Thomson J.L. Stritzel | USGS Science

USGS2010

Growth rates of rainbow smelt in Lake Champlain: Effects of density and diet

We estimated the densities of rainbow smelt ( Osmerus mordax ) using hydroacoustics and obtained specimens for diet analysis and groundtruthed acoustics data from mid‐water trawl sampling in four areas of Lake Champlain, USA–Canada. Densities of rainbow smelt cohorts alternated during the 2‐year study; age‐0 rainbow smelt were very abundant in 2001 (up to 6 fish per m 2 ) and age‐1 and older were abundant (up to 1.2 fish per m 2 ) in 2002. Growth rates and densities varied among areas and years. We used model selection on eight area–year‐specific variables to investigate biologically plausible predictors of rainbow smelt growth rates. The best supported model of growth rates of age‐0 smelt indicated a negative relationship with age‐0 density, likely associated with intraspecific competition for zooplankton. The next best‐fit model had age‐1 density as a predictor of age‐0 growth. The best supported models ( N = 4) of growth rates of age‐1 fish indicated a positive relationship with availability of age‐0 smelt and resulting levels of cannibalism. Other plausible models were contained variants of these parameters. Cannibalistic rainbow smelt consumed younger conspecifics that were up to 53% of their length. Prediction of population dynamics for rainbow smelt requires an understanding of the relationship between density and growth as age‐0 fish outgrow their main predators (adult smelt) by autumn in years with fast growth rates, but not in years with slow growth rates.

New York, Vermont↗

USGS2006

Hydroacoustic separation of rainbow smelt (Osmerus mordax) age groups in Lake Champlain

Separate assessment of young-of-year (YOY) and yearling-and-older (YAO) fish is desirable from both ecological and management perspectives. Acoustic assessments provide information on fish population size structure in the target strength (TS) distribution, but interpretation of TS distributions must be done carefully, as single age groups can produce multiple TS modes. We assessed the ability of in situ TS distributions to identify Lake Champlain rainbow smelt (Osmerus mordax) age groups in June, July, and September of 2001 using mobile and stationary surveys, knowledge of vertical distribution preferences, and predicted TS from trawl catches. YAO rainbow smelt (93-179 mm total length) had wide TS distributions between -60 and -35 dB in all 3 months with two modes at approximately -50 and -40 dB. Most stationary survey single-fish tracks attributed to YAO had targets in both TS modes and a wide TS range often over 15 dB. Between June and September, YOY rainbow smelt TS increased, but single-fish tracks were unimodal, and the TS range was smaller (6 dB). Overlap in TS attributed to YOY and YAO increased from no overlap in June (YOY TS -76 to -61 dB, 15-25 mm) to moderate overlap in July (-76 to -50 dB, 25-63 mm) to considerable overlap in September (-68 to -45 dB, 33-80 mm). In June and July, the TS distribution changed abruptly at the thermocline, indicating almost complete separation of the two groups. A more gradual TS transition was evident in September, indicating substantial overlap between YOY and YAO. Separate estimates can be obtained in September by decomposing TS overlap into components attributed to YOY and YAO rainbow smelt. However, this decomposition introduces additional uncertainty and an assessment in July or possibly August is preferable to obtain separate abundance estimates of YOY and YAO. ?? 2006 Elsevier B.V. All rights reserved.

Fisheries Research↗