MEPS 353:147-156 (2008)  -  DOI: https://doi.org/10.3354/meps07184

ABSTRACT: Recognizing the relationships between oceanographic variables and population descriptors has been a major step in understanding geographical range limits. In the present study, we investigate which oceanographic features are responsible for the decrease in fitness and consequent northern limit of the scallop Psychrochlamys patagonica distribution in Uruguayan waters of the SW Atlantic Ocean (SAO), based on information compiled from fishing and oceanographic surveys. P. patagonica showed clear population responses to physical variables, including a marked decrease in abundance, individual size and muscle weight towards the northern edge of the range. A generalized additive model (GAM) significantly explained 78.7% of the deviance in scallop abundance and retained all 4 descriptors in the model as significant, showing a decrease in abundance at lower latitudes with higher temperature and at lower values of chl a and water depth. The performance of GAMs was compared with that of the equivalent generalized linear models (GLMs) and it was concluded that the extended flexibility offered by GAMs resulted in better overall fits. Spatial variations in scallop abundance were accurately predicted by the GAM, including the 2 peaks at latitudes 36°20’S and 36°50’S. These spatial variations were closely related to the shelf break front around the 100 m isobath, where abundance was higher. The northern range limit of the species in the SAO may be caused by a sharp decrease in habitat quality, indicated by temperatures higher than 9°C (above species tolerance limit) associated with the influence of the Subtropical Water. Whereas physiological tolerance to temperature appears to be an important mechanism driving the range boundary of P. patagonica, food availability (reflected by the highest concentrations in chl a) is a key explanatory factor of the spatial variations in scallop abundance.

KEY WORDS: Psychrochlamys patagonica · Species borders · Environmental constraints · Generalized additive model · GAM · Southwestern Atlantic · Uruguay