arXiv subjects

Håvard Homleid Haugen

Publications and source records attributed to Håvard Homleid Haugen.

7 recordsLinked to original sources

arXiv2026

Varda-single-1.0: deterministic data-driven weather forecasting at 1 km resolution over Switzerland's complex topography

We present Varda-single-1.0, a medium-range data-driven weather prediction system built for the Alpine domain. It provides hourly deterministic regional forecasts on a mesh of 1 km resolution and global forecasts on a 31 km mesh. The system comprises two independently trained stretched-grid Graph Transformer models with encoder-processor-decoder architecture, developed in the Anemoi framework: a 6-hourly autoregressive forecaster and a temporal downscaler reconstructing hourly forecasts between the forecaster's steps. Its training curriculum includes pre-training on ERA5 reanalysis data, followed by training on a 20-year kilometre-scale regional reanalysis, and finally fine-tuning on operational kilometre-scale analyses. Verified over one year against operational analyses and surface station observations, Varda-single is competitive with or improves on MeteoSwiss' operational numerical weather prediction baselines for most headline scores and variables. It broadly matches the skill of the high-resolution 1 km ICON-CH1-EPS control at lead times up to +33 h and generally outperforms the 2 km ICON-CH2-EPS control at lead times up to +120 h. Despite competitive aggregate scores, Varda-single underestimates some local wind maxima and produces overly smooth convective precipitation fields, consistent with the smoothing associated with squared-error training. To gain insight into the model's behaviour, we investigate three case studies beyond the aggregated headline scores, and find particular weaknesses in Varda-single's representation of local winds over complex terrain. Varda-single represents an important step in the development of high-resolution ML forecasting over complex terrain, in complementing the operational regional numerical weather prediction models of MeteoSwiss with data-driven models and in providing a pretrained model for researchers and user-specific applications.

physics.ao-ph↗

arXiv2026

High-Resolution Probabilistic Data-Driven Weather Modeling with a Stretched-Grid

We present a probabilistic data-driven weather model providing ensembles of high spatial resolution realizations of 87 variables at arbitrary ensemble size and forecast length. The model uses a global stretched grid, dedicating 2.5 km resolution to our Nordic region of interest and 31 km resolution elsewhere, with 6-hour temporal resolution. Unique ensemble members are generated by a stochastic model architecture, and we train it using a loss function based on the Continuous Ranked Probability Score (CRPS) evaluated in grid-point and spectral space. The spectral loss component is shown to be necessary to create fields that are spatially coherent, which is not the case when training with mean-squared error loss, nor CRPS in grid-point space only. We evaluate the forecasts against observations from surface weather stations and compare them to high-resolution operational numerical weather prediction forecasts from the MetCoOp Ensemble Prediction System (MEPS). The model shows lower CRPS than MEPS for 2 m temperature and mean sea-level pressure, with average improvements of 13\% and 10\%, respectively, while differences for wind speed and precipitation are smaller. For Storm Dave, the model captures the location and structure of strong-wind systems, but underestimates the peak winds.

physics.ao-ph↗

arXiv2026

HourGlass: A probabilistic data-driven temporal downscaler for global and regional weather forecasting

Many forecast applications require high frequency temporal resolution, yet most state-of-the-art data-driven weather forecasting systems operate at 6-hourly resolution. Although direct hourly forecasting is possible, it suffers from error accumulation and temporal inconsistency. We introduce HourGlass, a probabilistic data-driven temporal downscaling method that reconstructs the evolution between forecast states. HourGlass is trained using variants of the continuous ranked probability score (CRPS) preserving small-scale spatial variability while encouraging temporal consistency. Unlike existing deterministic temporal downscaling approaches, which tend to produce overly smooth fields, HourGlass generates realistic probabilistic forecasts. Training on forecast trajectories rather than reanalysis or analysis data also avoids the temporal inconsistencies present in datasets used by previous methods. We evaluate HourGlass in two settings: AIFS-HourGlass, applied globally to ECMWF's AIFS-Single and AIFS-ENS forecast systems, and Bris-HourGlass, applied regionally to MET Norway's high-resolution stretched-grid ensemble model, Bris. Verification against observations shows that both models retain the skill of their underlying forecasting systems while producing temporally coherent hourly forecasts with realistic small-scale variability. Case studies demonstrate physically consistent evolution during rapidly developing weather events, including extratropical cyclones and organised convection. Hourly precipitation remains challenging: HourGlass improves the spatial realism of precipitation fields but still underestimates the most intense extremes, a common limitation of data-driven weather forecasting models. These results demonstrate that HourGlass effectively bridges the gap between 6-hourly data-driven forecasts and the hourly products required for operational regional and global forecasting.

physics.ao-ph↗

arXiv2026

Enhancing a high resolution data-driven weather prediction model with surface descriptors

We study the importance of surface characteristics when forecasting near-surface variables with a data-driven weather prediction model. To target the challenge of predicting small-scale weather conditions at high resolution, we introduce a range of surface descriptors in the training of a state-of-the-art data-driven model. The input data includes surface descriptors inherited from the numerical weather prediction model used to produce the training dataset and topographic neighbourhood indices. We found that errors of 2-metre temperature and 10-metre wind speed forecasts were reduced by 1.9% and 3.0% respectively compared to a baseline model over the model domain. Over certain surfaces, the improvements were significantly larger. For example, we found a 12% reduction of temperature mean absolute errors over urban areas when the urban fraction was included in the model input. Furthermore, we investigated how the model responded to removal of glaciers, resulting in an increase of temperature. This indicates that 1) the model produce a physically reasonable response and 2) input datasets can be updated without the need to retrain the model. The latter suggests a great benefit for operational systems as training is expensive compared to running these models. This study highlights the importance of including surface conditions in the prediction of near-surface variables.

physics.ao-ph↗

arXiv2024

Regional data-driven weather modeling with a global stretched-grid

A data-driven model (DDM) suitable for regional weather forecasting applications is presented. The model extends the Artificial Intelligence Forecasting System by introducing a stretched-grid architecture that dedicates higher resolution over a regional area of interest and maintains a lower resolution elsewhere on the globe. The model is based on graph neural networks, which naturally affords arbitrary multi-resolution grid configurations. The model is applied to short-range weather prediction for the Nordics, producing forecasts at 2.5 km spatial and 6 h temporal resolution. The model is pre-trained on 43 years of global ERA5 data at 31 km resolution and is further refined using 3.3 years of 2.5 km resolution operational analyses from the MetCoOp Ensemble Prediction System (MEPS). The performance of the model is evaluated using surface observations from measurement stations across Norway and is compared to short-range weather forecasts from MEPS. The DDM outperforms both the control run and the ensemble mean of MEPS for 2 m temperature. The model also produces competitive precipitation and wind speed forecasts, but is shown to underestimate extreme events.

physics.ao-ph↗

arXiv2021

First order superconducting phase transition in chiral $p+ip$ system

We use large-scale Monte Carlo computations to study the phase transitions of a two-component chiral p-wave superconductor in zero external magnetic field. We find a first order phase transition from the normal state to a chiral superconducting state, due to interplay between vortices and domain walls.

cond-mat.supr-con↗

arXiv2021

Thermal fluctuations and vortex lattice structures in chiral $p$-wave superconductors: robustness of double-quanta vortices

We use large-scale Monte-Carlo simulations to study thermal fluctuations in chiral $p$-wave superconductors in an applied magnetic field in three dimensions. We consider the thermal stability of previously predicted unusual double-quanta flux-line lattice ground states in such superconductors. In previous works it was shown that, neglecting thermal fluctuations, a chiral $p$-wave superconductor forms an hexagonal lattice of doubly-quantized vortices, except extremely close to the vicinity of $H_{c2}$ where double-quanta vortices split apart. We find dissociation of double-quanta vortices driven by thermal fluctuations. However, our calculations also show that the previous predictions of hexagonal doubly-quantized vortices, where thermal fluctuations were ignored, are very robust in the considered model.

cond-mat.supr-con↗