Search our site

Custom Search

Friday, 1 July 2016

How were Mars’ moons formed?

The formation of Deimos and Phobos, the moons of Mars, is still somewhat of a mystery. They were discovered by Asaph Hall in 1877, and observed in 1971 by Mariner 9, a NASA spacecraft orbiting Mars[1]. Although not the smallest moons, they are much smaller in comparison to Mars than Earth’s moon is to Earth.

Color composite of Phobos and Deimos
Deimos and Phobos have mean diameters 280 and 154 times smaller than Earth’s moon respectively. Images courtesy NASA (Phobos) and NASA/JPL-Caltech/University of Arizona (Deimos), composite by TWDK

Possible Theories of Formation: Asteroid Capture

Despite being known for so long, there is no accepted theory regarding their creation. They appear to be made of “...carbon-rich rock mixed with ice”[1], and are oddly shaped, which led to the idea that they are captured asteroids. This would also explain their heavy cratering and small size.

An asteroid is captured when it passes a larger mass (in this case, a planet), and is “caught” by the planet’s gravitational field and is forced into orbit. This means that the orbits of captured asteroids are expected to be very eccentric ellipses, meaning that the asteroids pass close by before swinging out further away. The orbits of Phobos and Deimos, however, are almost circular. Because of this, we can’t consider asteroid capture to be the definitive theory of the formation of Mars’ moons.

In 2003, A. V. Ivanov, a researcher at the Verdansky Institute of Geochemistry and Analytical Chemistry at the Russian Academy of Sciences, suggested a timeline for the capture of an asteroid by the protomars circumplanetary nebula (the nebula surrounding early Mars). In the suggested scenario, Phobos originates as an asteroid composed of carbonaceous chondrites formed the external zone of the asteroid belt[2]. Chondrites are materials containing chondrules (often found in stony meteorites), which are small (millimetre scale) spherical silicate objects.

Ivanov proposed that this could have occurred approximately 4.56 billion years ago, after which Phobos travelled to the solar system[2]. In his paper, he then suggests that Phobos was captured by the nebula surrounding early Mars (around 1 million years after the formation of chondrules) and then later began orbiting Mars[2].

Another paper by D. M. Hunten, a physicist at the University of Arizona, discusses an alternative theory for the capture of Phobos and Deimos. He considered two possible methods that could have brought asteroids near to Mars: “drag by the solar nebula” and “scattering by proto-Jupiter”[3]. The solar nebula is the disc-shaped cloud of gas and dust left over from the Sun’s formation, and proto-Jupiter is another way of referring to early Jupiter. He suggests that it is unlikely that Mars’ moons were “formed in orbit around it” as he assumes they are carbonaceous asteroids, and that, as a result of his assumption, they have a very different composition to that of Mars.

In order for Phobos and Deimos to have the orbits we see today, Hunten theorizes that they would have to have come towards Mars with a speed below a particular limit that would require “favourable circumstance” , or for the orbit of Mars to have had a much smaller eccentricity than it has now, to be achieved[3].

Adapting the work of two other scientists (Goldreich and Ward), Hunten finds that the orbital lifetime of Phobos and Deimos is long enough to allow them to travel to Mars from the asteroid belt. Although he concedes that it would not have been impossible for Phobos and Deimos to have been brought to Mars via drag by the solar nebula, he does suggest that it does not seem likely.

Possible Theories of Formation: Giant Impact

Another hypothesis of their formation is that they were created by a large object colliding with Mars, which dislodged a small amount of Mars’ mass and sent fragments of it into orbit. The fragments would then combine and, over a period of time, form moons[4]. If this were the case, then we would expect Phobos and Deimos to be composed of the same materials as the crust of Mars, however the composition of these moons is another Thing We Don’t Know.

HD 131488 Collision
Many scientists consider the Earth's moon to have been formed as the result of a giant collision, but is this also true of the moons of Mars? Image credit: Gemini Observatory/AURA/Lynette Cook

Mars’ 25-hour day is already considered evidence that it experienced a large impact, as the rotation speed needed to have a day of that length can’t be achieved just by material coming together ... from a uniform disk[5]. In models of impact scenarios, variables deduced from the spins of Earth and Mars lie well within the [expected] range[6]. A paper presented by Canup and Salmon at a recent Lunar and Planetary Science Conference tested the idea that such an impact created Deimos and Phobos.

Canup and Salmon simulated several impacts and “satellite accretion” situations (basically how the fragments combine together) to see whether they could find a scenario which results in two moons of approximately the right size orbiting at approximately the right distance. Their conclusion, essentially, is that they haven’t yet but might in the future[4].

The main problem with the idea of formation by large impact is that it seems very unlikely that Deimos and Phobos would survive. They will not be able to survive forever, though, as Deimos is slowly spiralling out of Mars’ gravitational pull, and Phobos is slowly spiralling inwards where it will eventually reach the Roche limit (the minimum distance to which a large satellite can approach its primary body without being torn apart by tidal forces) and break up into small fragments.

Any orbiting masses too large would collide with Mars, and a previous study found that all moons formed from fragments within a certain limit would be lost[4]. The paper presented at the conference considers the hypothesis where impact produces a radially extended disk whose outer edge is comparable to Deimos[4].

They ran many impact simulations to find impact angles and speeds that would lead to the approximately 25-hour day that we know Mars has. They found that, due to collisions, the fragments would end up at a characteristic distance. They found that situations that lead to correct length days produced disks whose outer edges ... broadly similar to Deimos’ orbital radius[4].

The next stage was to simulate how the disks would evolve (the accretion simulations), and look for any situations that produced moons equivalent to Deimos and Phobos. They found that orbitally expanding moons had the potential to combine with all of the smaller fragments, meaning that the matter that could have become Deimos and Phobos would have become part of a larger moon[4].

In their study, Canup and Salmon only considered the case where there were no satellite tides where tidal effects cause the decay of an orbit - such as the tidal deceleration causing Phobos’ orbit to spiral slowly inwards. In this limiting case, the only moons remaining after 10 million years had much larger masses than Phobos, and would experience tidal deceleration causing them to collide with Mars or reach the Roche limit and break up into small fragments.

Because only one limiting case for satellite accretion was considered, we cannot rule out the idea that Deimos and Phobos were formed by a large impact; but due to their inconclusive result, we also can’t consider it to be the definitive theory of their formation.

All hope for this theory is not lost, however, as Canup and Salmon state two effects which could lead to a model that predicts the formation of Deimos and Phobos-like moons: stronger planetary tides, and the inclusion of satellite tides. Stronger planetary tides would have the potential to increase the probability that small moons could survive instead of becoming part of larger moons, and the inclusion of satellite tides would increase their stability. They are investigating these cases now.

This article was written by Alice Wayne, a physics undergraduate from Royal Holloway, University of London as part of the annual TWDK-SEPnet internship scheme.

why don't all references have links?

[1] Dunford, B., and McKissick, K. “Mars: Moons”, Solar System Exploration, NASA (web)
[2] Ivanov, A. V. “Is the Kaidun Meteorite a Sample from Phobos?”, Solar System Research, 2004, Vol.38(2), p97-107, doi: 10.1023/B:SOLS.0000022821.22821.84
[3] Hunten, D. M. “Capture of Phobos and Deimos by photoatmospheric drag.” Icarus, vol.37(1), p.113-123, 1979. doi:10.1016/0019-1035(79)90119-2
[4] Canup, R. M., and Salmon, J. "On an Origin of Phobos-Deimos by Giant Impact." In Lunar and Planetary Science Conference, vol. 47, p. 2598, 2016.
[5] Dones, L., and Tremaine, S. “On the Origin of Planetary Spins.” Icarus, vol.103(1), p.67-92, 1993. doi:10.1006/icar.1993.1059
[6] Tremaine, S., and Dones, L. “On the Statistical Distribution of Massive Impactors.” Icarus, vol.106(1), p.335-341, 1993. doi:10.1006/icar.1993.1175