Another problem about satellites:
See the following image:
http://www.astra-aps.de/worldskies/satellites/footprints/NSS-5-340E-Global.jpgIt is the footprint (area of reception) of NSS-5, located @ 20 degrees West, supposedly above the Equator. The picture uses a flat projection (Mercator?) of an assumed spherical Earth. Due to the projection, the boundary looks like a "squircle" - the same image projected on a sphere would show the boundary as a true circle. Projecting it on the disc map commonly used on this forum, yields a completely nondescript shape.
This particular radio source (satellite? tower?) radiates nearly omnidirectionally, so if from a point on Earth, you have a line of sight to it, then a big-enough dish should allow you to receive the signal. Hence the distinctive perimeter shape (to which all other supposed satellite signals tend to converge - often, a more directional beam, when mapped, will look like it is "cut off" by this shape) - at the perimeter, the radio source can be seen touching the horizon; and crossing the perimeter causes it to drop below the horizon, preventing further reception.
My question is: Why is the shape of the perimeter (the curve at which the radio source appears to touch the horizon) shaped the way it is?
If we assume that the Earth is round, then we can easily explain it as an object in geosynchronous orbit (a satellite) emitting the signal. Any point-like object that emits radiation isotropically will always light up an area on any nearby sphere that will be bounded by a circle. The relative radius of the bounding circle is a function of the radius of the sphere, and the distance between the emitter and the center of the sphere.
However, on a flat Earth, we get no such model. Instead, translating the above image onto the disc map will yield a footprint with a perimeter that is impossible to achieve with a single emitter. (Remember, it is not enough to just lose signal exactly at the perimeter - the cause of the signal loss must be the radio source going below the horizon, for the theory to be consistent with observation.) The idea of multiple emitters may come up, but it is easily disproven by the fact that there are no reports of geographical fluctuations/discontinuities in the received signal strength of this source.
(On top of that, I suspect that not even multiple emitters can yield the above footprint on a flat disk.)
EDIT: Here is another radio source showing the same footprint:
http://staging.satbeams.com/footprints?beam=6321(The map projection is different, but it is the same basic footprint.)
Unlike the other one, this "satellite" broadcasts a continuous DVB-T transmission of Benin's Canal 3 TV channel, so it should be easier to confirm.