OSIRIS detects hints of ice in comet’s neck

This post is based on information provided by the OSIRIS team at the Max Planck Institute for Solar System Research (MPS) in Germany.

If it could be seen with the naked human eye, the nucleus of Comet 67P/Churyumov-Gerasimenko would be a dark grey all over. With its array of specialised filters, however, Rosetta’s scientific imaging system OSIRIS can discern tiny differences in reflectivity at different wavelengths across the comet’s surface. In turn, these differences can reveal clues as to the local composition of the comet.

The image shown here focuses on the Hapi region of the comet. Hapi is located in the neck between the comet’s two lobes and has in the past months proven to be particularly active, the source of many of the spectacular jets of dust and gas seen in the wider-view images.

 False colour image showing the smooth Hapi region connecting the head and body of comet 67P/Churyumov-Gerasimenko. Differences in reflectivity have been enhanced in this image to emphasise the blueish color of the Hapi region. The scientific data was acquired on 21 August 2014 by the scientific imaging system OSIRIS with the filters centred at 989, 700 and 480 nanometres. During these observations Rosetta was 70 km far away from the comet, and the corresponding spatial resolution 1.3m/pixels. Credits: ESA/Rosetta/MPS for OSIRIS Team MPS/UPD/LAM/IAA/SSO/INTA/UPM/DASP/IDA

False colour image showing the smooth Hapi region connecting the head and body of Comet 67P/Churyumov-Gerasimenko. Differences in reflectivity have been enhanced in this image to emphasise the blue-ish colour of the Hapi region. The scientific data was acquired on 21 August 2014 by the scientific imaging system OSIRIS with broad-band filters centred at 989, 700, and 480 nanometres. These images have been combined here as red, green, and blue, respectively, and the composite has been processed to enhance the slight colour differences. During these observations Rosetta was 70 km from the comet, and the corresponding spatial resolution is 1.3m per pixel. Credits: ESA/Rosetta/MPS for OSIRIS Team MPS/UPD/LAM/IAA/SSO/INTA/UPM/DASP/IDA

Three separate OSIRIS images were taken through three broad-band filters centred at 989, 700, and 480 nanometres, respectively. These were then combined as red, green, and blue, respectively, to make a colour image.

Because the images were taken sequentially, there was some comet rotation and spacecraft motion between them, meaning that accurately aligning the images before they are combined is non-trivial. Small errors in this process can lead to small-scale colour features which are not real, so it is important to focus on the large-scale colour variations between Hapi and neighbouring regions as an indication of broad compositional differences.

It is clear from this image that indeed Hapi reflects red light less effectively than most other visible regions on the comet and thus appears slightly blue-ish. Importantly, this blue-ish colouring might point to the presence of frozen water ice at or just below the dusty surface of Hapi.

“Even though the colour variations on 67P’s surface are minute, they can give us important clues,” says OSIRIS Principal Investigator Holger Sierks from the Max Planck Institute for Solar System Research (MPS) in Germany.

The scientists believe that the distinct reflectivity properties of Hapi hint at a higher abundance of frozen water at or near the surface in this region. The colour differences are very slight and could be explained by the presence of just small amounts of water ice peeking through the dusty shroud that covers much of 67P/C-G at the moment. While some dust present on the surface may be due to local depositions of dust associated with activity, some is also likely left over from the comet’s last passage around the Sun. Much of this dust shroud is expected to be blown away as the comet becomes even more active closer to the Sun later this year.

Previous missions had observed similar features on comets Hartley-2 and Tempel-1, and also associated a blue-ish spectrum with the presence of frozen water. While OSIRIS can only take images in a limited number of spectral bands, Rosetta is equipped with other instruments including VIRTIS, capable of making an unambiguous identification the spectral signature of frozen water molecules in infrared reflection.

“We are excited to see whether our suspicion will be confirmed,” says Sierks.

As the smoother surface of the Hapi region gives way to the more rugged terrain of the surrounding areas, the reflectivity changes, too.

“We see that the reflectivity properties are closely correlated to the surface morphology,” says OSIRIS scientist Sonia Fornasier from the Paris Observatory.

The Hapi region differs from the rest of 67P/C-G’s surface in many respects. Not only is it much smoother, it is one of the main sources of activity in the comet’s northern hemisphere at the moment – early dust jets originated there.

“During perihelion, when 67P heats up significantly, Hapi will be hidden in northern polar night. Outbound on the comet’s orbit, from March 2016, Hapi will receive solar heat again,” says Fornasier.

“At this large distance from the Sun, it will then be very cold. Hapi might therefore be a region that has been able to retain ices on its surface during past orbits around the Sun and has thus enough ‘fuel’ left to create the fireworks of activity we have witnessed in the past months.”




  • A.Cooper says:

    This finding is not a surprise when viewed in the light of stretch theory. The Hapi and Hathor regions constitute the interior of the comet that has been exposed due to the head lobe shearing away from the body lobe during a spin-up event.

    Since the undisturbed crust of the two lobes is likely to be largely spent of volatiles, it would be expected to show up as being redder. In contrast, the exposed core would contain ices that had never been anywhere near the surface before the shearing event, so they would never have had a chance to sublimate. Some portions of the neck would have been as deep as nearly 1km before being exposed.

    That is why ice is seen in higher abundance in the neck region: the depletion process in this region still hasn't run its course to the extent that it has occurred in the crust.

    Spin-up is not controversial. It is discussed in the following pdf by Paul Weissman of JPL, who cites several prominent scientists in the field, some of whom are involved in the Rosetta mission. The pdf corroborates the comments Marco and I have made here and elsewhere about spin-up rotation periods, multiple fragmentations and the delta V required for departing fragments to escape:

    This is my piece on the shearing and stretch event:

    One of the 7 papers published in Science Magazine on 23rd January 2015 referred to the Hapi/Hathor region as being the interior or core of the nucleus due to their findings. These papers are now paywalled so I can't check which one it was. It was clearly dawning on them that there is something intrinsically different about the evolution and consequent morphology of this region when compared to the rest of the surface.

    • Harvey says:

      Certainly an interesting presentation.
      But there is one assumpton in the modelling that bothers me for 67P, namely:
      "Nucleus is gravitational aggregate – no binding forces"
      I think its pretty clear that is not true for 67P from its morphology; if those binding forces are significantly greater than the -very small - gravitational forces, it invalidates the model.
      Also *fragmenting* & 'stretching' are very different. The former could be unbound, or brittle fracture. The latter requires plastic deformation, & its unclear to be that any of the candudate material for 67P will undergo plastic deformation in the temperature range of the comet.
      That said, it clearly supports the importance of spin-up in some cometary situations; but I'm not so sure it supports a spin up theory for 67P's shape really.

      (Gutierrez, (slightly confusing re which one - there seems to be more than one author of this name in cometary science & no initial given) one of the main authors cited, is also an author of a recent Science paper with the following abstract:
      "Images from the OSIRIS scientific imaging system onboard Rosetta show that the nucleus of 67P/Churyumov-Gerasimenko consists of two lobes connected by a short neck. The nucleus has a bulk density less than half that of water. Activity at a distance from the Sun of >3 astronomical units is predominantly from the neck, where jets have been seen consistently. The nucleus rotates about the principal axis of momentum. The surface morphology suggests that the removal of larger volumes of material, possibly via explosive release of subsurface pressure or via creation of overhangs by sublimation, may be a major mass loss process. The shape raises the question of whether the two lobes represent a contact binary formed 4.5 billion years ago, or a single body where a gap has evolved via mass loss."

      • Marco says:

        I'd be quite interested in the difference in morphology under the surface compared to the surface or say the top 50 metres. Stretch requires the surface to become hard and brittle(over the time of stretch) to create a kind of exoskeleton which holds firm to compression (and thereby shear forces) for when stretch has gone its course and gravity takes over. Underneath the hard brittle exterior, for the comet to stretch at some future point, the interior would need to retain "gravitational aggregate" properties, which would be far weaker to tensile forces than to compressive forces. Yes, a homogenous mass with consistent properties and predictable internal temperatures could not have these properties. If future stretch is found to occur, that would tell us all we need to know about whether there is the required plasticity.

        Regarding spin up the following research on the spinup of comet Temple2 shows that the spin period can be known accurately enough for the acceleration to be measured rotation by rotation.


        Note well that Comet Temple had an overall tendency to spin up, but for part of its orbit, it is slowing down. 67P is likely to show something similar. It may slow down during Equinox, for instance, then speed up rapidly during the southern summer.

        • Marco says:

          That should be Temple 1 9P for that article link. The spin down "might" be a proxy for stretch just slowing down due to the conservation of AM when stretching. This to explain the "anomalous" slowdown through an otherwise consistent spin up the rest of the orbit of 9P (and we should look out for similar with 67P)
          Of course, I would be happy to see alternatives.

          • mario filho says:

            marco, can you check the image 'Hapi_and_Hathor' cause i just find something strange there, like a image of a ET rising the mountain. I guess it is not paredolie. Thanks. mariosmf2@yahoo.com.br

          • Marco says:

            Hi Mario,

            What is the date of the photo of which you refer to?

        • Marco says:

          I must be psychic. On the 18th of March, I mention that 67P may be slowing its spin even though through last perihelion it was spinning up. Then on the 20th of March, Andrea tells the BBC that Lo and behold, surprise surprise, 67P is slowing its spin now at 1 second per day vs 33millisec per day last year.
          If Temple 1 is anything to go by, 67P will revert back to more rapid spin up by Perihelion, much more than reversing the cumulative slowdown happening now. If both slowdown and spin up are from asymmetrical outgassing, we should see jets rapidly change from being predominantly counter-rotational to pro-rotational. Importantly, neck jets are likely to have limited effects, and may even die down.

    • Harvey says:

      To state that this is 'not controversial' rather overstates the case; there is a very limited literature on the subject.
      The best paper is probably this one:
      "Effect of the tensile strength on the stability against rotational breakup of icy bodies" I Toth & C M Lise 2010. It references the very limited earlier work. However the paper, now 5 years old, has *never* been cited, which does not indicate any wide acceptance. Its principle conclusion is that "Accorging (sic) to this study most of the observed comets, centaurs, TNOs, and
      MBCs are stable against rotational breakup, with a few notable exceptions. E.g., we suggest
      that the rotational fission is a likely scenario for the Haumea-family in the Kuiper belt."

      General title searches under a wide variety of title terms found nothing further other than the references in Toth et all 2010.

      The reference cited
      is in fact a very 'weak' reference. It is a (possibly draft - the references it contains are incomplete) power point. It is undated, not attributed to a specific conference, and undated. We do not know if it was refereed; it is not cited in the standard databases, we dont even know it was actually presented. The author did not, as far as I can see, publish a follow-up full paper. Basically it says that 'if you spin up a purely gravitationally bound object so that 'centrifugal force' exceeds the gravitational forces, it will fly apart'; that is hardly a surprise! The earlier work is extremely limited, much of it refering back to Boehnhardt, H, 2004 - which is about 30 lines & one formula on this subject in a 671 page book!

      So on the basis of a fairly careful literature survey
      - the idea is represnted very thinly
      - it is supported for a very limited class of objects by the -uncited- Toth et al paper
      -most of the earlier work is *extremely* limited & applied to purely gravitational bound object - 67P is manifestly not purely gravitationally bound as a glance at it shows.

      Furthermore, this all applies to *breakup*, not *stretching*; I can find nothing in the literature at all supporting stretching, which is significantly different.

      Despite all this, I would *not* completely rule it out. We dont know enough about the properties of the strange, porous materials 67P is made from, or its history, to reliably do that. But there is essentially nothing in the scientific literature which directly supports it.

      • Marco says:

        There is a lot of evidence of breakup in comets in general. With 67P we have an opportunity to see up close if there is any evidence of "missing slabs" or how a comet can break up into two or more comets. Sure, we are unlikely to see it actually happen, but we can look for evidence of it happening in the past, or how it may happen in the future. There is evidence of monoliths breaking away from surface features around the amphitheatre and ending up a few hundred metres away. Sublimation and erosion just cannot explain their movements, while stretch theory can. Erosion and collision just cannot explain matching features and the angled alignment of strata between the lobes. It is contact binary and erosion which is being falsified by the data. Stretch theory is merely difficult to believe. Contact binary is as ridiculous as EU, and erosion explanations are not much better.
        Regarding not finding anything in the literature: there hasn't been good evidence of stretching until this mission, and yes (inadequate) models have not shown it to happen.

        • Harvey says:

          Dismissing a paper as 'inadequate' without saying why is a rather poor form of argument. The Toth and Lise paper presents arguments for their conclusions; which of them is 'Inadequate' and why?
          The literature analysis simply shows, as I clearly stated, that is is simply untrue to say that rotational breakup is 'not controversial', if we take that to mean 'widely accepted as true'. That is not the case. The concept, just barely, exists in the literature, and is supported by one uncited paper for a small class of objects.
          That does not mean the concept is wrong; but it does mean it is not 'not controversial' as was claimed.

          • Harvey says:

            Oh, re contact binaries.
            I have myself previously expressed some puzzlement about this, at first sight it seems improbable.
            But I haven't dug out and read the papers which support the concept. It seems to be established in the literature, clearly a significant number of qualified people who have thought about it far more than I have think it's credible, I take account of that.
            So personally I'm not going to call it 'inadequate' or 'ridiculous' until I've read the papers and have cogent , specific arguments to refute it.
            At least it doesn't break multiple laws of physics and fly in the face of a stack of direct observation, as the ludicrous EU 'theory' does.

          • Marco says:

            I guess I didn't quite make myself clear. The paper is really very good science. However, good science with incorrect assumptions may give inadequate answers when pertaining to comets. The assumption of homogeneity is the main one that I think is a problem with comets. The test of whether we should take the results to mean stretch cannot (or very probably cannot) happen is not in the quality of science, but how well it predicts the shapes that we have noted for comets. Surface evidence of the processes mooted should also be looked at. If there is no surface evidence of the processes in the models that do give a bilobed shape, we should hold the model(s assumptions) to be inadequate for the object at hand.

          • A.Cooper says:


            (2 attempts at posting this comment so far. I think it's getting eaten by the comments box)

            You said,

            "The literature analysis simply shows, as I clearly stated, that is is simply untrue to say that rotational breakup is 'not controversial', if we take that to mean 'widely accepted as true'."

            I think your assumption that I meant 'widely accepted as true' is key to a misunderstanding of what I meant by 'not controversial'. By that statement, I meant that at least the notion of spin-up for comets is being discussed by experts in the field and this is not sparking any sort of heated debate. In other words, I detect no sense of trepidation for those astronomers who bring the subject up. Their ideas may be discussed and countered by some, yes, but there is no sense of stepping over a line whereupon they'll be rounded on by their peers.

            The opposite of 'widely accepted' is 'not widely accepted'. A proposition can be 'not widely accepted' whilst remaining uncontroversial if the critics of that proposition recognise the theoretical feasibility of the process, yet raise objections to its practical realisation in the course of reasoned debate.

            No one can object to the theoretical feasibility of spin-up causing fragmentation, especially when you start from the premise of a solely gravitationally bound body with no tensile resistance. After all, you said yourself,

            "Basically it says that 'if you spin up a purely gravitationally bound object so that 'centrifugal force' exceeds the gravitational forces, it will fly apart'; that is hardly a surprise!"

            Any sensible person would agree and they might go further to say that although we all know there will be some sort of tensile resistance, it makes perfect sense to start with a gravitationally bound body only, with no tensile resistance so as to set the lower threshold for spin rate. That's because if spin-up to fragmentation somehow didn't occur, then the theory would be dead even before tensile resistance was invoked. Having established that minimum rotation threshold, tensile resistance can be introduced and the spin increased to overcome that resistance. Further tensile resistance will require a still faster spin rate.

            It is only at this point that reasoned discussion might take place over the proposition of spin-up and those discussions would centre on the tensile strength of the comet and the rotation period needed to overcome that tensile resistance. I doubt such discussion would be at all heated or that the proponents of lower tensile resistances or higher spin rates would be ostracised from their field. That is why I said that spin-up is not controversial. It's simply an observation of the fact that spin-up is at least being discussed and without any trepidation on the part of the proponents that it it might spark controversy.

            The reason for saying spin-up isn't controversial was because one would think that in the Rosetta camp it actually is controversial to mention it. It hasn't been mentioned once, as far as I know, by any of the hundreds of scientists involved whether in press releases, papers or presentations at AGU14, AAS15 and LPSC15. Granted, the 20-minute spin-up was mentioned in the Mottola paper (Sept 14) but that's not going anywhere near the suggestion of spin-up to fragmentation.

            Hopefully, this will clear up any misunderstanding. I also hope I'm not accused of resorting to semantics. I just feel obliged, in the light of your comments, to make it clear that I simply used the word 'controversial' in its obvious, everyday sense. I most certainly didn't set out to say anything "untrue". I may be into spin for comets but not the other sort for misleading people. It's not my style.

            I'll try to get back to the science issues in another comment but I'm a bit behind at the mo. Maybe in a future Rosetta post.

          • Gerald says:

            A.Cooper, spin changes are nothing fundamentally new, so there is no need to mention it in a paper, unless there are actual measurements about the precise amount.

            See also

    • Gerald says:

      Spinning up an only gravitationally bound object usually leads to a flattened ellipsoid, then to a (Dyson) ring or a core-ring system, then to a two-body system:

      • Marco says:

        Plenty of assumptions go into those referenced models. One word in particular that is hard to escape is "homogenous", a word rarely heard about when talking about comets, especially this comet. Heterogenous is the word I hear all the time. Yet when thinking about possible internal and external material composite properties, we think that we will get model results faithful to the reality. I'm not saying that the models are "wrong" but how many of them come up with shapes like the comets we see?

        • Gerald says:

          Well, the model calculations show, that the shape of the comet cannot be achieved by spin-up of a more or less homogenious just gravitationally bound system, unless you spin up to the formation of a two-body system, which may then spin down to a contact binary.
          Although this question is investigated since 130 years, there seems to be no spin-up model resulting in a duck-shaped configuration.
          Brittle objects just disrupt, when spun up.
          So being confident, that a model will be found, is nice, but you'll need lot of good luck and probably lots of very narrow constraints to find a solution which hasn't been considerd before.

          • Marco says:

            I'm not sure why you would believe the negative results of models(ok 130 years of models, only the last few years having computer simulations of a sufficient resolution) yet disbelieve the physical evidence before us telling us the opposite. We can't even propose a realistic model until we *know* the interior properties of the comet. I can assure you none of the models done over 130 years have considered a hard brittle outer layer and a soft plastic inner, but that is exactly what the circumstantial evidence is telling us.

          • Gerald says:

            You'll need a formal description of your suggested initial conditions, then run computer simulations to see whether it's possible to get the observed shape, or at least some result resembling the a two lobes configuration.

            I see this as the only way to get beyond baseless speculations.

          • Marco says:

            Let's see. Which speculation is baseless. One speculation is based on newly observed evidence and what mechanisms can explain the new evidence without reference to models whose assumptions are as yet unverifiable with respect to whether they apply to comets, as we don't know their dynamic physical properties.

            Another speculation is based purely on the theory of subtraction. If you erode or ablate an initially spheroid shape, and the less volatile elements are within a roughly peanut shape, then erosion would cause the peanut shape.

            Another speculation is based on the theory of addition. If two roughly spherical objects collide softly and stick, they would cause a bilobed shape.

            I'd love to believe all of these theories. It's not as if any of these are positively suggested by models. Perhaps sometimes if you put the right volatiles in the right place, or assume that two bodies are going to collide softly at some point, then sure. There is hardly a ringing endorsement to any of them despite really good science being put to work on the theory of spinning bodies. I would definitely lean to where physical evidence can tell us *what happened* rather than *what the models tell us may happen*

          • Gerald says:

            As far as I see, the first attempts to find clear evidence for the contact binary version failed.
            One has been a clear compositional heterogenity signal discerning the two lobes; this doesn't seem to be the case. The other has been a tomography by CONSERT data; that's incomplete thus far.
            If Phiae awakes from hybernation CONSERT may be completed. So there is some hope to resolve the contact binary question.

            The evaporation version could be resolved by comparing pre- and post-perihelion shape models to estimate the mass loss / shape change per orbit to extrapolate the mass loss back in time.

            So the observational basis for both options is thin, but has good chances to improve during the months to come. Maybethe sublimation version has some advantage, since we could consider the observed sublimation rate and extrapolate; but more reliable is waiting for shape data well after perihelion.

            Both models have been shown to be physically possible.
            The stretch scenario is lacking an according model. To be able to compete, this gap needs to be closed.

            What actually happened in the past with the comet can only be inferred by modelling, not observed, unfortunaltely.

          • Marco says:

            Hi Gerald, you say "Both models have been shown to be physically possible."

            Let's start with contact binary: the problem is not whether a bilobed structure can be formed by two spheroid objects impacting at slow velocity. The problem is that in the Kuiper belt and other regions comets are and have been for billions of years, low speed collisions are virtually impossible given the huge empty spaces and high relative velocity of comets.

            Secondly with erosion/ablation. The question is not whether erosion can explain the shape given that the areas around the neck could have higher percentage of more volatile species and thus erode quicker. The question is how the distribution of material could possibly have come about to be that way in the first place.

            Having models that are physically possible with required starting features that are not likely to be possible is at the very least a naive way to decide on working model possibilities.

          • Marco says:

            Hi Gerald, you say "What actually happened in the past with the comet can only be inferred by modelling, not observed, unfortunately."

            That theory probably explains why none of the features of the comet are looked at as "evidence" , but just as ornaments with unknowable history or reason for their shape or features.

          • Gerald says:

            Comets are only fast in the inner solar system. Further out they are rather slow. In the outer Oort cloud near aphelion just a few hundreds of meters per second, or less.
            (use the formulas here: http://en.wikipedia.org/wiki/Orbital_speed to verify, the orbital velocity is inversely proportional to the orbital radius for circular orbits, about 30 km/s for Earth, http://en.wikipedia.org/wiki/Earth, hence 300 m/s at 10,000 AU, less than 100 m/s at 100,000 AU)
            Binaries are not that rare, see e.g. a list of asteroids with moons:
            Actually comets formed from smaller pebbles.

            The composition is by densifying cosmic dust in the protoplanetary disk:

          • Gerald says:

            Hi Marco,

            the crater-like structures are presumed to be kind of sink holes due to loss of volatiles. but that's still under ongoing investigation. I'm sure we'll know more later in the Rosetta mission.

          • A.Cooper says:


            You mention hundreds of metres per second for far-flung orbits. The following is an extract from a blog post of mine a couple of weeks back. It uses Sedna's aphelion velocity to make some calculations on the likelihood of the two bodies meeting up in deep space and managing to stay together. Sedna is used because it's aphelion of 936 AU is a long way past the generally accepted radius of the Scattered Disc (which is where JFC's are thought to come from, not the Oort cloud). This conservative value gives the lowest possible orbital speed for the two bodies that supposedly coalesced.

            "Furthermore, they [the two bodies that supposedly joined] would have to approach each other at less than 1m/sec while both orbiting at speeds of anywhere between 400m/sec (per Sedna’s aphelion velocity at 936 AU) and 6km/sec (Pluto’s perihelion velocity inside Neptune’s orbit). This requires an absolute velocity match to within 0.25% at the very least and an approach angle of no more than sin-1(1/400). That would be the angle whose sin is 1/400, which is 0.15 degrees. This slimmest of chances would in fact be their best chance to come together and would apply only at 900 AU, for a few hundred years every few thousand years. In other words, both bodies would have to be, essentially, in identical orbits to have any chance of coalescing.

            If 67P really is a contact binary, then this virtually impossible close approach dynamic for the two lobes had to satisfied or they would have simply sailed right on past each other."

            I failed to mention in that post that whilst the speeds are most favourable at aphelion (though still very high for coalescing) the potential for meeting up is greatly reduced by the decrease in density of the whole Scattered Disc in that region. At perihelion all Scattered disc Objects pass moderately near to Neptune's orbit so they bunch up in comparative terms, but at aphelion they can be anywhere from fifty to hundreds of AU out and also extend up to 100AU above or below the ecliptic. This means the density reduces by several orders of magnitude just at the point where the aphelion velocities might give a vanishingly small chance of two such bodies coalescing.

            There's a very big clue that the above analysis points to. The two bodies have to have an identical absolute speed and approach at no more than 0.15° (or possess up to 1m/sec speed difference, maximum, and approach at an even smaller angle). This not only means they have to be in essentially the same orbit but also be at exactly the same point in that orbit. The best fit for this scenario is two bodies that were always flying along in the same place in the same orbit i.e. two bodies that were stretched apart from what was once a single body.

          • Marco says:

            Hi Gerald,
            The "collisional problem", that is, having two similarly sized bodies collide at not hundreds of metres per second but just a handful of metres per second relative velocity and sticking, is not resolved by moving the problem back in time and space to the outer Oort Cloud.
            There being plenty of binaries does not lend itself to the probability or otherwise of having a soft, sticky collision. Falsifiability is certainly problematic, as it is easy to point to early or outer conditions being favourable, without being able to even roughly model or judge whether that improves the probabilities. At least stretch theory, being in the here and now, as something that would have happened recently, and may continue to happen, information may be able to verify or falsify aspects within the next year of the Rosetta mission. If Doppler data or similar shows stretch in the range of even half a metre or so, there is little reason to use models to demonstrate plausibility. The data would have already done that for us.

      • Marco says:

        A very plausible fractionation process on an initially homogenous, pristine icy body would result in a body with a hard, brittle outer shell, and a soft gooey centre. A view from a chemist:


        • Gerald says:

          Well, Marco, this model is many parts rather close to the assumptions the Rosetta mission has been based on. Partiularly a crust of unknown thickness, but estimated up to several decimeters has been assumed. The harpoons have been designed to penetrate this crust. Below this crust a soft porous ice/dust mix has been anticipated.
          One assumption of the referenced blog probably needs an adjustment: The core temperature has probably never been above 25 K. So liquids might form temporarily near the surface, but that's not (yet) evidenced.
          See this LPSC2015 paper:
          The yet unresolved part of "stretch theory" is the spin-up without disruption, as expected for the brittle model, and without the formation of a spheroid/ring system/binary/n-body system as excessively investigated since many decades for the liquid (or rubble pile) model.

          • Marco says:

            The referenced paper is very solid, but it says nothing about deeper subsurface temperatures below the crust.

            "In this study, measurements of coma volatiles are employed to try to infer shallow subsurface nucleus temperatures of the poorly illuminated, southern winter hemisphere of comet 67P."

            The crust on 67P appears to be very close to 50 metres on both the head and body lobes, judging by the obvious "lip" on the head lobe and many other parts of the surfaces. This very crust (hard and brittle) which you agree exists, and the "soft porous" mantle and core that is assumed together fits neither the "brittle model" nor the "rubble pile model" and will not behave in any of the ways all these models can show. Show me a model with an outer brittle crust and rubble pile interior spin up, and then we might see what comets actually do.

          • Gerald says:

            Marco, that's the point of your opportunity to earn merits by providing an appropriate model consistently describing the observations.
            You are now aware of existing models. I'm too busy with other projects to fill in the gap of conclusively proving/disproving your approach.

            My educated guess is, that the stretch interpretation is very unlikely, although not quite impossible.
            This may become relevant, if the contact binary or the sublimation/erosion scenario are tried to be shown as definitively correct.
            In this case the stretch theory may need to be shown as exceedingly unlikely.
            For this purpuse it would maybe useful to write a two- or four page article summarizing the key ideas of the stretch scenario in a way consistent with physics. Don't try to claim anything to be applicable to 67P, unless it's really waterproof (not just in your opinion). Just describe your scenario, and show the gap in previously investigated scenarios. List the papers you can find about scenarios already investigated, to show that you've read the existing work of others.
            Show the non-applicability of published models to certain scenarios.
            Read similar papers; try to write the same style and rigor.
            If possible, give it to a scientifically educated friend who disagrees with you about your interpretation, to check the concept for weak points.
            Try to submit it e.g. to arXiv.
            That way you provide an opportunity to be referenced.

          • Gerald says:

            I think the comet fits the "brittle" model (in a microgravity range), since at very low temperatures water ice is hard and brittle as rock. Abundance of all possibly ductile or liquid species at low temperatures taken together is probably too low to make the nucleus ductile.

            There is one argument for a not perfectly brittle scenario: I'm not aware of an instability of the rotation axis of the nucleus. Brittle objects are more likely to "tumble" than those following partly a "liquid" model.

            About core temperature: Internal heat sources (mainly radioactive decay heat, today mainly of K-40) may have been relevant for small bodies early in the history of the solar system. But by now for a 4km diameter object I doubt it, but possible for some models. It depends much on the thermal conductivity, hence the small-scale porosity of the interior.

      • A.Cooper says:


        You may wonder why my Sedna value for its 936 AU aphelion speed is circa 400m/sec whist your 10,000 AU speed is only 100m/sec less despite being 10 times furthers out. This is because although there is indeed a reciprocal relationship to r, that exact relationship is to the root of 1/r. If you compare your Earth speed value with the other planets' average speeds and use this relationship, it will fit. The very small anomalies you find are due to their small eccentricities skewing the average speed from that of a perfect circle of that radius.

        Concerning binaries being common, asteroids are much more susceptible to collisions than TNOs, although doubtless some TNO's shed material this way. However, the chances of two similarly sized chunks remaining is less than that for small fragments in relation to the primary. All the known TNO's with moons are relatively massive bodies with escape velocities from 30-500 metres per second. This makes it much easier for them to capture passing bodies than for 67P. I realise their large size is skewed towards what we can see with current telescopes but I would wager a much smaller number of the smaller TNO's have moons because of their inability to hold on to passing bodies (i.e. needing ultra slow approach at extremely fine orbital path angles). They would only start shedding fragments if they won the JFC lottery and got close enough to the sun for spin-up, either via asymmetrical outgassing or YORP (or a Roche pass at Jupiter).

        There's only one way I can think of that a body as small as 67P could have gained a moon when out in the Scattered Disc. That would be via a relatively high velocity impact (and so not needing the speed and angle constraints) from a much smaller body. Most of the ejected material would escape but, by definition, some fragments would leave the surface at a range of speeds which are between the lowest possible orbital speed and escape speed. All these fragments will orbit the primary.

        However, a high speed collision where the impacting body was the head lobe itself, which then was somehow supposed to end up orbiting, gives rise to a problem. Such a collision would have enough energy to obliterate the two bodies before that sedate choreography could play itself out. So we're back to the two supposedly disparate bodies drifting together less than 1m/sec and not impacting at speed. That is the only scenario for a coalescing contact binary that is posited by the teams because they can see detailed structural morphology that hasn't been obliterated.

        Asteroids are also thought to shed material due to YORP spin-up. A paper came out a week ago that added greatly to the evidence for YORP spin-up to fragmentation for asteroids:


        Seeing as the asteroid belt is much more susceptible to both YORP and collisions, there will be quite a preponderance of asteroids with moons so I agree that binaries are not uncommon. The problem for 67P is that if it's a contact binary, it is posited that it happened when it was in the Scattered Disc. And due to the arguments I've outlined above, I think that for such a small object with such a small escape velocity, this is a highly unlikely scenario, virtually a zero chance.

        As for pebbles, other parts in my blog cite papers on pebbles, one paper specifically related to 67P. All are cited in the course of making a case for 67P having a malleable core that can stretch.

        • Gerald says:

          I'm not really a fan of the contact binary option. But nevertheless to advocate for that possibility, consider the number of objects involved.
          For a given pair of objects the probability to meet appropriately within a short period of time is exceedingly low.
          But now consider a trllion objects over a billion years as second component.
          "The outer Oort cloud may have trillions of objects larger than 1 km"
          A randomly occupied phase space of orbital parameters, say semimajor axis, and two angles.
          Considerig just one angle with 0.1° difference would still leave 1e12/3600 = 2.8e8 objects matching in terms of angle. This remaining number needs now to be distributed over the remaining orbital parameters. Orbital phase probability can be set close to 1 over a billion years in the non-resonance scenario.
          If I follow the hypothesis "The most widely accepted hypothesis is that the Oort cloud's objects initially coalesced much closer to the Sun" the cloud must have been denser than today. So the current density is a lower bound of the actual density; there has also taken place a loss of comets due to ejection and destruction in the inner solar system.

          Now distribute 2.8e8 objects over a 100,000 AU aphelion parameter range, such that the orbits don't intersect. Linearly spaced, that's 2,800 objects per AU, or 1 object each 50,000 km.
          But attention, the orbits are not allowed to intersect anywhere on the up to 100,000 AU.
          Two angles remain, you can play with, but not unconstrained, it's essentially one degree of freedom due to the 1-dimensionality of the orbits.

          This one considered object may skip through the meshes, but you have one trillion objects to consider, the famous birthday problem: http://en.wikipedia.org/wiki/Birthday_problem
          Now it's getting really crowded in the parameter space, such that you will get a significant ratio of soft collisions.

          The collision probability is higher for non-uniformly distributed objects.

          The smaller the objects the higher the probability of "soft" collisions. It may turn out that this mechanism is actually underlying the formation of comets and other small bodies in the solar system(s).
          Since the collision probability dereases with lumping of a given mass, this "soft accretion process" slows down at some equilibrium point in terms of object size, the size distribution we are now observing.

          • Gerald says:

            Questioning the slow collisions would raise the question, how comets could form at all.

          • Marco says:

            I'm sorry, but I don't buy it. The whole point is that when they are forming, they are gravitationally bound pebble piles, and thus if they collided, softly or not, they would just form a bigger pebble pile at that point. It is when planetesimals become more structurally rigid somewhere down the line in their evolution is when they need to collide softly to form this shape. In fact, the exact issues with mechanisms for stretch having the shape are mirrored in the precise statistical requirements for contact binaries. Ie. If they are "soft, ductile" pebble piles colliding, they would merge into a larger spheroid. If they are hard and brittle, they would either bounce off each other or shatter. They need damping to absorb the impact, and rigidity to maintain the shape. What sort of material is that?
            Moving on to erosion mechanisms. It too is full of "premechanism" problems. How do the different materials get sorted internally such that their differential volatility makes it the shape it is? Pre-erosion, it must have that shape internally of the less volatile substances.

          • Gerald says:

            Sublimation occurs at the areas of maximum solar illumination near perihelion, either near a pole, or near the equator, depending on the axial tilt of the comet.
            The axial tilt may shift due to mass loss.
            The nucleus may start close to homogenious on the larger scale, with ellipsoid/spheroid or "potato" shape.
            Best to extrapolate back by analysing current mass loss of 67P during / near perihelion.

          • Marco says:

            Or we could extrapolate the equinox stretch if any. Let's not jump to conclusions. Let's hope there are accurate length measurements, neck width measurements and good 3D models of surface features that may or may not erode.

          • logan says:

            As a primigenial dust condensate grow in size, also grows in fragility. A primigenial dust condensate never leaves the nursery.

            It is progressively 'rounded and compacted' as the pet's fur in an um-vacuumed house.

            Soft impacts. Impacts dictate if the condensate persists, compact to a new, stronger generation or is destroyed. This apply all the way to 67P.

            It is very important to account for ice as condensation/sublimation surface. There are currents within the cloud.

            A current going from a warm corner toward a cold one could be surface-trapping a lot of molecules, a lot more than sublimating.

            A current going opposite way could be sublimating a lot more than condensating. [Breathing].

            Low generation cometesimals could be formerly very low on dust. Extraordinary universe symmetry is that Just as dust is very good at trapping gas, once fluffy ice start to form around it, the later become very good at trapping dust.

            Next is about cosmology: There is an upper limit to mass. Upon that the comet is no longer gravitationally bounded to her nursery. Nursery dictates the average size of its comets. A voyager is born.

            Solar systems are another story 😉

            [This is cometary fiction].

        • Gerald says:

          The proportionality of orbital velocity to the inverse square root of the distance from the sun is clear.

        • Gerald says:

          The contact binary forming after break-up is an option I could imagine. Although I wouldn't call it a (bilobic) stretch. It's either due to collision, due to YORP spin-up (brittle disruption or liquid-model ring system / n-body decomposition), by asymmetric sublimation spin-up, by close encounter within the Roche limit, by gradual sublimation/erosion, or by explosive disruption driven by sublimation.

  • Sovereign Slave says:

    "It is a capital mistake to theorize before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts."
    Arthur Conan Doyle (1891) A Scandal in Bohemia

    Have quoted this before, and still, can't help but think that as long as those investigating have to find ice, all data will be interpreted in that way, to the exclusion of open minded discovery.

    • logan says:

      Error is not to theorize with so few data. Preliminary theories are vital. Actions have to be taken in the present, about future scenarios. Error is 'to fall in love' with particular theories. A theory is just a model. And models we can build, as many as we can handle. We start to throw them away only when they grow to big and complicated as to maintain them updated in view of new, conflicting data.

    • logan says:

      In the long term, all theories all preliminary.

  • Phil Stooke says:

    Wow, that is really spectacular. Thanks!

  • Gerald says:

    Very accurate channel match, really great work of the OSIRIS team!
    Looks easy in the release, but it's rather challenging to get there.

    Looking forward to the confirmation of the suspicion of the OSIRIS team by VIRTIS, and to subtle refinements.

  • Graham Hall says:

    “We are excited to see whether our suspicion will be confirmed,” says Sierks

    So, 6 months on from when the data was acquired, you don't know, Holger, what VIRTIS will come up with? Well now let me predict - they will confirm your hypothesis!

    This is a bit like a political party announcing the same policy time and time again as being new.

  • Robin Sherman says:

    Thank you Holger and team, no easy task to create such a wonderful image. 67P seems to have a "Dust Cycle" much like the Earth's Water Cycle, at least in the Northern "Hemiduck" anyway. Sublimating volatiles carry the dust up into the coma, cool and crystallise around the tiny dust grains, increase in mass and fall back to the surface. The growing "halo" around the comet shows most of the gas and dust is leaving the surface at very low velocity, a large amount of it must return to the surface during the hours of darkness.

    Even though the lighting accentuates the blue colour towards the bottom of the image, this is the area of the comet's North Pole and as such is likely to act as a big cold trap for any volatiles remaining close to the comet's surface during darkness. In essence a stock pile of volatiles builds up over the Northern Winter, around Perihelion, very close to the surface mixed in with the dust fallout from the coma. One would think that the whole Hapi Valley in the higher Northern latitudes would act as a cold trap in this way. Certainly volatiles that much closer to the surface would enable earlier sublimation and the earliest activity and those volatiles actually condensed onto the top of the surface could explain the burst of activity in April/May 2014. I commented recently that the activity in the neck region appears to have diminished, maybe most of those concentrations of easily sublimated volatiles have now gone.

    I notice that the dust free, dark, exposed comet material of the head lobe cliffs around this area also has a blue tinge as if water ice has condensed onto the colder, darker, solid material. This might be created more by the false colour making process though. The Amphitheatre is a less orange colour than higher up the side of the valley and there do seem to be greyer, colder traps in some of the flat circular features. This is possibly how they collect more dust, the dust being coated in ices as they condense from the coma as a form of "snow".

    Again there are clear signs of some sort of fluid or semi-fluid having flowed, well oozed, along the valley. There are two very clearly defined "frozen lava flow" like formations. At their head, the clear profile of a very large surface deformation in the form of a massive subsurface pressure build up. The layers of the comet can still be seen as part of a big arch in the centre of the image. The explosive collapse of such a huge pressure bubble could well be the explanation for the huge gash created in the comet that precipitated the formation of the comet's neck.

    • THOMAS says:

      " One would think that the whole Hapi Valley in the higher Northern latitudes would act as a cold trap in this way."

      Robin, if Hapi Valley is indeed the "cold trap" which mere thermal-energy-based logic requires, than why does the VIRTIS temperature data prove it to be by far the warmest region on the whole comet surface? See https://www.flickr.com/photos/130179313@N03/16524318826/ and https://www.flickr.com/photos/130179313@N03/15927644524/in/photostream/

    • logan says:

      Adding your "Dust Cycle" for suborbital speed particles into my hat. Robin.

      As commented by Harvey, gas is in molecular flow -free path- and don't cool down until impacting. So not so convinced of dust actually trailing back ice to surface.

      On favor of your argument will say that MIDAS showed and COSIMA suggested this fluffy dust could easily act as as a gas 'trap'.

      Just speculatively going against, estimate actual ballistic dust is sublimating a lot more than condensating when in day -gas surrounded- mode. At night, dust in ballistic mode has little gas to trap, as most of it is already 'lost in space'.

      Endorsing your 'dust trailing ice' in formative and stable comet phases. Really needed to 'layering' models.

      That fluffy dust has to be very resilient [ethernal?] to flows and whippings of energy present at space.

      • Harvey says:

        Logan, it is Im afraid rather complicated.
        Whether gas flow is effectively molecular or viscous depends on both the pressure and the scale size of the object. One has to be very careful with concepts of 'temperature', which is an equilibrium concept requiring collisions to establish it; how many depends on whether we are talking translational & rotational temperature (few) or vibrational temperature (many typically.) There may well not be a single 'temperature'; you can need several different temperatures to describe one molecular species.

        In a very simplified, crude way, the molecules diffuse, move randomly, once you are in molecular flow. In the near comet environment, as the pressure rises due to outgassing, there may well be some viscous flow effects at larger scales; further out it will become molecular.
        Individual gas molecules typically do not easily 'stick to each other' very easily in these circumtances. But they do readily stick to solid surfaces. However the situation is dynamic; the cold surface 'traps' some molecules which hit it - and releases some which evaporate, depending on the temperature, saturated vapour pressure of the material, & local pressure of that material (partial, not total pressure.) Its a dynamic equilibrium of 'pumping' and 'outgassing'.
        There is just no way to simplify this stuff, it is complicated.
        A serious problem here is some writers who, bluntly, clearly know very little real physics making authoritative sounding statements which are simply wrong, or so oversimplified as to be totally missleading.

        • logan says:

          Accepting your comment about actual complexity, Harvey. The 3 different temperatures which bring along quite different behavior to surface-gas interactions. Understand that even 'vibrationally' hot particle's surfaces can 'lock' a mono-layer or bi-layer of gas molecules. [All that learned from your comments 🙂 ]. Understand too that bigger particles are more on free path than small ones. Those 1-5 micrometer particles falling back near jet sources should be trailing back ices as Robin model says.

          Nowadays, a preliminary model is needed to leverage other phenomena models. Hope any soul dare to propose one.

          Also in favor of Robin's argument, remember a comment from COSIMA Team about some particles apparently 'jumping' when already seated on the plates.

    • logan says:

      Could any soul dare to speculate the molecular 'porosity' of this MIDAS photo?


      Just being joyfully irreverent.

      • Gerald says:

        0.81249382 (+0.151337 / - 0.427108) with a 2 sigma confidence.

        But: Starting with an assumed mean porosity of 0.75, and an assumed 0.75 dust by volume we get a porosity of 0.8125 for the volatile-freed comet material.

        • logan says:

          Still a strong comet. Why is so unresponsive to EM solar energy? Born as a water comet? 🙂

  • logan says:

    Bluer zone next to the tide-grinder-pouring zone.

    Apparently was wrong referring Hapi as the dryer geography of Coraline.

  • logan says:

    OSIRIS brings again a 'puff pastry' visual allegory to memory. Flaking of surface is a lot more evident in the high resolution provided. Notable the strong green at body. Could that be related to active 'calderas' ? 😉 Curious thing about 67P is that erosion in general is manifested as mass loss, collapsing and migration, in that order. More confident now that some dunes are related to tangential jets.

  • logan says:

    Hapi region offer some clues about how dust turn again into layers and develop 'ducting'. Pixels 720,1001 and 761,951 of


  • originalJohn says:

    This is all hopeful speculation Emily, although I am sure you are simply the messenger. Do your colleagues not have the VIRTIS result by now, six months later, the unambiguous spectral signature. They have to be careful though, because if water was being generated at these spots then it would not be unambiguous. And you could suggest to them that it wouldn't be a bad idea to measure the temperature of the spots, and perhaps whilst they are at it the ion concentration in the vicinity.

  • Dave says:

    I am not sure how this theory lines up with the data that shows the neck region to have a hotter temperature running right Along it.
    Ignoring the fact that the neck is in shade a lot of the time, so we might not of expected this area to be hotter,
    It seems at odds with the theory of ice being exposed in the neck, as the hotter area might be expected to drive any possible sublimation, such that this should not be an area where surface ice would be expected.
    Any one have any views to reconcile the data?

    • Gerald says:

      A series of VIRTIS images showing the temperature changing over time of day, and with solar angle would certainly simplify the interpretation.
      In the meanwhile you may try it with your imagination.
      For simplicity take the Sun light as the only heat source; neglect solar wind and other minor contributions; its more than 99.9% heat by solar illumination.
      Try to correlate high temperature on the VIRTIS images with the Sun near the zenith for those locations as a first approximation. To get a better fit, correlate dust cover (low thermal inertia) with high temperature for most of the remaining discrepancy to the first approximation; black dust is readily warmed by the Sun.

      Preparing a proper visual 3d simulation intuitively showing these correlations is beyond my time budget. Maybe someone else finds time to prepare it.

    • Gerald says:

      One (not necessarily the only) straightforward interpretation of the neck is faster sublimation than elsewhere. Once exposed near the surface the ice sublimates away and deepens the neck. The saddle structure of the neck results in a surface area perpendicuar to the most intense solar illumination.
      Additional black dust accumuated at the gravitationally deepest zone may increase heating further.
      From the water phase diagram, you may notice, that the partial pressure of water vapor increases about exponentially in the temperature range between 200 K and 270 K. So it's crucial, whether the peak temperature is 200 K or 220 K. Vertical solar illumination and a low thermal inertia black dust cover make the difference for water ice to subimate substantially.


    • Gerald says:

      "Water production varies both with
      location on the nucleus and time-of-day, with maximum
      outgassing rates associated with the "neck" region
      of the nucleus being illuminated by the Sun. Gas
      column abundances drop by an order of magnitude as
      our line-of-sight moves towards un-illuminated regions"


    • logan says:

      Hi Dave. Only occurs to me this is a 2014ago situation. Wanting to point the detail that blue is just a very small portion of neck surface. Maybe University Teams should give this little zone another name. Curiously, that's the one still active. Other jetting zones have become dominant. Betting on neck's activity being sort of the same as then [just a little more]. Lower contrast photos have less to show.

    • Ramcomet says:

      Grinding, deep down between the two lobes. Mortar and Pestle-like grinding creating heat from mechanical friction.

  • ianw16 says:

    Ice, you say? Not rock? But we have people on here all the time telling us there is no ice, and that the jets are due to electrical thingy whatsits. Who'd have thought it 😉

  • mario filho says:

    Please, can you check the image Hapi_and_Hathor cause i just can see something strange there, like a image of a being climbing the mountain. I guess it is not paredolie. Thanks.

  • Dave says:

    Thanks Graham that's an awful lot of info.
    But we know the neck is hotter all the way along, we don't know the temperature of this area because they perspire seems to have saturated the instruments.
    So I still find it difficult to determine a sensible mechanism got the possibility of ice in this region

    • Gerald says:

      The instruments aren't saturated that easily. The temperature has been up to 220K = -53 centigrades. They adjusted the scale for the graphics to cover the measured temperature range. The instrument is insensitive to the very low temperatures at the night side of the comet.

      The comet is made of silicate dust, water ice, ices of carbon dioxide, several other volatiles, less volatile organic compounds, several other less abundant minerals.
      Water ice starts sublimating above about 200 K. Carbon dioxide at lower temperatures, carbon monoxide even much lower, closer to absolute zero at a few 10s of Kelvins.
      The ice has always been there due to the formation process of the comet.

      Going back the history:
      The origin of the matter the comet is made of is interstellar dust and gas; the (known) comets of our solar system formed a little more than 4.5 billion years ago together with our solar system.
      The temperature of the dust has been a little above 3 K, but low enough to freeze water ice, carbon dioxide, and most other volatiles. (A little above the current 2.7 K mean temperature of the observable universe, since the universe was younger and a little warmer than today.)
      The origin of interstellar dust is supernova explosions or otherwise ejected material, mostly from aging stars.
      Most of the hydrogen is primordial, some of it may have temporarily been part of stars, but remained unaltered.

      The supernovae are exploding massive stars at the end of their life.
      Before those supernova explosion the stars fuse hydrogen, and some helium, to other chemical elements like oxygen or silicon, like other stars do.
      Initially most of the hydrogen, and some helium formed shortly after the Big Bang, some 13.8 billion years ago.

      Which part needs more detailed elaboration?

      • Harvey says:

        Gerald, Im not sure about your deduction that the dust grains will be ~3K; you are assuming equilibrium with the cosmic background. But if there is a star even very distant from the dust grain, its (much shorter wavelength) radiation will readily exceed the cosmic background long wavelength power density surely, due to the T^4 in the Stefan-Boltzmann equation.
        Obviously this depends on the distance to, size of & temperature of the nearest star. But it seems to me this might ofteen lead to a significanty warmer dust grain.
        *HOWEVER*, I doubt it would be warm enough to alter your deductions & conclusions in any significant way, as the temperature is still going to be easily low enough to freeze out the volatiles.

        • Harvey says:

          I just tried to put some numbers in.
          Wiki quote the CMB energy density as 4*10^-14J/m^3, so the power density is 1.2*10^-5 W/m^2
          Now for the sun, we have a power density at 1AU of around 1.35kW/m^2
          So the solar power density would equal the CMB at around 10^4 AU
          Taking the supposed scale of the Oort cloud as a yardstick, its said to extend from a few thousand AU to 10^5 AU.
          So I think solar power density dominates in the inner part, but your CMB vallue is correct inn the outer part (which, of course, has far more volume.)

          • Gerald says:

            Thanks a lot Harvey for narrowing this down!

          • Gerald says:

            Refined estimate:

            For a low albedo object at 1 AU take a planetary equilibrium temperature of 273 K. Since the equilibrium temperature is proportional to the inverse of the square root of the distance to the Sun, 3 K planetary equilibrium temperature for this object are at (273K / 3K)² AU = 8281 AU.

            With 2.5 lightyears = 2.5 * 6.3 * 10^4 AU = 1.575 * 10^5 AU as a conservative estimate of the mean distance to the next star (most of them colder than the Sun) in the Milky Way galaxy, we get a ratio of 1.575 * 10^5 / 8281 = 19 by radius, and 6880 by volume for the space within the Milky Way galaxy, where warming by the CMB dominates over warming by stars.


          • Gerald says:

            ... There are of course galactic microwave sources to be taken into account when considering interstellar dust at temperatures near 3 K.
            That might be relavant for some super-volatiles, like nitrogen or hydrogen; I don't think, that hydrogen would freeze out at 3 K near vacuum. I know, that helium doesn't freeze out at 3 K near vacuum.
            The presence or absence of some of the super volatiles may tell, at which temperature the cometesimals formed.

        • logan says:

          A dense dust cloud should be 'vibrationally' warmed at surfaces close to near stars. But just within, averaging processes between the other two temperatures should be happening between dust particles. Guessing stars are needed to 'compact' by wind and photon pressure a dust cloud enough as to create a 'deft' zone. [As mist create 'deft' zones]. A 'deft' zone makes 'sticking' of dust particles plausible.

          • logan says:

            CO2 cometesimals are 'sticked' together far far away as you say.

          • logan says:

            ERRATA: Says "...to create a deft zone". Should say "...to create a deaf zone". A zone 'cushioned' from most of the energetic 'cacophony' of the galaxy.

          • Harvey says:

            Logan, the comments about differing temperatures really apply purely to molecular species, not dust particles. They will simply reach an equilibrium temperature dependent on the radiation field they are immersed in.

          • Gerald says:

            Cold interstellar molecular molecular clouds tend collapse gravitationally when reaching a critical density.
            This may lead to the formation of stars with protoplanetary disks, the birth place for planets and comets.
            The collapse of a molecular gas cloud may be triggered by a nearby supernova shock wave, but not necessarily.

          • logan says:

            Thank for the guide, Harvey and Gerald 🙂

          • logan says:

            A mind experiment: An errant, solitary star at an empty corner of galaxy goes supernova. How could this explosive dust stop eventually and start condensing?

          • logan says:

            While molecule-molecule collisions are highly elastic, particle-particle are not. Particles start to kinetically average, -collision after collision-. Start to 'cool down' relative to each other. Start to 'dance' along. Brownian Motion, gravity an other attractive forces overcome the disintegrating forces of the wild, open space. Here we could have our initial particle condensates,

            Energetically dampened nurseries are needed.

          • Gerald says:

            The thought experiment with a solitary star: That's not the scenario usually leading to protoplanetary disks; but collision (or radiation pressure, etc.) with older interstellar gas and dust.
            Supernovae occur after a rather short lifetime of a massive star. The star then may still be within or near the interstellar cloud where it formed.

            But just for fun: Before a massive star explodes as a (type-ii) supernova it undergoes expansion, and ejects parts of its outer shell with moderate velocity. The supernova explosion shock wave later collides with those older gas and dust remnants leading to locally densifed (and heated) remnants, although insufficient to form a new solar system.

          • logan says:

            Thanks Gerald 🙂 Then supernovas use to result from gigantic, unstable, low on metals [blue?] stars?

          • Gerald says:

            There are several types of supernovae. There is an overview in Wikipedia:

            Almost all O type stars and probably some B type stars will end up as supernovae.
            The type of supernova depends a bit on metallicity. But most important is mass.
            Type Ia supernovae work a little different, since they are thought to form from binaries, one component a white dwarf, which implodes as a supernova Ia, if it gets above the Chandrasekhar mass limit for white dwarfs by accreting mass from its giant companion.

          • Gerald says:

            I should clarify this a bit: spectral type O and B stars are hot, most of them massive, but surface temperature doesn't striclty imply high mass, it's just mostly.
            More about spectral types:

      • originalJohn says:

        All of what you say Gerald is consensus theory, without any verification from observation. This includes the composition of comets, the history of the solar system, the source and life cycle of stars, the concept of the big bang and the primordial formation of elements. It is of no value in describing and understanding what is going on with a comet before our eyes.

    • Gerald says:

      Regarding the temperature determined by VIRTIS, you may like to read this LPSC2015 paper:

      • originalJohn says:

        Thanks Gerald.

        "Comet 67P is shown to be everywhere
        rich in organic materials with little to no water ice visible on the surface"

        The results in this VIRTIS paper and subsequent results would normally instigate an automatic contingency to examine the neck area in greater and more highly resolved detail, particularly the jets temperature profile and the temperature of the previously discussed "blue" spots. Why would this not be done in blaze of publicity as it would provide answers to crucial questions and end speculation in the absence of data. Then all the discussion about ice or rock and sublimation or surface chemical reaction would be ended. It would also have major implications for the subsequent programme of investigation.

        • Gerald says:

          They did a close flyby, but they risked Rosetta to go in safe mode. The dust is too dense to identify stars; that's needed for navigation.
          Everyone wants to know these details; probably rather more is already collected, just not yet prepared for pubishing. But even closer exploration isn't possible with reasonable risk for Rosetta at the current activity level .

      • Harvey says:

        The saturated VP of hydrogen is given as around 10^-11 Torr at 3K rising rapidly to 10^-7 Torr at 4K. (sorry, old fashioned units.) so the exact temperature is very important.
        (You can't cryopump hydrogen in a UHV system.)
        There is a complication I've not looked into regarding ortho v para hydrogen. I'm not sure if the SVP depends on this, (if so I assume the value given is for equilibrium o/p) and whether the 'early hydrogen' can be assumed to be o/p in equilibrium?
        However that is the SVP; the pressure over anything that can bind to it could be much lower, tiny binding energies would be significant.

        • Harvey says:

          Ps. After He, which of course won't freeze out, and H2, with its o/p complications, the next most volatile gas is neon. The other 'volatiles' such a N2, CO, CH4 are frozen out to extremely low pressures at even 10K. (They are very efficiently cryopumped in UHV systems, whereas He and H2 are not.)

        • Gerald says:

          The density of cold molecular clouds can be up to 10^6 molecules per cm³; this corresponds to a pressure of about 2.8*10^-11 Torr (if my calculations are correct).

          During gravitational collapse, these clouds should warm up adiabatically; therefore I think odds are against freezing out hydrogen ice.

          From papers about cosmic dust as a catalyst for the formation of H2 from H I'm presuming, that molecular hydrogen is readily desorbed by cosmic dust. Formation of hydrogen ice on cosmic dust particles doesn't seem to be considered:

          • logan says:

            Think that H and He are a latter -gravitational- addition to big bodies.

          • harvey says:

            Your sum is correct.

            So I'd guess you may well be right, it only needs to warm up a few degrees K to prevent condensation of bulk hydrogen. Nucleation would be a big issue for sure.

            Just a little wary however that if we are taking monolayers, bilayers on dust, the VP could be significantly lower owing to surface binding effects; at such low temperatures very low binding energies would change things.

            Maybe the whole ortho/para thing is a red herring.
            It would come into play if the molecular H2 is generated at a higher temperature & then condensed out. I'm not sure the relevant conversion rates are even known, in that in the lab wall effects etc may dominate & catalyse conversion, even in the liquid phase.

          • logan says:

            Maybe suns start as planets 😉

  • Dave says:

    I Gerald you have gone too far now. The std birth of the comet and the birth of our solar system is so full of holes and myth that it's impossible to believe. Even a scant look at other solar systems tells us our evolution is wrong. Just look at Saturn an Jupiter in our own system they are way off the script. I prefer the discussions of the features of the comet and the different theories people have. Also it's pointless counting your chickens with any theory we might have before the data is here.
    Data from 6 months ago may look very different today.
    It will be a while before we can hear champaine corks as there does not seem to be one theory on the blog that can tick all the boxes. Good luck with yours though.

    • Gerald says:

      Details about the formation of Jupiter and Saturn may be a little off-topic when thinking about the origin of comets. But I can't duplicate the issue you see with Jupiter. It's just the second-largest body in our solar system after the Sun.
      After (or while) the Sun formed, most of the remaining material accumulated to form Jupiter. Above a certain mass, the escape velocity of a planet is high enough to prevent hydrogen (the most abundant element) from escaping, such a planet even attracts hydrogen. That way we get the gas giants. Where is the issue? That's standard planetology since decades, and hasn't been seriously questioned since then.

      Then we've Saturn and Uranus following a similar logic with most of the remaining matter in our planetary system; the smaller rocky bodies in the inner solar system were too small to accumulate giant atmospheres.
      Neptun is in-between. Due to the distance to the Sun it needs less gravitational attraction to keep it's atmosphere.

      There are several models describing the dynamics of our solar system, not quite undisputed, one family of which are the Nice models:
      So far the current standard.
      Now we are looking at the refinements by data provided by the Rosetta mission.

      • dave says:

        Indeed Gerald it looks a reasonable model as the std models go, but that does not mean its right.

        A recent Horizon programme had main stream cosmologists discussing the fact that the planets have changed orbits over time, sure they still stayed close to the std model with Jupiter hoovering up all the available material as it moved closer to the sun. Some think based on observations off other solar systems the gas giants should of been closer to the sun so maybe the fact that they may change orbits over time can settle that anomaly.
        However, its also an indication that the std model is not yet resolved, I guess this may or may not affect comet formation,
        I believe it also means that the faith and the dogma with which the std model is used may need to be taken with a little pinch of salt.

    • Gerald says:

      Regarding exoplanets: The transit method as well as the Doppler method are biased towards the detection of large planets close to their star The transit method, since perfect alignment Earth - exoplanet - star is more likely for planets close to their star, and ususally you need three transits for an evident discovery of an expolanet; due to limited observation time that's only possible for short-period planets. The Doppler method needs observation periods long enough to detect wobbles of the star; again that's only possibe for large planets close to the star.

  • Cometstalker says:

    The filters are infrared red and blue so the combination is not realistic to the normal human eye, just a nice way to show some contrast in a colorized scale. There ar other filters available that could have been used as well but the Osiris team seem to ignore this possibility. Once upon a time in this mission images from earth was made in a perfect colorization. Try again please.

  • Harvey says:

    If unfamiliar with ortho/para H2 start here:
    This remarks that there is very significant heat yield on conversion at 20K of 670kj/kg:
    Small vapour p differences are known at 'high' temperatures, tens of K, resulting from a small difference in latent heat:
    But be very careful with extrapolation, at lower temperatures where the curve becomes very steep the difference might be much bigger, needs plugging into the Clausius Clapeyron equation.

    Im not sure if this whole issue is significant or not, but it could be. If the H2 forms in equilibrium at higher temperatures, much more ortho, how, when, if the o/p ratio changes as it cools and condenses might matter; as might the nature of anything it sticks to (Fe oxides catalyse conversion.)

    • Gerald says:

      Thanks for the links to the summaries!

      In interstellar space, when thinking about possible solid hydrogen at low temperature (< 20K), the only relevant form of molecular hydrogen is probably the para-H2.
      Particularly, since molecular hydrogen uses to form at catalysts form atomic hydrogen.
      Although there may occur temperature fluctuations for small dust grains. But those will probably immediately evaporate traces of possible solid H2.

      • harvey says:

        If the H2 is actually formed at 20K, yes it probably forms as the equilibrium para form.
        If the H2 forms at higher temperature, & later condenses at 3K, then you get the ortho/para issues of possibly different vapour pressures & 'unexpected' significant heat input.
        In an extremely low collision rate environment, it could remain ortho potentially for a *very* long time, as its days in the liquid phase & even that may be wall catalysed.
        Buts its probably not important, just caught my interest; specific to H2 of course.
        (O/p versions of other molecules exist, but the thermal differences are minute.)

Comments are closed.