Modelling spatial variation in an exploited species across marine reserve boundaries

An increase in exploited species within marine reserves is often expected to benefit adjacent fished areas through the movement or ‘spill-over’ of across reserve boundaries. However, harvest at and near boundaries can also decrease abundances inside reserves, forming ‘edge-effects’. Spatial analysis gradients abundance have been used interpret extents edge-effects spill-over. While shape these will vary depending on mobility fishing intensity boundaries, they largely assumed be gradual quantified using continuous modelling strategies. This study shows that strategies lead overestimations spill-over by underfitting rapid transition occur from areas. The spiny lobster (Jasus edwardsii) two north-eastern New Zealand were catch survey data collected over 2 years. Several approaches used, predictive performance each approach was compared. Abrupt changes found best represented with models incorporated a stepwise function where modelled independently. Despite mobile nature this species, we little evidence surveyed which may reflect low population status reserves. In contrast, standard overestimated extent both Policy implications. how accurate assessments evaluate effectiveness existing protecting species. These findings emphasise importance comparing several different when evaluating ensure appropriate model fits. Although such as generalised additive present simple solution exploring spatial variation ecological data, questioning applicability assessing change protected area