I'm confused. The first two images below are from this NASA/JPL page, the third is from The Planetary Society Blog.
The sizes indicated for Hartley vary widely. Which is it?



Edited: to reflect a change in The Planetary Society Blog image which now is in rough agreement with the far left image. The center image is still a problem which is strange since it and the left image are both on the same web page. It looks like that center image should read 0.5km rather than 0.5 miles?

I think that 1.25 miles/2 km is length of longer axis and 1.2 km is medium diameter.

Puts me in mind of a young Amanita

http://etc.usf.edu/clipart/14800/14815/yng...ta_14815_lg.gif

Movie of the flyby imagery here, seems to show some images we haven't seen.
http://epoxi.umd.edu/3gallery/vid_20101104_approach.shtml
(requires quicktime)

That is what is known as a "mistake." I'd gotten the Photoshop file ready for inclusion of Hartley 2 before the Arecibo observations, so I didn't have the new diameter, and I forgot to update it in my hurry today. There's a new version of the image now; thanks for pointing out the error.

I've edited my post to reflect your change. But there still seems to be a problem with agreement between the two NASA/JPL page images. Looks like that 0.5 miles on the center image should be 0.5km?

Just be patient. They haven't had time yet to come to agreement on an official new diameter for Hartley 2. They have to get the final navigation solution to be sure of range to the target, figure out what angle they're seeing the long axis at, etc etc. There's no point in them issuing a number that they'll have to correct later. They're using whole- and half-kilometer estimates and then doing low-precision conversions to Imperial units. Just be patient and we'll eventually see an official diameter announced. Consider everything between now and then to be imprecise, plus or minus 50%.

Returning to the question of the Hartley 2 shape and whether sublimation can cause such a shape, I ran a simple simulation; starting from a spherical shape and solving the sublimation differential equation in time for different rotation axis angles. For alpha=0 (i.e. axis perpendicular to the orbital plane, the arrow points to sun), the object elongates and then develops a waist. For alpha=90, it morphs into a discoid shape. For alpha=30 and 60, the shape looks like a bottle and a bullet, respectively. Of course, this is a very simplified model (starting from a perfect sphere, no shadowing...), but it shows that sublimation is one of the forces shaping a comet shape.

Nice work siravan!

A couple of amazing facts about Hartley 2.

http://scienceblogs.com/startswithabang/

And btw. great, great science blog!

Boy, no kidding nice work, Siravan! Heck, I'd call it fundamental work! The bowling-pin shape seems much less mysterious now.

the only problem that I see with the model is the assumption of starting with a sphere. what would happen with other, more likely shapes?

From the blog:
"Unlike other comets, with their tails made mostly of ice, melting into water, the tail of Comet Hartley is mostly rocky dust and carbon dioxide....
Because it's not a traditional comet; it's an asteroid that simply got too close to the Sun!"

Really? Is this an accepted interpretation?

full inline quote removed - ADMIN

yeah, would any asteroid behave like a comet if "coming too close to the sun"?!?

it's the first time I read this suggestion and to me, it makes no sense. why would other asteroids that venture far closer to the Sun behave like asteroids then?