NASA held a telecon/press conference on November 7 to announce the results from the MAVEN, Mars Express (MEX) and Mars Reconnaissance Orbiter (MRO) observations that were obtained during the encounter between Mars and comet Siding Spring. The broadcast can be found here: (http://www.ustream.tv/nasajpl2). During this event, there were presentations that showed three fascinating pieces of evidence indicating that a significant amount of comet dust was deposited into Mars' atmosphere during the encounter.
Mars Orbiter Observations
First, the MAVEN Imaging Ultraviolet Spectrometer (IUVS) obtained spectra of Mars' atmosphere showing highly anomolous levels of magnesium and iron. The explanation for the presence of these materials is that they came from comet dust that burned up when passing through the atmosphere, leaving these metals at altitudes of 80-100 km. Given some simple assumptions about comet dust composition, the MAVEN team suggest that "thousands of kilograms" of dust may have reached Mars.
Second, the MAVEN Neutral Gas and Ion Mass Spectrometer (NGIMS) detected eight different types of ionized metals in the atmosphere, with a spike in the density appearing almost a day after closest approach. As with the IUVS, these results are attributed to the deposition of metals in the atmosphere from cometary material.
Finally, the Mars Express Radar (MARSIS) also detected ions in the atmosphere, but they obtained three measurements that provide some temporal information. The first, at close approach, showed no unusual activity, while the second, 7 hours later, showed a layer of ions at ~100 km altitude. Seven hours after that, the ions had disappeared again. The middle measurement was interpreted as a transient layer of ions produced by the vaporization of dust grains from the comet. The lack of a detection in the close approach observation suggests that the influx of dust occured between the time of close approach and 7 hours afterward.
These observations represent unique data regarding comet Siding Spring and its interaction with Mars. Their analyses promise to reveal detailed information and new insights about the composition of dust grains from a dynamically new comet, the dynamics, structure and density of the dust, and the interactions of large amounts of dust with a terrestrial planet's atmosphere, making this unprecedented encounter a boon to science.
The Siding Spring Dust Models
After the presentations about the observations themselves, there were discussions about the surprising and unexpected amount of dust that hit Mars, and some speculation about the safety of the Mars orbiters and the potential meteor storm that might have been visible from the surface. Much of this discussion involved questions about whether the cometary dust models were representative of what was observed or not.
As the leader of one of the teams that did the dust hazard analyses for the spacecraft, I feel that it is appropriate to address some of these issues with respect to our dust models. (For background regarding the models and why we predicted what we did, see the blog I wrote back in August: http://www.cometcampaign.org/tony/so-close-yet-so-far). In short, a look at our analyses shows that the models were actually very successful in predicting the dust influx that was inferred from the Mars orbiter observations.
First, I acknowledge that our nominal model for the comet did predict "very few" impacts, which is clearly inconsistent with the spacecraft data. The nominal model was based on the best observations we had, primarily between 3 and 5 AU. Those observations contain inherent uncertainties, and then we need to propagate the conditions backward and forward in time, and to larger grain sizes than we could detect, with errors involved in each of these steps.
In recognition of these errors, we developed a second, more extreme model, pushing the uncertainties to their limits to find the maximum number of impacts that we might expect for the comet observations we had. It was this model that was presented to the spacecraft teams and used to define the potential risks. Results from this model, still indicate that there would be no impacts around the time of closest approach because radiation pressure rapidly sweeps the grains away (see my blog above for more details on this and the following discussion). This appears to have been the case, as the MARSIS data at close approach showed no detectable ion formation.
On the other hand, our upper limit model did predict that Mars would encounter dust grains ~100 min after closest approach, when it crossed the comet's orbital plane. Mars would cross a particular part of the dust trail, and that region would be populated with grains 1-3 mm in radius. The column density in this region suggested that there would be one grain in every 10 square kilometer cross section along Mars' path.
For a spacecraft a few meters across, this density presents an almost negligibly low (one-in-ten-million) chance of getting hit. Given this probability, the spacecraft teams opted for simple and easily implemented measures to protect the spacecraft, which mainly amounted to phasing their orbits to put them behind Mars during the crossing of the comet's orbit plane.
Mars, on the other hand, is not a few meters across; with a radius of 3400 km, it has a cross section of 36 million square km. Even for our low predicted dust column density of 1 grain per 10 square km, Mars could be hit by as many as a few million dust grains! If those grains are 1 to 3 mm in size, then this amounts to a total mass of 500-1000 kg of material, which is in line with what the MAVEN results are suggesting entered the atmosphere.
Now, 1000 kg may sound like a lot, but it actually corresponds to a cube of material only about a meter to a side. That's not very much when it is distributed over half a planet, but it is enough to produce the strong signal detected by the spacecraft.
So what does this mean regarding the discussions at the press conference?
First, the amount of dust that hit Mars should not have been a complete surprise. A mild one, maybe, since the numbers seem to be at the upper limit of the models, but they were the numbers that the spacecraft teams were working with in their hazard analyses.
Second, because our dust hazard model is consistent with the observations from the orbiters, our original calculations for the probability of a spacecraft impact remain valid. Given the cost/potential benefit of the action, it was prudent for the teams to position the spacecraft behind the planet during the peak hazard, but if they had not been placed there, the chances of them suffering any damage would remain extremely small. It is notable that Mars Express was exposed during the encounter, yet there have been no reports of damage. Furthermore, there are several annual meteor showers on Earth with comparable or higher rates, and few satellites ever experience damage from them.
Third, what would these grains, burning up in the atmosphere, have looked like from the surface of Mars? This is an interesting question, that I addressed, in part, in my August blog. If we assume that meteors on Mars act in a similar manner to those on Earth, then the 1-3 mm grains would produce respectable trails, comparable to typical meteors seen in most of Earth's meteor showers. They would be bright enough to be seen at night (barring a Martian dust storm), though they would fall well short of being fireballs. They would not be bright enough to be seen in the daytime.
As for numbers of meteors, although there are a few million dust grains hitting the atmosphere, they are distributed over the entire leading face of the planet. Given the curvature of Mars' surface, an observer at any particular site could only see a section of the sky covering an area on the order of a hundred square kilometers or so. Since our model gives a column density of ~1 particle per 10 square km, this suggests that an observer on the surface would see a few tens of meteors during the plane crossing. In comparison to some of the annual meteor showers on Earth, this is not a particularly spectacular event and would certainly not reach the status of a meteor storm.
Keep in mind that the spacecraft results are still preliminary, with a lot of work to be done to finalize the analyses. But as things stand now, our upper limit dust model appears to be in line with both the comet observations and the Mars observations. The biggest surprise in that respect, is that the comet parameters appear to be at the extreme end of their uncertainties. However, another benefit that arises from the Mars orbiter results is that they provide us feedback about the comet's dust environment, specifically regarding parameters that we were not able to measure before the encounter (e.g., large grains). This information can be folded into our analysis so that we can improve our nominal model, to characterize the comet's overall dust environment. In this effort, we can work with the spacecraft teams to develop a comprehensive interpretation that explains all observations of the comet and Mars.