Human‐facilitated jump dispersal of a non‐native frog species on Hawai'i Island

Abstract Aim The coqui frog (Eleutherodactylus coqui) was introduced to the Hawai'ian archipelago in the late 1980s and became established as a widespread species on Hawai'i Island over a short timespan, suggesting that humans are facilitating their movement. To determine the importance of human facilitation, we assessed dispersal patterns and genetic structure of coqui populations using microsatellite data. Location Hawai'i Island. Methods We obtained genotype data using seven microsatellites from coqui specimens collected from 25 populations on Hawai'i Island. The dispersal mechanism was examined using a Mantel test in GenAlEx and a genetic distance tree analysis in Phylip. Allelic diversity, measures of equilibrium, and genetic structure were analysed in GenAlEx and Arlequin. The correlation between genetic distance and geographical distance was used to distinguish between diffusion dispersal (positive correlation) and jump dispersal (zero or negative correlation). Results The Mantel test for isolation by distance found no significant correlation between genetic and geographical distance (r2 = 0.002, P = 0.4401). The genetic distance tree topology is consistent with this result and exhibited a pattern expected if population establishment occurred through jump dispersal. Migration rates were high (NM = 4.228), inbreeding was high, genetic differentiation between populations was low, and significant genetic structure was detected among populations (4% of total variation, P < 0.002). Main conclusions Genetic distance is not correlated with geographical distance, suggesting that humans are important facilitators of coqui dispersal. Migration rate was high, indicating that the rapid expansion of coquies on Hawai'i Island was human‐facilitated, while high levels of inbreeding and significant genetic structure suggest low post‐establishment dispersal. If this is the case, early detection of coqui populations will be crucial for management due to their propensity to be spread through human‐facilitated jump dispersal, followed by slow rates of diffusion dispersal from these newly established populations.