Climate Modeling¶
-
Exploring the potential of AlphaEarth and TESSERA embeddings for Fine-scale Local Climate Zone Mapping: A case study across five cities in Switzerland¶
Understanding urban spatial morphology is critical for climate modeling, risk assessment, and sustainable urban design, and Local Climate Zone (LCZ) mapping provides the basic framework for this....Understanding urban spatial morphology is critical for climate modeling, risk assessment, and sustainable urban design, and Local Climate Zone (LCZ) mapping provides the basic framework for this. However, many cities still use coarse ~100-m resolution LCZ records, which are unsuitable for fine-scale urban research. In this study, precomputed embeddings from TESSERA (Feng et al., 2025) and AlphaEarth (Brown et al., 2025) are compared to traditional Sentinel-1/2 (S1S2) composites in five Swiss cities to see if they can upscale coarse LCZ maps to 10-m resolution using an attention-based U-Net. Three experiments assess multi-city transferability, the impact of higher-resolution reference data, and temporal robustness to year-to-year phenology changes. We find that all datasets achieve strong performance with test data Intersection-over-Union (IoU) ranging from 0.59-0.69 and 0.77-0.82 in the first two experiments. TESSERA consistently outperforms both S1S2 and AlphaEarth across both settings As expected, we find that the transfer of embedding-based models from one year to another remains an open challenge. Overall, however, our results demonstrate the promising potential of embeddings derived from EO foundation models to reduce time consuming preprocessing, respectively, manual feature engineering tasks and to guide a universal deep learning-based LCZ mapping workflow. When combined with a simple location-aware attention U-Net architecture, the embeddings enhance regional transferability and scalability, supporting the development of comprehensive and reproducible fine-scale LCZ maps for global urban climate applications Improving reference data quality remains the strongest lever for further accuracy gains. more
-
Optimal scenario design for climate emulation¶
As deep learning for physical systems continues to grow in popularity, efforts to improve generalizability have primarily focused on designing architectures that embed physical constraints. However,...As deep learning for physical systems continues to grow in popularity, efforts to improve generalizability have primarily focused on designing architectures that embed physical constraints. However, for machine-learning surrogate climate models (emulators), we show that the low structural diversity in existing scenarios commonly used to generate training data places a ceiling on predictive skill. Here, we examine whether training datasets themselves can be optimized to improve generalization. We introduce a method to create datasets that produce emulators capable of generalizing to new, structurally different scenarios absent from the training data. We use a differentiable Simple Climate Model (SCM) to calculate the sensitivity of emulator loss to perturbations in the training data, iteratively updating the training data to maximize emulator skill. For an SCM, training on one scenario optimized in this fashion outperforms an emulator trained on six standard ScenarioMIP pathways. We achieve this higher predictive skill despite training on a smaller dataset, finding that our emulator successfully isolates distinct physical behaviors of different climate forcing agents (e.g., greenhouse gases vs. aerosols) without single-forcing runs. We then demonstrate that scenarios optimized using an SCM, when used to drive an intermediate-complexity climate model, produce a training dataset that yields a more skillful emulator than training on ScenarioMIP outputs. Our results suggest that, in the compute-constrained environment of running full-scale climate models, generating a small number of dynamically rich scenarios provides greater marginal value for emulation and characterizing system responses than expanding the suite of traditional emissions pathways. more
-
Investigating Inductive Biases for Machine Learning Emulation of Sudden Stratospheric Warmings in Idealised Isca Simulations¶
Machine-learning emulators are increasingly used for weather prediction and have the potential to extend skill on subseasonal-to-seasonal timescales by learning dynamically important sources of...Machine-learning emulators are increasingly used for weather prediction and have the potential to extend skill on subseasonal-to-seasonal timescales by learning dynamically important sources of predictability. A key challenge is whether the models can exploit predictability anchors, such as stratospheric variability, that influence tropospheric circulation beyond short lead times. We test how architectural inductive bias affects emulation of sudden stratospheric warming (SSW) dynamics using paired idealised Isca simulations that differ only in an imposed wave-2 heating perturbation. Across convolutional, transformer, and graph-based architectures trained for one-step prediction, model differences are modest when the stratosphere is dynamically quiet but widen substantially when SSW-like variability is active. Our results identify explicit three-dimensional vertical coupling as a key inductive bias for machine-learning emulation of stratospheric dynamics. However, Eliassen-Palm flux diagnostics show that low forecast error does not guarantee physically faithful wave-mean-flow interaction, with coherent errors remaining in stratospheric wave-driving structure. more
-
Regional Climate Model Emulation with Diffusion Approaches: What is the Added Value of Generative Machine Learning?¶
Emulators provide a cost-effective alternative to regional climate models (RCMs) by capturing their dynamical downscaling function. They link large-scale predictors simulated by global climate models...Emulators provide a cost-effective alternative to regional climate models (RCMs) by capturing their dynamical downscaling function. They link large-scale predictors simulated by global climate models (GCMs) to RCM-simulated high-resolution fields of the target variable, here precipitation. Machine learning methods, typically deep learning, are cheaper than running RCMs in computation time and energy. Among them, generative models are appealing because they can simulate ensembles of local high-resolution fields consistent with the predictors. This ensemble, which we call the uncertainty envelope, remains to be properly assessed for added value. Here, we make three contributions. First, we introduce ParamDiffusion, a new two-stage diffusion-based framework, and compare it with a state-of-the-art diffusion approach. Second, we expand standard validation through a comprehensive framework aligned with climate-science needs, examining specific precipitation events, including extremes. Third, within this framework, we assess the added value of diffusion approaches relative to deterministic methods. We intercompare four deep-learning models: a deterministic model designed to capture the precipitation tail; a parametric probabilistic model based on it; a recently proposed diffusion approach; and ParamDiffusion, which couples the parametric model with a diffusion model. Our results show that diffusion-based approaches reproduce climatological precipitation statistics with high skill, including distributional tails and spatially compounded extremes, while generating spatially detailed fields. However, none of the assessed models consistently accounts for the most extreme RCM-simulated events within its uncertainty envelope. Diffusion models are therefore promising for probabilistic RCM emulation, but progress is still required before they can reliably represent high-impact precipitation extremes. more
-
Scalable Deep Learning Framework for Global High-Resolution Land Use Reconstruction¶
Uncertainty in the terrestrial carbon cycle remains a major constraint in climate projections, partly driven by the uncertainties affecting the land surface representation and variability in Earth...Uncertainty in the terrestrial carbon cycle remains a major constraint in climate projections, partly driven by the uncertainties affecting the land surface representation and variability in Earth system models. To address this limitation, we present a data-driven framework AI4Land, for generating high-resolution historical reconstructions and future projections of key land surface variables. The framework follows a two-phase approach using a U-Net architecture. In the first phase, which is the focus of this work, it reconstructs annual land use and land cover by integrating coarse-resolution scenario data with static geophysical features. In a planned second phase, the resulting high-resolution maps will be used to predict dynamic biophysical variables, particularly leaf area index, at finer temporal scales. Trained on Earth observation data, the models learn to reproduce spatially explicit and physically consistent land surface patterns, extending temporal coverage to periods lacking direct observations. AI4Land was developed and trained on MareNostrum5, demonstrating how GPU-accelerated HPC infrastructure enables global-scale climate AI pipelines. The final product is a suite of open-source emulators designed for real-time coupling with digital twin platforms, such as those developed under the Destination Earth initiative. By delivering realistic and evolving land surface conditions on demand, this work aims to reduce critical uncertainties and improve the predictive power of next-generation climate simulations. more
-
MMClima: A Framework for Multimodal Climate Science Data and Evaluation¶
Climate change research increasingly requires AI systems that reason across text, dynamic visual content, and scientific figures, yet existing climate QA benchmarks are small, mostly textual, and...Climate change research increasingly requires AI systems that reason across text, dynamic visual content, and scientific figures, yet existing climate QA benchmarks are small, mostly textual, and cover a narrow range of models. We introduce MMClima, a large-scale multimodal climate question answering framework with 104k+ expert-validated question-answer pairs spanning articles, video transcriptions, and figures across five core climate science domains. MMClima is constructed via automated claim extraction and QA synthesis with human-in-the-loop validation to ensure both scale and reliability. Using MMClima, we benchmark state-of-the-art multimodal language models on tasks requiring factual recall, visual interpretation, and cross-modal synthesis. We additionally fine-tune on the textual split to produce mmclima-70b-txt, a domain-adapted baseline that outperforms strong open- and closed-source models on textual QA. We release the dataset, evaluation pipeline, fine-tuned model weights, and data creation framework to support standardized multimodal evaluation for climate science. more
-
U-Net-Accelerated Quality-Diversity Optimization for Climate-Adaptive Urban Layouts¶
Optimizing urban layouts for climate adaptation requires balancing building density with cold-air ventilation. Because physics-based climate simulations are computationally expensive, planners...Optimizing urban layouts for climate adaptation requires balancing building density with cold-air ventilation. Because physics-based climate simulations are computationally expensive, planners typically evaluate fewer than ten manual designs. qd algorithms offer a way to systematically illuminate the design space, but they require surrogate models to be practical. In this paper, we replace a slow, regulatory physics simulator with a spatial deep-learning surrogate (U-Net) inside an offline MAP-Elites loop. We systematically compare this spatial approach with a traditional gp surrogate across different training-data strategies (quasi-random Sobol sampling vs.\ active qd bootstrapping). Our results reveal that scalar gp surrogates fail catastrophically when trained on random samples, requiring expensive, actively generated qd archives to generalize. In contrast, the spatial inductive bias of the U-Net allows it to learn the underlying physics mapping robustly (\(R^2 = 0.996\)), completely independent of the training data source. This allows offline qd optimization to achieve highly accurate fitness rankings (\(ρ= 0.994\)) using only a one-time batch of random training samples. The resulting pipeline, deployed in the open-source OpenSKIZZE tool, generates thousands of diverse, climate-evaluated building layouts in under ten minutes. more
-
Probabilistic storyline attribution using machine learning¶
A fundamental goal in climate attribution is to estimate how forced climate change contributes to observed extreme weather events. The storyline attribution method compares an observed weather event,...A fundamental goal in climate attribution is to estimate how forced climate change contributes to observed extreme weather events. The storyline attribution method compares an observed weather event, conditional on its atmospheric dynamic state (i.e., atmospheric circulation), in the current, 'factual' climate to an event with very similar circulation conditions in a hypothetical, 'counterfactual' climate. However, physical climate models cannot directly transfer these storyline counterfactuals across different climate forcing states. Statistical and machine learning techniques may overcome this limitation; yet, emulating circulation-conditional extreme events under different climate states is challenging. Here, we demonstrate distributional autoencoders (DAEs) as a versatile method for generating climate counterfactuals. They model the full distribution of spatially resolved European temperature fields conditional on the atmospheric circulation state and the mean global warming level. These distributions allow for deriving meaningful conditional probability ratios, which is a particular advantage of the DAE-based storyline approach. We train DAEs on fully coupled climate model simulations and we evaluate the modelled distributions across different factual and storyline-based counterfactual climate model simulations. In an illustrative case study, we revisit the 2003 European heatwave and we generate counterfactuals for a hypothetical ‘2003-like European heatwave’ using ERA5 circulation, which we hypothesize to occur a quarter century (2028) and a half century (2053) after 2003. The conditional intensity would increase from 29.3 °C in 2003, to 30.3 °C and 32.1 °C in 2028 and 2053, respectively and conditional probability ratios would be 2.1 and 3.2 when compared to 2003. more
-
Evaluating Skill and Stability of ArchesWeather and ArchesWeatherGen under Multi-Decadal Climate Simulations¶
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....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. more
-
Probabilistic bias adjustment of seasonal forecasts using generative machine learning: A case study of Arctic sea ice predictions¶
Seasonal climate predictions support planning and risk management by offering early information of the most likely-to-occur climate conditions in the coming months, and associated uncertainties....Seasonal climate predictions support planning and risk management by offering early information of the most likely-to-occur climate conditions in the coming months, and associated uncertainties. Ensemble forecasts enable this by simulating many plausible outcomes, allowing predictions to be expressed as usable probabilities. Large ensembles and high-resolution forecasts strengthen this guidance by better sampling uncertainty and capturing finer-scale processes but come with significant computational cost. Moreover, forecast ensembles drift and exhibit systematic biases and spatio-temporal errors that grow with lead time, requiring careful post-processing and calibration. A probabilistic post-processing framework based on conditional Variational Autoencoders (cVAEs) was developed at the Canadian Center for Climate Modeling and Analysis to generate large ensembles of bias adjusted seasonal predictions of Arctic sea ice. The generative model was designed to learn the observational distribution conditioned on the biased model prediction. This enables generation of arbitrarily large ensembles of well-calibrated, bias corrected forecasts with improved skill. Here, we extend this framework to address the loss of fine-scale energy and the characteristic blurriness in predictions, a known limitation of standard cVAEs. Specifically, we employ a generator in place of the Gaussian parametrized decoder in the cVAE and use Continuous Ranked Probability Score in the objective function instead of the Mean Square Error. We further use a higher resolution target dataset compared to the raw forecast. We show that the adjusted forecasts are better calibrated, more consistent with the observational distribution, and exhibit smaller errors than benchmark predictions, while also enhancing the resolution of the raw forecasts and improving sharpness and spectral power relative to the standard cVAE. more
-
No Epoch Like the Present: Robust Climate Emulation Requires Out-of-Distribution Generalisation¶
Climate emulation is an out-of-distribution (OOD) projection task. This is precisely the challenge where modern Machine Learning (ML) methods are most prone to failure. Consequently, while current ML...Climate emulation is an out-of-distribution (OOD) projection task. This is precisely the challenge where modern Machine Learning (ML) methods are most prone to failure. Consequently, while current ML emulators trained on present climate achieve high in-distribution performance, their future reliability under the inevitable distribution shifts of a changing climate remains a critical, poorly understood blind spot. Addressing this challenge requires a fundamental shift in how we understand, evaluate, and design climate emulators. In this work, we first confirm that climate change drives a statistically significant and progressively growing shift in atmospheric state distributions, rendering standard evaluation protocols insufficient. We empirically establish that seasonal variation serves as an effective proxy for these long-term climate shifts, providing access to \(\textit{real-world}\) distribution shifts without recourse to heuristics like synthetic perturbations. Motivated by this link, we introduce a novel evaluation framework that leverages seasonal shifts as a rigorous, zero-overhead testbed for emulator robustness. Our systematic characterisation confirms that current state-of-the-art hybrid-ML emulators degrade significantly under these realistic shifts. Finally, we chart a path forward by identifying compositional generalisation, the ability to form novel combinations from observed elementary components, as a principled route towards robust climate emulation. We demonstrate that physically motivated decompositions substantially improve OOD performance with only modest trade-offs against in-distribution performance, providing an avenue towards ML-driven climate emulators robust to an unknown future. more
-
Blending machine learning and physics-based approaches for weather and climate: a typology¶
The integration of machine learning (ML) with traditional physics-based models is reshaping the landscape of weather and climate prediction. On their own, ML-based and physics-based approaches each...The integration of machine learning (ML) with traditional physics-based models is reshaping the landscape of weather and climate prediction. On their own, ML-based and physics-based approaches each have significant benefits - but also challenges. Deploying both these approaches side by side has the potential to accelerate the pull through of emerging science in a trusted and practical way. But there are many choices that can be made to how we "blend" ML and established physics-based modelling systems to get the optimal benefits. This paper aims to provide a typology of blended modelling approaches and discusses some of the strategic benefits that come with them. It can be used not just to classify modelling systems, but also identify routes to gradual, incremental or wholesale development and implementation of new and emerging capabilities. These approaches provide a practical path to innovation by combining the speed and adaptability of machine learning with the robustness, trust, and interpretability of physics-based systems. By adopting a structured vocabulary and outlining the benefits and limitations of each approach, this framework supports informed decision-making and strategic planning, and can be used by the wider community to navigate the transition to next-generation prediction systems. more
-
Spherical Harmonic Optimal Transport: Application to Climate Models Comparisons¶
Optimal transport provides a powerful framework for comparing measures while respecting the geometry of their support, but comes with an expensive computational cost, hindering its potential...Optimal transport provides a powerful framework for comparing measures while respecting the geometry of their support, but comes with an expensive computational cost, hindering its potential application to real world use cases. On manifolds, convolutional algorithms based on the heat kernel have been proposed to alleviate this cost, but their theoretical properties remain largely unexplored. We establish that the heat kernel cost converges to the optimal transport cost as time vanishes in the balanced and unbalanced cases. In the specific case of the 2-sphere \(\mathbb{S}^2\), we ensure that the associated Sinkhorn divergences retains the desirable geometric and analytic properties of classical optimal transport discrepancies. Moreover, we leverage the harmonic structure of the sphere to derive a fast Sinkhorn algorithm, requiring only \(\mathcal{O}(n)\) memory and \(\mathcal{O}(n^{3/2})\) time per iteration, with fully dense GPU-friendly operations. We validate its computational efficiency on synthetic data, and discuss its potential use in the evaluation of global climate models, providing both spatial and seasonal insights into models performances. more
-
Learning Displacement-Robust Representations for Landslide Early Warning under Rainfall Forecast Uncertainty¶
Rainfall-induced landslides pose a growing risk worldwide as climate change intensifies extreme rainfall events. To provide sufficient evacuation time, landslide early warning systems (LEWS) for...Rainfall-induced landslides pose a growing risk worldwide as climate change intensifies extreme rainfall events. To provide sufficient evacuation time, landslide early warning systems (LEWS) for real-time disaster monitoring must estimate near-future landslide risk by integrating observed rainfall with short-term rainfall forecasts from spatio-temporal environmental data streams. Although recent landslide prediction methods have improved predictive performance using statistical and deep learning approaches, most assume accurate rainfall inputs. In operational settings, however, landslide prediction relies on rainfall forecasts, which often contain spatial displacement of rainfall fields due to forecasting uncertainties. Such displacement can alter local accumulated rainfall and degrade prediction accuracy. To address this challenge, we propose a novel LEWS robust to rainfall field displacement. The key idea is to learn latent representations from rainfall and terrain data that remain stable under displacement in rainfall field motion, enabling reliable geospatial data integration for landslide risk estimation. The landslide prediction model is trained using Rainfall-Motion-Aware Contrastive Learning (RMCL), which introduces temporally correlated rainfall field perturbations to emulate forecast-induced displacement in rainfall-driven spatio-temporal environmental data streams. Experiments were conducted using two years of rainfall and terrain data across Japan, covering 19 regions with landslide events. The proposed system achieved up to 37% higher precision than state-of-the-art baselines. These results demonstrate that modeling rainfall as a moving spatial field and addressing rainfall field displacement during learning significantly improve the reliability of short-term landslide prediction in operational early warning systems. more
-
Emulating the Forced Response of Climate Models with Flow Matching¶
Global climate models are essential tools to simulate past and potential future pathways of climate change, as well as associated climate impacts. Shared Socioeconomic Pathways (SSPs) describe a...Global climate models are essential tools to simulate past and potential future pathways of climate change, as well as associated climate impacts. Shared Socioeconomic Pathways (SSPs) describe a range of future scenarios of global economic and demographic development. These SSPs are intrinsically linked to changes in climate forcings, the external drivers, such as greenhouse gas and aerosol emissions, which in turn lead to the human impact on the energy balance of the Earth over time. These forcings are fundamental boundary conditions in climate models in order to gain insight into the potential climatic impacts of these changes described by each SSP. Running a climate model, however, is extremely computationally expensive, conflicting with the need for large ensembles of simulations for each model to give, e.g., more robust estimates in the presence of internal variability (the inherent, chaotic fluctuations within the climate system) and scenario uncertainty. Recent research has demonstrated the ability to capture climate model dynamics using machine learning when conditioned on forcings from different climatic scenarios. We here train a Deep Learning (DL) model on multiple SSPs and successfully generate scenarios unseen during training. Our emulator is validated against MESMER-M, a statistical emulator of land surface temperature. Our research demonstrates the capacity to generate such changing climate states in response to a variety of simultaneous climate forcings (e.g., carbon dioxide, methane, nitrous oxide, sulphate aerosols, and ozone). In particular, our ablation studies underline a need to include a range of different forcings to represent long-term atmospheric trends with a DL emulator. more
-
Assessment of cloud and associated radiation fields from a GAN stochastic cloud subcolumn generator¶
Modern Earth System Models (ESMs) operate on horizontal scales far larger than typical cloud features, requiring stochastic subcolumn generators to represent subgrid horizontal and vertical cloud...Modern Earth System Models (ESMs) operate on horizontal scales far larger than typical cloud features, requiring stochastic subcolumn generators to represent subgrid horizontal and vertical cloud variability. Traditional physically-based generators often rely on analytical cloud overlap paradigms, such as exponential-random decorrelation, which can struggle to capture the complex, anti-correlated behavior of non-contiguous cloud layers. In this study, we introduce a novel two-stage machine learning subcolumn generator for the GEOS atmospheric model, utilizing a Conditional Variational Autoencoder combined with a Generative Adversarial Network (CVAE-GAN) and a U-Net architecture. Trained on a merged CloudSat-CALIPSO height-resolved cloud optical depth dataset, the ML generator creates 56 stochastic subcolumns representing cloud occurrence and optical depth profiles. Evaluated against the established Räisänen, the ML approach accurately reproduces bimodal cloud overlap distributions, significantly reduces biases in grid-mean statistics, and halves the root-mean-square error in ISCCP-style cloud-top pressure and optical thickness joint histograms. The improvements brought by our deep generative models translate into more accurate offline radiative transfer calculations, reducing the global-mean shortwave top-of-atmosphere cloud radiative effect bias by a factor of three. Provided that the generator can be accelerated on CPUs, this offers a practical pathway to reduce structural errors at the cloud-radiation interface. more
-
A Wasserstein GAN-based climate scenario generator for risk management and insurance: the case of soil subsidence¶
According to the United Nations Office for Disaster Risk Reduction (2025), the average annual cost of natural catastrophes increased from 70--80 billion USD between 1970 and 2000 to 180--200 billion...According to the United Nations Office for Disaster Risk Reduction (2025), the average annual cost of natural catastrophes increased from 70--80 billion USD between 1970 and 2000 to 180--200 billion USD between 2001 and 2020. Reports from organizations such as the IFOA and the WWF highlight the need for the insurance sector to adapt to this rapidly evolving context by developing medium- to long-term strategies that go beyond the one-year horizon of prudential regulations such as Solvency II. This paper introduces an artificial intelligence framework based on Conditional Generative Adversarial Networks (Conditional GANs) to generate future spatio-temporal trajectories of climatic indices. The approach focuses on the Soil Wetness Index (SWI), a key indicator used in France to assess drought severity. Drought accounts for approximately 30% of the indemnities paid under the French natural catastrophe insurance scheme. The proposed model, SwiGAN, simulates plausible drought propagation patterns up to 2050 for a region of France particularly exposed to this hazard. By generating realistic sequences of SWI maps, SwiGAN provides insights into drought dynamics under climate change scenarios and supports the design of adaptive risk management and insurance strategies. The methodology is also generalizable to other climate-related perils and actuarial applications such as economic scenario generation. more
-
Interpretable Neural Networks to Predict Momentum Fluxes of Orographic Gravity Waves¶
State-of-the-art Earth system models (ESMs) cannot explicitly resolve many small-scale atmospheric processes such as atmospheric gravity waves, and thus must represent, or parameterise, their effects...State-of-the-art Earth system models (ESMs) cannot explicitly resolve many small-scale atmospheric processes such as atmospheric gravity waves, and thus must represent, or parameterise, their effects on the resolved state. Machine learning (ML) has the potential to improve these parameterisations. In our study, we train neural networks (NNs) on ERA5 reanalysis data to predict momentum fluxes of orographic gravity waves as a function of the state variables at the resolution of a coarse ESM. Employing a full year of data, we extract inertia-gravity waves using the software MODES, which applies linear theory for wave filtering, and train ML models on data coarse-grained to the ESM's target resolution. We consider four different cases: the full spectrum of inertia-gravity waves resolved in ERA5, or just the part of the spectrum that is subgrid-scale in the target ESM, both over all land or just over mountainous terrain. Our NNs successfully predict momentum fluxes, with a global coefficient of determination (\(R^2\)) ranging from 0.72 to 0.56, depending on the case, when evaluated offline with data from another year. An analysis of our models using SHAP values, an explainable AI technique, suggests that the networks learned physically meaningful relationships. In addition, we give a comparison with the physics-based parameterisation scheme by Lott and Miller. This work forms the basis for the development of operational ML-based parameterisations to improve the representation of gravity waves and their effects in climate models. more
-
Probabilistic Classification and Uncertainty Quantification of Sahara Desert Climate Using Feedforward Neural Networks¶
Climate classification plays a vital role in agricultural planning, hydrological studies, and climate science. One of the most widely used systems for classifying global climate zones is the...Climate classification plays a vital role in agricultural planning, hydrological studies, and climate science. One of the most widely used systems for classifying global climate zones is the Köppen-Trewartha (KT) classification. However, the KT classification is fundamentally deterministic, offering discrete labels to spatial locations without accounting for uncertainties in classification. In this paper, we provide a framework for probabilistic modeling of climatic zones. We implement a feedforward artificial neural network (ANN) for classification, allowing for efficient, uncertainty-aware categorization of climatic regions, thereby offering a more nuanced understanding of transitional climate zones compared to traditional deterministic methods. We apply this method to the Sahara Desert region over the 30-year period of 1960 - 1989, using data at more than 400,000 space-time locations from the first 11 years to train our model. We assess the model's short- and long-term classification capabilities to evaluate its stability and accuracy over time. We also compare the probabilistic classification from our model with the traditional KT classification. In addition, we use fluctuation analysis methods to highlight the temporal evolution of climatic zones across the Sahara region and identify areas undergoing significant flux of probabilities of their climate classes, providing insights into broader trends in desertification. more
-
Towards accurate extreme event likelihoods from diffusion model climate emulators¶
ML climate model emulators are useful for scenario planning and adaptation, allowing for cost-efficient experimentation. Recently, the diffusion model Climate in a Bottle (cBottle) has been proposed...ML climate model emulators are useful for scenario planning and adaptation, allowing for cost-efficient experimentation. Recently, the diffusion model Climate in a Bottle (cBottle) has been proposed for generation of atmospheric states compatible with boundary conditions of solar position and sea surface temperatures. Crucially, cBottle can be guided to generate extreme events such as Tropical Cyclones (TCs) over locations of interest. Diffusion models such as cBottle work by approximating the probability density of the training data. Here, we show use cases of the probability density estimates of atmospheric states obtained from this climate emulator. Most importantly, these estimates allow us to calculate likelihoods of extreme events under guidance. When guiding the model towards states including TCs, comparing the probability density under the guided and unguided model enables us to quantify how much more likely the guidance has made the TC. We show how these odds ratios allow us to importance-sample from the TC distribution, reducing the standard error of the probability estimate compared to simple Monte Carlo sampling. Furthermore, we discuss results and limitations of the application of model probability densities to extreme event attribution-like experiments. We present these early but encouraging results hoping they will spur more research into probabilistic information that can be gained from diffusion models of the atmosphere. more
-
Physics-Informed Neural Learning for State Reconstruction and Parameter Identification in Coupled Greenhouse Climate Dynamics¶
Physics-informed neural networks (PINNs) have recently emerged as a promising framework for integrating data-driven learning with physical knowledge. In this work, we propose a coupled PINN approach...Physics-informed neural networks (PINNs) have recently emerged as a promising framework for integrating data-driven learning with physical knowledge. In this work, we propose a coupled PINN approach for the joint reconstruction of indoor temperature and humidity dynamics in greenhouse environments, together with simultaneous identification of key model parameters. The method incorporates a reduced-order physically motivated model into the learning process, enabling consistent estimation under sparse and noisy observations. The artificial intelligence contribution lies in the development of a coupled physics-informed neural learning framework that integrates governing dynamical constraints into neural network training, while the engineering application focuses on greenhouse climate state reconstruction and parameter identification. The proposed framework is evaluated on a controlled synthetic benchmark that mimics diurnal forcing conditions. Compared with a purely data-driven neural network baseline, the coupled PINN achieves improved reconstruction accuracy, reducing temperature and humidity errors while maintaining high coefficients of determination. The improvement is particularly pronounced in the humidity channel, where latent moisture dynamics are more difficult to infer from limited measurements. In addition to accurate state reconstruction, the method successfully recovers the dominant physical parameters governing the system dynamics, demonstrating its ability to learn interpretable representations beyond data interpolation. These results highlight the potential of physics-informed learning for greenhouse climate modeling and, more broadly, for data-scarce environmental systems. more
-
Leveraging Climate Services to Build Climate Resilient Power Systems¶
We explore the crucial interplay between climate change and power system planning, highlighting the urgent need to systematically integrate climate information into energy system studies. Climate...We explore the crucial interplay between climate change and power system planning, highlighting the urgent need to systematically integrate climate information into energy system studies. Climate change impacts the energy sector on multiple fronts. Short-term weather variability drives daily and seasonal fluctuations in supply and demand. Long-term trends and increased frequency of extremes pose risks to infrastructure performance, asset lifetimes, and system adequacy. Representing compound events and spatial correlations across borders is a complex challenge, and uncertainties persist due to uncertainties from different models, scenarios, and downscaling methodologies. The Pan-European Climate Database (PECD4.2), developed in partnership between ENTSO-E and C3S, marks a change in how energy system planning is conducted. The PECD4.2 integrates historical reanalysis and six climate models across four SSP's, providing harmonised, openly available datasets tailored for power system studies. The physical conversion models for wind and solar energy better reflect technological progression than machine learning methods trained on historical data, improving robustness under changing future conditions. Despite these advances, challenges remain. Particularly in hydropower modelling and the lack of public harmonised energy datasets that are required to train these models. Complex processing chains from raw climate data to actionable insights and the lack of standardized integration of climate information lengthen lead times for energy-sector adoption. This leads to diverging approaches and variable consideration of climate risks. Closer, more generalised collaboration and communication between climate service providers and energy stakeholders are therefore necessary, as are the development of user-friendly tools for data manipulation and analysis and robust feedback loops. more
-
Amplified Urban Climate Extremes from Global Warming-Urbanization Synergy: A Physics-Informed Intelligence Paradigm¶
The nonlinear synergy between global warming and urbanization is amplifying extreme climate risks in cities worldwide. While observations and simulations confirm these compounding effects, two...The nonlinear synergy between global warming and urbanization is amplifying extreme climate risks in cities worldwide. While observations and simulations confirm these compounding effects, two fundamental bottlenecks impede predictive understanding: (1) fragmented, case-specific perspectives that hinder the discovery of universal mechanisms, and (2) a methodological divide between computationally prohibitive high-resolution models and AI-based tools that lack physical interpretability at urban scales. This article advocates for a paradigm shift toward the deep integration of physical principles with data intelligence. To this end, we propose a transformative "Classification-Mechanism-Inference" (CMI) framework. Classification involves establishing a global urban "climate-morphology-development" typology to enable systematic comparison beyond isolated case studies. Mechanism advocates for physics-informed machine learning (PIML) as the core engine to develop efficient, physics-constrained surrogate models for uncovering nonlinear interactions. Inference leverages these models for high-throughput, tailored risk projection to directly inform context-specific adaptation planning. The CMI framework aims to bridge the cognitive and methodological gaps, thereby advancing urban climate science from phenomenological description towards mechanistic, predictive, and decision-relevant science, which is crucial for building climate-resilient cities globally. more
-
Deep Clustering for Climate: Analyzing Teleconnections through Learned Categorical States¶
Understanding and representing complex climate variability is essential for both scientific analysis and predictive modeling. However, identifying meaningful climate regimes from raw variables is...Understanding and representing complex climate variability is essential for both scientific analysis and predictive modeling. However, identifying meaningful climate regimes from raw variables is challenging, as they exhibit high noise and nonlinear dependencies. In this work, we explore the use of Masked Siamese Networks to discretize climate time series into semantically rich clusters. Focusing on daily minimum and maximum temperature, we show that the resulting representations: (i) yield clusters that reflect meaningful climate states under our modeling assumptions, offering a simplified representation for downstream use; (ii) enable sampling and analysis of specific climate scenarios; and (iii) exhibit statistical associations with El Niño events, underscoring their scientific relevance. Our findings highlight the potential of self-supervised discretization as a tool for climate data analysis and open avenues for incorporating richer climate indicators in future work. more
-
Assessing Emulator Design and Training for Modal Aerosol Microphysics Parameterizations in E3SMv2¶
Toward the goal of using Scientific Machine Learning (SciML) emulators to improve the numerical representation of aerosol processes in global atmospheric models, we explore the emulation of aerosol...Toward the goal of using Scientific Machine Learning (SciML) emulators to improve the numerical representation of aerosol processes in global atmospheric models, we explore the emulation of aerosol microphysics processes under cloud-free conditions in the 4-mode Modal Aerosol Module (MAM4) within the Energy Exascale Earth System Model version 2 (E3SMv2). To develop an in-depth understanding of the challenges and opportunities in applying SciML to aerosol processes, we begin with a simple feedforward neural network architecture that has been used in earlier studies, but we systematically examine key emulator design choices, including architecture complexity and variable normalization, while closely monitoring training convergence behavior. Our results show that optimization convergence, scaling strategy, and network complexity strongly influence emulation accuracy. When effective scaling is applied and convergence is achieved, the relatively simple architecture, used together with a moderate network size, can reproduce key features of the microphysics-induced aerosol concentration changes with promising accuracy. These findings provide practical clues for the next stages of emulator development; they also provide general insights that are likely applicable to the emulation of other aerosol processes, as well as other atmospheric physics involving multi-scale variability. more
-
climt-paraformer: Stable Emulation of Convective Parameterization using a Temporal Memory-aware Transformer¶
Accurate representation of moist convective sub-grid-scale processes remains a major challenge in global climate models, as traditional parameterization schemes are both computationally expensive and...Accurate representation of moist convective sub-grid-scale processes remains a major challenge in global climate models, as traditional parameterization schemes are both computationally expensive and difficult to scale. Neural network (NN) emulators offer a promising alternative by learning efficient mappings between atmospheric states and convective tendencies while retaining fidelity to the underlying physics. However, most existing NN-based parameterizations are memory-less and rely only on instantaneous inputs, even though convection evolves over time and depends on prior atmospheric states. Recent studies have begun to incorporate convective memory, but they often treat past states as independent features rather than modeling temporal dependencies explicitly. In this work, we develop a temporal memory-aware Transformer emulator for the Emanuel convective parameterization and evaluate it in a single-column climate model (SCM) under both offline and online configurations. The Transformer captures temporal correlations and nonlinear interactions across consecutive atmospheric states. Compared with baseline emulators, including a memory-less multilayer perceptron and a recurrent long short-term memory model, the Transformer achieves lower offline errors. Sensitivity analysis indicates that a memory length of approximately 100 minutes yields the best performance, whereas longer memory degrades performance. We further test the emulator in long-term coupled simulations and show that it remains stable over 10 years. Overall, this study demonstrates the importance of explicit temporal modeling for NN-based parameterizations. more
-
Connecting the forward problem to the inverse problem in uncertainty quantification of Earth system models using fast emulators¶
Quantifying and reducing uncertainty in Earth system model parameterizations is essential to improving their reliability in decision-making. Forward uncertainty propagation is used to derive...Quantifying and reducing uncertainty in Earth system model parameterizations is essential to improving their reliability in decision-making. Forward uncertainty propagation is used to derive parameter sensitivity but requires physically plausible parameter distributions first be learned from observations. Bayesian inference offers a principled approach but can become ill-posed when observations weakly constrain parameters--a condition difficult to know prior to inference. Addressing this gap, we show that parameter sensitivity results from forward uncertainty quantification can guide a non-iterative strategy for identifying observations informative to Bayesian calibration. We explore both forward and inverse uncertainty quantification for parameterizations of atmospheric turbulence in the Weather Research and Forecasting (WRF) model. To overcome the computational bottleneck of \(\mathcal{O}(10^5)\) model evaluations required for both analyses, we leverage Gaussian process emulators trained on several hundred WRF simulations. Using these emulators, we conduct a global sensitivity analysis across observation space, investigating how parameter contributions to output variance depend on quantity of interest, atmospheric stability, time-averaging length, and spatial location. We then introduce nondimensional diagnostic measures that systematically identify regions where a parameter's contribution to output variance exceeds observational noise and its independent effect exceeds interaction effects. We demonstrate that observations from these regions serve as a strong proxy for accurate Bayesian calibration and reduced posterior uncertainty. Through emulator-aided Bayesian inversion with synthetic observations, we show how parameter uncertainty can be systematically reduced by leveraging sensitivity information. more
-
AI-based Waste Mapping for Addressing Climate-Exacerbated Flood Risk¶
Urban flooding is a growing climate change-related hazard in rapidly expanding African cities, where inadequate waste management often blocks drainage systems and amplifies flood risks. This study...Urban flooding is a growing climate change-related hazard in rapidly expanding African cities, where inadequate waste management often blocks drainage systems and amplifies flood risks. This study introduces an AI-powered urban waste mapping workflow that leverages openly available aerial and street-view imagery to detect municipal solid waste at high resolution. Applied in Dar es Salaam, Tanzania, our approach reveals spatial waste patterns linked to informal settlements and socio-economic factors. Waste accumulation in waterways was found to be up to three times higher than in adjacent urban areas, highlighting critical hotspots for climate-exacerbated flooding. Unlike traditional manual mapping methods, this scalable AI approach allows city-wide monitoring and prioritization of interventions. Crucially, our collaboration with local partners ensured culturally and contextually relevant data labeling, reflecting real-world reuse practices for solid waste. The results offer actionable insights for urban planning, climate adaptation, and sustainable waste management in flood-prone urban areas. more
-
Enhancing AI and Dynamical Subseasonal Forecasts with Probabilistic Bias Correction¶
Decision-makers rely on weather forecasts to plant crops, manage wildfires, allocate water and energy, and prepare for weather extremes. Today, such forecasts enjoy unprecedented accuracy out to two...Decision-makers rely on weather forecasts to plant crops, manage wildfires, allocate water and energy, and prepare for weather extremes. Today, such forecasts enjoy unprecedented accuracy out to two weeks thanks to steady advances in physics-based dynamical models and data-driven artificial intelligence (AI) models. However, model skill drops precipitously at subseasonal timescales (2 - 6 weeks ahead), due to compounding errors, systemic model biases, and the chaotic nature of the atmosphere. To counter this degradation, we introduce probabilistic bias correction (PBC), a machine learning framework that substantially reduces systematic error by learning to correct historical probabilistic forecasts. When applied to the leading dynamical and AI models from the European Centre for Medium-Range Weather Forecasts (ECMWF), PBC doubles the modest subseasonal skill of the AI Forecasting System and improves the skill of the operationally-debiased dynamical model for 91% of pressure, 92% of temperature, and 98% of precipitation targets. We designed PBC for operational deployment, and, in ECMWF's 2025 real-time forecasting competition, its global forecasts placed first for all weather variables and lead times, outperforming the dynamical models from six operational forecasting centers, an international dynamical multi-model ensemble, ECMWF's AI Forecasting System, and the forecasting systems of 34 teams worldwide. These probabilistic skill gains translate into more accurate prediction of extreme events and have the potential to improve agricultural planning, energy management, and disaster preparedness in vulnerable communities. more
-
Capturing Aleatoric Uncertainty in Climate Models¶
Internal climate variability arises from the climate system's inherently chaotic dynamics. Quantifying it is essential for climate science, as it enables risk-based decision-making and differentiates...Internal climate variability arises from the climate system's inherently chaotic dynamics. Quantifying it is essential for climate science, as it enables risk-based decision-making and differentiates between externally forced change and internal fluctuations. In statistical terms, natural variability corresponds to aleatoric uncertainty, i.e., irreducible stochastic variability. Despite this close conceptual alignment, the link between internal climate variability and aleatoric uncertainty has not yet been formalized. We establish a theoretical link by showing that member-to-member differences in single-model large ensembles provide a direct representation of aleatoric uncertainty. To quantify the spatio-temporal structure of aleatoric uncertainty, we employ generalized additive models. The proposed framework is validated through comparison with ERA5-Land reanalysis data, demonstrating that ensemble-derived estimates reproduce key spatial and temporal patterns of real-world variability. Applied to the water balance over the Iberian Peninsula, our approach reveals coherent variability structures and pronounced regional heterogeneity. We find a decline in variability in drought-prone regions and seasons, a pattern that strengthens under +3 °C global warming, implying an increased risk of persistent summer drought conditions. Beyond this application, the framework is climate-model agnostic and transferable to other variables and spatial scales, providing a statistical basis for quantifying internal climate variability as aleatoric uncertainty. more