Oceanic Lagrangian Coherent Structures have been shown to 15 deeply influence the distribution of primary producers and, at the other 16 extreme of the trophic chain, top predators. However, the relationship be-17 tween these structures and intermediate trophic levels is much more ob-18 scure. In this paper we address this knowledge gap by comparing acous-19 tic measurements of mesopelagic fish concentrations to satellite-derived 20 fine-scale Lagrangian Coherent Structures in the open ocean. The results 21 demonstrate unambiguously that higher fish concentrations are signifi-22 cantly associated with stronger Lagrangian Coherent Structures, and we 23 observe that these features represent a limiting condition for high fish 24concentrations. A model, specifically built for mid trophic levels with 25 realistic parameters, provides a possible mechanism of fish aggregation, 26 and is coherent with the observations. These results may help to inte-27 grate intermediate trophic levels in trophic models, which can ultimately 28 support management and conservation policies. 29 Introduction 30 Marine biomass distribution is highly patchy and variable in time across the 31 entire trophic web Bertrand et al. (2014); McManus and Woodson (2012). Dis-32 cerning the factors underpinning ocean patchiness is fundamental to understand 33 how they influence biogeochemical reactions and ecosystem stability Martin 34 (2003); Lévy and Martin (2013). These issues are pivotal for conservation pur-35 poses Gaines et al. (2010) and for assessing the impact of climate change on the 36 marine environment Hoegh-Guldberg and Bruno (2010).37 One of the origins of the heterogeneity of the biotic fields is the dynamic nature 38 of the ocean environments, which transforms the water masses on a large range 39 2 Fine-scale fronts as hotspots of fish aggregation in the open ocean.of temporal scales, including those of ecological relevance. In this regard, a spe-40 cial role is fulfilled by the mesoscale and submesoscale McGillicuddy (2016), now 41 commonly referred to together as "fine-scales", which span a spatial range from 42 a few to hundreds of kilometers.
43One fruitful approach for capturing the structuring effect of fine-scale dynamics 44 is the extraction of so-called Lagrangian Coherent Structures, and in particu-45 lar Lagrangian fronts Haller (2015); Lehahn et al. (2018). Lagrangian Coherent 46 Structures (LCSs) provide several types of information regarding flow proper-47 ties, such as the location of fronts, barriers to transport Boffetta et al. (2001), 48 or retentive and coherent regions d'Ovidio et al. (2013). One of the most com-49 mon Lagrangian diagnostics used to determine LCSs is the Finite-size Lyapunov 50Exponent (FSLE, d'Ovidio et al., 2004). This measures the exponential rate of 51 water parcel deformation and has maximal values over frontal regions. 52 By shaping and elongating water patches, Lagrangian Coherent Structures have 53 been demonstrated to set the frontiers of phytoplanktonic patches in terms 54 of chlorophyl...