Rosetta Ma’at 03 Landing Site- 03 Site Matched to 02 Seating


(See also the update with gifs at greater accuracy. It’s at the end).


The pit known as Ma’at 02 (also Deir El-Medina) is at lower left and Ma’at 03 is at upper right. Ma’at 03 used to nest into Ma’at 02.

Large orange dots in the first header- these show (a) the final landing site for Rosetta in the pit known as Ma’at 03 (upper right) and (b) its former seating point in Ma’at 02 (lower left, within the cluster of frames). This is because 03 was originally seated on 02. 

Small orange in the second header- two lines extending from the landing site and seating dots. These show the seating, within the blue cluster of frames at Ma’at 02, that correspond to the orange dots at Ma’at 03. The terrain photographed in the red frames to the lower left of the orange dots at Ma’at 03 would not have had its seating photographed at Ma’at 02.

Fine yellow lines- a ridge at Ma’at 03 and its corresponding seating at Ma’at 02 


This photo was posted on the Rosetta blog on 15th December 2016. Here’s a link to the post.

The original shows all the pictures that were taken on the approach to Rosetta’s final landing site. That’s why the frames get smaller and smaller. They run from dark blue to red in chronological order. 

The dark blue and light blue frames are clustered over Ma’at 02. The frames then stretch out to the landing site at Ma’at 03, getting smaller and turning red. 

Since this blog has dealt with the delamination of Ma’at 03 from Ma’at 02 (and both from Ma’at 01), it seems appropriate to cross-correlate the landing site in Ma’at 03 with its former seating position nested in Ma’at 02. The header shows this (the large orange dot). 
The second header shows (using extra orange dots) the sweep across Ma’at 03 that corresponds to the blue frames showing the seating for this line at 02. To the lower left of the last orange dot in Ma’at 03, pictures were taken but they would correspond to a seating position that is outside the cluster of blue frames at 02.

It’s important, of course, to know that these two strips of terrain, denoted by lines of orange dots, correspond to a layer (containing Ma’at 03) and its seating (Ma’at 02) because all manner of comparisons can then be made using the close-up data collected during the last few minutes of the incoming trajectory. 

This post will be updated over the next few days to show the actual layers the pits are sitting on. Although the pits look to be translational matches anyway, this is by no means the basis on which the nesting of the two pits is based. The adjacent layers they sit in will themselves will be proof enough. Once the layers are slid back in a reversal of the ‘stretch movie’ the holes simply cannot help but nest over each other whether they look uncannily similar or not. 

UPDATE 19thDECEMBER. Here’s the rest of the post as promised:

Photo 2- Ma’at 01, 02, 03 from above, without the OSIRIS frame footprints. ‘Original’ is an old annotation hence Ma’at 02 still being dotted. 


Light blue- from left to right, Ma’at 01, 02 and 03. 

Orange- landing site (on the right, 03) and seating (left, 02). 

Yellow- matching ridges.


Photo 3- the red layers.
Copyright ESA/Rosetta/üs – CC BY-SA IGO 3.0/A.COOPER

Red- these are real, original layers that delaminated along the long axis direction at the shear line on the body. This happened before the head sheared from the body but it was herniating from it at the time. We know this because the head layers are angled upwards and flared-out in a bell rim shape. These red layers will be matched to their body counterparts in photo 12. They are stated as being real layers as opposed to the quasi-layers that are described further down (bright green in later photos). Although they’re at an angle to the long axis, and the quasi-layers, their delamination vector was in line with the long axis. This is because that was the principle tensile vector of stretch due to the centrifugal forces of spin-up. 

Photo 4- a side view from Ma’at.
Copyright ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0/A.COOPER

Red- as before. The upper line (now on the right) is only apparently split in two. It’s continuous, but it goes round the promontory and reappears in the distance. 

Photo 5- a view from above. 

Regarding the right hand line in photo 5, from this viewing angle this line is the upper of the two lines when viewing 67P in ‘upright duck’ mode. However, it’s on the right in this view from above the head. Where this line progresses above Ma’at 03, which is below and to the left in this photo, it becomes almost indiscernible, especially in other photos as it goes across the dust towards the outcrop above it in the foreground. But that line is clearer here. It’s a very spidery line outlining what are more rounded zig zags. If you’re not convinced you can move over to Ma’at 03 and trace its more obvious zig zags round its front lip to a point where this spidery line starts (easier to do in photo 7). You can then disregard the spidery line, carrying on round the pit rim on the stronger zig zags, then parallel to the spidery line. If you did this same operation for 02 in photo 7 as well as for 03, you’d find yourself tracking at exactly the same angle and distance from their respective red lines, which is telling in itself. 

Photo 6- another top-down view. 

Photo 7- This is the Rosetta landing site ellipse photo from a Rosetta blog post. 
Copyright ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0/A.COOPER

As you can see in photo 7, the left hand annotated orange dot denoting the actual landing site is inside the ellipse so it was on target. This is the best photo for tracing certain parts of the ‘upper’ line as viewed in upright duck mode. That would be the left-hand line from this side view and its runs between 02 and 03. If you view the original, unannotated version, there’s a stronger line between 03 and the its bottom end than in other photos. This line is tightly zig zagged like much of the rest of this line and the other one. This clearer section happens to run parallel and close to the yellow landing trajectory line. Also, the upper section as described for photo 5 is clearer here than in most other photos. Notice the lower red line bifurcates at its lower end in this photo. The small dots denote the line as shown in photos above. The larger ones running down from the bifurcation follow another line which is now more favourably lit. This isn’t cherry-picking. This line looks clearer and the old one really is a bifurcation because everything was delaminating along this vector including mini-delaminations being dragged out from the main layer front as the next layer slid away from it. There are several such mini-delaminations along the rim section just below this area. 

Photo 8- same as photo 4 but with quasi-layers added in bright green. 
Copyright ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0/A.COOPER

Bright green- the quasi layers. These tore across the real layers when the head finally sheared from the body. They recoiled at 90° to the head rim and their counterparts recoiled in mirror fashion on the body. This is why Babi on the body below looks scalped and is known to be a gravitational low. It also has a mass of bunched-up terraced material at its far end, which is the recoiled body material. That’s why it’s bunched-up. This is dealt with in Part 41. The red lines weren’t quite as well identified in the photos for that part and they’re light blue, not red, for various reasons. 

The recoil happened in three stages: first, the top layer tore along the shear line and sprung back. That’s the lower-right bright green line in this view which is the upper line (up the head lobe) in upright duck mode. Then as the head herniated more, the second layer tore along the shear line and recoiled less. That’s the middle bright green line. Then the entire head lobe finally tore from the body. That was the third stage and so the upper-left bright green line is the actual head rim. So we have three lines denoting 2 recoiled layers and one head rim. In this sense, the head rim denotes the bottom of a quasi-layer as well because it too tore across the real layers. It didn’t ride up the head lobe because it constituted the bottom of the entire head lobe, which was moving up as one lump after head shear. At the instant of shearing, all tensile stress had been transferred to the stretching neck.

This three-stage scenario was also laid out in Part 52 which goes into more detail for the mirror-image Babi recoil and the full narrative runs from Part 38 to Part 41. 52 is a summary of the bright green recoils with an emphasis on the matching spot on the body for Ma’at 02. This summary of the recoils in this part is with respect to how they cut across the real layers on the head lobe. It’s only included because the quasi layers look more real than the real layers that are supposed to have dominated proceedings during stretch before the head sheared. Of course, the real layers did dominate because they were long-axis delaminations and the quasi layers didn’t exist until just before the head sheared fully. After shear, the real layers presumably would have concertinaed in on themselves somewhat, along their length as they were relieved of tension after the tear. In other words, they would have gone flaccid as they bunched up along their length as if guided by the layer fronts as rails (the red lines). That may be why they have zig zagged fronts. This bunching-up certainly appears to have happened to the real-layer components within the Babi quasi layers- it’s probably why they’re tightly terraced, i.e. they went flaccid too. 

On shearing, each of the two head lobe (and body lobe) quasi layers failed, one after the other. They simply failed at whatever depth the shear resistance couldn’t resist the tensile force of the herniating head. That depth is visible today as the rebate that runs along under the two right hand bright green lines in photo 8, above. 

Photo 9- as photo 8 with annotations added.
Dark green- this is an obvious smoking gun for the quasi-layers tearing at the shear line and riding up the head (towards the right in this view). The zig zag links the first quasi-layer to the head lobe rim. This particular nesting together of quasi-layer to rim has been made in previous parts (34 and 62) and by other means (matches along the rim and quasi layer). This zig zag was first mentioned in Part 35 except it was only the bottom component or V-shape that was recognised. It’s now known that that Ma’at 01 was actually formed by concertinaing open to form the second zig zag. That’s why it can be seen here sitting inside the second V-shape. It sits directly above Ma’at 01A (Part 62) and both were supplied volatiles from the 01A source already identified along the shear line in Part 62. The first zig zag was very well-matched in Part 35. It clearly collapses to one line when you reverse the ‘stretch movie’. 

Yellow- three promontories which collapse together when we reverse the stretch movie. The two left hand (lower ones in upright duck mode) collapse by virtue of being part of the zig zag. The upper promontory on the right rifted cleanly as part of the upper quasi-layer but also as part of a real (red) layer. That’s because the real layers cross the quasi layers in a lattice and the top two promontories are the intersection points where our two red lines and upper two quasi layer green lines crossed. Indeed, it may be the reason for them being promontories at all. There’s more on this phenomenon below, namely, the fact that the long-axis rifting of red layers dominated for this upper promontory and explain it not dragging up a zig zag as well. These three yellow dots nested at the head rim at the end of the zig zag. Regular readers will know this to be the right-hand wing tip of the much-discussed gull-wings (Parts 5, 7 and 8 to name a few). And, of course, they nested at the shear line on the body. Their seating point is shown in photos below. 

Terracotta- the head rim along the bottom of Ma’at and on the viewer’s side of the so-called cove which hosts the zig zag. The bright green line leading from its top end in this view is actually part of the Ma’at rim on this side as well but it’s marked bright green here because it’s intimately related to the middle bright green line on the other side of the cove, as we’ll see. It’s all Ma’at head rim but the join between terracotta and bright green is the point on the rim where the lower quasi layer started to rift from- except it became head rim on this side of the cove. That’s why there are only two bright green lines on this side of the cove and three on the other side. All will become clear in the following photos. 

Just to recap, the right hand (upper) bright green line in photo 9 used to nest to the head rim which is the left hand bright green line with the dark green gull wings included and a small amount of undotted cove (usually yellow). This means the two red lines, denoting the real layers, were sandwiched under the upper quasi layer and above the lower quasi layer. This was already dealt with in Parts 52 and 62. Above the upper quasi-layer line (right hand bright green line) we return to the large expanse of Ma’at that hosts real layers right across it, as you’ll see in photos 10 and 11. All these layers are concentric onion layers and are perfectly in line with the red lines identified in this part. They’re also parallel like the red lines and have roughly the spacing apart. This makes sense because the red lines depict real layers as well. It’s just that their signature was smudged by being buried one quasi-layer down as well as extra smudging via the head herniation and consequent recoil of the quasi-layers. That’s why they’re less obvious until we zoom in, as we’re doing here. 

The real onion layers across Ma’at may look like straight, flat slabs because the lines are very parallel. However, Part 28 explains why they’re concentric onion layers. 

Photo 10- A long view of the above annotations. 
Copyright ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0/A.COOPER

Photo 11- same as photo 10 with additions. 
Dark green- the zig zag.

Yellow- the nesting promontories.

Dark blue- these are the random points at which the lines in photos 8 and 9 disappear over the horizon. That is, their own local horizon. Ma’at 03 finds itself beyond an imaginary line drawn through all four blue dots which suggests it should be over the horizon in photos 8 and 9. The reason it is actually in view in those photos is that there are no local humps or ridges between it and the viewer even though it’s quite close to the horizon. 

Small bright green- this small curve is what was dotted in photos 8 and 9, where the upper, bright green quasi-layer appears to favour snapping to the real layer for that short stretch. That’s due to red and bright green crossing at this point, as mentioned above for the upper promontory. Photo 10 shows the obvious average quasi layer line from a distance and does so with the larger bright green dots, i.e. without doing the jump or snapping to the red line. But the small dots here in photo 11 are included since that line was traced in the earlier photos. 

Photo 12- a close up of photo 11 with additions.
Red- the main thing to look for in photo 12 is the two red lines on the body. These are the seating points for the two red lines on the head. 

Bright green- this line on the body is where the rim was seated (Part 1). However, it’s also where the lower quasi layer hosting the two red lines on the head also once sat. We know this because the middle bright green line nests to the bottom line which is the head rim. It does so via the zig zag collapsing and it’s also proved via the matches in Part 62. So the left hand section of bright green line on the body matches to the middle bright green line vertically above it on the head. The curved piece of bright green is usually yellow and that’s the cove on the head and its body seating, the so-called ‘horseshoe crater’. Cove and horseshoe were matched in Parts 3, 5, 8, 18, 36, and 62, one of the most exhaustively matched parts of the comet. 

Dark green- on the body, this is the seating on the body for the zig zag. It’s therefore the seating for the gull wings. That match was made in Part 5 and is a side-on match confirming the top-down matches in Part 3.

Light blue- on the body, these are the seatings for Ma’at 01, 02 and 03, obediently respecting their correct positions on the red layers. Notice how 01 is a semicircle because the source is under the gull wings. That’s how the gull wings came to be two pushed-up humps (Part 7). The curves of the humps are whited out here (see Part 5 for close ups). 

Small light blue- Ma’at 02 perimeter and its source on the body. This source looks like a complete guess here so for verification you’d need to see Part 62, also 52. The photos in those parts show a well-defined semicircle with a provenance linking it to the 01/01A source before long-axis delamination dragged it along the shear line (hence Ma’at 02 being set back from 01 as well as being above 01). Light blue also shows a ridge running from Ma’at 02 that’s matched at its seating and on the head rim that was sandwiched between the two. This triple-match is shown in detail in Part 62 (Yellow annotations in that part). 

Yellow- five dots, four on the head and one on the body. Three of the four on the head are the points of the zig zag hence their being promontories. These cant help but concertina back together when reversing the stretch. In doing so, they collapse the entire cove to the head rim (Part 34). This collapse includes the ‘scallops’ defined by ‘ribs’ running round the cove from each yellow promontory. The ribs run to a similar set of promontories on the other side of the cove. Part 34 was stumbling very slowly towards understanding the quasi-layers. It was realised at the time that the scallops in the cove seemed related to these two very long lines that ran past the cove either side and piggy-backed along the ribs of the cove as they traversed it. That was all that was known. The long-axis delaminations weren’t known about at all at that time so the reader needs to bear it in mind that they have superior knowledge to the the writer at the time of writing. Long-axis delaminations were discovered soon after Part 34 and rolled out in Part 38. 

The fourth yellow dot on the head is the upper one which rifted away without dragging a zig zag with it. This probably has something to do with being on a red line, or real layer, and therefore subjected to a genuine long-axis delamination along the shear line before the quasi layers even existed. Only after that clean, longways shunt did it then move cleanly upwards with the first (upper) quasi-layer. The zig zags join two quasi layers which would be the lower layer to the head rim and, further above, we designated the head rim as a notional quasi layer. These two layers are not real layers and so the zig zag displacement scenario (sandwiched layers opening upwards, up the cove, into zig zags) was very different from long-axis delamination of the red layers. This is despite the portion of the red line above the zig zag taking on a the zig zag match (the yellow promontory) by virtue of rifting from the same original point at the shear line. 


This part shows the small areas of Ma’at 03 that match to Ma’at 02 and that also got photographed by Rosetta as she descended to her final resting place. This may lead to some interesting data comparison for these two areas. 

The rest of this part is dedicated to proving the existence of the red, long axis layers that delaminated along the shear line before the head lobe sheared. This is in order to prove that Ma’at 03 delaminated from Ma’at 02 and therefore prove the assertion that Rosetta took pictures of matching areas for the two pits. 

In proving the existence of the red delaminations that are real, long-axis layer delaminations, it was necessary to explain the quasi layers which seem to dominate the area in question. 

The quasi layers did actually facilitate pit delamination along a different vector from the red delaminations but that is beyond the scope of this post- they didn’t cause the 03 from 02 delamination, only the long-axis vector for the red layers did that. 

Quasi-layer pit delamination will get its own post in due course. In the meantime, Part 52 is almost entirely related to quasi-layer pit delamination. It just doesn’t explicitly say so because it’s fairly obvious. The long-axis delaminations weren’t an issue for that part because we were analysing the displacement of pits up and down the comet from their shear line outgassing sources. The shear line sources were set in stone, so to speak, after long-axis delamination. Therefore, long-axis delamination e.g. 03 from 02 wasn’t within the scope of that part: 02 and 03 had already delaminated along the shear line, along with their delaminating sources prior to the quasi layers distributing them as a ready-made pair up the head lobe. 

UPDATE 26th December 2016​​

Above are two gifs showing the Ma’at 02 seating more accurately. The first one is two zoomed-in crops from the header photo. They’re zoomed in by the same degree so we know they match in terms of size. This sizing confirmation has then been used for two different photos in the second gif. These have been cropped and resized so as to match up the same features as in the first gif. This second gif allows us to see the Ma’at 02 seating without the superimposed photo frames and also see both 02 and 03 closer up. However, the perspective is different in the two photos, rather in the same manner as the perspective anomaly described in Part 67, photo 3. This means there’s a small amount of drift between the two gif frames but the angle change is about 40°-50° not 100° like in Part 67.

The white-dotted ridge and other colours are due to this being a photo from Part 67. The white ridge is mentioned below. 

The location of the seating in Ma’at 02 for the red squares shown in Ma’at 03 can be inferred simply by looking at any red square and then noting the terrain that replaces it in the Ma’at 02 frame component in the gif. 

These gifs show the seating to be a few metres further into the pit of Ma’at 02 than in the header photos for this post. This is because a different fiduciary point was chosen along the ridge by checking the Ma’at 02 overhead close-up from Part 67 and comparing it to the Ma’at 03 zoomed-in crop from the header. By moving the fiduciary point along the ridge by 15-20 metres, it moved the Ma’at 02 seating for the last square (orange dot in header) from the pit rim, down the slope, and onto the so-called rugged terrace. That’s the lower section of 02’s consolidated material, at the bottom of the main slope. The rugged terrace’s smaller cliff borders the Ma’at 02 dust but the rugged terrace itself is for all intents and purposes, inside the pit. 

You can see that the last red square, which represents the last photo taken at 20 metres’ altitude, is just into the dust of the Ma’at 03 pit. Another six or eight red squares along the track leading up to that point also traverse the dust near the 03 perimeter. For their Ma’at 02 seating, all these frames pass over the consolidated material of the rugged terrace in the second gif. For the first gif, the end of the line of squares is a couple of squares’ width into the Ma’at 02 pit dust as explained below. The accuracy of the seating for the last square is two or three square-widths but that’s still pretty accurate. It corresponds to about 15-20 metres and is commensurate with the moving up and down the ridge to different fiduciary points and also sizing and aligning the 03 crescent and other 03 features with their seatings. However the first gif confirms these error bars because it shows the seating as being at the end of the crescent (upper swallow tail tip) as well but just past the rugged terrace and into the dust of Ma’at 02. In other words, it’s less than two square widths away from the second gif’s seating. Since this is a work in progress, it will be revisited as other photos are found that allow better constraining of the fiduciary points. 

The accuracy of the line as a whole seems more accurate than the debate over the last square’s seating. This is because the end of the line is drifting around, left-right in this view, by around two squares but the line itself runs solidly through the middle of the crescent in both gifs. This is fortunate because it’s more important for data comparison to nail down the line of red squares to its seating line than knowing exactly where the line ends. 

It’s always best to hold a pen tip against a particular feature to see how well it matches. It’s usually a closer match than it appears to be because changing shadows lull you into thinking the features are being displaced by the same amount as the shadows. 

This is a work in progress, hence the matches aren’t annotated. The main matches are:

-The crescent shape which the red frames run through the middle of in their path towards the dust in Ma’at 03. This crescent can be seen in the Ma’at 02 frame as well, especially in the second gif. This match is especially the case for the crescent’s top perimeter that’s at a 45° angle in the frame (down-left to up-right). This crescent shape is the swallow tail for the flow feature described at length in Part 67. The swallow tail tips are on the rugged terrace. The apex of the crescent formed by the swallow tail is halfway up the slope between rugged terrace and the main pit rim. That is, the main rim that gives way to the flows in Part 67. 

– another 45° line just above the 45° line mentioned above. The two short, parallel lines together form a matching feature between the gif frames. 

-The ridge cutting across the top of the frame.

-The curve leading down vertically from that ridge and arcing round to the left. 

– A faint flow feature/depression at lower right. This is on a slope which you probably know if you’ve read Part 67 so it’s not a gravitational low depression, just a slight dip stamped into the slope. 

– A very faint hint of the white-dotted ridge in Ma’at 02 apparent in the Ma’at 03 image. Just the faintest of shadows. 

-Other spidery features which are left to the reader to spot. 




Copyright ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0

To view a copy of this licence please visit:

All dotted annotations by A. Cooper.