Weird. I did a little googling, and I came up with horizon distance formulas with the constant anywhere from 1.17 to 1.32 (I may have made a mistake with this last one, but I think that that is the equivalent). What I did was use 1.22--I derived it similarly* to how it's done on this wiki page where the formula is actually given as sqrt(1.5 * height in feet). They say "the Imperial one is actually quite precise"

*The square of the distance (straight line, not along surface, but that's close) to the tangent is equal to the height times (twice the radius plus the height, but the height is negligble in this sum)