Explore the vast range of titles available.

Ensemble weather forecasts provide a probabilistic description of the future state of the atmosphere and give users flow-dependent estimates of forecast uncertainty. Here, we introduce AIFS-CRPS, an ensemble variant of the machine-learned Artificial Intelligence Forecasting System (AIFS) developed at ECMWF. Its loss function is the almost fair Continuous Ranked Probability Score (afCRPS). It is based on a proper score, the CRPS, but approximately removes the bias in the score due to finite ensemble size yet avoids a degeneracy of the fair CRPS. The trained model is stochastic and can generate as many exchangeable members as desired. For medium-range forecasts AIFS-CRPS outperforms the physics-based Integrated Forecasting System (IFS) ensemble for the majority of variables and lead times. For subseasonal forecasts, AIFS-CRPS outperforms the IFS ensemble before calibration and is competitive with the IFS ensemble when forecasts are evaluated as anomalies to remove the influence of model biases.

We present version 1.1.0 of ECMWF's Artificial Intelligence Forecasting System (AIFS Single), operational since 25 February 2025. The revised system introduces a bounding-layer framework that enforces physical constraints, such as non-negativity and internal consistency within precipitation and cloud cover variables, alongside expanded training data, revised loss weighting, and an extended set of surface and atmospheric variables. Overall skill improves by 4 %–6 % in the upper air and near-surface variables without degradation of spatial variability. A controlled comparison shows that training data expansion is the dominant source of upper-air skill gains, highlighting the importance of frequent model updates. The bounding framework delivers the largest precipitation improvements, up to 12 % and an approximately 1 d advantage using a categorical measure of skill. We further show that enforcing precipitation non-negativity resolves a gradient ambiguity at the zero-precipitation boundary under MSE training, explaining the reduction in drizzle bias and the improvements in precipitation.

V2028 Cyg shows a B[e] phenomenon. Due to the presence of both cool (K7III) and hot (B4III) components in the spectra, it is supposed to be a binary. Our modelling of the time variability of Hα line bisectors shows, however, that this hybrid spectrum can originate in one star, which is surrounded by a disc.

We introduce a probabilistic diffusion-based method for global atmospheric downscaling implemented within the Anemoi framework. The approach transforms low-resolution ensemble forecasts into high-resolution ensembles by learning the conditional distribution of finer-scale residuals, defined as the difference between the high-resolution fields and the interpolated low-resolution inputs. The system is trained on reforecast pairs from ECMWF IFS, using coarse fields at 100 km to reconstruct fine-scale variability at 30 km resolution. The bulk of the training focuses on recovering small-scale structures, while fine-tuning in high-noise regimes enables the generation of extremes. Evaluation against the medium-range IFS ensemble target shows that the model increases probabilistic skill (FCRPS) for surface variables, reproduces target power spectra at small scales, captures physically consistent multivariate relationships such as wind-pressure coupling, and generates extreme values consistent with those of the target ensemble in tropical cyclones.

We formulate an XAI-based model improvement approach for Graph Neural Networks (GNNs) for node classification, called Explanation Enhanced Graph Learning (EEGL). The goal is to improve predictive performance of GNN using explanations. EEGL is an iterative self-improving algorithm, which starts with a learned \"vanilla\" GNN, and repeatedly uses frequent subgraph mining to find relevant patterns in explanation subgraphs. These patterns are then filtered further to obtain application-dependent features corresponding to the presence of certain subgraphs in the node neighborhoods. Giving an application-dependent algorithm for such a subgraph-based extension of the Weisfeiler-Leman (1-WL) algorithm has previously been posed as an open problem. We present experimental evidence, with synthetic and real-world data, which show that EEGL outperforms related approaches in predictive performance and that it has a node-distinguishing power beyond that of vanilla GNNs. We also analyze EEGL's training dynamics.

Abstract We present a preliminary analysis of our six-year observation campaign of the B[e] stellar system V2028 Cyg (MWC 623). The time variability of spectral features is described. [PUBLICATION ABSTRACT]