9484 - Laserfiche WebLink

Home Browse Search

December 2002 FLOW-SEDIMENT EFFECTS ON RIVERINE FISH TABLE 1. Threshold discharges (m3/s).for initial motion of the bed (Q.) and the bankfull discharge (Qb) in the study strata. Stratum Q? Qn 11 246 623 10 211 580 9 278 608 8 548 979 7 519 1021 6 497 1320 5 609 1429 4 650 2929 3 561 2012 2 659 1543 1 nat 1788 Notes: The threshold for full mobilization of the bed occurs at the bankfull discharge in strata 6-11, but probably at lower discharges in strata 1-5. Initial motion of the bed occurs at very low flows in stratum 1 due to the unconsolidated nature of the predominantly sand bed. t Not applicable. stream (ANCOVA, F,,92 = 0.79, P = 0.38). However, mean grain size of coarse material in riffles did sig- nificantly decrease (ANCOVA, F1,70 = 4.22, P = 0.044) in a downstream direction. Grain size of the gravel- cobble portion of the river bed declined downstream in both habitat types between strata 11 and 6, then increased downstream through stratum 3. Differences between the longitudinal patterns of the D50 (all sediments) and those of the mean sizes of coarse particles (gravel-cobble) resulted from the varying proportion of fine sediment (<2 mm) present in the bed (Fig. 2e); this fine sediment fraction was greater in runs than in riffles (ANOVA, F,,,, = 27.86, P < 0.00001) and in runs it increased downstream (ANCOVA, F,95 = 54.69, P < 0.00001), whereas in riffles it did not (ANCOVA, F,,75 = 0.85, P = 0.36). This fine sediment was responsible for a major increase in percent em- beddedness in downstream runs (ANCOVA, F1,486 = 209.4, P < 0.00001). In contrast, percentage of em- beddedness in riffles did not significantly change lon- gitudinally (ANCOVA, F,,711 = 2.77, P = 0.10). Per- 1727 centage of embeddedness (Fig. 2f) was higher in runs than in adjacent riffles (ANOVA, F1,724 = 357.3, P < 0.00001) and this difference increased substantially downstream of stratum 6. The quantity of fine sediment in the streambed also was reflected in measurements of total DFS and inter- stitial void volume. DFS was less in runs than in riffles (ANOVA, F1,719 = 154.7, P < 0.00001). In runs, DFS decreased significantly (ANCOVA, F,,,,, = 70.25, P < 0.00001) in a downstream direction (Fig. 2g). How- ever, in riffles, although DFS declined between stratum 10 and 8, and was similar in strata 6-8, it increased downstream of stratum 6, and the overall downstream trend was not significant (ANCOVA, F,,389 = 28.51, P = 0.28). As expected, the longitudinal pattern of void volume resembled that of DFS (Fig. 2h), and void vol- ume in riffle substrates was significantly greater (AN- OVA, F,,726 = 82.91, P < 0.00001) than in run sub- strates. Several physical parameters were highly correlated across riffles and runs (Table 2). Principal component analysis was used to characterize sites according to their streambed attributes (including detritus dry mass and midcolumn velocity) with a single variable to fa- cilitate comparison with periphyton and invertebrate mass. The factor scores of the first principal component provided a single variable that contained 58% of the information of the original seven bed-sediment param- eters (Table 3). Other principal components explained <17% of the variability in physical parameters (range 1.7-16.6%). The loadings of the first principal com- ponent variable suggest a strong positive relationship (all same sign) with Dso, void volume and total DFS, and a lesser positive relationship with midcolumn ve- locity and detritus dry mass (Table 3). Strong inverse relationships (inverse sign) with percentage of sub- strate <2 mm and percentage of embeddedness were also found for the first principal component variable. These relationships indicate that the score of the first principal component relate to the amount of sedimen- tation of the streambed at a location: higher scores TABLE 2. Pearson correlation coefficients for various physical parameters, chlorophyll a concentrations (mg/m2), and in vertebrate dry mass (mg/ml). Parameter Chl a Inverte- brates Detritus Sub <2 mm D50 Voids % Embedded DFS Invertebrates 0.60 Detritus 0.41 0.54 Sub <2 mm -0.50 -0.51 -0.45 D50 0.43 0.51 0.46 -0.83 Voids 0.35 0.44 0.42 -0.76 0.91 % Embedded -0.35 -0.54 -0.47 0.76 -0.72 -0.70 DFS 0.43 0.53 0.39 -.0.76 0.79 0.76 -0.80 Velocity 0.17 (0.016) 0.51 0.33 -0.37 0.28 0.22 (0.001) -0.50 0.40 Notes: All data were loge transformed for analysis. All correlations were significant at P < 0.00001 for r = 0.0, except the two indicated in parentheses. Parameters in order are: chlorophyll a concentration (Cbl a), invertebrate dry mass (In- vertebrates), detritus dry mass, fraction of substrate particles <2 mm (Sub <2 mm), median particle size of the surface layer (D50), volume of interstitial void space within the substrate, percentage of the surface area consisting of fines (% Embedded), absolute depth of free space (DFS), and midcolumn water velocity.