​

 

Millions of people live in active volcanic areas around the World and setting multi-parametric permanent monitoring systems is mandatory to reduce the risk in these areas. Integration of geophysical (e.g., seismic, deformation, infrasound signals) and geochemical data has shown to be the best approach to understand the volcanic system and identify precursors of activity. Installation of permanent gas geochemistry monitoring stations in volcanic areas became more frequent in the last two decades with different type of networks worldwide, such as NOVAC and DECADE. These networks measure essentially active plumes that release sulphur dioxide (SO₂). In long dormant volcanoes, without SO₂ release, the recognition of precursors is challenge and the permanent networks are mainly based on the continuous emission of carbon dioxide (CO₂) through the volcanic soils.

Since 2001 a permanent soil CO₂ flux station network started to be set up in the Azores archipelago and nowadays seven stations are automatically running in four dormant volcanoes. These stations have coupled several environmental sensors to understand how external variations may interfere with the gas fluxes. Several studies, based essentially in regression and spectral analysis have showed that soil gas fluxes vary up to 60% only due to external environmental changes. Recognition of these correlations between soil gas fluxes and environmental variables is a key factor to understand the volcanoes behaviour and identify precursors of unrest. The development of advanced algorithms for Data Science and machine learning turned out to be a powerful tool used in several research topics of the Earth Sciences domain. This research project pretends to apply the recent developed algorithms (e.g. data mining and neural networks) to the long soil CO₂ flux time series recorded in the Azores in order to recognize patterns that can contribute to identify precursors of volcanic activity.

The candidate will integrate the scientific unit of “Gas Geochemistry” of the IVAR, mainly focused on the study of volcanic/hydrothermal gases for both seismovolcanic monitoring purposes and risk assessment. The main objectives of this scientific unit are to define the baseline behaviour of the volcanic emissions from active volcanic systems in order to recognize signs of unrest; model and understand degassing paths in volcanic areas, as well as discriminate the origin of the gases released; quantify the emissions of volcanic gases to the atmosphere and to estimate the heat fluxes released in fumarolic grounds; characterize and quantify the indoor toxic/asphyxiating gases that may cause not only health problems to population but also interfere with all biological systems.

The candidate will develop research using long CO₂  time series recorded in volcanic environments. The study will be developed in the Azores archipelago and the main goal will be the application of advanced algorithms of data mining and neural networks to the data recorded by the permanent soil CO₂ flux stations installed in the Azorean volcanoes since 2001. This mathematical modelling will allow to highlight patterns on the CO₂ flux time series that can be used as indicators of changes on the volcanic systems.

Candidates must have a degree in Mathematics, Physics, Engineering, Geology, Geophysics, or in a similar field.

The candidate should have the following characteristics: