DEF: Diffusion-augmented Ensemble Forecasting
This addresses a bottleneck in machine learning for weather prediction by providing a generalizable stochastic method, though it is incremental as it builds on existing diffusion and ensemble techniques.
The paper tackles the lack of general methods for generating initial condition perturbations in machine learning for weather prediction by proposing DEF, a diffusion-augmented ensemble forecasting approach that transforms deterministic neural systems into stochastic ones, showing improved predictive performance and reduced error accumulation in long-term forecasts on the ERA5 dataset.
We present DEF (\textbf{\ul{D}}iffusion-augmented \textbf{\ul{E}}nsemble \textbf{\ul{F}}orecasting), a novel approach for generating initial condition perturbations. Modern approaches to initial condition perturbations are primarily designed for numerical weather prediction (NWP) solvers, limiting their applicability in the rapidly growing field of machine learning for weather prediction. Consequently, stochastic models in this domain are often developed on a case-by-case basis. We demonstrate that a simple conditional diffusion model can (1) generate meaningful structured perturbations, (2) be applied iteratively, and (3) utilize a guidance term to intuitivey control the level of perturbation. This method enables the transformation of any deterministic neural forecasting system into a stochastic one. With our stochastic extended systems, we show that the model accumulates less error over long-term forecasts while producing meaningful forecast distributions. We validate our approach on the 5.625$^\circ$ ERA5 reanalysis dataset, which comprises atmospheric and surface variables over a discretized global grid, spanning from the 1960s to the present. On this dataset, our method demonstrates improved predictive performance along with reasonable spread estimates.