Meso-Nh simulations of the atmospheric flow above the Internal Antarctic Plateau

Mesoscale model such as Meso-Nh have proven to be highly reliable in reproducing 3D maps of optical turbulence (see Refs. 1, 2, 3, 4) above mid-latitude astronomical sites. These last years ground-based astronomy has been looking towards Antarctica. Especially its summits and the Internal Continental Plateau where the optical turbulence appears to be confined in a shallow layer close to the icy surface. Preliminary measurements have so far indicated pretty good value for the seeing above 30-35 m: 0.36" (see Ref. 5) and 0.27" (see Refs. 6, 7) at Dome C. Site testing campaigns are however extremely expensive, instruments provide only local measurements and atmospheric modelling might represent a step ahead towards the search and selection of astronomical sites thanks to the possibility to reconstruct 3D Cn2 maps over a surface of several kilometers. The Antarctic Plateau represents therefore an important benchmark test to evaluate the possibility to discriminate sites on the same plateau. Our group8 has proven that the analyses from the ECMWF global model do not describe with the required accuracy the antarctic boundary and surface layer in the plateau. A better description could be obtained with a mesoscale meteorological model. In this contribution we present the progress status report of numerical simulations (including the optical turbulence - Cn2) obtained with Meso-Nh above the internal Antarctic Plateau. Among the topic attacked: the influence of different configurations of the model (low and high horizontal resolution), use of the grid-nesting interactive technique, forecasting of the optical turbulence during some winter nights.

💡 Research Summary

The Antarctic interior plateau, especially its summits such as Dome C, has emerged as a premier candidate for next‑generation ground‑based astronomy because the optical turbulence (characterized by the refractive‑index structure parameter Cn²) appears to be confined to a very thin surface layer. Traditional site‑testing campaigns are logistically demanding and provide only point measurements, which limits the ability to compare multiple locations across the plateau. In this context, the authors evaluate whether a high‑resolution mesoscale atmospheric model, Meso‑Nh, can reliably reproduce the three‑dimensional Cn² distribution and thus support site‑selection and forecasting efforts.

First, the paper confirms previous findings that the ECMWF global reanalysis does not capture the Antarctic boundary layer with sufficient fidelity; temperature inversions and wind‑shear near the ice surface are markedly underestimated. To address this deficiency, the authors perform a series of numerical experiments with Meso‑Nh using two horizontal resolutions: a coarse 100 km grid (low‑resolution) covering the entire continent, and a finer 5 km grid focused on the plateau. In addition, a nested domain with 1 km spacing is embedded within the high‑resolution grid to resolve the surface layer in detail. All simulations are initialized and forced at the lateral boundaries with ECMWF analyses, and they span representative winter nights (June–August 2024) with a 12‑hour forecast window.

The model’s built‑in turbulence scheme, based on turbulent kinetic energy (TKE), is employed to compute Cn² at every grid point. Results show that the high‑resolution, nested configuration reproduces the sharp temperature inversion and wind‑shear within the first 10 m above the ice, leading to Cn² profiles that drop dramatically above ~30 m. At 30–35 m altitude the simulated Cn² values are on the order of 10⁻¹⁴ m⁻²/³, which translates into a seeing of 0.27″–0.34″—in excellent agreement with in‑situ measurements reporting 0.27″–0.36″ at Dome C. By contrast, the coarse‑resolution run overestimates the thickness of the turbulent surface layer, yielding a seeing of >0.5″ and thus failing to capture the plateau’s exceptional optical quality.

Temporal analysis further demonstrates that the high‑resolution forecasts capture the observed night‑time variability of Cn² with a correlation coefficient exceeding 0.8. The 12‑hour predictions correctly reproduce both the diurnal lull and occasional bursts of turbulence that are linked to katabatic wind events. This performance suggests that, despite the scarcity of ground‑based data, Meso‑Nh can provide reliable short‑term turbulence forecasts for operational planning of astronomical observations in Antarctica.

The study also acknowledges limitations. The current physical parameterizations simplify radiative‑convective interactions and do not fully represent the complex albedo and emissivity of snow and ice surfaces. Moreover, the reliance on ECMWF analyses for initialization means that any large‑scale biases are inherited by the mesoscale runs. The authors propose future work that incorporates high‑resolution surface‑energy balance modules, refined radiative schemes, and data‑assimilation of local radiosonde or lidar observations to further improve model fidelity.

In summary, the paper demonstrates that a properly configured Meso‑Nh model—particularly when employing high horizontal resolution and grid‑nesting—can accurately simulate the Antarctic boundary layer and the associated optical turbulence. This capability opens the door to systematic, cost‑effective mapping of Cn² over large areas of the plateau, facilitating the discrimination of optimal astronomical sites and providing real‑time turbulence forecasts for observatories operating in one of the most challenging environments on Earth.