Search

Environmental monitoring guides conservation and is particularly important for aquatic habitats which are heavily impacted by human activities. Underwater cameras and uncrewed devices monitor aquatic wildlife, but manual processing of footage is a significant bottleneck to rapid data processing and dissemination of results. Deep learning has emerged as a solution, but its ability to accurately detect animals across habitat types and locations is largely untested for coastal environments. Here, we produce five deep learning models using an object detection framework to detect an ecologically important fish, luderick (Girella tricuspidata). We trained two models on footage from single habitats (seagrass or reef) and three on footage from both habitats. All models were subjected to tests from both habitat types. Models performed well on test data from the same habitat type (object detection measure: mAP50: 91.7 and 86.9% performance for seagrass and reef, respectively) but poorly on test sets from a different habitat type (73.3 and 58.4%, respectively). The model trained on a combination of both habitats produced the highest object detection results for both tests (an average of 92.4 and 87.8%, respectively). The ability of the combination trained models to correctly estimate the ecological abundance metric, MaxN, showed similar patterns. The findings demonstrate that deep learning models extract ecologically useful information from video footage accurately and consistently and can perform across habitat types when trained on footage from the variety of habitat types. [ABSTRACT FROM AUTHOR]

Understanding how salinity affects marine parasites is vital to understanding their ecology and treatment, particularly for host-parasite systems that traverse marine and freshwater realms such as the globally important Atlantic salmon (Salmo salar), salmon louse (Lepeophtheirus salmonis) system. Growing concerns for wild fish populations, and decreased efficiencies and burgeoning costs of lice treatments for farmed fish has necessitated more environmentally and socially acceptable delousing procedures, such as hyposaline treatments. The effect of brackish water on L. salmonis following primary attachment is largely unknown, with experimental evidence derived mostly from unattached or newly attached copepodids, or adult stages. We aimed to understand how attached lice respond to hyposaline environments to assess effectiveness as a parasite management strategy and to help better define delousing areas used by wild fish. Louse development at 4, 12, 19 and 26 ppt, and survival at 4 ppt, decreased as exposure times increased, but survival was otherwise unaffected. Subjecting salmon to fluctuating, repeat exposures did not influence efficacy. We confirm that free-swimming stages are susceptible, and show that attached copepodids were more tolerant than previously predicted based on experiments on alternate development stages. These results improve our understanding of the utility of hyposaline treatments in aquaculture and self-treating in wild fish, and could apply to other fish-lice parasite systems. Further, these data are important for models predicting host-parasite interactions and can contribute to predictive models on the transmission dynamics of sea lice from farm to wild fish. [ABSTRACT FROM AUTHOR]