This thesis documents an internship at the Regional Environmental Modeling Center of ARPA FVG, focusing on the analysis and evaluation of the Aurora Foundation Model (Microsoft) for atmospheric, air quality, and ocean wave forecasting applications. Activities included studying the technical documentation, testing selected datasets, and comparing local and cloud infrastructures. The work provided expertise in HPC, workflow automation, software versioning, and AI tools in Python, providing insights for a strategic evaluation of the use of foundational models in the environmental field.
La tesi documenta un tirocinio svolto presso il Centro Regionale di Modellistica Ambientale di ARPA FVG, incentrato sull’analisi e valutazione del foundation model Aurora (Microsoft) per applicazioni di previsione atmosferica, della qualità dell’aria e delle onde oceaniche. Le attività hanno incluso lo studio della documentazione tecnica, test su dataset selezionati e confronto tra infrastrutture locali e cloud. Il lavoro ha permesso di acquisire competenze su HPC, workflow automation, versionamento software e strumenti AI in Python, offrendo spunti per una valutazione strategica dell’uso di modelli fondazionali in ambito ambientale.
Valutazione del foundation model Aurora per il supporto alle attività operative di ARPA FVG
STEFANUTO, GIOVANNI
2024/2025
Abstract
This thesis documents an internship at the Regional Environmental Modeling Center of ARPA FVG, focusing on the analysis and evaluation of the Aurora Foundation Model (Microsoft) for atmospheric, air quality, and ocean wave forecasting applications. Activities included studying the technical documentation, testing selected datasets, and comparing local and cloud infrastructures. The work provided expertise in HPC, workflow automation, software versioning, and AI tools in Python, providing insights for a strategic evaluation of the use of foundational models in the environmental field.| File | Dimensione | Formato | |
|---|---|---|---|
|
Stefanuto_Giovanni.pdf
accesso aperto
Dimensione
1.44 MB
Formato
Adobe PDF
|
1.44 MB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/91750