The Ross Sea Region Marine Protected Area (MPA) was implemented to understand ecosystem dynamics and ensure that the Antarctic toothfish industry was not impacting key toothfish predators in the Ross Sea, Antarctica. Weddell seals, a named focal species of the MPA, are studied to help understand the impacts this fishing industry may have in the Ross Sea. This highlights the importance of understanding and quantifying Weddell seal population dynamics over space and time.

Weddell seals are one of the most abundant seal species found in Antarctica, with McMurdo Sound colony populations among those most frequently studied since the 1950s. Seal tagging, ground counts, field camera and aerial surveys are among the most common methods employed to monitor long-term populations, with more recent approaches allowing for seal counts to be derived from satellite images and citizen science platforms. While these methods have provided a wealth of spatio-temporal data on the species, they can be particularly time- intensive, costly, and not always feasible due to logistical and accessibility constraints.

Here, a novel method is proposed that utilises the spatio-temporal benefits field camera images offer, while greatly enhancing the seal counting process through the use of a semi- automatic image object detector. Satellite and citizen science-derived seal counts were also analysed to assess their efficacy in providing valuable population data. Field camera images were trialled using images collected from three well-known seal colonies during the summer months (Big Razorback in 2010, Turtle Rock in 2014, and Scott Base 2018-19 and 2019-20) and implemented with different camera setups.

Results show that the object detection counts were not significantly different from the reference counts manually conducted by experts. For each different survey location, the R2 ranged from 0.889 to 0.997, with no absolute errors showing a strong correlation to the number of seals in each count. Detection accuracy typically dropped when the distance between seals and the camera varied greatly within an image and oblique camera angles made the gap between individual seals undiscernible.

The Big Razorback survey had the lowest object detection accuracy of all the surveys (R2 = 0.890), followed by Turtle Rock (R2 = 0.988). The most accurate detections came from the two Scott Base surveys from 2018-19 and 2019-20 (R2 = 0.995 and 0.992, respectively), when comparing object detection counts against reference counts. These results, when compared against the location and proximity of the camera relative to the seals suggest that camera setup is the most important factor in improving detection rates, rather than a limitation of the object detector’s capabilities.

The Turtle Rock dataset was also counted by citizen scientists, producing the lowest accuracy when compared against reference counts (R2 = 0.967). While the citizen science counts generally overestimated the number of seals, they did produce the same overall haul-out pattern as the object detection counts. The study also found that citizen science projects are beneficial for outreach purposes, and with some alterations to the project setup, could be implemented for future monitoring programmes.

The field camera datasets used in this study also provided a consistent set of images that allow for observations to be made of the physical environment, such as sea ice distribution and extent. This provides the opportunity to compare environmental variables against the spatio-temporal population data to understand any underlying factors that may affect haul-out and population variability. In this way, it was evident that an early breaking up of sea ice around the Dellbridge Islands led to a very large increase in Weddell seal numbers present in front of Scott Base during the 2018-19 summer.

Knowing the timing of maximum and minimum haul-out is crucial for estimating abundance from very-high resolution satellite images. As suitable satellite images are rarely available across survey periods, the timing of peak haul-out must be considered before any valid conclusions about Weddell seal abundance can be made. Satellite images were taken when seals were approximately halfway to peak haul-out, indicating that current abundance reports are likely underestimating the total number of Weddell seals. This study’s findings suggest that a combined approach of very-high resolution satellite images coupled with object detection and field camera images would provide the best approach for long-term Weddell seal monitoring for Ross Sea Region Marine Protected Area objectives.