Improving blended probability forecasts with neural networks
Fuente:
Wiley Open Access
Guardado en:
| Autor principal: | |
|---|---|
| Formato: | Artículo Open Access |
| Publicado: |
Wiley
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1867021396570275840 |
|---|---|
| author | Belinda Trotta |
| author_facet | Belinda Trotta Belinda Trotta |
| collection | Wiley Open Access |
| contents | Improving blended probability forecasts with neural networks Belinda Trotta Meteorological Applications AbstractOperational forecasting systems often combine calibrated probabilistic outputs from several numerical weather prediction (NWP) models. A common approach is to use a weighted blend, with the more accurate models having higher weights. We show that this approach is not ideal and that using a simple neural network to combine forecasts yields better results. The sharpness of the forecast is increased, so that extreme events are more likely to be predicted. Improvements are also observed in accuracy as measured by the continuous rank probability score (CRPS) and reliability. The proposed neural network model has a simple architecture with few parameters, and training and inference can easily be done using a central processing unit. This makes it a practical option for improving the accuracy of blended operational forecasts. 10.1002/met.70021 http://creativecommons.org/licenses/by-nc-nd/4.0/ |
| doi_str_mv | 10.1002/met.70021 |
| format | Artículo Open Access |
| id | wiley_oa_10_1002_met_70021 |
| institution | Wiley Open Access |
| license_str_mv | http://creativecommons.org/licenses/by-nc-nd/4.0/ |
| publishDate | 2024 |
| publisher | Wiley |
| record_format | wiley_oa |
| spellingShingle | Improving blended probability forecasts with neural networks Belinda Trotta Meteorological Applications Improving blended probability forecasts with neural networks Belinda Trotta Meteorological Applications AbstractOperational forecasting systems often combine calibrated probabilistic outputs from several numerical weather prediction (NWP) models. A common approach is to use a weighted blend, with the more accurate models having higher weights. We show that this approach is not ideal and that using a simple neural network to combine forecasts yields better results. The sharpness of the forecast is increased, so that extreme events are more likely to be predicted. Improvements are also observed in accuracy as measured by the continuous rank probability score (CRPS) and reliability. The proposed neural network model has a simple architecture with few parameters, and training and inference can easily be done using a central processing unit. This makes it a practical option for improving the accuracy of blended operational forecasts. 10.1002/met.70021 http://creativecommons.org/licenses/by-nc-nd/4.0/ |
| title | Improving blended probability forecasts with neural networks |
| topic | Meteorological Applications |
| url | https://rmets.onlinelibrary.wiley.com/doi/10.1002/met.70021 |