2009-02-06

Spirit Sol 34

Today we get to do real work. The scientists are very happy. Last night they ran the IDD sequences I'd built a couple of days before -- the ones that didn't run because the sun-find failed -- or some variation on them. They worked splendidly, and produced surprising scientific results (always the best kind).

So the scientists are excited. The RAT lightly brushed Adirondack just as desired, revealing a very dark surface ("We've struck oil," Art Thompson announces). This is a surprise because it means the surface of Adirondack was likely very dusty, and it wasn't supposed to be -- they chose a location high on the rock face specifically so that the dust layer would be minimal. It also seems that the rock layer below crumbled very easily, another surprise because, given its chemical composition, we expected it to be very hard. Its behavior makes it much like sandstone, only softer -- it offered almost no resistance. (Must be French.) Steve Gorevan, the RAT PUL, says the rock -- or at least its outer covering -- may have the compressiveness of Dover chalk, maybe less.




Courtesy NASA/JPL-Caltech. The RAT-brushed surface of Adirondack, as seen by the PANCAM.



As a result, someone asks if the next RAT can grind as deeply as possible into the rock. This would help us decide whether the crumbliness is just an outer layer or applies all the way through Adirondack. Gorevan says that that's a natural impulse but would be a bad idea: overly aggressive RATting can cause an IDD fault, as attempting to drill into a hard rock can end up skipping the IDD off to the side. So you have to take it in stages, and we won't be at this rock that long.

Scott McLennan presents some fascinating results on comparisons between the Gusev and Meridiani soils. (There I go again. Dirt is not fascinating, dummy. Except, well, maybe sometimes it is.) "You might want to get your 3-D glasses out," he warns, and people do. Where are my five-dollar 3-D glasses? In my office. I grab a handy pair of the cheap ones.

Anyway, it turns out that both sites show significant microtopography -- look at it close up, and the sand has structure. Gusev's microtopography might be a little stiffer: when we placed the MB at Gusev, the microtopography retained its shape; at Meridiani, it collapsed. Later, he shows a very cool picture of the post-MB soil at Meridiani: the MB left an imprint on the sand, clearly visible in the subsequent microscopic image -- you can even see a bump in the center that corresponds to an indentation in the instrument. Meridiani's soil, unlike Gusev's, also has a bimodal distribution of grain sizes: a few large grains (the ones that looked like pebbles in the early MI image) and a lot of smaller ones that are like the ones we see at Gusev.




Courtesy NASA/JPL-Caltech. Opportunity MI image showing part of the MB noseprint.



Then we find out why we needed the 3-D glasses. His next slide is a cool 3-D image of the soil, showing vertical tubes and honeycomb-like structures. He points out multiple generations of growth along the tubes and a cross-cutting relationship between them. There's clearly some kind of cementing agent in the soil.

What does this tell them? The broadly similar soil compositions and microstructures suggest a global process. Their working hypothesis is that the incipient weak layer at Meridiani and the duricrust observed at the Viking site are likely points along a continuum of the same global process. There are already competing theories of this process -- one that works top-down, and one that's bottom-up. We might be able to pick a winner by using the rover to dig through the upper layer, as Jeff and Mark were doing last night.

Because the sequences ended with an overnight tool change, we don't have telemetry indicating the final state of the rover -- and we won't have it for hours yet, which makes planning the next day's activities more difficult. Jeng hacks up something reasonable and we hope for the best.

It turns out that most of the work we did in the testbed last night was for nothing. Oh, well, I still wouldn't trade it for anything; it's more funny than frustrating. We spent nearly all of the time testing a sequence that would take us away from White Boat, but the scientists have decided they want to go to White Boat after all. The broad outlines of the drive are known: we'll MI Adirondack one last time, stow the IDD, back up to MTES the RAT hole(s), back up further to image the area we've been sitting over for two weeks (to see what happened to the dust under the rover during that time), then drive to White Boat much as we've already planned. But at least Mark and I did remember to test the White Boat drive just before finishing for the night, so we can say it's been run in the testbed -- except for the minor changes they've requested today.

The SOWG meeting starts reasonably well. We have a lot of time for science, and effectively no limit on data, since the flash was cleared out when it was reformatted. But we're somewhat constrained on energy, since today's RATting will eat a lot of what's available.

The SOWG meeting gets more contentious than any I've seen before. There's a move afoot to skip imaging the RAT magnets before grinding, since, once everything else is taken into account, this will take more time and energy than we can easily allocate. In order to image the RAT magnets clearly, we'll need to move the IDD to bring them into view. And we can't just wait and do it later, because the main point of the RAT magnets was to see what accumulates on them between the first brush (which happened today) and the first grind (which is up tomorrow), then again after the first grind. If we don't get the between-brush-and-grind image, one of the normally placid scientists complains, we've missed almost the whole point of sending their experiment to Mars. For him and his team, it will be four years of work, mostly down the tubes. To make things worse, they're an international partner -- and nice people.

But it looks like they're going to be a victim of the pressure to move on. The SOWG chair is calling for a vote, and it's going to go against the RAT magnet guys. But while they're arguing, I'm working furiously in RSVP, and I find a way they can get the image without any extra IDD motion -- we just have to stop in the middle of a motion we have to do anyway, and snap a photo. As it turns out, when I look at the shadowing, it's even better than the image they were hoping to get originally -- the RAT will be perfectly situated in sunlight and pointed nearly dead-on at the camera. I grab a microphone, have Jeff put RSVP up on the projection screen, and show them.

Scott saves the day. They even applaud me.

I'm helping!

After the sequences I built the other day, these are cake. This time I make it my goal to finish an hour earlier than last time -- to be done by the time my RP-2 comes in. It's aggressive, but except for some information I have to wait for other people to supply, I make it. As always, there will be cleanup work to do, but the less time the RP-2 has to spend finishing the sequencing, the more time he can spend checking my work, and that's a good thing.

So I'm done plenty early, but I end up sticking around late anyway. At first it's just to get my daily pix fix -- I'm two days behind in images from both rovers -- but a better reason shows itself just as my shift is about to end. The scientists have now started giving us "midnightly" science briefings, similar to the lunchtime briefings the SMSA has been getting for weeks. I missed the first one, last night, since I didn't know about it (and I was in the testbed all night anyway), but that's a mistake I won't repeat. Tonight's briefing is from Matt Golombek, who talks about the implications of the RAT results from today. He points out the sandblasting effect you see on Adirondack's face, the little scallop-shaped chunks, which tell you a lot about the soil: its grains are sand-sized (small enough to be blown around by the wind, large enough that they're not just lifted up into the air like dust, and cohesive enough that they'll take tiny chunks out of the rock instead of just breaking up harmlessly when they hit it).

And I learn a couple of rules of thumb about secondary impact craters such as Bonneville: they're about 0.1 as deep as they are wide, and their rims are about 0.02 of the diameter. Since Bonneville is 150m across, it should be about 15m deep, and its rims about 3m above the surrounding terrain. Bonneville is also filled with white sand dunes; one of the reasons the scientists want to go there is to find out what they're made of. We don't think they could be quartz (white sand on Earth is quartz), but we don't know why they're white. They might be the same stuff as White Boat, but we'll have to go there to find out.

Incidentally, I also learn something interesting about sand dunes: when you see one, you know it's made of sand. Which is more newsworthy than it seems. They're never made of dust (particles smaller than sand) because that will blow away quickly, and they're never made of particles larger than sand, because those weather relatively rapidly as well. The process of sand dune formation depends almost entirely on frictional forces that apply at sand-sized scales only (0.6mm to 2mm, I think; Matt's rule of thumb is "about 1mm").

So that's all groovy, and I'm happy and about to go home so that I can come back a couple of hours early for a meeting, when I remember something important. I was supposed to set up a navigation memory parameter for the drive, so that Mark can accumulate more data for the drive to White Boat. For complicated reasons, this needs to be done a day early -- today. And I completely forgot about it -- I made a note, but I forgot to look through my notes as I usually do. D'oh! It wouldn't be a disaster if it didn't happen, but Mark's already not getting the drive he really wants; I'd hate to disappoint him further. Fortunately, he and I (read "he") had already made a canned sequence to set everything up the way he wants it, so I break that out, deliver it, and head home.

No comments: