Papers for

weather prediction teams

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Forecast nonlinear systems with calibrated probabilistic Koopman models

Learning calibrated probabilistic forecasts using Koopman mode decomposition and the continuous ranked probability score

Abstract: Well-calibrated forecasting for non-linear dynamical systems is essential in science and engineering applications such as fluid dynamics and weather prediction. We present a forecasting method combining probabilistic machine learning with the structured linear dynamics of Koopman mode decomposition. Within this framework, the observation, linear dynamics, and state reconstruction operators are constructed using standard machine learning architectures, producing predictive distributions via the pushforward of injected noise samples. The model is trained using a proper scoring rule, the continuous ranked probability score, to directly target calibrated, accurate forecasts. The approach is validated on two non-linear ordinary differential equation examples and two partial differential equation examples, vortex shedding and cavity flow, using both large and small, noisy datasets. Results demonstrate that known Koopman eigenfunctions and spectra can be identified and used for accurate forecasting. The model can also produce multi-modal predictive distributions, demonstrated for a trajectory initialised on the boundary between two basins of attraction. Uncertainty is captured by the sample ensemble spread, with spread--skill ratios close to one over the training horizon in several experiments, although for extended rollouts and noisy datasets, the ensemble may become under- or over-dispersed. In low-noise experiments, temporal rollout is generally accurate for at least twice the training horizon, and up to twenty times in some cases. Approximation quality is comparable to a deep ensemble approximation over the training horizon, but at a reduced computational cost and with better robustness in identifying Koopman eigenvalues.

Sun 8 NovMachine Learning
The gist
Predicting how complex systems change over time, like weather or fluids, is tricky because they behave nonlinearly and unpredictably. The authors combine a method called Koopman mode decomposition with machine learning to create probabilistic forecasts that not only predict future states but also express uncertainty accurately. Their approach learns to produce trustworthy probability distributions of possible outcomes, improving reliability especially when data is noisy or limited. They tested this on several simulated systems, including fluid flows, showing it can capture different future possibilities and remain accurate for extended periods.
Open → 2611.11257v1

Weather downscaling improved by learning operators and normalizing guidance

Operator-Aware Posterior Sampling for Multivariate Weather Downscaling

Abstract: Statistical downscaling is the task of reconstructing high-resolution meteorological fields from coarse inputs. A principled approach formulates downscaling as a Bayesian inverse problem, using a diffusion prior over high-resolution fields and incorporating the coarse field only at inference through a likelihood. This decouples the prior from the coarse source, allowing one pretrained model to adapt to different sources. However, two challenges limit this flexibility: (1) when coarse and target fields come from different reanalysis systems, no fixed coarsening operator can capture their relationship; and (2) guidance gradients can differ by orders of magnitude across variables, making a single guidance strength ineffective. This paper addresses both challenges within the guidance loop, leaving the prior unchanged. We learn the forward operator from paired data and normalize the gradient per channel. We show that this normalization corresponds to steepest descent under a block-max norm, allowing a single guidance strength across variables. Our code is publicly available at https://github.com/CarloSaccardi/operator-aware-weather-downscaling.

Thu 29 OctComputer Vision and Pattern Recognition
The gist
It is hard to turn low-resolution weather data into detailed, high-resolution weather maps, especially when data sources vary. The authors found problems with current methods when the data comes from different systems and when different weather variables influence the process unevenly. They fixed these by teaching the model how to translate between data sources and by adjusting how the model balances different weather data. Their improvements work without changing the fundamental model used to create the detailed maps.
Open → 2610.39592v1

Local sparse methods improve fluid flow predictions using kernel learning

Local and Sparse Kernel Operator Learning for Incompressible Flows: An Operator Lagrange Formulation

Abstract: We present a local and sparse kernel method for learning the solution operators of incompressible flows. It builds on our earlier property-preserving kernel method, whose global output reconstruction and dense operator regression scale poorly. Our method replaces the global output reconstruction by local divergence-free interpolants with divergence-free polynomials of any degree, solved in parallel on the GPU. These interpolants are divergence free, periodic on periodic domains, and, with a multiscale kernel, approach Kolmogorov statistics under refinement. A compactly supported kernel, positive semidefinite in any input dimension, makes the operator regression sparse. Its support radius, and hence the sparsity, is fitted to data, and optional affine polynomials in the input capture global trends. Operator Lagrange weights join the two. They cost one sparse solve per new input and act on the training fields before the output reconstruction, so their local coefficients are never formed. We show that, unlike direct evaluation, Lagrange evaluation does not amplify solve errors by the cancellation among spatial coefficients. On ten fluid benchmarks, validation selects sparse matrices nearly everywhere, with densities that mostly decrease as the training set grows, and affine augmentation lowers the error by more than an order of magnitude for forced isotropic turbulence. These sparse models are at least as accurate as our earlier dense method and more accurate than neural operators. Lagrange evaluation is as accurate as direct evaluation with local output reconstructions, and up to nearly three orders of magnitude more accurate with global ones. We also present GPU timings and storage.

Wed 21 OctMachine Learning
The gist
Simulating how fluids move and flow is important but hard because it needs to follow special rules like incompressibility. The authors developed a new way to learn these fluid behaviors using local and sparse mathematical tools that keep these rules intact. Their method uses small, manageable pieces that work together, and can run fast on GPUs. This approach reduces errors and storage compared to earlier methods and works well on many fluid flow examples.
Open → 2610.29009v1

Mesh-free convolution improves long-term predictions for spatial equations

Mesh-Free Convolution: Learned Spectral Attenuation and Transport

Abstract: Learning solution operators for partial differential equations (PDEs) requires models that can represent spatial structure across grids and scales. This becomes especially important in time-dependent problems, where the learned operator is often used autoregressively: each predicted state is fed back as input to predict future states. In this setting, small spatial errors can be repeatedly propagated through the learned dynamics, leading to instability and drift, vitiating long-term solution fidelity. We propose Mesh-Free Convolutions (MFCs), filters defined in continuous space and discretized only at evaluation time. This allows the same learned filters to be applied across resolutions. Rather than imposing a grid-dependent stencil or a fixed spectral cutoff, MFCs learn smooth scale-dependent filtering and include directional terms for transport-like behavior. Across fluid-dynamics and weather-forecasting benchmarks, MFC-based models produce stable long-horizon rollouts, transfer reliably across resolutions, and better preserve long-time statistics. These results suggest that continuous, mesh-free filtering is a useful inductive bias for PDE operator learning, especially in autoregressive prediction.

Sun 11 OctMachine Learning
The gist
Predicting how things like fluids or weather change over time can be tricky because small mistakes can build up quickly. This paper introduces a new method called Mesh-Free Convolutions that works smoothly across different scales and resolutions without relying on fixed grids. The authors show that their method makes long-term predictions more stable and accurate for problems like fluid dynamics and weather forecasting. This approach helps models remember complex spatial patterns better over time.
Open → 2610.15931v1

Scalable method improves high dimensional Gaussian dependency estimation

SCORE: Spectral Correlation Estimation for Multivariate Gaussians

Abstract: Neural network-based predictive modeling with high-dimensional structured Gaussian targets requires an efficient and numerically stable, yet expressive approximation of the covariance matrix. We propose SCORE: a scalable framework, combining scoring rule training with an expressive covariance approximation learned in spectral space. For $d$-dimensional data, the learning task is decomposed into learning the marginal distributions and learning a structured correlation matrix, which enables dense dependencies with linear storage and $\mathcal{O}(d\log d)$ cost. We utilize the closed form Gaussian kernel score for training, which remains defined even for degenerate covariances and admits bounded gradients during optimization. We characterize kernel scores under invertible transforms and prove exact invariance under unitary transforms. At population level, our two-level objective recovers the true marginals and projects the target correlation onto the representable class; finite-sample PAC bounds show that the errors of the two stages enter additively. We evaluate our model on a variety of tasks with a commonly assumed Gaussian domain: Time-series forecasting, monocular depth estimation, and spatial weather prediction, showing improved performance at lower computational cost.

Thu 8 OctMachine Learning
The gist
Modeling complex relationships in data that follow a Gaussian pattern requires estimating how variables relate in a stable and efficient way. The authors propose SCORE, a method that breaks down the problem to learn simple parts of the data and its correlations separately, using a spectral approach that is fast and uses less memory. Their method can be trained reliably even when some data aspects degenerate and ensures consistent results across certain transformations. They tested it on tasks like predicting weather, estimating depth from images, and forecasting time series, finding better accuracy while saving computation.
Open → 2610.12096v1

Spherical spiral network improves global weather forecast accuracy

S$^3$N: A Spherical Spiral Scanning Network for Weather Forecasting

Abstract: Machine learning-based weather prediction (MLWP) has achieved strong performance in global weather forecasting. Recent Hierarchical Equal Area isoLatitude Pixelation (HEALPix)-based methods use the HEALPix (HP) grid to avoid area distortion near the poles of conventional latitude-longitude (LL) grids. However, existing HP-based approaches often use pointwise mapping methods and process HP pixels within separate base faces or local windows. Consequently, the mapping may introduce reconstruction errors and cross-face communication depends on handcrafted boundary handling or shifted windows. We propose the Spherical Spiral Scanning Network (S$^3$N) to address both limitations. First, L2Proj provides a bidirectional method for mapping atmospheric fields between the LL and HP grids through an $L^2$ projection of their continuous finite-element representations. Second, the Attention-Guided Quad-Spiral State-Space Scanning (AQSS) block uses cross-latitude attention to guide selective state-space updates along four global pole-to-pole spiral paths. This design enables continuous information propagation across HP base-face boundaries without additional boundary-processing mechanisms. Experiments show that S$^3$N achieves better results at 4-, 7-, and 10-day lead times, and exhibits slower error growth in long-range forecasting.

Sat 3 OctMachine Learning
The gist
Weather forecasting models use grids to divide the Earth and make predictions, but traditional grids can cause errors near the poles. The authors introduced a new method that uses a spherical spiral pattern to better connect grid points and share data across regions smoothly. They also created a precise way to convert data between curved and flat grid views. Their approach leads to more accurate weather forecasts up to 10 days in advance with slower error growth over time.
Open → 2610.04338v1

Probabilistic forecasting improves time series predictions using Koopman autoencoders

Variational Augmented Invertible Koopman Autoencoder for probabilistic time series forecasting

Abstract: Neural Koopman autoencoder models have been shown to successfully build a latent embedding with linear dynamics for arbitrary dynamical systems, enabling strong performance in long-term time series forecasting. However, these models usually work in a deterministic setting, which does not allow the quantification of the uncertainty of their predictions. Thus, we propose the new Variational Augmented Invertible Koopman AutoEncoder (VAIKAE), in which the latent embedding follows a Gaussian distribution instead of being deterministic. A key property of the VAIKAE architecture is that it leverages normalizing flow models, enabling the use of likelihood computations in the state space of dynamical systems for training a model. We further propose new strategies for uncertainty-aware latent data assimilation with a trained VAIKAE model. The effectiveness of our methods is demonstrated in a series of experiments on long-term time series forecasting benchmarks.

Tue 29 SeptMachine Learning
The gist
Making good long-term predictions from time series data often assumes the future can be guessed precisely, but uncertainty is common. This paper proposes a method that uses a special kind of neural network to not only predict future states but also estimate how confident those predictions are. By treating the hidden parts of the system as probabilities rather than fixed values, the method can better handle the unpredictability of real-world data. The authors also introduce ways to update their model’s predictions as new information comes in, improving accuracy over time.
Open → 2609.37435v1

Machine learning weather models struggle with energy flow and error growth patterns

Butterfly Effect and the Kinetic Energy Cascade in Probabilistic Machine Learning Weather Prediction Models

Abstract: This study analyses kinetic energy (KE) spectra, difference kinetic energy (DKE) spectra, and signatures of KE transfer across spatial scales in four state-of-the-art probabilistic machine learning weather prediction (MLWP) models: NeuralGCM-ENS, FourCastNet 3, AIFS-ENS, and GenCast. Results are compared with those from the physics-based numerical weather prediction model IFS-ENS. While NeuralGCM-ENS successfully reproduces the expected upscale transfer of KE, noise injection at its encoder stage underestimates mesoscale KE. Conversely, AIFS-ENS, GenCast, and FourCastNet 3 produce realistic KE spectral magnitudes but do not capture the expected upscale transfer of KE. In particular, AIFS-ENS and GenCast, which employ spatially uncorrelated stochastic perturbations, exhibit enhanced accumulation of KE at high wavenumbers. All examined models exhibit upscale error growth, reflected by the progressive shift of the DKE spectral peak toward larger wavelengths over time. However, the MLWP models struggle to reproduce the rapid initial growth of ensemble spread at small spatial scales associated with the butterfly effect. The results show that MLWP models can misrepresent the known scale transfer of kinetic energy despite producing skilful weather forecasts.

Wed 16 SeptMachine Learning
The gist
The study looks at how well several advanced machine learning weather models handle the flow of energy at different scales in the atmosphere and how errors spread out over time. The authors find that while some models mimic certain energy patterns well, others do not capture important ways energy moves between scales. All models have trouble showing the quick early spread of small errors known as the butterfly effect. This means that even though the machine learning models can predict weather well, they don't fully represent how energy and errors behave in the real atmosphere.
Open → 2609.18489v1