Impacts of submarine volcano eruptions on the shallow ocean from satellite imagery and Machine learning

Abstract
Although there are more volcanoes on the sea floor than on land, submarine eruptions are very hard to observe (only 300 eruptions are known in the past 10,000 years, relative to 8000 on land). Their impacts on the ocean are therefore hard to observe, but are known to be important: influencing ocean. geochemistry and providing important nutrients to the ocean ecosystems. In recent years, there have been frequent eruptions of submarine volcanoes (e.g., Volcano in 2019, Lateiki in 2019, and Kavachi, 2022). Although the general volcanic process of submarine volcanic eruptions is like that of subaerial eruptions, many factors are less well understood, such as impact of hydrostatic pressure on an eruption column. Both sub-aerial and submarine eruptions can cause the increase of nutrients in the water – which act as phytoplankton fertilizer and are necessary for the rapid recovery of a marine ecosystem (sometimes critical in oligotrophic oceanic regions).
Detection of submarine eruptions is a complex process due to their remote location associated with the unpredictability of these events. The subgroup of submarine eruptions that have expressions at the ocean’s surface, are mainly explosive and have shallow vents – for example, Havre Seamount is a submarine volcano that last erupted in 2012. Although discovered at a depth greater than 800 meters, the eruption’s explosiveness and other factors caused volcanic products from this event to be detected on the ocean’s surface. Hence, these eruptions can have many expressions at the ocean’s surface, like discolouration plumes, ephemeral islands, pumice rafts, ash clouds, and even volcanic explosions. It is, therefore, vital not only to detect and monitor the eruptions but also to track their associated products in time. The latter can pose significant risks for passing ship vessels, air traffic, and nearby populations. Monitoring is a step that helps mitigate the hazards concerning submarine eruptions, such as Tsunamis (as was the case of the January 2022 Hunga Tonga-Hunga Ha’apai eruption). Measuring the impact of explosive submarine eruptions on ocean chemistry and ecosystems is also critical to understand their role in ocean environment.
Satellite imagery has proven helpful in volcano monitoring and related volcanic hazard assessment. Submarine volcanoes release elements, including iron (Fe), aluminium (Al), and silicon (Si), depending on their geochemical activity, and these chemicals change the colour of the surrounding ocean. Generally, a higher proportion of Fe produces a yellow or brown colour, whereas a higher proportion of Al or Si produces a white/green colour. Nevertheless, this identification is not easy since the discoloured volcanic seawater is determined not by the absolute values of Fe, Al, and Si but by the relative mixing ratio.
Hence, this PhD project has the objective of using various case studies of submarine eruption time series, such as El Hierro (2011-2012) and Fukutoku-Oka-no-Ba (2005, 2010 and 2021) example, to develop a tool that can automatically detect submarine volcanic events, as well as monitoring the evolution of the eruption, volcanic products, and changes in the near-surface ocean waters. Twenty-three years of MODIS/AQUA data (from 2003 to 2026) will be used along with Sentinel-2 imagery for each case scenario – other sensors can be used for comparison purposes. The inter-relation of the volcanic plumes with Ocean Colour (OC) parameters, such as the Diffuse Attenuation Coefficient for Downwelling Irradiance at 490 nm (Kd490), Particulate Organic Carbon (POC), Particulate Inorganic Carbon (PIC), Chlorophyll-a Concentration (Chla), and other reflectance (Rrs) products will be analysed. Additional satellite derived geophysical parameters such as sea surface temperature (SST) and AVISO altimetry products (e.g., Sea Surface Height (SSH), Sea Surface Anomaly (SSA), currents) may also be used to help detect and follow some of these events. Artificial Intelligence (AI) and Machine Learning (ML) algorithms shall be applied to the RS data and the model(s) created shall be validated in areas where submarine eruptions are confirmed. It is expected that the model successfully separates discolouration plumes, especially near the source, from other targets, such as Chla blooms and that the evaluation analyses can extract information about the time-period of eruptions, surface and near-surface ocean changes and their evolution, and associated biological (e.g., phytoplankton) responses.
Overall, this PhD project will create a tool that allows the detection of submarine volcanic events and trace their impacts on the ocean. The model would do this through the combined analysis of current meteorological and oceanic data, which allows monitoring these events and their products in near-real-time (NRT), helping to mitigate associated hazards. This study should also increase the knowledge of submarine volcanic processes and products, opening many doors for future applications for other events. Furthermore, by combining RS with AI and ML techniques, it will be possible to provide additional decision support for public authorities in the events of future submarine volcanic eruptions.

Candidate’s Profile
Strong preference is given to candidates with advanced scientific degrees (MSc or equivalent) in the Geosciences field, with a background in Remote Sensing applied to Submarine Volcanism. Candidates should have proven experience in Ocean Colour data analysis software and some programming language knowledge. Relevant experience in research and scientific publications in these fields can constitute an advantage.

Name(s) of supervisor(s)
Ana Maria Martins – Department of Oceanography and Fisheries, Faculty of Sciences at the University of the Azores (Portugal) and Okeanos-UAc – Oceanography and Remote Sensing
Susanna Ebmeier – School of Earth & Environment, University of Leeds – Geophysics, Volcanology and Remote Sensing
Issah Nazif Suleiman – Eyecon Group – Applied Mathematics and Remote Sensing

Identification of PhD Program

PhD Programme in Sciences of the Sea, University of the Azores, Portugal, Universidade dos Açores | (uac.pt)

Notice of the call (english version)

Notice of the call ( portuguese version)

We are no longer accepting applications for this scholarship. Thank you.