot of that would be almost instant – that would be a seriously big explosion. That is probably what happened in 1790 and the results of what that would do to a human (or a few hundred of them) are quite well known…

    It is starting to look like a really bad idea to reopen the park now…

Comments are closed.