9484 - Laserfiche WebLink

Home Browse Search

1724 D. B. OSMUNDSON ET AL. Ecological Applications Vol. 12, No. 6 the same dimensions, covered with water, and the water volume measured. To derive the percentage of substrate G2 mm, five interstitial substrate samples (500 mL) were collected near the Hess samples with a 3 cm di- ameter, clear, PVC, core tube; oven dried in the lab at 68°C; and sieve separated into size fractions. Mid-col- umn water velocity was measured at each Hess sample location with a Model 201 portable water current meter (Marsh McBirney, Gaithersburg, Maryland, USA). To derive median substrate particle size (total Dso and Dso of particles >!2 mm), substrate size distributions were quantified from pebble counts (Wolman 1954); one count was made parallel to shore at each run and riffle sampling site in the vicinity of the Hess sampling de- scribed above. Mapping was used to quantify the surface area, with- in each reach, of seven major mesohabitat types (riffles, runs, shoals, backwaters, low velocity, slack water, and vegetated), with two to six possible subtypes. This was done once during base flows of fall 1996. Habitats were drawn in the field on. hard copies of airborne videog- raphy, orthocorrected, and transferred to a GIS base map for interpretation. Additional parameters were measured during a syn- optic survey that immediately preceded each sampling period, and included light extinction, dissolved or sus- pended nutrients, and turbidity. This survey was con- ducted in one day and data were collected at one site within each sample stratum. Light was measured at multiple depths with a model LI-193SA spherical quan- tum sensor (LICOR, Lincoln, Nebraska, USA) that col- lected scattered as well as surface light. Water samples were collected from midchannel at a depth of - 15 cm. One 1-L water sample was split into two parts: one part was filtered in the field with a 0.45-µm mesh type HA Millipore filter for ammonia, nitrate, nitrite, and orthophosphorous analyses; the other part remained as a whole-water sample for total nitrogen and total phos- phorous analyses. All water samples were acidified with concentrated sulfuric acid. Time between collec- tion and laboratory analysis varied but was always within holding periods specified in Standard Methods (American Public Health Association et al. 1992). All nutrients were analyzed in the laboratory with a Spec- tronic 301 (Milton Roy Company, Rochester, New York, USA), utilizing a 5-cm cell. Methodologies for nutrient analyses followed Standard Methods. Turbid- ity was measured in situ (midchannel) with a Hydrolab Surveyor 3 (Hydrolab, Austin, Texas, USA). Main- channel temperatures were monitored year-round at seven sites within the study area (strata 1, 6, 7, 8, 9, 10, and 11) as part of another study and methods and results were previously reported by Osmundson et al. (1998). Biological parameters Biomasses of periphyton, benthic macroinverte- brates, and benthic detritus were estimated at each riffle and run sample location. Benthic macroinvertebrates were collected at the same five Hess-sample sites de- scribed in the preceding section. All substrate particles above the level of embeddedness were dislodged or rubbed by hand to loosen macroinvertebrates as par- ticles were removed from the Hess sampler for the interstitial void volume analysis described above. In the laboratory, formalin-preserved samples were sorted and dry mass of invertebrates was estimated from dis- placement volumes using family-specific regression equations developed for this study. Detritus, or coarse particulate organic matter (CPOM), collected with the invertebrate samples was oven dried at 68°C and weighed with an analytical balance. Chlorophyll a was used as a relative index to live periphyton biomass (Steemann Nielsen and Jorgensen 1962); this was because periphyton samples contained dead tissue, detritus, and deposited silt particles that could not be easily separated. Five cobble-sized rocks adjacent to each Hess sample were selected to provide periphyton samples. Periphyton scraped from a 2.5 cm diameter circle (5 cm2) scribed on each rock was cov- ered in tin foil to exclude light, frozen in the field with dry ice, and stored frozen until analyzed in the lab. Chlorophyll a concentrations were measured with a Model 111 fluorometer (Turner Associates, Palo Alto, California, USA) within 30 d of collection. Electrofishing catch rates were used as an index of relative abundance of main-channel fish species. We assumed these rates were proportionally related to the total fish biomass within a study reach. Only fish ?100 mm in total length (TL) were targeted under the as- sumption that pikeminnow ?550 mm TL require forage items of at least this size (Osmundson et al. 1998). Within each study reach, both shorelines were elec- trofished in a downstream direction with a 5 m long, hard-bottomed, electrofishing boat. In reaches contain- ing rapids, a 5-m rubber raft outfitted for electrofishing was sometimes used. Each craft was equipped with a Coffelt VVP-15 (Coffelt Manufacturing, Flagstaff, Ar- izona, USA) that produced pulsed DC. In strata 7-11, where fish were abundant, two people were required to dipnet stunned fish from the bow of the boat; in strata 1-6, where fish were few, only one person was needed to net all fish. Each shoreline within each reach was treated as a separate sample, resulting in six samples per stratum. Netted fish were transferred to one of two live wells on the boat and held until a shoreline sample was com- pleted: fishes from run habitats went in one and fishes from riffles went in the other. Elapsed shocking time (s) through each habitat type (counted on the VVP me- ter) was recorded. Fishes were identified, measured for TL (to the nearest 1 mm), weighed with an electronic balance (to the nearest gram), and released. Adult Colorado pikeminnow densities per stratum were derived from mean annual population estimates (see Osmundson and Burnham 1998 for methods), the