There seem to be a few unofficial photos of the landing floating around (e.g., screenshots of computer monitors showing photos apparently taken near the end of the descent). Is there any reason why these have not been officially released yet? As far as I can tell, only one ROLIS descent image has been officially released (the 3km one).

Livestream at some points had over 400,000 viewers (and constant 300k+ after ca 4 pm CET), which does speak of people tolerating some lack of polish.

They said images would be released if there was no obviois new science discovery in the images, and the PI owner agreed.

Not everything on twitter is correct or true! This being said, it's hard to get hard facts from ESA.

Proper adjustment / processing needs some time.

In the case, that there is no contact to the ground, and the flywheel is near the center of mass, yes (Philae is almost symmetric to the z-axis). Otherwise the motion (precession / nutation) can be rather complicated.

Here's a very quick and dirty tweak of the higher altitude ROLIS frame:
Click to view attachment
Based on comparison with this landing site image with stated 500 m radius circle, I find a width of about 38.5 metres for the ground visible in this frame. With a ROLIS fov of 57.7 deg, that gives a distance from the ground of about 35 metres for this image.

Here's a very dirty job regularizing the closer screenshot image to a roughly rectangular shape. Can't be sure about the aspect ratio since it appears some of the image is missing on the bottom:
Click to view attachment
Comparison with the previous frame gives a width of about 9.1 metres for this image, with corresponding distance of about 8.3 metres from the ground. In other words, still well before landing.

There are big caveats here, of course: the original 500 metre scale was only approximate, the leaked images may only be crops of the full images, and any errors compound as you go from one image to the next.

Landing site Agilkia from different imagers and in different scales.
The last one "published" ROLIS image has resolution ~4 cm/pix.
I suppose that full image will be (was) downloaded at 1 Mpix resolution, then full resolution will be ~1 cm/pixel from distance 10 meters.

The BBC has released a video of Stephan Ulamec's statements in the ESA press briefing:

http://www.bbc.co.uk/news/science-environment-30031531

I found it curious that those last two ROLIS images depict some apparent rotation which should not have been the case if both were during the same descent. Is it possible that the last image is a post-bounce image that now represents some translation effects of the flywheel? If so, my eyes might be deceiving me but there is perhaps an imprint of a length of leg in the upper right on a "sandbar."

Scientists at DPS14 discussing your image https://twitter.com/maxmutchler

Click to view attachment

No, apparently philae was spinning once every 9 minutes during the descent. It was mentioned earlier in the stream I think.

They did detect a slow rotation during decent, which apparently surprised them. They also think it might have done a little bounce turn when it landed. Also the flywheel was designed to disengage upon touch down. The bar in the upper right is one of the lander legs that's in the camera's field of view.

I missed the earlier comment about rotation, so thanks for the reminder. I'm aware of the hardware in the corner. The very faint mark I saw is beneath the "shark tooth" shadow on the middle right (upper as in quadrant I but very low in it). I'm not convinced it is real, particularly given the poor resolution and possibility that the photo is still during the original descent. It just happened to match nicely with the meter-length legend below it. I see that conspiracists are seeing alien shadows as well now, so I'm even more removed from my tenuous hypothesis now! Next topic...

Fantastic work with the landing site location, Machi!

Phil

...still no images from the surface..

Images could be delayed if there is something of scientific interest.

I would be surprised if we see anything more before the press conf tomorrow afternoon CET.

FWIW - reading very much between the lines ( the notion that after the first touchdown there was an approx 2hr period of the spacecraft exhibited rotation and then came to a stop which infers a long bounce ) - my basic math suggests a 2hr bounce would reach about 530m altitude and start with a 0.3m/sec rebound.

And with the speed of the nucleus's rotation, the surface was moving underneath a great distance too! Hopefully they can track down where it ended up.

2 hrs would put us 1/6th of a comet rotation apart and therefore the second touchdown might be off the 'head' and closer to the back of the neck of the comet. Or it might have got a direction to the rebound sending it another direction completely. I'm not very confident that a 500m 2hr bounce really did occur - but it's a thrilling thought. 2 landings for the price of one

Fingers crossed OSIRIS can spot Philae on the surface and we can then compare that to Machi's exceptional location chart.

The damping system would have also damped some of the relative lateral component of the lander, so I expect less drift. Several bounces may have occurred during that settling time, further reducing down all those effects exponentially. One hopes that the last landing was still upright and flat! I will be tuned to the social channels for tomorrow morning's (8 a.m. Eastern) briefing.

Just imagine the view from the lander during that 2 hr bounce... It would be like watching almost the entire 2001 Space Odyssey without blinking.

If Philae really bounced 500 m. into space would it then autonomously orient itself legs down for a second landing?

Not if the flywheel had been turned off on the first landing...

The flywheel turned off at first contact, so no. If its right side up on the second contact its due to luck as much to engineering. (Edit: beaten).

A dual landing is special indeed, and make this another unintended first.

With the bounce it seem plausible that the thruster did not work after all.
Yet it might be good news about the harpoons if they did launch and has been rewinded or was not used at all, regardless of those alternatives one attempt to anchor the lander might be tried.

Its just past midnight in Darmstadt so there will not be anything official until tomorrow. Spacedaily got this story "European probe lands on comet, fails to anchor".

Worth keeping in mind from one of Emily's tweets: "Ulamec emphasizes this is all speculation. The story may turn out to be quite different tomorrow." I take this to include the possibility that even the "movement" may be due to other things. I recall Rob Manning, during one of the MER landings, saying "It's still rolling" when the real effect had more to do with DSN reception angles or some such.

Apologies if this has already been addressed, but was Rosetta imaging the target site during the time of the landing? Could it have resolved Philae bouncing off the comet, if so?

Tweet from @Philae_ROMAP
magnetic field analysis revealed 3 landings at 15:33, 17:26 and 17:33 UTC

Awesome.

Is there a source for the two ROLIS images in this montage yet (beyond "random unsourced tweets")?

Are either of them from the CNES webcast? (Anyone got some time offsets for "good bits" in that?)

Here's my guess at the rough location of the 'landing' site that machi identified in the officially released 3km ROLIS image.
Click to view attachment

when the lander made contact for the first time and the flywheel turned off,
it transmitted some or all of its angular momentum to the body of the lander,
hence the rotating. That would've acted as a gyroscope, keeping the lander
upright - however imperfectly - with respect to the local vertical.

It would be a different story if it veered too far (kilometres) from its intended landing site, though.

Actually, the honor of unintentionally bouncing goes to Surveyor 3. First intentional second landing goes to Surveyor 6. And that's not counting Luna 9 and Luna 13, since they were intentionally hard landers.

If it did indeed do some very slow bounces, could it have picked up some of the rotational momentum (not sure of correct technical term) of the comet upon each touchdown? So maybe it's not too far away from its original landing, as it got nudged in the right direction with each bounce into an extremely low orbit that matched the rotation, somewhat, of the surface just below.

We'll see when Rosetta reacquires the signal in a few hours when the site rotates back into view.

You are absolutely right, I should have said '...first on a comet.' Since that was what I had in mind.
Now PaulH51 have mentioned it might be 3 and we still have no final word, Philae might soon compete for the grand title of most landings ever. =)

Strictly speaking...Spirit and Opportunity bounced quite a few times in their airbags...

Just a quick check-in from Darmstadt. What a ride it's been today. It's been occasionally frustrating in the media room -- ESA are always much more focused on the VIPs than the wretched news media (we were actually 2 floors down from the room where they were doing updates). But it was fun anyway.

I anticipate little news coming out of the mission except maybe in the form of the kinds of tweets we just saw from ROMAP before the 2pm CET briefing. Everybody is just too tired. Even the news media. We all need to sleep in.

Except I have one more interview on BBC World News tonight

So just to be clear? Do we know if the lander is stationary on the surface of the comet?

Tidbit of news from elakdawalla (who I thought was already totally done for the night, but just blogged -- thanks!) that I hadn't noticed mentioned already:

Wow, three landings. I'm amazed it made it past the second - after two hours I'd expect the lander could have drifted into any orientation relative to the surface and been damaged on the second landing. But clearly it survived if we have telemetry from ROMAP for the third landing. I don't know if we have confirmation of any telemetry after the third landing, so it's current status is unclear.

Still, I suppose we may get telemetry even if the lander came down upside down. That's got to be a very real possibility now. I wonder how much science could still be done in that scenario.

About drifting between landings, that depends strongly on the ratio of horizontal to vertical velocity after the first landing, and that's clearly very uncertain now. For a short bounce, zero horizontal velocity means no drift, even though the comet is rotating. For a higher bounce, Coriolis effects due to the rotation should become important, but it's not clear off the top of my head what drift might result.

Ulamec said two hours after the first touchdown signal, the indications of "turning" due to the flywheel stopped, but that they still had a radio link and data past that point. Sounded like it was gone as expected by the time of the media briefing.

Since the fluctuations of the signal were so visible during the turning, I wonder if that means the lander finally settled right side up? I don't know anything about the radiation pattern of Philae's antenna, but wouldn't the signal be weaker if it were on the side or top plate?

Just got home from work, and have only been able to follow events sporadically. Question/thought: If the lander is indeed upright, would it be possible to firmly secure Philae by just embedding the sampling drill without retrieving a sample?

Fully understand that would be a MAJOR sacrifice in science return, but the trade-off would presumably be allowing the rest of the instruments to execute a full campaign.

Not anyone's preferred option, obviously. But is this an option at all?

It's a chicken-and-egg problem: you need the other anchoring in place to be able to use the drill. Used by itself, the drill might spin/rattle the body around instead. If the craft is upright after next contact, they'll assess the use of the screws again and possibly the harpoon, depending on ability to reuse them. There is plenty of contact, imaging, and sounding science to be done while they can.

Possibly, but it only needs to reach Rosetta, since the craft are set up as a relay system. The main problem in this case would be that most of the instruments and cameras would be pointed the wrong way. Hope for the best!

I realize that. I was just trying to find comfort in the fact that Ulamec thought it worthwhile to mention the fluctuating signal during the bounce, but didn't mention anything about the final steady signal being weaker than what they had been seeing during descent. Again, though, I guess the significance of that depends on whether or not the antenna's radiation pattern varies with orientation.

I imagine that the actual location of the lander is another variable that might affect expected signal strength.

EDIT: MarsInMyLifetime, I forgot to add that my drill idea assumes at least one footpad anchor was embedded (somewhat, anyhow).

That does make a difference. But in that case, if they were able to drill with only one anchor, I wonder why not finish the sampling? Was anchoring required for any other tests? I hope that the bounces did not break off or compromise the other instruments that depended on contact/compression data.

The ice screws use the energy from impact to drive into the surface - Reference
Maybe they were able to dig in at the third landing.

The harpoons use a gas generator, so they are certainly single-use.
Dr. Ulamec seemed certain at first touchdown that they had fired. So they are probably not an option, unless he was mistaken.


Edit: Someone asked for info on SD2, here is a paper on it
Click to view attachment

Thank you for that link. It answers many questions and reminds us why anchoring is critical. Fingers crossed!

For comets specifically, yes, but not planetary landings. A lot of landers bounce (Pathfinder especially);
Huygens bounced/skidded out of the hole it made. But more to the point one of the Surveyors bounced a bit IIRC,
and one Surveyor was even commanded to take off again and hop a little to one side.