Part 66- The Final Approach Ma’at 02 mosaic 




This post is concerned with the viewing perspective of one of the last photos taken by Rosetta as she approached the comet on her controlled crash trajectory. This occurred on the 30th September 2016 at about 10:20 UTC and the photo was of the pit called Ma’at 02, which was the focus of Parts 62 to 64. The mosaic in the header is in effect a single photo of the area around Ma’at 02 and in this sense it’s is a bit confusing because the constituents are overlapped. It’s even confusing once it’s been stitched together because we’re looking at Ma’at 02 ‘upside-down’ in relation to the usual views we have of this pit. It’s something of an illusion and will be explained below. 
Ma’at 02 is also known as Deir El-Medina. That’s its recent new name but its still referred to as Ma’at 02 for clarity and brevity on this blog.

The upside-down illusion regarding Ma’at 02 will be crucial in understanding an upcoming post which is to do with two similar photos of Ma’at 02 that were used in a scientific paper. 

For readers wondering what happened to the rest of the Ma’at pit delamination series, this will be resumed in due course with at least two more parts to add to the three parts mentioned above, 62-64. They will be presented with a reminder that all five are linked as a series. 


Photo 1- the mosaic (header reproduced).


The mosaic of Ma’at 02 is essentially a radially exploded photo. The pit is in the middle and every photo to the left, right, up and down overlaps with the last one towards the pit at the centre as you progress to the edge of the frame in four directions. Put another way, if you cut out all the photos and then concertinaed them together carefully so there were no overlaps, you’d see a single, faithful image of the pit and its surrounding terrain (see photo 2).

Photo 2 (2nd header, reproduced)- this is a stitching together of the constituents of the mosaic as described above. H/T to Gerald, a Rosetta blog commenter, for doing this so neatly. Gerald did just the stitching. All later annotations on this stitched version using coloured dots are mine, and of course the usual ESA/OSIRIS credit applies for the photo mosaic itself. 


It might seem as if that’s that and we can move on to other things. After all, if we toggle between the stitched view and the classic overhead view of Ma’at 01, 02 and 03 that was presented in Parts 62-4, the two seem to look as if they’re from a similar perspective, one viewpoint tipped up just a bit with respect to the other:

Photos 3 and 4- toggling between the stitched and classic views. Ma’at 01, 02 and 03 are dotted blue and 02 is in the middle. 



However, looking more carefully, we can see that the stitched viewpoint is actually upside-down when compared with the usual ‘upright’ views we get of Ma’at 02. 

These classic views are shown in photos 5 to 7. They’re classic in the sense that the duck-shaped comet is upright i.e. in ‘upright duck’ mode with the head lobe above the body lobe. 

Photo 5- this is a NAVCAM photo taken from above the head lobe.
Copyright ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0/A.COOPER
Ma’at 02 is marked with a single light blue dot in photo 5. This photo shows the head rim as being brighter than the body which helps differentiate the two. The differentiation is less clear in photo 6, which is almost the same view but more detailed because it’s an OSIRIS photo. 

Photo 6- another one from above the head lobe. The toggling view in photo 4 is a close up of this photo. Ma’at 02 is light blue again. 

Photo 7- the classic side view. 
Copyright ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0/A.COOPER

Photo 8- this is the photo 7 view with the two viewing perspectives from photos 3 and 4 superimposed.
Copyright ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0/A.COOPER

As you can see, the two views are wildly off with each other, over 90° in fact, and certainly not similar viewpoints. We haven’t gone ‘upside-down’ yet so it might appear a bit confusing until you get to photo 10, which comes soon. 

Red is the classic overhead viewing direction or vector (the viewpoint in photo 3) and yellow is the stitched view (photo 4). Arrows show the direction of viewing. 

Since we’re viewing both vectors from almost overhead, it’s not showing the vectors’ angles with the surface (~40°) so much as their angle to the latitude and longitude lines of the comet if those lines were superimposed. The angle with the local average surface is known as the altitude angle and the angle with the lat/long lines is known as the azimuth angle. So photo 4 shows primarily the azimuth angles. But since we’re off to one side slightly (to the left) we can see under the two lines just a bit which betrays a little of the altitude angle. Photo 9 shows the altitude angles very well. 

Photo 9- this is a side view which shows the altitude angles of both viewpoints. Yellow is the viewing vector for the stitch. Red is the viewing vector for the classic top-down view. 
Copyright ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0/A.COOPER

But it doesn’t quite end there. Photo 10 gives the true viewing orientation as we look at Ma’at 02 along the yellow stitch vector (and the classic red view for comparison). 

Photo 10- The viewer’s position is upside-down. 
Copyright ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0/A.COOPER

This means that what looks like a similar viewpoint in photos three and four is only apparently similar and in fact 100 degrees off. 

Understanding the capacity for this Ma’at 02 perspective phenomenon to trip us up will be crucial to the next part. It’s going to be instrumental to showing that evidence for recent morphological changes, presented in a scientific paper, is just a mirage.


Photos 11 and 12- photo 11 (top) is the stitched version rotated 180° to the classic upright duck orientation. It now appears very close to the viewpoint of the classic side view (photos 7 and 16, below) but a little lower because you can see under the ledge that’s somewhat below Ma’at 02. Photo 12 is the original upside-down stitch, included for toggling.
Photos 13 and 14- this is just the pit, Ma’at 02. Photo 13 (top) is the original stitch, unrotated and photo 14 is the classic top-down view from photo 4. They look uncannily similar despite being viewed from completely different angles. 

Photos 15 and 16- photo 15 is the stitch, rotated, and photo 16 is the classic photo 7 view.

This pair shows that the views are now almost the same just by rotating the stitch to the familiar ‘upright duck’ view. This is the equivalent of inviting our upside-down lady in photo 10 to stop hanging upside down and view Ma’at 02 down exactly the same viewing vector but sitting ‘upright’ like the rest of us. 



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. 



Part 65- OSIRIS Map Anomalies and Corrections



This post is regarding the OSIRIS maps presented in this Rosetta blog post:

There are several errors in the maps. These were outlined by me and by Marco Parigi in the comment thread for the above-linked post. I found eight errors and Marco found one of them independently. Our two comments are linked at the bottom of this post and mine is reproduced in full a little further below here so that you can refer back to it and mentally tick off the eight points as they’re corrected in this post.

Marco’s find was one of the more large-scale and therefore important anomalies. He has already done a blog post on it so I won’t have to address that particular one here. It concerns the Sobek/Anuket border being placed a few hundred metres apart in different renditions (or viewpoints) of the map. Link here:

Marco’s post corresponds to my point 3 in the first list of four in my comment.


(This is reproduced in full for clarity. However, you can skip it and go straight to the separate points of the comment that are addressed below it).

Hi Claudia

I’ve identified several rather important contradictions on the maps presented in this post as well as several smaller contradictions. The more important contradictions are:

1) While Bes is clearly marked in the first map it disappears in the second one, taken over by Geb and Anhur. It’s not perspective, you can see the same features coloured differently. 

2) In the montage, right hand column, middle image, “Seth” and “Babi” are actually “Anhur” and “Bes” respectively. 

3) On the third map, Anuket is shown including the second heavily shadowed scarp along its border with Atum and even incorporates the third (non shadowed) scarp. These two scarps are incorporated into Sobek in the fifth map. In that map, the border kisses only the first scarp which is the scarp where Vincent et al’s jets 27 and 31 were located. This constitutes a 4-500 north-south drift in the Anuket border. 

4) Staying with the same Anuket border in 3, above, its border with Neith is similarly southerly in the third map (it starts half way along the Maftet border along the head rim. But in the fifth map, the Neith border has crept north by several hundred metres almost to the beginning of Maftet where that dog-leg in the head rim is. In fact, the parallax of the head rim rebate diminishes the effect. It’s as good as in line with the dog leg. 

There are several smaller contradictions:

1) In the top-right montage, the border of Seth with Babi has been truncated with respect to the older ESA regional map. On that map; Seth used to have a finger protruding 400 metres along the Hapi rim and into the Babi area. This isn’t a contradiction between these recent maps and so might have been an intentional change in the border line.

2) The Seth border is also shown as being too far down Ash in the bottom-left montage frame (difficult to judge with the acute perspective but nevertheless a few hundred metres off because of that acute angle. 

3) In the bottom-left montage frame there’s a green blob on the Serqet rim. This is the smooth area between the pillars of the C. Alexander Gate. It was part of Serqet in the old regional map but this green blob implies that it’s thought to be a part of Anuket peeping round from under the head rim. That’s not possible from this angle- the green blob is definitely the smooth area between the pillars of the C. Alexander Gate. Although this is certainly a contradiction with the old ESA maps, it also appears to contradict the top-left and bottom-right montage frames too, although this area is very nearly over the horizon in both frames. 

4) On the third map, a finger of green depicting Apis extending up into Ash but not there in the bottom-right montage map. This finger is not in any previous map either. 

There are several other anomalies to do with Geb/Bes, Wosret/Maftet and (possibly) Seqet/Ma’at but they are too involved to explain here. 

Hopefully, these contradictions could be corrected, especially the first four. They could lead to future papers inadvertently causing confusion by citing these areas that others interpret as different areas. This would be due to referring to a different map in this presentation (or referring to the old ESA regional map instead).


As Marco has dealt with point 3 in the first list, I’ll deal with the other three of the four as they’re the most important. The second list may have to wait a while but it’s less important. We start with point 1 and then the rest of the first list, i.e. points 2 and 4 will be presented soon in updates. So point 1 will go out right away, on its own, with the updates following over a few days. 


From my comment:

“1) While Bes is clearly marked in the first map it disappears in the second one, taken over by Geb and Anhur. It’s not perspective, you can see the same features coloured differently.”

Here are the two maps. We’ll call the upper one ‘A’ and the lower one ‘B’. They’re not labelled as such in order to keep them close together for toggling. I’ve annotated them with fiduciary points that match morphological features from one to the other. The key follows.


Key (A is upper photo; B is lower photo)

Yellow- 3 dots that are at the ends of three distinctive ridges. The top-left one is on the north-western border of Geb in A but well within Geb in B. The other two are on the Geb-Bes border in A but well within Geb in B where there is no representation of Bes. Geb has therefore taken over Bes in B. However there’s some of Bes that remains in shadow along the bottom in B (i.e. below the brown dots which are described further below).

Orange- a ridge that’s entirely within Bes in A but entirely within Geb in B. 

Light blue- two dots. One is a boulder or protrusion at upper-left in both photos and the other is a boulder or ice patch at lower-right in both photos. The upper left one is within Bes in A and on the Geb Anhur border in B. The lower left one is well within Bes in A but well within Anhur in B. 
Mauve -this sharp turn is correctly positioned at the exact top-right point of the scarp as identified in the related paper but this point is on the Bes-Anhur border in A and on Geb-Anhur in B.

Red- a distinctive feature that’s well within Bes in A but well within Anhur in B. 

Brown- this is top edge of the strange protrusion at the south pole. It sits well within Bes in A but runs along the bottom border of Anhur in B.


From my comment

2) “In the montage [header for this part], right hand column, middle image, “Seth” and “Babi” are actually “Anhur” and “Bes” respectively.”

This is clearly a simple mislabelling but is put here for completeness in covering the first list of four points. I’m sure the paper authors are completely familiar with the Seth and Babi positions. For other readers, Babi and Seth are on the right in the montage image referred to in point 2. Babi is light brown and sitting above the ‘A’ of ‘Aten’. Seth is out of sight beyond it except for a tiny blob of red peeping over the horizon in the distance. That blob seems to show, correctly, the southwestern rim of the Aswan crater at Seth. You can see both regions clearly in the top-left image of the montage. In that image, Anhur and Bes are hidden on the other side of the comet. 

POINT 3 (see update below that complements Marco’s post on this).

3) [covered in Marco Parigi’s post, linked again below but included here for completeness] “On the third map, Anuket is shown including the second heavily shadowed scarp along its border with Atum and even incorporates the third (non shadowed) scarp. These two scarps are incorporated into Sobek in the fifth map. In that map, the border kisses only the first scarp which is the scarp where Vincent et al’s jets 27 and 31 were located. This constitutes a 4-500 north-south drift in the Anuket border.”

POINT 3 UPDATE (8th October 2016).

After completing point four below, it’s now possible to pin down Marco’s Sobek border shunt to a good degree of accuracy. 

Here are a couple of annotated photos, reused from point 4 but with the extended Sobek border shown. The first photo (named update 1) shows the official map with Sobek in light brown and Neith, the subject of point 4, in blue. Both regions extend a long way northward or leftward towards the green of Anuket when compared with the second photo (named update 2). 

The second photo shows a yellow-dotted line tracing what is the line of the more northerly border for Sobek in update 1. The blue line does the same for Neith but is presented in more detail in point 4. 

The two photos are placed together for toggling up and down and have a joint key below. 



Thick beige or light brown line- (above the ‘k’ of ‘Sobek’). This is an arbitrary line past which Sobek extends down into shadow i.e. across where it says “Sobek”. The beige line is covering a yellow line that is on the official map and looks a bit like an actual border of a very small Sobek. That’s a little confusing, especially when drawing in an extension to Sobek. So it’s been drawn over in the same light brown colour as for the Sobek region so as to depict the lower (western) edge of this small illuminated part of Sobek. You can see parts of the old yellow line peeping through but there shouldn’t be any line here that looks like a border. It’s just a place where the top (east) of Sobek gives way to the shadowed bottom of Sobek.  

Medium yellow- (in update 2) this is the line of the Sobek border as it’s shown in update 1. It starts on the right and follows the actual update 2 line for Sobek. That portion of the border is mostly in shadow in update 1 but assumed to follow the same line as update 2 for this short stretch. The most important part of the yellow-dotted line is the section running along the blue line (actually contiguous with it) and then dropping vertically down the neck, past the two big yellow dots and into shadow. The green area enclosed under this line is officially designated as being part of Anuket (green) in update 2 but is deemed as being Sobek (light brown) in update 1. The section in shadow is an informed guess based on the track of the lower (western) Sobek border in update 1. Of course, it doesn’t follow any carefully identified fiduciary points through the shadow as the rest of the yellow and blue lines do so it’s just for general guidance. 

Blue- (in update 2) this is the dotted line that does exactly the same thing for the Neith border as the yellow line does for the Sobek border. It’s presented in detail in point 4, along with a number of extra photos. 

Large yellow- these are the Vincent et al. 2016 outburst locations numbered 27 (upper) and 31 (lower) in that paper. They’re described in detail in part 4 and are used here as fiduciary points for the yellow line to skim past. It skims south of 27 and north of 31. Since the dots themselves are placed on the same fiduciary features in both update 1 and update 2, it’s not just randomly placed dots but actual features that are constraining the yellow line. 31 is admittedly in shadow in update 2 but it’s very close to what for me and Marco are very familiar features. This means the error in its placing on the actual feature in the shadow is no more than half a dot width. The feature is the third point that collapsed en masse without crumbling. It did so over perihelion (probably a little before the actual perihelion date) and is now less visible through not casting a shadow as an overhang. It’s described in point 4. 

Other colours- please see point 4 photos and keys. These features act as fiduciary points to constrain the Neith border shunt. They were carefully identified so the Sobek border shunt uses these as well to get an accurate placing for its yellow dotted line. 

The extent of the northerly shunt of the Sobek border (including the northerly Neith shunt that it wraps around as a finger) is at least 300 metres.


This is added before point 4 because it affects it. It’s a significant movement of the Maftet/Wosret border of 100-150 metres and so it’s really an anomaly in its own right. It’s residing in this list as the fifth point but placed before point 4 due to being related to it. This was the Wosret/Maftet anomaly that was mentioned in the main comment but which was deemed too complicated to explain there. It in fact confused me for articulating point 4 so that explanation isn’t fully accurate. Point 4 still stands but one aspect of it is adjusted as a result of understanding this point, 4A. It’s interesting that despite knowing about 4A, it still tripped me up so it’s certainly going to confuse scientists using or citing this map. 

Photo 4- a close up of the Maftet-Wosret border which is at odds with most renditions of this border, including its position in the photos shown below.

Photo 5- The Maftet-Wosret border from a distance 

Photo 6- A close up of photo 5 with annotations. 

Photo 6 is annotated with the portion of the Maftet-Wosret border that’s visible in photo 4. It’s marked in medium pale orange. This is the true border that isn’t actually marked in photo 4 but its line is nevertheless visible in the frame. The anomalous border in photo 4 is marked in small pale orange dots in a fairly straight line that’s notionally parallel to the true border. It’s some 100 metres or more beyond the true border. 

I’ve also used small pale orange dots for a short curve below the official line. I trace this curve in the lower photos as being the right hand side of that ‘m’ shape plus the curve after it. It would look strange not to annotate the photos below in this manner because this lower curve really does look like part of Maftet in those photos. So this short curve isn’t part of the 4A anomaly which is the longer, straighter section of pale orange above and beyond the official border. 

Photo 7- this is photo 4 again but with the correct Maftet-Wosret border as shown in photo 6.

Photo 8- this is the original used for colouring regions onto photo 7. 

The originals are clearer than the coloured versions so they come in useful for tracing the border lines more accurately. Key follows.

Medium pale orange- the true Maftet border as far as it can be extended (up and right from our viewpoint) along Wosret before going off-frame; and straight up towards the top of the frame, which is along the very defined edge of the head rim before again going off-frame. 

Bright green- two fractures or gouges that are often annotated in this blog as fiduciary points in this area (see Parts 17 and 19). They will also be shown in photos below to ensure we’re anchored in the right place with our various lines. 

Photo 9- same as photo 8 but with the anomalous Maftet-Wosret border in small pale orange. 

Photo 10- this is the Part 17 header, used for extra context from another angle. 
Copyright ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0

Red- on the head lobe, this is the head rim. So Maftet and Wosret sit above it and are divided by the medium pale orange line. The red line on the body is relevant to Part 17 and not this post. 

Small orange and bright green- as for other photos above. 

Photo 11- this is one of the photos that will be used in point 4, along with photo 4 as the comparison photo. It shows the same annotations as for the photos above. 

Photo 11 depicts what is in effect a shunting of the Maftet-Wosret border northwards along the head rim. This reduces the degree of the point 4 anomaly, at least with regard to where the Neith border meets the head rim. This shunt is what tripped me up in point 4, below. The Neith border is still nevertheless shunted quite far north when looking further down the neck. 

Photo 12- this is the original for photo 11. Again, it’s clearer. You can see the left hand (northern) Maftet border being traced more accurately along the head rim and the right hand bright green dot has found its home. 


This is a minor correction to the head rim line in photo 4 using its original, photo 8. This caused some extra confusion as explained further down. The two photos are reproduced here together as photos 13 and 14 with new annotations. 


In photo 13, the usual pale orange line shows the Maftet border, including its run along the head rim as in photo 8. In that photo, my pale orange head rim line doesn’t dip in sharply like its coloured-in twin does. Similarly, in photo 13 above, it runs along the true rim and completely ignores the green triangle that’s trying to be part of Anuket. That triangle is dotted yellow and is thereby reclaimed as part of Maftet. That’s because it’s sitting solidly above the sharp head rim line. You can see the head rim line (just) in the much clearer original, photo 14. In that photo, just the apex of the yellow triangle is depicted for orientation. You can imagine the yellow triangle extending to the head rim line. It’s clearly extending across the head with the ‘green’ Anuket neck separated, well away, under the head rim overhang. 

This caused more confusion for explaining point 4 because the only sharp dip in along this portion of the head rim is at the northwest corner of Maftet (just out of frame at the top). That’s the dog-leg referred to in point 4 and most obvious in photo 10. 

Although this anomalous dip looked too small to be that dog-leg, it blends into a ridge that looks uncannily like the frilly ridge just beyond it and off frame (that’s because it’s a delamination from that frilly line and so it’s a translational match). Beyond the off-frame frilly line, there’s a section of Anuket that does indeed seem to encroach on the head rim. So I assumed that this visible section beyond the frilly line was the encroaching part that’s off-frame and that Maftet was now deemed to include this patch. 

If I’d seen the original photo at that time, I would’ve noticed the ‘bright green’ gouge that’s covered by the ‘Wosret’ label and seen it was too close to my supposed dog-leg. But I was hooked by the anomalous sharp turn and knew that it ‘had to be’ the dog-leg because I know there are simply no other dips this sharp in the head rim until the one 800 metres to the north at mid-Serqet (which is the orange match in this blog).


Point 4 had a slight error to do with the dog-leg as explained above in Point 3 but is quoted here from my Rosetta blog comment in full:

“4) Staying with the same Anuket border in 3, above, its border with Neith is similarly southerly in the third map (it starts half way along the Maftet border along the head rim. But in the fifth map, the Neith border has crept north by several hundred metres almost to the beginning of Maftet where that dog-leg in the head rim is. In fact, the parallax of the head rim rebate diminishes the effect. It’s as good as in line with the dog leg.”

Photos 15 and 16- photo 15 corresponds to the third map referred to in the point 4 excerpt above and photo 16 corresponds to the fifth map referred to. The only thing you have to note here is that the northern (left hand) border of Neith shunts itself north in the photo 16 map. It’s therefore dotted blue in photo 15 to show where it ‘should’ be in that map. Neith itself is coloured blue in both photos.


Photos 15 and 16 are reproduced further below with a full key for the other dotted features.

The northern border of Neith is officially marked in both maps. However they are in different places, over 100 metres apart. In photo 15 there’s a blue-dotted line that corresponds to the official line as it’s depicted in photo 16. You can see that the dotted blue line is a long way north of the official photo 15 line next to it. That official line should overlay the blue-dotted line perfectly because the maps should show the regional borders in the same place, following the same features. 

Just for clarity, although Neith remains blue in both maps for photos 15 and 16 there’s a difference in the colours used on the official maps for defining the shunted border in question. The official photo 15 border line is in fact a green line (not blue) that’s bordering the blue of Neith. In photo 16, the north-shunted version of the line is blue (not green) and again, it borders the blue of Neith. That’s why my dotted line is blue. It was chosen for annotating photo 15 so as to show where the blue line of photo 16 should be if it were similarly placed. However, that official blue line in photo 16 (now changed to green in photo 15) is a hundred metres away and notionally parallel to my blue-dotted line. 

References below to “up, down, left and right” are with reference to ‘upright duck’ mode with the head lobe at the top and the body lobe at the bottom. The neck then runs ‘vertically’ between the two lobes. Wosret and Maftet are on the head, Geb is on the body. Anuket is the neck. 

Once the dog-leg issue is resolved, the point at which the Neith line reaches the head rim is roughly the same in both maps. However, the border line in the photo 16 map starts substantially further north at the bottom end of Neith, halfway down the the neck. The anomaly shadows the Sobek northward anomaly outlined by Marco (see his blog post linked in point 3, above). This is because the Neith border going up the neck stems notionally from the northern Sobek border in both versions of the border in the two maps. The fact that the Neith border is essentially an extension of the Sobek border means that the green Anuket area beyond them is a straight vertical sweep up the neck from bottom to top. 

The northward Neith border in photo 16 then bulges out as it progresses up the neck. That’s after we start at at the bottom of Neith, along the top of Sobek, and turn the corner to go up the neck. This bulge is also seen in the more southerly counterpart in photo 15. The bulge makes the two lines roughly parallel for the whole extent of their run up the neck and 100-150 metres apart. 

So the anomaly between the two officially drawn borders in the two maps is a 100-150 metre north-south anomaly. The line in photo 16 is truncated near the top by being obscured by the substantial head rim rebate at this point. We’re looking down at the head and neck somewhat so the rim rebate obscures the top 100 metres or so. If you extrapolate the dotted blue line in photo 15 to the rim, it would touch it just a little way north of the more southerly component. The southerly component does a little left turn at the top before reaching the rim which helps to bring the two lines almost together at the rim. 

Here are some more photos to nail down the exact path of the more northerly version of the two lines. I may do the more southerly one in due course but the northerly one is easier to trace. That’s because it was overlaid on a more detailed, hi res photo in the first place. 

Photo 17- this is photo 15 reproduced. It also includes the version without the blue-dotted line for toggling and comparing (but see photo 18 for a more detailed blue line). Originals or very similar versions for toggling aren’t designated separate photo numbers. 


Photo 18- the exact line in smaller blue dots. It’s laid on the original version of the photo used for the map. Includes non-dotted version for toggling and a more zoomed-out version for context, showing more head rim etc.


Photo 19- this is the same as photo 16 and its key for all the other colours follows. 


(this key includes fiduciary points and lines that are shown in later photos from a different viewpoint so as to show we’re tracing the correct blue-dotted line in photo 15). 

Pale orange- the Maftet border (medium pale orange) and the anomalous Maftet border (small pale orange). This was outlined in points 4A and 4B. 

Light blue- two fiduciary boulders (both near the red line).

Red- a distinctive ridge. 

Green- various curved and winding ridges that enclose dips. The green curve that straddles the border of Neith is an important fiduciary point for getting the line right in later photos. The two ends of the curve extend all the way down to the second and third ridges that are arrowed by the OSIRIS team in the original below. This means these two ridges are joined in a curve at their top ends. Only the curved end is relevant here. There’s also an isolated green line next to the yellow dots and not very visible so it’s dotted with bigger dots. This corresponds to the first or left hand ridge, arrowed in the original. This is a later photo than photo 15 and its annotated relatives below. The obvious protruding, overhanging points on the earlier photos have collapsed in this version (see ‘yellow’ below). 

Yellow- these two dots denote the Vincent et al. 2016 outburst sources numbered 27 (upper yellow) and 31 (lower yellow). 27 sits on the site of the most easterly collapsed point that was identified by Marco Parigi (see link below). 31 sits at the tip of the collapsed point that I identified. Both points used to be pointed overhangs. These collapse identifications were made before the Vincent et al. 2016 paper was published. Marco’s point collapsed fully into a pile of rubble. My point collapsed en masse, retains its original form and is almost completely camouflaged, being flat on the terrain it once overhung. These two collapses makes the ridge appear both shorter and ‘blunter’ in the sense of losing its three points. Marco identified the middle point’s collapse too but that’s a smaller collapse and Vincent et al. don’t show an outburst location for that. These three collapses makes comparison of this ridge in pre- and post-perihelion photos very challenging. The reason for explaining this here is that Marco and I are intimately aware of the morphology changes in this area. It follows that any apparent misidentification of this ridge in the lower photos isn’t in fact a misidentification. It’s the same ridge. There are pointed overhangs for this ridge in those other photos below and no pointed overhangs in this photo. Once we’re aware of this, it furnishes us with a very useful fiduciary point to check on and constrain the other two green lines. 

Photo 20- this is the original used for the photo 19 map. Both are zoomed-in versions of the true original further down, at the bottom of point 4.

All colours are the same as for photo 19. 

Unannotated versions for photos 19 and 20 (not numbered). 


Photos 21 and 22- photo 21 is the same as photo 18 but with the same annotations as photo 19. Photo 22 is photo 19 reproduced for toggling between the two so as to check all the fiduciary points match. 


Photos 23 and 24- this is the same set-up as for 21/22, above but 22 is replaced by its original with the even clearer annotations for toggling and checking the fiduciary points match. 


Photos 25 to 28- the zoomed-out originals used for point 4.



This post completes the presentation of the more important anomalies in the recent OSIRIS map updates. These were presented as points 1 to 4 in my pasted Rosetta blog comment near the top of this post. They were dealt with in detail and in sequence in this post but with two extra points numbered 4A and 4B sitting between 3 and 4. This was because 4A and 4B were pertinent to 4 and needed presenting before point 4 for that reason. However, they were anomalies in their own right. They therefore augment the list of the more important anomalies from four to six. 

The second list (of four less important anomalies) was also in the Rosetta blog comment. This list will get the same treatment in due course but will be left for now owing to a large backlog of stretch theory posts. It will be in a separate, twinned post because this one is now rather long. It will be linked below this conclusion. So if there’s no link, it hasn’t been done yet. 

If, however, anyone including OSIRIS scientists, independent scientists or citizen scientists would like the other list dealt with earlier, please leave a comment below or tweet/DM me: @scute1133. I shall then do it sooner rather than later. 


Marco Parigi’s blog post on the collapsing points along the first ridge:


 Link to my comment:

Link to Marco Parigi’s comment which was below mine in the same comment thread:

Part 64- The Evidence for Long-Axis Stretch Along the Head Lobe Shear Line at Hapi/Babi



The header is the same photo as photo 2 in Part 62 and this post is a follow-on from Parts 62 and 63 which dealt with the delamination of the three Ma’at pits when the head lobe was still clamped to the body lobe. This could therefore be considered the third part in an ongoing mini-series whose underlying theme is always the delamination of the Ma’at pits. More posts will follow, totalling at least five for the mini-series. They build up evidence for the collateral mechanism for the delamination of the holes: long-axis stretch along the head lobe shear line and the consequent sympathetic delamination of layers along the shear line. 


It should be noted that although the term Hapi/Babi is used for the Ma’at 01, 02 and 03 pit delamination line, some of that line was along the short finger-protrusion of Seth. Since it’s confusing to say Hapi/Babi/Seth, this short finger is counted as being part of Babi in this part and in Parts 62 and 63. This expedient is all the more compelling when we consider that the Seth finger is very much a part of the Babi morphological evolution by virtue of finding itself on the Babi side of the mirror-symmetry line described in Part 63. 

The Seth finger was included as part of Seth in the ESA regional map only because of visual similarities on its surface with no known reason for that similarity. The reason is that the head lobe originally sat on the finger, and this caused its characteristic rectangular shape by virtue of the furious outgassing escaping from underneath it (Parts 1, 5, 7, 8, 20). And apart from a tiny extra area next to the finger, described further below, the head lobe sat nowhere else on the official Babi area.

Photo 2-The ESA regional map. The finger is the pink protrusion coming towards us along the Hapi rim (next to the Hapi label). This is counted in this blog as being very much a part of the Babi morphology. 



The originals aren’t part of the numbering scheme. Dots should act as a guide only before verifying the dotted features on the originals. This is because the dots can sometimes partially obscure the detail they’re trying to show. 

Some keys are narrative keys and so some key colours get divided into paragraphs. These longer keys end with ‘/////’.

Photo 3- the header reproduced. 



Green- these two lines are the same shape, a translational match. They show a massif on the right that has slid from its seating on the left. The seating and massif correspond to the second and third bumps respectively. Those are the bumps or mini-volcanos in Part 62, that both delaminated along the shear line. They delaminated along with their respective layers, both delaminating ultimately from the first bump. The left hand green line here is sitting on top of the second bump or mini-volcano. The right hand green line is kissing the bottom rim of the third bump. In this view, it’s the bottom end of the line that’s kissing the upper perimeter of the bump rim. In this sense, the slid massif is only notionally related to the third bump. It will be shown in future parts that it probably was sitting on the third bump originally, but the main point here is to show that it’s a slid massif and that it slid in a direction that’s along the long axis of the comet. Furthermore, it will be shown that it ground along the head lobe rim as it slid i.e. it sheared from the head rim due to the shear gradient set up across the long-axis tensile forces. This shear gradient was described in Part 62.

Red- these aren’t exactly slide tracks but are a proxy for the slide track of the massif. The bottom red line is in line with the section of head lobe that’s 1000 metres directly above it in reality but slightly below it and to the right in this view due to Rosetta’s slight parallax. In other words, Rosetta wasn’t quite vertical over the head rim and shear line when this photo was taken. So the lower red line in question is in line with that straight section of head rim just before it curves round and downwards somewhat towards the right. This match will be depicted in lower photos, as it was in Part 62. 

The reason the bottom red line is a proxy for a slide track is that a) it’s in line with the bottom ends of both the matching green lines and b) it’s actually the ragged line that the matching section of head rim tore away from. So the massif that slid to the right was kissing the future head lobe rim along this line when the head was still seated on the body. It then tore away from the future head rim under the influence of long-axis stretch along the shear line. And of course, the shear line is, by definition, where the head rim originally sat. So the massif slid along, or ground its way along the head rim before the head fully detached ‘upwards’ as viewed in upright duck mode. Upwards is towards us in this overhead view. 

This means the massif didn’t exactly tear ‘away’ as mentioned above but tore ‘along’, i.e. it sheared under the shear force, described in Part 62, which was focussed along the shear line. The massif would have slid only during head herniation and before head detachment because all the tensile forces of stretch would have been transferred almost instantaneously to the incipient neck once the head sheared. That left little or no residual tensile force in the Babi layers to continue fuelling the long-axis stretch and further widen the delaminations. And there was certainly no residual tensile force at all once angular momentum had been conserved at today’s head lobe displacement (T= circa 5.8 hrs, see ‘Spin-up Calcs’ page).

The upper red line is certainly in line with the slid massif and its seating and may be a simple slide track. However, it may have delaminated from the other red line on head shear i.e. at 90° to the long-axis stretch (see Part 62 for this orthogonal delamination scenario). So it would have delaminated from the lower red line towards the top of the frame, which is along the same vector as the Babi slide (Part 40). 


Photo 4- same as photo 3 with additions.


Yellow- on the right, it depicts the back of the slid massif. On the left it depicts the bottom of the second bump. Together with their respective green lines, the yellow lines enclose similar shapes. 

Mauve- a mini match within the slid massif and within its seating. 

Photo 5- the delaminated layers. All annotations except green have been temporarily removed, just for this photo. 


Green- these are just three of the five delaminated layers that are depicted in the header of Part 38. The green line along the front of of the slid massif is the same as in photo 3. So it turns out that the front edge of the massif constitutes the front edge of the third delaminated layer. Its green line is extended round in a dog-leg although the real layer may be under that other terraced or bunched-up detritus beyond it (above it in the frame). That’s the material from the Babi slide (Part 40). 

The seating of the slid massif isn’t shown here because we know where it is. It’s right next to the very obvious line running along the front of the second bump which is at the front edge of the second delaminated layer. The seating is so close to the second layer front edge that we can say that the massif, i.e. the third layer, delaminated from the second layer along the direction of the long axis. This is in keeping with what was stated in Part 62. The second layer’s front edge does a fairly sharp turn and continues down Babi (upwards in the frame) to join another green line on the left. 

That left hand line leads back up to the Hapi rim at which point it describes two waves. Those are actually up/down waves and from the front (viewing from the left and very low down), they look like gull wings. So these are the gull wings which Part 62 describes as the first bump or mini-volcano from which the second and third bumps delaminated. But the bumps are just the ends of their respective layers, perched on the Hapi rim. So if the bumps delaminated, it means the entire layers delaminated too: second from first and third from second. 

The fourth layer is off-frame to the right and technically beyond the scope of this post. However, in Part 63, it’s implied as the southern perimeter of the rift at this end of the long-axis delaminations along Hapi. So it tore rather than delaminating, just as happened at the Aswan/Seth end of the long axis delaminations.


We can see from photo 5 that the bumps delaminated along the long axis in a straight line and yet their respective layers seem to fan out from one point. The first line even has a three-pronged trident shape pointing at the gull wing waves that is replicated at the second bump. The trident staffs are the delaminated layers marked green but the two sets of three prongs are a definitive match (see them annotated much more clearly in Part 38, photo 6). The replication on the second layer is at the place you’d expect, which is the with the prongs pointing at the base of the second bump. And they’re even slightly flipped round by the differential forces within the fan because the prongs are that much closer to the true shear line with the much steeper shear gradient. 

The trident match is less obvious here but is annotated and described in detail in Part 38, photo 6. Here’s a link to Part 38:

The trident movement, arcing over on its staff from the axis at bottom of the fan is strong evidence for the fanned delaminations. It seems highly unlikely that a trident shape would be reproduced at both bumps. 

Photo 6 below is the Part 38 header (not Part 38’s photo 6 with the trident annotated but you can see it here in its header, unannotated). It shows the full fan effect. The delaminated fronts making up the fan in this photo are in light blue because ridges were being denoted in light blue back then):
Photo 6- Part 38 header, reproduced. 

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

Light blue- these are the ridges that betray the the fanned-out delaminations. 

Dark green dot pairs- the three delaminated bumps (lower-left) and two more that were presented in Part 38 (upper-right). 

Photo 7, below, is reproduced from Part 62. It shows the Ma’at pits and their sources on the body as well as many matches from head to body. However, the main point in reproducing it here is to show that the shear line on the body (terracotta) runs contiguously with the lower of the two red slide lines. We’ve already established that the slid massif was originally attached to this line. Therefore, it must have also been attached to the portion of head rim that was seated there. That portion of head rim is that straight section of rim just before it curves round and downwards somewhat towards the right. It’s offset just to the right and downwards from the bottom red line in this view due to parallax (as mentioned above in the ‘red’ description in photo 3). If in any doubt, you can use the additional version with two lines of arrows showing the match or the beige dots that align in one line when the head is reseated. 

Photo 7- the head rim was seated along the lower red seating line for the slid massif. Therefore the slid massif was originally attached to the head rim before shearing along it and sliding to its current position. 


Dark green- as for photo 3 but there’s an extra fan rib or delaminated layer on the left. It’s one of the two extra ones shown in the Part 38 header above. The other one is off-frame to the left. This extra one that’s shown also extends to the common ‘stalk’ of the four fanned lines while, if you recall, the fifth one is detached from the fan in the form of our slid massif. The first line, the off-frame one, branches to the left from a point that’s a fraction off-frame in the top corner. 

Notice that the faint dark green dots on the head rim match to the wavy line on the first green line on the body. This is the gull-wing match, matched in 3D (from the side) in minute detail in Part 5. Except, this is a slight cheat because what we’re seeing on the head rim is a second, curved-up gull wing delamination that sits directly above the true set it delaminated from. It bent upwards during head lobe herniation and now exhibits a gentle curve. It constitutes the upper component of the first zig zag in Part 62. So there are actually two sets of gull wings on the head that sandwiched down onto one set on the body. This was described in detail in Parts 5 and 7. 

Although the sideways, 3D match of the gull wings isn’t strictly related to this post, they were sitting right on top of the Ma’at 01/01A source (Part 62) and that’s why they were raised into two gull wings. You can even see the two slurry piles that oozed neatly out in front of each wing on the body’s set of wings. They’re semicircular piles, each one the same width as its wing. They exhibit more concentric semicircles within their semicircular perimeter. Any engineer would identify that as classic signature of slump. 

So, since the 3D gull wing match is somewhat related to this post and intimately related to Ma’at 01, various links to the gull wing 3D match have been put at the end of this post along with a photo of the match with exquisitely detailed fiduciary-point matches. Please see the appendix. 

Red- still with photo 7, this is the massif’s seating against the head lobe which doubles as its implied slide track. The slide track betrays the long-axis shear force vector causing the layers to delaminate along the long axis. This led to the pits delaminating along with their gas sources being emitted from each of the layer fronts. The gas sources uplifted the layer fronts into bumps or mini-volcanos before exiting through the actual pits. One of those uplifting zones was the two gull wings constituting the first mini-volcano. On head shear, the pits recoiled up the head and down the body (Part 62; also Parts 40 and 41). Their gas sources were left marooned at the shear line and are clearly visible today directly below Ma’at 01, 02 and 03 (Part 62). 

Orange- this is the now-bunched-up layer that slid away from the shear line after head lobe shear. This layer used to sit over the bumps and the fan shape. It fully delaminated as opposed to undergoing a fan delamination. That’s how it was loosened enough to slide. The very fact it sat over the fanned layer meant that it was overburden that increased the tensile strain resistance of the fanned layer. That’s why the fanned layer resisted full delamination and didn’t end up joining its overburden layer behind the orange line. 

Other colours- these are as for this same photo in Part 62 (photo 17).


Photos 8 and 9- these are a comparison of the header and a close-up of the Part 38 header. They exhibit exactly the same annotations and so are presented together with the same key. The focus is on the intricacies of the attachment of the massif to the red line. 


Photo 9: Copyright ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0

Red- the attachment and implied slide line.

Dark blue- the outside curve of the massif (not annotated in other photos further above) and its seating on the ‘outside’ of the red line. In reality, it’s simply at the bottom of a long, thin sloping area that constitutes a widened version of the red line. The green line runs along the top of that widened line. 

Green- the front of the slid massif and its seating on the second bump (as for photo 1).

Light blue- fiduciary positions of certain larger boulders.


Photo 10- the 3D gull wing match. Fullest zoom required for all detail.
Copyright ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0

The photo above is the finer-detailed match from Marco Parigi’s summary post of my Part 5. 

My Part 5 is the first link below, which shows how I traced the sideways view of the gull wings on head and body and checked meticulously that they were indeed the same features as those we see for the green gull wing match in photo 7 above in this post.

Marco’s summary post is the second link for which I did the more detailed version of the match shown above. A quote from his post sums up why this match alone, made in December 2014, proves almost beyond doubt that the head lobe sheared from the body. This, despite it being around 1% of the evidence to date for head lobe shear.

“These mini matches make the original match conclusive, because if the large scale match was a coincidence based on large features, there is extremely low expectation of the constrained smaller section also matching in the small scale and/or in the third dimension.”

Quote from above link by Marco Parigi, October 24th 2015.



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. 


Part 63-The Mirror-Image Pit Delaminations Along Hapi/Seth


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

This is the second part in a mini-series regarding the long-axis delamination of pits along the shear line. Stretch theory is not a mainstream theory and is not entertained by the Rosetta mission scientists. Therefore, the pit delaminations described below are at odds with the current thinking on pit morphological evolution. If you are interested only in peer-reviewed literature this will save you further reading. 

The two header photos are reproduced further below with their keys and originals. ESA’s regional map is at the bottom.

Everything applying to the long-axis stretch and delamination of pits along the Hapi/Babi rim (Part 62) also applies to pit delamination along the Hapi/Seth rim at Aswan. These two sections of the head lobe shear line underwent the same process in mirror fashion. It will be some time before this Aswan section is written up in full hence this overview to show how the Ma’at 01, 02 and 03 pit delaminations at Hapi/Babi were consistent with the overall picture across this north pole side of the comet. 

Three pits delaminated along the Hapi/Seth rim in a mirror image to the delamination of Ma’at 01, 02 and 03 along the Hapi/Babi rim. That’s why there were two half-pits identified along the Aswan rim at Hapi/Seth in the original sinkhole paper by Vincent et al. (July 2015).

The large pit in the Seth region is the mirror image of the Ma’at 03 source described in Part 62. That source is the third mini-volcano and it was itself once an outgassing pit on the body. It has the largest source diameter of the three Ma’at pit sources. It’s that majestic, undercut curve overhanging Hapi. That diameter is about the same as the main Seth pit. And the Seth pit is the largest of the three pits that delaminated in the mirror image of Ma’at 01, 02, 03. The other two are the half-pits on the Aswan rim. 

The Seth pit and the Ma’at 03 source pit lie at either end of the very straight long-axis stretch line along Hapi that will be described in the next part. Both pits are larger than any of the others because they share a unique morphological evolution as directed by the long-axis tensile forces of stretch. 

For both pits, this evolution proceeded in an identical and mirrored fashion. Both pits are sited where the long-axis stretch vector turned to make its way in a straight line along the side of the herniating head lobe. The stretch vector did this in an attempt to make its way around the herniating head lobe. This ‘line of least resistance’ behaviour is patently clear, stamped on the comet’s surface in the form of the red triangle (Part 26, signature 6) and is the raison d’être for Serqet and Nut. So both pits are sited at the exact point where the long-axis tensile force vector joined the herniating head lobe rim. 

Both pits opened up as a result of 200-metre-wide rifts careening up the long axis tensile force vector for 1.6 kilometres and into the future head lobe rim, passing underneath it before the rim sheared from the body. For the Seth rift see Parts 48 and 49. The two points at which the two rifts went under the head lobe opened up into a 200-metre-wide notional square with the Hapi rim acting as a third side and seated Hathor matrix as the fourth side. The head lobe acted as a loose lid. 

So the Seth pit and Ma’at 03 pit are sited where there was both head lobe shear and a 200-metre rift opening up. Only the Hapi rim stayed (relatively) still while the three-way movement of two rift perimeters and head rim induced furious outgassing. This is why the Seth pit is by far the largest identified pit on the comet and the same 200 metres width as the rift it sat inside. The Ma’at 03 pit is the same diameter and is effectively an unidentified, dormant half-pit like the two along the Aswan rim.

Both pits, Seth and Ma’at 03 source pit, then delaminated away from the shear line after head lobe shear. The Ma’at 03 pit suffered a single delamination across Babi (Part 40) while its original Ma’at 03 source pit stayed put at the shear line as you might expect a hole to do. 

However, the Aswan pit actually slid on the onion layer below it. To repeat, this hole did actually slide: a sliding hole within its sliding matrix which was the entire Aswan basin onion layer. This is why the Seth pit is flat at its base. The base is the next onion layer down, looking up at us and it’s the layer the Seth pit slid on. It’s also why the base of the pit is at exactly the same level as the lower lip in Hapi at the base of the Aswan cliff. That’s the layer over which Aswan along with its pit slid. This sliding scenario is absolutely dictated by the conclusions drawn from photo 7 in Part 49. There’s a line of boulders on this lower lip that is a translational match to the base of the Aswan cliff line. This is strong corroborative evidence for the Aswan-plus-pit slide but photo 7 in Part 49 is even more compelling:


Larger Red- the 200-metre-wide rift

Bright green- the extension of the rift that went under the head rim before the head sheared. 

Light Blue- the Seth pit and the quasi square location where it first opened up before sliding (between the bright green lines). 

Small red- red line at right: the Aswan cliff line (including its line in the shadow from other photos); at middle: a red line showing a boulder deposition line that matches to the cliff base; at left: the original seating of the cliff base. The Aswan cliff slide is almost the same distance as the Seth pit slide but certainly apparently less. In reality it has to be the same because the cliff is joined to the pit. The difference will be explained in a future part because the full extent of the sliding behaviour in this small area is beyond the scope of this post. What we see annotated and explained in this photo is 90% of the full picture anyway. 

Slate blue- at top: a pair showing the upper half pit (or second half-pit if equating it to its mirror image Ma’at 02 source pit) along with a very obvious match to its seating next to the blue quasi-square; at bottom: the first half pit which equates to Ma’at 01’s source pit. 

 After the Seth pit slid, it delaminated into three across Aswan (Part 32).

There’s much more to say on the Aswan delaminations, along with copious evidence in the form of translational matches on the body and head and mirrored matches to the head rim underside. Some of this evidence has been noted in passing in Parts 37, the 48/49 twins and Part 50. The unavoidable conclusion is that all these pits that are today on Aswan, Seth, Babi and Ma’at were once delaminating their way in opposite directions along the Hapi shear line. 

Most tellingly of all, this mirrored, long-axis delamination behaviour along the Hapi rim is symmetrical about the north pole point in Hapi. Since the north pole is where the rotation axis pierces the comet’s surface, this is a significant signature that the mirrored delamination process was driven by spin-up of the comet:

Photo 2- the mirror-image delamination of the holes across Seth/Hapi. 

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

Brown- this is the paleo rotation axis adjusted up the comet’s x axis by 700 metres so as to make a comparison with today’s rotation axis (dark blue). The paleo rotation axis held sway before head shear. Only then did the axis precess 12°-15° to today’s axis location. The paleo axis is derived by taking the line that’s orthogonal to the inferred paleo rotation plane (see the Paleo Rotation Plane Adjustment page). 

Notice how the paleo axis runs down the middle of those two horseshoe craters and also directly between Aswan and Babi. It’s the line of symmetry between them. It’s also the line of symmetry between the two red rifts running from the original Seth pit seating and from the Ma’at 03 source. 

And most importantly of all, it represents an almost perfect line of symmetry between the two sets of three delaminating pits (light blue). You can see the three Ma’at pit sources delaminating one way (left) along the long axis from the brown line. And you can see the three Seth pits delaminating (and in their case, tearing and stretching) to the right along the same long-axis line at Seth. 

The paleo pole is the larger brown end dot. It’s equidistant from the original Seth pit seating and the Ma’at 03 source pit. The original Seth pit seating is where the symmetry of tensile forces originally held sway before the pit slid. The paleo pole is therefore the centre point origin along the symmetry line for all the symmetrical features described above. This is strong evidence that spin-up of the comet was the mechanism by which this mirror-symmetrical pattern arose and the consequent long-axis delamination at Hapi occurred.

Photo 3- the ESA regions




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. 



Part 62- The Morphological Evolution of Ma’at 01, 02 and 03


This post has been written with the hope that one or some of the end-of-mission planners might be able to read it. It concerns the morphological evolution of the pits that will be examined by Rosetta as she descends to her final resting place on 67P. That will happen at Ma’at on September 30th 2016 and the moment she lands will be the end of the two-year mission. 

The evolution of the pits as laid out below would perhaps point to the consideration of tweaks in the way the data is taken on the way in to land. 

The pit evolution scenario described below is according to stretch theory which isn’t accepted by the mainstream. It proposes that 67P was spun up via asymmetrical outgassing which itself isn’t controversial. The controversial part is the suggestion that this spin-up caused the comet to stretch into an ellipsoid, then allow the head lobe to herniate from the body and finally, shear away from the body, rising on a growing neck (the neck we see today).

The following is related as if it’s fact so as to spare the reader endless qualifications and conditional statements. It is, of course, a hypothesis. However, there’s an abundance of evidence supporting it.


The originals aren’t part of the photo numbering system. The main post follows the photos. It’s a summary of the photo explanations including further background explanation and links to other stretch blog posts. If you have any doubts or objections to what is presented in this ‘photos’ section, they should hopefully be resolved in the further explanation lower down. This is a very complex area so it would be a surprise if you got through the photos without any further questions. Hopefully the photos will be compelling enough to make you read on. 

The ESA regional maps are at the end of this photo section if you need to get orientated (photos 19 and 20).

Photo 1- the Ma’at pits 01, 02 and 03. 

Photo 2- An overhead view with the three pits (larger blue dots) and the exact outgassing sources for them, dotted blue and sited on the body below.


In photo 2, the majestic curve of the source for 03 is half hidden behind the head rim so it’s extrapolated with smaller dots. The evidence for outgassing from these three sources is documented in many blog parts, mainly Parts 5 and 7 and 8. There are also numerous head/body mirrored matches for this small area in Parts 1, 2, 5, 7, 8 and many more parts up to 52 and this Part. These matches completely constrain the body outgassing sources to match the 01, 02 and 03 hole positions (when the head was seated on the body). This is in addition to each source exhibiting signs of furious outgassing. 

Photo 3- same as photo 2 but with the pits numbered and a new one added, dubbed 01A for convenience, and just a little bit hidden here hence the arrow. The left hand outgassing source (as viewed here from above) fed 01A as well as 01. 01 then delaminated from 01A upwards towards us and to the right. Notice how 01 is facing its body source.

Photo 4- the bumps on the body (four dark green dots) that are a tell-tale sign of uplift from outgassing along the Hapi/Babi cliff rim. 
Copyright ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0/A.COOPER

These bumps are just next to the body outgassing sources depicted in photos 2 and 3. The bumps supplied the outgassing sources (see their relationship to the sources in photo 11). The bumps for the 01A/01, 02 and 03 sources are the three on the right. The fourth sitting on the left is beyond the scope of this post. Known as the ‘gull wings’ in this blog, each bump is neatly sited at the join of separated layers: four bumps betraying five layers. The separated layers are long-axis delaminations and the gases emerged from under each join hence their translational symmetry and equidistant separation. 

All the bumps are more pronounced in this photo. They are slightly less so in the following photos due to a less favourable angle. However, if they’re pronounced here then they really are very prominent even if apparently less so in other photos. 

Photo 5- a long shot of the head and body lobes showing the three relevant bumps and the Ma’at pits they supplied. The pits are the larger blue dots with their perimeters dotted in smaller blue (except 02 shows an extension which matches the body, shown later).

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

Photo 6- as above but with the pits numbered and 01A arrowed. Again 01A isn’t very obvious but will become so in subsequent photos.

Photo 7- a close-up of the so-called cove on the head lobe showing 01A and 01.

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

In photo 7, 01A and 01 are dotted light blue. It can be seen that 01A is just the floor of the original pit before it delaminated. Pale orange denotes a massif attached to 01 that slid from its seating with 01. The seating is also pale orange. The 01A floor can be extrapolated from the pale orange massif slide, attached to 01, which means 01 slid too by the same distance and on the same direction vector. So the massif dragged 01 from the extrapolated 01A position to its current position. The extrapolation is also fully constrained by the dark green and mauve annotations. Green are delaminated layers which, when sandwiched back together, follow the pale orange massif slide vector. Sandwiching them back together is reversing the stretch movie, so to speak. Once clamped back together, they form the lower perimeter of 01A when combined with the ridge above the upper green line that was dragged with it (see also photo 8). Mauve is a suspected dyke which supplied 01A/01 when 01 sat on 01A. The gasses forced their way between the two green layers and out into the 01A hole when 01 sat there. This may be why 01A and the area between the green lines (the mauve gas track) looks scoured and not typical of the smooth dust that dominates Ma’at. 01A’s upper perimeter is constrained by the cove delamination (Part 34) and is beyond the scope of this post. 

Photo 8- As photo 7 but with the slide/delamination vector arrows added. 

In photo 8, the right hand arrow depicts the orange massif/blue 01 slide vector in reverse, that is, if you reversed the stretch movie. So it’s pointing from where the pale orange massif is today to where it used to be. When the massif slid up and back, it dragged its 01 hole with it from the original 01A floor. The middle arrow shows the end of the upper green delamination being seated to the corresponding end of its lower twin, again in reverse. The green dotted line has now been extended further up into a zig zag. That zig zag can be seen to have concertinaed upwards in an extended version of the two green lines described above. This zig zag is due to the cove delamination in Part 34. The left hand arrow is pointing downwards, showing that the tip of the upper zig zag point sat on the corresponding tip of the lower zig zag point. And the upper two lines of the zig zag sandwich onto themselves like the lower two described above. Therefore, the whole zig zag collapses down when you reverse the stretch movie. The two sandwiched zig zags used to be clamped to their body match. This body clamping therefore constrains the body gas sources either side of the sandwiched tip to supply the holes either side of the sandwiched tip on the head (i.e. source 01A/01 feeding both 01A and 01 and source 02 feeding 02). The zig zag delamination of course implies that the entire layer hosting Ma’at 01, 02 and 03 delaminated upwards from the lower (head rim) layer too. The zig zag is, after all, the end-on cross-section of those two layers. These two layers indeed exhibit mini-matches along from their zig zag points as well as to their expected body seating (see photos further down).

Photo 9- a simple version of photos 7 and 8. 

Photo 10- this is the same as photos 5/6. It has red lines depicting the notional paths of gasses from the uplifted body bumps to their respective pits on the head. 

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

You can see that the right hand bump supplied both 01A and 01 which delaminated upwards from 01A as described above. The 02 source actually delaminated from the 01A/01 source. There’s lots of evidence for this in Parts 37-39: the opening up of the so-called shallow crater, as well as the ‘gull wing slide’ and the ‘India shape slide’ in those parts absolutely constrain the 02 source to delaminate that distance along that same slide vector from the 01A/01 source. The 03 source then delaminated in turn from the 02 source. Only then did the pits that the 02 and 03 sources were supplying, delaminate upwards with the zig zag described above. Those were of course, the Ma’at 02 and 03 pits. That’s how three sources supplied four holes (or supplied three holes and the original hole base, 01A, that they delaminated from). Of course, the gas supply happened only when the head and its pits sat right down on the bumps. It’s now 1000 metres above the bumps so most of the red lines, across the shadowed expanse of the neck, are just notional links. The red lines kissing the bump tips is also notional because the exact gas sources are just to the right of the bumps (in this view). This is because the gases were exiting to the right of the bumps i.e. out from between the delaminated layers in the reverse direction to that in which they delaminated. The exact sources are shown in the next photo. 

This large amount of outgassing was along the shear line where the head rim was herniating upwards under the tensile forces of stretch and about to shear away. That’s what caused the zig zag: upward delaminating layers. The main source of gas was along the shear line (the Babi/Hapi rim) and it found its way out between the long axis delaminations hence the bumps being like mini-volcanos that are sited both on the shear line and the long-axis delamination interfaces. The difference between long-axis delaminations and upward head herniation delaminations is laid out after this photo section. 

Photo 11- this is the same as photo 10 but with the exact source positions added next to the bumps or mini-volcanos. The sources are actually just out of sight because the gasses ran up the side of the Babi cliff that drops into Hapi. So we’re seeing the very tops of the sources where the bottoms of the holes or pits were originally clamped.

Photo 12- this is the same as the above two but it has a small addition. The beige dots curve up the volcano bump of the 03 source and continue up that curved half-bell shape on the head. The head rim bell shape nests perfectly over the 03 source. The 03 source looks like a scoured-out cave under the volcano bump. The bell shape on the head rim is a chimney leading to 03 and it was supplied from the cave albeit in a much more concertinaed-down configuration. The current chimney length is due to head lobe herniation which stretched the chimney. This is possibly responsible for the pockmark holes dotted up the chimney side that may be collateral gas escape routes. 

Photo 13- the Part 38 header, adjusted. The pits are numbered as above but so too are their sources, dotted with blue lines on the body. 

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

In photo 13, all four bumps are annotated in dark green. They’re pairs of green dots because the first set at the top is the first set from which the other four delaminated. That’s the classic set of so-called gull wings so there’s one dot for each gull wing. The other three sets retain vestigial double bumps in sympathy. The fourth set, almost in shadow is of course irrelevant here. Yellow denotes key fiduciary points that match from head to body. You can now see the bell-shaped chimney from above (albeit in shadow but it is there) and see how it clamps over the 03 source. The full head rim to body shear line match is shown in photo 17 and the chimney is fully visible.

Photo 14- the same as 13 but without the numbering. 

Photo 15- this is the same as photo 2 but with additional matches between the first two layers of the head lobe that delaminated with the zig zag in photo 8. It also shows matches between the head rim and body including triple matches for both layers to the body (in yellow). 


Yellow nearest to us (lower right) is the Ma’at 02 pit’s perimeter extension that was marked in light blue in photo 5. This has a mauve feature adjoining it. The same yellow and mauve lines are seen on the head rim layer below the upper layer and from which the upper layer with the pits delaminated. This is a bled match (features repeated through layers). It’s not very apparent on the head rim layer in this photo (see original). But it is there in many photos. It’s most robust for the mauve feature and for the yellow line up to half way which is why its upper portion has smaller yellow dots. They depict a gentle trough rather than a defined line. The upper-left long spidery triangle is on the body and matches both head layer yellow lines including that extra short line on the head rim. Notice how the body yellow match kisses its light blue source semicircle in the same manner as the yellow extension meets 02 on the head. All the yellow lines are angled the same way. 

The larger bright green dot on the body is the 02 source bump or gull wing set and the small green dots show the extent of the bump. The bump nested into the niche shown in bright green on the head rim. There’s a rope-like ridge running down the body from the 02 source bump. It turns and runs towards top-right in this view and casts a sharp shadow along its length. This is the rest of the delaminated layer of which the 02 source bump is a part. In part 38 this is confirmed by noticing that this line matches to the sharper ridge leading to the upper left yellow lines (the ‘trident shape’, not quite so obvious here). These are mini-delaminations of a thicker onion layer. They’re perhaps 15 metres thick. The others are visible here but beyond the scope of this post. 

Photo 16- this is the same as photo 15 but includes the head rim to body shear line match in various colours. This match is traditionally done in terracotta but with yellow for the cove match. In addition here, we have the bright green 02 source bump which would normally be terracotta and also dark green going up and down over both gull wings for the first set (bump for 01A/01 source). The dark green head match is a cheat- it’s the upper line of the first zig zag, not the lower line on the true head rim. But they’re perfectly in line from this above view. The cove and its seating is yellow. The 01A/01 source is blue as before but since it was squeezing up at shear line inside the cove on its seating, it’s usually yellow. The head match for the 01A/01 source is 01A itself and that’s hidden behind the pale orange massif attached to 01 (visible here but not dotted pale orange). 

Photo 17- this is the same as 15 and 16 with two more additions. It shows the chimney for Ma’at 03 in beige dots (tracing its centreline) and the continuation of the beige line down the 03 source bump on the body. You can now see how the 03 chimney’s curved base nested over the curved base of the bump on the body. The three beige dots on the body run upwards towards us from the base of the bump (a mini volcano, like the other bumps, through being uplifted). The three dots run up to the true rim of Babi overlooking Hapi. This is a spectacular, curved overhang over Hapi. The area under the three beige dots is deeply undercut into a cave which we can’t see here. That’s more evidence for this being a gas source for 03. And of course the smaller blue dots extend, as in photo 2, to beyond where the head rim obscures the majestic, overhanging curve. That curve neatly matches (is concentric with) the curved base of its volcano as well as the curve of its chimney base, now 1000 metres directly above. 

Still with photo 17, the apparent offset of the head lobe rim to its body match here is almost entirely due to Rosetta’s viewpoint parallax but technically speaking there’s a fraction of offset due to head tip. Head tip also contributes to the head terracotta/bright green line being 4% shorter due to foreshortening. However some of that 4% may be due to the fact that the dark green/terracotta corner on the head rim was frilled upwards and backwards which is why its two yellow lines don’t progress to a joined-together point like their body match. The frill needs ironing out so that it extends out to match the green/terracotta corner on the body. Careful analysis of the mauve match shows that the second layer pinned the back of the frill down on the body while the outgassing from under the first bump (the main gull wings) frilled up the edge on that corner. Even Ma’at 01’s perimeter did some of the pinning because it’s part of the zig zag. There much about the frill in Parts 5 and 7. It extends along the head rim further than is visible here. 

There’s also a suggestion of a collateral hole, supplied by the beige 03 chimney. It’s nested up under the second head layer and in line with the chimney and Ma’at 03. This suggests Ma’at 03 delaminated upwards from this larger hole so that both were once one hole, clamped over the 03 source. That would be like 01 sitting on 01A and both being clamped over their respective source. 

Photo 18- this is the same as 5-6 and 10-12 with a few alterations.

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

The red lines are now running more faithfully from their true sources on the body to their respective pits on the head. Before, they were running notionally from the green-dotted bumps. The long lines across the shadow are largely dispensed with in this version. There are still short lines running up from the body sources but they’re just the rough distance of gas travel from the sources to the holes when they were seated on the body and herniating upwards. The 03 source line has three dots running down its volcano bump. They depict the gas flow, hidden on the other side and flowing out of the undercut cave. The 02 and 01A/01 sources aren’t undercut. 

Photo 18 also shows the 03 gas path through the chimney more faithfully and a short offshoot to the intermediate hole tucked under the second layer rim. It also shows a tentative position for an intermediate hole along the 02 gas path. This is only because both the 01 and 03 paths have intermediate holes. Seeing as this comet is almost obsessively symmetrical when delaminating the suggested hole is annotated just as a suggestion. It’s not readily apparent as a hole but may make itself apparent as a narrow collateral dyke. There’s already a tiny collateral dyke within the yellow match line just below the suggested position and that tiny dyke has the 02 path running right under it. 

Finally, the dimple just above the head rim, midway between the 02 and 03 paths is the position of the bright green dot that fits over the 02 bump in photo 15. The 02 bump is dark green here, in keeping with the others as that’s their usual colour. 

Photos 19 and 20- the ESA regional maps.


This Part should be regarded as a twin to Part 52. Reading that part will give more background as would Parts 38-41. However, this part can still be read and understood in isolation. As is often stated here on the stretch blog, the process by which stretch theory explains the delamination of the pits known as Ma’at 01, 02 and 03 is described below as if it’s fact so as to spare the reader endless qualifications and conditional statements. Of course, it’s a hypothesis but after two years of scrutinising this small area (hundreds of photos for very many hours) I’m sure the evidence is highly compelling. It’s so extensive that only a small amount of that evidence can be presented in this post with the rest left to links to other posts. 


The Rosetta mission is going to land the Rosetta orbiter on 67P/C-G on 30th September 2016. At that instant, all transmissions will stop and the mission will end. 

On the way in to land, Rosetta will analyse the pits at Ma’at on the head lobe near the head rim. They’re named Ma’at 01, Ma’at 02 and Ma’at 03 although Ma’at 02 has recently been renamed ‘Deir El-Medina’. Rosetta is slated to land between 02 and 03. I’ll generally refer to the three pits as 01, 02, and 03 for brevity here with occasional reference to their longer names. 

I’m posting this because I think the largely accepted view that the pits are probably sinkholes might affect the decisions on how to take the data on the way in to land. For instance, it might affect the decisions regarding attitude orientation (where Rosetta is pointing) and which instruments are switched on for taking data. At the time of posting this, there are intense discussions as to which instruments can be allowed to operate and to what extent because of acute power and telemetry constraints. This Rosetta blog post published recently outlines the constraints:

Of course, the mission scientists will be attempting to point the instruments into the pits so you might think the pointing issue is academic. But the pits have delaminated from each other. The knowledge of their delamination might prompt the pinpointing of certain specific areas, perhaps at the stratum of the delamination in successive pits. Or extra scrutiny of the boulders in Ma’at 02, knowing some might’ve fallen from the sides of Ma’at 03 as it delaminated. 


There’s also a fourth pit on the head, next to the other three, that hasn’t been identified as such because doesn’t look like pit. It’s the most most important of the four because it’s the original one the other three delaminated from and today it’s just the old, scoured-looking floor with no walls. It’s lying right under the landing flight path (I believe) and data will probably be taken from above it. But surely it would be a good thing to know in advance that it’s not the comparatively boring flat expanse it seems to be, but key to everything Rosetta is analysing in the last moments of her 2-year mission. 


The long-axis delaminations at Babi/Hapi are presented in more detail in Part 38:

There’s additional information on the long-axis delaminations in Part 52:

Part 52 was linked in a comment I made on the “Celebrating Two Years at the Comet” post in the hope that a low flyby of the Babi slide would provide data on the jet there that’s related to the 02 source and therefore to Ma’at 02 itself. The Babi slide itself (Part 40) is beyond the scope of this post but puts it in perspective. 

According to stretch theory, pits 01, 02 and 03 came about as a result of onion layers delaminating along the Babi/Hapi rim under the influence of long-axis stretching of the comet. 67P stretched into a quasi ellipsoid and continued stretching even more along the long axis of that ellipsoid. This eventually caused the diamond shape we see today that defines the body and, partially so, the head lobe. The long-axis delaminations along Babi (and Seth) were therefore trying to accommodate the stretch. Long-axis stretch is why Hapi and Sobek are longer than Bastet and Anuket, resulting in a neck with an elongated cross-section that’s aligned with the long axis of the comet. The cross-section and its long-axis alignment can be seen in Hirabayashi, M., et al 2015 extended data figure 2:

The tensile force vectors operating during the stretching process would have been brought about by spin-up via asymmetrical outgassing. 

The long-axis stretch and delamination process happened when the head lobe was still attached to the body, prior to shearing. The shear line runs along the Babi/Hapi rim where the delaminations are most apparent. 

Since the layers were delaminating along Babi, any particular feature on the original layer that delaminated would be reproduced in the successive delaminations. This would include the pits/holes and the uplifted mini-volcanos supplying them. There are also other features such as a an obvious trident shape. This is why the mini-volcanos in photo 4 are so similar and near to being equidistant:

Photo 21 (4 reproduced).
Copyright ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0/A.COOPER

Since the tensile force vector for the long-axis delaminations was in line with today’s Babi/Hapi rim and the head was still attached, it meant that the delamination layer fronts would be expected to be at 90° to that vector, or at least notionally so, all other factors being equal. However, this is notwithstanding the initial layering configuration prior to delamination and indeed this appears to make the Babi delaminations angled forward towards Bastet somewhat. This may be a primordial arrangement but it might instead be due to the pinning of layers at the proto-Anuket region, causing the V-shaped configuration we see on the head today and, rather disguised, on the body too (see Part 27, ‘Signature 3’). 

However, notionally speaking, the delaminated layer fronts were ‘draped’ across the long axis tensile stress vector. Since the head was still seated on the body during long-axis stretch, these draped layers ran from proto-head to the body, across the future head lobe shear line at the Hapi rim. 


The head lobe shear line itself runs in the long axis plane or to be more exact, in a plane that is parallel to the long-axis plane. The shear line is also the line along which the proto-head started herniating. So it was a weak spot. The head chose to herniate along, and eventually shear away from this particular line because it was literally a shear line: a sharp tensile strain gradient occurred across it causing differential strain forces. Whilst the long-axis tensile forces were the same across this line (no gradient) the differing tensile force resistance (strain resistance) of the matrix either side of the line resulted in the strain gradient. This caused shear along the Hapi border with both Babi and Seth and it’s indeed the very reason for all three regions exhibiting different morphologies. Furthermore, it’s the reason for Seth and Babi being rough mirror images of each other fanning out from the straight portion of the Hapi rim. That straight portion was where the shear force was most potent, causing the most defined head shear. At Bastet and Anuket the shear turns into a yanking-up which is why the neck looks similar at both ends (Bastet: Part 21 and Anuket: Parts 24,25).

On one side of the shear line there were the more susceptible Babi delaminations; on the other, the proto-Hathor cliff, still clamped face-down on Hapi, was delaminating more subtly: it delaminated into narrower layers (the strong striations we see today) and did so in a fan shape. This is why both Hathor and Wosret are fan-shaped.

The greater resistance to strain of the Hathor cliff matrix when seated on Hapi was presumably due to the matrix being colder and more brittle. This property would become more pronounced with the depth of Hathor under the seated head and lead to increasing strain resistance towards the central area. This translates to less stretch at depth and would be an elegant explanation for the fan shapes at Hathor and Wosret: less stretch at the bottom of the fans and more stretch at the top couldn’t help but form a fan shape. 

This gradation of strain resistance with depth would also account for the apparent directing of the long axis tensile forces around the proto head lobe and the consequential sudden strain gradient at the Hapi rim where the fan-shape strain mechanism gave way to full-blown layer delamination across Babi. That would explain the location of the shear line and by extrapolation, the shape the head rim traces. Thus, the symmetrical shape of the head when you look down on it is explained (yet again, aligned with the long axis).

Moreover, the nodule sitting centrally at the base of the Hathor cliff could represent the threshold at which no stretch at all was possible (100% strain resistance). It would be like trying to stretch and deform an avocado. The flesh would stretch and deform around the unyielding stone and the brittle skin would crack. If made of several layers, the skin would delaminate as well or instead. The Hathor cliff matrix would then represent an intermediate phase of reluctant, partial stretch as would happen if the avocado stone had a thick, semi-ductile shell. For 67P, this translates to less stretch at depth with the tensile forces being directed around whatever doesn’t want to stretch or wants to stretch less. 

If 67P was a perfect ellipsoid the tensile forces wouldn’t have a proto-head to get forced around and work free via shearing along Hapi and Sobek. However as soon as any proto head lump started to herniate under stretch, it could be visualised as an avocado with two stones or a peanut-shaped stone. The tensile forces would start to be directed around the girdle of the peanut, this causing shear at the surface in line with the girdle i.e. at the shear line/head rim line we see today. 

This is the first mention of the Hathor fan-shape explanation on this blog as well as the fact of the shear line being an actual shear line where shear forces acted either side of it. These will of course get their own posts. They’ve been brought forward here because of their relevance to the long-axis delaminations. 


The Ma’at pits were caused by outgassing, yes, but in a shorter, sharper catastrophic episode than through gentle sublimation. They never had roofs that fell in after a cavity had gently eroded away underneath. They acted just like volcano calderas and even had their uplifted mini-volcanos that supplied them with the fast-sublimating gasses emerging at the shear line during head lobe herniation. These are clearly visible on the body today, aligned 1000 metres below 01, 02 and 03. When the head was still attached, the already-delaminated pits sat on (or rather, just next to) the mini-volcanos.

Whilst the Ma’at 02 pit probably did sit on or very near its bump at the time of being supplied, the bump delaminated a little further from the outgassing source signature we see today. So it continued delaminating longways after 02 delaminated upwards. This is evidence that a small amount of long-axis delamination continued even as the two ‘zig zag’ layers were delaminating upwards under the influence of head lobe herniation. 

This slight continuation of long-axis delamination after head herniation commenced is perhaps corroborated by the 03 path. 03’s outgassing source (the undercut cave) hugs its volcano very faithfully but, together, they are slightly further displaced along the body from today’s 03 pit on the head. The 03 source has the strongest 3D match to 03 itself: a bell-shaped chimney, starting at the third volcano which is seen to curve up across the first layer of the head lobe to meet 03.

So the gas source for 02 is slightly displaced from its volcano. The other two sources are hugging theirs, just to one side. This is consistent with the gases at the shear line finding their way out firstly from the shear line itself but also from under the draped ‘lasagne layer’ in the reverse direction of delamination as you’d expect. 


The mini-volcanos are what are known as the ‘gull-wings’ in this blog because the first set that was matched from body to head looked like a pair of gull wings. That was in Part 5, December 2014. The other three sets of gull wings delaminated from it, along the long axis, so they partially preserve the gull wing shape. It wasn’t known in 2014 that they had delaminated but there was much evidence of former outgassing from in front of all three sets that was noted at the time. I noted the outgassing evidence from in front of sets one and two, suggesting it was the reason for the uplift. Rosetta blog commenter, Robin Sherman, often referred to the possible ejection of material from the cave under the third set because it looked scoured-out somehow. Part 5 talks of the first set being uplifted, hence their gull wing shape. Parts 7 and 8 discuss the path the gasses took from the soon-to-be Hapi that was still trapped under the head lobe. That discussion incorporated the second set of gull wings. Hence, all three sets were identified as experiencing catastrophic outgassing and therefore being uplifted. And of course, it has to be remembered that this entire section of body had already been matched to the head rim underside anyway. That was done in Parts 1 and 2. But still, it wasn’t realised that the gull wings or volcanos were delaminated from each other. That came in Part 38:

You can click through to Parts 39-41 from Part 38 for the recommended extra background reading. 

The long-axis delaminations and how they match to the head layers is explained in much more detail in Parts 38-41. Regular readers may notice that the numbering scheme for gull wings above is different from in Part 38. The numbering above is for clarity regarding the matching of mini-volcanos to Ma’at holes in this post.


Photo 22- the cross-cut strata. 

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


Yellow- true strata. The body components look wiggly due to being flaccid after recoiling backwards along their length. 

Red on the head- quasi strata. These are the head lobe herniation ‘strata’ which are really just passive chunks of matrix that yielded to head lobe herniation. They just happened to yield to that upward tensile stress at a ~50-metre depth, tore from the shear line and recoiled up the head lobe. In so doing they hid the real strata within them (see below). The two upper red lines are the layers that recoiled. The third bottom one is the head rim. The top red line is the one at which the true head lobe strata (yellow) appear to stop dead. In fact, the head strata carry on within and across the red quasi strata but are almost hidden except in the head cove. Red and yellow can be seen cross-cutting in the cove. The uppermost red line hasn’t been discussed so far in this part because it was the very top head herniation layer and it slid up past the Ma’at holes (Ma’at 02 is dotted blue, just below this line). The layers above the first (rim) line and second line are the two that we’ve been discussing and which have the zig zag at their ends. The uppermost line appears to have been set back from the shear line when on the body which is why there’s a blue dot on the body. That’s the Ma’at 02 body delamination. So the head took the bottom two layers of the Ma’at pits that sat on their Ma’at sources. And the body kept the very top layer. That blue dot on the body (or rather its surrounding walls) sat on top of where Ma’at 02 is today.

Red on the body- these are quasi-strata as well. The same principle applies as for the lines for the red quasi strata on the head but the signature of the layers is less obvious. The red lines on the body were originally joined to the lines on the head: bottom body to top red; top body to bottom head; and middle to middle. 

Blue- top blue is Ma’at 02; middle is the 02 source; bottom, on the body, is the jet that originally sat on top of Ma’at 02. Its adjacent red line wrapped round it and clamped to the red line above Ma’at 02 when they were sitting on their common 02 source at the shear line. 

The two-way delamination explains the paradoxical, different-angled ‘strata’ in this location at eastern Ma’at. They delaminated in two directions: along the long-axis stretch vector and along the upwards head herniation vector. They’re at a notional 90° angle to each other as mentioned above. That’s the best way to visualise it close-up. From a distance it’s a bit more nuanced.

The two-way delamination is the reason the strata appear to cross-cut inside the head lobe ‘cove’ above the border between Seth and Babi. And it’s why head strata appear to stop dead in their tracks at a straight line running all the way from western Ma’at, through the cove, and on round to Bastet in the east. That impressive line is the upper layer of the head herniation delaminations. These delaminations could be viewed as sliding or even recoiling upwards after having torn at the shear line due to the herniating head pulling on them. 

Together, they form that impressive green zig zag at their ends (photo 8). The zig zag betrays the delamination. But these two upward herniation layers are not real strata. They’re just layers of passive comet matrix which were torn at the shear line by head lobe herniation and recoiled (Part 41). They just happened to fail at that particular ~50-metre thickness, one after the other, as the herniation progressed. But the most important thing to remember is that both upward-recoiled layers contain all four true strata that originally delaminated along the long axis before tearing right across their width at the shear line and recoiling upwards. Those are the true strata and they supplied all the material for the two upward-recoiling layers which appear in photo 22 as quasi-strata. The quasi strata almost completely hide the true, long-axis-delaminated strata. The true strata became almost totally hidden due to tearing across their widths, succumbing to the quasi strata’s 50-metre-thickness requirement and becoming flaccid as they recoiled up the head and along their length.

But these hidden, true strata are traced in Part 41. And most crucial of all for this post, the true strata are the long-axis-delaminated layers that contain the delaminated holes, Ma’at 01, Ma’at 02 and Ma’at 03. 


I hope the mission scientists can take this into consideration in their deliberations on how to take the pit data as Rosetta descends to the surface on 30th September 2016. 



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. 


Part 61- Close-up gifs of the Bastet Pancakes Matching to Aker

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


Part 51 header for context. The two left-hand mauve features on Bastet and Aker (head and body respectively) are the same as the much closer-up mauve features in the “head” and “body” gif components. 
Copyright ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0


This post is a follow-on part to Part 51 ‘The Bastet Pancakes and Their Relationship to Neighbouring Slides’. That part showed how the pancakes, which are thought to be splatted cometesimals are in fact slid pieces of crust which used to to sit in the two dips at Aker. 

Part 51 showed the matches in several photos and included two gifs. One was a bit crudely dotted and they weren’t close-ups showing the multitude of subtle matches. This post shows a gif that’s a closer-up version of those two gifs. 

For the sake of absolute transparency, the nature of the Part 51 header, a NAVCAM photo, didn’t lend itself well to cropping for the gifs. Aker was the problem. It was to do with it enlarging when rotating it to horizontal and moving out of frame. So an OSIRIS picture was found which was from the same angle. However, small discrepancies in distance meant it had to be resized by about 10% to fit the NAVCAM Bastet component. Seeing as the matches had already been made on the single header photo where no resizing was needed, this seemed a reasonable course of action because we know that the matches are the same size in that photo. Any objection to this approach would have to contend with the fact that the head lobe is showing a multitude of matches that are all 90% the size of their body counterparts. 

The viewing angle issue was negligible and that’s betrayed by the fact that the matches fit in two dimensions when resized. If there was a perspective discrepancy, they would be the same length on resizing but not the same width. 

OSIRIS still used for gif:




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

To view a copy of this licence please visit:



All dotted annotations by Andrew Cooper

Part 60- 150m Massif Slides 250m on Ma’at 

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

(It’s just this one photo in this post with crops, annotations and gifs. Full credits at the bottom). 


Just one of many small-scale slid layers that are sitting in the queue of posts. This one is on Ma’at, just above Nut. It does betray matches from the perimeter of the slid layer to its seating if you look carefully (especially a saw-tooth line down the left perimeter). However it betrays bled matches very well. Bled matches are features that are repeated down through the layers in the third dimension. Somehow they’ve bled through to the surface. There’s a lot on bled matches in Part 45. It’s akin to spilling wine on a newspaper and the stain bleeding through the pages- except it presumably happened in reverse through sublimating gases finding the line of least resistance through the layers and possibly depositing refractory material on the way through. 


The original NAVCAM photo is first, then a closer version zooming into the area in question, then a succession of stills and different speed gifs constructed from the stills. It’s fairly self-explanatory.

The coloured versions are fiduciary points to direct you to where the bled matches are. They’re by no means exhaustive e.g. the yellow line continues all the way round that lighter feature and there are two more v shapes below the mauve-dotted one. Also, once you’ve familiarised yourself, you can return to the main non-gif photo and see the subtler matches to the right, along that wavy finger extension. The two bright green lines are not strictly on the main slid section but on another delaminated layer under it because everything was delaminating at Ma’at (see Part 29). 

It’s pointless just looking at the dotted versions only. They are just an initial guide. 

It’s advisable to hold the tip of a pen to any point of interest. It’s surprising how little it moves from gif image to gif image when you do this. The shadowing and surrounding features make it look as if it’s not aligned so well but it’s an illusion. The residual non-alignment is due to the slid section being at a slightly different profile angle (over the curving head) and human error in constructing the gif. You can even trace your pen tip down a certain line as the gif is running. 

Using the fast gifs is recommended for getting your bearings. Then slow for main analysis then fast again for better confirmation of what you’ve found. Then back to slow, and so on. 



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.