Browsing Benthos Ecology by Subject "Logistic regression"
Now showing items 1-1 of 1
Environmental factors associated with invasion: modeling occurrence data from a coordinated sampling programme for Pacific oystersDocumenting establishment and spread of invasive species requires extensive co-ordinated sampling programmes. Identifying the factors promoting or inhibiting local establishment of an invasive species can improve capacity to predict further spread and underpin strategies to limit spread. Here, a structured sampling programme was used to assess the current distribution of feral populations of Pacific oysters, Crassostrea gigas, in Ireland. Sixty-nine sites were sampled using a standardised protocol combining semi-quantitative and quantitative approaches. Sites were chosen to represent variation in proximity to aquaculture and a range of environmental variables. Oyster populations were found at 18 locations, with densities ranging from single individuals to nine individuals per m2. The broad size range of oysters found is indicative of more than one recruitment event. Logistic regression indicated that feral oysters were positively associated with the presence of hard substrata or biogenic reef, long residence times of embayments and large intertidal areas. There was also a tendency for oysters to occur disproportionately in bays with aquaculture, but >500 m from it. Small-scale analysis within sites showed that oysters were almost exclusively attached to hard substrata and mussel shell. The approach taken here provides a rigorous repeatable methodology for future monitoring and a detailed basis for the prediction of further spread.