Landmark Tagging Rework Suggests Atlantic Menhaden Die Less Naturally Than Managers Assumed

One of the most consequential numbers in East Coast fisheries management may have been wrong for years. Atlantic menhaden, the small, oily forage fish that supports one of the largest commercial fisheries on the U.S. Atlantic coast while simultaneously feeding striped bass, bluefish, seabirds, whales, and other predators, may be considerably less resilient to harvesting than the official stock assessments have assumed. That is the central conclusion of a new peer-reviewed analysis published on September 16, 2026 in the journal Fisheries Research, which finds that a key estimate of natural mortality, the rate at which fish die from causes other than fishing, was substantially overstated in the assessments currently used to set harvest policy.

The study was led by Jerald S. Ault, professor emeritus at the University of Miami Rosenstiel School of Marine, Atmospheric, and Earth Science, together with co-author Jiangang Luo, a scientist at the Rosenstiel School’s Cooperative Institute for Marine and Atmospheric Studies. Rather than collecting new field data, the pair returned to the raw records of a landmark federal tagging experiment conducted from 1966 through 1971, an enormous undertaking that spanned the Atlantic coast from Massachusetts to Florida and involved more than one million tagged menhaden and more than 100,000 recovered tags. By reconstructing that historic dataset from the ground up, the researchers uncovered discrepancies that, once corrected, fundamentally changed the mortality estimate that flows through modern management models.

Natural mortality is one of the most influential and least directly observable inputs in any fish stock assessment. It represents deaths from predation, disease, old age, and environmental stress, and it is typically estimated rather than measured. When that rate is set too high, the mathematics of assessment models effectively attribute a larger share of total deaths to nature and a smaller share to the fishing fleet. The result is an optimistic picture: the spawning stock appears larger and more productive than it really is, and the models generate harvest advice that the population’s underlying biology may not be able to support. For a forage species sitting at the base of the coastal food web, that optimism ripples outward to every predator that depends on it.

The stakes are illustrated by the numbers themselves. The new analysis estimates natural mortality for Atlantic menhaden at approximately 0.50 per year, meaning that in an unfished population roughly half of the fish would be expected to die each year from natural causes. That figure is less than half of the 1.17 estimate reported in a 2019 reanalysis by Liljestrand and colleagues, which was incorporated into the SEDAR 69 benchmark stock assessment used by the Atlantic States Marine Fisheries Commission. It is also substantially below the corrected value of 0.92 adopted in a 2025 stock assessment update. In practical terms, the difference between these values determines how much of the population’s decline managers are willing to blame on fishing rather than on nature.

Reconstructing the historic tagging experiment was a forensic exercise. Over approximately three years, Ault and Luo rebuilt two independent versions of the historical dataset, cross-checked the counts of fish released and tags recovered, reconstructed the fishing effort records for the era, and examined in detail how metal tags were actually detected when harvested fish arrived at processing plants. The investigation identified discrepancies involving the numbers of released and recovered tagged fish, the reconstruction of historical fishing effort, the treatment of tag-detection efficiency at the plants, and aspects of the statistical model used to interpret the recovery data. Each of these issues, the authors found, had the potential to bias the mortality estimate, but one proved decisive.

That decisive factor was the performance of the magnets used to recover the metal tags embedded in harvested menhaden. In the original experiment, tagged fish were detected as they moved through reduction plants along the coast, where conveyor-belt magnets were supposed to pull tagged carcasses out of the processing stream. Detection efficiency, however, varied widely from plant to plant and from month to month, depending on equipment, throughput, and operating conditions. Earlier analyses that assumed a uniform detection rate across this heterogeneous system would have systematically miscounted how many tagged fish were actually recovered, distorting the apparent survival of the tagged population and inflating the estimated rate of natural death.

When the researchers built statistical models that allowed tag-detection efficiency to vary by location and by month, the fit to the recovery records improved markedly. Both of the independently reconstructed datasets produced natural mortality estimates near 0.50 per year, a striking convergence given that the two reconstructions were built separately from the original federal records. The team then ran additional computer simulations to test how sensitive the results were to uncertainty in the underlying assumptions, and the best-fitting outcomes consistently supported a similar value. The estimate also aligned with earlier research and biological evidence indicating that Atlantic menhaden historically lived at least ten years, a longevity that is difficult to reconcile with a very high natural mortality rate, since fish dying at a rate of more than one per year would rarely survive that long.

The authors are careful to frame what the finding does and does not mean. The study concerns a historical estimate of natural mortality, not a newly observed change in the survival of today’s menhaden stocks. It does not establish a new catch limit, and it does not claim that recent population trends have been misreported. What it does argue is that the input value used in the assessment framework should change. Ault and Luo recommend that future stock assessments evaluate natural mortality values centered at 0.50 per year while explicitly accounting for the uncertainty surrounding that estimate, rather than anchoring the models to the substantially higher rates adopted in the 2019 reanalysis and the 2025 update.

The management implications are potentially far-reaching. If natural mortality is high, fishing appears to account for only a small share of population change, and harvest limits can be set generously without much apparent risk. If natural mortality is closer to 0.5, as this study and earlier historical estimates suggest, fishing may account for a considerably larger share of the losses, and the same level of harvest represents a heavier burden on the stock. For Atlantic menhaden, a species that is both heavily harvested by one of the region’s largest reduction fisheries and ecologically central as prey for managed predators such as striped bass, that distinction matters not only to the menhaden fishery itself but to the entire chain of coastal ecosystems and fisheries that depend on abundant menhaden schools.

The study, titled Investigation of Atlantic menhaden mortality rates, demonstrates how revisiting the raw data behind decades-old experiments can still reshape contemporary science policy. Tagging studies from the 1960s were conducted with different technology, different quality controls, and different statistical tools than those available today, and the assumptions baked into their original analyses can persist unexamined through successive rounds of assessment. By showing that a single overlooked source of measurement error, the uneven performance of recovery magnets at processing plants, could shift a pivotal mortality estimate by more than half, the work underscores a broader lesson for fisheries science: the numbers that guide harvest decisions are only as reliable as the historical records and statistical assumptions beneath them, and those foundations deserve periodic re-examination, especially when the species in question anchors an entire coastal food web.

Subject of Research: Reanalysis of historical tagging data to re-estimate natural mortality rates of Atlantic menhaden for fisheries management

Article Title: Study challenges estimates used to guide Atlantic menhaden fishing

Article References: Study challenges estimates used to guide Atlantic menhaden fishing. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: Atlantic menhaden, natural mortality, fisheries management, stock assessment, tagging study, forage fish, Fisheries Research, Atlantic States Marine Fisheries Commission, SEDAR 69, tag detection efficiency, coastal food web, University of Miami Rosenstiel School