Inferring connectivity range in submerged aquatic populations (Ruppia L.) along European coastal lagoons from genetic imprint and simulated dispersal trajectories
Triest, L.; Sierens, T.; Menemenlis, D.; Van der Stocken, T. (2018). Inferring connectivity range in submerged aquatic populations (Ruppia L.) along European coastal lagoons from genetic imprint and simulated dispersal trajectories. Front. Plant Sci. 9: 20. https://dx.doi.org/10.3389/fpls.2018.00806
In: Frontiers in Plant Science. Frontiers Media: Lausanne. e-ISSN 1664-462X, more
Coastal salt- and brackish water lagoons are unique shallow habitats characterized by beds of submerged seagrasses and salt-tolerant Ruppia species. Established long-term and large-scale patterns of connectivity in lagoon systems can be strongly determined by patterns of nearshore and coastal currents next to local bird-mediated seed dispersal. Despite the importance of dispersal in landscape ecology, characterizing patterns of connectivity remains challenging in aquatic systems. Here, we aimed at inferring connectivity distances of Ruppia cirrhosa along European coastal lagoons using a population genetic imprint and modeled dispersal trajectories using an eddy-resolving numerical ocean model that includes tidal forcing. We investigated 1,303 individuals of 46 populations alongside subbasins of the Mediterranean (Balearic, Tyrrhenian, Ionian) and the Atlantic to Baltic Sea coastline over maximum distances of 563–2,684 km. Ten microsatellite loci under an autotetraploid condition revealed a mixed sexual and vegetative reproduction mode. A pairwise FST permutation test of populations revealed high levels of historical connectivity only for distance classes up to 104–280 km. Since full range analysis was not fully explanatory, we assessed connectivity in more detail at coastline and subbasin level using four approaches. Firstly, a regression over restricted geographical distances (300 km) was done though remained comparable to full range analysis. Secondly, piecewise linear regression analyses yielded much better explained variance but the obtained breakpoints were shifted toward greater geographical distances due to a flat slope of regression lines that most likely reflect genetic drift. Thirdly, classification and regression tree analyses revealed threshold values of 47–179 km. Finally, simulated ocean surface dispersal trajectories for propagules with floating periods of 1–4 weeks, were congruent with inferred distances, a spatial Bayesian admixed gene pool clustering and a barrier detection method. A kinship based spatial autocorrelation showed a contemporary within-lagoon connectivity up to 20 km. Our findings indicate that strong differentiation or admixtures shaped historical connectivity and that a pre- and post LGM genetic imprint of R. cirrhosa along the European coasts was maintained from their occurrence in primary habitats. Additionally, this study demonstrates the importance of unraveling thresholds of genetic breaks in combination with ocean dispersal modeling to infer patterns of connectivity.
All data in the Integrated Marine Information System (IMIS) is subject to the VLIZ privacy policy
