Fish Aggregating Devices could enhance the effectiveness of blue water MPAs

In the past two decades, drifting fish aggregation devices (FADs) have revolutionised pelagic fisheries, and are now responsible for the majority of tuna purse seine catches. Here, we argue that by taking advantage of the same proven aggregative properties, FADs could be used to enhance the benefits provided by blue water Marine Protected Areas (MPAs). Using models of commercially-targeted fish populations, we explore the potential benefits that could be achieved if unfished conservation FADs were positioned within blue water MPAs. Our results suggest that conservation FADs could deliver benefits, both to target species and the broader ecosystem. By increasing the residence time of exploited species, conservation FADs will reduce average mortality rates inside MPAs. By increasing the local density of species whose populations are depressed by exploitation, FADs can also improve the function of ecosystems in blue water MPAs. Conservation FADs could therefore amplify the benefits of blue water MPAs. We find this amplification is largest in those contexts where blue water MPAs have attracted the most criticism - when their area is small compared to both the open ocean and the distribution of fish stocks that move through them.


INTRODUCTION 25
The open-ocean is under unprecedented threat from human activities. The creation of blue water marine 26 protected areas (MPAs; Figure 1) has been an important part of the response to this threat (Wagner 2013). Blue 27 water MPAs are large-scale (>100,000 km2) protected areas that encompass open-ocean, pelagic ecosystems, 28 although they are often centred upon oceanic islands, reefs, or seamounts. The number and extent of blue water 29 MPAs has accelerated rapidly in recent years and is likely to continue to accelerate in support of international boundaries. This is especially true for species targeted by commercial fisheries: including tuna and swordfish 39 3 (Boerder et al. 2019), whose migration patterns may encompass whole oceans. Even the largest blue water 40 MPAs could not protect individuals of these species from fishing mortality throughout their lives, a key design 41 objective of coastal and reef MPAs (Green et al. 2015). As a consequence, the ability of blue water MPAs to 42 conserve pelagic species and ecosystems is a question of intense debate (Game et     water MPAs, rather we imagine them as self-powered devices capable of remaining semi-stationary.

72
cFADs could deliver benefits to species that experience direct fishing mortality, and also to the broader 73 ecosystem. By increasing the residence time of highly-mobile and migratory species inside the protected area, 74 cFADs decrease the exposure of these species to fishing mortality. Despite the temporary nature of this 75 protection, it could nevertheless enhance stock levels, via the same mechanism as temporary closures. An  The deployment of cFADs inside blue water MPAs could therefore help to counterbalance the extensive use  However, as far as we are aware, the potential for cFADs to amplify the benefits of blue water MPAs has not 82 previously been explored, theoretically or empirically. The goal of this study is to explore the logic of this 83 proposal, by introducing cFADs to standard theoretical models of exploited fish species with MPAs.

85
Nonspatial model 86 The first model describes the effects of cFADs as simple, spatially-implicit redistributions of stock numbers.

87
Following ideas from optimal foraging theory (Visser & Fiksen 2013), we assume that each FAD attracts fish 88 by offering them greater benefits than the surrounding seascape. As a consequence, all FADs simply 89 concentrate existing animals from the surrounding seascape. This aggregative role is independent of their 90 spatial location -they aggregate fish both inside and outside protected areas -and this means that dFADs 91 could play a positive conservation role when they drift inside the MPA.

92
A stock of fish has population X, which is distributed across an area N, of which an area M has been placed 93 into a blue water MPA. The area contains ! fishing dFADs, which move at random across the entire stock 94 distribution (including some time spent inside the MPA). The area also contains " stationary cFADs, all of 95 which are permanently inside the MPA.

96
Fish gain some density-dependent benefit from being in proximity to any FAD, and the population therefore 97 aggregates in increasing density at each FAD until the marginal benefit of joining a FAD aggregation falls to 98 zero (24). We use the parameter ≪ to denote the proportion of the total fish stock that is found in the 99 vicinity of each FAD. As increases, the aggregating effect of the FADs increases, meaning a higher density 100 of fish will be found in the vicinity of each device. Those fish that are not associated with any FAD will be 101 distributed evenly across the stock distribution area N. 102 In the absence of any FADs, the time-averaged number of fish that will be found inside the MPA will then be:

Equation 1
105 Equation 1 represents the essential role of an MPA -to protect a proportion of the fish stock distribution from 106 mortality. It is also the basis of the primary critique of blue water MPAs: that their size is small relative to the 107 area of a mobile pelagic stock (i.e., ≪ ). If cFADs and dFADs are present, then the time-averaged 108 6 proportion of the stock that will be found inside the MPAs (and therefore protected from excess fishing 109 mortality) is: The first term on the right-hand side of Equation 2 represents fish that are not associated with any FAD and 113 are protected by the MPA, while the second term represents those fish that are associated with any FAD, and 114 which are currently inside the MPA. We also assume that ≤ /( ! + " ), to ensure non-negative fish 115 populations.

116
Thus, the presence of " conservation cFADs amplify the stock benefits flowing from an MPA by a factor :

Equation 3
119 For example, an amplification factor of = 1.5 indicates that the addition of cFADs increases the benefits of 120 the MPA by 50%, compared to the MPA without any cFADs. 121 We parameterised the nonspatial model with area N = 1 and a normalised stock abundance of = 1. We allow 122 the MPA to represent 2%, 4%, and 8% of the stock distribution, ranging between the current global coverage  nonspatial model in Figure 2A. 147 Since dFADs compete to attract the same fish as cFADs, our intuition suggested that more dFADs deployed 148 for fishing would result in a net export of fish from the MPA. We therefore expected that a larger number of 149 dFADs would reduce the amplification. However, this turned out not to be the case -the solution for  Figure S1. Natural mortality is assumed to be = 195 0.05, fishing mortality is % = 0.1, and density-dependent Ricker mortality is governed by parameter = 0.1.

196
The model assumes that fishing mortality in cells without a dFAD is zero, but we also investigated the effects and ecosystem functions that benefit from high fish densities (e.g., hunting seabirds), this variation makes it 225 more likely that benefits will be available at some location in the MPA.

227
The world's tuna fishing fleets increasingly rely on dFADs to enhance their catches. By producing and

244
Foraging efficiency is tightly linked to breeding success, placing a premium on consistent access to resources 245 within the foraging range of the species.

246
The effectiveness of FADs as fishing gear may be a reason for caution when deploying them for conservation 247 goals. The presence of semi-stationary cFADs in blue water MPAs may make them more attractive to illegal 248 fishers, for example. This risk seems low, however, since illegal fishers would find FADs hard to locate Our results provide a theoretical rationale for deploying cFADs to blue water MPAs. However, the models 253 that we use here are limited in a number of important ways that may affect the potential benefits offered by 254 cFADs. First, cFADs will increase the densities of any species that is attracted to pelagic floating objects, and 255 not all of these will be negatively affected by fishing effort or bycatch mortality. Over 55 bony fish species are 256 frequently found around dFADs used by the tropical tuna purse seine fishery, and many of them will be 257 abundant, fast-growing, and of low conservation concern (Amandé et al. 2008). Second, our models are based 258 on the assumption that the aggregating effects of cFADs and dFADs are equivalent. dFADs move passively 259 with ocean currents, which may place them to current boundary locations that offer better conditions for 260 environmental factors and food. Third, our models are not parameterised for any particular fish species or 261 location; as a consequence, they cannot accurately predict fish densities around cFADs, nor how these densities 262 would change with the number or location of the cFADs. Given the novelty of the cFAD concept, anticipating 263 parameter values for a particular location and species would be very difficult. Our models' goal was to provide 264 a theoretical justification for the experimental trials that would be needed to estimate these values.

265
The immediate policy question for conservation FADs in blue water MPAs is whether their benefits are large 266 enough to justify their deployment. Moving forward, we believe that the best method to answer this question