Corrected calculation of star trails caused by differential atmospheric refraction

Abstract

Differential atmospheric refraction does not cause a net rotation of a field of stars thus a correctly tracking equatorial telescope does not require rotation of the detector during observations. Ira S. Bowen was mistaken in 1966 in his calculation of star trails. In the publication of his lecture titled "Future Tools of the Astronomer" after being awarded the Gold Medal from the Royal Astronomical Society in 1966, on page 18, he said "At this point I should digress to make a few remarks about the ultimate limitation on the size of field we may hope to reach. For many purposes this will be set by differential refraction in the atmosphere". It shows trails for a ring of stars at a declination of 60 degrees for a site at 30 degrees latitude, for an 8-hour exposure over a 2 degree field diameter. The tangential component of the East and West trails is 3 seconds of arc (arcsec). The correct separation is 0.4 arcsec in the opposite direction which balances the 0.4 arcsec clockwise trails of the North and South stars. Also his 0.6 arcsec radial trail should be 1.1 arcsec. The source of the errors is in the application of spherical trigonometry.