banner-qai-volc.png

QAI_Volc: High-Performance Computing Infrastructure for Volcanic Early Detection and Nowcasting with Quantum AI Technologies

QAI_Volc is an infrastructural and methodological initiative within the ROSE (Reinforcement of the Observational Systems of the Earth) infrastructural project of INGV, funded by the Italian Ministry of University and Research. The initiative aims to substantially strengthen the operational capabilities of the Etna Volcano Observatory for the analysis and near-real-time forecasting of complex volcanic phenomena, by integrating: (i) a reinforced High-Performance Computing (HPC) environment, (ii) advanced AI-based early detection systems, and (iii) emerging Quantum Artificial Intelligence approaches for volcanic nowcasting.

Launched in 2026, QAI_Volc is conceived as a 24-month program combining infrastructure upgrades, operational tool consolidation, and frontier research and training activities, with direct relevance for monitoring activities at Etna and Stromboli.

Key objectives

QAI_Volc pursues one overarching goal: deploying an advanced computational infrastructure that couples HPC and Quantum AI to improve early detection and nowcasting of volcanic activity, with a clear operational orientation.

The objectives are structured into three Work Packages:

HPC infrastructure reinforcement (WP1)
Consolidate and evolve the Etna Observatory computing center into a resilient, scalable hub supporting real-time data ingestion, physics-based modeling (digital-twin-oriented workflows), deep-learning pipelines, and hybrid classical–quantum experimentation.

ETNAS enhancement and extension (WP2)
Consolidate and upgrade the ETNAS early-detection system for rapid identification of lava fountains and magmatic intrusions at Etna, and assess transferability to Stromboli, through multi-source data integration and robust AI models.

Quantum AI methodologies for volcanic nowcasting (WP3)
Explore and test quantum and hybrid QC–HPC models for short-term forecasting of dynamic volcanic phenomena (lava flows, ash clouds, eruptive evolution), with rigorous benchmarking against classical deep-learning approaches.

 Fig_1_-_Logo.png

Scientific rationale

Volcanic phenomena such as lava fountains, magmatic intrusions, lava-flow emplacement and propagation, ash-cloud dispersion, and ground deformation are typical examples of nonlinear, multi-scale, and uncertainty-dominated systems, where time-critical decision support requires advanced computational and data-driven capabilities.

Quantum AI combines quantum-computing paradigms with AI/ML algorithms and offers a forward-looking route to handle high-dimensional representations, complex optimization, and hybrid physics-data integration problems. QAI_Volc leverages consolidated expertise at the Etna Observatory in AI pipelines and operational multi-source monitoring, while building readiness for quantum-enabled workflows through a controlled, progressive research plan.

 

Technologies

QAI_Volc integrates enabling technologies across three coordinated layers.

High-Performance Computing and infrastructure readiness

  • Expansion of heterogeneous compute resources (CPU/GPU) and high-performance storage
  • Low-latency internal networking and scalable workflows
  • Integration with Quantum Cloud services via software gateways and hybrid execution pipelines
  • Containerized environments and orchestration tools for reproducibility and portability
  • Dedicated technical support (HPC/Research Software Engineering profile) to ensure sustainability and operational continuity

AI for early detection (ETNAS)

  • Deep-learning models for rapid detection of paroxysmal activity and magmatic intrusions
  • Multi-source data fusion (geostationary satellite data, visual/thermal networks, multiparametric monitoring)
  • Robustness improvements (false positives reduction, onset sensitivity, handling missing data)
  • Generalization testing and adaptation to different eruptive regimes (Stromboli)

Quantum AI for nowcasting

  • Variational Quantum Circuits (VQC), Quantum CNN, quantum kernels
  • Quantum autoregressive approaches for short-term prediction
  • Hybrid QC–HPC workflows with systematic benchmarking versus classical deep learning
  • Application-oriented proof-of-concepts on real monitoring scenarios at Etna and Stromboli

 

Quantum Artificial Intelligence School

QAI_Volc contributes to the Quantum Artificial Intelligence School, an advanced training initiative organized under the scientific leadership of the DEMETRA research line of the ROSE project, whose dedicated web page is available on the TechnoLab website.

The School is designed as a transversal enabling action supporting multiple ROSE research lines and aims to strengthen internal competences in high-performance computing, artificial intelligence, and emerging quantum technologies applied to Earth and environmental sciences.

The Quantum AI School is developed in close collaboration with the University of Catania. Within this framework, QAI_Volc provides a full scientific and infrastructural contribution, particularly in relation to volcanic monitoring, early detection, and nowcasting applications.

Training activities combine lectures with hands-on laboratories based on real INGV datasets and operational workflows, with access to HPC infrastructures and Quantum Cloud resources. The School supports the development of skills in:

  • HPC and hybrid HPC–Quantum Cloud infrastructures
  • AI-based monitoring and early detection systems (including ETNAS)
  • Quantum AI methodologies and QC–HPC integration for volcanic nowcasting pipelines

The School contributes to the long-term sustainability of ROSE by fostering interdisciplinary expertise and training researchers, technologists, and students at the interface between geophysics, artificial intelligence, and quantum computing.

Key information

  • The Quantum AI School will take place at the end of September 2026.

  • Participant registration will open in early March 2026.

 

Project organization

QAI_Volc Principal Investigators (INGV -  Etna Volcano Observatory, Catania):

  • Flavio Cannavò (ETNAS team)
  • Claudia Corradino (TechnoLab)
  • Salvatore Mangiagli (Operations Room Unit)

ROSE Project Coordinator

Ciro Del Negro (INGV -  Etna Volcano Observatory, Catania)

Contributions:

QAI_Volc involves coordinated contributions from the Etna Volcano Observatory, including:

  • the Operations Room Unit (validation/testing in operational conditions),
  • the TechnoLab (HPC/AI/Quantum AI development for nowcasting applications),
  • the ETNAS development team (evolution and integration of the operational early-detection system).

Workpackages (WP)

WP1 – HPC infrastructure reinforcement and Quantum Cloud integration (Lead: Salvo Mangiagli)
Infrastructure upgrade, hybrid readiness, workflow scalability, and dedicated specialist technical support.

WP2 – ETNAS: development, consolidation, and extension (Lead: Flavio Cannavò)
Enhanced AI algorithms, multi-source fusion, improved operational robustness, and transferability testing to Stromboli.

WP3 – Quantum AI for volcanic nowcasting (Lead: Claudia Corradino)
Development of quantum and hybrid models, QC–HPC integration, benchmarking against classical approaches, and experimental applications to real volcanic scenarios.

  

Contact

For scientific and organizational information on QAI_Volc:
Email: flavio.cannavoThis email address is being protected from spambots. You need JavaScript enabled to view it.; This email address is being protected from spambots. You need JavaScript enabled to view it.; This email address is being protected from spambots. You need JavaScript enabled to view it.