I just finished listening to one of the most fast paced, data flying talks I’ve seen so far. In 15 minutes, dozens of PowerPoint slides flew furiously as J.C. Andrews-Hanna presented tantalizing new results that indicate that Mars may have been hit by a 2230km diameter impactor early in its history (for perspective, Mars’ diameter is ~6800km – what hit it was ~1/3 its current size!).
Here are the details. Anyone who has looked at a topographical map of mars (above right) has probably noticed that the planet has a split topography with one pole being significantly lower elevation than the other. Along with the lower elevation, the crust in these regions is also much thinner. This is referred to has the Mars dichotomy. People trying to understand this weirdness, not seen on such large scales anywhere else in the solar system, have struggled because the boundary of this giant basin are cut into by the Tharsis Bulge – a giant lava flow punctuated with three volcanoes. In order to fully understand this problem, it is necessary to see beneath the lava. Luckily, this can be done with gravity mapping. By comparing the topographic maps with gravity maps, it is possible to build to build theoretical models of lava and crust geometries that reproduce the mass-densities required to recreate the gravity seen in gravity maps and confined to the observed geometries in the top maps. Through these models, this team finds that the basin boundary is continuous beneath the Tharsis bulge, and smoothly spans adjoining features.
When looked at globally, these newly traced boundary allows them to see that the basin is a beautiful elliptical shape super-imposed on a (formerly) spherical planet.
Prior to this dataset, two different and competing models existed to explain this thin crust – thick crust dichotomy. One model was based on internal (called endogenic) theories and impact theories (in fact, the prior speaker had the misfortune of presenting one of these endogenic models). Internal dynamics, such as large upwelling of materials, can alternatively thicken crust if the act in one set of ways, or thin crust if the act in another set of ways. No matter how they work, however, they don’t generally create elliptical shaped effects on the surface. This makes the new data very hard to explain with endogenic theories – they aren’t completely knocked off the table, they just require a lot more add-ons (ones that don’t currently exist) to fit the data.
Impact models, however, can easily (if you have a big enough supercomputer) reproduce an elliptical basin. Infact, several large basins on various objects in the solar system have elliptical basins that look very very similar to the thin crust – thick crust boundary when Tharsis bulge is corrected for.
More tantalizing evidence that this is an impact event exist in Arabia Terra. This regions northern and southern boundaries are parallel and match the expected spacing for an impact the size of this basin. These boundaries could be the double ridge of a of a partial ring boundary, where the rest of the boundary has been erased through cratering, much of it during the age of heavy bombardment. These ring sections also match the structures found with smaller craters when they are properly scaled.
The dynamics required in this type of a collision are quite scary. According to work being done by Marinova et al (Nature, in revision), models with a 45 degree impact angle and a 2230km diameter impactor match the observations. There is precedence for this type of event however – both Earth and Mercury seem to have been hit by objects that were similarly large fractions of the planets mass.
It really looks like the north – south hemisphere dichotomy in crustal thickness is the result of an ancient impact, and the boundary is an isostatic, ellipticalboundary “separating two provinces of distinctly different curstal thickness, clearly preserved to the present day.” While endogenic models are killed off, it is much harder to require this was created by endogenic effects.
So, I think it is safe to say, in the early days of the solar system, Mars got womped in a really cool way.
At the end of the talk someone asked (huge paraphrase): Why isn’t their a resultant moon? Look at the Earth-Moon system as an example. Why didn’t the same thing happen at Mars?
Andrews-Hanna: Formation of Moon not inevitable. Mercury may also have had the same [type of impact event] happen, and in fact and in models a moon isn’t produced. Moon lack of a moon being inevitable as a problem for people originally trying to model production of the moon. This isn’t a problem.