How to capture fish in a school? Effect of successive predator attacks on seabird feeding success.

Prey aggregations, such as fish schools, attract numerous predators. This typically leads to the formation of multispecific groups of predators. These aggregations can be seen both as a place of increased competition and as a place of possible facilitation between predators. Consequently, the functional role of such predator-prey aggregation is uncertain, and its effect on individual feeding success is virtually unknown. Using underwater film footage of different predators feeding on fish schools during the sardine run in South Africa, we directly measured the in situ feeding success of individual Cape gannets Morus capensis in different foraging situations. We determined the types of Cape gannet attacks (direct plunge dive or plunge dive followed by underwater pursuit) and we measured the occurrences and timing of attacks from the different species (mostly Cape gannets and long-beaked common dolphins Delphinus capensis). We also estimated the size of the targeted fish schools. These observations were complemented with a simulation model to evaluate the cumulative effect of successive predator attacks on the prey aggregation structure. The probability to capture a fish in one feeding attempt by Cape gannets averaged 0·28. It was lower when gannets engaged in underwater prey pursuit after the plunge compared to direct plunge (0·13 vs. 0·36). We found no effect of the number of prey on gannets' feeding success. However, the timing and frequency of attacks influenced strongly and positively the feeding success of individuals. The probability to capture a fish was the lowest (0·16) when no attack occurred in the few seconds (1-15 s) prior to a dive and the highest (˜0·4, i.e. more than twice) when one or two attacks occurred during this time window. The simulation model showed that a prey aggregation disorganized just after an attack and that the maximum of disturbance was obtained a few seconds after the initiation of the successive attacks. Our study suggests that, in multispecies predator assemblages, the cumulative effect (through disorganization of school cohesiveness) of the multiple species attacking a prey aggregation may increase the feeding success of each individual. Therefore, facilitation between predators is likely to overcome competition in these multispecific assemblages.

Linking local retention, self-recruitment, and persistence in marine metapopulations.

Three indices of larval retention have been used in the literature to assess the tendency for self-maintenance of local marine populations: local retention (LR), self-recruitment (SR), and relative local retention (RLR). Only one of these, LR, defined as the ratio of locally produced settlement to local egg production, has a clear relationship to self-persistence of individual sites. However, SR, the ratio of locally produced settlement to settlement of all origins at a site, is generally easier to measure experimentally. We use theoretical, simulation, and empirical approaches to bridge the gap between these different indices, and demonstrate that there is a proportional relationship between SR and LR, for metapopulations close to a stable state and with lifetime egg production (LEP) approximately uniform over space. Similarly, for systems where larval mortality rates are a relatively uniform function of release site, RLR (defined as the ratio of locally produced settlement to all settlement of local origin) and LR will also be proportional. Therefore, SR and RLR provide information on relative rates of LR for systems satisfying these conditions. Furthermore, the ratio between LR and SR can be used to evaluate global persistence of metapopulations, and therefore provides valuable information not necessarily available if only LR is considered.

A new Sargassum drift model derived from features tracking in MODIS images.

Massive Sargassum stranding events affect erratically numerous countries from the Gulf of Guinea to the Gulf of Mexico. Forecasting transport and stranding of Sargassum aggregates require progress in detection and drift modelling. Here we evaluate the role of currents and wind, i.e. windage, on Sargassum drift. Sargassum drift is computed from automatic tracking using MODIS 1 km Sargassum detection dataset, and compared to reference surface current and wind estimates from collocated drifters and altimetric products. First, we confirm the strong total wind effect of ≈3 % (≈2 % of pure windage), but also show the existence of a deflection angle of ≈10° between Sargassum drift and wind directions. Second, our results suggest reducing the role of currents on drift to 80 % of its velocity, likely because of Sargassum resistance to flow. These results should significantly improve our understanding of the drivers of Sargassum dynamics and the forecast of stranding events.

Minimum drift times infer trajectories of ghost nets found in the Maldives.

This study explores methods to estimate minimum drift times of ghost nets found in the Maldives with the aim of identifying a putative origin. We highlight that percentage cover of biofouling organisms and capitulum length of Lepas anatifera are two methods that provide these estimates. Eight ghost nets were collected in the Maldives and estimated drift times ranged between 7.5 and 101 days. Additionally, Lagrangian simulations identified drift trajectories of 326 historical ghost nets records. Purse seine fisheries (associated with Korea, Mauritius, the Philippines, Spain, France and Seychelles) and gill nets from Sri Lanka were identified as 'high risk' fisheries with regard to likley origins of ghost nets drifting into the Maldives. These fisheries are active in areas where dense particle clusters occured (drift trajectories between 30 and 120 days). Interestingly, ghost nets drifting less than 30 days however, remained inside the exclusive economic zone of the Maldivian archipelago highlighting potential illegal, unreported and unregulated fishing activity is occuring in this area. This study therefore points to the urgent need for gear loss reporting to be undertaken, especially by purse seine and gill net fisheries in order to ascertain the source of this major threat to marine life. This should also be coupled with an improvment in the data focused on spatial distribution of the abandoned, lost or discarded fishing gear originating from both large- and small-scale fisheries.

Modeling larval dispersal for the gilthead seabream in the northwestern Mediterranean Sea.

To investigate dispersal and connectivity between spawning and lagoon nursery habitats of the gilthead seabream, Sparus aurata, in the Gulf of Lions (northwestern Mediterranean Sea), we modeled the potential transport of the species' larvae between its supposed main spawning site in the region (the Planier Island) and two of its main local nursery areas (the coastal lagoons of Thau and Salses-Leucate). Passive larval drift simulations using a dispersal biophysical model showed a large variability in the possible trajectories from spawning to nursery areas and in the predicted ages for larvae arrival on the two nursery sites. The most common ages at arrival obtained in the simulations (20-60 days) are broadly consistent with previous modeling studies but contrast with the actual ages of the S. aurata post-larvae collected in 2016 and 2017 at time of the lagoon entrances (60-90 days, from otolith readings). The period between 25 and 70 days being critical for gilthead seabream larvae to acquire sufficient swimming, osmoregulatory, and olfactory abilities to enter coastal lagoons, we argue that ontogenic development plays a crucial role in the transport and local retention of S. aurata larvae in the studied region, explaining the discrepancy between simulation results and observed data.

Use of Lagrangian simulations to hindcast the geographical position of propagule release zones in a Mediterranean coastal fish.

The study of organism dispersal is fundamental for elucidating patterns of connectivity between populations, thus crucial for the design of effective protection and management strategies. This is especially challenging in the case of coastal fish, for which information on egg release zones (i.e. spawning grounds) is often lacking. Here we assessed the putative location of egg release zones of the saddled sea bream (Oblada melanura) along the south-eastern coast of Spain in 2013. To this aim, we hindcasted propagule (egg and larva) dispersal using Lagrangian simulations, fed with species-specific information on early life history traits (ELTs), with two approaches: 1) back-tracking and 2) comparing settler distribution obtained from simulations to the analogous distribution resulting from otolith chemical analysis. Simulations were also used to assess which factors contributed the most to dispersal distances. Back-tracking simulations indicated that both the northern sector of the Murcia region and some traits of the North-African coast were hydrodynamically suitable to generate and drive the supply of larvae recorded along the coast of Murcia in 2013. With the second approach, based on the correlation between simulation outputs and field results (otolith chemical analysis), we found that the oceanographic characteristics of the study area could have determined the pattern of settler distribution recorded with otolith analysis in 2013 and inferred the geographical position of main O. melanura spawning grounds along the coast. Dispersal distance was found to be significantly affected by the geographical position of propagule release zones. The combination of methods used was the first attempt to assess the geographical position of propagule release zones in the Mediterranean Sea for O. melanura, and can represent a valuable approach for elucidating dispersal and connectivity patterns in other coastal species.

Spatial synchrony in host-parasitoid models using aggregation of variables.

We consider a host-parasitoid system with individuals moving on a square grid of patches. We study the effects of increasing movement frequency of hosts and parasitoids on the spatial dynamics of the system. We show that there exists a threshold value of movement frequency above which spatial synchrony occurs and the dynamics of the system can be described by an aggregated model governing the total population densities on the grid. Numerical simulations show that this threshold value is usually small. This allows using the aggregated model to make valid predictions about global host-parasitoid spatial dynamics.

Influence of Biological Factors on Connectivity Patterns for Concholepas concholepas (loco) in Chile.

In marine benthic ecosystems, larval connectivity is a major process influencing the maintenance and distribution of invertebrate populations. Larval connectivity is a complex process to study as it is determined by several interacting factors. Here we use an individual-based, biophysical model, to disentangle the effects of such factors, namely larval vertical migration, larval growth, larval mortality, adults fecundity, and habitat availability, for the marine gastropod Concholepas concholepas (loco) in Chile. Lower transport success and higher dispersal distances are observed including larval vertical migration in the model. We find an overall decrease in larval transport success to settlement areas from northern to southern Chile. This spatial gradient results from the combination of current direction and intensity, seawater temperature, and available habitat. From our simulated connectivity patterns we then identify subpopulations of loco along the Chilean coast, which could serve as a basis for spatial management of this resource in the future.

Integrative biology: linking levels of organization.

Biological systems are composed of different levels of organization. Usually, one considers the atomic, molecular, cellular, individual, population, community and ecosystem levels. These levels of organization also correspond to different levels of observation of the system, from microscopic to macroscopic, i.e., to different time and space scales. The more microscopic the level is, the faster the time scale and the smaller the space scale are. The dynamics of the complete system is the result of the coupled dynamical processes that take place in each of its levels of organization at different time scales. Variables aggregation methods take advantage of these different time scales to reduce the dimension of mathematical models such as a system of ordinary differential equations. We are going to study the dynamics of a system which is hierarchically organized in the sense that it is composed of groups of elements that can be themselves divided into further smaller sub-groups and so on. The hierarchical structure of the system results from the fact that the intra-group interactions are assumed to be larger than inter-group ones. We present aggregation methods that allow one to build a reduced model that governs a few global variables at the slow time scale.

Optimal number of sites in multi-site fisheries with fish stock dependent migrations.

We present a stock-effort dynamical model of a fishery subdivided into fishing zones. The stock corresponds to a fish population moving between these zones, on which they are harvested by fishing fleets. We consider a linear chain of identical fishing zones. Fish movements between the zones, as well as vessels displacements, are assumed to take place at a faster time scale than the variation of the stock and the change of the fleet size. The vessels movements between the fishing areas are assumed to be stock dependent, i.e. the larger the stock density is in a zone the more vessels tends to remain in it. We take advantage of these two time scales to derive a reduced model governing the dynamics of the total harvested stock and the total fishing effort. Under some assumption, we obtain either a stable equilibrium or a stable limit cycle which involves large cyclic variations of the total fish stock and fishing effort. We show that there exists an optimal number of fishing zones that maximizes the total catch at equilibrium. We discuss the results in relation to fish aggregating devices (FADs) fisheries.

Continuous cycling of grouped vs. solitary strategy frequencies in a predator-prey model.

We present a model of predator and prey grouping strategies using game theory. As predators respond strategically to prey behavior and vice versa, the model is based on a co-evolution approach. Focusing on the "many eyes-many mouths" trade-off, this model considers the benefits and costs of being in a group for hunting predators and foraging prey: predators in a group have more hunting success than solitary predators but they have to share the prey captured; prey in a group face a lower risk of predation but greater competition for resources than lone prey. The analysis of the model shows that the intersections of four curves define distinct areas in the parameter space, corresponding to different strategies used by predators and prey at equilibrium. The model predictions are in accordance with empirical evidence that an open habitat encourages group living, and that low risks of predation favor lone prey. Under some conditions, continuous cycling of the relative frequencies of the different strategies may occur. In this situation, the proportions of grouped vs. solitary predators and prey oscillate over time.

Migration frequency and the persistence of host-parasitoid interactions.

This paper analyses the effect of migration frequency on the stability and persistence of a host-parasitoid system in a two-patch environment. The hosts and parasitoids are allowed to move from one patch to the other a certain number of times within a generation. When this number is low, i.e. when the time-scales associated with migration and demography are of the same order, host-parasitoid interactions are usually not persistent. When this number is high, however, persistence is more likely. Moreover, in this situation, aggregation methods can be used to simplify the proposed initial model into an aggregated model describing the dynamics of both the total host and parasitoid populations. Analysis of the aggregated model shows that the system reaches a stable steady state for some regions of the parameter domain. Persistence occurs when the movement of the parasitoids is asymmetrical, i.e. they move preferentially to one of the two patches. We show that the growth rate of the host population is a key parameter in determining which migration strategies of the parasitoids lead to persistent host-parasitoid interactions.