Source: https://www.youtube.com/watch?v=KLufSkz-et0 If there were no atmosphere, and somehow the Earth still had oceans, then the curvature of those oceans would be readily apparent, with a pin-sharp horizon cutting off distant objects with mathematical precision. You'd be able to zoom in clearly on ships vanishing over the horizon, and calculate the radius of the earth with great precision. But back in the real world, we have an atmosphere. That atmosphere is generally denser as it gets lower down, and this makes the light bend down a bit, which lets you see beyond the mathematically calculated horizon. This is normally illustrated with a side view, with the camera on one side and the target object on the other, and a greatly exaggerated curve of the Earth, like this: This is where the explanation starts to get lost on people. This abstract side view is something that often causes confusion. So I'm seeing if I can get some kind of physical demonstration of what is going on. I've got a fishtank partly filled it with water, layered some sugar on the bottom so it would dissolve and create higher refractive index the lower down you went, and I've got a laser with a beam splitter: The idea is that from the side it's like the diagram. There's rays of light, they curve slightly, and there's the slight curve of the earth: The curve is slight, but there. More visible with some compression. At the other end of the tank, I put a photo of Toronto, and flipped the laser to a single beam and put it behind it. I then moved the camera up and down. This compression of the scene near the horizon is very similar to that observed in the Jenna Fredo video (@jenna1789), discussed in this thread: https://www.metabunk.org/views-of-t...rt-niagara-illustrate-earths-curvature.t8149/ And this setup can be duplicated in the refraction simulator, with a very cool lower layer, taking it from 80 feet to 1 foot above the water: https://www.metabunk.org/refraction/?~(profile~(~12.41~0~13.084~6.017~17.352~32.883~17.315~15.205~16.48~287.6~16.52875552~262.976)~useRefraction~true~useStandard~false~useFlat~false~useNarrow~false~useNight~false~showSideView~true~showSideGradient~true~useDebug~false~showEyeLevel~false~vFOV~0.022689280275926284~tilt~0.0008726646259971648~showEveryLines~10~viewerHeight~1~viewerOffset~773~minX~12~maxX~18~minY~-10~maxY~300~RH~50~wavelength~550~computedParams~false~name~'Toronto*20From*20Hamilton*20Beach~src~'Toronto*20From*20Hamilton*20Beach*2032*20miles.png~sourceURL~'https*3a*2f*2fwww.youtube.com*2fwatch*3fv*3dAFJnrMZT7KA~targets~(~(distance~168960~height~1815~name~'Toronto~multiple~0~gap~0)))_ So, a good practical results. But I wonder how best to communicate it to people. I think the key concept is how the laser line is the same as a line of sight. If you shine a laser in a particular direction it will hit a spot. If you look in exactly the same direction your were shining the laser, and from the same position, then your line of sight will follow that same (possibly curved) line, and you will be looking directly at the spot.