In which I finally write up last week's Deep Impact Hartley 2 press briefing

On Thursday, November 4, at 13:50 UTC, Deep Impact flew within 700 kilometers of comet Hartley 2. Hartley 2 is the smallest and most active of the five comets that have been directly by a spacecraft, and the first to be visited within the lifetime of its discoverer. The Deep Impact spacecraft performed flawlessly, with no errors in its imaging, and missing its aimpoint in time and space by only two seconds and three kilometers, respectively. (This is exceptionally good.)

The pictures from the closest approach revealed a lumpy, clumpy, two-lobed nucleus that invited comparison to a dog bone, a fried chicken drumstick, and a "shaved poodle." More seriously, it was described as looking like Itokawa with the jets of comet Halley. It won't win any beauty contests, but the members of the mission team who talked about the encounter at a press conference on Thursday afternoon were thrilled by its appearance and by how well their spacecraft performed, and continues performing.

The panel was opened by Ed Weiler, representing NASA Headquarters. I usually don't find a press panel's introductory comments to be worth writing down, but Weiler had several good points. He said he felt a particular fondness for Deep Impact because he was the administrator who'd selected it to go forward as a Discovery mission a decade ago. "It was going to carry a copper bullet and slam into a comet," he said. "Somehow this survived peer review." He got a chuckle from the audience. He also claimed that "for about 10% the cost of the Discovery mission, we got a second Discovery mission."

Now, this statement might be easy to dismiss, but I think he's right. Few people would argue with the fact that it's good to use spacecraft that are already in space and finished with their primary missions to go on to perform extended missions. But Deep Impact's first mission not only flew past a comet, it slammed an impactor into the comet and studied its aftermath. How could you call the Deep Impact flyby of Hartley 2 a whole 'nother Discovery mission if it was only a flyby?

In flying by Hartley 2, Deep Impact did a couple of things that it didn't do at Tempel 1. The main difference between the two encounters is that they were able to take much more data at Hartley 2 near the closest approach. Here's why. The mission didn't know what to expect in terms of the environmental hazards at Tempel 1, so they had to be very conservative, turning the spacecraft to a safe "shield" orientation for half an hour around closest approach, an orientation that prevented Deep Impact from capturing images at that critical time.

But they learned through analysis of the Tempel 1 data -- in particular, the data captured by the impactor on its way down to its crash -- that they needn't have been so conservative. So, as a direct result of what they learned at Tempel 1, they were able to place far fewer restrictions on what the spacecraft was allowed to do at Hartley 2, and were able to take images throughout the flyby. Not only did this result in higher-resolution pictures, but it also resulted in the kind of data they need to produce a really detailed model of the shape of the comet with far more fidelity than they can produce for Tempel 1.

They also were able to send their data back to Earth at a much higher rate during the Hartley 2 encounter than they were able to at Tempel 1. That translates directly into much higher time resolution on their data. The Deep Impact spacecraft has rather limited onboard data storage; it could only take data for so long before having to relay it back to Earth. So there were repeated cycles of take data -- turn to Earth to relay -- take data -- turn to Earth for relay. During the flyby, the orientation of the comet and camera meant that the high-gain antenna couldn't be pointed to Earth; they didn't regain Earth pointing until about twenty minutes after the flyby. At that point, they were able to take data and relay it to Earth simultaneously. They used a "leaky bucket" approach, draining data from the recorder, and erasing it, at the same time as they were writing new data to the recorder. They kept their onboard storage nearly full for days after the flyby, still taking data as they continued draining it; for all I know, the recorder may still be nearly full. Because of all of this, the rate that they can communicate the data back to Earth is very important -- the faster you can drain the bucket, the higher a rate you can continue to refill it. Hartley 2 was closer to Earth during the encounter than Tempel 1 was, so Deep Impact's transmitter could send the data back about twice as fast this time around; and, moreover, they were able to test their downlink rates during their exoplanet observation phase and determine the very highest data rate that they'd get away with during the encounter.

OK, enough preliminaries, on to the actual results from the encounter. It was a very happy Mike A'Hearn and Jessica Sunshine who presented their very, very preliminary remarks on the science results of the flyby. They're happy because the results were great, but also, I think, because there had been significantly less stress in the days leading up to this extended-mission, bonus encounter than there was with the seemingly ridiculous (but ultimately successful) hit-a-comet-with-a-bullet scheme of the primary mission. Mike spoke about the observations that they made as they approached the flyby, how they watched the jets burst out from this very active comet. He showed how the amount of carbon dioxide that was jetting out of the comet varied hugely with time:

Mike said that the variation with time very likely means that there's one particular region on the comet that is enriched in carbon dioxide ice and is actively jetting, and that every time it rotates into the Sun, the outburst flares up. But Hartley 2 has a complex rotational state, where it's likely spinning around one axis and rolling about a secondary axis, so that each time around, not quite the same spot on the comet has the Sun overhead; this explains why there isn't the same amount of carbon dioxide bursting out every time around. One of the press asked where in this cycle Hartley 2 was at the moment of closest approach, and Mike said they weren't certain yet, but that he was pretty sure that the area that is most sunlit in the closeup images and also jetting most violently is, in fact, the area that is responsible for these puffs of carbon dioxide. I asked them where the rotational pole was in the closeup images, and both Mike and Jessica laughed and replied "if you can figure that out, please tell us."

Jessica began her part of the presentation by cracking a joke that I wasn't able to write down; but what she said, paraphrased, is that she hardly knew what to say about this amazing looking comet, but it's a press briefing and she had to say something, so here goes. She spoke about what they saw in the comet at closest approach. She said that while Hartley 2 is the smallest comet that's yet been visited, it's the most interesting she's seen, and for its size, the most active. "The dominant geologic signature is that you have two rough ends and a smooth middle. Where you see the jet activity is the rough areas." She said that the rough area on the sunlit end is where "we are fairly confident is the end of the nucleus where the dominant carbon dioxide and dust is coming from. There are also jets on the other end. In many cases jets seem to be correlated to specific topographic features.

"In the middle we see fine-grained material that has been redistributed and collected in a topographic low. We also see places where we see clumps of material. We think that those might be remnants of activity." She showed the image of the night side of the comet, stretched to reveal the jets sprouting out from the darkness and silhouetting the limb of the nucleus against the sky. "Not only do we have major activity toward the bottom, but along the terminator there is a line of jets so that we can see the silhouette of the nucleus. We have jets in the nighttime, we have jets along the edge, and we have jets in the Sun. This is just spectacular. This is probably a good time to admit that we have a lot of work to do" to understand just what is going on.

She told me after the press briefing that after they had finished writing the sequence of commands that they planned to send to the spacecraft, instructing it to downlink those five precious close-approach images in the minutes right after the flyby, she realized unhappily that she'd forgotten to include with the downlink of those first five pictures a longer-exposure image to show the comet's jets, as had been done with Tempel 1. She figured that on the day of the encounter she'd be disappointed not to be able to see the comet's jets, that it would just look like an asteroid. So Jessica was euphoric to see the jets in those five images, while still being able to see all that detail on the surface of the nucleus. It's really the first time, she said, that she's been able to see in a single exposure jets that go all the way to the surface, and yet see variability on the comet's surface at the same time.

I asked her about those clumps and mentioned that they reminded me of the boulders on Itokawa. Jessica's reply was cautious. She pointed out that the processes that are thought to result in the surface of Itokawa being covered with boulders -- which largely have to do with impacts -- are not thought to be the same as the processes that shape Hartley 2's surface. Hartley 2's surface probably results mostly from its intense outburst activity. So she is not calling the things on the surface "boulders," because there's no way of knowing if they are internally cohesive; she's sticking with the term "clumps." She said they were "quite a surprise that we weren't expecting to see. We didn't see them on Tempel for example. And there are surprises are yet to come in compositional side."

There is so much yet to come from this data set. In fact, the spacecraft is still collecting data, and will continue to do so for another couple of weeks. In total, there will be 120,000 images from the three main instruments. There is an awful lot going on at Hartley 2. There are things that are familiar -- its shape looks like Borrelly and Halley and even like several other bodies that have been imaged by radar. But as a package, it's a totally new thing.

asked what is going to become of this comet -- how long can it possibly last, jetting away so actively? Mike cautioned that " We don't know how comets die, so we don't know how long it will last," but he said that it loses roughly 1 to 1.5 meters' worth of material with every passage close to the Sun, and it's only about 500 meters across to start with. Any way you look at it it's just a few thousand years until this thing should be used up.

What's next for Deep Impact? Sadly, it does not have enough fuel left on board (only about 4 kilograms remain) to perform the large trajectory correction burn that would be necessary to send it to a third comet. But that amount of fuel is plenty for pointing the spacecraft and maintaining communications with Earth from its current orbit for years to come. Since the spacecraft is in such good shape, I think it's quite likely that we'll hear of another mission extension similar to the extrasolar planet follow-up studies that Deep Impact performed while it was cruising to the Hartley 2 encounter.

A final note: I just took the 72 images that were released in the hours after the flyby last Thursday and assembled them onto an easy-to-browse index page.