Relating fish recruitment to stock biomass and physical environment in the Black Sea using generalized additive models

Relationships between recruitment, parental stock biomass and physical environment variables in four Black Sea species: sprat Sprattus sprattus, whiting Merlangius merlangus, anchovy Engraulis encrasicolus and horse mackerel Trachurus mediterraneus, were investigated using generalized additive models (GAMs). Sea surface temperature, wind speed, wind stress, wind mixing, sea level atmospheric pressure and river run-off data were used as indices of interannual variability of the Black Sea environment. The environmental variables were studied by means of the principal component analysis, which tended to account for the major patterns in the time series and approximate the main environmental forcing agents in the sea. Recruitment responses to the considered environmental predictors in the four cases were compared and possible interpretations were discussed. Significant correlations appear between fish recruitment, stock biomass and physical environment. Patterns of the recruitment response to wind variables and sea level atmospheric pressure were found similar in the four species. The relationships indicated similarities between anchovy and whiting, whose reproduction is related to the coastal habitat on the one hand, and between sprat and horse mackerel, which are open sea spawners on the other hand. Recruitment in the latter species appeared to be less dependant on the parental stock biomass and the river discharge and was negatively related with SST. GAM was found to be a suitable tool for fisheries and environmental data analysis, providing a flexible and powerful way to explore and model non-linear relationships. (C) 1999 Elsevier Science B.V. All rights reserved.