Demography and movements of spawning American shad (Alosa sapidissima) in the Penobscot River, Maine, prior to dam removal

The Penobscot River is currently the subject of an intensive river restoration effort. Under the Penobscot River Restoration Project (PRRP), the two lowest dams in the system, Veazie Dam (rkm 48) and Great Works Dam (rkm 60), will be removed. Great Works Dam was dismantled in the summer of 2012, and Veazie Dam is scheduled for demolition by 2014. This study examined the migratory movements, and age and spawning structure of American shad (Alosa sapidissima) in the Penobscot River prior to dam removal. Although shad were historically abundant in the system, little is known about the current-day population which was presumed to be small. This decline is attributed to a lack of migratory connectivity and accessible habitat; American shad do not pass the Veazie fishway, and freshwater spawning habitat (15 rkm) and brackish rearing habitat (~50 rkm) are limited below the dam. It is anticipated that the Penobscot River shad will benefit from restored connectivity and access to upriver freshwater habitats. In the spring of 2011, a Dual-Frequency Identification Sonar (DIDSON) was used to record footage of fish approaching the entrance of the Veazie Dam fishway. Fork lengths (FLs) from high-quality images were measured, and the resulting FL distributions were compared against known length distributions of river herring, American shad, and Atlantic salmon using a Bayesian mixture model. The model classified over 76 % of fish observations as American shad, and attributed 16 % to Atlantic salmon and 8 % to river herring. These results indicate shad presence at the base of dam. However, because the imaged fish were not uniquely identifiable and may have been imaged repeatedly, abundance at the base of the dam cannot be inferred from these proportions. A combination of radio and acoustic telemetry was used to investigate the movements of migratory American shad in the spring of 2010 and 2011. Radio telemetry results indicate that few tagged shad (5 - 8%) approached the Veazie Dam, but that those who did investigated the dam on at least 2 days. Tagged fish exhibited 3 main movement patterns in freshwater: using the upper end of the radio array above rkm 43, using the lower end of the array near rkm 34, or using the entire array. Mean freshwater residence time in freshwater ranged from 9.1 to 14.0 days. “Hotspots” where shad congregated included Eddington Bend (rkm 47) and the Bangor Dam headpond (rkm 44), and spawning activity was observed at the latter site. Freshwater survival and survival to the estuary were at least 71%. The high survival rate was confirmed the ageing and spawning histories obtained from shad scales, which indicated that 75 to 95 % of the sampled fish were repeat spawners, and that adult migrants ranged from age 4 to 9. Acoustic tagged fish exhibited a series of upstream and downstream reversals upon entering the lower estuary. These movements were previously unreported for American shad, and may be related to osmoregulatory acclimatization for re-entry into saltwater, a resumption of post-spawn feeding activity or both