Okay, as one of the ISS people here on this forum, I will try to clarify our position. First, as Emily pointed out, we had the raw images page on the JPL site thrust upon us, there was definitely some strong criticism of that decision by a number of people on the team. The reasons vary, but primarily it took away the ability of the team to announce major discoveries, to an extent. For example, everyone here knew about the spokes (and the rumors abounded within the space science press) for a week before we actually announced it. This jives with what John said on Slashdot, it puts us at a disadvantage when we are scooped by someone who woke up earlier than us. Now, in many ways, I both agree and disagree. I have no problems with some of the stitching work done here. Due to the JPEG artifacts and stretching, I don't feel I have anything to worry about, and it provides good press for Cassini. However, I do agree that any talk of publication based on the images on the raw images page or from press release images is down right unethical and because of the way the data has been manhandled before you see them (except my mosaics, they are treated with TLC or someone gets a caneing), that no peer-review publication would even touch it. But, that may not stop someone from working on a paper then submitting it once the data officially hits the PDS. And in that way, we are put at a disadvantage.

Now, what does this mean for people like jmknapp. Well, his work isn't as constricted. His work was a simulation of particle fall out from a plume near the south pole of Enceladus. Now, you would get into some problems with the fact that the seed locations were along tiger stripes that have not been described in a peer-review publication. So, I would dissuade you from going forward until that publication is out. But still, it may be best to make the simulation more generic, maybe simulate fallout patterns from random locations in the region, and see where that gets you.

I'm moving all the Cassini Titan tour tweaks discussion to another thread.

Perhaps that's one ancillary (though perhaps unintended) benefit, but I always thought the primary purpose of releasing raw, unvalidated imagery on the Internet was for public relations (viz., "You [the public] see the images before we do! This shows how cooperative we are and it also shows we're not airbrushing out the Little Green Men.").

At any rate, I somewhat agree with Phil Stooke regarding this issue. If, for example, NASA/JPL/ISS doesn't want the public "analyzing" the raw, unvalidated imagery (whatever that means), then they should explicitly say so. Frankly, though, I don't think that will happen since, as a practical matter, any such an approach would quite likely blow up in their collective faces. While there may be some latent hostility and grumbling among the team members, the gatekeeping process of peer-review that I alluded to (though "pat" described more clearly) serves as the ultimate check.

That said, I don't think this is much of a problem. Except for the kooks, who are easily dismissed, or honest amateurs with a real interest in the process, there aren't many people rushing in to "analyze" the raw, unvalidated data. For instance, MER released the raw, unvalidated Pancam, Navcam, and MI imagery from the outcrop at Eagle Crater and conducted detailed press briefings months before the first papers were published in Science. In fact, the story at Eagle Crater unfolded slowly over several weeks and, despite the great amount of speculation on the Internet about what it all meant, I didn't see anyone rushing to publish ahead of Squyres et al. Indeed, their recent, detailed publication in EPSL, over a year after the initial discovery, still outstrips all of the discussion I've seen.

I attribute this to a variety of reasons, but chief among them is that the vast majority of the public doesn't have the faintest idea of (or interest in) how to do science in the first place, so they don't even try, lest they embarass themselves in the attempt. And the small segment that does know how to do science "plays by the (unwritten) rules" of etiquette.

Wasn't the orginal intention to have one picture from Cassini released per day, Monday - Friday?

I agree that the ISS team members get screwed with this approach, while other teams get to hold on to their data ("I demand access to the T8 MAG data now!"), and this would engender some hostility as a result. And you're right, any scientist with ethics wouldn't try to scoop the team in the first place by publishing in peer-reviewed journals.

However, journalists, bloggers, and other armchair amateurs are different. For example, Emily's lament that she had been "yelled at a few times...[for] crossing the ISS team" shows how ridiculous this has become. It's as if the developers of bird seed at an animal food company are complaining to the birds for eating seeds placed in the bird feeder because the sales/marketing department released it against their objections.

BTW, here's an animation of the incoming Enceladus flyby, Dec. 24:

Enceladus Dec. 24 inbound (2.2MB AVI)

The yellow dot shows where the ORS platform is pointed, based on the SPICE CK kernel. Not sure what the scanning is all about.

Please help me wrap my software engineer brain around the concepts in this discussion.



Does this mean that the tiger stripes do not "officially" exist, even though images of them have been released to the public by NASA, until a peer reviewed article has been published? And Joe would have problems because his simulation involves these features that do not "officially" exist?

Not trying to be mean or sarcastic - I just want to understand the process...

I don't see why this needs to be officially described. Any sentient being with tentacles and antennae can see plume, fractures and subdued high-albedo areas and put Bwrdd and Iaith together to get Gymraeg. Joe- and a lot of others, I'm sure- are assuming that the plumes are here, so lets assume this is happening, crunch numbers and get _this_.

My 10^6 Zlot, FWIW.

--Bill

To put another two cents in: There really shouldn't be all _that_ much of a problem regarding major announcements getting scooped, just as long as our noodling around with the raw images doesn't work its way into the mainstream press. There are what, a few hundred people here? That's really not much when compared to the millions, or even tens of millions, who are going to see the official press releases on Google News.

Also, I agree with the folks who have pointed out that most of the amateur analyses being done here would never get through peer review anyways. For example, I've played around with some images for the fun of it, and even written a few Matlab scripts to "analyze" raw images. But the techniques I used consisted of ad hoc, strap-it-on-and-go reasoning, of the sort that I wouldn't send out for peer review if my life depended on it. That can hardly compete with a conference abstract.

There is of course another, less easy to implement, way to alleviate the problem, and that would be to speed up the peer review process. (If you really want to hear an angry, bitter diatribe, try asking me about the paper that I've had in review at "a journal" for over two years now. It has just been scooped by someone else, who independently came up with the exact same idea, submitted it to "a different journal" over a year after mine went in, and then got published first because they got through review in six months. Experiences like this will give you some appreciation for the frustration felt by the imaging team.)

They have only been discussed in press releases and at DPS/AGU. So, yes, they "officially" exist. And the plumes have been shown in a press release, so they "officially" exist. Using the exact location of the tiger stripes is a problem because officially, that info isn't available, and you are using press release images to derive those locations. Now, if Joe wanted to do a simulation with random seed locations throughout the south polar region, that would be fine. It may even be more useful...

Gotcha - thanks!

Bringing us back to the plumes, the VIMS team has released some specta of the plume/jets taken during the Nov. 27 non-targeted encounter:



Enceladus Plume

During a non-targeted flyby by the Cassini spacecraft of Saturn's moon Enceladus on Nov. 26, 2005, Cassini's visual and infrared mapping spectrometer measured the spectrum of the plumes originating from the south pole of the icy moon.
The instrument captured a very clear signature of small ice particles in the plume data, at the 2.9 micron wavelength. This image of Enceladus, taken with the visual and infrared mapping spectrometer, shows not only the plume over the south pole, but also the dark side of the moon, silhouetted against a foggy background of light from the E Ring.
The bottom graph shows the measurements of the spectrum, of this background light. It shows a very similar signature of small ice particles to that in the plumes, confirming earlier expectations that Enceladus is indeed the source of the E ring.
Preliminary analyses suggest that the average size of the particles in the plume is about 10 microns (or 1/100,000 of a meter). The particles in the E ring are about three times smaller. The sunlit surface of Enceladus itself, visible as a thin crescent at the bottom of the image, is also composed of water ice, but with a much larger grain size than the plume.
The Cassini-Huygens mission is a cooperative project of NASA, the European Space Agency and the Italian Space Agency. The Jet Propulsion Laboratory, a division of the California Institute of Technology in Pasadena, manages the mission for NASA's Science Mission Directorate, Washington, D.C. The Cassini orbiter was designed, developed and assembled at JPL. The visual and infrared mapping spectrometer team is based at the University of Arizona.
For more information about the Cassini-Huygens mission http://saturn.jpl.nasa.gov. The visual and infrared mapping spectrometer team homepage is at http://wwwvims.lpl.arizona.edu.

On that note:

Think it'd be a good idea to take all the posts in this thread that relate to publication, and put them in their own thread? This has been one of the better discussions I've seen in this forum, but it _is_ a bit off topic for the Enceladus Plume Search (although of course it didn't start out that way).

Hm. It may be misguided of me to put my two cents in, but here goes anyway: I can see two possible outcomes. Either you get the same patterns from ejections from random places between the pole and South Latitude X degrees. Then fine, it really doesn't matter where they come from. Or they come out differently, and then -- if there's the time to run enough simulations -- you could constrain the locations off the simulation itself, saying "in order to get a pattern mapping to such-and-such an area [assuming that that area officially exists!] the ejection has to take place from an area within this part of the southern polar region.

Now that would probably be pretty arduous, but if jmknapp could *predict* the existence of the tiger stripes based on that, it would be a pretty remarkable result.

Ouch--a quick heads up: I'm not exactly sure yet, but I think the results of the sim are suspect.

I've been testing the basic code the last couple days and figured a good test of the integration would be to see if it could simply predict the orbit of Enceladus, using the SPICE-reported positions as a gold standard. If the simulated Enceladus starts out at the same position as the real (SPICE) one, the positions diverge over time such that after 6 hours of sim time, the positions are 200 km different. This error was the about the same regardless of the integration step size, meaning the integration code itself is probably not at fault. What I think is happening is that the orbit of the real Enceladus is precessing which is not modeled by a simple point-mass integration that my sim is doing.

So in the sim the SPICE library calls fix the position of Saturn and Enceladus accurately, but the test particles lack the precession component. With particle times of flight upwards of 10 hours, the error is significant. Anyway, regardless of the cause, the predictor seems to be faulty.

I modified the code to test this further. Instead of using the SPICE library, I determined the Enceladus position also with the integrator. So both Enceladus and the test particles would follow idealized, point-mass trajectories. That sim is still running (takes a long time especially with the added calculations) but I can see by the results so far that the pattern seems to go away. Will post tomorrow...