Annual variability in upstream migration of glass eels in a southern USA coastal watershed

We investigated the environmental factors that affected temporal variability of eel recruitment and upstream migration in a freshwater coastal river along the southeastern US. Glass eels Anguilla rostrata were collected through ichthyoplankton sampling in the lower Roanoke River, North Carolina. Monthly samples were taken from fixed stations from May 2001 through June 2003. There was no evidence of consistent seasonal migration patterns for glass eels in Roanoke River. From May through December in 2001, glass eels were captured only during August. In 2002, glass eels arrived in February and remained in ichthyoplankton samples through October, with the exception of samples from September. Peak catch occurred in March at 4.02 ± 1.2 and declined through June to 0.18 ± 0.07 (#/1,000 m³). By August, the mean density increased to 0.96 ± 0.82 and to 3.59 ± 2.77 by October. In 2003 from January through June, glass eels were captured only during February and March. Glass eels were routinely collected when river discharge rates were <150 m³ s⁻¹. River discharge rates >650 m⁻³ s⁻¹ resulted in no glass eels in our samples. Upstream migration during 2002 was not correlated with water temperature or related to lunar phase. Glass eel freshwater upstream migration was initiated when water temperatures exceeded a threshold range of 10°C to 15°C; however, glass eels continued to migrate when water temperatures approached 30°C. The overall negative effect of river discharge suggests that changes in the water release schedules of upstream hydroelectric facilities during glass eel migration could strongly influence their recruitment success.