Nowcasting the turbulence at the Paranal Observatory

TL;DR

This paper explores machine learning methods to predict short-term optical turbulence at Paranal Observatory, aiming to improve scheduling and decision-making for astronomical observations, especially for adaptive optics systems.

Contribution

It introduces a turbulence nowcasting approach using historical data and machine learning to forecast atmospheric conditions over the next two hours.

Findings

Machine learning can predict turbulence conditions with certain accuracy.

The approach shows potential for operational use in telescope scheduling.

Limitations include data quality and model generalization.

Abstract

At Paranal Observatory, the least predictable parameter affecting the short-term scheduling of astronomical observations is the optical turbulence, especially the seeing, coherence time and ground layer fraction. These are critical variables driving the performance of the instruments of the Very Large Telescope (VLT), especially those fed with adaptive optics systems. Currently, the night astronomer does not have a predictive tool to support him/her in decision-making at night. As most service-mode observations at the VLT last less than two hours, it is critical to be able to predict what will happen in this time frame, to avoid time losses due to sudden changes in the turbulence conditions, and also to enable more aggressive scheduling. We therefore investigate here the possibility to forecast the turbulence conditions over the next two hours. We call this "turbulence nowcasting", analogously with weather nowcasting, a term already used in meteorology coming from the contraction of "now" and "forecasting". We present here the results of a study based on historical data of the Paranal Astronomical Site Monitoring combined with ancillary data, in a machine learning framework. We show the strengths and shortcomings of such an approach, and present some perspectives in the context of the Extremely Large Telescope.