Chapter AI and Machine Learning to extend Meteo-Marine Station Observations into the Future
The real-time availability of data from coastal meteo-marine stations is crucial for various stakeholders, including port authorities, government agencies, researchers, and the general public. While observation data is fundamental, short-term forecasts can significantly enhance planning and decision...
Tallennettuna:
| Päätekijä: | |
|---|---|
| Aineistotyyppi: | Online |
| Kieli: | englanti |
| Julkaistu: |
Firenze University Press
2025
|
| Aiheet: | |
| Linkit: | ONIX_20250801T173835_9791221505566_231 |
| Tagit: |
Ei tageja, Lisää ensimmäinen tagi!
|
| _version_ | 1869526560622510080 |
|---|---|
| author | Azzopardi, Joel |
| author_browse | Azzopardi, Joel |
| author_facet | Azzopardi, Joel |
| author_sort | Azzopardi, Joel |
| collection | Directory of Open Access Books |
| description | The real-time availability of data from coastal meteo-marine stations is crucial for various stakeholders, including port authorities, government agencies, researchers, and the general public. While observation data is fundamental, short-term forecasts can significantly enhance planning and decision-making processes. This study explores the application of Machine Learning (ML) techniques to predict hourly values of air temperature, wind speed, atmospheric pressure, and humidity for the next 24 hours. We evaluate three ML models: Long Short-Term Memory Network (LSTM), Random Forest (RF), and Multivariate Linear Regression (LR). The models were trained using Python libraries and Optuna for hyperparameter tuning on datasets of varying lengths from stations in the Malta-Sicily channel. Additionally, we investigated transfer learning with the ERA5 dataset, which provides hourly values over an 83-year period, to address the challenge of limited data availability. The results show that models trained on longer datasets generally achieve better performance. Furthermore, the models demonstrated considerable generalizability, particularly across nearby stations, allowing models trained at one station to be effectively used for predictions at other proximate stations. To support further research and practical application, we have made our models and tools publicly available. |
| format | Online |
| id | doab-20.500.12854ir-163374 |
| institution | Directory of Open Access Books |
| language | eng |
| publishDate | 2025 |
| publishDateRange | 2025 |
| publishDateSort | 2025 |
| publisher | Firenze University Press |
| publisherStr | Firenze University Press |
| record_format | ojs |
| spelling | doab-20.500.12854ir-1633742025-08-02T05:05:03Z Chapter AI and Machine Learning to extend Meteo-Marine Station Observations into the Future Azzopardi, Joel Machine Learning Artificial Intelligence Transfer Learning Meteorology Prediction The real-time availability of data from coastal meteo-marine stations is crucial for various stakeholders, including port authorities, government agencies, researchers, and the general public. While observation data is fundamental, short-term forecasts can significantly enhance planning and decision-making processes. This study explores the application of Machine Learning (ML) techniques to predict hourly values of air temperature, wind speed, atmospheric pressure, and humidity for the next 24 hours. We evaluate three ML models: Long Short-Term Memory Network (LSTM), Random Forest (RF), and Multivariate Linear Regression (LR). The models were trained using Python libraries and Optuna for hyperparameter tuning on datasets of varying lengths from stations in the Malta-Sicily channel. Additionally, we investigated transfer learning with the ERA5 dataset, which provides hourly values over an 83-year period, to address the challenge of limited data availability. The results show that models trained on longer datasets generally achieve better performance. Furthermore, the models demonstrated considerable generalizability, particularly across nearby stations, allowing models trained at one station to be effectively used for predictions at other proximate stations. To support further research and practical application, we have made our models and tools publicly available. 2025-08-02T05:05:02Z 2025-08-02T05:05:02Z 2025-08-01T15:54:32Z 2024 chapter ONIX_20250801T173835_9791221505566_231 2975-0288 https://library.oapen.org/handle/20.500.12657/104781 9791221505566 https://directory.doabooks.org/handle/20.500.12854/163374 eng Monitoring of Mediterranean Coastal Areas: Problems and Measurement Techniques open access image/jpeg Attribution-NonCommercial-ShareAlike 4.0 International https://library.oapen.org/bitstream/20.500.12657/104781/1/43718.pdf Firenze University Press 10.36253/979-12-215-0556-6.73 10.36253/979-12-215-0556-6.73 2ec4474d-93b1-4cfa-b313-9c6019b51b1a 9791221505566 12 Florence open access |
| spellingShingle | Machine Learning Artificial Intelligence Transfer Learning Meteorology Prediction Azzopardi, Joel Chapter AI and Machine Learning to extend Meteo-Marine Station Observations into the Future |
| title | Chapter AI and Machine Learning to extend Meteo-Marine Station Observations into the Future |
| title_full | Chapter AI and Machine Learning to extend Meteo-Marine Station Observations into the Future |
| title_fullStr | Chapter AI and Machine Learning to extend Meteo-Marine Station Observations into the Future |
| title_full_unstemmed | Chapter AI and Machine Learning to extend Meteo-Marine Station Observations into the Future |
| title_short | Chapter AI and Machine Learning to extend Meteo-Marine Station Observations into the Future |
| title_sort | chapter ai and machine learning to extend meteo marine station observations into the future |
| topic | Machine Learning Artificial Intelligence Transfer Learning Meteorology Prediction |
| topic_facet | Machine Learning Artificial Intelligence Transfer Learning Meteorology Prediction |
| url | ONIX_20250801T173835_9791221505566_231 |
| work_keys_str_mv | AT azzopardijoel chapteraiandmachinelearningtoextendmeteomarinestationobservationsintothefuture |