Monitoring coastal seal colonies

Developing a broadly applicable, drone-based system for monitoring seal colonies along the entire Norwegian coastline.

Goal

The goal is to develop a broad system for monitoring seal colonies along the entire Norwegian coastline.

It will be reached in two steps:

moving beyond small-scale operations with small drones launched locally, towards longer-range operations using autonomous flight operations.
building a data pipeline for automation of most steps from data acquisition, archiving, image classification and results generation. One key component of this is the development of machine learning approaches to seal detection in drone-derived imagery. This is done through close collaboration with SeaBee Data Analysis team

Partners

The application is led by Martin Biuw, from Institute of Marine Research (IMR). The work is in collaboration with NTNU, NR and NINA.

SeaBee Technology

For local operations, IMR uses use small drones (mostly DJI Mavic 2 Enterprise Dual), which can be hand-launched and retrieved from small boats. This allows testing of infrared sensor technology to facilitate seal detection.

In 2022, we began exploring options for larger platforms operating over longer ranges. Here, a Mugin-2 Pro 2930MM H-Tail Full Carbon Fiber VTOL (Vertical Take-Off and Landing) drone was used, operated from a ground station and obtaining RGB images with a YYY camera along pre-programmed transects covering entire groups of skerries where harbour seals are known to haul out.

For the data pipeline, we will use the approach developed by the seabird team at NINA. For automatic detection of seals, we will collaborate with NR to modify the existing system developed for seals breeding in the Arctic pack ice (harp and hooded seals).

Current results

As part of IMR’s ongoing efforts to monitor coastal seal colonies, a large dataset of drone-based images has been gathered over the past 7 years. These images have already been used to further train the existing machine learning system for seal detection on ice floes, so that it is better tuned for detecting seals on rocky shores.

In 2022, tests of long-range BVLOS operations were carried out in collaboration with NTNU at Tarva, an area where harbour seals are known to occupy, in Trøndelag. We used the SeaBee equipment mentioned above. The team carried out three flights, covering two known seal localities within the Undtarva archipelago, obtaining a total of 1196 images (see example below). Flight operations were highly efficient and telemetry of data from the aircraft to the ground station proceeded without issues, showing the reliability of the system, and the feasibility of BVLOS operations for large-scale coverage.

Datasets

As datasets for this application become available, they will be published at the GeoNode Server.

Next Steps

Further training of the automated seal detection system will be carried out in 2023 based on an even greater collection of aerial photographs, also including earlier images collected from manned aircraft.

Further developments of the BVLOS approach are being undertaken, including the purchasing of a higher-resolution RGB camera and an infrared sensor for improved seal detection.

Relevant policies

The approach described here has the potential to greatly improve the coverage and frequency of seal assessments carried out by IMR in support of the national coastal seal management plan. This is highly relevant, also in view of increased developments of human maritime activities along the coast (e.g. offshore wind, offshore aquaculture, fishing), and the associated risks of negative impacts on ecosystems, including seals.

Contact Us

If you have any comments, questions, or concerns, please fill out the form below and we will get in touch with you shortly.