Seasonal migrations of the American eel, Anguilla rostrata, in the Upper Salmon River, New Brunswick

Abstract

Stable isotope analysis of American eels (Anguilla rostrata) captured in the Upper Salmon River, New Brunswick, revealed isotopic patterns suggesting that yellow-stage eels may be migrating between saline summer foraging grounds and freshwater overwintering habitat. We examined the seasonal movements of yellow eels in the Upper Salmon River using passive integrated transponder (PIT) and radio telemetry. Comparisons of isotopic signature and seasonal movements of 288 individual eels captured in fresh water in the spring of 2009 and 2010 showed that eels with enriched vs. depleted isotopic ratios of carbon (13C/12C) and nitrogen (15N/14N), were more likely to migrate to the estuary in spring following release where they remained before exhibiting the reverse migration in fall; confirming that seasonal amphidromous migrations were driving the observed isotopic variation. Summer habitat residency in the estuary was confirmed in 2009 and 2010 using a manual PIT detection survey which showed that 36% and 34%, respectively, of spring-captured yellow eels PIT tagged and released in fresh water were identified resting in dense mats of sea lettuce (Ulva intestinalis). Freshwater overwintering of amphidromous eels was confirmed via radio tracking of select eels in the winter of 2009 and 2010. An examination of the environmental correlates of amphidromous eel migrations using count regression models determined that temperature and photoperiod were positive and negative correlates, respectively, of spring downstream migration. In addition, temperature and discharge were identified as negative and positive correlates, respectively, of fall upstream migration. Fall migration was also found to correlate significantly with the lunar cycle, with two peaks in migration occurring prior to the new and full moons. Almost all migration occurred during periods of darkness, centered about 3.3 and 4.0 h post-sunset in the springs of 2009 and 2010, and 5.9 h post sunset in the fall of 2009. In all cases, migration was centered about the midpoint of the dark period. Spring eel migrations peaked 3.7 and 6.3 h following high tide in 2009 and 2010, respectively. However, fall upstream migration was randomly distributed with respect to tide. Mark and recapture models estimated that 10,220 (97.5% CI: 6139-16,540) and 3022 (97.5% CI: 2158-5073) yellow eels >20 cm in length migrated downstream in spring 2009 and 2010, respectively. Fyke nets and rotary screw traps (RST) sampled contrasting components or size classes of the migrating eel population. An RST retention test determined that eels <20.6 cm in length were likely to escape the RST holding box. In contrast, fyke nets did not appear efficient in capturing larger eels (i.e., >20 cm in length). RST modifications are warranted for future monitoring programs.