MODELING WITH ATMOSPHERICAL NON-HYDROSTATIC MESOSCALE MODELS

The characterization of the optical turbulence done with mesoscale atmospherical models plays an absolutely crucial role for the success of the future ground-based astronomy and, in general, of the high-angular resolution technics for at least three fundamental reasons:
(1) mesoscale models can provide 3D maps of the optical turbulence (CN2 profiles) above a surface of a few tens of kilometers with sub-kilometric horizontal resolution.

(2) mesoscale models can forecast the optical turbulence, necessary requirement to implement the flexible-scheduling. The flexible scheduling is a term employed in astronomy to indicate the method used to schedule the scientific programs to be carried out at the focus of ground-based telescopes. Using this method the scheduling of the scientific programs is done following criteria that consider the excellence of the scientific programs as well as the quality of the turbulence condition.

(3) mesoscale models can perform a climatology of the optical turbulence extended on the time scale of the order of decades to identify trends at large temporal scale of the seeing and other integrated astroclimatic parameters that should be very difficult to identify with measurements. This might be of great impact for the astronomical observations. It can happen, indeed, that, climatological changes of some atmospheric parameters can induce changes in the features of the optical turbulence. (An example of these phenomena is the deterioration of the seeing values observed above Cerro Paranal after the 1990, time in which this site has been selected for the VLT. Observations put in evidence during the nineties that the seeing measured by the seeing monitor got worse monotonically).

No other tools of investigation, able to achieve these scientific goals, exist at present. For this reason the atmospherical mesoscale models are a very appealing tool for research in astronomy.

A non-hydrostatic atmospherical mesoscale model (Meso-Nh) has been employed for the first time to predict the vertical distribution of the optical turbulence (CN2 profiles) above an astronomical site by Masciadri et al. 1999a, 1999b. In these first studies the Astro-Meso-Nh-Code has been implemented in the Meso-Nh model and it has been proven that the model could reconstruct CN2 profiles well correlated to measured 
CN2 profiles from a qualitative (CN2 shape) and quantitative (total turbulence budget) point of view. In Masciadri et al. (2001) it has been proved, among other, that the peculiar turbulence scheme of Meso-Nh, that takes into account the Prandtl number variability calculates CN2 profiles better correlated to measurements. In a successive study (Masciadri & Jabouille, 2001) it has been proposed a method to calibrate the model for the prediction of the optical turbulence  and a statistical validation of the model has been published later on (Masciadri et al. 2004) on a sample of 10 nights. The model has been applied, for the first time in autonomous way, after calibration by Masciadri & Egner (2006). The CN2 profiles have been simulated for 80 nights, uniformly distributed along one year and a complete characterization of all the main astroclimatic parameters has been performed identifying, for the first time, new features related to the seasonal variations of the CN2 profiles at the jet-stream level (α effect,  for an extended description see also Masciadri et al. 2010).

The ForoT activities mainly concentrated above two sites: Mt. Graham and the Internal Antarctic Plateau.

 

Internal Antarctic Plateau (Dome C)

The Astro-Meso-Nh-Code has been validated above Dome C using a grid-nesting configuration. Using a sample including all the measured CN2 profiles (15 nights) during the winter time (Trinquet et al. 2008), we proved that the Astro-Meso-Nh-Code correctly calculates the mean surface layer (PBL) thickness hsl and the median seeing above and below the surface layer (Lascaux et al. 2009, 2010).The latter are the first estimates of the optical turbulence in the free atmosphere ever done with a mesoscale model above Dome C.

Fig.1: (from Lascaux et al. 2010): CN2 temporal evolution during 18 hours above Dome C during a night in winter 2005. (a) CN2 in the first 300 m, (b) CN2 in the [1 km, 12 km] range. The turbulence appears very concentrated near the ground in this case. In the free atmosphere, it is well visible how the CN2 evolves in time and space. Such a feature is an indication of the model variability even in the high part of the atmosphere, one of th emost critical issue for such a kind of models.


Fig 1: (from Lascaux et al. 2010): Total seeing and seeing in the free atmosphere calculated by the model Meso-Nh and measured by balloons (all the balloons launched during at Dome C during the winter - Trinquet et al. 2008). Calculations are done using two different configurations for the Meso-Nh: a grid-nesting configuration using three imbricated models with the highest horizontal resolution  equal to 1 km (Meso-Nh high) and a simple model with a horizontal reoslution equal to 100 km (Meso-Nh low). The height of the surface layer is calculated, in all cases, as the height below which 90% of the turbulence developed in the whole atmosphere is concentrated. (a) total seeing, (b) seeing above the height of the surface layer (median value). Black dots: Meso-Nh low. Red dots: Meso-Nh high. As can be seen in the Table, only the configuration at high resolution provides a good correlation with measurements for the three most important parameters: turbulent surface layer height,seeing in the free atmosphere and in the surface layer.

The excellent  agreement with measurements, statistically quantified in Lascaux et al.(2009, 2010), guarantees us about the reliablity of the model at these extreme latitudes and gives us confidence in employing the model above the Internal Antarctic Plateau to identify the best locations for astronomical applications. Extended analyses are on-going above other critical locations of the plateau potentially interesting for astronomical applications. A couple of animations for the seeing and the isoplanatic angle related to calculations done above Dome C are shown here.


       (a) 1/8/2005: seeing                                                                  (b) 1/8/2005: isoplanatic angle


(a): Horizontal map, extended on a surface of 80 km x 80 km and a horizontal resolution of  1km, of the seeing calculated integrating the turbulence from an height of 50 m from the ground up to  21,422 km above sea level. Units: arcseconds. Temporal evolution of the simulation: (00:00 - 12:00) UT. Temporal sampling: 10 min. The table of color shows the seeing included in the range (0.2-2) arcsec. The black isolines represent the isolevels of the topography. The higher is the lines density, the higher is the ground steepness. The cross indicates the location of the Dome C where the Franco-Italian  Concordia Base is located.
(b):
Horizontal map, extended on a surface of 80 km x 80 km and an horizontal resolution of 1km, of the isoplanatic angle calculated integrating the turbulence from the ground up to  ~ 21,422 km above sea level. Units: arcseconds.

Further Model Outputs (Antarctica)

Mt. Graham

         (a) 29/2/2008: seeing                                                                  (b) 29/2/2008: isoplanatic angle


(a):
Horizontal map, extended on a surface of 20 km x 20 km and a horizontal resolution of 500 m, of the seeing calculated integrating the turbulence from an height of 20 m from the ground up to  ~ 24,095 km a.s.l. The height of 20 m corresponds to the height of the Vatican Advanced technology Telescope (VATT) telescope where the instrument for the optical turbulence measurement (Generalized Scidar) has been located. Units: arcseconds.
Temporal evolution of the simulation: (00:00 - 12:00) UT. Temporal sampling: 10 min. The table of color shows the seeing included in the range (0.2-2) arcsec. The black isolines represent the isolevels of the topography. The higher is the lines density, the higher is the ground steepness. The cross indicates the  Large Binocular Telescope (LBT) located  at the MGIO Observatory of Mt. Graham (Arizona, US).
(b):
Horizontal map, extended on a surface of 20 km x 20 km and a horizontal resolution of 500 m, of the isoplanatic angle calculated integrating the turbulence from the ground up to  ~ 24,095 km. Units: arcseconds.

Instantaneous values of integrated astroclimatic parameters above Mt. Graham (29/2/2008)


Acknowledgments: This work is funded by the Marie Curie Excellence Grant ForOT - MEXT-CT-2005-023878



E.Masciadri, 11/2009