DANIELE TRAPPOLINI

Dottore di ricerca

ciclo: XXXVIII


supervisore: Fabio Galasso
co-supervisore: Elisa Tinti

Titolo della tesi: Artificial Intelligence in Geoscience Applications

Earthquakes remain challenging to analyze in real time because waveforms are noisy, source pa- rameters are uncertain, and shaking patterns evolve rapidly. This thesis advances the seismic monitoring pipeline with deep learning (DL) methods that improve data quality, reduce latency, and raise predictive accuracy on operational datasets and networks. First, we introduce CDiffSD, a cold diffusion model for seismic denoising that learns non-Gaussian degradation processes directly from station noise. On the STEAD-based test, CDiffSD increases median signal-to-noise ratio and cross-correlation over a strong deep-denoiser baseline, while markedly increasing P-pick recall within ±50 samples (from ∼ 0.60 to ∼ 0.90 at T =300). Second, we propose a masked Graph Convolutional Network for rapid shaking prediction in Italy. Using only the first 10 s of three- component waveforms from national-scale networks (INSTANCE), the model yields up to 6% and 5.5% mean-squared-error reductions for PGA and PGV, respectively, near-zero median residuals across intensity measures, and mitigates biases relative to GMM (an Italy-specific ground-motion model). Third, we develop LLM4Geo, a domain-adapted large language model that estimates epi- center, hypocenter, origin time, and magnitude within five seconds of first detection. On 3,094 test events, average errors are ∼4 km (epicenter), ∼6.8 km (hypocenter), ∼0.6 s (origin time), and ∼0.2 magnitude units, outperforming early-stage automated INGV solutions while using fewer picks. Fi- nally, we present SeismicAE, a waveform autoencoder that transfers audio-domain representation learning to seismology. The learned embeddings improve few-shot classification and reduce error in ground-motion regression relative to models trained on raw traces. In the end, we build a pro- totype AI agent that combines different AI models for specific tasks (expert) in a single workflow (orchestration), giving compatible results for the evaluation of operational seismic risk.

Produzione scientifica

11573/1755389 - 2025 - Masked graph neural network for rapid ground motion prediction in Italy
Trappolini, Daniele; Oliveti, Ilaria; Faenza, Licia; Jozinović, Dario; Michelini, Alberto - 01a Articolo in rivista
rivista: SEISMICA ([Montréal]: McGill University Library, [2022]-) pp. - - issn: 2816-9387 - wos: (0) - scopus: (0)

11573/1716903 - 2024 - Cold Diffusion Model for Seismic Denoising
Trappolini, Daniele; Laurenti, Laura; Poggiali, Giulio; Tinti, Elisa; Galasso, Fabio; Michelini, Alberto; Marone, Chris - 01a Articolo in rivista
rivista: JOURNAL OF GEOPHYSICAL RESEARCH. MACHINE LEARNING AND COMPUTATION (Hoboken NJ: Wiley Periodicals LLC, 2024-) pp. - - issn: 2993-5210 - wos: (0) - scopus: 2-s2.0-105030151073 (10)

11573/1675145 - 2023 - DiffSD: Diffusion models for seismic denoising
Trappolini, Daniele; Laurenti, Laura; Tinti, Elisa; Galasso, Fabio; Marone, Chris James; Michelini, Alberto - 04d Abstract in atti di convegno
congresso: European Geoscience Union General Assembly (Vienna)
libro: EGU European Geoscience Union General Assembly 2023 - (9781510812253)

11573/1664528 - 2022 - Mesoscale precipitation nowcasting from weather radar data using space-time-separable graph convolutional networks
Trappolini, Daniele; Scofano, Luca; Sampieri, Alessio; Messina, Francesco; Galasso, Fabio; Di Fabio, Saverio; Silvio Marzano, Frank - 04d Abstract in atti di convegno
congresso: European Geoscience Union General Assembly (Vienna)
libro: EGU General Assembly Conference Abstracts - ()

© Università degli Studi di Roma "La Sapienza" - Piazzale Aldo Moro 5, 00185 Roma