There have been many questions about the candidate tracks we have identified here and are using in the new Calibration Movies. We'd like to give some insight into the selection of these candidates.
First, a discussion about the high-angle candidates.
We won't know what the high-angle candidates are until we extract a few and analyze them. We think that they are unlikely to be interstellar candidates, for two reasons:
(1) these tracks do not show the characteristic flaring of hypervelocity particles. This flaring is caused by shock compression and heating. When a particle impacts moving much faster than the speed of sound in the aerogel, a shock wave is formed (like a "sonic boom" from a supersonic aircraft). This shock compresses and vaporizes the aerogel, causing the track to be approximately ten times the particle diameter near the top of the track. As the particle slows, the shock weakens and eventually disappears. At subsonic velocity, the particle is more like a snowplow, just making a "whisker" track. The candidate high-angle tracks appear to be subsonic.
(2) They are all pointing in directions that make them consistent with an impact of a micrometeoroid on the solar panels.
However, we have been surprised before (many times) so these could turn out to be much more complex than this simple explanation. Only detailed analysis after extraction will tell for sure.
Second, a discussion about the interstellar candidates.
There are several points to be made about these candidates:
-- Some, all, or none of these may be actual interstellar dust particles. We won't know until we extract them and analyze them in detail. We can be pretty sure that not all of these are interstellar. From the dust measurements on the Ulysses and Galileo spacecraft missions, we expect perhaps 20 interstellar dust particle should have been collected in the area that we have scanned so far. If this is correct, then only 40% of these candidates are real. But there are many uncertainties in these measurements, and it is possible that there are substantially more (or less) than the dust instrument measurements of those past missions imply.
-- None of these candidates look just as we expected. This is not in itself surprising, because we can't simulate the impact of complex interstellar
particles at hypervelocities in the laboratory. The Phase 1 calibrations were based on the best simulations that we have (submicron carbonyl iron at about 20 km/sec, done at the Heidelberg dust accelerator), but the projectiles are different from interstellar dust and the aerogel used in the simulation is different from the Stardust mission interstellar aerogel. Because we measured the efficiencies of the "dusters" using the calibration movies, and found that the efficiency was very high, we can be confident that if there had been tracks that look just like our laboratory simulations, they would have been found.
-- The identification is entirely subjective. We don't have enough information about what interstellar dust tracks should look like to be able to define objective criteria. When we have some "ground truth" we expect to identify tracks with more confidence. The fact that dedicated "dusters" identified these extremely subtle features is extremely impressive and may turn out to be very important. We might have been able to train a computer to search for features like our laboratory simulations, and if we had we might not have found any of these candidates. So we regard this as vindication of the Stardust@home approach.
-- We are scanning tiles while they are still in the tray. The reason we do this is that we want to make sure that we don't lose critical trajectory information. If we extract the tiles, we run the strong risk of losing this information. We are therefore imaging these tracks in the worst possible optical conditions. We can illustrate this with tracks from the cometary side of Stardust, which are much bigger.
First, we show an image of track 22 from the cometary side, imaged while it was still in the aerogel tile, and with the aerogel tile out of the tray, so transmitted illumination was used. (So these are much better optical conditions than we have on the interstellar side, where we are forced to use reflected illumination!)
To be sure, this is not a focus movie, but it would not be a lot more impressive if it were. You can see a mosaic of the entire
tile at http://curator.jsc.nasa.gov/stardust/sa ... t1_20x.jpg
Now we contrast this to what we can see after this same track was extracted in a "keystone" in our lab at Berkeley.
The particle that made this track was about 1,000 times more massive than the typical interstellar tracks that we're looking for. So this illustrates the challenge that we have!
About the candidates themselves.
We have reviewed most of these candidates in the high-resolution, raw movies. Generally, we have selected candidates that look like tracks. They have circular, near-circular, or elliptical mouths and appear to extend below the surface. Since the track may be obscured by the mouth, it is often difficult to be confident that there is a track there. In the compressed movies on the VM, it can be even more difficult. We are not posting the raw movies for viewing because they are enormous in file size. In a couple of cases, the track is so small that the shape of the mouth is impossible to determine; all one sees is the structure
extending below the surface.
Some have found that they can see the most subtle candidates if they increase the resolution on their monitors, so if you just don't see the candidate you might try that.
We emphasize that this is only a partial list. We are still selecting new candidates, using different methods for analyzing the search data and looking at people's reports on the forum. If there is a candidate that you think is particularly promising, and you don't see it on the candidates list, please post it on the forum here.
Please have patience with us; we're working as hard as we can!
Some of these have plausible alternative explanations. For example, several may be just two inclusions that are coincidentally one on top of the other. Others may be little fragments of aerogel that are partially embedded in the surface. But we are deliberately casting a wide net to be more likely to find the real interstellar stuff, and we don't want to throw the baby out with the bathwater (sorry to mix metaphors).
There is no doubt at all that some of these will turn out to be something else. We have quite a wide variety of different shapes and morphologies here. As we extract these candidates, we will get images like that of the keystone above. These will immediately teach us a lot. One thing that we can almost guarantee is that we're in for some surprises!
This search might have come up with nothing. Fortunately, it didn't turn out that way. We are very excited about extracting these candidates and analyzing them. These extractions require a huge investment in time and careful, meticulous effort. We wouldn't go to this effort if we weren't excited about what YOU have found through your truly impressive effort.
Stay tuned!
27 Sep 2007 - Discussion of Candidate Tracks
Moderator: Stardust@home Team