3 ms·

It's not an unreasonable distance for a refraction mirage, however. My math reached the same number as yours, but it does assume the light goes in a straight l

• agf 5y ago
  The 170km number given in another comment (and by the calculators) is meant to take into account "normal" light-bending; the distance here is 250km. While I'm not an expert, that seems like a lot of extra distance / angle to make up.
  • someguydave 5y ago
    you can use shortwave radios to communicate 10,000 miles using ionospheric refraction. It is reasonable to think there are similar refraction phenomena in the troposphere, probably associated with thermal inversions