6042 - Laserfiche WebLink

Home Browse Search

7 Near-shore drift collections were made using 0.5-m diameter conical plankton nets (Wildlife Supply Company, Saginaw, Michigan) mounted on 0.5- x 0.3-m rectangular steel frames and fitted with 33-cm long (lO-cm diameter) removable PVC collection buckets with a threaded cod end. Each net had a 560-micron Nytex nylon mesh length of 4.0 m, and an open mesh to net mouth ratio of 11:1. Filtration efficiency approaches 100% when the open mesh area is more than three times the area of the net mouth (Faber 1968; Tranter and Smith 1968). In the Colorado River, a removable four-point, steel-cable bridle assembly terminating in a spring-loaded carabiner was attached to each net frame. Nets were deployed either by staking the net frames to the river substrate in shallow areas or fastening the bridle carabiner to a polypropylene line fixed to either an instream boulder or a metal post driven into the shore at deeper sites. Mid-channel sets were attempted but discontinued for safety reasons and because of possible boat traffic during collection times. In the Yampa River, drift-nets were deployed off rafts tied to shore during runoff periods when depths and currents made wading difficult. During lower water periods, nets were attached directly to metal posts driven into the river substrate. At all sets, a safety line was attached to each net frame and to a post driven into the shore. Three drift-nets were deployed along the shoreline at each site just below the water surface. Water volume passing through each net was calculated from velocity readings made with either a Marsh-McBirney (Model 201) or pygmy (Gurley Model 625 F) current meter. Water temperature (C) was measured at each sampling time. In order to evaluate possible diel periodicity in larval fish drift, samples were taken at sunrise, noon, one-half hour after sunset, and at midnight. Sampling duration at each period ranged from 1 to 2 hours depending upon the suspended solids load. The contents of each net were rinsed into l-gallon plastic jugs, preserved in 10% buffered formalin, and returned to the Larval Fish Laboratory for processing. All specimens were counted, measured to the nearest 0.1 mm (TL), and assigned to a developmental phase according to Snyder (1976). This procedure permitted evaluation of larval fish drift periodicity both within a particular developmental phase (e.g., protolarval) and between phases (e.g., protolarval vs mes~larval). Numerical densities were computed as numbers of larvae per 1,000 m. Two-way analysis of variance (2-ANOVA) was used to compare drift densities (response variable) among the four sampling times and between day and night samples (i.e., combined sunrise and noon = day; combined sunset and midnight = night) on individual sampling dates over the entire sampling season. The 0.05 probability level was used to determine significance in F values. Flow and temperature data were obtained from United States Geological Survey (USGS) readings when available and supplemented with measurements taken during sampling and CRFP thermograph data. All field and laboratory data were recorded to be compatible with data collected by NW Region (CDOW) and CRFP personnel with any differences resulting from specific requirements of early life history studies. Data were generally compatible with the United States Fish and Wildlife Service MANAGE database program. All field data has been filed on dBase II and stored at CDOW, Denver. To estimate spawning dates using back-calculated age of seine- and drift-net-collected young-of-the-year (YOy) Colorado squawfish, two predictive age equations (Haynes and Muth 1984) were derived using Hamman's (1981) total length at known age data for hatchery cultured and reared Colorado squawfish larvae (Fig. 2).