Big data, such as vessel monitoring system (VMS) data, can provide valuable information on fishing behaviours. However, conventional methods of detecting behaviours in movement data are challenged when behaviours are briefer than signal resolution. We investigate options for improving detection accuracy for short-set fisheries using 581 648 position records from 181 vessels in the Gulf of Mexico bandit-reel fishery. We first investigate the effects of increasing VMS temporal resolution and find that detection accuracy improves with fishing-set duration. We then assess whether a feature engineering approach—in our case, changing the way pings are labelled when training a classifier—could improve detection accuracy. From a dataset of 12 184 observed sets, we find that the conventional point-labelling method results in only 49% of pings being correctly labelled as ‘fishing’, whereas a novel window-labelling method results in 88% of records being labelled as ‘fishing’. When the labelled data are used to train classifiers, point labelling attains true-positive/balanced-accuracy rates of only 37%/66%, whereas window labelling achieves 68%/83%. Finally, we map fishing distribution using the two methods, and show that point labelling underestimates the extent of fishing grounds by ∼33%, highlighting the benefits of window labelling in particular, and feature engineering approaches in general.

Abstract. We report on patterns of abundance, recruitment, and predation on the blue mussel (Mytilus edulis) in three human-dominated estuaries in the northeastern United States. Through replicate field experiments and observational studies at multiple sites nested within each of the three estuaries, we investigated the relative influences of local and regional scale variation in select bottom-up and top-down factors on blue mussel populations on wave-protected rocky shores. The most striking result was the decoupling between adult abundance and recruitment: mussel recruitment rates were highest in the most northern estuary, Casco Bay, while adult abundances were highest in the most southern estuary, Long Island Sound. We detected evidence of top-down forcing on adult abundance by consumers in the two more southern estuaries, Narragansett Bay and Long Island Sound, but not in Casco Bay. Finally, we observed some indications of bottom-up forcing on mussel abundance and recruitment at the withinestuary scale, but these signals were not consistent among estuaries or across the responses measured (e.g., adult abundances and recruitment rates). Our results support previous work demonstrating the importance of both top-down and bottom-up influences on rocky shore populations, and also highlight how future research-particularly integrating studies of the different ontogenetic stages of mussels-could further advance understanding of biological population dynamics in this and other systems.