RainODE: Continuous-Time Precipitation Forecasting with Latent Neural ODEs
This work addresses the need for high-temporal-resolution precipitation forecasting, which is critical for applications like flood warning and water resource management.
RainODE reformulates precipitation forecasting as a continuous-time dynamical system using a Neural ODE in latent space, combined with a Brownian Bridge stochastic module to handle non-advective intensity changes. It achieves consistent improvements over multiple temporal intervals on SEVIR and a new RAPID dataset.
In precipitation forecasting, not only accuracy but also temporal resolution is critical. However, increasing temporal resolution is constrained by observational limitations and the computational cost of dense discrete modeling. To overcome this limitation, we reformulate precipitation forecasting as a continuous-time dynamical system and propose RainODE, a framework that models precipitation evolution in latent space using a Neural ODE. This formulation enables derivative-consistent temporal dynamics and captures the dominant large-scale advective motion of precipitation systems. Nevertheless, a purely deterministic ODE struggles to represent non-advective intensity changes such as localized growth, decay, and sub-grid variability, often leading to over-smoothed predictions. To address this issue, we introduce a stochastic source modeling module based on a Brownian Bridge formulation, which refines residual intensity variations and restores fine-grained structures while preserving advective consistency. By combining deterministic continuous dynamics with stochastic refinement, RainODE enables arbitrary-time inference while maintaining sharp predictions. Experiments on SEVIR and the newly introduced Radar-based Precipitation Integrated Dataset (RAPID) demonstrate consistent improvements across multiple temporal intervals and precipitation regimes. The code is available at https://github.com/SeongYE/RainODE.