Application of a Bayesian Spatial Tag-Return Model to Historical Atlantic Menhaden Tagging Data

More than one million Atlantic menhaden were tagged from 1966−1970 by the National Marine Fisheries Service in one of the largest tag-return experiments ever conducted. Tagged fish were released in five areas of fishing activity. Fish were tagged internally with batch-marked ferromagnetic tags that were recovered by industrial magnets at menhaden factories where landings were processed. Early analyses of tag-return data indicated relatively low survival rates and a general northward migration pattern; however, a quantitative analysis that accounted for age structure and important model assumptions was never completed. We recently analyzed this historical dataset using a Bayesian spatial tag-return model to estimate mortality in each area of fishing activity and movement probabilities between regions. Corroborating early analyses, our spatial tag-return model estimates of instantaneous fishing mortality rates ranged from 0.87 in Florida to 1.59 in New York with higher probabilities of northward movement than southward movement in older fish. Significant spatial variations were estimated for both fishing mortality and for probabilities of movement within the Atlantic menhaden stock.  These estimates may inform spatially-structured assessments of the Atlantic menhaden stock in the near future.