Limits of resolution also play an important role.
If the resolution is only 8 seconds, the building will be blocked.
like in the video below.
Now that I have a chance to watch the video I can comment on it more (at it is a massive waste of time).
Notice the picture you show 10 seconds in?
The gap between the blocks is still quite easy to resolve, all the way up to 40.3 km away. You can easily just move the image of the first over to get a decent idea of what to expect:
You should easily be able to see the lower sections of the building. So why can't you?
Even in the latter ones when you can't resolve it, that is due to the atmosphere causing distortions. But again, that works equally for all sections, not magically more for the bottom. Neither explain why the angle is wrong, why the top of the building is lower than it should be.
You can also see the horizon is much closer as it is not significantly distorted.
If limited resolution or limited visibility through the atmosphere was going to be an issue, you wouldn't have that nice clear horizon. You would have a blur. It would look more like this:
No clear horizon, just a foggy blur until everything is lost.
You then decide to chose 8.25" for no reason other than it coming close to fitting your nonsense.
If you wanted to use the human eye you would want to use 1 minute of arc, no 8.25 seconds.
But they weren't, they were taken with a camera which clearly a very decent angular resolution.
You then just draw a line as if it matches, scaling the building to fit your nonsense while completely ignoring reality.
Notice that the function you are focusing on is linear?
That means twice as much should be obscured at twice the distance.
At 25 km, barely any is obscured. Your angle predicts roughly 1 m, yet almost the entire lower block is missing.
If you were to double this to 50 km, it should double, giving 2 m based upon your angle or ~2 blocks from the first observation.
But what do we see at a significantly lesser distance of 40 km (where less than the 50 km hidden should be hidden)?
We see ~4 blocks hidden.
Notice how that doesn't match your linear function at all? (Also notice how it doesn't match your claimed measured missing distance? So not only are you blatantly lying about how much should be obscured you are also blatantly lying about how much is hidden).
That shows that that is not the reason, and that you are just presenting the data however you please to pretend it works.
The lines you are drawing don't even make sense.
The blue line you plot is meant to be the angular size of the building. That means the red line should be horizontal, as it is the same angle consistently.
Quit with the BS.