Could The Gimbal Video Show an Atlas V Launch?

Just to answer the question about visibility of a rocket launch from Jacksonville operating areas, rocket launches at night are easily visible from Jacksonville. I grew up there and saw numerous night launches in the 1970s. Some of them were quite spectacular. There were even more readily visible from the beach, where you did not have problems with trees or other obstructions. It is an interesting theory (atlas rocket launch) but I think the only way for this to be the source would have been if the rocket was downrange at the point the video was recorded. The angular motion would have been significantly reduced at that point.
 
I just realized that I was in Cocoa Beach that night for the launch. I had taken some vacation time to visit family and we went to see friends in Merritt island. It went on a ESE trajectory as I had previously stated. IMG_0282p.jpg
 
Wow, @Tim Printy , would love to see a launch someday. My parents used to live in the Bahamas from 68-72. My father used to watch the Apollo launches on TV and then after 3 minutes into the flight he'd go outside and watch them fly overhead. Stories like that have fed my interest in space & satellite observing (and correctly identifying the weird things we see). Anyway... here's some of my thoughts on this now
It is an interesting theory (atlas rocket launch) but I think the only way for this to be the source would have been if the rocket was downrange at the point the video was recorded. The angular motion would have been significantly reduced at that point.
I agree, the gimbal object appears stationary. It would need to have been recorded at a time when when the launch vehicle was approaching the horizon, which would have been when it was significantly downrange,

I see from your post #41
Some points here:
The launch of the rocket was at 0104 UTC on the 21st, which means 2004 EST on the 20th.
The inclination of the launch was 16.7 degrees. You can calculate the launch azimuth by using the latitude for Cape Canaveral (28.4 degrees) and the desired inclination. A rough calculation put this at an azimuth of about 95 degrees, which is towards the ESE away from the Jax op areas. While it is an interesting theory, one would need the actual time of the video, and the plane would have to have been facing towards the South or Southeast for the rocket theory to be plausible.
From https://forum.nasaspaceflight.com/index.php?topic=35158.160 page 9 we get the following image from the launch broadcast. The resolution isn't great, but it does show the ground track of the Atlas V on an azimuth heading SE, and it confirms your data. Like you said, as time after launch increased the angular motion would reduce for an observer from an F-18 in the JAX OPAREA.

1637923141178.png

Also @Leonardo Cuellar said in post... #266
Thank you for providing me with this information that I did not know about. If the fighters were 300NM from the coast, they were evidently further east of the carrier group. According to Graves, they observed GIMBAL when they decided to turn around to return to the USS ROOSEVELT. This would suggest that the direction in which they were looking at the object was towards the coast.
To me, this suggest that they were heading towards the coast when they first saw the object - but we know from the gimbal video that they pointed the ATFLIR 54° left at the start and changes to 7° right as Graves does the stern conversion manoeuver - which suggests that the object could have been in a more southerly (or south easterly) direction. This makes my understanding as follows: (Base Map taken from post #273)

1637926041926.png

So, on the surface, it all seems to fit. But the problem is confirming exactly where the F-18 was and exactly when the video was recorded. Not really sure how to do that.
 
Last edited:
I agree, the gimbal object appears stationary. It would need to have been recorded at a time when when the launch vehicle was approaching the horizon, which would have bben when it was significantly downrange,
Since my hypothesis was articulated through the narrative of the witnesses, I cannot take into consideration what Ryan Graves said about the first contact with ATFLIR, which was within 10NM. This would definitively cancel the Atlas V hypothesis. However, this datum would create many problems for the geometric and environmental evaluation of the Gimbal Video. If it really had been a stationary object at that distance, it would not only have been a 10 meter long object (initially a slant size for a 0.35 ° FOV), but with a speed of only 120kt. The elevation of the object might have also undergone a change as the fighter was approaching the object. And this doesn't appear. And looking down would have placed the object's altitude about 2000ft lower. And at this altitude it seems even more complicated to find a 120kt westerly wind in that area throughout January 21, 2015.
I must admit that the situation is very complex.
 
@Leonardo Cuellar It is indeed very complex with lots of data points missing. The problem we all have is making assumptions and deductions upon this missing data, which makes any conclusions weak and lacking in provenance.

To me - the weirdness of the object's movement seems to be due to a mix of 2-dimensional infra red imagery and parallax effect.

As yet, I am not totally convinced, but the data I've seen - and particularly the time correlation - are consistent with the rocket theory. I'll look into your comments above later and see if they change my mind.
 
@Leonardo Cuellar It is indeed very complex with lots of data points missing. The problem we all have is making assumptions and deductions upon this missing data, which makes any conclusions weak and lacking in provenance.
I have a feeling that some data wouldn't meet our expectations. I am referring in particular to elevation. In my opinion, when the ATFLIR fails to lock onto the object, because it's too far away or not in strong contrast with the background, it no longer gives accurate polar position information on the object itself, but more on a field of view where it points. I must say that the movement of the clouds also leaves many interpretations. As identified by @Mendel, we can see that these stop, in their apparent parallax motion, at the end of the video. There is a lot of uncertainty about this interpretation, which needs to be deepened.
 
If it really had been a stationary object at that distance, it would not only have been a 10 meter long object (initially a slant size for a 0.35 ° FOV), but with a speed of only 120kt. The elevation of the object might have also undergone a change as the fighter was approaching the object. And this doesn't appear
You're assuming that the "stop", which Graves reports as visible on the radar, is also visible in the public short infrared clip, and there's no evidence of that.

I'm not pursuing this further because I've convinced myself that it's not the Atlas, because
a) cloud parallax is wrong for a faraway UAP,
b) it's a daytime video,
c) the Navy would've identified already from the data they have.
Plus what Graves said.
 
a) cloud parallax is wrong for a faraway UAP,
b) it's a daytime video,
c) the Navy would've identified already from the data they have.

If you've already explained these points in the thread, please forgive me, but...

a) hmm, the cloud parallax looks right to me for a far away object and closer clouds. Why do you think it is wrong?
b) how have you determined the IR video is daytime?
c) possibly.
 
If you've already explained these points in the thread, please forgive me, but...

a) hmm, the cloud parallax looks right to me for a far away object and closer clouds. Why do you think it is wrong?
b) how have you determined the IR video is daytime?
c) possibly.
a) If you look out the left window of your car (or a train) at a distant mountain, the mountain seems to not move, but the scenery in front moves right-to-left. The scenery in Gimbal moves left-to-right.
So, either the UAP is close, or it is moving very quickly to the left, but in that case, how can the F-18 get behind it so quickly (needs to fly a very big circle segment)?

b) The clouds have shadows in a way that makes me think they're lit from above, but during the launch, there was neither sun nor moon out. If we're looking at the IR emitted by the clouds themselves, we'd be looking at a big light and shouldn't see those kinds of shadows on them.
 
b) The clouds have shadows in a way that makes me think they're lit from above, but during the launch, there was neither sun nor moon out. If we're looking at the IR emitted by the clouds themselves, we'd be looking at a big light and shouldn't see those kinds of shadows on them.
But in the white-hot segment, the tops of the clouds are black, which would be expected if they're losing heat to space, as happens at night. If they were lit from above, I'd think the tops of the clouds reflecting the heat of the Sun would be lighter in white-hot mode. The "shadows" just seem to be contrast between the cold tops and the warmer sides. They look more like shadows in the black-hot segment, where the valleys between clouds are darker, and therefore warmer. (It's possible the Sun was on the right side of the aircraft and shining at a low angle.)

Screen Shot 2021-11-26 at 2.41.03 PM.png
 
Last edited:
a) If you look out the left window of your car (or a train) at a distant mountain, the mountain seems to not move, but the scenery in front moves right-to-left. The scenery in Gimbal moves left-to-right.
So, either the UAP is close, or it is moving very quickly to the left, but in that case, how can the F-18 get behind it so quickly (needs to fly a very big circle segment)?

Agreed. But in both those examples the vehicle is moving in a straight line. The the gimbal video the aircraft is moving in an arc (part of a circle with a central point of rotation) as the pilot tries to get behind the object. Depending upon which side of the circle the clouds in frame are, they will appear to move in different directions.
iMarkup_20211126_231400.jpg
(Drawing not to scale, obviously.)
 
Last edited:
a) If you look out the left window of your car (or a train) at a distant mountain, the mountain seems to not move, but the scenery in front moves right-to-left. The scenery in Gimbal moves left-to-right.
So, either the UAP is close, or it is moving very quickly to the left, but in that case, how can the F-18 get behind it so quickly (needs to fly a very big circle segment)?
I agree this seems to be one of the the biggest obstacles for the Atlas hypothesis.

Agreed. But in both those examples the vehicle is moving in a straight line. The the gimbal video the aircraft is moving in an arc (part of a circle with a central point of rotation) as the pilot tries to get behind the object. Depending upon which side of the circle the clouds in frame are, they will appear to move in different directions.
This is wrong, as the lines of sight all focus on the UAP, Anything in front of it will appear to move to the left unless it has sufficient leftward motion to stay ahead of the parallax.

We are looking left of the plane's heading, so the camera is moving right. The clouds move right, so they are behind the object (unless, as noted, it's moving very fast to the left)
 
This is wrong, as the lines of sight all focus on the UAP,
I think that's right for objects that are within a certain range. However as the object (in the atlas hypothesis) is effectively at infinity, doesn't that make all the lines of sight effectively parallel..? (Or at best they have a very small convergent angle).

@Mick West
We are looking left of the plane's heading, so the camera is moving right. The clouds move right, so they are behind the object (unless, as noted, it's moving very fast to the left)
I disagree. The plane's heading is constantly changing as it turns. The camera is pointing at a position beyond the turning point (centre of turning circle) which will cause the nearby layer of stratus clouds (within the FoV) to appear to approach the nose of the aircraft as it turns towards the object at infinity. The nose is moving right to left, so the relative motion of the clouds will be left to right. This illusion is compounded by the narrow field of view of the NAR IR camera. The rotational movement is lost to the video observer and the brain thinks the clouds are moving linearly.

I think.

Edits: typos
 
But the camera's heading isn't, it stays pointed at the UAP.
And because of that, the apparent motion of a far distant UAP through near clouds would be left-to-right:
But the video shows the opposite:

Hmm. I'll have to think about that. (this is melting my brain!)
 
Apart from the discussion about the headings and motions etc, what about the fact the object does not show any (heat) tail, which I would assume would be visible knowing the incredibly powerful rocket motor's exhaust-trail? I cannot imagine the heated gasses coming from the rocket instantly cool down to ambient temps..

Just a thought.
 
Apart from the discussion about the headings and motions etc, what about the fact the object does not show any (heat) tail, which I would assume would be visible knowing the incredibly powerful rocket motor's exhaust-trail? I cannot imagine the heated gasses coming from the rocket instantly cool down to ambient temps..

Just a thought.
I'd thought this too about the heat trail - but I think it could be that the 'object' we see is the rocket exhaust. Maybe we cant actually see the rocket body due to the glare from the rocket motors. It kinds reminds me of this video which is said to show a saucer crash at White Sands missile test range. In fact, the 'saucer' that we see is the plume from a missile being test fired. It just looks like a silvery saucer.
 
I'd thought this too about the heat trail - but I think it could be that the 'object' we see is the rocket exhaust. Maybe we cant actually see the rocket body due to the glare from the rocket motors. It kinds reminds me of this video which is said to show a saucer crash at White Sands missile test range. In fact, the 'saucer' that we see is the plume from a missile being test fired. It just looks like a silvery saucer.
Yes, but there is clearly a trail/tail/path. This is not seen in the gimbal video. If there is one specific thing with rockets, it is the massive exhaust..
 
Last edited:
Hmm. I'll have to think about that. (this is melting my brain!)
In your sketch, when the plane is at the top of the circle, with the camera looking straight ahead, the parallax motion stops, and then starts the other way as the camera turns to the right side of the aircraft; and that continues to the bottom of the circle where it reverses back.

If you fixate a closer object, you get crossing sight lines, and then the background moves the other way, but for a distant rocket, it wouldn't work like that.
(Mauro does assume a close Atlas V with the F-18 pitched up a lot; I expect that's why he keeps disagreeing with my posts. I'm explicitly talking about a different situation, though.)
 
Yes, but there is clearly a trail/tail/path. This is not seen in the gimbal video. If there is one specific thing with rockets, it is the massive exhaust..
But we aren't seeing the object (which in this case would be the rocket plus its exhaust flume), we are seeing local glare at the camera. The apparent size of a distant, very hot object could be a small fraction of the apparent size of the local glare, in which case there would be no shape to the flume, just a blob surrounding it.

Like others here I'm moving in the direction of thinking that Atlas is unlikely. A possibility still remaining is that one of the critical data is wrong — for example, if someone made a mistake in reporting that the sighting was in the Jacksonville operating area, when it was actually much farther north. I estimated way back that an Atlas would have to be some 700 NM away and at an altitude of at least 28 NM for these observations. But this, too, presents other problems.

In an unrelated note, I don't think anyone has done a 3D analysis based on replicating the cloud movement. I'd like to give it a shot. I believe we could get an estimate of the distance to the clouds, and that estimate could be checked based on the scale of their largest features, but I'll have to do some research for that. And from there, distance to the object.
---
Edit: Just getting started on the Blender. The reference cube at right is 10 statute miles across. The sphere below it is the Earth. The gray layer (a segment of a sphere) is stratocumulus cloud cover that tops out at 7,000 feet. On the z axis (blue vertical line) is our aircraft at 25,000 feet, with a 2° sightline (orange). Notice that the sightline never intercepts the clouds, falsifying an earlier claim that the sightline would be looking at the ground. The sightline's closest approach to the cloud bank is at a range of about 130 statute miles. So, assuming the cloud bank goes past the horizon, that should be roughly the distance to the clouds in the video. However, this model makes me question that assumption, since the video doesn't show clouds receding deep into the distance. They just end.

Screen Shot 2021-11-27 at 2.11.08 PM.png

Here's a zoomed-out look that goes to 700 statute miles. At that range, the sightline has an altitude of some 190,000 feet (~31 NM).

Screen Shot 2021-11-27 at 2.11.45 PM.png
 
Last edited:
@Edward Current, the clouds may be much higher than 7000 ft, if you look at my other thread on the meteorological data on Jan 20-21, there were areas of high clouds, up to 5000m or more. I think these clouds may in fact be the top of cumulonimbus.
 
We could be in a situation like this, with high clouds (5000m) being at 40-50 NM.

Screenshot_20211128-003106-140~2.png
With clouds stopping before the line of sight gets under the clouds elevation. What intrigues me is that the FOV of the ATFLIR being so narrow, what are the chances it captures the cloudlines, and not just full sky or full clouds? But that what we have here.
 
My post was not related to the rocket, I was just replying to Edward because I think it's a good idea to make a reconstruction using the clouds information.
 
Hmm. I'll have to think about that. (this is melting my brain!)
It's very difficult to think about - I constantly have to stop and think about which way things are moving, and usually with a "close one eye and move your head" test.

Three points I find helpful to keep in mind are:

1) The camera is always pointing at the object, because it's locked on to it.
2) Distant objects move WITH the camera (relative to closer objects)
3) Parallax is entirely from relative positions, not view angles (rotating the camera moves everything equally)

Unfortunate none of these are themselves super intuitive things either - best to try to set up a practical recreation of the scenario, or at the very least draw all the lines of sight in your diagrams.

#2 implies a static scene, but if an object is moving, then it's still useful, as the parallax (apparent) motion is added to actual motion. Thus a plane landing in the opposite direction to which you are driving can have it's motion cancelled out by parallax, relative to nearby trees.

In the context of Gimbal we see the clouds move right and we are looking (initially) left 54°. The clouds are moving WITH the camera which implies (point #2) the clouds are more distant than the object.

2021-11-28_03-50-13.jpg

For the object to be far behind the clouds, it would have to be moving left fast enough to overcome the rightward parallax effect (which would be WITH the camera). A strong point against this is that as the camera's rightward motion decreases (as it turns to fly more directly towards the object), then the object's apparent motion relative to the cloud decreases, when parallax dictates it would increase.

I don't think this is possible with the Atlas hypothesis.

(Or course we also have the weird object scenario, where it's just flying over the clouds, then stopping and hovering - coincidentally as any parallax is small, but that's obviously not Atlas)
 
Understood.
I was confused because this is the rocket thread; I'd have expected non-rocket-related geometry in https://www.metabunk.org/threads/gi...mates-using-lines-of-bearing-and-or-dcs.11836 .
I posted the screenshots in the rocket thread because I think this is the first time we have a graphical representation of the constraints on the object's altitude. For all practical purposes, this seems to falsify the Atlas hypothesis except in the case that it's >500 miles away — and that's falsified by the direction that the clouds move.

@Edward Current, the clouds may be much higher than 7000 ft, if you look at my other thread on the meteorological data on Jan 20-21, there were areas of high clouds, up to 5000m or more. I think these clouds may in fact be the top of cumulonimbus.
They don't look like cumulonimbus. Cumulonimbus form in unstable air and are irregular and dynamic. Stratocumulus form in stable air, typically under an inversion layer that keeps their tops at a very consistent altitude, which is what we see in the video. Stratocumulus are also more common.

I looked at the meteorology thread, but I don't think we know enough about the aircraft's or the object's location to make guesses about the cloud cover that overrules the observation of a stable, consistent cloud cover in the video, which to me looks like run-of-the-mill stratocumulus.

Further, as you mentioned, the 0.35° FOV comes into play. Here are the widths of a 0.35° FOV at various ranges:
10 NM - 370 feet
20 NM - 740 feet
40 NM - 1,480 feet
80 NM - 2,960 feet
100 NM - 3,700 feet
120 NM - 4,440 feet
The larger features (the "puffs") of stratocumulus are ballpark on the order of 800 feet across. So, if these are stratocumulus, these figures seem to suggest they are at a range of something like 120 NM. Since the FOV is also ~4,400 feet tall at this range, this would also be consistent with clouds visible in the bottom one-third of the FOV.

Further modeling can tighten the constraints on the clouds, but I'll post that elsewhere.
 
Last edited:
They don't look like cumulonimbus. Cumulonimbus form in unstable air and are irregular and dynamic. Stratocumulus form in stable air, typically under an inversion layer that keeps their tops at a very consistent altitude, which is what we see in the video. Stratocumulus are also more common.
I agree they look like stratocumulus. But the meteorological data shows there were high clouds in the area too at this time, and I wonder how they would look from above. Searching for "altocumulus plane" in Google Images, I see clouds that do not look very different from what we see in the video. Maybe a pilot or a cloud expert could provide his opinion ?

The larger features (the "puffs") of stratocumulus are ballpark on the order of 800 feet across.
Just curious, where do you get this number from ?

So, if these are stratocumulus, these figures seem to suggest they are at a range of something like 120 NM. Since the FOV is also ~4,400 feet tall at this range, this would also be consistent with clouds visible in the bottom one-third of the FOV.
I agree the larger the FOV, the more likely it is to see the junction between clouds and sky.

Further modeling can tighten the constraints on the clouds, but I'll post that elsewhere.
Looking forward to see what you come up with !
 
I agree they look like stratocumulus. But the meteorological data shows there were high clouds in the area too at this time, and I wonder how they would look from above. Searching for "altocumulus plane" in Google Images, I see clouds that do not look very different from what we see in the video. Maybe a pilot or a cloud expert could provide his opinion ?
Stratocumulus (SC) and altocumulus (AC) are the same type of cloud, ie generated by turbulent mixing of humid layers of air just above the level of condensation. The difference is between the vertical region of the atmosphere where it forms. Looking at the satellite images they seem more AC. Usually quoted at 5-7km with thicknesses up to about 1 km.
 
Just curious, where do you get this number from ?
I searched and searched for a size estimate of stratocumulus clouds, to no avail. But several sources for amateurs and kids suggested to hold your hand up to the sky. If the puffs are the size of your fist, they are stratocumulus clouds; if they're more like the width of your thumb, they're altocumulus. I extrapolated 5" of apparent size out to a mile (roughly their altitude) to get the very ballpark figure of 800 feet.

Altocumulus (6,500–20,000 feet) are a good possibility. Much more likely than cumulonimbus. Although they tend to be more spread out, as if they were shaken apart, here's some altocumulus which if viewed at a low angle might look just like the clouds in the video:
110625578-altocumulus-cloud-by-day-may-be-used-as-background-.jpeg

I won't assume the altitude of the clouds. I'll start with the stratocumulus layer at 7,000 feet but will try a range of altitudes and see how they variously constrain the numbers.
 
Clouds are not the only information in the stitched version of the video. You can also see layers of hotter and colder air above the clouds and I do wonder if the darker, warmer background actually is the ocean:
Screenshot_2021-11-30-13-47-18-134~2.jpeg



Also note the concave shape of the stitch, which seems to indicate constant gimbal rotation.
 
Clouds are not the only information in the stitched version of the video. You can also see layers of hotter and colder air above the clouds and I do wonder if the darker, warmer background actually is the ocean:

At 25,000 feet, the horizon is looking down -2.8 degrees.
ATFLIR is looking down -2 degrees. Taking into account the inaccuracy in the display, it could be -2.49 (~-2.5) degrees.

The FOV for NAR 2X is 0.35 degrees. So the vertical image covers -2.5+-0.17 degrees (from -2.67 to -2.33). The bottom of the image is still looking above the ocean.
 
At 25,000 feet, the horizon is looking down -2.8 degrees.
ATFLIR is looking down -2 degrees. Taking into account the inaccuracy in the display, it could be -2.49 (~-2.5) degrees.

The FOV for NAR 2X is 0.35 degrees. So the vertical image covers -2.5+-0.17 degrees (from -2.67 to -2.33). The bottom of the image is still looking above the ocean.
At what angle would the ocean be visible? Because I doubt whether the angles indicated on the ATFLIR are as accurate as assumed in this calculation.
For targeting, a laser range finder is used that is actively kept optically aligned with the ATFLIR LOS, so the ATFLIR angles are not needed for targeting.
They may be used for slaving the ATFLIR to the radar, but that would work as long as both radar and ATFLIR have the same systematic error.
So I wonder if the angle at which the ocean would be seen still falls within a reasonable margin of error.
 
Ah, sorry I missed your first remark. It is 2.8 degrees then. So the ATFLIR only needs to be off by about 1 degree.
 
Also note the concave shape of the stitch, which seems to indicate constant gimbal rotation.
I'd say radial is perhaps a more accurate word than concave. I think that's just a result of the aircraft turning in a circle as the stitch covers about 40° of the horizon.1638285737619.png.
 
How would that work? If I stand on a tower and film the horizon with an extreme zoom lens while turning my body 40 degrees keeping the camera level, the stitched result would be flat, not concave.
The ATFLIR primary mirror in the gimbal makes a pendulum-like movement when the gimbal rotates. This would create the concave effect of the stitched video.
 
How would that work? If I stand on a tower and film the horizon with an extreme zoom lens while turning my body 40 degrees keeping the camera level, the stitched result would be flat, not concave.
The ATFLIR primary mirror in the gimbal makes a pendulum-like movement when the gimbal rotates. This would create the concave effect of the stitched video.
That's a good question. I guess its to do with, in the Gimbal example, the camera is actually moving in a circle at at the same time as rotating, whereas in your example the camera is rotating on the spot. (maybe)
 
When I simulate it in Blender, with a camera following a circular path tracking a distant sphere I also notice a fixed horizon object (plane representing the cloud layer) apparently move up and down from the camera's perspective.
 
At 25,000 feet, the horizon is looking down -2.8 degrees.
ATFLIR is looking down -2 degrees. Taking into account the inaccuracy in the display, it could be -2.49 (~-2.5) degrees.

The FOV for NAR 2X is 0.35 degrees. So the vertical image covers -2.5+-0.17 degrees (from -2.67 to -2.33). The bottom of the image is still looking above the ocean.
Thw horizon is -2.6⁰ refracted. And there's always some leeway over water. See https://www.metabunk.org/curve/
Also note the concave shape of the stitch, which seems to indicate constant gimbal rotation.
I believe that's an artifact of rectilinear perspective when the camera is angled down, it would bend the other way if it was angled up? The intersection of the camera viewport with a vertical wall is a trapeze.

And why is this discussion not in the other thread?
 
Back
Top