I recorded the azimuth reading every 100 frames (3.3 seconds) and plotted it as a function of time:
As you can see, the lateral drift rate is not consistent. It's very low at the beginning, but accelerates towards the end of the video.
To quantify this acceleration, I fit a quadratic to the azimuth reading as a function of time. The fit was:
θ = -1.86e-3 * t^2 - 2.7e-2*t + 4.1
Taking the derivative, we get the angular speed:
ω = -3.7e-3 * t - 2.7e-2
Here is a plot of the angular speed we get from the equation. It increases from -0.05 deg/s at the beginning to -0.30 deg/s at the end (the negative sign means the UFO is drifting left instead of right). What could account for this acceleration?
One possibility is that the UFO got 6x closer. But if it gets 6x closer, the elevation will also get multiplied by 6, as will the apparent size of the object. None of that happens in the video. Another possibility is that the UFO turns to the left during the recording.
Let's take one specific scenario and run with it. Let's assume the UFO has an initial distance of 50 km, size of 25 m, and true airspeed of 250 m/s. It is also turning. At the beginning of the video, it is heading 10 degrees to the left of the fighter; by the end, it is heading 80 degrees to the left of the fighter, for a turn rate of 1 deg/second. At the beginning, it would have an apparent lateral drift rate of 250 * sin(10 deg) / 50,000 = 0.05 deg/s; by the end, it would have a lateral drift rate of roughly 250 * sin(80 deg) / 48,000 = 0.30 deg/s.
What about the closing rate? At the beginning, the UFO is actually running away from the fighter at 250*cos(10 deg) - 195 = 51 m/s. By the time the turn angle gets to 40 degrees, 30 seconds into the video, the fighter starts closing on the UFO. By the end of the video, the fighter is closing on the UFO at 150 m/s. In total, if you do the integration, the fighter closes on the UFO by 2 km--hardly anything.
Is this rate of turn realistic, or too fast? 1 degree per second is 1/3 of the
standard rate of turn. From that article:
A rate half turn (1.5° per second) is normally used when flying faster than 250 kn. The term rate two turn (6° per second) is used on some low speed aircraft.
So 1° per second is about how fast airplanes usually turn. We can also compute the centripetal acceleration of the UFO. The turning radius is R = v/omega = (250 m/s) / (pi/180 rad/s) = 14.3 km. The centripetal acceleration is v^2/R = 4.4 m/s^2, or about 0.4 g.
In this interpretation, the apparent increase in size of the UFO is not because it's getting closer, but because it's turning left and presenting a wider cross-section. This interpretation would also explain why the UFO looks completely different in IR mode at 2x zoom at 00:15 compared to 01:10:
Perhaps the IR image at 00:15 doesn't represent the outline of the UFO at all, but is instead the glare from the UFO's engines.
Lastly, I want to point out an error in my first post. I said that the UFO in the video was not detected on radar. At least according to the
TTSA report, this is not true:
According to the radar display, the initial tracks were approximately 30-40 nm to the south of the aircraft. LT [redacted] was controlling the radar and FLIR and attempted multiple times to transition the radar to Single Target Track (STT) mode on the object. The radar could not take a lock, the b-sweep would raster around the hit, build an initial aspect vector (which never stabilized) and then would drop and continue normal RWS b-sweep. When asked, LT [redacted] stated that there were no jamming cues (strobe, champagne bubbles, "any normal EA indications").
From
this Fightersweep article:
The WSO first picked up a contact on the radar around 30nm away while it was operating in the RWS scan mode.
So the UFO
did show on on radar; the radar just had difficulty tracking it. 30nm is 55 km. This is in the ballpark of the 50 km distance that I said was most probable based on the FLIR footage alone. If the UFO is turning, that could potentially make it harder to track on radar, but I don't know enough about radar to say this confidently.
Difficulties with this explanation:
- Earlier, I also said that this interpretation explains why the elevation indicator drops from 6 degrees to 5 degrees at 00:23--the UFO is pulling away from the fighter for the first 30 seconds. However, I realized it doesn't explain why the elevation doesn't increase again to 6 degrees for the rest of the video. Maybe I'm also doing what I said Mendel was doing, which is taking the elevation indicator too literally instead of allowing for a realistic amount of error.
- Why would the UFO be turning? Most airplanes don't turn for the vast majority of their flight. This is especially true for commercial airliners, which maintain a constant heading and speed for pretty much the entire flight. However, there is one place where all aircraft typically turn: near the airport where they take off or land. So, did this encounter take place near an airport?
What is the time and location of the encounter?
I tried to determine when and where the encounter took place as accurately as possible. According to the TTSA report:
The time was approximately 1500L. The flight checked with the E-2C and separated. Their jet proceeded south to the southern cap.
If you look at the video, there's a timer at the bottom that reads 41:35 at the beginning and 42:50 at the end. I assume this is the time since the beginning of the flight. If this is true, the encounter happened
November 14, 2004 at around 15:42 local time (plus or minus probably 20 minutes. I assume local time means Pacific. This is backed up by the
Nimitz deck logs, which say the time zone is "+8U", where U is the military time zone corresponding to Los Angeles.
Where did the encounter happen? Again according to the TTSA report, the first encounter happened at N31 20' W117 10'. The second encounter happened at the "southern CAP point". The
Fightersweep article says:
Though low on gas, the FASTEAGLE flight swung through the CAP position to try and pick up the AAV again. They saw nothing more on their way back to Nimitz.
This implies that the CAP point is roughly in between the Nimitz and the point where the first encounter happened. Since this is the southern cap, I assume the CAP is to the south of Nimitz, and the first encounter is to the south of the CAP.
In
this podcast with Lex Fridman, Fravor says (1:28:30):
'Hey sir, you're not going to believe this, but that thing's at your CAP.' [...] This is like, 60 miles away, and it just re-appears.
So the CAP is 60 miles to the north of N31 20' W117 10'. That would put it roughly here:
However, maybe you shouldn't trust this too much. The accounts by Fravor, Wingwoman, and the TTSA report appear inconsistent with respect to the location of the first encounter:
Fravor says (1:19:10 on that podcast):
And we're basically off the coast of...if you look at a map and go to San Diego and Ensenada, Mexico, we're kinda in between that and we're probably about 80, 100, I don't know, it's pretty far off the coast.
I labelled this point on the map above, as "Fravor: 80-100 mi west of SD/EN". Wingwoman's
account says:
“Source” and OK-1’s aircraft was approximately .3 NM behind OK-2 and OK-3’s aircraft and both F-18s proceeded East at an altitude of approximately 10,000 to 20,000 feet towards San Clemente Island.
This implies they were west of San Clemente Island and flew east to get to the first encounter. San Clemente Island is on the map, so this would be on the line between "Nimitz?" and "60 mi east of Nimitz?" I plopped the Nimitz pin where it is because Wingwoman's account also says:
On 14 November 2004, the U.S.S. Nimitz Battle Group was conducting a training mission in U.S. Navy Operating Area ‘Whiskey’, approximately 80 nautical miles (NM) West from the coast of San Diego.
Finally, the TTSA report says the first encounter happened at N31 20' W117 10'. This is labelled on the map as "First encounter".
So I'm not really sure where the first encounter took place, since the three accounts seem contradictory. This would put into question my reconstruction of where the second encounter (aka the southern CAP point) is located.
If my reconstruction is correct (a very big if), the southern CAP point would be within 40-50 km of San Diego airport, which has one of the busiest airways in the world.