Spatio-temporal Nowcasting of Atmospheric Optical Turbulence for LEO Satellite-to-Ground Laser Communication Networks

The stability and efficiency of high-capacity Low Earth Orbit (LEO) satellite-to-ground laser communication (SGLC) networks are severely challenged by the highly dynamic and non-stationary nature of atmospheric optical turbulence during rapid transit windows. Addressing the critical necessity for network scheduling, this study moves beyond static atmospheric characterization by deriving high-resolution Fried parameter fields to capture transient atmospheric states. We propose the Diffusion Schrödinger Bridge Graph Transformer (DSBGT) for precise minute-scale spatiotemporal nowcasting. The DSBGT architecture employs a Graph Wavelet Network-based Transformer (GWNformer) backbone to resolve complex spatiotemporal dependencies via dynamic graphs and parallel attention mechanisms. Crucially, to tackle the inherent non-stationarity of turbulence, a Diffusion Schrödinger Bridge (DSB) is incorporated to establish precise probabilistic optimal transport mappings from current features to future Fried parameter distributions. Experimental results demonstrate that DSBGT achieves a superior Mean Absolute Error (MAE) of 0.22 cm, significantly outperforming existing benchmarks. These findings highlight the model's critical value for enabling proactive operational link scheduling and dynamic resource optimization in future LEO SGLC networks.