PhD students

PhD students and projects at DTU Aqua within the research area Marine Living Resources.

Georgina VIchery. Photo: DTU Aqua

Georgina Vickery

Title of the PhD project
Applying new technologies to increase efficacy and reduce impact of fisheries stock assessment

Background of the project
Fisheries stock assessment is integral to facilitating sustainable management of this valuable marine resource. However, all sources of assessment data have limitations, from size or species selectivity of fishing gears to the spatial and temporal coverage of both fisheries dependent and independent data. Furthermore, there is increasing focus on reducing environmental impact of scientific activities and increasing the use of non-invasive sampling. If successfully integrated into existing timeseries, autonomous underwater vehicles (AUVs) and other imaging platforms coupled with AI have the potential to become a low-impact solution to the limitations of existing data collection techniques.

About the project
Combining new technology from Institute for Marine Research, Norway, with academic expertise from DTU, this PhD will integrate new technologies and methods into the three main steps of stock assessment. Species detection models with tracking and length estimation will be developed to process high intensity synthetic aperture sonar (HiSAS) data from AUVs as well as video data from towed and in-trawl cameras. These data will be integrated into existing timeseries, with a focus on optimising spatial-temporal coverage and accuracy of data collected. The final step of the project is to apply management strategy evaluation (MSE) to elucidate how the incorporation of new data from AUVs and other new technologies impacts stock assessment and the subsequent management advice.

Perspectives
Globally, fisheries institutions are transitioning to increased use of autonomous systems. This PhD project will demonstrate the full pathway of deploying an AUV to collect fisheries-independent data suitable both for use in stock assessment and to be integrated into existing timeseries. This revolutionary leap forward will facilitate sustainable management of fisheries and create a template for optimal deployment and assimilation of these systems into stock assessment processes.

Supervisors
Carsten Hvingel and Fletcher Thompson, DTU Aqua & Fabian Zimmermann, Institute for Marine Research, Norway