“My advice is to follow your passion”
Professor Andy Hooper is a geophysicist at the University of Leeds, UK who uses satellite data and machine learning to monitor volcanoes and earthquakes. Andy is also a Director of SatSense, a start-up that aims to make high-resolution ground-deformation data more accessible
Tell us about your work
My research focuses on detecting small movements of Earth’s surface and interpreting them to better understand volcanic plumbing systems and earthquake processes. These measurements can be made from radar satellites that, despite being hundreds of kilometres above us, are able to sense movements of just a few millimetres, which still amazes me. I first became fascinated by volcanoes when climbing a number of them in Indonesia and South America. Being able to now contribute to mitigating the risks posed by volcanoes, is very motivating.
What’s a typical day for you?
A typical day usually includes discussing research with PhD students and colleagues, but beyond that it is very varied, and might also include advising the European Space Agency on new satellite missions, teaching students, or making field measurements on volcanoes.
Tell us more about your work
Forecasting volcanic eruptions largely relies on human interpretation of the activity measured at individual volcanoes. However, volcanoes can behave in unexpected ways not previously seen at a specific location and most volcanoes are not instrumented, meaning we have few on-ground measurements. A key indicator of volcanic activity is deformation of the surface of a volcano, due to magma migrating beneath it. Thanks to satellite monitoring, and in particular the European satellite mission Sentinel-1, we can now measure surface movements as small as a few millimetres and are generating an incredible global dataset of deformation at volcanoes worldwide. In a project that I am leading called DEEPVOLC, my team and I aim to take advantage of this global dataset to enhance forecasts of volcanic activity. In a new approach, we are applying the latest deep learning algorithms to satellite data to combine knowledge from all volcanoes that have been active in the satellite-monitoring era – and the algorithms will continue to improve as they ingest data from new volcanic activity. This will enable us to use our knowledge of how volcanoes behave globally to forecast how unrest at a volcano may evolve locally. Importantly, we will be working closely with different volcano observatories to deliver tools that can be used operationally to monitor and forecast volcanic activity.
What advice would you give to someone hoping to work in your field?
My path to researching volcanoes was convoluted: I started a degree in geology in my late twenties. After that it took me a few years to decide that geophysics was where my interests lay, and I started a PhD in crustal deformation. My advice is to follow your passion and don’t be afraid to switch direction as your interests crystallise.
What’s your favourite thing about your work?
Travelling to scenic places like Iceland and witnessing volcanic eruptions is, of course, fantastic and an incredible privilege. But I also love interrogating and modelling data with students and collaborators – plotting results to create a figure on the screen and seeing something new for the first time is a real buzz.
Hooper, A. My advice is to follow your passion. Geoscientist 32 (3), 54, 2022; https://doi.org/10.1144/geosci2022-030