Laura Sandri is one of those researchers who loves what she does. Since childhood, she has been fascinated by rocks and geology, which led her to study various forces of nature such as earthquakes and volcanoes. A mother of two grown-up children and two cats (named after New Zealand volcanoes, Ruapehu and Ngauruhoe), she holds a PhD in Geophysics and a degree in Physics.
Serving as a senior researcher at INGV in Bologna, Italy, a key partner of the Centre of Excellence (CoE) for Exascale in Solid Earth (ChEESE), her work primarily focuses on Italian volcanoes such as Campi Flegrei, Vesuvio, and Etna, her expertise extends globally, encompassing volcanic studies in countries like New Zealand and Peru.
In this second interview of the Meet the Researcher series organized by the Dissemination team of the ChEESE, we delve into Laura’s professional journey within the field of geophysics. We’ll explore her notable contributions to the Project and how she leverages her expertise in probabilistic volcanic hazard assessment to mitigate risks associated with volcanic activity.
What are INGV’s main tasks in the project?
I am leading Task 3 within Work Package 6 on Scientific Cases that need capacity in computation. It means that I am coordinating the efforts made in the project in which HPC is requested to run a large number of simulation instances to explore the effect of different boundary and initial conditions on the output from a simulator of a natural process. For example, we may need to run a simulator of a tsunami due to many different earthquakes, characterized by different hypocenters, earthquake mechanisms, and magnitudes, which are all unknown for the next seismogenic tsunami.
However, as scientists, we need to account for uncertainties and that is how we do it: we consider different plausible “scenarios” (the initial and boundary conditions”) and we assign a probability to each of them. The output of HPC simulation from each scenario considered is weighted with the probability of the scenario itself, and this is how we produce a probabilistic hazard map. This process requires capacity, that is a large number of computationally heavy simulations. And this is where ChEESE enables this.
Can you explain how modeling and methods such as Bayesian approaches, pattern recognition, and Machine Learning play a role in hazard assessment?
Hazard assessment is made by weighting many different simulations with the probability of each scenario. Bayesian approach is the mathematical framework that justifies this approach. Pattern recognition and machine learning are useful, in my opinion, in the forecasting of hazardous events, as they may be valuable tools to explore large datasets of data for example of monitoring data. They are able to find patterns and schemes in data that the human eye cannot detect. However, they need a massive amount of data and this may not be the case for rare geophysical events (for example, for Mount Vesuvius we have no pre-eruptive monitoring data since the last eruption occurred in 1944).
Are there any particularly challenging aspects or obstacles that you’ve encountered in your research in the project, and how have you overcome them?
HPC is challenging. We need to deal with huge amounts of simulations over large domains at high resolution. We need to put together different expertise, from earth scientists and geophysicists as I am, to computational scientists and mathematicians and engineers as some of my colleagues are.
Can you discuss any recent achievements or milestones that your team has reached within the framework of the project? How do you foresee the impact of your research on European society?
One of the things I am doing in the project, within the task I am coordinating, is to start setting up a European hazard map for volcanic tephra. This map would enable Europe to take a large step in terms of preparedness to manage emergencies. For example, during the famous 2010 Eyjafjallajokull eruption in Iceland, European civil aviation was disrupted and we were largely not prepared for such type of disruption. Going towards a hazard map for tephra for the whole continent would be important for civil aviation purposes.
“One of the things I am doing in the project, within the task I am coordinating, is to start setting up a European hazard map for volcanic tephra. This map would enable Europe to take a large step in terms of preparedness to manage emergencies”
Laura Sandri
As a senior researcher in Volcanology, what inspired you to pursue this specific area of research, and what excites you most about your work?
I have always been fond of rocks, geology, and ultimately by all the displays of the huge force of Nature such as thunderstorms, earthquakes. So the majesty of volcanoes and their activity is still fascinating me utterly. The funny thing is that in the end I did not study geology but physics. But I managed to go back to geophysics during my PhD.
What would you be if you were not a researcher?
Not sure, but maybe I would work as a sports teacher. That was my B plan.
For aspiring scientists or individuals interested in a career in geophysics and HPC research, what advice would you give to inspire and guide them in pursuing this field?
I would tell them to take care of their curiosity and keep it alive.
Beyond the realm of science, can you share a cultural piece (film, book, podcast) that has significantly influenced your perspective or approach to your work?
I don’t think there is any book or film (and podcasts are so young!). What I remember was a photo exhibition from “exotic” travels I saw when I was something like 12, it must have been in the second half of the 80’s. There was a session on Iceland and I got totally crazy about it: Iceland at that time was very far and very remote, I did not know anyone who had been there or came from there, but I was completely taken by the beauty and majesty of mountains, volcanoes, ice, rivers, and waterfalls. I told my parents that one day I wanted to go there, and I did, since I did my ERASMUS year there in 1996-97.
By Aerton Guimarães
From the ChEESE-2P Dissemination Team
