I don't know if they've addressed that publically anyway. Maybe Enceladus is way too small (diameter about the distance between Los Angeles and San Jose) to have such a core?

On another matter regarding the plumes: may we assume that if the plumes are sometimes sending ice particles into space with escape velocity, the plumes also sometimes emit particles with less than escape velocity that will impact somewhere downrange on the moon?

If so, then is there then effectively a rain of ice particles impacting the surface, resulting in accumulation and even erosion in the preferred impact spots? After all, the particles would impact with about the same velocity they left the plume with, up to 240 m/sec. Truly a "hard rain" falling.

This is not the first time a magnetic field has shown up where it was least expected: Ganymede was a big eye opener. The difference I can see, is as you have elluded: There is a possible pretender to a magnetic core in the water/ice/vapour that is escaping. This is just a guess, and not a well rooted one.

It's been assumed since Voyager that this is the likely explanation for Enceladus' extremely high albedo -- it's refrosting itself, all over its surface. (Presumably, when such impacts vaporize some of the ice in the particles, it quickly refreezes again a short distance away.)

Seems like the process is becoming a lot more constrained, knowing that the ice jets are coming from (approximately) the south pole.

A while back ugordan and I were discussing the escape velocity & I came up with a little simulation predicting what would happen to a particle emitted from the south pole. I refined the program some more recently, and now it allows prediction of the path of a particle emitted from any point on the surface, and in any direction (azimuth and elevation from the "vent" location). Some interesting asymmetries are evident resulting from the moon's rotation and the three-body situation involving Saturn's strong gravity.

Take the case of a vent located at 180W, 80S (10 degrees from the south pole) and with a plume direction of 70 degrees elevation. Rotating that plume 360 degrees in 5-degree steps (think of it kind of as a rotating sprinkler) results in a surprising asymmetrical pattern. This map shows the predicted impact points, varying the plume velocity from 0-220 m/sec (approx. escape velocity) in 1 m/sec steps (click for larger version):



The impact points are preferentially towards the center, which is the anti-Saturn (180-degree) meridian, offset to the west a little due to the moon's rotation.

That's just the result with the plume at one location, but interestingly a simlar pattern persists even if the plume location is moved (click on any for a larger version):

0W 80S:


90E 80S


270E 80S:


Not sure if it's my imagination, but it seems like the patterns line up with some of the features on Enceladus. Interesting anyway.

Outstanding work, Joe!
I'm curious: are those Matlab simulations or something you programmed for yourself?

One thing that could be done to further investigate the patterns would be to integrate across all tiger stripes and all angles/speeds and see if the "fallout" map would preferentially exclude some areas of the moon and then compare with the albedo and crater saturations of those regions.
That would probably be grounds for a scientific paper, though

What you did already suggests there indeed are some areas in the north that should be very depleted of snowfall.
Once again, great work!

I programmed it in C, using the CSPICE library to determine the positions of Enceladus and Saturn over time, and a Runge-Kutta integration to determine the particle trajectory. One thing about doing a lot of integrations though across the tiger stripes or whatever: each map takes a few hours to generate on my machine! So I'm wondering how exactly to plan it out that would give relevant results and not waste a week of number-crunching.

Thanks for the magnetometer vector slide, Jim; most interesting!

One thing that I find particularly striking is that the field vectors--rough as the data may be for this comment-- do not seem to be "dragging" due to Enceladus' orbital motion, as you might presumably expect if the field source was in fact the eruption cloud. In fact, such an effect might also be expected from transfer and diffusion of eruption material (and charge) throughout the E-ring; instead, the field X-Y plane seems orthogonal to Saturn and therefore centered on the mass of the moon itself. Might that imply an endogenic, internally generated field for Enceladus as a reasonable interpretation of this dataset instead, or am I overestimating the effects of orbital motion and cloud behavior?

It's been a while since I studied different integration methods, but isn't there a time step that's adjustable for different levels of precision? Or are the stability constraints too tight? I'm probably talking jibberish here...
Out of curiosity, what machine are you running the sims on?

Yes, I used a step of 3 seconds, which is maybe a little overkill but not sure how far I can push it. I'll look into that. The main time hog is calling the SPICE library every step to find the positions of Saturn & Enceladus--very accurate but definitely overkill. Maybe the whole thing could be sped up quite a bit.

It's running currently on a Linux system.

I don't think you would lose much in terms of accuracy if you simply assumed Enceladus was in a circular orbit with a given orbital period. Given the angular momentum, it would be a no-brainer to calculate Enceladus' position at any given time.
That way there'd be no need for extensive calls to the SPICE library at all. All those function call overheads would simply go away...

Just as a sidenote to the main thread: There's been some discussion here regarding a small iron core for Enceladus. Assuming for the moment that the observed magnetic anomaly is internally generated, it seems to me that a small _solid_ iron core, below the Curie temperature, is a better contender for the source of the magnetic field than a liquid core.

I just have trouble believing that a dynamo could get going in such a small body (and that the core could be molten). Of course I haven't been through the physics so this is just conjecture, but it seemed worth mentioning.

So the question then becomes, why don't we see similar fields at Tethys, Dione or Rhea? This argues _against_ internal generation of the field.

There are at least three major differences between Enceladus and the other icy satellites relevant to this discussion:

1. Enceladus is on the receiving end of orbital resonance 'squeezing' from Saturn and Dione, and it doesn't have much overall mass to dissipate the resultant heat.

2. Enceladus formed closer to Saturn than the other icy moons cited, and therefore may be enriched in metals (including radionuclides) by comparison, and again has limited mass for use as a heat-sink.

3. Cassini proved that Enceladus alone among the icy moons has a present internal heat source of sufficient magnitude to drive surface vulcanism.

Systemically, these factors (if valid; the actual magnitude of #1 and any other tidal interactions seem poorly defined right now, and only #3 is a confirmed fact) may yield the right combination to produce at least a semi-molten iron core.

Another passing thought/wild theory:
What if Iapetus' dark side was caused by something similar to Enceladus' plumes - a massive "tiger stripe" on the one side, which is now the ridge? It could have spewed out material long ago, coating one side of the moon with dark material, and at the same time, formed a volcanic ridge, similar to the mid-Atlantic ridge on Earth.

Well, SOP for optimizing a slow calulation like that is to simply do it once, create a lookup table, and reference that every time. If each run starts at the same time, they can use the same Saturn/Enceladus location information. In fact, you could even create the lookup table off-line and load it at the start of each execution.

I just had a thought: What if there is a "flux tube" between the Saturnian magnetosphere and Enceladus, as there is between Jupiter & Io? If so, perhaps the south pole of Enceladus is the impact point, which might add another increment of local heating and explain why vulcanism is apparently confined to that location.

Presumably this would still require Enceladus to have an endogenic magnetic field, though.