Videos showing the convexity of water should account for refraction, providing information about temperature and humidity all over the filmed area. I don't want to be demanding, but in some cases refraction can be a real deal, allowing to light to travel in non-straight paths. In the video it is possible that refraction made the boat disappear.
In any experiment you want to take into account as many variables as you can, but in the real world you will never fully account for all of them. No matter how many spot measurements you take across the lake, it's always possible for you to have missed a patch of cold air, and the wind constantly blowing means you can't be sure things haven't changed.
So even if you took a boat back and forth with sensitive temperature measurements while filming it is still
possible that refraction could be causing it. No matter how careful you are you can't rule out the possibility.
But what you can rule out is how
likely it is. Anything is possible after all. It's possible the camera has a software bug that just happened to make the boat disappear but it's extremely unlikely.
How likely is it that refraction caused the boat and trees behind it to vanish exactly as the math indicated? Refraction certainly can cause light to bend that much, but how likely is it that on that day it refracted things exactly to match the predictions of a round Earth? This video also has the advantage of a nice background of trees instead of an empty blue sky. This would make any refraction effects very visible, you would see the trees distorting and changing shape. So for refraction to be in play here, it would have to also be extremely even and also smoothly change in altitude to duplicate the effects seen when raising and lowering the camera. Through the video you can also see the waves are sometimes high and sometimes lower, indicating the weather conditions changed, but the effects of raising the camera up and down did not.
That's a lot of coincidences required for refraction to be the cause.
It's not impossible of course, but when you add up everything required it is extremely good evidence against it.
The rest is literally just him moving the camera up and down at the edge of a lake. I can’t believe how much footage he’s included. But anyway...
I'm going to guess he included all the footage because it's so gorgeous. Plus, more evidence is always good. This is a great video, extremely well done and such amazing scenery.