Scalable non-separable spatio-temporal Gaussian process models for large-scale short-term weather prediction
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866918332818522112 |
|---|---|
| author | Gyger, Tim Furrer, Reinhard Sigrist, Fabio |
| author_facet | Gyger, Tim Furrer, Reinhard Sigrist, Fabio |
| contents | Monitoring daily weather fields is critical for climate science, agriculture, and environmental planning, yet fully probabilistic spatio-temporal models become computationally prohibitive at continental scale. We present a case study on short-term forecasting of daily maximum temperature and precipitation across the conterminous United States using novel scalable spatio-temporal Gaussian process methodology. Building on three approximation families - inducing-point methods (FITC), Vecchia approximations, and a hybrid Vecchia-inducing-point full-scale approach (VIF) - we introduce three extensions that address key bottlenecks in large space-time settings: (i) a scalable correlation-based neighbor selection strategy for Vecchia approximations with point-referenced data, enabling accurate conditioning under complex dependence structures, (ii) a space-time kMeans++ inducing-point selection algorithm, and (iii) GPU-accelerated implementations of computationally expensive operations, including matrix operations and neighbor searches. Using both synthetic experiments and a large NOAA station dataset containing more than one million space-time observations, we analyze the models with respect to predictive performance, parameter estimation, and computational efficiency. Our results demonstrate that scalable Gaussian process models can yield accurate continental-scale forecasts while remaining computationally feasible, offering practical tools for weather applications. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_03609 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Scalable non-separable spatio-temporal Gaussian process models for large-scale short-term weather prediction Gyger, Tim Furrer, Reinhard Sigrist, Fabio Applications 60G15, 62M20, 62R07, 68T09 Monitoring daily weather fields is critical for climate science, agriculture, and environmental planning, yet fully probabilistic spatio-temporal models become computationally prohibitive at continental scale. We present a case study on short-term forecasting of daily maximum temperature and precipitation across the conterminous United States using novel scalable spatio-temporal Gaussian process methodology. Building on three approximation families - inducing-point methods (FITC), Vecchia approximations, and a hybrid Vecchia-inducing-point full-scale approach (VIF) - we introduce three extensions that address key bottlenecks in large space-time settings: (i) a scalable correlation-based neighbor selection strategy for Vecchia approximations with point-referenced data, enabling accurate conditioning under complex dependence structures, (ii) a space-time kMeans++ inducing-point selection algorithm, and (iii) GPU-accelerated implementations of computationally expensive operations, including matrix operations and neighbor searches. Using both synthetic experiments and a large NOAA station dataset containing more than one million space-time observations, we analyze the models with respect to predictive performance, parameter estimation, and computational efficiency. Our results demonstrate that scalable Gaussian process models can yield accurate continental-scale forecasts while remaining computationally feasible, offering practical tools for weather applications. |
| title | Scalable non-separable spatio-temporal Gaussian process models for large-scale short-term weather prediction |
| topic | Applications 60G15, 62M20, 62R07, 68T09 |
| url | https://arxiv.org/abs/2602.03609 |