The 4th MSL Landing Site Workshop: Day 3 – Final Discussion

We wrapped up the landing site workshop on wednesday afternoon by revisiting each of the four sites and discussing them in turn. Unfortunately, the way that we did this was very disappointing, and made for a frustrating afternoon.

The discussion was centered around a word document that was projected up on the screen in the room. Over lunch, the meeting leaders had conferred and listed what they thought were the key points for each site that everyone agreed with, along with areas of future work. But they should have known that getting a room of 200 scientists to agree with something is nearly impossible, and it didn’t help that their list of points was a mish-mash of actual, irrefutable observations (e.g.  “there are strong clay mineral signatures at Mawrth”, or “there is a huge layered mound at Gale”) and complete speculations (e.g. “the delta at Eberswalde preserves organic molecules” or “the layered rocks at Holden are likely lake deposits”) and everything in between.

The result was that we spent most of the afternoon going through this document line by line and debating word choices and phrasing and grammar. It was incredibly frustrating, to the point where I almost had to leave the room. By the end, there were probably only five or so people actively participating, while everyone else in the room silently suffered through watching this document being written by committee.

After the discussion, we did end up with a central hypothesis for each of the sites, along with a list of future work.Gale was the first site we discussed, so a lot of the time was spent figuring out how we were going to do this document editing as a group. I also was on the edge of my seat because my adviser was off the the side telling me that I should chime in. I always get stressed out when I’m expected to comment or ask a question when I don’t really have a comment or question to make. In the end I asked about the wording of one of the bullets that said Gale has a “relative paucity of fluvial channels”. Because it actually has tons of them. But apparently “relative paucity” meant “there is more volume in the mound than could have come from those channels” so the sentence was not edited. Those two things mean different things to me, but I didn’t press the point.

Anyway, all of that meant that I didn’t take careful notes for Gale, but the final hypothesis was something along the lines of “how do the environmental and mineralogical signatures preserved in the layered rocks reflect the aqueous processes in the crater?” Although someone else suggested a more concise and to-the-point version: “Did the big-ass mound form in a lake?” Future work for Gale included getting a better handle on the age of the deposits and identifying specifically where MSL would go to look for organics.

For Mawrth the process was a little smoother, but still painful. I don’t know how professional scientists can have so much trouble stringing together a coherent set of words. in the end, the hypothesis they came up with was: “Mawrth records geologic processes during early martian history when  aqueous phyllosilicate-forming processes were pervasive and persistent and provides the opportunity to understand early habitability.”

Future work for Mawrth was to figure out the timing of the formation of the stratigraphy and the mineralogy, and just generally getting a better grasp on the possible depositional settings. There was also the question of what MSL would do on an extended mission, and Jean-Pierre Bibring called for astrobiologists to take a look at the site and try to fill in that part of the story.

By the time we got to Holden, the process for the discussion was in place, and the site advocates had had time to write down a hypothesis. unfortunately, it was wordy and awkward and immediately had to be edited, but in the end it boiled down to: “Holden preserves evidence of a fluvial-lacustrine system that provides the opportunity to apply a systems approach to evaluating a sustained, habitable environment.”

For Holden, the future work was to see if we can learn anything more about the light-toned layered deposits to see if they were formed in a lake. Also, looking at the orientation of the layers in the site was mentioned specifically.

Finally, Eberswalde benefitted from watching all of the other sites go first, and the fact that it has the clearest hypothesis to test of any of the sites. Sanjeev Gupta delivered this hypothesis: “Eberswalde crater’s stratigraphy, geomorphology and mineralogy records the evolution of a crater lake and associated fluvio-deltaic systems and additionally represents a habitable sedimentary environment that is favorable to the preservation of organics.” This one was clear enough and well worded enough that people actually applauded when they realized we wouldn’t have to agonize over the wording.

I don’t have notes for future work for Eberswalde, but I am pretty sure it included looking more carefully at the minerals seen from orbit and identifying specifically where to go to access the “bottomset” beds, which should have the best chance of preserving organics (the Eberswalde presenters repeatedly answered this question – you go the the bottom of any preserved stratigraphy – but people seemed determined not to hear this answer)

There was also a repeated call for all of the sites to identify specific targets for MSL to investigate, which can be fed into the rover drivers’ attempts at estimating how long it will take to get to the primary targets of each site.

So, that wraps up the 4th MSL landing site workshop. We didn’t vote any of the sites “off the island” this time, but I can tell you the impression that I got. I think Eberswalde is the clear leader among the four sites because, if you’re going to look for organics on Mars, it is the only place where we can be confident that there was a low-energy depositional environment that would concentrate and preserve organics. Based on the vibe in the room, I think Mawrth would come in second place if we had voted. Even though we have no idea how the rocks there formed, they have the undeniable advantage that you land on your primary target. Also, their team has always been a very cohesive and outspoken group at these meetings. They can be annoying in their fanaticism over their site, but I think their disciplined messaging may have had some influence.

I think Gale crater would come in just behind Mawrth. It has a lot of great qualities: including clear detections of both phyllosilicates and clays, and a very thick stratigraphic sequence that would tell you about a lot of environments. But Gale is a go-to site, and even though I showed a lot of cool stuff in the ellipse, people are a bit uncomfortable with the go-to sites. Gale particularly worries people because driving up a big mound sounds risky. We saw that based on slopes, there looks to be a clear path up the mound, but i wouldn’t be surprised it the rover drivers end up telling us it’s going to be extremely slow to climb it. The other issue is that we can’t say for sure what the mound is made of. It might have formed in a lake, but then again it might not have. This is true for all the sites except Eberswalde, which is pretty clearly a delta, but it seems to be emphasized most for Gale.

Finally, I think Holden would come in last. The Holden presenters did a great job highlighting all the good science that can be done on the alluvial fans in Holden, and with the go-to targets. But Gale has many similar features, only more-so. At both sites you land on an alluvial fan and then drive to layered rocks. Except the rocks at Holden are 100s of meters thick and the rocks at Gale are 1000s of meters thick. Holden does have nice examples of pre-impact breccia as well as the flood deposits, neither of which you would get to see at Gale, but those targets are toward the end of a long traverse at Holden, and may not tell us much about habitability.

So, that’s how I think the voting would go if we had voted at this meeting. My own ranked order shuffled around throughout the meeting: each site made a very convincing argument, so tended to jump higher on my list until the next site presented its case. Now, with a few days of perspective, I would rank the sites in this order: 1. Eberswalde, 2. Gale, 3. Mawrth, 4. Holden. I personally put Gale ahead of Mawrth because even though we don’t know the depositional setting for either site, at Gale there is enough stratigraphy that you get multiple chances to find a habitable environment.

Of course, there are caveats to this. If Eberswalde really did form from melted runoff caused by the Holden impact, that might knock it down on the list. If we can figure out the depositional environment at Mawrth and it’s not just impact ejecta, that might bump it up in the ranking.

The bottom line coming out of this meeting is that there is still a lot we can learn about all of  these sites. Some of it can only be learned by a rover, but some of it just requires a deeper look at the data we already have. MSL is a phenomenally powerful (and phenomenally expensive) machine, and we only get one. I hope this meeting gets people digging deeper to learn about these sites so that we send MSL to the best place possible, whatever that may be.

Advertisement

The 4th MSL Landing Site Workshop: Day 2 – Holden Crater

The drainage system that Holden and Eberswalde interrupted. Image excerpted from Ross Irwin's LSWG presentation.

The second site that we discussed yesterday was Holden Crater. Ross Irwin gave the first, overview presentation. Holden is a 155 kilometer crater that formed right in the middle of a huge drainage system that spans from the Argyre basin to the northern plains, and at Holden you would land on a bunch of coalescing alluvial fans on the western crater floor and then drive southeast to access some nice phyllosilicate-bearing light-toned layers. Irwin emphasized that even though the light-toned layers are a key target, Holden is all about visiting a bunch of diverse water-related settings. He also made the interesting argument that MSL is likely our only chance, at least within the next decade or two, of sending a plutonium-powered rover to Mars that is capable of landing at a site as far south as Holden and Eberswalde, whereas Gale or Mawrth could be accessed by a later, solar-powered rover. Irwin also gave extremely detailed lists of what science MSL could do at each stop along the traverse. These sorts of lists are exactly what we need to make for all the sites because they really get us thinking in detail about MSL’s capabilities and what we would learn.

Next up was Kelin Whipple, who is an expert on alluvial fans here on Earth. Kelin described in more detail what fans can tell you about the climate when they were deposited (a lot!) and showed that the fans at Holden could be made by mud flows or water flows and that either way they would be really interesting.

After that, James Wray, a fellow Cornell grad student and Goddard ’06 NASA Academy alum, gave a presentation about the mineralogy at Holden. He particularly looked at some of the rocks in the walls of the crater, which would be present in the materials of the fans in the landing site. James found a pretty diverse set of minerals, including two types of olivine, two types of pyroxene, some phyllosilicates and some possible salts or high-silica minerals. James also pointed out that there are some mounds of material in the eastern ellipse which may be outcrops of pre-impact material.

A topo map of Holden crater. MSL would land on the gently sloping alluvial fans in the western portion of the crater and the drive southeast to study light-toned layers, deposits from the flood that breached the crater rim, and ancient bedrock.

Following James’ presentation, Debra Buczkowski discussed the results of work with Kim Seelos studying what looks like a really extensive layer of phyllosilicate-bearing rock just below the surface  in the region to the northwest of Holden and Eberswalde. They found that the clays in Holden that appeared to be in-place were different from those in this extensive layer, but that some of the Holden clays that looked like they had been transported might be the same stuff.

Finally Peter Buhler gave a presentation of his work on a small trough to the southeast of Holden which may also have been a system of lakes, and emphasized that Holden and Eberswalde could give us a “contextualized” hydrology of the broader region.

The discussion started off with a question from Dawn Sumner about whether the alluvial fans on the wall are interbedded with the light-toned layers, or whether the fans are just on top of the light toned stuff. This is a tough question to answer because as alluvial fans form, they advance outward into the basin, hiding everything underneath. Next my adviser, Jim Bell, asked whether all the great science described for the Holden fans could also be done on the fan at Gale Crater, to which Kelin Whipple said “Yes!” although he pointed out that it’s hard to tie the fans to the stratigraphy of the mound at Gale. Another question about Holden was what the compositional variation of the fan material would be: in other words,  how far would we have to drive to get samples of different stuff? The short answer was: we don’t know until we get there. But James showed that there was a variety of rock types up in the crater wall. Whipple explained that if the fan was made by mud flows, which tend to come from a very localized source, they each layer in the fan would be different, whereas if it was formed by a more watery runoff flow then it would have a more blended composition.

After hearing the presentations for Holden I felt a lot better about it as a site than I had before, but I still have the nagging thought that it’s very similar to Gale: you land on alluvial fans, then traverse to visit nice layered stratigraphy, except at Gale the stratigraphy is much more diverse and much thicker than at Holden. If I had to guess, based on the vibe in the room, I would guess that Holden and Gale are the bottom two sites in the ranking. Part of that may have been the timing of the presentations for Holden (after lunch lull) so I’ll see how it looks after the more detailed discussion today. As for my own personal ranking of the sites, it has been jumping all over the place as I hear the presentations from each group. I’m very interested to hear what the engineers tell us today, and to see how the giant discussion this afternoon plays out.

The 4th MSL Landing Site Workshop

Well folks, I’m off to Pasadena to help the Mars community decide where to send its next rover. Long-time readers will recall that i’ve been to a couple of these things before and they’re always fascinating. I was going to post a reminder of what the four finalist sites are, with pros and cons and all that, but it turns out I don’t have to! My friend Lisa Grossman, a former Cornell astronomy major, is now a science writer for Wired! She interviewed me and my adviser earlier this week and put together a nice article summarizing the sites. I’m quoted in it quite a bit, so rather than repeat the same stuff, I’ll just point you over to her piece.

There are a few points of clarification that I should mention. First, the article says that MSL is searching for life, and that’s not really true. MSL is searching for signs of habitability. Obviously finding life would be a pretty good sign! But habitability is broader than just the search for life or even the search for organic molecules. Evidence for habitability could come from the texture of a certain rock telling us that it was deposited in water, or from the detailed chemistry revealing that the minerals in the rock could only form in benign liquid water.

Also, she’s right that some of the clays at Mawrth are kaolinites, which tend to form on earth in tropical soils. But to clarify, I don’t think anyone is saying that the huge amount of kaolinite clays at Mawrth are the result of tropical conditions. They do suggest that there was a lot of water involved though, which is why Mawrth is so interesting.

A final clarification: in the article, it mentions that it will take “several days of hard driving” to get to some of the go-to sites, where the really interesting stuff is outside the ellipse. If it were several days, that would be no big deal. It is going to be more like a year or two. A lot of people are really nervous about landing, only to have to buckle down and drive drive drive to get to the main target of the mission. Of course, all of the sites have good science to do in the landing ellipse, but that is a blessing and a curse for a go-to site. On the one hand, you get some results early on in the mission, but if you don’t have a lot of discipline, you can spend all your time staring at the rocks at your feet and never get to go climb the mountains in the distance.

With that, I’ll let you go read the article. You can also check out my old blog posts about the sites from the last time one of these workshops was held (Gale, Holden, Mawrth and Eberswalde). I’m going to do my best to take notes and blog about the meeting, and Emily Lakdawalla of the Planetary Society will be there for part of the meeting as well. We’ll do our best to keep you informed!

PS – You should totally check out the comments on the Wired article, where someone calls me out for saying that there was water on Mars and says that we Mars scientists are either stupid or have ulterior motives. Someone hasn’t been paying attention to every press release about mars for the past decade or so.

LPSC 2010 – Day 4: Mars Oceans, Titan Lakes, Astrobiology and Asteroids

Thursday started off with a couple of talks about the possibility of oceans on Mars. The first one, given by Gaetano DiAchille looked at possible locations of deltas all over Mars to try to figure out the water level of a past ocean. Deltas form when a river hits a standing body of water and drops its sediment, so they are a reliable marker of the water level. DiAchille found that “open deltas” – that is, deltas that do not end in a closed basin like a crater, all appear at the same elevation. This might mean that they all fed into a large northern ocean.

A map of valley network density on Mars and the possible extent of a northern ocean.

In the second talk, Wei Luo described his work mapping where all of the valley networks on Mars are and found that the northern limit of the networks fits with elevations that had previously been considered as possible ocean shorelines. The valley networks also matched with locations that atmospheric models predict would get the most precipitation.

Neither of these studies is conclusive evidence for a northern ocean on Mars, but they are interesting and they suggest that the “ocean hypothesis” is becoming popular again after years of little interest.

Later that day I saw a talk by Nick Warner describing the possible thermokarst lakes that he discovered in Ares Vallis on Mars. I wrote an article on Universe Today about this discovery when it was first announced a couple months ago.

I ducked out of the Mars talks to go see a talk by my friend Debra Hurwitz about a lava channel in a crater in Elysium Planitia. The channel was formed when lava breached the rim of the crater, flowed down the inner wall and ponded in the bottom. She calculated that the lava probably flowed at about 17-35 meters per second and that 6,000 cubic meters per second flowed down the channel for about 15 days. She also found that the channel could have been eroded mechanically without the need for the lava to actually melt the underlying rock very much.

A sketch of the lava channel filling the crater in Elysium Planitia.

After that, I headed over to the Titan session to hear a talk by Ralph Lorenz about waves on Titan lakes. Most of what we know about the surface of Titan, including the presence of liquid hydrocarbon lakes, is based on radar images from Cassini that measure roughness. The lakes show up as perfectly smooth (and therefore dark) surfaces, which is weird because radar images of lakes on earth usually have slight roughness due to waves. On Titan the gravity is lower, so you would expect bigger waves. It’s possible the lack of waves is due to the viscosity of the lakes, which might be increased by bigger “tar-like” molecules dissolved in the thinner ethane and methane, but it might also be due to a lack of wind. The Cassini mission will be watching as the seasons at Titan change to see if the wind changes and kicks up any waves.

A (suggestively colored) radar map of lakes on Titan.

I did a lot of session hopping on Thursday! The next stop was the astrobiology session. Oleg Abramov presented some results of his investigation of what intense impacts might have done to early life on the earth or Mars. He found that even during the Late Heavy Bombardment, the crust is not sterilized by the impacts, and in fact it might be more habitable for early life because impacts deliver organic molecules and cause widespread hydrothermal activity!

The talks I was really interested in were two talks on the magnetite crystals discovered in the famous ALH84001 meteorite. I posted a while back about a new paper that claims these crystals are evidence of life on Mars, and these two talks were focused on the claim. The first talk, by Allan Treiman gave some good background on the debate over whether ALH84001 preserves evidence of life and then addressed some of the new claims about the magnetite crystals. He said that most of the attributes of biological magnetite crystals, such as their size, lack of flaws, and precise crystal structure were not observed in the ALH84001 crystals. The big question is why the crystals are so pure. Allan argued that you can get pure crystals just from the heating of iron carbonate, which is found in the meteorite.

The following talk was by Kathy Thomas-Kleptra, whose paper Treiman was responding to. She showed that Treiman had probably made an error in calculating the breakdown temperature for iron carbonate. She also pointed out that the crystals are found in carbonates without much iron and that there is no graphite observed, but it is also a byproduct of heating the carbonates.

I don’t know enough about petrology and geochemistry to know who is right here, and I was very disappointed that both Kathy and Allan used up all of their time talking, so there was no chance at all for questions! I wasn’t the only one. When the moderator said that there was not time for questions and that they had to get on with the next session, most of the room groaned and protested. But alas, the talks pressed onward.

Biogenic magnetite crystals inside a bacterium one Earth.

I zipped back over to the Titan talks in time to catch the end of one pointing to features that they claimed were “deltas” in one of the lakes. I was very skeptical of this because the quality of the radar images is so low. What they avtuall observe is a dark branching channel that ends at a peninsula in one of the lakes. That’s not evidence for a delta in my book. This talk made me realize how spoiled I am with HiRISE, CTX, MOC and other high-resolution data on Mars!

Finally, I stopped by the asteroid session for two talks. The first was by Dan Scheeres and he talked about the role that tiny forces might play in holding asteroids together. He showed that Van Der Waals forces, normally ignored for all but the tiniest particles, actually might be important in holding particles together in asteroids. He made the analogy to powders like flour or cocoa powder on earth. These can clump together and when they are stressed the form fractures even though they are made of loos grains. The same thing might happen on a much bigger scale with the gravel and boulders in low-gravity asteroids!

It's possible that the fractures in objects like Phobos are more like the cracks you see in flour than like cracks in a solid, fractured rock.

The last talk I caught on Thursday was by my friend Seth Jacobson, who showed some simulations of asteroids that spin so fast they break apart. He showed that the ratio of sizes between the two bodies make a big difference in how the binary asteroid evolves. In some cases, the secondary asteroid even swings so close to the primary that it splats apart and forms a short-lived three-body system!

LCROSS preliminary results

Hey remember when we bombed the moon? Here’s an interesting article about some preliminary results from LCROSS. I was especially surprised when they said that there may be mercury at the impact site. They say they’re seeing spectral lines that could be produced by iron, magnesium or mercury, but then the article goes on as if mercury is the likeliest candidate! I’m skeptical. Fe and Mg are common in lunar rocks. Mercury: not so much. Oh well, it’s an interesting update anyway, and it sounds like the real juicy results are still in the works.

PS – Have you voted today for my MSL: Mars Action Hero article over at the scientificblogging.com science writing competition? Remember, you can vote daily!

To the Moon! Zoom, Bang!

As I write this, there is a NASA spacecraft on an unstoppable collision course with the moon.

Early on Friday morning it will impact a crater near the moon’s south pole at 9000 km/hr, causing an explosion that will excavate 350 tons of lunar rocks, blasting them up into space and leaving a 66 foot-wide crater.

Of course, this is all intentional. The LCROSS mission will use the upper stage of the rocket that launched the Lunar Reconnaissance Orbiter to the moon as a missile to blast the possibly ice-bearing crater Cabeus in much the same way that the Deep Impact mission blasted a hole in comet 9P/Temple. The hope is that the ejecta from the LCROSS impact will reveal that the crater does indeed contain ice.

In addition to the “shepherd” spacecraft that will follow the big Centaur stage to its explosive death, snapping pictures all the while, scientists all over the world will be watching the moon with their own telescopes. In fact, even large (12 in or 30 cm) amateur telescopes may be able to see the impact!

Even if you don’t have a big telescope to peer through, NASA TV will be covering the impact. The impact is planned to occur on Oct 9, 2009 at 11:31:30 UTC (04:31:30 am PDT), and NASA TV coverage will begin at 10:15 UTC (03:15 am PDT).

Stay tuned! It promises to be a blast! (Sorry, the pun had to be made.)

Big Picture: Mercury MESSENGER

If you’re not already following the Boston Globe’s Big Picture blog, you should be. They always have spectacular photos, often of current events, but also quite often of space-related stuff! Today’s post is about the MESSENGER mission to Mercury. Go check it out.

Impact Crater

In my posts about our field trip to Arizona, I showed my best pictures of meteor crater, but really none of them come close to expressing the feeling of standing on the brink of such a feature and trying to imagine an explosion big enough to carve it out. I just came across a photo by Stan Gaz that does a much better job than my snapshots (click to follow a link to a bigger version):

(Hat tip to Bad Astronomy)

Meteor Crater, Walnut Canyon, and Red Mountain

Meteor Crater is the best preserved (and the first recognized) impact crater on Earth.

(This is day 5 of a week-long planetary geology field trip to Arizona. Get caught up with days 1,2,3,4)

Today was a long and awesome day. We started out at meteor crater, the youngest and best preserved impact crater on Earth! Our guide today was Shaun Wright, a colleague from the Hawaii field workshop, among other places. He showed us infrared images of the crater taken from an airplane and we walked around the rim trying to identify the main compositions detected. Meteor crater is especially nice for this because it excavated into three distinct layers: the red Moenkopi siltstone (the surface of the surrounding plains), the yellowish Kaibab limestone (normally beneath the Moenkopi), and the white Coconino sandstone (below the Kaibab).

Back in the early 1900s, people were trying to dig and find the iron meteorite that they thought was buried under the crater. (it turns out the meteorite was blasted into thousands of pieces upon impact) Luckily, the mining work carved a notch in the rim that lets you see the three units of the crater where they have been overturned by the impact. When a large impact occurs, it lifts up the ground and forms an “overturned flap” at the rim. You can see in the picture that the Moenkopi goes from relatively solid-looking to very fractured-looking, and is then overlain by blocks of Kaibab and Coconino.

At the rim of the crater, the impact has reversed the sequence of layers. The red Moenkopi would normally be on top but here it is overlain by blocks of Kaibab limestone and Coconino sandstone that have been excavated by the impact.

Another very interesting part of the crater is that the impact pulverized the coconino sandstone, crushing the sand grans into powder. This powder was actually mined for a while because it is a very high grade silica “rock flour” used in things like makeup. Amazingly enough, even though it has been subjected to one of the most violent forces imagineable, the crushed sandstone still maintains its original structure, and you can even see crossbeds preserved!

The shocked sandstone still preserved very fine cross-bedded layers, but can be crumbled into a power with your hand.

After Meteor Crater, we made a short stop at Walnut Canyon, where the Coconino sandstone is not shocked and the crossbeds are displayed prominently. Remember, cross-bedded layers typically form when sand dunes are lithified in place and turned into sand stone, preserving the layers within the dune. For  more info about crossbeds, check the USGS site about them.

Crossbeds at Walnut canyon are essentially fossilized sand dunes from when Arizona was a coastal desert. The direction that the layers are tilted tells us that the prevailing winds blew from north to south, although the various sets of layers in this image actually reflect several wind directions.

Finally, after Walnut canyon we drove up to Red mountain, which is a cinder cone volcanoe that has been carved open by erosion. Not only does it give a great view of the interior structure of the cone, it also erodes into a very bizarre landscape that looks like it belongs in a Dr. Seuss book.

The interior of Red mountain cinder cone. The layers are from different stages of the eruption that deposited cinders with slightly different composition or weathering properites. The bizarre shapes are due entirely to erosion, mostly by water.

That’s all for today. Tomorrow we are off to Grand Falls and the nearby dune field!

The footprints of a moonlet’s demise…

A cool paper just came out in Icarus this week claiming that a crater in the northern plains of Mars may be the result of the impact of a small moonlet of Mars, possibly just smaller than Deimos. Here’s the crater:

(Chappelow and Herrick, 2008)

Because the crater is so elliptical, and because of the “blowouts” on the eastern edges of the craters (where ejecta actually blew back through the crater rim), it’s likely that these craters were formed by 2 impactors entering the atmosphere at a very shallow angle, traveling to the west. The impactor probably fractured into 2 pieces as it entered the upper atmosphere. Usually, these kind of impacts are attributed to comets or to asteroids that enter the atmosphere at an unusually shallow angle, but the small separation between the craters implies that the impactor wasn’t going fast enough to be an asteroid when it hit the atmosphere.

So, that means that the most likely impactor that formed these craters was a moonlet. One idea that’s been floated is that Mars used to have a good size population of captured asteroids in orbit as moonlets, but that Deimos and Phobos are the only ones that haven’t succumbed to tidal decay. At least, not yet – Phobos will probably impact the surface sometime in the next 50 million years, and might even leave behind it’s own set of footprints!

Correction: The blowouts are actually due to the impactor and debris travelling forward from the impact, which means that the impactor was travelling west to east. This is the same direction that Mars rotates, which would put the moonlet in a prograde orbit, like Phobos and Deimos. (Thanks JEC!)