In Search of Climate Effects on Atlantic Croaker Stock off the U.S. Atlantic Coast Using Bayesian State-Space Biomass Dynamics Models

Bayesian state-space biomass dynamics models (BSSBDMs) without the minimum winter estuarine temperature (MWETE) from the Chesapeake Bay region (Model 1, M1) and with MWETE (Model 2, M2) were used to investigate whether these models capture MWETE effects on the Atlantic croaker (Micropogonias undulatus) population dynamics off the U.S. Atlantic coast. MWETE was introduced into the intrinsic rate of population increase, r, assuming it controlled croaker’s productivity via growth, recruitment, or both processes during the pre-recruit stage. M1, M2 and related sensitivity configurations (i.e., using shrimp trawl fishery bycatch or an alternative prior probability density function for r) lacked fishing effort data. They only relied on fisheries removals (1972-2008) and fall biomass indices developed from the NMFS Northeast Fisheries Science Center (NMFS-NEFSC) Multispecies Trawl Survey (1972-2008) and the Southeast Area Monitoring and Assessment Program (SEAMAP) Coastal Trawl Survey (1990-2008). All BSSBDMs fitted indices equally and yielded similar results. Biomass residuals from base and sensitivity M1 showed slightly positive but significant correlations with MWETE. This suggested that warmer winters would enhance croaker biomass production. Model goodness-of-fit and comparisons indicated that base and sensitivity M2 were not supported by the data. If biomass dynamics models were envisaged for croaker stock assessments, it may be reasonable to perform them without MWETE effects owing to the previous inconclusive, somewhat conflicting results. All BSSBDMs pointed to a croaker stock that was not overfished in the 2000s.