Search papers, labs, and topics across Lattice.
DeepMindCU BoulderEuropean Centre for Medium-Range WeatherINRIAOtto-von-Guericke-University MagdeburgUniversity of PotsdamMay 28, 2026arXiv:2605.29976This paper evaluates the climate simulation capabilities of two ML models, ArchesWeather (deterministic) and ArchesWeatherGen (probabilistic), by adapting them as forced atmospheric models conditioned on monthly mean SST and SIC, following the AIMIP Phase 1 protocol. The study demonstrates that these models, originally designed for weather forecasting, can produce stable long-term climate simulations and capture key climate variables' drift when forced with appropriate boundary conditions. The models accurately reproduce ERA5 climatology, large-scale circulations, and interannual variability.
Weather models can do climate, too: ArchesWeather and ArchesWeatherGen, originally built for short-term forecasting, show surprisingly strong performance in multi-decadal climate simulations when forced with SST and SIC.
We evaluate the climate simulation capabilities of ArchesWeather and ArchesWeatherGen, two machine learning models originally trained for weather forecasting and evaluated up to a 10-day lead time. ArchesWeather is a deterministic model, while ArchesWeatherGen is a probabilistic flow-matching model leveraging ArchesWeather's forecasts, enabling ensemble-based uncertainty quantification. In this work, we adapt these models to act as forced atmospheric models by using additional conditioning on the monthly mean sea surface temperature (SST) and sea ice cover (SIC) as boundary conditions. In particular, we follow the AI Model Intercomparison Project (AIMIP) Phase 1 protocol, which, analogous to the Atmospheric Model Intercomparison Project (AMIP), proposes a standardized experimental setup to evaluate the climate skill of ML-based forced atmospheric models. We present a comprehensive evaluation of both models under these conditions, including comparison against numerical climate models, ablation studies that examine key design choices in the extension, and an analysis of forced versus unforced configurations. Despite being originally developed for weather forecasting, we demonstrate that forced configurations of ArchesWeather and ArchesWeatherGen produce stable long-term climate simulations, have a stable annual cycle, and capture the drift of many climate variables. The models faithfully reproduce ERA5's climatology, large-scale circulations and interannual variability, and they capture the tails of the distributions.
