Efficient deep learning for radar precipitation nowcasting using spatiotemporal encoding and two-dimensional reconstruction

Short-term precipitation nowcasting, the prediction of rainfall over lead times from a few minutes to about an hour, remains challenging because radar-derived precipitation fields evolve not only through motion but also through rapid, non-linear changes such as growth, decay, and structural reorganization. Classical extrapolation methods are efficient yet struggle to represent these intensity and morphology changes, while many learning-based approaches become costly when scaled to large, high-resolution radar grids. 

Our approach treats temporal learning and spatial reconstruction as two separate problems. A compact 3D convolutional encoder processes a short radar sequence to capture how precipitation structures evolve over time. We then convert the encoder feature volumes into 2D skip representations through depth aggregation and channel compression and use a lightweight 2D decoder to reconstruct full resolution forecasts. We benchmark against persistence and a strong 2D convolution baseline. 

The framework is evaluated on the RYDL dataset derived from the German Weather Service radar composite, providing 2D radar fields every five minutes over Germany at 1 × 1 km resolution on a 900 × 900 grid. Performance is benchmarked against persistence and a strong 2D convolutional baseline using complementary verification measures, including mean absolute error, critical success index at multiple intensity thresholds, and fractions skill score with spatial tolerance. Across benchmark lead times, the proposed approach reduces MAE from 0.22 to 0.20 at 5 min, from 0.35 to 0.28 at 30 min, and from 0.44 to 0.42 at 60 min relative to the 2D baseline, indicating improved robustness at intermediate horizons while retaining competitive short-range accuracy. These results suggest that combining explicit spatio-temporal encoding with efficient two-dimensional reconstruction offers a practical route to scalable radar nowcasting on large domains. 
Keywords: Radar nowcasting, precipitation forecasting, deep learning, spatio-temporal representation learning, forecast verification