Kristian Strommen

Abstract:

Machine-learnt weather prediction (MLWP) models are now well established as being competitive with conventional numerical weather prediction (NWP) models in the medium range. However, there is still much uncertainty as to how this performance extends to longer timescales, where interactions with slower components of the earth system become important. We take GenCast, a state-of-the-art probabilistic MLWP model, and apply it to the task of seasonal forecasting with prescribed sea surface temperature (SST), by providing anomalies persisted over climatology (GenCast-Persisted) or forcing with observed SSTs (GenCastForced). The forecasts are compared to the European Centre for Medium-Range Weather Forecasts seasonal forecasting system, SEAS5. Our results indicate that, despite being trained at short timescales, GenCast-Persisted produces much of the correct precipitation patterns in response to El Ni˜no and La Ni˜na events, with several erroneous patterns in GenCast-Persisted corrected with GenCast-Forced. The uncertainty in precipitation response, as represented by the ensemble, compares favourably to SEAS5. Whilst SEAS5 achieves superior skill in the tropics for 2-metre temperature and mean sea level pressure (MSLP), GenCast-Persisted achieves higher skill in some areas in higher latitudes, including mountainous areas, with notable improvements for MSLP in particular; this is reflected in a slightly higher correlation with the observed NAO index. Reliability diagrams indicate that GenCast-Persisted has little skill relative to climatology, whilst GenCast-Forced produces forecasts with reliability comparable to SEAS5. These results provide an indication of the potential of MLWP models similar to GenCast for the ‘full’ seasonal forecasting problem, where the atmospheric model is coupled to ocean, land and cryosphere models.

Abstract:

Understanding how fast atmospheric variability shapes slow climate variability and sensitivity is a central challenge in Earth-system science. Recent advances in machine-learned (ML) atmospheric models have demonstrated remarkable skill on weather timescales, but their emergent behaviour in a fully coupled climate system is largely unexplored. We present results from a new hybrid modelling framework that couples a machine-learned atmosphere to a dynamical ocean model. We report on a set of 70-year coupled simulations (1950–2020 historical forcing and fixed-1950s control) in which the ACE2 ML climate emulator is interactively coupled to the NEMO ocean model. These experiments represent, to our knowledge, the first multi-decadal integrations of a machine-learned atmosphere interacting with a full-depth dynamical ocean. We assess the behaviour of the coupled system, with particular focus on low-frequency tropical variability and the climate response to greenhouse-gas forcing. Preliminary results indicate realistic emergent El Nino-like variability and a physically plausible climate sensitivity, suggesting that key atmosphere–ocean feedbacks can be captured within a hybrid ML–dynamical framework. These results evaluate the possible role of entirely machine-learned components in next-generation Earth-system models.

Abstract:

Uncertainty in projections of future regional climate change remains large, driven by structural differences among Earth System Models and the influence of internal climate variability. Existing uncertainty-reduction approaches, including emergent constraints and Bayesian variants, primarily focus on forced climate responses derived from simple aggregate metrics, thereby requiring strong assumptions and exploiting only low-dimensional climate information. Here we propose a data-driven deep-learning framework that directly forecasts spatially and monthly resolved decadal mean climatologies of surface temperature anomalies from the 2030s to the 2090s, using only recent monthly trajectories spanning 1980-2025. The training ensemble contains 265 historical+SSP2-4.5 simulations, distributed across 40 ESMs from 25 different families (i.e., modelling centers) over which the cross validation is performed. The architecture couples pluri-annual to multi-decadal temporal convolutions with a spatial U-Net encoder-decoder and is evaluated on CMIP6 simulations using a leave-one-model-family-out cross-validation (LOMFO-CV) design to ensure generalisation across separately developed ESMs. Predictive uncertainty is quantified via LOMFO-CV errors, yielding conservative and reliable ranges that incorporate irreducible internal variability and systematic model shifts.To further evaluate the predictive capacity beyond the CMIP6 distribution, we evaluated the network on historical+SSP2-4.5 simulations from a recent HadGEM3-GC5 model hierarchy developed within the European Eddy-Rich ESMs (EERIE) project, the European contribution to HighResMIP2 for CMIP7. In particular, the eddy-rich GC5-HH configuration explicitly simulates mesoscale ocean dynamics that are absent in CMIP6-type models, providing a rigorous test of generalisation to richer and more realistic physical representations. Despite these substantial differences, the network successfully reproduces warming trajectories and future climate patterns for all three model configurations (GC5-LL, GC5-MM, GC5-HH), with forecast errors largely contained within empirically calibrated uncertainty bounds from the LOMFO-CV, both globally and locally. These results, notably for GC5-HH and its more realistic physics, strengthens confidence in the applicability of the framework to real-world data.When applied to observations, the extracted end-of-century global-mean surface temperature and its uncertainty range are consistent with prior estimates from Bayesian frameworks. At local scales, the network reduces uncertainty by 40% (2030s) to 30% (2090s) on average, and by up to 75% in some regions for all future decades. Importantly, these uncertainty estimates account not only for uncertainty in the forced response (as emergent constraint methods do), but also for errors associated with predicting different realisations of internal variability, providing a physically meaningful reduction of local and global climate uncertainty.

Abstract:

This study evaluates the performance of high-resolution (grid sizes of 9-28 km for the atmosphere; 5-13 km for the ocean) global simulations from the EERIE project in representing the persistence of the Southern Annular Mode (SAM), a leading mode of Southern Hemisphere climate variability. Using the decorrelation timescale of the SAM index (τ), we compare EERIE simulations with CMIP6 models and ERA5 reanalysis. EERIE simulations reduce long-standing biases in SAM persistence, especially in early summer, with τ values of 9-20 d compared to CMIP6's 9-32 d and ERA5's 11 d. This improvement correlates with a more accurate climatological jet latitude (λ0). EERIE atmosphere-only AMIP runs outperform the coupled simulations in both τ and λ0, showing smaller biases and ranges of variability, underscoring the critical role of sea surface temperature (SST) representation in shaping atmospheric circulation. In these AMIP experiments, the atmospheric eddy feedback strength, combined with the damping timescale estimated via friction, correlates more strongly with τ than λ0. We speculate that the well-captured jet position (biases < 1° relative to ERA5), due to prescribed SSTs, limits λ0's explanatory power for τ differences, allowing other processes to dominate. Using a finer model grid (9 km vs. 28 km) of the same AMIP model reduces τ, though the mechanism remains unclear. Finally, motivated by the importance of oceanic eddies in the Southern Ocean, we conducted sensitivity experiments that filter transient mesoscale features from the SST boundary conditions. The results suggest that oceanic eddies may enhance summertime SAM persistence (by g1/4 2 d), though this signal is not statistically significant and is absent in the single 9 km run, pointing to a subtle role of mesoscale ocean-Atmosphere interaction that remains to be explored.

Abstract:

Weather regimes and weather patterns, here jointly referred to as circulation types, are used to generate forecasts for a variety of applications, such as energy demand and flood risk. However, there are usually many different choices available for precisely which circulation types to use. Ideally, one would like to use circulation types that are both highly informative for the application and also highly predictable, but in practice, there is often a tradeoff between informativity and predictability. We present a simple, general framework for how to construct a circulation type forecast that optimally balances these factors by segueing between different choices of circulation types at different lead times based on information‐theoretic considerations. As an example, we apply this framework to the case of forecasting energy demand in Great British winters. We compare a set of 30 weather patterns produced by the UK Met Office with the much simpler two‐state framework consisting of a positive and negative North Atlantic Oscillation (NAO) regime and show how to optimally combine the two across a winter season.