Research report: Migration dynamics of acoustic-tagged hatchery and natural atlantic salmon smolts

Acoustic tags attached to hundreds of young Atlantic salmon have revealed a surprise in the waters of Maine: hatchery-raised smolts survived their journey to the sea at higher rates than their river-reared counterparts, a result that runs counter to much of the published literature on hatchery versus wild fish. The finding comes from a three-year telemetry study in the Penobscot River Estuary and Bay, where researchers tracked migrating smolts of both origins across roughly 100 kilometers of river, estuary, and coastal waters during the springs of 2010 through 2012.

Natural smolts were collected at the downstream bypass of Weldon dam, the first mainstem dam the fish encounter during emigration, while hatchery smolts came from the USFWS Green Lake National Fish Hatchery, which produces age-1 smolts through an accelerated growth program.

Behavioral differences between the two origin groups proved inconsistent across years. Movement rates in the tidal river did not differ significantly between hatchery and natural smolts in 2010 or 2011, but in 2012 natural smolts moved significantly faster, at 2.6 kilometers per hour compared with 1.6 kilometers per hour for hatchery fish. Rates increased from the tidal river to the estuary for both groups, and within the bay postsmolt movement slowed with seaward progression, with the only significant difference again in 2012, when natural fish were faster. Overall migration duration through the entire array showed no significant difference between origins in 2010 or 2011, and in 2012 hatchery smolts took significantly longer, at 5.8 days versus 4.2 days for natural smolts. Across all years combined, the median emigration duration was 4.2 days for hatchery smolts and 4.0 days for natural smolts, a difference that was not statistically significant.

Path choice was similarly consistent between groups. In the middle bay, more than half of each origin group used the eastern middle bay channel, and nearly 90 percent of both hatchery and natural postsmolts traveled through the western outer bay on their way to the Gulf of Maine. No postsmolts were detected at the northeastern outer bay exit point during the study. On the western outer bay curtain array, significantly more hatchery postsmolts were detected on the western side, while natural postsmolts showed a similar but statistically nonsignificant pattern. The study’s authors note that this corridor information can help managers make informed decisions about infrastructure and renewable resource projects within Penobscot Bay.

Survival, however, told a different story. Using a Cormack-Jolly-Seber mark-recapture model, the researchers found that rearing origin and river discharge were the primary factors influencing smolt survival. The best-fitting model included an interactive effect of reach, origin, discharge, and year, and received overwhelming statistical support over the next-best model. Cumulative survival for hatchery smolts ranged from 0.40 to 0.69 across the study years, while natural smolts ranged from 0.32 to 0.64, with hatchery fish showing greater cumulative survival each year. A likelihood ratio test confirmed that origin explained significant variation in apparent survival after accounting for time, year, and discharge.

Discharge emerged as a key environmental covariate with a negative effect on survival, meaning that within a given year and reach context, higher river flow was associated with lower survival. The study period spanned dramatically different water conditions: in 2010, discharge hovered under 400 cubic meters per second for much of the study, while 2011 featured the highest overall volume with peaks reaching 1,600 cubic meters per second, and 2012 showed an intermediate pattern with a single prominent peak near 1,000 cubic meters per second. The authors caution that because discharge was assigned at the fish-day level and was partially confounded with release date, the observed relationship may partly reflect differences among release cohorts and should not be interpreted as causal. Still, they suggest that relative flow conditions could help inform release timing, pending validation across additional dates and years.

Survival patterns varied by habitat. In the tidal river, hatchery smolts exhibited moderate-to-high per-kilometer survival of 0.97 to 0.99, while natural smolts showed 0.92 to 0.97. Within the estuary, survival was broad for both groups, with hatchery fish at 0.93 to 0.99 and natural fish at 0.87 to 0.99, and the lowest estuary values occurring just before the inner bay. Postsmolt survival declined with seaward progression through the bay: in the inner bay, hatchery fish ranged from 0.92 to 0.96 and natural fish from 0.98 to 0.99; in the middle bay, both groups ranged from roughly 0.76 to 0.90; and in the outer bay, hatchery fish ranged from 0.65 to 0.84 and natural fish from 0.60 to 0.81. Mortality notably increased at the river mouth, an area characterized by consistent salinities of at least 20 parts per thousand and significant temperature gradients, a pattern that aligns with findings from a nearby river system where rapid inundation with cold, high-salinity waters coincided with initial postsmolt losses.

The result that hatchery fish outperformed natural fish contrasts with many head-to-head evaluations of tagged hatchery and wild smolts, which generally indicate that natural-origin salmon with longer riverine residence perform as well as or better than their hatchery counterparts. The authors hypothesize that the collection method for natural smolts at the dam bypass facility, which involves novel handling and dam interaction, may have contributed to their lower survival and potentially masked any inherent advantage over hatchery smolts. During the three study years, 11.5 percent of natural smolts collected at the bypass trap were excluded because they were dead or moribund, and only fish without visible injury or irregular behavior were selected for surgery, so the survival estimates represent a minimum-impact scenario. Interactions with dams can result in both immediate and latent mortality, and nearly all natural Atlantic salmon smolts in the drainage must pass at least one and potentially up to eight dams before reaching the estuary.

Physical characteristics differed markedly between the groups. Hatchery smolts had a median fork length of 195.0 millimeters and a mass of 74.4 grams, while natural smolts measured a median 172.0 millimeters and 47.4 grams, differences that were significant each year. Hatchery smolts also carried a lower average transmitter burden of 3.9 percent of body weight compared with 6.1 percent for natural smolts. Smolt scores, a physiological proxy for readiness, were higher for hatchery fish, with 85 percent scoring 5 on a six-point scale, whereas roughly half of natural smolts scored 4. Higher physiological readiness has previously been shown to correlate with increased probability of successful estuary passage. Although smolt size has been linked with migration success in other studies, it was not a significant covariate in the best-fitting model here, and models incorporating rearing origin consistently fit better than those using fork length, suggesting that rearing origin is a stronger predictor of survival.

The findings carry weight for conservation of the Gulf of Maine Distinct Population Segment of Atlantic salmon, which is protected under the U.S. Endangered Species Act and represents the last remaining population of the species in the United States. Adult returns to Maine remain static at 1,000 to 1,500 fish annually despite hatchery supplementation, and the smolt-to-adult return rate for multiple sea-winter adults has been below 0.5 percent over the past two decades. The authors conclude that managers need to balance two approaches: increasing the number of smolts reaching the ocean through hatchery supplementation, and continuing to improve passage and mitigation measures to raise the survival of natural smolts. Improved dam passage, they argue, is critical for increasing overall restoration efforts, and improvements would justify reseeding vacant habitats in the upper Penobscot River, which could produce approximately one million additional natural smolts.

Subject of Research: Migration dynamics and survival of acoustic-tagged hatchery and natural Atlantic salmon smolts in the Penobscot River Estuary and Bay

Article Title: Migration dynamics of acoustic-tagged hatchery and natural atlantic salmon smolts

Article References: Hawkes, J. P., Goulette, G. S., & Molina-Moctezuma, A. (2026). Migration dynamics of acoustic-tagged hatchery and natural atlantic salmon smolts. BMC Zoology, 11(1), Article 34. https://doi.org/10.1186/s40850-026-00278-y

Image Credits: AI Generated

DOI: 10.1186/s40850-026-00278-y

Keywords: Atlantic salmon, smolt migration, acoustic telemetry, hatchery supplementation, Penobscot River, mark-recapture survival, river discharge, dam passage, Gulf of Maine, fish conservation, estuary ecology, postsmolt survival