Linking bio-oceanography and population genetics to assess larval connectivity

Abstract

Marine reserves (areas closed to fishing) have been advocated for the management of many species, including the rock scallop Spondylus calcifer in the northern Gulf of California (NGC), Mexico. We developed an explicit coupled biological-oceanographic model (CBOM) to assess connectivity among fished subpopulations of S. calcifer. We focused on the Puerto Peñasco corridor, located in the northeastern portion of the NGC. We validated CBOM's outputs through 2 different techniques: population genetics with 9 microsatellite loci and measurements of spat abundance on artificial collectors. We found strong demographic connectivity between the corridor and southern sources. Sampled localities showed low levels of genetic structure; however, we identified 2 subtly differentiated genetic clusters. On average, the spatial scale of demographic and genetic connectivity is in agreement, suggesting that connectivity decreases when the spatial scale is >100 km. We observed a gradient of higher values of both predicted particles and observed densities of settled spat for the northern and southern sites and lower values for the central sites. Larval recruitment within the corridor could be linked to a large spatial scale of larval inputs, including local sources and subpopulations further south. The absence of a strong barrier to migration suggests that the siting of marine reserves along upstream sites would likely benefit downstream subpopulations. The spatial scale of connectivity (~100 km) could be used as a reference for the strategic siting of marine reserves in the study area. CBOMs and population genetics are powerful complementary tools to assess the relative strength of connectivity among sites. KEY WORDS: Spondylus calcifer · Larval dispersal · Genetic structure · Marine reserves · Gulf of California Resale or republication not permitted without written consent of the publisherMar Ecol Prog Ser 463: [159][160][161][162][163][164][165][166][167][168][169][170][171][172][173][174][175] 2012 grounds (Gell & Roberts 2003, Beukers-Stewart et al. 2005, Pelc et al. 2009).Benthic bivalve species are structured in metapopulations in which separate subpopulations of sessile juveniles and adults are connected through larval dispersal (Lipcius et al. 2005, Orensanz et al. 2006. The extent to which these subpopulations are linked by the exchange of larvae is termed connectivity and can have multiple and different patterns (Palumbi 2003, Orensanz et al. 2006. Thus, effective design and implementation of marine reserves requires knowledge about the spatial and temporal dynamics of biological connectivity between subpopulations, particularly those that are sources of larvae that can populate other habitats (Fogarty & Botsford 2007, Pelc et al. 2010.Depending on how connectivity is measured it can have different meanings and implications (e.g. for fisheries management, conservation of species and their evolutionary potential) (Lowe & Allendorf 2010). Demographic connectivity is the degree to which the subpopulation growth ...

Ecological guidelines for designing networks of marine reserves in the unique biophysical environment of the Gulf of California

“…Later versions emphasized the need to design networks of multiple interconnected reserves to scale up their benefits and allow for multiplicative properties that are not present in individual reserves (e.g. the demographic coupling of populations in different reserves: Gaines et al 2010;Sale et al 2010;Jessen et al 2011), or to more carefully consider constraints imposed by local human activities (Fraschetti et al 2009). …”

“…Biophysical models and genetic studies indicate that larval dispersal kernels in the GOC are not spatially symmetrical but are highly constrained in particular routes by the direction of the currents driven by the narrow shape of the Gulf, particularly for species that spawn during a single season (Soria et al 2012;Munguia-Vega et al 2014, 2015a, 2018TurkBoyer et al 2014;Lodeiros et al 2016;AlvarezRomero et al 2018) (Fig. 5).…”

Designing connected marine reserves in the face of global warming

“…We used the ecosystems inhabited by each focal species (Table S6) to represent their spawning and/or recruitment habitats, and assumed they all have, potentially, equal per unit area contributions of larvae. We did not distinguish between spawning and recruitment habitats because for our targeted species both are mainly areas within rocky reef systems (Aburto-Oropeza et al, 2007, Soria et al, 2012.…”

“…Given the lack of reliable downscaled projections of physical variables that force the model, our dispersal model did not explicitly include potential changes in currents due to ocean warming (e.g., Andrello et al, 2015), which is an area of active research. Further details of the oceanographic modeling can be found in previous studies , Munguía-Vega et al, 2015, Soria et al, 2012 and are summarized in the Supplementary Material.…”

“…Our measure of the value of planning units as stepping stones was distance-weighted fragmentation , which calculates the effect of removing a node on the overall connectivity of the network (Table S7). Given our aim was to maximize whole-of-network connectivity, and that only very few previously identified locations within our planning region showed modeled and empirical evidence of high self-recruitment , Soria et al, 2012, we did not target areas for self-recruitment. In fact, the location of sites showing high local larval retention in the region are unique, and our larval dispersal model shows they are few (Table S8) and associated to small eddies that form near the northern pointy ends of large islands or coastlines.…”