Inside Rosetta’s comet

This article is mirrored from the main ESA Web Portal.


Mosaic of images of Comet 67P/Churyumov-Gerasimenko taken by Rosetta’s navigation camera between August and November 2014. Credit: ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0

There are no large caverns inside Comet 67P/Churyumov-Gerasimenko. ESA’s Rosetta mission has made measurements that clearly demonstrate this, solving a long-standing mystery.

Comets are the icy remnants left over from the formation of the planets 4.6 billion years ago. A total of eight comets have now been visited by spacecraft and, thanks to these missions, we have built up a picture of the basic properties of these cosmic time capsules. While some questions have been answered, others have been raised.

Comets are known to be a mixture of dust and ice, and if fully compact, they would be heavier than water. However, previous measurements have shown that some of them have extremely low densities, much lower than that of water ice. The low density implies that comets must be highly porous.

But is the porosity because of huge empty caves in the comet’s interior or it is a more homogeneous low-density structure?

In a new study, published in this week’s issue of the journal Nature, a team led by Martin Pätzold, from Rheinische Institut für Umweltforschung an der Universität zu Köln, Germany, have shown that Comet 67P/Churyumov-Gerasimenko is also a low-density object, but they have also been able to rule out a cavernous interior.

This result is consistent with earlier results from Rosetta’s CONSERT radar experiment showing that the double-lobed comet’s ‘head’ is fairly homogenous on spatial scales of a few tens of metres.

The most reasonable explanation then is that the comet’s porosity must be an intrinsic property of dust particles mixed with the ice that make up the interior. In fact, earlier spacecraft measurements had shown that comet dust is typically not a compacted solid, but rather a ‘fluffy’ aggregate, giving the dust particles high porosity and low density, and Rosetta’s COSIMA and GIADA instruments have shown that the same kinds of dust grains are also found at 67P/Churyumov-Gerasimenko.

New Norcia ground station

ESA’s 35 m-diameter deep-space tracking station at New Norcia, Australia. Credit: ESA

Pätzold’s team made their discovery by using the Radio Science Experiment (RSI) to study the way the Rosetta orbiter is pulled by the gravity of the comet, which is generated by its mass.

The effect of the gravity on the movement of Rosetta is measured by changes in the frequency of the spacecraft’s signals when they are received at Earth. It is a manifestation of the Doppler effect, produced whenever there is movement between a source and an observer, and is the same effect that causes emergency vehicle sirens to change pitch as they pass by.

In this case, Rosetta was being pulled by the gravity of the comet, which changed the frequency of the radio link to Earth. ESA’s 35-metre antenna at the New Norcia ground station in Australia is used to communicate with Rosetta during routine operations. The variations in the signals it received were analysed to give a picture of the gravity field across the comet. Large internal caverns would have been noticeable by a tell-tale drop in acceleration.

ESA’s Rosetta mission is the first to perform this difficult measurement for a comet.

“Newton’s law of gravity tells us that the Rosetta spacecraft is basically pulled by everything,” says Martin Pätzold, the principal investigator of the RSI experiment.

“In practical terms, this means that we had to remove the influence of the Sun, all the planets – from giant Jupiter to the dwarf planets – as well as large asteroids in the inner asteroid belt, on Rosetta’s motion, to leave just the influence of the comet. Thankfully, these effects are well understood and this is a standard procedure nowadays for spacecraft operations.”

Next, the pressure of the solar radiation and the comet’s escaping gas tail has to be subtracted. Both of these ‘blow’ the spacecraft off course. In this case, Rosetta’s ROSINA instrument is extremely helpful as it measures the gas that is streaming past the spacecraft. This allowed Pätzold and his colleagues to calculate and remove those effects too.

Whatever motion is left is due to the comet’s mass. For Comet 67P/Churyumov-Gerasimenko, this gives a mass slightly less than 10 billion tonnes. Images from the OSIRIS camera have been used to develop models of the comet’s shape and these give the volume as around 18.7 km3, meaning that the density is 533 kg/m3.

Extracting the details of the interior was only possible through a piece of cosmic good luck.

Given the lack of knowledge of the comet’s activity, a cautious approach trajectory had been designed to ensure the spacecraft’s safety. Even in the best scenario, this would bring Rosetta no closer than 10 km.

Unfortunately, prior to 2014, the RSI team predicted that they needed to get closer than 10 km to measure the internal distribution of the comet. This was based on ground-based observations that suggested the comet was round in shape. At 10 km and above, only the total mass would be measurable.

Then the comet’s strange shape was revealed as Rosetta drew nearer. Luckily for RSI, the double-lobed structure meant that the differences in the gravity field would be much more pronounced, and therefore easier to measure from far away.

“We were already seeing variations in the gravity field from 30 km away,” says Pätzold.

When Rosetta did achieve a 10 km orbit, RSI was able to gather detailed measurements. This is what has given them such high confidence in their results, and it could get even better.

In September, Rosetta will be guided to a controlled impact on the surface of the comet. The manoeuvre will provide a unique challenge for the flight dynamics specialists at ESA’s European Space Operations Centre (ESOC) in Darmstadt, Germany. As Rosetta gets nearer and nearer the complex gravity field of the comet will make navigating harder and harder. But for RSI, its measurements will increase in precision. This could allow the team to check for caverns just a few hundred metres across.

“A homogeneous nucleus for Comet 67P/ Churyumov–Gerasimenko from its gravity field,” by M. Pätzold et al. is published in the journal Nature.



  • A. Cooper says:


    I’m a bit confused. The article says there are no large caverns inside 67P but the last sentence says:

    “But for RSI, its measurements will increase in precision. This could allow the team to check for caverns just a few hundred metres across.”

    That implies that the large caverns so far ruled out would have to be bigger than a few hundred metres across to have been detected and there could still be caverns of less than a few hundred metres. I’d call a few hundred metres a huge cavern. But I suppose it’s all about definitions of big and small holes in a porous nearly zero gravity matrix.

    The abstract also says it’s highly homogeneous and so it seems it too is ruling out caverns of several hundred metres by implication.

    Seeing as I think it stretched to its current shape and the body stretched to its current diamond shape before the head sheared, I would expect there to be little scope for large caverns too.

    I was also wondering if the paper on the head lobe being less dense than the body was going to be coming out soon. It looks as though the resolution from RSI would allow that to be established. I think they said about 10% less dense from memory. And I wonder whether the definition of the head includes the neck. It would be interesting to know what the density of the neck alone is. I should think it’s noticeably less than both head and body.

    • ianw16 says:

      @A Cooper,

      The only thing I can see in the letter that might relate to any remaining uncertainty is the following:

      “With knowledge of the bulk parameters only, it is still not possible to distinguish clearly between micro- and macro-porosity, that is, between the inherent porosity of fluffy dust particles embedded in the ice matrix or the porosity that would result from large voids within the nucleus body originating from the formation of the nucleus. It has been argued that the micro-porosity of the dust material already accounts for most of the nucleus porosity.” ^16

      16 refers to a paper: Greenberg, J. M. From comets to meteors. Earth Moon Planets 82, 313–324 (2000).

      • A. Cooper says:


        Thanks for delving in! Seeing as you’re good at finding relevant papers (same applies to Gerald), have you seen any references to the neck as a separate entity with regard to dimensions, mass, density or porosity? I’ve only seen the neck differentiated for things like insolation, temperature and sublimation rates. I have seen only one paper that dealt with anything related to structure in the neck specifically and that was Thomas et al proposing a tensile strength of 10-20 pascals as opposed to 40 pascals for overhangs on the body lobe. I expect the neck to be rather different in structure from the lobes and even within itself, with Hathor area being of a different density and/or porosity than both Anuket and Bastet.

        Even the ESA dimensions page has only the dimensions of the lobes and not the neck. I had to work it out by a convoluted subtraction method using known dimensions as a yardstick. I wonder if it has something to do with the centre of gravity not being pinned down and not being sure whether it is indeed that or the fact that the neck might be stretching or contracting. There’s a chicken and egg situation there and I know they were having some difficulty with the c of g location. The RSI results would suggest that it really has been pinned down- but not if the resolution/uncertainty is still for caverns of a few hundred metres. Such a cavern would be about 1% of the head lobe mass and mimic the c of g shifting by a few metres as Rosetta orbited. I seem to remember the discrepancy being cited as “a few metres” still in one of the October 2015 papers but don’t quote me on that.

        • ianw16 says:

          Sorry AC, only just seen your post.
          I can’t recall anything that specifically refers to the porosity or density of the neck region. Mind you I’ve got a lot of papers from this mission (~ 70 at last count), not to mention a shed load from other missions and observations.
          The only things that immediately spring to mind regarding the neck are:
          1) The bluer spectral slope, meaning there is certainly more ice in that region close to the surface.
          2) The fracturing. I don’t have a dog in the stretch/ contact/ evolution fight, but have recently read material that suggests the higher fracturing seen in the neck region could be due to the higher thermal stress experienced there, due to shadowing effects (no link; will keep looking)
          3) I’ve also seen it hypothesised that it is due to rotational and/ or orbital stress, which is mentioned in :
          4) Another claim, which I can’t find a link to, is that the neck could be evolutionary, due to a pre-existing dip (maybe a small impact crater) and the shadowing effects then cause the pit/ crater to evolve to the point we see today (again, related to the thermal stress of shadowing).

          Doesn’t really answer your questions, but I shall keep my eyes open.
          Something that does seem to be coming through from what I have read, is that the fracturing and layering seen on the comet as a whole may well be evolutionary rather than primordial.
          I suspect a lot of models are going to be run in the years following this mission, and that the reason for the shape of the comet is far from a settled matter. Like Harvey, I’m not totally sold on the contact hypothesis, purely on intuition.

          • A. Cooper says:


            Thanks for that. I read the paper you linked. It chimes with Marco’s thinking on BLEVE’s but he cites trapped liquids in slightly pressurised pockets being released. Higher temperature required is due to a long-term convection of liquids (and/or gases, I think) under the sealed layer. I can’t help but see signs of large-scale slurry release all round the comet, especially around the ‘rocky’ outcrop on the body below and to the left of the crack in this paper (what I call the green match or anchor). And also slurry around the north pole on the body.

          • Marco Parigi says:

            Hi Ianw16,

            Yes. I can’t say that I agree with the premise of 32K to 38K being the temperature inside the neck, but otherwise, the cracks reaching down and exposing material which then evaporates explosively fits anyway.
            Bubbles of “mud”, which could be anything from liquid ethane to liquid water mixed with all the other dust and volatiles, would evaporate explosively leaving hardening slurry at the margins, which I suspect are the ridges that are parallel to the cracks.
            The clues that suggest liquids over supervolatiles are subtle and therefore not conclusive enough to override the assumptions of 32K to 38K.

  • dave says:

    Maybe there is a paper to read that makes things clearer, I am confused by a few statements.

    – No large caverns inside the comet, this seems logical from Consert, but we see huge caverns everywhere, certainly more than a few tens of meters in all dimensions. It says 67ps head is homogeneous certainly to a few 10s of meters re Consert Data, but we see caverns much deeper than this?

    – I find the comments on dust ambiguous, the comet is far from all dust and we have several different materials visible on the surface, some seemingly resistant to erosion, presumably chunks of the surface are not fluffy aggregate?

    – Finally there is still so much unknown about the ice where it is and how does it get to the surface, we have seen the water/ ice cycle and some suggestions how it gets to the surface but how deep does this mechanism apply, and does the density of the ice change with depth?

    Finally, ‘Comets are the icy remnants left over from the formation of the planets 4.6 billion years ago’, surely by now there is some doubt about this, there have been so many surprises, not least the wrong water and now the abundant O2 being released from the surface. It seems we do not know how this got there if we use accretion model with out envoking changes to the model, so do we have any idea how old the comet is? and even maybe where it was formed?

    I may be having a cynical morning, but it all looks a bit makeshift.


    • ianw16 says:


      1. The pits observed would have had ices beneath the surface. The tensile strength of the overlying material would be insufficient to keep it in place above such a void. After sublimation, violent or gentle, you are left with a pit.

      2. Unfortunately, the only paper I have that seems to mention this is paywalled:
      However, it mentions sintering of ice/ recondensation of vapour, whch hardens the surface, and develops a thin dust layer above. The KOSI experiment is mentioned, so may be worth a search.

      3. This is only from memory (too many papers to search!), but I think the diurnal skin depth was about 10cm. Of course it’ll be deeper for CO or CO2. To get to the surface it merely has to be heated to the sublimation temperature of that particular ice. As a gas it will simply seep out through the overlying dust. A sort of similar thing happens in the Arctic with methane deposits in permafrost.

      4. If anything, the discoveries at this comet are helping to constrain the timing and conditions under which comets formed. Nothing found would suggest a later genesis; the conditions, once the Sun is fully ‘formed’ rule it out.
      There was recently a paper on the complex organic molecules (COMs) found in comet Lovejoy. Lots of them. I had a quiet night in comparing what had been found in comets, compared to what had been found in interstellar dense molecular clouds, and hot corinos. All of the COMs in comets have been seen in these star forming regions.
      There is no way of comets incorporating these materials in the solar system as it currently is. Ditto with the O2.

      • Dave says:

        Ianw16 pits & ices observed

        Thanks for the reply,Just a few comments

        1. There are many examples on the comet where the roof of a cavity seems to have blown off (no debris in pit suggesting collapse. Also with the ice ‘so say’ rising from below the surface and refreezing, it seems odd that Consert would not find large shallow caves close to the surface where the ice that has frozen has formed then sublimated leaving a cavity, I agree there are some areas where there has been a sink or a removal of dust.
        2. Yet to get some data related to your link.
        3.You remember the same papers as me, I could not remember the depth – However the crusty parts we can see (presumably not eroding because they are not Ice) all seem much thicker than what was in the paper. I believe the paper was suggesting sublimation did not come through this layer, hence it was hard to detect. I maybe wrong I am looking for it to read again.

        4. a lot of info in 4, but the O2 is the oddity, if we accept the comet was formed with the rest of what we can see, how come there was not enough heat or friction to get this reacting, what was it that froze it in place if it was in such a violent birth. I know a soft coming together of dust has been mooted, so that O2 is permitted from birth, but the explanation was not that convincing.
        Maybe it did not get made at this time, or born at that time but not by the accepted method, or maybe there is some other method for it to get there.
        At the moment there are so many loose ends – I find that exciting because it may be something new, although its gone very quiet.


    • logan says:

      Sadly perceive this paper as more of a Statement.

  • Gerald says:

    km-sized building blocks probably would have left over large voids. This scenario looks less likely now.

  • Sovereign Slave says:

    Looked up to see what other material might have a density of 533 kg/m3, and found that the dry density of the wood of the Douglas Fir tree is this exact density. This obviously indicates that comets are made of wood, right? And we all know that drill bits don’t work on wood, right?

    • Gerald says:

      Must be failed drill bits, which cause the jets.

    • sjastro says:

      Perhaps if they used the term bulk density it would remove the need for sarcasm..

      • Sovereign Slave says:

        Yes, it’s been my experience that as soon as the word “bulk” is introduced to any situation, all sarcasm immediately ceases. But, given that in the article above it’s stated with certainty that “Comets are known to be a mixture of dust and ice” (they obviously didn’t think to include an arborist on the Rosetta team), what is the greatest density possible for any given part of the comet? Ice is 917 kg/m3. So, is it possible for any large or small part of the comet to be denser than ice? And again, the Rosetta team couldn’t have anticipated the drill needing to drill into ice? And if the comet is made of dust and ice (and not Douglas Fir, or Oregon Pine for that matter), and I’m wrong about ice being the densest material possible on the comet, what specifically accounts for that greater density within the ice and dust primordial solar system soup formation theory?

        • Gerald says:

          Sovereign Slave,
          if 67P is only remotely similar to other asteroids and comets already investigated, it’s highly heterogenious on the microscopic scale. The densest microscopic grains are probably those of iron-nickel alloys with a density near 9,000 kg/m³. More abundant are probably dust grains made of silicates, like olivine or enstatite with a density somewhere near 3,000 kg/m³.
          Then there are organics, water ice, dry ice, etc., all mixed somehow to end up in the bulk density. near 500 kg/m³.
          Due to the 4-fold abundance (by mass) of the dust portion with respect to the ices, there needs to be much void space inside the comet nucleus, on microscopic and/or on a macroscopic scale.
          This pdf goes into some detail regarding grain composition in asteroids and comets:

          The nucleus should therefore be fluffy, overall. Fluffy material uses to be soft to very soft. Some crust on the nucleus has been anticipated when Philae was designed. But close to nothing was really known about the strength of the surface. Since an encounter of a soft surface has been considered as more likely, the drill was designed for such a soft surface. Since the payload is limited, it’s not possible to be prepared for all unknowns.
          My personal fear was more, that Philae would be swallowed by a thick layer of fluffy dust and snow. But the designers in the 1990s already anticipated a crust, although not quite as hard as it turned out to be.

          • logan says:

            This ‘fluffiness’ has to be very ‘crackly’, according to latest ‘outburst’ models..

          • logan says:

            Maybe more on the field of quickly hardening ‘transient foams’

        • Harvey says:

          Retry, Capta……

          Nothing limits the density of the constituents to that of ice.
          The overall observed density is simply a weighted average of the materials present, including voids.
          But the more dense the material, the less there can be of it; so nothing rules out tungsten, uranium or lead from a density viewpoint; there just can’t be much of the dense material.

        • ianw16 says:

          Of course we could just derive an experiment that would give us a good idea of the overall properties of at least the upper few tens of meters; we could smash a ~360 kg impactor into a comet, at ~10 km/s, and see what size of crater was left behind. If it was rock, any online cratering program would say you should get a ~7m diameter crater.
          Damn, just realised that we already did that.
          You should read the results. It wasn’t 7m, by the way. Not by a long stretch.
          Of course, if one is used to getting ones information on comets from a certain retired electrical engineer and his mythologist counterpart, then that might have passed one by. They merely concentrated on a non-existent electrical flash, whilst completely omitting any mention of crater size and the solid ice ejected therefrom.
          Still, it is probably due to some sort of academic snobbery and self-serving reasons that those guys are not taken seriously. I mean, what’s not to believe about comets being blasted off of rocky planets by interplanetary lightning bolts, possibly at a time when Earth was orbiting Saturn?

        • sjastro says:

          I must be missing something here. Given that ice has been detected by VIRTIS and MIRO, dust detected by COSIMA and GIADA, plus the coexistence of gas (water vapour) and dust in the tail originating from the nucleus, why should there be doubt cast on the comet being a mixture of ice and dust?
          Given that the density of an individual silicate dust particle can be around 3000 kg/m^3 I suppose one can technically state the highest possible density found on the comet is this value. Realistically however one doesn’t think in these terms do they?
          Similarly given that the density of 533 kg/m^3, is calculated result based on the mass and volume of the comet, it is equally implausible to suggest an isolated sample measurement from the comet is indicative of the comet’s density.
          A density of 533 kg/m^3 is a realistic value subject to experimental error in the data generated for the mass and volume calculations.
          Given this value is much lower than the individual densities of ice and dust, it is an indication of the bulk density of a matrix composed of ice, dust and trapped gas which is supported by CONSERT data.

        • Sovereign Slave says:

          Sorry, my post above was confusing related to the drill bit and density, so let me restate. If comets are composed of basically an even mixture of dust and ice, granted, the different types of dust could account for different densities, but dust is still dust…it’s not a solid material except for the ice that is holding it together. If the frozen volatiles weren’t there, one dust pile would be just as easy to drill through as another, regardless of the density of the material the dust is made of. Similar to saying that mercury would be just as easy to drill through as water, even though it’s much more dense. So, I guess I’m referring to hardness, not density, when saying that the ice must be the HARDEST material on the comet. And if the scientists knew the drill had to drill into ice at whatever hardness it would be in a vacuum, why didn’t they design it to do so?

          • sjastro says:

            SS wrote,

            Given that Philae was not anchored to the surface blame Newton’s third law for the drilling problems.

          • Gerald says:

            Sovereign Slave,
            when Philae was designed, nobody remotely knew, how the surface of the comet would be structured or composed. The MUPUS hammer was designed on the basis of a best guess. A crust of organic-rich material, a few decimeters thick on top of soft interior was considered. But it could have been a brittle field of dust ot debris, as well.
            Not to provide at least a chance to drill the comet, simply because the surface properties aren’t known, would probably have been assessed as stupid, if at hindsight it would have turned out that it would have been possible. So the decision has been to provide a chance, despite not being sure that it would work. But being prepared for all conceivable possibilities wasn’t possible. So, some decisions needed to be made. Some worked well, others less. At least some hardness information is now available. That’s a result, too.
            Another thing: Don’t forget, that Philae landed on a shadowed place with surface temperatures of about -160°C. At those temperatures water ice is much harder than close to 0°C. Initially landing at a sun-illuminated location had been planned. An icy crust would have been softer under the planned conditions, at least near the surface, if ice could have survived near the surface, at all.

    • ianw16 says:

      ‘Bulk’ density, i.e. average. On Earth it is 5500 kg/m^3. You’d struggle to drill through that. Take the same drill to Dover, and try it on the White Cliffs, and you’d have more success!

  • Guili says:

    Long delayed Osiris images of the landing phase were finally promised in January. We are February the 3rd. These images are 16 months old. Will the joke still go on for a long time ?

  • Harvey says:

    Sovereign Slave:
    But it could be Oregon Pine!
    Or even Sycamore, those wretched things spring up everywhere.
    But one thing is absolutely clear; comets are the left overs of a cosmic logging operation, no doubt about it.
    The out gassing is obviously would alcohol, sawdust jets, outside nicely carbonised, fits ALL the data.
    Do I need a :-). ?

    • Sovereign Slave says:

      Well, I’m sure with a bit of fancy math and a few hundred simulations, there may be some grant money in it for us. And something has to be fueling the sun, right?

    • dave says:

      Re Harvey Oregon pine.

      I’ll drink to that.


  • logan says:

    Applauding Martin Pätzold, Andert, Hahn and allies effort. Found the document open [at Local Campus] today 🙂

    On quick view find myself unable to assess the paper, at this moment.

    Salvos should be consented on making stances on Gravitational Doppler methods with New Norcia’s 35m dish, receiving signals from ROSETTA’s 2m antenna, orbiting an extremely irregular 3000m cosmic object.

    Lots of good stances can be made based in this work [As having -for the first time- a raw 3D mapping on the density of a comet. Or no core, no mantle, see You all 🙂 ]. Wander if this particular statement, is spin journalism related.

    How much confidence do you have on the stability of the clock, along all of the duration of the sampling, at ROSETTA transmitter? Or continuous particle impacting introducing noise in the experiment?

    Considering the multiple clues of layered structures at 67P. Chambers could be long and wide, and of little height.

    Accepting this -extremely provisional- statement..

    • Dave says:

      logan re multiple clues of layered structures.

      I also think that based on the visible layering and the method suggested bringing ice up to the surface where it can sublimate into space, there must surely be some rather thin caverns or discontinuities with large areas. Thinking of something like the look of carbon flakes in unmodified cast iron.

      So is it that Consert could not see this type of cavern, especially when hampered by Phillae not avilable.


    • logan says:

      Thank Dave, for the courage to express your idea in view of contradicting statements from the Qualified Souls.

      Regularly I accept the concept of having to live with multitude of scenarios.

      Maybe I’m starting to value my ‘mind investment’.

      The ‘un-investment’ would go as far as my totally accepted Gerald’s idea that primigenial grains are hydrophilic and hydrophobic…

  • Uwe Dietzel says:

    Hi, I’ll make two points with my comment: 1) The comet will be gone more or less within 1300 years and more importantly 2) Earth could still collect water from space and maybe didn’t need direct impacts of comets to do that in the past

    When I looked through the photos, that the Rosetta probe took from Churyumov–Gerasimenko I noticed an interesting detail: During the comets flyby 2015 near the Sun (perihelion) Rosetta observed a loss of 20 meters (!) surface material in form of gas and water vapor.

    But the comet measures only 4.1 km in its widest extent and 1.1 km in its lowest. In 1959 the comet had an encounter with Jupiter, which changed its orbit around the Sun to a very close rotation of only 6.44 years. That means, if Rosetta’s observation represents the usual rate of evaporation/sublimation, the comet shrinks 20 m every 6.44 years and will be completely degassed in less than 1320 years! It will have lost all its water, and the leftover will probably be only a bunch of small-sized rubble.

    If this can happen so fast, isn’t it likely, that the inner planets get a lot of water vapor from comets? There is still the question, where the water on Earth came from. Maybe this question is wrong, and the process is more dynamic than previously thought!?

    Comets only develop their tail inside the orbits of Mars and Earth. The solar wind probably pushes part of the emerging vapor and gas outward again. But with the many comets circulating around the Sun, maybe it doesn’t need a direct impact to transport water to Earths surface? Earth with its comparably large gravity can probably collect gas pretty well from the surrounding space…

    • Harvey says:


      Water, or any other gaseous material that evaporates into space, then diffuses out three-dimensionally due to the thermal velocity of the molecules. The fraction of this that would be intercepted by a planet is infinitesimal. This is ‘molecular flow’ and quite unlike the ‘viscous flow’ we are used to.

      There are also issues around whether it would, or would not, be ‘swept up’ by a planet anyway. That’s not straightforward due to the photo dissociation of H2O to H and OH and photoionisation too. For H at least it would thermally escape earth gravity.

      But the amount the planet would ‘encounter’ is so minute is hardly worth considering those issues in detail.

  • logan says:

    Hi Uwe. No doubt our Scientists will eventually give a ‘solid’ answer to those, very important questions..

    Preferred answer as of today is that most of water was already integrated in the accreting material of the planets, not discarding Cometary Material as part of that primigenial recipe.

    As every person, Planets have their individual histories. And our young Sun behaved quite different.

    To collect gas in space, to a measure of relevance, you need a quite ‘cool and quiet’ environment, or ‘steal’ it from your neighbors 🙂

  • logan says:

    Clues of a former cavernous chamber at neck, at the beginning 8-12 frames of this gif. This is a repost.

    Could also account as very rounding and selective erosion.

    Lots of another explanations also plausible..

    • logan says:

      Any void this size, forming at what actually remains of 67P, impossible.

  • ianw16 says:

    By the way, for anybody still holding on to the vain hope that this comet has a significant fraction of rock in its make up, they should also note that MIRO is easily capable of distinguishing between rock, dust and ice.
    As seen at the asteroid Steins, the thermal inertia of rock is very different to that of ice and dust. Not that MIRO has sampled every square centimeter of the surface, but one would think it would have found something by now.
    Ice is pretty much transparent to MIRO, dust comes in at around 10 – 60 J/(K m2 s^0.5), and rock at ~ 650 J/(K m2 s^0.5).….4410201H

    If that link doesn’t work (adsabs links don’t always copy well) just Google ‘Millimeter and submillimeter observations of asteroids from the Rosetta spacecraft’,’ on Google Scholar.

  • originalJohn says:

    Interesting. The previously widely quoted density figure was in the 400 s of kg/m3. Now it has leapt up to 533 kg/m3, a 25 % increase, without eliciting any comment or explanation from the authors. So, was the measurement so wildly out before or has the density actually “changed” since the 400 days.

    And once again the comets as precursors of a 4.6 billion year old solar system is stated as fact. It is a hypothesis, a belief, dogma.

    Then this statement appears :

    “The most reasonable explanation then is that the comet’s porosity must be an intrinsic property of dust particles mixed with the ice that make up the interior.”

    For “the most reasonable explanation” read “an interpretation”. It should then be comparatively easy to sample some of the emitted dust particles and examine them closely to see if they exhibit a porous structure or not. This would be a good way of assessing the reasonableness of the explanation rather than just leaving it as an assumption, as we are there with the orbiter.

    • Kamal says:

      OriginalJohn: The last density figure I remember was 0.47, which was provisional to the southern hemisphere becoming visible, at which time its volume would be computed. It seems from the new number of 0.533 that the southern hemisphere has turned out a lot flatter than earlier expected. It is quite a difference, it brings C-G closer to other comets like Halley.

      Combustion of hydrocarbons is in consonance with lower densities than required for the electric machining of rock. Perhaps the surface is something like coal, which superfiicially looks rocky, and can be hypothesized to burn.

      • originalJohn says:

        I presume Kamal that you mean S G ( specific gravity) which is what we are talking about, rather than CG ( perhaps centre of gravity).
        As I have said elsewhere I am surprised at how far out the volume assumption was. And bringing the specific gravity closer to that of other comets such as Halley is of no value. All of those were based on estimates (assumptions) of both volume and mass, derived from the Whipple comet hypothesuis. This one uses a true measurement of mass and is more reliable.

        I don’t get the point you are making Kamal about combustion and lower densities. Why should any particular density be required.
        I have not seen any results on the properties of the hydrocarbon layer from this mission, other than it being confirmed as a layer which coats the whole surface. It would be surprising if it resembled coal which is carbonised wood, or carbonised some other fibrous organic material. I have, I must admit without justification in the absence of evidence, imagined the layer to be solid but more like a frozen supercooled liquid, if not just liquid. A hydrocarbon oil in fact and not necessarily vegetable in origin. It would be interesting to see more attention being paid to it.

      • Kamal says:

        OriginalJohn: And what would you hypothesize is under the layer of oil-based hydrocarbons? Because that is key to density.

    • Kamal says:

      The burning hypothesis goes back to Lord Kelvin’s ideas of explaining the Sun using thermodynamics, leading to an age of 20 million years for the Sun. These ideas were shown to be wrong by Charles Darwin, who never published his calculations.

    • Gerald says:

      Originaljohn, the first density estimates were 470 kg/³. The corrected 533 kg/m³ are 13.4% more, not 25%.
      The reason has been the incomplete shape model. Part of the nucleus had been in shadow, and wasn’t visible before recently, for the first time. So part of the surface needed to be estimated, with some implied uncertainty for the volume, and hence the density.

      The dust particles have been investigated in high resolution, and they show high porosity.

      • originalJohn says:

        Yes Gerald, interesting. At the time the 470 figure was being bandied around I do not recall any doubt at all being attached to it. It was held as an irrefutable certainty in the mass and density discussions yet it was based on a wrong assumption of the shape. This reflects the fact that it is impossible to measure the volume of an irregular shape hanging in space. If you can’t get your hands on it and get it to displace water it can only be estimated. Of course this should be emphasised at all times in density discussions but unfortunately rarely is if it does not suit the argument.
        And this in turn reflects the prevalent attitude of the dogmatists that has prevailed throughout this Rosetta mission. Leaping on the most convenient superficial explanation with regard to the dogma and conveniently forgetting that for that explanation to be definitive all other possible explanations must be eliminated. Otherwise the explanation remains an assumption.
        The same is happening now with the dust porosity question. You for example assert:

        “The dust particles have been investigated in high resolution, and they show high porosity.”

        This is a convenient distortion of the truth Gerald. Convenient because the density argument requires all the dust to exhibit high porosity. But it in fact does not does it Gerald. In the post you link to it is summarised in this statement:

        “While more fluffy particles were detected than compact ones, their size distribution reveals that they only contribute a minor fraction of the total mass of dust being lost by the comet.”

        I will stick in an aside here and say that this reminds us that the comet nucleus is losing matter all the time. To maintain the density figure with later measurements this loss must always be included in the model, and the factor stated.

        Getting back to the dust particles, your linked reference therefore clearly reported two types of dust particles, compact ones and fluffy ones, stating that the density of the compact ones fell within the range 800 -3000 kg/cubic metre.
        To register a density of 533 you would not want too many of these compact ones in the mix. Yet the above quote suggests they comprise most of the dust emitted and probably therefore most of the dust present. The argument for most of the dust being fluffy and therefore porous, as needed, begins then to look rather weak.

        Then the question arises why would there be two distinct types of dust, types that were so physically different, in the same body. It seems to me this has not been addressed at all, again because it does not help the argument.

        Might I suggest that the compact dust is removed by an energetic process and in a form which represents the physical structure of the comet, and the fluffy dust is generated in a process which is known to commonly produce such a physical form, electric discharge. This is not of course an assertion. It is a suggestion.

        Nevertheless it is clear there are huge gaps in our knowledge of the relationship between the physical characteristics of the dust and the density of the comet nucleus. The tiny amount published so far, whilst highlighting some of the issues, cannot be held to answer the dust and density questions categorically. Yet for you Gerald it appears to be an open and shut case. Many more aspects need to be investigated and in far greater detail before we can have any certainty.

        • ianw16 says:

          “To register a density of 533 you would not want too many of these compact ones in the mix. Yet the above quote suggests *they comprise most of the dust emitted* and probably therefore most of the dust present. The argument for most of the dust being fluffy and therefore porous, as needed, begins then to look rather weak.”


          “During the study period, a total of 193 compact particles were detected, impacting the GIADA detectors at an average speed of 3 m/s. A total of 853 detections of fluffy particles were made…..”

    • ianw16 says:

      You should read some of the papers produced by the team that come out now and again.
      The density has changed due to the more accurate shape model that has been produced since the southern hemisphere has been illuminated. Mass is the same, volume slightly less. Therefore, as density = mass/volume, the density has gone up.

      As for the 4.6 Ga age of the comet, perhaps you’d like to tell us how this could have been formed at any later time, given the conditions prevalent in the solar system once the Sun is ‘lit up’. Have a look at all the COMs found on comets, and correlate that with where they are found in surveys outside of our solar system. Or the Nitrogen ratio etc. Come up with a *scientifically valid* explanation of how a comet could form in ‘recent’ times.

      ” It should then be comparatively easy to sample some of the emitted dust particles and examine them…….”
      They did. Again, you haven’t been following the published literature:
      “GIADA was able to distinguish different types of particles populating the coma of 67P: compact particles and fluffy porous aggregates.”

      • originalJohn says:

        Believe it or not ian I do read some of the output of the “team” that comes out now and again. And I realise that there are a lot of complexities in this density issue. It is far from being the clear cut situation that you would like to convey. Of course if you change your volume assumption the density goes up. I am surprised though it is by such a degree and that the original assumption imagined such a large lump of nucleus that was not there.

        This is probably not the place to get into detail about a solar system formation hypothesis. I referred to it because it was presented once again as fact, when it is simply a belief. Neither you nor anyone else has any way of knowing what conditions were present in the solar system when the sun lit up, or indeed whether there was a system then at all, or whether any accretion of dust or gas was involved. The whole thing is a story and the whole question is one of plausibility. I can come up with plausible alternatives but they have no effect on whether the mainstream model is plausible or not. So on that plausibility alone it is not plausible at all, and on many counts. The accretion model for example requires the gas disc to remain extremely cold yet comets are found to contain minerals which can only form at high temperatures. A glaring inconsistency conveniently ignored. There is also no mechanism, with gravitational attraction ( almost nothing with dust particles of equal size) or electrostatic attraction that can ensure they stick together. No model in other words that works. To stick together you need not just contact but a mechanism of adhesion to overcome the surface energy, and once again one that works at extremely low temperatures. Perhaps you can come up with one. That would help the story a great deal.
        As for COMs which you chose to initialise in a jargony way I cannot for the moment guess what you are referring to so cannot comment.
        Chemical factors such as isotope ratios could only be categorical if you could assume that nothing could happen within the hypothesised 4.6 billion years to affect that ratio. This is of course a rather ridiculous assumption of uniformity. In other words once again the common belief procedure is to make the facts fit the dogma, which proves nothing, except perhaps the strength of your belief.
        When the plausibility does not stack up I prefer then to consider alternative explanations and it may not surprise you that I have some which you would not accept. Not because they are any less plausible than yours ( they are more plausible) but because your belief in the dogma is categorical.

        For that reason it is not possible for us to argue about it and I am not prepared to do it here. I can however say there are some very interesting ideas out there if you can break away from the consensus and use your intellect, which I would highly recommend.
        I have addressed the dust porosity issue in my reply to Gerald and to avoid repetition I refer you to that.

        • ianw16 says:

          COMs = complex organic molecules. As anybody with an interest in such matters would know.
          They need certain conditions to be able to form. Those conditions are not prevalent in inner solar system space as it currently is.
          However, they are found in dense interstellar molecular clouds. the precursors of solar systems. And also in protoplanetary disks

          “Dogma”? Pot calling the kettle black there, I think!!
          Perhaps you could sum up the evidence for the electric comet woo as it currently stands. Or do you no longer hold on to that idea quite so dogmatically?

          As for my intellect, that is precisely why I don’t have any time for the woo dreamed up by people who simply don’t know what they’re talking about. If you’d like to link to these ideas within the scientific literature, I’ll be happy to read them. If the authors haven’t dared to risk peer review, then don’t bother.

    • Harvey says:

      There is nothing strange or mysterious about this.

      The mass of the comet is deduced from its gravitational pull on Rosetta, which is very small, and gets smaller with distance.
      So the measurement becomes more accurate as Rosetta gets closer for that reason.
      But also the longer you measure it for, the more accurately you can determine the result for two reasons. You can integrate up the signal and reduce the noise, and build a more sophisticated model including shape effects.
      So it’s entirely predictable that early figures will get revised, and the result become progressively more accurate.
      Note the increase from one significant figure, 400 to three significant figures, 533.
      Now if it changes *again* significantly, that would need explanation.
      But these changes look routine.

    • Booth says:


      I must say, “Interesting” is an understatement! Think about it, a bulk density with three sig figs? Freakin’ AWESOME is more like it! Rosetta continues to add the most extraordinary detail to our understanding of comets!

      Of course, … you always seem to have some kind of problem with the Rosetta data! Regardless of your beliefs, you can’t ignore the Rosetta data!

      Regarding the new density value, you write, “So, was the measurement so wildly out before or has the density actually “changed” since the 400 days.”

      No! The bulk density of 67P has not changed! Our knowledge of the mass and volume have improved over time. Remember, bulk density is a derived or calculated value based on an object’s mass and volume. Perhaps a short (and grossly oversimplified) history lesson in how the bulk density has been refined over the last 18 months is warranted ….

      Brief aside – spacecraft safety demands that we know where the hard comet edges are at any given moment in time (i.e., volume) and the forces acting on the spacecraft to either push it away (i.e., outgassing – which will be negligible at rendezvous) or pull it in (i.e. gravity as a function of mass).

      Approximately one week before rendezvous, the Flight Dynamics team observed the gravitational effects of the comet on the spacecraft’s trajectory. Because Rosetta approached the comet from the sun-ward side (thinking safety first), it was not possible to use RSI in the earliest mass calculations. Hence, the initial mass of 67P was determined using an “olde skool” technique – Newtonian gravity! GRAVITY! Love that word! GRAVITY! This method, of course, resulted in a very rough “estimate” of mass which could include a significant error term – think about 67P’s odd, rotating shape. Regardless, this initial mass value was used by the Flight Dynamics team to plan Rosetta’s orbital course to allow RSI operations. As soon as the spacecraft could be positioned correctly, RSI was employed to refine the mass value for 67P. IMPORTANT POINT – before the first bulk density value was ever reported, the mass had already gone through a number of iterations or refinements! Volume was determined by examining OSIRIS and NAVCAM images to establish “heights” relative to an arbitrary datum – in this case, the center of mass. This allowed the scientist to build the first “3D shape model” of the comet. Now, recall that at rendezvous the winter hemisphere was not illuminated, meaning that a great deal of volumetric information was going to be missing! Restated – the initial volume of 67P was going to be … wrong! As it turns out (based on the results reported in this thread, and previously calculated bulk densities) the initial volume was going to be dramatically overestimated.

      Thus, the first bulk density was reported as (10^13 kg) / (25 km^3) = 400 kg/m^3 – Note the date!

      The Rosetta science teams NEVER once claimed that this bulk density was absolute! Nor did they sit idle. As time went by, RSI built a database of mass measurements. At the same time, the volume value was being refined. First, MIRO was used to take off-limb temperature measurements to find the edge of the cold, unlit southern hemisphere. Not very accurate, but a significant improvement on a “guess.” These improvements in mass and volume yielded a second bulk density of 470 kg/m^3. Following sunrise on the southern hemisphere, the volume was again readjusted based on new OSIRIS and NAVCAM data to give a bulk density of ~500 kg/m^3. And now …?

      The more precise density value of 533 ± 6 kg/m^3, as reported in this thread, is well within the value range expected of comets composed of ices and dust!!! Remember the KOSI experiments?

      Is this the TRUE bulk density of 67P? No! Probably not! But it is closer to the actual bulk density than that which was calculated at rendezvous. How much closer? To answer that we need to continue measuring the mass and improving the shape model.

      Now, as we say in the business, comedy is timing, and you write, “The previously widely quoted density figure was in the 400 s of kg/m3. Now it has leapt up to 533 kg/m3, a 25 % increase, without eliciting any comment or explanation from the authors.” WHAT THE …?

      What do you think this thread, and the associated Nature paper is all about? REALLY!!! Seems like the authors are providing a perfectly acceptable explanation! And how much more commentary do you need? Given your attitude towards Rosetta science, would any additional explanation help?

      You write, “And once again the comets as precursors of a 4.6 billion year old solar system is stated as fact. It is a hypothesis, a belief, dogma.” WRONG! WRONG! and WRONG!

      First of all, the article states that, “Comets are the ICY REMNANTS LEFT OVER from the formation of the planets 4.6 billion years ago.” Completely different context and meaning!!! And you can’t argue with the decades of evidence and the Rosetta data that supports this claim! However, I am certain that you will, because you are always right and thousands of scientific investigations over many decades are completely wrong! Do you have any EU/EC evidence you would like to share with the class at this time? No? Didn’t think so!!!

      Next, you quote one sentence directly from the article and then immediately proceed to ignore the OBVIOUS Rosetta data that follows. Allow me to refresh your memory! This is what the article actually says …

      “The most reasonable explanation then is that the comet’s porosity must be an intrinsic property of dust particles mixed with the ice that make up the interior. In fact, earlier spacecraft measurements had shown that comet dust is typically not a compacted solid, but rather a ‘fluffy’ aggregate, giving the dust particles high porosity and low density, and Rosetta’s COSIMA and GIADA instruments have shown that the same kinds of dust grains are also found at 67P/CG.”

      How could you miss the obvious reference to COSIMA and GIADA, and then absurdly state that, “It should then be comparatively easy to sample some of the emitted dust particles and examine them closely to see if they exhibit a porous structure or not. This would be a good way of assessing the reasonableness of the explanation rather than just leaving it as an assumption, as we are there with the orbiter.”

  • Kamal says:

    As 67P heads for Jupiter (now they are in the same region of sky) I wonder whether the gravitational effects of Jupiter on the nucleus can be measured. While I was looking up the sky map I also realized that 9P/Tempel I is also in the same region. 67P and 9P were quite close to each other on February 2, perhaps around a million kilometres.

    • Gerald says:

      Since NASA’s DSN can measure velocities with an accuracy of about 10 micrometers per second, e.g. according to slide 15 of this pdf
      I’d be very surprised, if they couldn’t measure the gravitational effect of Jupiter.

      • Kamal says:

        Thanks, Gerald, that is an interesting document.

        So perhaps one should be able to calculate tidal effects due to Jupiter. If there is any surface on the nucleus where these effects can be seen one could make some predictions. If there is liquid somewhere below one can hypothesize whether Charon-like fracturing might happen on the surface above it.


      • Kamal says:

        On a different note, we should be reaching the equinox soon, after which the southern hemisphere will start going into shadow. So if the summer heating has had a cumulative effect on substances below the southern hemisphere surface, one should see something now. After that one would look to Jupiter to cause some action.

    • ianw16 says:

      From memory, Rosetta took measurements of the Tempel 1 impact in 2005.

  • sara says:

    good post

1 Trackback

Comments are closed.