Could you link to sources so I can read some more, I'm a little confused.
I don't have any sources at the moment, but basically light follows "straight" lines through space time, where "straight" means that it follows the curvature of space time.
In flat space this is easy to visualise, just being a straight line.
In non-flat space, it is harder.
Using polar space, where coordinates are typically specified by (r,theta), where space curves in 1 of the 2 dimensions, there are many straight lines through one point (just like normal space).
In one case, where theta remains constant and thus you are travelling along a path through space which doesn't curve. As such, this "straight" line will be a straight line.
In another case, you project the light in a direction perpendicular to the first. In this case you start off purely travelling in the theta direction, with no r component at all. This is where the effect of curvature may be considered to be the greatest (for a given r anyway) and as a result the "straight" line follows the curvature of space, producing a circle. One such circle would be akin to the surface of your "flat" Earth.
For angles in between there will be a vertical component and a horizontal component. The horizontal component is still effected by the curvature of space and thus results in the "straight" line curving, producing a spiral.
If you have dr/dtheta be constant, then it would be a linear spiral (r=k*theta, dr/dtheta=k), but that would mean it is at different angles to vertical as you go out.
If instead you keep that angle constant it follows an exponential spiral.
That could also be stated as keeping dy/dx constant where y is the vertical component and x is the horizontal component (at that instant), where you have r=e^(k*theta), dr/dtheta=k*e^(k*theta)=k*r, r=y, theta=x/r, dr/dy=1 thus dr=dy, dtheta/dx=1/r thus dtheta=dx/r, from before dr/dtheta=k*r=dy/(dx/r)=r*dy/dx thus dy/dx=k).
Perhaps a better question would be how are you determining the curvature of space?
I also found multiple computer programs which are used to calculate scattering of light in the upper atmosphere which rely on spherical geometric projections of the Earth - surely these couldn't work if light behaves abnormally in spherical space?
They don't actually use projections of Earth, instead they treat Earth as a sphere.
Some papers can have misleading titles.
Once again, in spherical space there is no difference in height.
But there is, not without bendy light.
In normal space you can consider it as 2 contributions, one is from the light moving sideways and moving up due to the angle of projection, and the other is due to the curvature of Earth.
In spherical space, you lose that second contribution and thus get a different height.
Overall, I think there's been a misinterpretation here. I do not mean to claim that the entire universe is in spherical space, I wouldn't have the faintest how that would work
I think it would be a bigger issue the other way around.
Having only Earth in spherical space would require it to be a pocket of spherical space in normal space. How do the 2 connect? What causes it?
If it only curves in 2D, then if you get far enough away from Earth the curvature becomes insignificant. We still don't know if the universe is truly flat or the 3D surface of a very large 4D sphere.
Could you elaborate on your conceptualisation of my theory?
Basically any point can be described by 3 variables:
r, theta, phi
You can consider it as an infinite number of spherical shells where the surface of Earth is one such shell.