WSPC06762 - Laserfiche WebLink

Home Browse Search

OJ(73) determine whether groundwater was a source of nutri- ents to surface water (McMahon et ai" 1994), Analyses of nutrient concentrations followed methods described in Fishman (1993) and Patton & Truitt (1992), Dissolved oxygen and temperature, pH, and specific conductance were measured in situ in surface and hyporheic waters. Values of altitude, sinuosity, slope and stream order were derived from 1:24000 scale topographical maps (Meador et ai" 1993)_ Data analyses ., Ordination of sites by taxa was performed using detrended correspondence analyses (DCA) on natural log-transformed abundance data using the FORTRAN program DECORANA (Hill, 1979), Analyses included the sixty-six taxa that were collected at more than one site and composed greater than 05% of the total abundance of organisms (taxa with asterisks in Table 2). These criteria for the inclusion of taxa in the analyses were used to decrease the effect of rare species. Ordination by DCA arranges sites with similar taxonomic composition to cluster more closely together and produces site scores that can be related to environmental variables. From the DCA ordination plot, four site groups were identified: mountains, plains/braided channel, plains/tributary, and plains/ downstream from point source. Invertebrate taxa, functional feeding groups, and environmental vari- ables were further defined based on these four site groups, Univariate analyses of environmental variables, comparison of linear regression analyses of DCA axis 1 and 2 to individual environmental variables, and analyses of variance (ANOVA) among site groups were made using SAS statistical programs (SAS, 1990), Natural log or square-root transformations were per- formed on environmental variables to achieve approxi- mate normal distribution of the data, A Shannon diversity index was calculated using the dominant taxa (excluding rare taxa) for each site (Shannon & Weaver, 1963), Phi values were calculated for the dominant and subdominant substrate sizes and the mean substrate size (Folk, 1980), All inverteb- rate taxa including rare taxa were assigned to func- tional feeding groups using Merritt & Cummins (1984) for insects and Pennak (1978) for non-insects, @ 1995 Blackwell Science Ltd, Freshwater Biology, 33, 439-454 Invertebrates of the South Platte River 443 Results A total of 104 invertebrate taxa were collected from all twenty-one sites in the stream survey (Table 2), Invertebrate density and number of taxa at a site ranged from 100 to 20 300 organisms m -, and six to thirty-six taxa/site (Table 1), Invertebrate density in streams of the Southern Rockies ecoregion (mountains) ranged from 1980 to 18100 m -2 and streams of the Western High Plains ecoregion (plains) ranged from 100 to 20 300 m -', The mean number of taxa at a site was greater in the mountains (mean = 25; range = 10-36, n = 9) compared with the plains (mean = 14; range = 6-26, n = 12)(ANOVA, P < 0,005, Table 1), The relative magnitude of eigenvalues for each DCA axis is an expression of the relative importance of the axis, DCA axis 1 (eigenvalue = 0,44) and axis 2 (eigenvalue ~ 0,28) accounted for about 72% of the variance in the data set, whereas, OCA axis 3 (eigenvalue = 0,16) and DCA axis 4 (eigenvalue = 0,10) together accounted for < 30% of the variance and were not strongly correlated to any measured envirorunental variables. In contrast, regression of DCA axis 1 and 2 scores with environmental variables indicates significant relationship with seven physical and seven surface- and hyporheic- water chemistry variables for DCA axis 1 and two physical and four surface-water chemistry variables for DCA axis 2 (Table 3), Thus, DCA axis 1 separated sites in the mountains (Southern Rocky Mountain ecoregion) from sites in the plains (Western High Plains ecoregion) based primarily on stream slope, water temperature, specific conductance, and surface water organic nitrogen + ammonia and total phosphorus concentra- tions; each variable accounted for 48% or more of the variance in DCA axis 1 scores (P < 0,05, ,2 '" 0,48; Table 3), DCA axis 2 separated sites in the plains based on surface water nitrate + nitrite concentrations and channel width, which accounted for 36 and 32%, respectively, of the variance in DCA axis 2 scores (P < 0,05, ,2 > 0.30, Table 3), Four site groups were then identified from plotting DCA axis 1 and 2 scores: mountains, plains tributaries near their confluence with the South Platte River (plains/tributary), plains streams with braided channels (plains/braided channel), and plains sites affected by wastewater- treatment-plant effluent (plains/downstream from point source) (Fig, 2), The South Platte River above eleven-mile Canyon I ~ , r ! f ~;l ~