Abstract

Historically, the recreational use of reefs has been assessed through on-water observations, which often provide limited spatiotemporal coverage, and boater interviews, which may have low response rates. To overcome these limitations, an object detection model was trained and deployed to identify small boats within high resolution (∼3 m/pixel) PlanetScope imagery at four North Carolina offshore artificial reefs during 2019. The resulting visitation time series consisted of 1319 stationary and 201 underway vessels detections. This is ∼12 times the number of visitations reported during 2019 from the Marine Recreational Information Program (MRIP) boat-ramp interviews and ∼20 times the number captured using Automatic Identification System (AIS) vessel-tracking data. Although the MRIP surveys do not report any visitations during the winter months, imagery-based detections show that the use of these reefs extends throughout the winter. Detections also show steady use of the sites on weekends, whereas the MRIP interviews indicate relatively fewer visitors on Sundays. Imagery-based detections and MRIP results show AR-315 to be the most visited reef. However, whereas the MRIP results indicate limited use of the other sites, the imagery-based visitation shows that AR-370 and AR-425 reefs are visited nearly as often, with AR-160 being under-utilized relative to these sites. These comparisons show how the low survey response rate and sparsity of tracking beacons on small vessels may lead to a biased assessment of visitation patterns. The clustering of detections within sites highlights the relative popularity of different reef materials among users. Leveraging the growing availability of high spatiotemporal resolution satellite imagery, this automated framework can be used to detect boats around other areas and times of interest. Visitation information is essential in assessing the socioeconomic impacts of these reefs, and important in the design and maintenance of current and future reefs to support local coastal economies and recreation.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call