Measurements of mirror seeing in the laboratory and at the telescope

We present a new technique for determining mirror seeing: this method is unobtrusive, and could be employed on a working telescope. We describe a series of experiments in the laboratory to determine how the correlation of thermal fluctuations between two points above a warm surface varies with separation. These fluctuations have been measured directly using an array of fast thermistors. By considering the attendant variations in refractive index of the air, we show how this relation, the structure function, may be used to estimate the Strehl ratio and the point spread function. We have also measured the structure function of the refractive index fluctuations directly by interferometric means, and compared the results from the two methods. We find that the fluctuations, and the derived phase structure function and consequent image broadening, are strongly influenced by the environment of the mirror. When care is taken to minimize all effects such as draughts, we find that the phase structure function is much flatter than the usual Kolmogorov 5/3 power law: indeed, there is strong evidence for a distinct outer scale. This will significantly reduce the importance of mirror seeing in the optical and, particularly, in the infrared. Extrapolating our results to a 10-m mirror, we find that the image diameter including 50 per cent of the optical energy increases by approximately 0.13 arcsec per K of excess temperature. The thermal sensors were then deployed on the William Herschel Telescope to measure the locally introduced thermal seeing when the primary mirror was not in thermal equilibrium with its immediate surroundings. Again we found that the environment of the telescope had a marked effect. However, during the time we were at the telescope the temperature of the primary mirror was close to ambient, or slightly lower, which resulted in negligible mirror seeing.