Arcetri towards an ambitious goal: the optical turbulence forecasting

It is known that, for the same wavelength, the ground-based telescopes have the advantage to reach higher angular resolution than the space-based telescopes. The angular resolution measures the ability of an optical system to distinguish details of an image. The higher is the resolution, the easier is the detection of close objects. If, for example, we intend to detect an extra-solar planet orbiting around its star, we need extreme angular resolution to distinguish two objects not only very close but also characterized by a very large contrast. The angular resolution is proportional to l/D (where l is the wavelength and D is the diameter of the telescope’s mirror). If we consider the new generation space telescope, the James Webb Space telescope (D=6.5m) and the new generation ground-based telescopes, the Extremely Large Telescopes (D=30-40m), we deduce that the angular resolution of the second one is better than that of the former of a factor 5-6. Unfortunately, this gain is only potential. The presence of the atmospherical turbulence around our planet induces fluctuations of temperature and, as a consequence, fluctuations of the refractive index of the atmosphere that finally induces perturbations of the wavefront of objects that we are observing getting the effective angular resolution of the telescope equivalent to that of a telescope of 10 cm in diameter.

To exploit the potentials of new generation telescopes it is therefore necessary to study optical turbulence (definition that is based on the effect of the atmospherical turbulence on the wavefronts), to know its characteristics to be able to correct its effects on the wavefront and to recover the original information contained in the luminous signal.

This is the reason why, in the last decades, a set of research lines, crucial for the astrophysics that are known with the name of "High Angular Resolution techniques" developed beside the most classical research lines of the astrophysics such as, for example, the galactic and extra-galactic astrophysics. These disciplines aim to study the nature of the optical turbulence (OT), how it is triggered, how it develops and how it disappears; how the turbulence affects the wavefront, how to measure the wavefront perturbations and how to correct them. In this framework, in Arcetri are concentrated two research groups that focus their activities on how to correct the wavefront perturbations (Adaptive Optics group) and to study the optical turbulence, to measure it, to model and to forecast it (Optical Turbulence group).

The problem is, indeed, that the efficiency of the adaptive optics is strongly dependent on the state of the turbulence and it is therefore fundamental to quantify and to know the distribution of the OT in space and time. Besides that, we remind that, many among the most challenging scientific programs require excellent turbulent conditions to be executed and, if we are not able to predict in advance these temporal windows, the risk is that these programs have in reality the lowest probability to be completed with the consequent loss of scientific potential of sophisticated instruments as well as of scientific observational programs. 

The scientific impact of the modern technology strongly depends therefore on our ability in selecting the right instrument to carry on the most suitable scientific program at the right moment. Considering that the cost of a night of observation at top class telescope of present time is around a hundred thousand of US Dollars and, considering that the turbulence is one of the most difficult physical phenomena still to be understood and decoded we can understand why such disciplines are so critical for the astrophysics of the third millennium.

 Optical turbulence forecast above the two most important sites of the European Astronomy

MOSE is a feasibility study funded by INAF and ESO having the goal to investigate the possibility to forecast the optical turbulence and a set of atmospherical parameters (temperature, wind speed and direction, …) relevant for ground-based observations above two among the most important sites of the European Astronomy for observations in the visible and infrared regimes: Cerro Paranal and Cerro Aramzones. The study uses an atmospherical non-hydrostatical mesoscale model (Meso-NH - Lafore et al. 1998) with a package for the reconstruction of the optical turbulence (Masciadri et al 1999). The PI of the project is a researcher of the Arcetri Observatory ( Elena Masciadri) who, with her collaborators, has published recently on MNRAS the first encouraging results obtained.

In a first paper (Masciadri et al 2013) it has been put in evidence that this technique permits to forecast the wind speed stratification along 20 km above the ground in a very satisfactory agreement with observations taken as a reference (relative error within 14% on the whole atmosphere). Such a good correlation has been observed in statistical terms as well as on individual nights. Because it does not exist at present an instrument based on remote sensing principles that is able to monitor automatically the wind speed stratification during the individual nights, it follows that the technique proposed by Masciadri and collaborators is at present extremely appealing to quantify the temporal evolution of the wavefront coherence time (τ0), one of the fundamental parameter for the adaptive optics that tells us at which frequency the wavefront corrector has to work. The method presented in this study permits therefore to identify when the limits within which an adaptive optics system can run efficiently are respected and it permits to optimize the management of the astronomical observations done with a telescope.

In a second paper (Lascaux et al. 2013) it has been put in evidence that the same model is able to reconstruct in the surface layer (~ 30 m) the absolute temperature with a median bias and a RMSE in the [0.03, 0.64]° Cand [0.64, 0.93]° C. This guarantees us to forecast with an accuracy better than the centigrade the temperature in the surface layer of the atmosphere in proximity of the telescope (To) and therefore to tune the temperature of the primary mirror (Tm) and inside the dome (Ti) to eliminate the most important source of image perturbation at the focus of a telescope: the dome seeing and the mirror seeing. This depends indeed on the thermal gradient between Tm, Ti and To. At the same time it has been possible to identify the right model configuration to reconstruct efficiently the wind speed and direction close to the surface. The strong wind speed near the ground is the main cause of vibrations of the adaptive secondary (a fundamental element of an adaptive optics system). It is known that, in presence of strong wind speed, it is fundamental to know the wind direction because vibrations on the adaptive secondary can have a great impact or be negligible depending on the wind direction.

Results on model performances in reconstructing the optical turbulence will be published in a forthcoming paper but we can anticipate that they are encouraging. At present the Optical turbulence group is engaged in defining the second phase of the project in which it will be defined an operational prototype.  

19122007 K 3 wind 05m seeing dom3 reduced  

2D map of the seeing obtained integrating the optical turbulence in the [5m, 20km] range above ground level. The map has a surface of 32.5km x 20km, the horizontal resolution is ΔX = 500m and the temporal sampling of the animation is 5 min extended on the interval [00:00-09:00] UT i.e. the local night. Letters P and A indicate the positions of Cerro Paranal and Cerro Armazones. Black lines indicate the isolines of the orography. The vector field indicate the wind at 10 m above the ground. Color bar indicates seeing values (arcsec).


Edited by Elena Masciadri and Anna Gallazzi