What limits the precision of ground-based stellar photometry and positional measurements?

Abstract

Nobody has reached the photometric precision better than 0.001 of a magnitude with ground-based telescopes. At the same time, we should spend barely some seconds of time to detect a million photons from any bright star and to attain a Poisson noise of one millimagnitude. The instrumental accuracy of coordinate determination with the ground-based telescopes equipped with CCD cameras is also no better than 0.1–0.2 arcsec. Many factors limiting the precision of stellar photometry and astrometry on the ground are superimposed and occur simultaneously. We show, however, that image motions in the focal plane of telescopes, recently discovered by the authors, may be in reality the barrier both to millimagnitude photometry and milliarcsecond astrometry with ground-based instruments. It is quite essential that stellar image motions (SIM) have a chance to be found out only by using synchronous observations with several telescopes. The measurements of SIM were carried out by taking registration of a star image near to diaphragm edge, which plays a role of an optical knife. The spectra of SIM show periodic variations on the scale from a few seconds to minutes typically having amplitudes of a few tenths of arc seconds. Clearly, image motions during the integration time can cause photometric errors and frustrate exact coordinate determinations. The nature of SIM remains obscure. An important point is that image displacements measured synchronously in different telescope reference frames show significant correlations. Emphasis is given to problems of the detecting of SIM and estimating their amplitude-frequency characteristics with the Synchronous Network of Telescopes.