Eleonora Amato

The Earth’s climate system is highly complex and responds to several forcing mechanisms, both natural and anthropogenic, with significant short- and long-term societal impacts. Among natural forcings, large explosive volcanic eruptions represent the dominant driver of abrupt cooling episodes over the past two millennia. However, the limited number of well-documented events and the substantial uncertainties in eruption source parameters, initial conditions, and aerosol forcing make the quantitative assessment of volcano–climate interactions particularly challenging. Addressing these limitations requires the integration of large and heterogeneous datasets, from satellite observations and in situ measurements to historical and paleoclimate archives, within modeling tools capable of capturing the nonlinear dynamics of the climate system. Advanced computing technologies, such as Artificial Intelligence (AI), High-Performance Computing (HPC), and emerging Quantum Computing (QC), offer new opportunities to overcome these constraints. AI and hybrid Machine Learning–physics approaches can emulate computationally expensive model components, improve the representation of aerosol–radiation processes, and accelerate sensitivity analyses, while HPC and QC can reduce the cost of large ensemble simulations and discover hidden patterns. Here, we highlight how these methodologies can enhance the study of volcano-climate interactions, improving model performance, enabling a more efficient exploration of uncertainties, and refining predictions of the climatic impacts of major explosive eruptions.

Physical phenomena evolve in space and time following governing laws with a high level of complexity and mathematical models can help to make accurate predictions of their behavior, describing complex fluids with a good balance between accuracy and computational costs. We have long used detailed Computational Fluid Dynamics (CFD) models to simulate complex dynamics with high accuracy, but these simulations typically entail high computational costs, resulting in long execution times and the use of expensive computational resources. To overcome these limitations, we have recently integrated CFD with Artificial Intelligence (AI), in the so-called Emulators, to expand the scope of fluid modeling, improving its performance. Here, we present an AI-based CFD emulator for Smoothed Particle Hydrodynamics (SPH) simulations, which uses an Artificial Neural Network (ANN) to enhance simulations of complex fluids with viscous and thermal components. We show the model capability to reproduce the spatio-temporal evolution of natural visco-thermal fluids. In addition, we demonstrate the emulator capacity to generalize its applicability to problems not encountered during the training phase. We also conduct a detailed error evaluation, showing that the minimal observed discrepancies do not compromise model accuracy and robustness, especially given the theoretical and computational advantages. These key points open this emulator to practical applications for natural fluids, e.g., oil, honey, or geophysical fluids such as lava, enhancing fluidmodeling performance and extending functionalities. The innovation of this method improves studies in the field of numerical simulations, for example in its use as a digital Twin of a physical phenomenon, for the study of the dynamics of the system without the need of large costs for field analysis or laboratory experiments.

Lava flows are complex, non-Newtonian fluids with visco-thermal dependencies that can overcome barriers, form tunnels, and significantly impact surrounding areas. Understanding and predicting these flows are critical for quantifying volcanic hazards. Computational Fluid Dynamics (CFD) models are indispensable tools for simulating lava dynamics, but they often entail high computational costs, limiting their real-time applicability. To address these challenges, we propose an AI-enhanced CFD emulator for lava flows, designed to improve modeling efficiency while preserving accuracy. Our approach integrates AI with CFD to capture the visco-thermal properties of lava and its intricate dynamics, including phase transitions, particle solidification, and the influence of air on thermal behavior. The emulator has been validated through simulations of diverse physical scenarios, demonstrating its capability to generalize across varying conditions. Additionally, we conducted a sensitivity analysis, exploring the influence of key parameters, such as effusion rate, on lava flow evolution and eruption styles. By incorporating satellite-derived estimates, we provide insights into eruptive behaviors while minimizing the risks of field observations. Our results showcase the potential of combining AI, numerical models, and remote sensing to enhance traditional volcanic monitoring approaches. This hybrid methodology enables faithful, near real-time simulations of lava flows, offering valuable tools for hazard assessment and risk mitigation.

The combination of Computational Fluid Dynamics (CFD) and Artificial Intelligence (AI) enables to expand the scope of fluid modeling and improve simulation performance. Eulerian methods have already been extensively integrated with AI, providing high-fidelity and reliable results (e.g., weather prediction); the application of AI to Lagrangian methods remains less consolidated. Smoothed Particle Hydrodynamics (SPH) is a Lagrangian mesh-less CFD numerical method with well-established advantages for the simulation of highly dynamic free-surface complex fluids. However, reliable SPH simulations require long execution times and large computational resources. This downside can be resolved using emulators, which can reproduce the dynamics of a physical system, speeding-up the simulations. In detail, an emulator is a model in which AI algorithms join or replace the equation-based mathematical representation of physics. Thus, the emulator learns from CFD simulations the behavior of the CFD reference model and reproduces it to solve fluid dynamics problems in shorter times. For the emulator to be trusted, it is important to assess the ability to generalize and correctly predict the behavior of the fluid in conditions that have not been presented during training phase. Here, we present the generalizability of an AI-based emulator for SPH, utilizing an Artificial Neural Network (ANN) to estimate the hydrodynamic forces between SPH particles. We demonstrate the capability of the emulator to reproduce particle interaction quantities, as opposed to their specific spatial configuration. Therefore, when a different problem is treated, as long as the variables values are in a range with physics meaning presented during training, the emulator will be capable to produce physically meaningful results. We discuss the salient points of designing this emulator, such as the choice of the AI model, i.e., the ANN, and the choice of the features. We show applications to different test cases, highlighting the emulator ability to generalize for problems with varying levels of complexity, and for changes in the model resolution.

The latest generation of high-spatial-resolution satellites produces measurements of high-temperature volcanic features at global scale, which are valuable to monitor volcanic activity. Recent advances in technology and increased computational resources have resulted in an extraordinary amount of monitoring data, which can no longer be so readily examined. Here, we present an automatic detection algorithm based on a deep convolutional neural network (CNN) that uses infrared satellite data to automatically determine the presence of volcanic thermal activity. We exploit the potentiality of the transfer learning technique to retrain a pre-trained SqueezeNet CNN to a new domain. We fine-tune the weights of the network over a new dataset opportunely created with images related to thermal anomalies of different active volcanoes around the world. Furthermore, an ensemble approach is employed to enhance accuracy and robustness when compared to using individual models. We chose a balanced training dataset with two classes, one containing volcanic thermal anomalies (erupting volcanoes) and the other containing no thermal anomalies (non-erupting volcanoes), to differentiate between volcanic scenes with eruptive and non-eruptive activity. We used satellite images acquired in the infrared bands by ESA Sentinel-2 Multispectral Instrument (MSI) and NASA & USGS Landsat 8 Operational Land Imager and Thermal InfraRed Sensor (OLI/TIRS). This deep learning approach makes the model capable of identifying the appearance of a volcanic thermal anomaly in the images belonging to the volcanic domain with an overall accuracy of 98.3%, recognizing the scene with active flows and erupting vents (i.e., eruptive activity) and the volcanoes at rest. This model is generalizable, and has the capability to analyze every image captured by these satellites over volcanoes around the world.

Timely forecasting of the evolution of lava flows is one of the key elements for assessing volcanic hazards. Lava flows are among the main hazardous phenomena during an effusive eruption, due to the possibility to reach urban areas and cause damage to infrastructure. Physical-mathematical models can be used to estimate the dynamics of a fluid or the fluid-solid interactions, in particular for the case of lava flows. However, high fidelity models require long execution times and large computational resources. Recently, artificial intelligence (AI) has been adopted to emulate physics-based models and deliver similar results, speeding up the simulations. We will discuss the possibility to use AI-based approaches to emulate highly complex numerical models used to simulate the spatio-temporal evolution of lava flows. Analyzing and treating the formal mathematical aspects of the models under analysis, we will verify and validate the models using test cases associated with the main features of lavas, discussing the accuracy and the performance offered by the two approaches.

Satellite remote sensing data are suitable to monitor global scale volcanic hazards in an efficient and timely manner. The development of monitoring systems which automatically collect and process satellite data is crucial during a volcanic crisis. The huge amount of multispectral satellite data available requires new approaches capable of processing them automatically and artificial intelligence (AI) addresses these needs. Machine learning, a type of AI in which computers learn from data, is gaining importance in volcanology. The combination of ML algorithms and satellite remote sensing in volcano monitoring has the potential of analyzing global data in near real-time for mapping and monitoring purposes. Here, an AI-based platform was developed to monitor in near real-time the volcanic activity from space. AI algorithms are used to retrieve information about the ongoing volcanic activity. Under this perspective, a key role is played by ML since it overcomes the issues related to hard coded/explicit rules by implicitly learning them from historical satellite data. Volcanic eruptions are then fully characterized in terms of their energy release, e.g. volcanic radiative power (VRP), effusive rate, quantification of the erupted products, i.e. volume, spatial extension, volcanic cloud composition. This task is achieved by combining a variety of freely available satellite datasets, i.e. infrared (IR) data with different spatial, temporal and spectral features.  In particular, both a geostationary satellite sensor, i.e. SEVIRI (Spinning Enhanced Visible and InfraRed Imager, on board Meteosat satellites), and several mid-high spatial resolution polar satellite sensors, e.g. MODIS (Moderate Resolution Imaging Spectroradiometer, on board Terra and Aqua satellites), VIIRS (Visible Infrared Imaging Radiometer Suite, on board the Suomi-NPP and NOAA-20 satellites), SLSTR (Sea and Land Surface Temperature Radiometer, on board Sentinel-3A and Sentinel-3B satellites), MSI (MultiSpectral Instrument, on board Sentinel-2), are adopted. We demonstrate the potential of this web-based satellite-data-driven platform during the recent eruptive events on Stromboli and Etna.