DiffScale: Continuous Downscaling and Bias Correction of Subseasonal Wind Speed Forecasts using Diffusion Models

TL;DR

This paper addresses the challenge of subseasonal wind speed forecasting by introducing DiffScale, a diffusion-based method for continuous spatiotemporal downscaling conditioned on a scaling factor $α$ and lead time $l$. It jointly performs upsampling and forecast bias correction by sampling from the conditional density $p_{α}(oldsymbol{x}^{(α)}|oldsymbol{x},τ)$, guided by weather priors and without autoregressive time stepping. DiffScale demonstrates improved deterministic and probabilistic skill relative to ECMWF S2S and climatology, with strong gains at shorter lead times and coarser scales, and geographic improvements in coastal and mountainous regions. The approach offers a flexible, data-efficient framework for high-resolution wind forecasts with potential impact on renewable energy planning, while acknowledging limitations from domain size and data availability.

Abstract

Renewable resources are strongly dependent on local and large-scale weather situations. Skillful subseasonal to seasonal (S2S) forecasts -- beyond two weeks and up to two months -- can offer significant socioeconomic advantages to the energy sector. This study aims to enhance wind speed predictions using a diffusion model with classifier-free guidance to downscale S2S forecasts of surface wind speed. We propose DiffScale, a diffusion model that super-resolves spatial information for continuous downscaling factors and lead times. Leveraging weather priors as guidance for the generative process of diffusion models, we adopt the perspective of conditional probabilities on sampling super-resolved S2S forecasts. We aim to directly estimate the density associated with the target S2S forecasts at different spatial resolutions and lead times without auto-regression or sequence prediction, resulting in an efficient and flexible model. Synthetic experiments were designed to super-resolve wind speed S2S forecasts from the European Center for Medium-Range Weather Forecast (ECMWF) from a coarse resolution to a finer resolution of ERA5 reanalysis data, which serves as a high-resolution target. The innovative aspect of DiffScale lies in its flexibility to downscale arbitrary scaling factors, enabling it to generalize across various grid resolutions and lead times -without retraining the model- while correcting model errors, making it a versatile tool for improving S2S wind speed forecasts. We achieve a significant improvement in prediction quality, outperforming baselines up to week 3.

Figures (18)

• Figure 1: Visual comparison of the ERA5 target, ECMWF S2S and DiffScale (lr-ws & sf) for the finest resolution considered in our method. Displayed is the mean of ws10m obtained over all forecast times.
• Figure 2: MAE and CRPSS metrics calculated across the different bins for each resolution. The different setups of DiffScale's experiments are shown as colored lines, while the ECMWF S2S model is represented as a dashed and dotted green line. The dashed black line corresponds to climatology.
• Figure 3: The DiffScale Diffusion model with classifier free guidance $\tau$. Arrows symbolize numerical solver steps for the respective score, solving the reverse process to generate $\hat{\mathbf{x}}^{(\alpha)}$ given $\mathbf{x}$.
• Figure 4: Accumulation of errors for ECMWF numeric predictions over lead time and respective bins.
• Figure 5: Evaluation of the MAE for different solver methods and number of function evaluations (solver steps) per sampling of downscaled $\hat{x}^{(\alpha)}$. Displayed is the mean over all scaling factors per bin and number of solver steps.