Here's a mashup of the image on the DLR website with high-resolution image PIA14313. I took the enhanced-color IR composite and broke it up into little chunks, then took each chunk and warped it to try to align with surface features in the high res image. Reassembled the color composite, then used that to blend to the high-res layer underneath. Full details on flickr (here)

Neat how the scarp and some of the craters and ejecta are greenish in the IR composite.

Click to view attachment

An update on Dawn trajectory:
in spite of engine running in last 20 hours, the probe is gently accelerating, progressively approaching a "circular orbit speed", as showed below:
Click to view attachment
The green triangles indicates definitive survey orbit; they are two because I'm not sure if 2700km nominal height is referred to the center or the surface of Vesta... can someone help me?
As additional info, now average acceleration toward Vesta is 3e-5 m/s² while angle between trajectory and Vesta-to-Dawn vector is approaching 70°.

Addendum: I do not share Emily's bad feeling about image release policy... look to the bright side, last released image was captured only 3 days before and also contrast/gamma is a lot better than previous ones!

Another interesting stat might be energy/kilogram. Although kinetic energy is increasing, potential energy is decreasing.

--Greg

The distance for that image was (or should have been) 15222 km (center of Vesta). You already know the angular extent of one pixel. You tell me where you are wrong...

We are working to improve the caption.

Again, I ask you if you think the enlargement of the last image was done badly. Reading your blog, I am not sure. I used the Mitchell-Netravali algorithm, which I think is appropriate. Remember, it is a tradeoff between blurriness and jagged edge.

Why is it being done at all? If (as we have been told) there is a constraint on the teams time in producing images for outreach...why waste some of that precious time in un-necessarily enlarging these images?

To be fair, that practice of magnifying otherwise poststamp-sized images is pretty much the norm for NASA image advisories, although they do tend to specify the magnification factor in most cases.

Enlarging images is standard procedure. NASA use that, almost every amateur processor use that. When one is working with mosaic from multiple images, then resizing is even necessary. When one have post-stamp size image, enlarged variant looks somewhat better, than original one.
"waste some of that precious time"
It's not, in IrfanView (my favorite software for resampling images) it's matter of seconds.
"I used the Mitchell-Netravali algorithm, which I think is appropriate."
Mitchell is fine, it's my favorite one, but I think, that it isn't what Emily is complaining about.
She simply need information about actual size of pixels, that's all.

South pole of Vesta in stereo. Cross eye and anaglyph version. Resolution is approx. 1 km/pix.
Stereo images were made from one published hi-res image and synthetic image (made from both published hi-res images).

machi Thank you for the 3D!

My kids appreciate it also!

Fantastic! Thanks.

Phil

Really nice Daniel.

Thank you all!

I prepared one more cross-eye/anaglyph stereo image, using second published hi-res image.

Today is Thursday, maybe we'll see images acquired on July 18 ?
http://dawn.jpl.nasa.gov/mission/status.asp
Dawn team usually publishes something on Thursday.

Awesome!

Thanks machi

Recent images would be of a narrow crescent, but in a few days we will be seeing the northern hemisphere... brand new territory again, and from closer range.

Phil

With such a lumpy surface, the high-phase images should be pretty dramatic.

I'm curious about the answer as well. I get a diameter of 615km trying to replicate the calculation, and I don't see anything wrong with how Emily is doing it. But that is well outside the biggest dimension of the oblate spheroid model. Is it just that the oblate spheroid is that lousy of a fit for vesta's true shape?

Hopefully this will be out of date very soon, but here's a composite view of all the closer images we have seen so far. Please let me know if I am missing any.

Phil

Click to view attachment

(EGD points out - I should have scaled the first to match the long axis dimension, not the polar axis!)

It was my understanding that a good portion of the northern hemisphere will be in darkness (winter) for the first months of the mission, am I mistaken in believing that?

-kap

Only the north polar area itself will be hidden - most of the northern hemisphere will be visible now.

Phil

Thanks for checking my math, and now I feel a little more confident in questioning the factor-of-2 enlargement. Here's the DPS abstract on Vesta's dimensions from Hubble data (289, 280, 229 km semi-major axes, or 578, 560, 468 for diameter), which refers to a previous conference abstract with a diameter based on an occultation (561 +/- 3 km). In no way are any of these consistent with any principal axis diameter above 600 kilometers. So either the enlargement factor or the range to the target has not been reported correctly.

I know I'm beating a dead horse here and I think the Dawn team now regards me as kind of a pest and really I am very excited about seeing a new world. But it's hard to do outreach when I know that some of the information that I'm getting must be wrong.

Regarding the enlargement, I like neither the blurriness nor the jaggedness of the limb. Both, I think, do a disservice to the FC which I think is supposed to be a very fine instrument. Now that details can be resolved, it's much better just to post images that are not enlarged. Consider the most recently released image of Vesta. Now consider a Cassini image of Phoebe with the same number of pixels. The Cassini image looks so much crisper, and a comparison between the blurry-looking FC image and really any other deep-space camera image makes FC look like a crappy instrument, when we know it's not. it's not a fair comparison, because the Cassini image started out with 800 pixels, while the Dawn one started out with only 400. But most people don't know that; they just see a blurry, pixelated image.

Please realise that at this stage, we face similar uncertainties. There is one thing about which I am 100% sure: the enlargement factor is 2. But then, there are so many possibilities:

1. I calculated the distance wrong, and my colleague made the same mistake
2. The SPICE kernel we used is wrong
3. We used the wrong SPICE kernel
4. The camera FOV has shrunk
5. Your measurement is correct
...

We are using trajectory *predictions*! Would it make a difference if I say that we (the FC team) are going through an extremely busy period, in which our main priority is to ensure the FC is in good health and taking images as planned?

I'm sure that's not the case.

...but if any of them do think like that, then I really think they should be grateful that a respected and accomplished science journalist like yourself is so excited by, and passionate about, the mission, wants to ensure its success by writing about it for a public audience, and is trying hard to get the facts straight in order to write accurate reports.

I apologize for being such a pest. I think I don't properly appreciate or understand the differences in navigational certainty between an ion-powered mission -- and one that's approaching a small body with relatively poorly constrained mass -- and the kind of orbital or flyby missions I'm accustomed to writing about. I'm glad to have a definitive answer about the enlargement factor, and now at least I think I understand which bits of information are the sources of the uncertainty. I had assumed wrongly that the range to the target was one of the more precisely known bits of information. So, sorry! I hope you'll take my pestiferousness as a sign of my real interest in and excitement about the mission!

another thing to question is the size of the error bars on the Hubble measurements of Vesta's size. The Thomas et al 1997 paper is online at http://www2.keck.hawaii.edu/inst/people/co...wg/icarus97.pdf. Skimming through that, they use images with a resolution of 52km per pixel. They estimate errors in the semi-major axes of +/- 5km, or 1/10 of a pixel. I don't know details of their fitting, but it doesnt seem crazy to me that they might have missed by a bigger fraction of a pixel than that. Getting sub-pixel info from images is hard.

What surprised me about that paper was that their definitive reference for Vesta's size based on occultation data was from 1989. My 5 minutes of trawling in ADS didn't turn up anything obviously better. Given Vesta's huge size and interesting shape I'm surprised there haven't been more, better-quality occultation studies since then.

EDIT: Looking at fig 2 of that paper, I think it can be stated pretty strongly that no semi-axis of Vesta is at all likely to be longer than 290 km.

New image! http://www.nasa.gov/mission_pages/dawn/mul...0721-image.html

OK, now it looks like Hyperion. Vesta's an asteroid of many guises

So *that's* what it was reminding me of!

I'm not so convinced. My understanding is those numbers are for best fit ellipses. But Vesta has a big chunk missing on one side. Fitting an ellipse to a shape which isn't an ellipse could give a systematic bias. The missing chunk would make all the best fit ellipses underestimate the true maximum dimension.

Has anybody looked at the effects of geometric perspective? Dawn is now close enough to Vesta that it should show a somewhat greater angular diameter as seen from Dawn than you would expect by simply dividing Vesta's diameter by its distance from Dawn. Vesta's edge, as seen by Dawn, is actually slightly closer to Dawn than the "great circle" perpendicular to the line connecting the centre of Vesta with Dawn, which would be slightly smaller as seen by Dawn (if it could be seen through solid rock). Therefore the number of pixels from one side of Vesta to the other would be slightly more than calculated using Vesta's diameter, and its distance from Dawn.

I do not know if the difference would be enough to account for the discrepancy described earlier, but somebody could check?

I have a question about Vesta's reference grid. A few days ago an image release specified as "The original image was map-projected, centered at 55 degrees southern latitude and 210 degrees eastern longitude." OK, the latitude is determined by Vesta's spin axis, of course, but how is the system of longitude determined? What feature on Vesta defines 0° (or some other standard) longitude?

The answer to that question is in the paper that tfisher linked to: http://www2.keck.hawaii.edu/inst/people/co...wg/icarus97.pdf

Thanks, Emily ...

Well, it's the difference between 2*r/d vs 2*arcsin(r/d), right? So if I take 10,500 km as the distance and 289 km as the radius, then I get an angular diameter of 3 degrees, 9 minutes, 16 seconds. As I figure it, your approximation is 2 seconds smaller.

In survey orbit, at 2700 km, the difference will amount to about a minute and a half. (Someone should double check me, though.)

--Greg

I think we have a new image just posted at JPL website. Taken on July 18th.

Not sure how to embed a thumbnail, so I'll just have to put in a link.

http://www.nasa.gov/mission_pages/dawn/mul...0721-image.html

Already being discussed for a page of this thread - first cited here

http://www.unmannedspaceflight.com/index.p...mp;#entry175945

Nice idea, Greg; this is the output in terms of energy per mass unit (on the left), with a zoom around the insertion time:
Click to view attachment
On the right, a plot of ratio between the two energies vs distance (obviously, I considered absolute vaue of potential energy, which is increasing while approaching Vesta); target is a 2700 km distance with a 0.5 ratio (a circular survey orbit).

I have been playing with a very rough map of Vesta from the released images. Please bear in mind this is VERY approximate and not controlled by any shape model or pointing information. It is intended just to show approximate image coverage and locations of major features. The tie to more distant images is very rough. Zero longitude in the Hubble map/shape model coordinates would be at the left end (and the right, I guess). A much earlier version of this, posted here earlier, used a different (arbitrary) zero longitude.

Phil

Click to view attachment

Dawn's looping round the north side now, so the 'where is Dawn' page is showing an illuminated crescent:

http://neo.jpl.nasa.gov/orbits/fullview4.jpg

(presumably that links to the current image, when you link to it, not the version I am looking at now)

Enough of the surface is visible to see that the texture map currently in use is really from Tethys! Penelope and the chain of craters to its west are visible right now.

Phil

Funny, and well spotted! I've attached the current screenshot, since it will have changed by the time some people here look at it.

Here's a crater - middle of this view - with dark markings inside and outside its rim. Other distant images show at least one other dark spot like this as well.

Phil

Click to view attachment

It looks like maybe a fresh impact inside an older crater.

I'd say excavation of dark subsurface material.

Phil

I'm curious if MYSTIC simulator use real data to create the image of Vesta ? Is this a true picture of Vesta? If it corresponds to reality, it is very interesting.
Click to view attachment
Ion engine is not operating from a few hours, I hope that the framing camera just works.

As Phil pointed out yesterday, that's Tethys.

If our images of Vesta were already that good, we'd have no need of a Dawn mission...

Tried to line up the IR image with this view. If I got it right, that crater is right in the middle of a "greenish" splatty zone. This is the same tint as the sharp scarp face. There are two other greenish splatty zones below the central peak in the image, but I'm almost positive this area is misregistered. I'm going to run with the idea that the green tint indicates fresher material.

Click to view attachment

BTW, here's the blink animation between the original black and white image, and the manually warped IR image:

Click to view attachment

[animated GIF: Click to animate]