Many fishers diversify their income by participating in multiple fisheries, which has been shown to significantly reduce year-to-year variation in income. The ability of fishers to diversify has become increasingly constrained in the last few decades, and catch share programs could further reduce diversification as a result of consolidation. This could increase income variation and thus financial risk. However, catch shares can also offer fishers opportunities to enter or increase participation in catch share fisheries by purchasing or leasing quota. Thus, the net effect on diversification is uncertain. We tested whether diversification and variation in fishing revenues changed after implementation of catch shares for 6,782 vessels in 13 US fisheries that account for 20% of US landings revenue. For each of these fisheries, we tested whether diversification levels, trends, and variation in fishing revenues changed after implementation of catch shares, both for fishers that remained in the catch share fishery and for those that exited but remained active in other fisheries. We found that diversification for both groups was nearly always reduced. However, in most cases, we found no significant change in interannual variation of revenues, and, where changes were significant, variation decreased nearly as often as it increased.diversification | risk | catch shares | fisheries

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.

Decision-making agents face a fundamental trade-off between exploring new opportunities with risky outcomes versus exploiting familiar options with more certain but potentially suboptimal outcomes. Although mediation of this trade-off is essential to adaptive behavior and has for decades been assumed to modulate performance, the empirical consequences of human exploratory strategies are unknown beyond laboratory or theoretical settings. Leveraging 540,000 vessel position records from 2494 commercial fishing trips along with corresponding revenues, here we find that during undisturbed conditions, there was no relationship between exploration and performance, contrary to theoretical predictions. However, during a major disturbance event which closed the most-utilized fishing grounds, explorers benefited significantly from less-impacted revenues and were also more likely to continue fishing. We conclude that in stochastic natural systems characterized by non-stationary rewards, the role of exploration in buffering against disturbance may be greater than previously thought in humans.