Forecasts of the atmospherical parameters close to the ground at the LBT site in the context of the ALTA project

TL;DR

This study evaluates the accuracy of mesoscale atmospheric models in forecasting ground-level parameters at the LBT site, aiming to enhance telescope operation planning through reliable predictions.

Contribution

It validates the Meso-Nh and Astro-Meso-Nh models for the LBT site, extending previous site-specific studies and supporting operational forecasting for astronomical observations.

Findings

Models show high accuracy in predicting wind speed and temperature.

Forecasting performance is validated against local measurements using contingency analysis.

Results support operational use of models for telescope scheduling and adaptive optics.

Abstract

In this paper we study the abilities of an atmospherical mesoscale model in forecasting the classical atmospherical parameters relevant for astronomical applications at the surface layer (wind speed, wind direction, temperature, relative humidity) on the Large Binocular Telescope (LBT) site - Mount Graham, Arizona. The study is carried out in the framework of the ALTA project aiming at implementing an automated system for the forecasts of atmospherical parameters (Meso-Nh code) and the optical turbulence (Astro-Meso-Nh code) for the service-mode operation of the LBT. The final goal of such an operational tool is to provide predictions with high time frequency of atmospheric and optical parameters for an optimized planning of the telescope operation (dome thermalization, wind-dependent dome orientation, observation planning based on predicted seeing, adaptive optics optimization, etc...). Numerical simulations are carried out with the Meso-Nh and Astro-Meso-Nh codes, which were proven to give excellent results in previous studies focused on the two ESO sites of Cerro Paranal and Cerro Armazones (MOSE Project). In this paper we will focus our attention on the comparison of atmospherical parameters forescasted by the model close to the ground with measurements taken by the observatory instrumentations and stored in the LBT telemetry in order to validate the numerical predictions. As previously done for Cerro Paranal (Lascaux et al., 2015), we will also present an analysis of the model performances based on the method of the contingency tables, that allows us to provide complementary key information with the respect to the bias and RMSE (systematic and statistical errors), such as the percentage of correct detection and the probability to obtain a correct detection inside a defined interval of values.