When investigating the effect of prey on predators’ foraging, we found that foraging efficiency and effort (BDCr, dive depth, dive numbers, bout length) were affected in a complex way for both species, as the interaction between prey depth and prey aggregation was retained in all models and always explained the most partial deviance. A. torda were most efficient when prey was at intermediate shallow depths (Fig 3A) and at intermediate aggregation numbers (Fig 3C). Dive efficiency decreased with prey depth in U. aalge (halved at deepest prey depth, thought the effect was not statistically significant; Fig 3A) but increased with aggregation numbers up to intermediate levels (1 aggregation/10° azimuth), after which the steepness of the curve declined (Fig 3C). In both species, the interaction between prey depth and prey aggregation translated into a higher efficiency at deeper depth when prey aggregation numbers were low (Appendix A11), though for both species overall the prey distribution patterns explained very little of the deviance (Fig 3). A. tordadecreased the number of dives with increasing prey depth (Fig 3B).U. aalge kept foraging at deeper distributions, while A. torda stopped diving when prey distribution got deeper than 22m. As the numbers of aggregations increased, so did the number of dives with a sigmoid curve for A. torda and almost linearly for U. aalge. These effects seem to be synergistic as both species performed more dives when there were high numbers of aggregations at shallow depth (Fig A11C&D), though for U. aalge high numbers of aggregations and deeper distributions also led to more dives (Fig A11D). Both species dove deeper when prey were deeper (Fig 3E). Dive depth positively increased until 24m prey depth for U. aalge, after which the depth stagnated and started decreasing. Interestingly, the same trend was visible in A. torda but for much shallower dive depths and a decrease that continued to higher depths. However, the response to increasing number of aggregations alone was negative, with decreasing depth already at lower numbers of aggregations for A. torda thanU. aalge (Fig 3G). Both species also dove extra deep if there were many aggregations and deep distributions (Fig A11E&F).