Help Shape the Future of Space Exploration

Join The Planetary Society Now  arrow.png

Join our eNewsletter for updates & action alerts

    Please leave this field empty
Facebook Twitter Email RSS AddThis

Headshot of Emily Lakdawalla

Rosetta imaged Philae during its descent -- and after its bounce

Posted By Emily Lakdawalla

17-11-2014 17:10 CST

Topics: Rosetta and Philae, pretty pictures, pics of spacecraft in space, comets, mission status, comet Churyumov-Gerasimenko, spacecraft

This morning ESA released a set of images of the Philae lander taken by the Rosetta orbiter during -- and after -- the lander's first touchdown. If the touchdown had happened nominally, one of these images would have shown Philae sitting on the surface. As it happens, the images contain evidence for the spot Philae first touched the comet, and a crucial photo of Philae's position several minutes into its first long bounce, showing that the lander took a sharp right turn after hitting the ground.  

OSIRIS sees Philae multiple times during landing

ESA / Rosetta / DLR / MPS for OSIRIS Team MPS / UPD / LAM / IAA / SSO / INTA / UPM / DASP / IDA

OSIRIS sees Philae multiple times during landing
The sharp-eyed OSIRIS camera on the Rosetta orbiter snapped numerous images of Philae as it descended toward its touchdown on the comet on November 12 at 15:34 UTC. Images documented the spacecraft rotating, and also saw evidence of the lander's touchdown on the comet surface. One final image, captured 9 minutes after the landing, sees the spacecraft bright against the shadowed surface, heading to the east on its first bounce.

Let's just pause for a moment to appreciate how awesome that series of images is. Especially the final one in the sequence, taken at 15:53, showing the sunlit spacecraft silhouetted against a black shadowed cliff. The lander is headed off in a completely different direction than the one from which it approached -- at least from Rosetta's point of view. (From the comet's point of view, the lander was descending vertically, then bounced east.)

According to navigational data supplied to the Jet Propulsion Laboratory by the European Space Agency, here is some detailed information about the positions of lander and orbiter at the moments those pictures were taken. There is a full minute-by-minute table of altitude and position for both Philae and Rosetta calculated by Joe Knapp from SPICE navigational data available at Data for Philae ends at the moment of first touchdown, because they don't yet know Philae's post-impact trajectory. It shows Philae on the ground at an altitude of 158 meters, which must be measured relative to a reference ellipsoid for Churyumov-Gerasimenko.

Date Time Philae altitude (km) Sub-Philae Longitude Sub-Philae Latitude Rosetta Altitude (km) Sub-Rosetta Longitude Sub-Rosetta Latitude
2014-11-12 15:14:00 1.337 149.62 -27.28 15.179 174.19 -27.54
2014-11-12 15:19:00 1.068 148.12 -26.72 15.176 172.21 -27.41
2014-11-12 15:23:00 0.848 147.05 -26.18 15.175 170.63 -27.30
2014-11-12 15:35:00 0.158 144.82 -23.44 15.181 165.90 -26.96

I wanted to get a wider view, so I plotted the spots where we see the descending lander on two context images, both taken September 14, the first by OSIRIS, the second by NavCam:

Philae's ground track on an OSIRIS context image

ESA / Rosetta / DLR / MPS for OSIRIS Team MPS / UPD / LAM / IAA / SSO / INTA / UPM / DASP / IDA / annotated by Emily Lakdawalla

Philae's ground track on an OSIRIS context image
Rosetta's OSIRIS science camera photographed Philae several times during its descent on November 12, 2014. This map shows the locations where OSIRIS saw Philae on a slightly wider view of the comet that was taken on September 14, 2014 from an altitude of 30 kilometers. The image scale is about half a meter per pixel. The intended landing site is marked with a small white cross.
Philae's ground track on a NavCam context image

ESA / Rosetta / NavCam / Emily Lakdawalla

Philae's ground track on a NavCam context image
Rosetta's OSIRIS science camera photographed Philae several times during its descent on November 12, 2014. This map shows the locations where OSIRIS saw Philae on a global view of the comet that was taken on September 14, 2014 from an altitude of 30 kilometers.

It's really difficult to imagine what it would have been like to ride on Philae as it climbed and climbed on its high bounce above the surface, drifting eastward over time. We know that its second bounce happened about two hours after the first one, and it was a short bounce, and when it came to rest it was wedged up against a cliff. Where, exactly? Hopefully OSIRIS will figure that out soon -- but the fact that Philae's solar panels only saw light for 1.5 hours per day, the orbiter has to be in the right place in the right time to see the shiny spacecraft sunlit.

Another cool thing about the OSIRIS images is that we can look at changes to the surface of the comet that were caused by the landing. The community at has been actively examining these images and trying to understand them. And I have to say that the post-landing changes are quite puzzling, at least to me. You see several circular spots that look like mini-craters, and want to interpret them as footpad locations. But there are too many of them, and at least one of them is too far away from the others for all of them to have been made at the same time. Are any of these footpad locations? Do these spots look circular by some trick of the light, but are actually deposits of darker material on the surface? It's hard to imagine how you darken a comet. It's all kind of confusing; I'll leave you with an animation that shows what I see, but I'll tell you I'm really looking forward to the ESA operations team's interpretation of these photos, because I don't think I'm quite getting it!

Puzzling out the changes at the Philae first landing site

ESA / Rosetta / Philae / ROLIS / DLR; ESA / Rosetta / MPS for OSIRIS-Team MPS / UPD / LAM / IAA / SSO / INTA / UPM / DASP / IDA; Emily Lakdawalla

Puzzling out the changes at the Philae first landing site
The two images in this animation come from Philae (the ROLIS descent camera image taken from an altitude of 40 meters) and Rosetta (an OSIRIS image taken from orbit just after the Philae landing). There are several visible changes between the ROLIS "before" and OSIRIS "after" images, but it's not clear what story they tell about the landing.
See other posts from November 2014


Read more blog entries about: Rosetta and Philae, pretty pictures, pics of spacecraft in space, comets, mission status, comet Churyumov-Gerasimenko, spacecraft


ethanol : 11/17/2014 06:00 CST

Those landing marks are pretty hard to wrap my head around. Are we sure that the harpoons didn't fire? I recall that they had been activated, along with the mechanisms for spooling them back in, but they decided that they hadn't actually fired. Was this conclusion from some other telemetry, or just from the fact that it didn't stick? It's just that if Philae somehow landed at an angle on two legs (but how would it end up like that after the most recent ROLIS pic?) I could almost justify the other marks as ineffectual angled harpoon shots. The only other possibility I can come up with is some sort of "walking" effect from precession of the lander as it tries to level with the surface against the resistance of the gyroscope (no, I don't fully understand how gyroscopes function). But then how would it have enough energy for the bounce? Count me as befuddled.

Brian: 11/17/2014 07:41 CST

Is it just me or is the lander spinning as it's descending? If it was spinning, and the flywheel was stopped at first contact, that could add up to quite the gymnastic vault at touchdown and tumble during the bouncing phase of the landing. Any data on spacecraft attitude available?

MadeKetir: 11/17/2014 08:44 CST

Philae took a 90 deg change of trajectory on that bounce. Must have been on a significant slope. The gyro worked to keep her level. So the 3 legs would have hit sequentially and not in any kind of neat triangle. Could the lander body have rolled out an imprint between the 2nd and 3rd leg contacts?

Wade: 11/17/2014 10:43 CST

I think it unlikely that all legs touched at the same instant, and remember, the vehicle is moving in slow motion. I'm guessing that the "but then what's this" crater was where the right leg hit. Assume a slight downrange velocity component, pointed roughly along the line made by the 3 rightmost circles in Emily's gif. If the top leg in the photo hit first, the vehicle would begin rotating CCW (it's dragging a wing), and pitch forward slightly, resisted by the flywheel. It would also kick rightward on the bounce. The trailing left leg then touches and drags. I'm not sure what happened with the screws, but I assume they deployed and didn't hold. If they even partially deployed, then there's another set of landing gear down there. The harpoons didn't fire but they hang down there somewhere too, and it's possible for the vehicle to bottom out. So there has to be something these items did to the terrain. Now, if the right leg started farthest from the surface, it hit last, and if the top leg had done it's thing and was already pushing right, and/or the misc parts under the vehicle just held it up, the right leg landed in the "but then what's this" crater. It's only 1 vehicle diameter beyond the perfectly vertical landing spot. Being the downrange leg, it would have borne the brunt of the impact, and that crater definitely is as deep as the others. Everyone gets 1 free guess, and that's mine.

johnt: 11/17/2014 11:57 CST

It cart wheeled after initial contact...simple... measure the distance between the lander feet and the center of the contact points and you will have your answer

Paul McCarthy: 11/18/2014 12:44 CST

Minor point: is there a slight mismatch between "the final one in the sequence, taken at 15:53, showing the sunlit spacecraft silhouetted against a black shadowed cliff" and the image labels which show 15.43? Emily: do they know for sure they got a drill sample yet? (As per my comment in your last Blog post).

Wade: 11/18/2014 01:04 CST

Paul MCCarthy: They know they did not get a drill sample. (from somewhere in tweetspace, from the COSAC PI)

Marco: 11/18/2014 01:12 CST

Hi Emily. I don't think Philae could have cartwheeled on impact as the fly wheel would have limited that. The forces acting on the lander would have been in the form of an impulse on a rotationally stabilised Philae.Ie. Philae would have tipped as one leg at a time hit the surface, but on a hard surface with tethering and jet failing, only a portion of the momentum would have been damped and the rest of the reaction would have turned back to kinetic energy of the lander. I think the protruding "rock" in about the landing spot would have given it its sideways kick.Possibly the body of the lander itself could have hit it and given it the sideways ricochet without the damping that the legs have inbuilt. The "what's this" could have been the harpoon or screw hitting separately.

stone: 11/18/2014 03:40 CST

The tweet from a journalist is a shortened version of a longer interview with the COSAC PI and the statement was not that there was no sample delivered, because COSAC has no means to know if there was sample, but the amount of organics in the sample is close to zero.

LJ: 11/18/2014 04:54 CST

What bothers me here is the "ground track". Rosetta was nowhere "above" the Philae when these pictures were taken. If it was, it would have crashed on the surface after. So all the pictures are taken from an angle, which was probably also changing with time. The position of Rosetta should be taken into account when interpreting these images, along with the rotation of the comet. So unless the bounce was in the direction of Rosetta, it wouldn't seem like a straight line in the pictures. Because of all that I don't think that Philae really was directly above those points on the comet surface, except on touchdown. It would only look like that from Rosetta's perspective. So in my opinion the ground track should be drawn from a different perspective, e.g. from the first touchdown point. It would look a lot different then.

lakeman: 11/18/2014 07:25 CST

Thanks Emily for all your hard work in the last week or so. I definitely got my subscription moneys worth. As for the trolls take no notice of them, i for one think you would make a very good job of it . Dave

gozlemci: 11/18/2014 07:34 CST

Hi Emily; you have provided best info so far on Philae flight. Thanks...

deezee: 11/18/2014 05:43 CST

Greetings,,, I'd like to complement you on your info-sharing and especially this visualization against larger views of the comet. I was asking collegues for any clarification on the approach path -relative to the comet's tumble and they offered much, including this page. One detail offered by them was that the lander was approaching in line with the lobe's motion around the axis. If true this would help to understand what may not be grasped in a static image. After TD1 the lander would have slowed but the lobe would continue-on, so might the apparent deflection and long duration of the first bounce be visualized as the lander lagging behind as the lobe advanced? If so, then might the lander have hooked back to the right and retreated toward the neck as it tried to catch up with a slower moving area nearer the axis? So to find it we u-turn back and to the right of the last position. For the first ten minutes after landing it lost its relative timing to intercept so now might the lobe move away and the lander takes a long plumet toward the departing lobe?

Marco: 11/18/2014 08:44 CST

He went Thatataway <*Points madly with fingers*>

Moe Ali: 11/18/2014 09:48 CST

It's my understanding that the gravitational escape velocity off the comet is 0.5m/s. Meaning if you were standing on the comet and decided to jump up, as long as your upward jump was under 0.5m/s, you would come back down at some point. The Lander was descending at twice that speed, so it's amazing we didn't lose the lander into space after it bounced the first time. If the harpoons went off, they might have slowed it down the descent and thus reducing the bounce back after the first impact. There was a small thruster that was in place to push the lander down and reduce the bounce, but that didn't fire. So maybe the harpoons did go off? I don't know….. but I also want to say that Emily did a fantastic job in covering this mission. You were so lucky to be there right to the last historic moment! Keep up the great blogging, it's very much appreciated.

dewege: 11/19/2014 03:31 CST

Looking at the circles in the gif, three more (unmarked) can be seen just upper right of them. They are not caused by the TD because they are there before. If I assume that the dust layer is similar to dry powder snow, such "craters" could be caused by a blowout of gas from the rock beneath the dust. So my interpretation is that the "But then what's this"-craters could be similar blowouts which might have been triggered by a kind of landing impact shock wave.

CosmosQuest: 12/08/2014 11:12 CST

Emily, and Karen This is just a test to see if the comments feature has been restored to the blog posts, and if it is working normally.

Karen Hames: 12/08/2014 05:56 CST

Thanks Mark! Everything seems to be working normally.

Leave a Comment:

You must be logged in to submit a comment. Log in now.

Space in Images

Pretty pictures and
awe-inspiring science.

See More

Join The Planetary Society

Let’s explore the cosmos together!

Become a Member

Connect With Us

Facebook, Twitter, YouTube and more…
Continue the conversation with our online community!