arXiv subjects

Daniel Holmberg

Publications and source records attributed to Daniel Holmberg.

9 recordsLinked to original sources

arXiv2026

Decadal wave reconstruction in the Mediterranean Sea with graph neural networks

Accurate simulation and prediction of ocean waves are essential for coastal risk management and climate studies. Deep learning has shown promising results for wave modeling, but most approaches still operate on regular grids and on forecasting time scales, and do not generalize to unstructured discretization or to long time horizons. Here we present WaveGraph, a model based on Graph Neural Networks (GNNs) that emulates basin-scale wave dynamics directly on unstructured meshes with high resolution along the coasts (up to 2-3 km). Trained on bias-corrected simulation data over the Mediterranean Sea, WaveGraph uses a multiscale architecture combining the unstructured model mesh with a uniform graph, allowing simultaneous representation of local coastal interactions and large-scale wave dynamics. The model reconstructs the evolution of significant wave height, mean period, and mean direction, and is applied autoregressively for a continuous 17-year period without reinitialization or drift. Validation against buoy and satellite observations shows skill comparable to the input data set, and ablation experiments indicate that wind forcing drives most of the long-term stability while wave history improves swell-driven and basin-scale dynamics. These results show that GNNs can provide stable and efficient emulators of spectral wave models on unstructured domains, enabling decadal wave reconstructions.

physics.ao-ph↗

arXiv2026

Njord: A Probabilistic Graph Neural Network for Ensemble Ocean Forecasting

Ocean dynamics are inherently chaotic, yet existing machine learning ocean models produce only deterministic forecasts. We introduce Njord, a probabilistic data-driven model for ocean forecasting, applicable to both global and regional domains. Njord combines a deep latent variable framework with a graph neural network architecture, enabling sampling each forecast step in a single forward pass. We apply Njord globally at 0.25{\deg} resolution and regionally to the Baltic Sea at 2 km resolution. To scale to these large ocean grids we introduce K-means cluster meshes that adapt to irregular sea surface geometry. Experiments demonstrate strong performance on both domains compared to deterministic machine learning baselines, while also providing uncertainty estimates from the sampled ensemble forecasts. On the global OceanBench benchmark, Njord achieves the lowest errors on average across upper-ocean variables when evaluated against real-world observations, with the largest improvements in surface temperature prediction.

cs.LG↗

arXiv2026

Deterministic and probabilistic neural surrogates of global hybrid-Vlasov simulations

Hybrid-Vlasov simulations resolve ion-kinetic effects in the solar wind-magnetosphere interaction, but even 5D (2D + 3V) configurations are computationally expensive. We show that graph-based machine learning emulators can learn the spatiotemporal evolution of electromagnetic fields and lower-order moments of the ion velocity distribution function in near-Earth space from four 5D Vlasiator runs, each driven by steady solar wind conditions. The upstream ion number density is systematically varied between the runs, while the grid spacing is held constant, to scan the ratio of ion inertial length to grid size. Using a graph neural network (GNN) operating on the 2D spatial simulation grid comprising 670k cells, we demonstrate that both a deterministic forecasting model (Graph-FM) and a probabilistic ensemble forecasting model (Graph-EFM) based on a latent variable formulation produce accurate predictions of future plasma states. A divergence penalty is incorporated to encourage divergence-freeness in the magnetic fields. For the probabilistic model, a continuous ranked probability score objective is added to improve the calibration of the ensemble forecasts. In terms of wall time per output step, the trained emulators run over two orders of magnitude faster on a single GPU than the Vlasiator simulations on 100 CPUs. Most forecasted fields have Pearson correlations above 0.95 at 50 seconds lead time. Fields that exhibit degenerate (near-zero) distributions in the 5D setting are more challenging for the emulator to keep well correlated. The ensemble forecasts remain underdispersive, with spread-skill ratios of approximately 0.2-0.3, and thus provide spatially structured relative uncertainty estimates. Overall, these results demonstrate that GNNs provide a viable framework for rapid ensemble generation in hybrid-Vlasov modeling and highlight promising directions for future work.

physics.space-ph↗

arXiv2025

Graph-based Neural Space Weather Forecasting

Accurate space weather forecasting is crucial for protecting our increasingly digital infrastructure. Hybrid-Vlasov models, like Vlasiator, offer physical realism beyond that of current operational systems, but are too computationally expensive for real-time use. We introduce a graph-based neural emulator trained on Vlasiator data to autoregressively predict near-Earth space conditions driven by an upstream solar wind. We show how to achieve both fast deterministic forecasts and, by using a generative model, produce ensembles to capture forecast uncertainty. This work demonstrates that machine learning offers a way to add uncertainty quantification capability to existing space weather prediction systems, and make hybrid-Vlasov simulation tractable for operational use.

physics.space-ph↗

arXiv2025

Accurate Mediterranean Sea forecasting via graph-based deep learning

Accurate ocean forecasting systems are essential for understanding marine dynamics, which play a crucial role in sectors such as shipping, aquaculture, environmental monitoring, and coastal risk management. Traditional numerical solvers, while effective, are computationally expensive and time-consuming. Recent advancements in machine learning have revolutionized weather forecasting, offering fast and energy-efficient alternatives. Building on these advancements, we introduce SeaCast, a neural network designed for high-resolution regional ocean forecasting. SeaCast employs a graph-based framework to effectively handle the complex geometry of ocean grids and integrates external forcing data tailored to the regional ocean context. Our approach is validated through experiments at a high horizontal resolution using the operational numerical forecasting system of the Mediterranean Sea, along with both numerical and data-driven atmospheric forcings. Results demonstrate that SeaCast consistently outperforms the operational model in forecast skill, marking a significant advancement in regional ocean prediction.

physics.ao-ph↗

arXiv2024

Regional Ocean Forecasting with Hierarchical Graph Neural Networks

Accurate ocean forecasting systems are vital for understanding marine dynamics, which play a crucial role in environmental management and climate adaptation strategies. Traditional numerical solvers, while effective, are computationally expensive and time-consuming. Recent advancements in machine learning have revolutionized weather forecasting, offering fast and energy-efficient alternatives. Building on these advancements, we introduce SeaCast, a neural network designed for high-resolution, medium-range ocean forecasting. SeaCast employs a graph-based framework to effectively handle the complex geometry of ocean grids and integrates external forcing data tailored to the regional ocean context. Our approach is validated through experiments at a high spatial resolution using the operational numerical model of the Mediterranean Sea provided by the Copernicus Marine Service, along with both numerical and data-driven atmospheric forcings.

physics.ao-ph↗

arXiv2024

Learning Developmental Age from 3D Infant Kinetics Using Adaptive Graph Neural Networks

Reliable methods for the neurodevelopmental assessment of infants are essential for early detection of problems that may need prompt interventions. Spontaneous motor activity, or 'kinetics', is shown to provide a powerful surrogate measure of upcoming neurodevelopment. However, its assessment is by and large qualitative and subjective, focusing on visually identified, age-specific gestures. In this work, we introduce Kinetic Age (KA), a novel data-driven metric that quantifies neurodevelopmental maturity by predicting an infant's age based on their movement patterns. KA offers an interpretable and generalizable proxy for motor development. Our method leverages 3D video recordings of infants, processed with pose estimation to extract spatio-temporal series of anatomical landmarks, which are released as a new openly available dataset. These data are modeled using adaptive graph convolutional networks, able to capture the spatio-temporal dependencies in infant movements. We also show that our data-driven approach achieves improvement over traditional machine learning baselines based on manually engineered features.

cs.CV↗

arXiv2023

Jet energy calibration with deep learning as a Kubeflow pipeline

Precise measurements of the energy of jets emerging from particle collisions at the LHC are essential for a vast majority of physics searches at the CMS experiment. In this study, we leverage well-established deep learning models for point clouds and CMS open data to improve the energy calibration of particle jets. To enable production-ready machine learning based jet energy calibration an end-to-end pipeline is built on the Kubeflow cloud platform. The pipeline allowed us to scale up our hyperparameter tuning experiments on cloud resources, and serve optimal models as REST endpoints. We present the results of the parameter tuning process and analyze the performance of the served models in terms of inference time and overhead, providing insights for future work in this direction. The study also demonstrates improvements in both flavor dependence and resolution of the energy response when compared to the standard jet energy corrections baseline.

hep-ex↗

arXiv2021

Interatomic Fe-Cr potential for modeling kinetics on Fe surfaces

To enable accurate molecular dynamics simulations of iron-chromium alloys with surfaces, we develop, based on density-functional-theory (DFT) calculations, a new interatomic Fe-Cr potential in the Tersoff formalism. Contrary to previous potential models, which have been designed for bulk Fe-Cr, we extend our potential fitting database to include not only conventional bulk properties but also surface-segregation energies of Cr in bcc Fe. In terms of reproducing our DFT results for the bulk properties, the new potential is found to be superior to the previously developed Tersoff potential and competitive with the concentration-dependent and two-band embedded-atom-method potentials. For Cr segregation toward the surface of an Fe-Cr alloy, only the new potential agrees with our DFT calculations in predicting preferential segregation of Cr to the topmost surface layer, instead of the second layer preferred by the other potentials. We expect this rectification to foster future research, e.g., on the mechanisms of corrosion resistance of stainless steels at the atomic level.

cond-mat.mtrl-sci↗