Modelling and forecasting the effects of atmospheric turbulence on the propagation of optical waves is crucial for the development of future free-space optical communication systems. This work explores the use of autoregressive models to provide short-term predictions (i.e., minutes to hours) of optical turbulence integrated parameters, such as the Fried parameter, isoplanatic angle and scintillation index. It exploits measurements from a Shack-Hartmann Image Motion Monitor (SHIMM) instrument at Paranal Observatory, as well as recent data collected in Barcelona during the TURBulence mOnitoring (TURBO) campaign. The study demonstrates that combining long-term forecasts from numerical weather prediction models with on-site measurements using autoregressive models significantly improve short-term predictions. Additionally, autoregressive models based solely on measurements also show good prediction performance. This work motivates future measurement campaigns at optical communication sites to enhance site characterization and optical turbulence modelling, while enabling more accurate short-term forecasts.
Quatresooz, F., Griffiths, R., Osborn, J., Vanhoenacker-Janvier, D., & Oestges, C. (2024). Application of autoregressive models for optical turbulence prediction at optical communication sites. Proceedings Volume 13699, International Conference on Space Optics — ICSO 2024. Published. International Conference on Space Optics - ICSO 2024, Antibes Juan-les-Pins, France. https://doi.org/10.1117/12.3075364