Characterising the dynamics around offshore wind farms by combining machine learning and oceanographic techniques (4465)

This project will use data science techniques and satellite data sets to better understand how offshore wind farms can impact the ocean. In particular, effects on the waves, water circulation and the movement of sediment can change the environment where wind farms are built and the oceans around them.

Until recently, data have mainly been gathered using boats and fixed buoys. However, scientists increasingly use satellites and autonomous vehicles to collect offshore data. These technologies are providing new data sets and this project will aim to use data science techniques, including machine-learning, to combine new, exciting measurements with established techniques leveraging their respective strengths.

The aim of this data fusion approach is to provide oceanographic data with resolution, accuracy and ease of access not previously obtained. These data sets will allow us to measure changes to the natural environment and identify the most suitable sites for offshore wind farms while protecting our oceans.

The rapid growth in offshore wind combined with the importance of commercial and environmental viability of proposed projects mean that this project can make a valuable and timely difference to spatial planning and the sustainable management of marine resources.

The project would suit someone with abilities or interest in any of the following fields, data science, the marine environment, offshore wind and remote sensing. The student will be based in the University of Exeter Penryn campus in Cornwall, South West UK. The project will include placements with 4 Earth Intelligence, also in Cornwall, where they will learn how the commercial data market operates for offshore projects alongside the commercial processes for purchasing and processing environmental satellite data.

During the final year of research a further placement with 4EI will be used to analyse how the systems and architectures developed for physical oceanographic parameters can be integrated with commercial systems.