		Site Map \| Contact Us \| Search:

JOIN

The Planetary Society Blog

By Emily Lakdawalla

LPSC: Wednesday afternoon: Deep Impact results

Mar. 17, 2006 | 16:03 PST | Mar. 18 00:03 UTC

We need your help.
Please donate to support our blog, website, and podcast.

Weblog Archive

My longest day at LPSC ended with a few talks on the composition and structure of Tempel 1 from the Deep Impact mission. Before I begin to summarize, I wanted to mention that I've been getting some mail complaining about the outcome of the Great Comet Crater Contest, in which we were forced to pick winners from among the large number of entrants who guessed a crater diameter between 100 and 250 meters. One commenter even suggested that the lack of knowledge of the crater size was "bad science" and that "based on the contest results, it was probably pure fortune rather than scientific and engineering foresight" that Deep Impact even managed to hit Tempel 1 at all!

Well, we're sorry to those of you who were so disappointed, but the fact is that if the results of planetary missions were perfectly predictable, there would be no point in sending the mission at all. True, Deep Impact's inability to see the crater was, in a sense, a failure of the mission, but that failure occurred in part because so little is known about the surfaces of comets -- the very point of launching Deep Impact in the first place! We certainly know a lot more about comets than we did before; and Deep Impact did accomplish its other objectives. This sort of partial success is hardly uncommon. When I get to it, I'll be writing up the first science results from Hayabusa -- a mission for which it will be nearly a miracle if it succeeds in its goal to bring back a sample from Itokawa. But Hayabusa has already returned on JAXA's investment by providing an incredible and unprecedented view of a tiny Earth-crossing asteroid.

Enough ranting, and back to the Deep Impact session. I came into the room as Seiji Sugita was in the middle of his presentation on mid-infrared wavelength observations of the impact from the Subaru telescope (here's the abstract). I wasn't in time to see his data but I jotted down a few notes from the end of his talk -- that he saw evidence for the dust that came out of the impact being accelerated by gas, which I think means that the vaporization of volatiles from inside the comet caused the dust to expand more rapidly than it would have otherwise; and that their observations suggest that the growth of the crater was gravity-controlled, rather than strength- or compression-controlled.

Next up was Casey Lisse, who presented spectrometric measurements from Spitzer (here's the abstract). His presentation wasn't all that different from his talk at DPS: apparently the dust that was ejected from the impact never got particularly hot, so very little of it could have melted and recrystallized. Instead, the impact dis-aggregated large particle aggregates into constituent crystalline particles "without doing chemistry to them." This is based on the observation that prior to the impact, Tempel 1's coma had an almost featureless spectrum; afterward, "we saw a huge increase in the silicate emission feature" in the spectrum. Lisse went on to point out a "hint of a PAH [polycyclic aromatic hydrocarbon, an organic compound] feature, lots of crystalline olivine and pyroxene, and 5-8 percent carbonates." Unfortunately, Lisse left no time for questions, so I'm unable to say what the audience thought of his model fits to the Spitzer spectrum.

Click to enlarge >

Maps and spectra of ice-rich areas on comet Tempel 1
Deep Impact studied Tempel 1 prior to the impact and took both images and spectrometric measurements of the nucleus. Images (a) and (c) are high-resolution and medium-resolution images of the nucleus. The light blue highlighted areas are places where a ratio of images in blue and red wavelengths showed a notably blue part of the surface. Graphs (b) and (d) are crude spectra of those spots, as seen through each of the camera filters. Fortunately, Deep Impact also carried a spectrometer that could scan across the surface of Tempel 1 (e) and produce much more detailed spectra (f). The spectra shown in (f) are normalized by the spectrum of a smooth, non-blue area (outlined by a red square) to make it easier to discern the characteristic spectral features of water ice absorptions. Credit: NASA / JPL-Caltech / UMD / SAIC / J. M. Sunshine et al.

Next came Jessica Sunshine, who presented some stuff I hadn't seen before on water ice on the surface of Tempel 1 (here's the abstract). I've written earlier that their temperature maps pretty much rule out abundant water ice near the surface, but Jessica showed "new de-convoluted high-resolution color images of anomalous areas that are bright in ultraviolet wavelengths." These anomalously bright regions are 30% brighter in the ultraviolet than in the broadband visible (which makes them merely less black than the other areas). Jessica then showed data from the spectrometer and indicated that there were water ice absorption features in these anomalously bright areas. She was also able to use stereoclinometry to determine that the bright spots were roughly 80 (give or take 20) meters below the surrounding areas.

But temperature remains a problem. "If we had pure ice on the surface, we would have to have a temperature of 200 Kelvin," the freezing temperature of water in a vacuum. "What we have is 280. So we can't be seeing large amounts of ice on the scale of the pixels. In other words, whatever ice is there, is thermally decoupled from whatever else we're measuring the temperature of." She attempted to model the spectrometer data with mixtures of ice particles of different sizes and found that about a 4% component of 20-micron-diameter ice particles best fit the data.

Jessica then moved on to infrared spectral imaging of the plume, and showed something else interesting. Looking at the plume in a wavelength that emphasizes dust, you see a broad, fan-shaped plume, just like in the visible images. But if you look at images that are processed to emphasize the presence of water, ice, you see something very different: "it is very collimated and is not broadening the same way that the dust was. Also, the strength of the water emission does not change very much" with distance from the impact site, "suggesting that the water is not sublimating very quickly." Again, Jessica used a model to constrain the particle sizes, and found that 5 to 10-micron-diameter particles were the best fit. "Compare that to the much larger ones we found on the surface," she pointed out. (Note that the particles in the plume are 1/4 to 1/2 the diameter of the surface ones -- which gives only 1/32 to 1/8 the mass. "The change in particle size strongly suggests disaggregation on impact," just as Lisse was saying. Concluding, Jessica said "there must be extensive subsurface water sources, near, but not at, the surface. This suggests that the dormancy of comets is not due to volatile loss" but rather due to the development of a refractory crust.

Next was a talk by Lori Feaga, employing the same sorts of spectral observations Jessica had, but at a greater distance and before the impact on the coma of Tempel 1 (here's the abstract). She found some striking asymmetries in the distribution of carbon dioxide and water in the coma. There was more water in the sunward direction and in a northern jet; more carbon dioxide in the southern direction, particularly in the southern, anti-Sun direction. I'm not sure if this asymmetry implies anything more profound than the fact that the comet is inhomogeneous.

I should note here that apparently, earlier in the session, someone (I'm assuming Mike A'Hearn) apparently showed some newly enhanced and processed images that reveal a pronounced jet or several jets visible on the northern side of the comet, just off the limb. Others showed those images and referred to them in passing. I'm looking forward to being able to get a closer look at them.

Finally, Kevin Housen delivered a talk that was intended to be cautionary to those who thought they might understand the impact mechanics (here's the abstract). Most people appear to be behind the conclusion that the crater growth was gravity controlled, meaning that the particles composing the surface of the comet were essentially strengthless, powdery and cohesionless like copier toner rather than having some stickiness or strength like mud, ice, glass, or rock. Strength is measured in units of pressure, Pascal; the current estimates for the strength of Tempel 1's surface is a few tens of Pascal, a tiny quantity. "There are two features that are used to argue for gravity scaling," Housen said: "that the plume remained attached to the surface, and the mass of the plume." The "plume attached" argument goes like this: particles near the center of the impact site are ejected with some high velocity. As you go away from the center of the impact site, the force of the impact attenuates, and the ejection velocity decreases. If the surface has any strength or cohesion, at some point before the ejection velocity goes to zero, the force is not enough to overcome the material's strength. The last stuff to be ejected is ejected with nonzero velocity, and at that point the expanding ejecta curtain detaches from the surface. With no strength, stuff keeps being ejected until the ejection velocity goes to zero, at which point the ejecta essentially hits the ground at the same moment that it rises; the curtain remains attached to the ground.

But Housen argued that laboratory experiments have shown that gravity-controlled growth is not required for the plume to remain attached to the surface, and he showed some experiments that got a murmur from the audience. "The strength of the surface could be as much as 10 kilopascal for the plume to still remain attached to the surface," Housen said; that's more than a thousand times the current estimate for Tempel 1.

Secondly, there's the issue of the mass in the plume. He showed some graphs that showed that although there is, indeed, more mass ejected initially from a gravity-controlled crater, that hours later after the ejecta has started falling back in to the surface there's no difference between a strength controlled and a gravity controlled crater. In point of fact, he said, neither strength-controlled or gravity-controlled cratering are consistent with accepted models of crater formation. "We have a bit of a problem," he said. The amount of mass that's been estimated by several observers -- more than a million kilograms -- requires some new process: "either an acceleration to allow more ejecta to escape but which is still consistent with the observed rate of growth of the base of the plume, or some nonstandard cratering mechanism."

It's fun when something crops up that nobody can explain! Scientists will be working on this problem for a while.

Phew! That's finally it for Wednesday. Next up: the poster sessions from Tuesday and Thursday evenings.

Post this page to:

Comments


Name

E-mail (Will not appear online)

Title

Comment


To prevent automated Bots form spamming, please enter the text you see in the image below in the appropriate input box. Your comment will only be submitted if the strings match. Please ensure that your browser supports and accepts cookies, or your comment cannot be verified correctly.

This comment form is powered by GentleSource Comment Script. It can be included in PHP or HTML files and allows visitors to leave comments on the website.

Emily's on Twitter! »