Data from: Self-recruitment in a Caribbean reef fish: a method for approximating dispersal kernels accounting for seascape

Characterizing patterns of larval dispersal is essential to understanding
the ecological and evolutionary dynamics of marine metapopulations. Recent
research has measured local dispersal within populations, but the
development of marine dispersal kernels from empirical data remains a
challenge. We propose a framework to move beyond point estimates of
dispersal towards the approximation of a simple dispersal kernel, based on
the hypothesis that the structure of the seascape is a primary predictor
of realized dispersal patterns. Using the coral reef fish Elacatinus lori
as a study organism, we use genetic parentage analysis to estimate
self-recruitment at a small spatial scale (<1 km). Next, we
determine which simple kernel explains the observed self-recruitment,
given the influx of larvae from reef habitat patches in the seascape at a
large spatial scale (up to 35 km). Finally, we complete parentage analyses
at six additional sites to test for export from the focal site and compare
these observed dispersal data within the metapopulation to the predicted
dispersal kernel. We find 4.6% self-recruitment (CI95%: ±3.0%) in the
focal population, which is explained by the exponential kernel y = 0.915x
(CI95%: y = 0.865x, y = 0.965x), given the seascape. Additional parentage
analyses showed low levels of export to nearby sites, and the best-fit
line through the observed dispersal proportions also revealed a declining
function y = 0.77x. This study lends direct support to the hypothesis that
the probability of larval dispersal declines rapidly with distance in
Atlantic gobies in continuously distributed habitat, just as it does in
the Indo-Pacific damselfishes in patchily distributed habitat.