Integrated turbulence parameters' estimation from NAOMI adaptive optics telemetry data

TL;DR

This paper presents a novel method for estimating atmospheric turbulence parameters using low-resolution adaptive optics telemetry data, achieving high accuracy on both simulated and real data from the VLT.

Contribution

It introduces a modified chi-squared modal fitting algorithm that estimates turbulence parameters and their uncertainties from low-resolution AO telemetry data.

Findings

Achieves 1% accuracy in seeing estimation within 20 seconds on simulated data.

Estimates median seeing of 0.76 arcseconds with 1.2% statistical and systematic uncertainties from on-sky data.

Spatial distribution of telescopes does not affect the seeing value.

Abstract

Monitoring turbulence parameters is crucial in high-angular resolution astronomy for various purposes, such as optimising adaptive optics systems or fringe trackers. The former are present at most modern observatories and will remain significant in the future. This makes them a valuable complementary tool for the estimation of turbulence parameters. The feasibility of estimating turbulence parameters from low-resolution sensors remains untested. We perform seeing estimates for both simulated and on-sky telemetry data sourced from the new adaptive optics module installed on the four Auxiliary Telescopes of the Very Large Telescope Interferometer. The seeing estimates are obtained from a modified and optimised algorithm that employs a chi-squared modal fitting approach to the theoretical von K\'arm\'an model variances. The algorithm is built to retrieve turbulence parameters while simultaneously estimating and accounting for the remaining and measurement error. A Monte Carlo method is proposed for the estimation of the statistical uncertainty of the algorithm. The algorithm is shown to be able to achieve per cent accuracy in the estimation of the seeing with a temporal horizon of 20s on simulated data. A 0.76" +/- 1.2%$|_\mathrm{stat}$ +/- 1.2%$|_\mathrm{sys}$ median seeing was estimated from on-sky data collected from 2018 and 2020. The spatial distribution of the Auxiliary Telescopes across the Paranal Observatory was found to not play a role in the value of the seeing.