Disruption of the lower food web in Lake Ontario: Did it affect alewife growth or condition?

Abstract

From the early 1980s to the late 1990s, a succession of non-native invertebrates colonized Lake Ontario and the suite of consequences caused by their colonization became known as "food web disruption". For example, the native burrowing amphipod Diporeia spp., a key link in the profundal food web, declined to near absence, exotic predaceous cladocerans with long spines proliferated, altering the zooplankton community, and depth distributions of fishes shifted. These changes had the potential to affect growth and condition of planktivorous alewife Alosa pseudoharengus, the most abundant fish in the lake. To determine if food web disruption affected alewife, we used change-point analysis to examine alewife growth and adult alewife condition during 1976-2006 and analysis-of-variance to determine if values between change points differed significantly. There were no change points in growth during the first year of life. Of three change points in growth during the second year of life, one coincided with the shift in springtime distribution of alewife to deeper water but it was not associated with a significant change in growth. After the second year of life, no change points in growth were evident, although growth in the third year of life spiked in those years when Bythotrephes, the largest of the exotic cladocerans, was abundant suggesting that it was a profitable prey item for age-2 fish. We detected two change points in condition of adult alewife in fall, but the first occurred in 1981, well before disruption began. A second change point occurred in 2003, well after disruption began. After the springtime distribution of alewife shifted deeper during 1992-1994, growth in the first two years of life became more variable, and growth in years of life two and older became correlated (P < 0.05). In conclusion, food web disruption had no negative affect on growth and condition of alewife in Lake Ontario although it appears to have resulted in growth in the first two years of life becoming more variable, growth in years of life two and older becoming correlated (P < 0.05), and growth spurts in year of life three. Copyright ?? 2008 AEHMS.