SCIDAR measurements at Pic du Midi

Abstract

We present SCIDAR measurements made at the Pic du Midi Observatory with PISCO. A series of vertical profiles of the refractive-index structure constant C 2 N (h) was obtained with the observations of the binary stars γ Del and γ Ari, with 30 second time sampling. In the free atmosphere, the turbulence was dominated by the contribution of a few well-separated atmospheric layers at altitudes in the range 8-20 km, with significant changes within a few minutes. We describe a new method based on the CLEAN algorithm applied on the 2-D covariance function of the irradiance fluctuations in the pupil plane and compare it with the widely-used Maximum Entropy Method applied on a slice of this 2-D function. Our method is much less sensitive to the noise. Vertical profiles of the wind velocity and direction were obtained with an automatic procedure working on the mean cross-correlation function of consecutive frames. Furthermore, we validate a model for the SCIDAR detectivity threshold in terms of distance to the turbulent layers.

First evidence of the finite horizontal extent of the optical turbulence layers. Implications for new adaptive optics techniques

“…We note that the differences in vertical C 2 N structure are observed, at least in the 21/22 night, near the ground and in the higher part of the atmosphere. Beside this, it has been shown that the statistical noise, that intrinsically affects the GS measurements, increases near the ground and in the high part of the atmosphere (Vernin & Azouit 1983;Tokovinin 1997;Prieur et al 2001). Therefore we performed a deeper study to confirm that the seeing variations estimated in our study are effectively due to a real different distribution of the turbulence and not to some noise.…”

Automatic Determination of Wind Profiles with Generalized SCIDAR

“…The SCIDAR technique has been the subject of many papers (e.g., Rocca et al 1974;Vernin & Azouit 1983;Caccia et al 1987;Avila et al 1997;Klückers et al 1998;Prieur et al 2001). Here we present only the guidelines of the method in order to introduce the quantities that are useful for this paper.…”

“…We have chosen to use an iterative approach based on the CLEAN algorithm, as we did for the inversion of profiles 2 C N for SCIDAR measurements (Prieur et al 2001). We wanted to take advantage of the experience acquired working with the interactive program developed by Avila et al (2001), which had a demonstrated efficiency in providing good measurements.…”

“…The determination of is thus an 2 C (h) N inverse problem, obeying equation (6), which is a Fredholmtype equation. It can be solved by various numerical methods; for instance, by using a maximum-entropy algorithm (e.g., Vernin 1992) or CLEAN (e.g., Prieur et al 2001). Vernin & Azouit (1983) showed that , which is the DR(h) equivalent radius (full width at half-maximum) of a correlation peak for a given layer at altitude h, is proportional to .…”

“…3 Although profiles of are routinely calculated from those obser-2 C (h) N vations, using maximum entropy methods (Vernin 1992;Klückers et al 1998) or CLEAN (Prieur et al 2001), few papers presenting wind profiles have been published (Klückers et al 1998;Avila et al 2001Avila et al , 2003Vernin et al 2000). The main reason is that wind parameters are rather difficult to retrieve from SCIDAR (or GS) data.…”