Assuming a perfectly spherical Earth (not true) of radius r metres and an
observer whose eye is h metres off the ground, then the angle at the centre
of the Earth between where that observer would measure the horizon and the
observer, x can be summed up in the following
cos x = r / (r + h)
x = arccos [r/(r+h)] (where arccos function is in radians)
The curved distance, D from the point on the Earth directly below the
observer to the observer's horizon would then be given by
D = rx
D = r arccos [r/(r + h)]
The straight line distance, d, from the observer's eyes to the observer's
horizon is also given by (by Pythagoras' Theorem)
d = sqrt(2rh + h^2)
For small values of h, this simplifies to d = sqrt(2rh)
For d in miles and h and r in feet, this boils down to
d = 1.2252 * sqrt(h)
e.g. For an observer standing on the Earth (r = 6378km) with eyes at 1.75
metres (5' 9")
Curved distance, D = 4.725 km (2.936 miles)
The other formula gives the same distance for straight line distance.
Your formula gives 2.935 miles for the distance to the horizon - very good
approximation.
Try another example
An aircraft is flying at 10,000metres (32,800 ft)
My formula, D = 356.92 km (221.78 miles); d = 357.20km (222.01 miles)
Your formula gives d = 221.89 miles - still a good approximation - only 200
yards out in over 200 miles.
A further example.
An orbiting spacecraft at 800km (500 miles) above the surface of the Earth.
My formula, D = 3,039.9 km (1,888.9 miles); d = 3,293.1 km (2,046.3 miles)
Your formula gives d = 1,984.6 miles - approximation begins to break down
Taking it to the extreme.
From the Moon, how far is Earth's horizon.
My formula, D = 9,914km (6,160 miles) ~ 1/4 of Earth's circumference.
d = 390,326 km (242,537 miles) - not surprising - this is just over the
distance from Earth to the Moon.
Your formula gives d = 43,480 miles - very large error!
To sum up, yes, your formula is good enough for most heights that you choose
to calculate, but the approximation gets worse as you go higher up.
Ric
> To sum up, yes, your formula is good enough for most heights that you
> choose to calculate, but the approximation gets worse as you go higher
> up.
Hey, thanks for all of that! I'll start using YOUR formula instead.
Ric
I say yes.
--
Night Boar
(Portugal)