We introduce deep generative diffusion for multivariate and regional
surrogate modeling learned from sea-ice simulations. Given initial
conditions and atmospheric forcings, the model is trained to generate
forecasts for a 12-hour lead time from simulations by the
state-of-the-art sea-ice model neXtSIM. For our regional model setup,
the diffusion model outperforms as ensemble forecast all other tested
models, including a free-drift model and a stochastic extension of a
deterministic data-driven surrogate model. The diffusion model
additionally retains information at all scales, resolving smoothing
issues of deterministic models. Furthermore, by generating physical
consistent forecasts, previously unseen for such kind of completely
data-driven surrogates, the model can almost match the scaling
properties of neXtSIM, which are also observed for real sea ice. With
these results, we provide a strong indication that diffusion models can
achieve similar results as traditional geophysical models with the
significant advantage of being orders of magnitude faster and solely
learned from data.