Asymmetric connectivity of spawning aggregations of a commercially important marine fish using a multidisciplinary approach

Understanding patterns of larval dispersal is key in determining whether no-take marine reserves are self-sustaining, what will be protected inside reserves and where the benefits of reserves will be observed. However, explicitly incorporating dispersal data into designing reserves for fisheries and conservation is still uncommon in many places around the world. We followed a multidisciplinary approach that merged detailed descriptions of fishing zones and spawning time at 17 sites distributed in the Midriff Island region of the Gulf of California (GC) with a biophysical oceanographic model that simulated larval transport at Pelagic Larval Duration (PLD) 14, 21 and 28 days for the most common and targeted predatory reef fish (leopard grouper Mycteroperca rosacea). M. rosacea is endemic to the GC and considered ‘Vulnerable’ according to World Conservation Union. We described metapopulation dynamics using graph theory and employed empirical sequence data from a subset of 10 sites at two mitochondrial genes to verify the model predictions. Our approach made sense of seemingly chaotic patterns of genetic diversity and structure, and provided a mechanistic explanation of the location of fishing zones. Most of the connectivity patterns observed were strictly asymmetric, except for a small region in the Southeast. The best-supported gene flow model confirmed a pulse of larvae from the Baja Peninsula, across the GC and northward up the Sonoran coastline, in agreement with the cyclonic gyre present at the peak of spawning (May). We found support that genetic diversity increased in sink sites that concentrated larvae from many sources at the time of larval flexion (PLD 14 days), while diversity decreased at important gateways identified at PLD 28 days with high betweenness centrality that are key for multigenerational dispersal and population resilience. Heavily targeted fished areas seem to be sustained by high levels of local retention, contribution of larvae from upstream sites and oceanographic patterns that concentrate larval density from all over the region. The general asymmetry in marine connectivity observed highlights that benefits from reserves are biased towards particular directions, that no-take areas need to be located upstream of targeted fishing zones, and that some fishing localities might not directly benefit from avoiding fishing within reserves located adjacent to their communities. We discuss the implications of marine connectivity for the current network of marine protected areas and no-take zones, and identify ways of improving it.