Skydiver films falling meteorite

Today the Norwegian broadcasting corporation NRK will air the popular science programme Schrödingers katt which features the incredible story of the skydiver Anders Helstrup. In a jump above Østre Æra airstrip near Rena in Norway on 17th June 2012 the totally improbable happens: Just after he deploys his parachute about 1000 metres above ground level a dark rock from above zips past him. His two action cameras document the incident. How can a rock suddenly appear this far above the ground? There is nobody above him, neither an airplane nor other jumpers. No other possible explanations seem to remain but the improbable: a meteorite just fell a few metres away.
Dark flight
(Photo: Anders Helstrup / Dark Flight, photomontage: Hans Erik Foss Amundsen)

Meteors enter Earth’s atmosphere at speeds ranging from 11 to 72 km/s. The bigger ones will appear as fireballs in the sky. Most are completely vaporised, but some fragments sometimes reach the ground as meteorites. Most of the speed is maintained until an altitude of about 30 km, but from that point they rapidly decelerate and they completely fade about 20 km above ground. The speed has then been reduced to a few km per second, but the meteor will continue to decelerate until it reaches terminal velocity as if dropped from an aircraft. When the fireball is no longer emitting visible light, any surviving fragments will enter their dark flight portion of their fall. These meteorites are usually small, a few kilograms or less, and will be moving at speeds around 300 km/h (depending on shape and density) near the ground. They are for all practical purposes invisible and cannot be photographed. But Anders has videos telling us otherwise.

Anders teamed up with a number of individuals including experts from the Geological museum of Oslo, the Norwegian Space Centre and the Norwegian Meteor Network, and a project to recover the meteorite was formed. The recovery of the meteorite would be the ultimate proof that the impossible really has been photographed. The videos have been carefully analysed in order to pinpoint the location of the skydiver at the time of the encounter. There have been many attempts to calculate the impact site, and hundreds of hours have been spent on the ground searching. It has been a really exciting project for those involved in the hunt. Yet, no meteorite has been found. The Norwegian broadcasting corporation NRK has documented parts of the hunt. Despite that the meteorite has not yet been found, Anders has now chosen to go public with his story and to publish the videos and analysis work done. This is, however, by no means the end of this story. We hope going public marks the beginning of a new chapter. Perhaps we missed important clues. And perhaps going public is the move required for the meteorite to be recovered. We sincerely hope that this story goes viral. All the material is copyrighted Anders Helstrup / Dark Flight and can not be used for commercial purposes without a written permit. Research institutes, schools and amateurs are welcome to use the material for free as long as the copyright holder is acknowledged. We hope that someone out there manages to crack the riddles and help us find the Dark Flight meteorite. Or prove that the rock is an illusion, a bowl of petunias falling from the sky, or something else. We guarantee that the videos are real and have not been manipulated.

Anders has two cameras that recorded the event. Another skydiver films Anders at a distance, but too far away for the meteorite to be detected in that recording. The camera mounted on the front side of the helmet records in 1080p30 and captures the best details and the meteorite is visible in 7 frames. The recording shows that the meteorite is spinning very quickly, apparently at least 15 revolutions per second. The animation below shows drawings based on the video with the corresponding original image in the top left corner. The images have been resized corresponding to distance.
Meteorite
(Photo: Anders Helstrup / Dark Flight, animation/drawing: Steinar Midtskogen)

The video mounted on the back of the helmet records in 720p50 with a wider lens and no details of the meteorite are visible. The video does confirm that something fell at a great speed towards the ground, and it shows the position of the meteorite relative to the photographer and the ground, but since the photographer is also moving, it’s a non-trivial task to calculate the impact site. The video also makes it possible to estimate the albedo of the meteorite, approximately 0.12 which is pretty typical for meteorites. The picture below shows the path of the meteorite (marked with yellow dots) relative to the background.
Back facing camera
(Photo: Anders Helstrup / Dark Flight, photomontage: Steinar Midtskogen)

The exact size of the meteorite is not known, but it appears to weigh several kilograms. We can assume that it’s falling at a speed greater than 220 km/h, more likely around 280 km/h. We also assume that Anders is falling vertically at a speed of at most 100 km/h, and the relative speed is correspondingly lower. We then find that the meteorite measures at least 7×9 cm weighing 900g (assuming a density of 3.2 g/cm³) zipping past 2.5 metres away. More probably it’s measuring 12×16 cm at 4.6 kg 4.6 metres away. The upper limit for the meteorite seems to be 18×24 cm at 20 kg zipping past at a distance of 6.5 metres. This means that the meteorite appears to be roughly the same size as the Valle meteorite, the most recent meteorite discovery in Norway. If Anders had been struck by the meteorite, the outcome would likely be lethal, and the accident investigators would surely be puzzled!
Size
(Photo: Anders Helstrup / Dark Flight, figure: Hans Erik Foss Amundsen)

Several independent approaches have been applied to pinpoint the exact location of the encounter. All of them point to a location almost directly above Kjølsætra, an old summer farm a bit south of the airstrip. But what if we made some mistakes? Perhaps we’ve done the most thorough search on the wrong side of Kjølsætra? We hope that others will study the videos with fresh eyes and come up with alternative solutions. The picture below shows the encounter as seen from the second jumper, Jon Vegar, and the figure shows what direction Anders must have been travelling in assuming that the meteorite falls dead vertically. Kjølsetra is seen at the bottom of the picture.
Jon Vegar's perspective
(Photo: Jon Vegar Andersen / Dark Flight, photomontage: Hans Erik Foss Amundsen)

The weather was fair with south-western winds around 5 m/s. The approximate local time of the incident was 14:06. There is an uncertainty of a couple of minutes. The wind is confirmed by both weather models and cloud movement seen in the videos. We have examined camera recordings showing the sky over Oslo at the same time looking for signs of a fireball, but it was more cloudy in Oslo. We have no reports of a fireball from witnesses, but it could have been impossible to see because of the clouds or the glare from the sun. Even seismic data have been examined for hints of a fireball without luck. And any people who might have heard something might easily mistake a sound from a fireball for thunder or construction work. It is not uncommon that meteorites fall in populated areas without reports about anything unusual, the Oslo meteorite in 2012 being the most recent example.
Wind profile
(Wind profile: Norwegian Meteorological Institute)

More information will shortly be published. All the videos have been posted on YouTube, and a dedicated website is in the making (not yet online at the time of writing) which will include videos and some information on the work that has been done so far.

Anyone who finds a meteorite in Norway can claim ownership of it, and a 5 kg meteorite will fetch a handsome price on eBay. However, this meteorite will be worth a lot more if it is bundled with the film footage and the story, and perhaps wrapped in a piece of the parachute when offered for sale. So we hope that the lucky finder will join the project to get the maximum reward. The meteorite should be made available for research and for the public, and we aim to give the Norwegian National Meteorite Collection the first right of refusal to acquire a piece of the meteorite.

Dark Flight project contacts:
Anders Helstrup, reisefotanders@gmail.com
Morten Bilet, mbgeotop@gmail.com
Hans E.F. Amundsen, ha@epx.no
Steinar Midtskogen, steinar@norskmeteornettverk.no

NRK story:

40 thoughts on “Skydiver films falling meteorite

  1. If you know the size of the sensor in the camera and the lens focal length you can calculate size of object versus distance and can better pinpoint the size of the object.

  2. I can do the math, including the aerodynamics. I would need the make/model of the camera and the focal length of the lens and the video. I cannot download it off Youtube and I need to do some frame captures from it to get distances etc.

    • The front camera is a Contour VholdR recording in 1080p30. Try their website for specs. It’s 110 degrees IIRC 135 degrees. The back camera is a GoPro Hero2 recording in 720p50. To estimate the size not just as a function of distance, you also need to know its relative speed, which is not known precisely. Our guess is that it’s dropping at a speed between 220 and 400 km/h and the skydiver has a vertical speed probably somewhat less than 100 km/h, which will have to be subtracted.

      Here are the individual frames: frame 1, frame 2, frame 3, frame 4, frame 5, frame 6, frame 7.

  3. In my opinion it is nearly impossible and unlikely that it was a pure vertical trajectury. Did you ever watched a rocket lauch? Did you see how the rocket seems to go to the side? Yet from inside the rocket the trajectury would still seem to be straight up vertical. Its the earth that moves so don’t forget to calculate that offset.

    • I’m not sure what you mean, but meteors (except large asteroids) quickly lose all their original momentum during the dark flight and the expected behaviour would be a near vertical drop at terminal velocity drifting with the wind. In the footage, however, it does not appear to fall dead vertically, but the speed and direction of the skydiver are complicating factors.

  4. I made also my own “flypass model”. In this I compare the velocity derived from the image (-model) to (dynamically) calculated velocity values for differen meteorite sizes. The criterion is the mutual fit of the object size and the velocity. I seem to have the drag coefficient somewhat smaller than you have this, which gives the solution value somewhat bigger. Yours may be more correct(?), but this is not my main point in here.
    Going either upwards or downwards in the mass (size) from the best value would give more poor fit between these. This would hold in the whole size range IF there were not the downward motion of the skydiver and camera.
    This down-velocity changes the situation in the smaller size. And this results that there is another solution to the size in the cm size range!
    The actual solution is dependent on the drag coefficient and the camera downward velocity ( among others like meteorite density ..).
    And there really exist such a possible solution would say in between 1 and 10 grams. I assumed the camera downwards velocity of 25 m/s , A smaller velocity shifts this solution to even smaller size but a bigger drag coefficient moves this to bigger sizes.
    This would mean that the small meteorite fragment passed the camera within around 1.5 meters only. And I see nothing that would be against this model (unless it might have then collided with the parachute). Small meteorite fragments are a lot more numerous than about kg size meterites, especially as fragments but also as individual cases. The camera lens aperture probaly was so small that no poor focusing could have resulted from this, as judged from some nearby things.
    So, I really suggest this as an explanation of why the meteorite has not been found. You probably did not seek a few grams meteorite, and even if you would have searched that is a lot more difficult to find than say a kilo size meteorite. If still tryong to find ths (speculated) small meteorite, then of course the wind drif model down to the ground would need to be calculated anew.

    Esko Lyytinen

    • Whether a meteorite of a few grammes would be easier missed or not on the ground would depend on where it landed. There are many wet, marshy places there where a heavier object more easily would bury itself into hiding than a lighter object. In any case it will be hard to find. The key is to narrow down the search area.

      Making the object very small will also make the object very slow. I would say 25 m/s relative speed would require pushing all the parametres somewhat, but this is also what we invite everyone to do. If you assume a small object weighing a few grammes, can you get it to match what is seen in the backwards facing GoPro Hero 2 camera?

      I do see some things supporting a small size and close encounter: The very rapid spin, and the sudden appearance in the backward camera well inside the frame. On the other hand, this is a 720p camera with a 170 degree field of view. A very small object would only be visible within a few metres.

      • Yes the relative velocity could be around 25 m/s or maybe even somewhat smaller.

        If the cameras were in the same postion except vertical difference and the target moving directly down in constant velocity, then the angular behaviour (both velocity and apparent diameter) for a big one and for a small one, would be (mutually) similar in the backward camera if they were mutually similar in the front camera. But I would expect that the backward camera has some horizontal shift as compared to the forward camera. And this would give some “parallactic” difference to the apparent paths. Maybe the tracks (or one of them) can be extrapolated to a common image background (gorund or clouds) and this would tell the azimuthal difference. This could tell, if it was quite nearby. I wolud like to try this.
        Also a very exact (better than one frame interval) mutual timing frame for the two vidoes would be useful, if possibly available. This could utiilize the cameras vertical different locations.

        (However I started the study of this case in the evening of thuersday and worked practically the full day of yesterday, so I may need some “time delay”,)
        I only worked with the forward camera. Actually in my quck look at the other video, I did not recognize the target at all. Could I get the original individual frames of this video, where it is visible.

  5. Folks, it was a pebble caught in the canopy that escaped upon opening. I’ve seen it myself with my own parachute, 30+ years ago.

    With a wide angle camera, assuming it *was* a pebble caught in the chute the last time it was packed, it was no more than 10 feet from the camera as it fell from about 10-15 feet above it. Then the motion is around 1 ft/frame and speed around 60 ft/s, plausible if the inflating canopy acted like a blanket toss and launched the rock downward.

    (Actually, it turns out I had a whole bunch of pebbles and dirt caught in the canopy [I hadn't checked the last flake as I packed it] and with my head resting on my reserve chute that rode high on my shoulders I was in the habit of watching the entire deployment sequence. That particular day I saw the pilot chute haul the bag away, then the canopy expanded outward- and a veritable explosion of sand and gravel shot past me on all sides, over and done before I could flinch. I laughed all the way to the ground, though.)

    • We encourage you to analyse the videos from Anders’ two cameras, do the math and find a solution that matches the videos while being consistent with an origin in the parachute. It’s pretty complicated and we ask for help.

  6. I agree with Esko. I get a 5-10cm sized object close (1m) to the camera. It appears to track horizontally about 2.4m/s which leads me to think it is more related to the canopy opening. I expect a meteor to be falling vertically and not tumbling by then. But I would tend to look for a smaller rock on the ground. I will have coordinates soon.

    • Thanks. Yes, I agree that there must be horizontal movement relative to Anders. However, it must also be taken into account that Anders is in no way still. He still has much speed and he almost makes a full turn. Jon Vegar’s video might be worth watching as well for clues, but careful synchronisation of the videos are required. Anders appears when he deploys in chute, then goes of the frame for a few seconds and then becomes visible just as the object passes.

    • I also get a good fit wit assumed non accelrating velocity. But The timing ( in the front video that I have analyzed) when it was visible is only 0.2 seconds. And I came to the conclusion that at least without a very detailed camera image projection calibration etc the two alternate possibilties can not be resolved with this. In 0.2 seconds the velocity would increase at most 2 m/s but because ( an assumed small stone) would already have a considerable velocity relative to the terminal free fall velocity, less than this. In my fit, the individual (between frames) velocity values deviate at maximum about 13 % form the average. There is an about sinusoidal variation of about one cycle ( at 9 Hz). This may be from the inaccuracies but might also tell of some (vetical) oscillating movement with an amplitude of about +-10 cm or smaller, assuming a small piece. But this was achieved with suitable detailed fit of the angles to the relative fligh direction. A minor change in this could probably make this suit for a slightly acclerating model, (maybe not remove the oscillation?).

      I have quite a lot of experince on accurate calibrating of camera imaging geometry even from quite scarce different type datas, but this is very time consuming and with this probably can not do this, at least now soon.

      I would still like to have the original back camera images where it is visible.

  7. I have worked on similar problems before. I have some questions that I hope might help, but unfortunately I have no time to work on this personally.

    1. Regarding the second, fainter falling object that has been identified in the video, have its terminal velocity, actual velocity, distance and size been calculated consistently, yet? It is either a very small object or farther away. if it is farther away, it is definitely not from the canopy and is certainly a meteor. (Maybe that should be quantified — how far can a rock be flipped off a canopy as it inflates?) On the other hand, if it is much smaller, then the speed it is falling may not be consistent with an object that tiny. Remember also it would have just fallen from the canopy so perhaps has not had time to achieve its own terminal velocity. This additional, faint rock may finally settle that these are definitely meteors and not canopy debris.

    2. If the second, fainter rock is smaller than the more obvious rock, and if they are fragmented from each other, then it should have been traveling behind the larger one rather than in front of it as seen in the video, because smaller objects would have slower terminal velocities. That would imply a third, larger object that the small fragment came from. (That is, the only way it could have gotten in front of the larger rock is if it had been riding upon an even larger one for a while.) On the other hand, if the faint rock is larger than the more obvious rock, then its trajectory being ahead is consistent with it being both larger and further away without needing to invoke a third (unobserved) rock. Has this been checked?

    3. To further rule out these being canopy debris, has anybody checked whether an object from the canopy would have reached terminal velocity so quickly? The comments above mention it possibly being only 1.5 m away and just centimeters in size, which might suggest canopy debris, but that calculation assumed the rock was at terminal velocity. Would a rock falling from the canopy be able to achieve terminal velocity so quickly, considering the dynamics of falling off the canopy? I would assume the canopy caught the wind and slowed for an instant to far below the speed of Anders as well as far below the terminal velocity of the fragment itself. Thus, it was pushing the rock against gravity to a very small velocity. Then, after being fully deployed, the canopy was suddenly jerked and accelerated downward by the cords as both Anders and the canopy rapidly achieved their new terminal velocity, with Anders decelerating in that process and the chute at first accelerating and then decelerating together with Anders. In that sequence of events the debris would have come loose and begun falling from the initially slow velocity of the chute prior to the cords jerking it. A comparison of the overall system might clearly rule out debris because the jerk the canopy got from Anders would have pulled it some distance from the rock and then the rock would have to accelerate from a slower velocity and from further behind, and so it would take some time to catch up. Perhaps an analysis of that would show that the rock was traveling far to fast (no matter its distance) to have come from the canopy.

    4. If the fainter fragment is consistent with it being a meteorite at terminal velocity, and if it is larger than the more obvious fragment, then you should be able to calculate how much backward in time they fragmented from each other by the time-separation they have achieved at their relative terminal velocities. A complete analysis would include the decelerations after fragmentation from the original terminal velocity that they shared together until they each achieve their new terminal velocities. Integrating the decelerations twice should produce the observed time delay between fragments as a function of when fragmentation occurred. Then, you could calculate the angular momentum at separation that would result in the observed relative horizontal offset of the two fragments, and that might explain the large spin of the more obvious fragment. In other words, the angular momentum should prove consistent. That’s something else to check to get the full story. And of course there could have been other unnoticed fragments that carried away angular momentum.

    5. The spin of the rock will produce a magnus lift force in the direction that is the cross product of the velocity and the spin axis. Since the rock is tumbling (which is generally the case) rather than simply spinning on a fixed axis, the direction of the magnus lift is constantly changing, but it should have an average direction. I don’t know if it is significant or not, but sometimes it does turn out to be significant and can affect the trajectory and where the rock ultimately lands. Since you have an excellent model of the tumbling already I think you can immediately calculate the magnus lift and see how it affects the impact location.

    6. Here is a paper I published a few years ago on a similar topic. I don’t know if it will be helpful but maybe it will inspire some new ideas.
    http://arxiv.org/abs/0910.4357

    7. Have you looked in the tops of the trees to see if it became embedded in the wood high above the ground? If you use quad-copters you may be able to examine the upper trunk of every tree in the main target area. After all, if you have already scoured the ground then maybe the meteorite never made it to the ground. It would be tedious but it might be possible. Maybe you can inspire a group of young students in robotics clubs to come out and do that for you — making it a contest for them with the agreement you keep the meteorite and they get to have an amazing educational experience working with scientists. Maybe they get a cash prize and a trophy.

    8. Has anybody asked the question whether this type of rock at this terminal velocity would have shattered if it struck a tree? Maybe the tree would still show a scar from two years ago and you could look for smaller fragments around the base of the scarred tree.

    I wish you the best and I hope you find it soon!

    • Thanks a lot, Phil.

      1, 2 & 4. If you mean the other object visible just before the bigger object enters the frame, that’s Jon Vegar, the other skydiver who arrives 4 seconds later. So that’s no meteorite, but because of his distance we can at least rule out that the main object was dropped by him.

      3. I think you’re touching the core of the important discussion here, and it’s getting complicated. There are some variables here and unknowns, such as the distance and size, and Anders’ velocity and direction. Calculations also need to match what is seen in Anders’ second video catching the outbound path. We think it is the same object. By making all the videos available we hope that somebody is able to offer a convincing solution. Jon Vegar’s video might also offer some clues. It shows the parachute deployment from a distance, then Anders disappears for a few seconds to return into the frame again at the time the object passes.

      5. We’ve considered the Magnus effect as we see both the incoming and outgoing paths curve. However, the curve also seems easily explainable if Anders and the object travel in different directions. I think perhaps the object isn’t visible for long enough to detect such effect, but we’ve been aware of the possibility of this effect and taken that into account in the searches on the ground. We’ve also searched well outside the expected impact location since there could be more fragments.

      6. Thank you. Will a look.

      7 & 8. It’s something we’ve discussed and we have also looked for clues in the trees. I think it’s unlikely but possible that it has become stuck in a tree. There are mainly pine trees. One curious feature of the trees in this area is that many of them have a split trunk. I guess it’s possible, though unlikely, that it could embed itself in a split. Below an image showing Anders in the middle looking for clues in the trees. It also gives an idea of the terrain. Also note the soft, mossy ground, which could swallow an object. Below this soft layer the ground is mostly made up of rubble.

    • As to:

      3. To further rule out these being canopy debris, has anybody checked whether an object from the canopy would have reached terminal velocity so quickly?

      I did consider this. Assuming that a small stone from the canopy has not reached the terminal velocity, my size/velocity model would give the fit (at the smalles sizerenge ) for a somewhat bigger stone, but guite small anyway. And the further velocity increase during the 0.2 second time can not be recognized with this level model, because there is some freedom in the angles, and in the frames where the target is in the upper part, even a very small change in the angles makes a big change.

  8. Here is the final, published version of the paper I mentioned above.
    http://www.sciencedirect.com/science/article/pii/S0094576509005104

    By the way, three physicists were discussing this at our workplace and if there is anything specific you would like us to calculate, we may be able to find some time to work on it. We all three have extensive experience in a broad range of projects calculating trajectories of objects that were incidentally caught on video camera to find the objects and/or to figure out what the objects were. But we are all very busy and we would be doing this in our own time away from work, so I can’t make promises. If you have anything very specific and can give us the data so that we don’t need to spend a lot of time becoming smart on this problem, then we might be able to find time to help. I would very much like to help but I am so over-worked these days! Again, I wish you the best and I hope you find it soon!

  9. If a pebble accidentally packed with the parachute is released along with the parachute (this is before the parachute fully deploys), it will instantaniously have the same speed as the falling skydiver and reach maximum fall speed very quickly.

    What happens next depends on the surface-to-weight ratio of pebble and skydiver. If the surface of the pebble is relatively big in ratio to its weight (and for small objects, that often is the case, depending on density of the pebble: note that a pebble need not be very dense depending on the material: a lot of road gravel for instance is quite porous), the still falling skydiver (parachute released but not yet open) might actually fall faster than the pebble. This stops as soon as the parachute is fully deployed, and then the pebble will overtake the parachutist again. Because it was realeased with the parachute while the skydiver, released-but-not-yet-open parachute and the pebble were all three basically in free fall, it will have a velocity close to or on its maximum fall velocity and there will not be much notable acceleration of the pebble.

    To my mind, it is suspicious that the event happens about right after the parachute fully deploys, not somewhere halfway parachute descend or while still in free fall. This might be an indication that it has to do with the release and deployment of the parachute.

    • I think you’re making some fair points here. A puzzle still is a physical explanation for how the rock and Anders apparently are travelling in different directions. The pebble would then have to be pulled out of the pack in the initial deployment of the parachute obtaining a relatively low vertical speed while Anders is still falling much faster, then the parachute fully deploys and after a short while it overtakes him again at near terminal velocity, and at the same time Anders must done a wide swing to explain the differences in direction. However, the curved path most easily seen in the back camera strongly suggests that Anders is travelling faster to the north, the original direction before the parachute deployment, than the pebble. How then could the pebble have overtaken him in the northward direction, it rather seems that Anders is overtaking the pebble in this direction?

  10. In your analysis you assume that Anders has a velocity of 148 km/h at the time of the fly by. Is it realistic, given that his chute has been opened for a few seconds already? In the same way, the -37° horizontal angle looks odd to me.

      • I can try, though Anders is best qualified to comment on what’s realistic, and the figure you’re referring to is a fit done by Hans E F Amundsen, but I’ve done some similar exercises. I think the speed and direction of Anders is an area of high uncertainty. See my reply to Phil a bit below. And it’s a reason for concern, since it gives much freedom in picking values for our variables since the speed and direction of Anders can be adjusted quite a bit to fit which opens for many solutions. The angle in particular has puzzled me a bit, and that applies no matter the origin of the object assuming it has a near vertical drop. The only explanation I have is that Anders swings pretty hard, which has some support in the videos. That speed adds on top of the general speed and direction that he has. Before deployment the angle can be estimated looking at the back camera, since the pilot chute deployed first almost point at the sun. Anders is also speculating that he then fell through a bit because the pilot initially failed to release the main pack, so his vertical speed increased and he fell a bit steeper.

    • I checked the animation on the Skylights blog. What the arrow is pointing at is Jon Vegar, the other skydiver who jumped just after Anders. He’s still too far away to become more than a faint dot, but that faint dot is not falling towards the horizon as a rock should, and the direction is a perfect match for where he’s ought to be, so I think there can be no doubt about that dot.

  11. To provide a check of your efforts I ran terminal velocity calculations using a model of coefficient of drag that is similar to the one that I developed for the paper linked in the above comments. I used an atmosphere model for the 1000 meter altitude and looked up ground weather conditions for Oslo on that day as the input to the atmosphere model. I used Sutherland’s formula for the air’s viscosity. I got the velocity and size relationships with their distance from this website tather than taking measurements from the videos. I assumed Anders is going 100 km/hr. The resulting plot is here:

    https://www.facebook.com/photo.php?fbid=10203621851610903&set=a.3349266734524.2161210.1354520875&type=1&theater

    The distance to the rock is the horizontal axis, in meters. The vertical axis is velocity, in meters per second. The red curve is the terminal velocity of an object of the observed size when you assume these distances from Anders. The dashed line is the observed velocity assuming the same range of distances. They overlap or are close to overlapping at only two locations. One is from 3 to 4 meters distance, which means the object is 7.4 to 12.5 cm in diameter. The other is about 13 m, which means the object is about 40.7 cm in diameter. It looks to me that the midrange distances are not feasible and so the rock is probably not any of the intermediate sizes.

    One possibility that I am checking now is that the rock was traveling at faster than its terminal velocity. If the rock is very small (gravel-sized) and fell from the parachute pack, it would start out traveling at Anders’ velocity before his chute opened, which is probably faster than the rock’s terminal velocity. The rock will take some time to slow down, and may have still been faster than terminal velocity as it passed Anders, who has now rapidly decelerated due to the opening of the chute. It looks like the droque chute was deployed 3 seconds before the main chute opened, and the rock passed by another 9 seconds after that. That would mean that Anders had gained a lot of distance in those 9 to 12 seconds and that the rock did not slow down too terribly much in the same time. I’m not sure this is feasible but I’m checking. If this doesn’t work out then I am probably reaching the conclusion is is a meteorite. Until then I am staying skeptical because of the severe coincidence of the event, which I’m sure you understand :) I’ll let you know these results as soon as they are done.

    • Thanks again, Phil. Very interesting.

      Before fully digesting the details, I’ll just rush in to say that the assumed 100 km/h vertical speed of Anders is an upper limit. We hold 60-80 km/h as the more likely range.

      His exact speed and direction has long seemed to be an important clue in order to understand what is happening and what we see, but the answer has been elusive. In order to shed light on the problem, last year Anders even made similar jumps with a lot of sensors attached: GPS, accelerometers, gyros and a pressure sensor logging at 25 Hz. However, we didn’t get much closer. The readings were a bit all over the place, suggesting that the sensors weren’t sufficiently accurate at small timescales, and every jump appeard to be different, consistent with what Anders is experiencing. I’ve put a visualisation of the data on YouTube of one of these jumps. I think we have to live with some uncertainty here.

      I think the big size (40cm) can be ruled out. That should have been visible on Jon Vegar’s camera, and it doesn’t match the sudden appearance of the object in the back camera. Further problems with that size includes, although perhaps not conclusive, that it would likely be found and it would have required a big fireball event which would likely have been noted.

      • When comparing Phil’s results to my own (of similar type in principle), I noticed the same thing than Steinar about the vertical velocity. And if you make the Anders vertical speed smaller, then the red line and the dashed line will cross at the still smaler side (for the smaller solution). And that would make the exact solution quite indetermine, down to about one cm in diameter. I also agree that the bigger side solution can be ruled out.

  12. I ran ballistics calculations for a piece of debris falling from the parachute pack at the time of the release of the drogue chute. If the debris was 3.3 cm in diameter then it will at first fall behind Anders, but then after Anders’ main chute opens it will pass him, 12 seconds after the drogue chute was opened, as seen in the video. Thus, to be consistent with the timing of the video, a piece of debris would have to be 3.3 cm in diameter. This is a very crude calculation and does not account for horizontal motion. It is only intended to show plausibility that the rock is a small piece of gravel from the parachute pack.

    The velocity of this gravel as it flew past Anders would be consistent with the velocity observed in the video if it passed by him very closely — at just more than a meter from the camera. However, I did not look at velocities along the entire trajectory in the video; I only used velocity values that were calculated through proportionality with distance from the camera, using the values you reported on the website. I do not know if the nonlinearities of the perspective will make my argument invalid or now.

    The graphs are on my facebook page, here:
    https://www.facebook.com/Philtill/posts/10203622350143366?comment_id=8404678&offset=0&total_comments=2&notif_t=feed_comment

    My calculations are also consistent with it being a meteor at a distance of 12 to 18 meters. However, considering the timing — just after parachute opening — and the relative probabilities of debris in the parachute pack versus experiencing a personal meteor flyby, I am leaning toward believing this is parachute debris: a small piece of gravel that flew very closely past the camera. This may be disproved by other factors I have not considered. As noted above, a more detailed measurement of the trajectory may not be consistent with it passing within 1 meter of the camera. I don’t know if it is or isn’t. Also, the focus of the camera might tell us something about the distance — or it might not.

    I have done way too much on this since I had no time really to spare. I still wish you eventually prove to the world it is a meteorite and that you find it! That would be amazing.

    All the best!

  13. …so can they find a very similar sized rock and put it in the parachute and then watch it pop out of the pack and see what happens (and how close this may match what did happen)…?

    Love it if they could find a similar shaped rock and drop it from space and confirm that Yes, it picks a definite angle and does NOT spin as shown in the video. Of course a rock that just started to fall and not yet reached terminal velocity would be MUCH more likely to spin or tumble. I just think that is your biggest clue… if from space, that rock would have a very, long time to orient itself vs. a rock that came from the pack would still be likely to tumble. That and a rock from space may have easily had a starting speed of 160,000 mph to 45,000 mph. so it could be really really (glowing) hot by the time it reached the skydiver’s height in the sky… no vapor trails… no smoke trails, so again, this points to simple rock in the pack (btw was there confirmation that skydiver packed this shoot on the ground, or in his house?) I am sure every skydiver is likely to let the parachute drag on the ground when they land and hundreds of small folds where a small rock like that could easily slip in. I think for fun you should study a few thousand small metorites that have been found, that have a fusion crust and see how they orientate based on shapes and why you never seen fusion crusts on 3-4 sides… only 1. Sure a meteorite could explode down low and then the smaller pieces are not going fast, or have time to create a fusion crust as larger piece took all the heat upon entry, but if this small piece came solo all the way from space, it would be very unlikely to be spinning at all and this one spins very fast from the pics/video. I think the spinning rock is your 300 lbs straw that breaks the camel’s back.

  14. I agree that the spin might be a clue that it was recently torqued in a way a meteorite cannot have been — but I am not sure yet. I’m not sure…is it true that there is nothing that will naturally spin a meteorite as it falls? For example, air flow is nonlinear and if an object tumbles slightly that might set up nonlinear air dynamics that enhance the oscillation instead of dampening it. Also, if the object is ablating in one area more than others then as it rotates the ablation will occur preferrentially in only some parts of its rotation, and that might amplify rather than dampen the rotation. So I’m not sure. I’m not a meteorite expert. Perhaps they already know the answer to that.

  15. If the meteor was a pebble packed in with the parachute it would have been falling at the same speed as Anders then carried on at the same speed when Anders was first pulled up past the pebble by the chute then slowed down. So it wouldnt have overtaken him after opening the chute.

    • This was more or less the thinking that we were stuck in as well. But when the chute is released, first the wind will catch it and it will fly off Anders’ back until the cords hold it back and pulls it with Anders. The pebble, however, might be loose and not kept back by the cords. It will then stay above the chute until it catches up again.

      This is a possibility, and such a possibility trumps the meteorite hypothesis when there is no strong evidence against the object being relatively small.

Comments are closed.