A Generative Framework for Probabilistic, Spatiotemporally Coherent Downscaling of Climate Simulation
This work addresses the need for accurate local climate information for impact assessment and decision-making, representing an incremental improvement over existing statistical downscaling methods.
The paper tackles the problem of generating high-resolution local weather dynamics from coarse climate simulations by introducing a generative framework using a score-based diffusion model, which produces spatiotemporally coherent weather patterns consistent with global climate output.
Local climate information is crucial for impact assessment and decision-making, yet coarse global climate simulations cannot capture small-scale phenomena. Current statistical downscaling methods infer these phenomena as temporally decoupled spatial patches. However, to preserve physical properties, estimating spatio-temporally coherent high-resolution weather dynamics for multiple variables across long time horizons is crucial. We present a novel generative framework that uses a score-based diffusion model trained on high-resolution reanalysis data to capture the statistical properties of local weather dynamics. After training, we condition on coarse climate model data to generate weather patterns consistent with the aggregate information. As this predictive task is inherently uncertain, we leverage the probabilistic nature of diffusion models and sample multiple trajectories. We evaluate our approach with high-resolution reanalysis information before applying it to the climate model downscaling task. We then demonstrate that the model generates spatially and temporally coherent weather dynamics that align with global climate output.