Species and habitats - Raja clavata - Mature - Potential habitat in July modelled by Quantile Regression and its uncertainty with BTS data

Date(s) (Publication)
Custodian(s): CHARM Consortium
Point of contact(s): Steve Mackinson
Affiliation(s) 1 : Center for Environment, Fisheries & Aquaculture Science
2 : CHARM Consortium
Credit CEFAS
Abstract Modelised abundance of species or prediction uncertainty.
Keywords Species data set, CHARM, Habitats et biotopes, ressource halieutique
Lineage In short, models based on GLMs predict the mean response of the species to environmental factors whilst models based on RQ predict the maximal response. When GLM uses abundance data, the preferential habitat is predicted, whilst the probable habitat is predicted when GLM uses binary presence-absence data. Generalised Linear Modelling (GLM) describes and predicts the "preferential habitat", i.e. the portion of the potential habitat that is used on average over time, or, in the case of
Utilisation research-only