Relative importance of fishing and natural mortality for spotted seatrout (Cynoscion nebulosus) estimated from a tag-return model and corroborated with survey data

The spotted seatrout (Cynoscion nebulosus) is one of the most economically important sportfish in the U.S. South Atlantic and Gulf of Mexico, including at its northern distributional extent in North Carolina and Virginia. The recent stock assessment for this region used an assumed fixed rate of natural mortality (M), obtained from a general life-history relationship based on weight. However, biased estimates of fishing mortality (F) could result if the life-history proxy failed to capture either the magnitude or temporal variation in M. Data from the first comprehensive tag-return study of spotted seatrout in this region were used in a Bayesian statistical modeling framework to estimate F and M. Both laboratory and field studies, including high-reward and double tagging, were conducted to obtain estimates of auxiliary parameters (i.e., tag-reporting rate, tag loss, and tagging mortality) necessary for the tag-return model. There was no measured mortality associated with tagging, but reporting rate and loss of internal anchor tags limited returns in this study. From 2008 to 2012, tag-return model estimates of bimonthly instantaneous mortality rates ranged from 0.003 to 0.067 2-mo(-1) for F and from 0.002 to 2.850 2-mo(-1) for M. Annual estimates of F were much lower than M for the three years studied, and annual M estimates were higher than those used for spotted seatrout in this region's recent stock assessment. Bimonthly estimates of total mortality rate (2) from tag-return data were similar to bimonthly estimates of Z from an independent analysis of concurrent gill net survey data, which corroborates the variability and magnitude of mortality estimates determined from tagging. A strong seasonal influence (i.e., winter severity) on annual loss of spotted seatrout was observed, suggesting that future assessments and management measures for this stock would be improved by explicitly accounting for temporal variation in M in models of fishery population dynamics.