Date of Graduation

2016

Document Type

Dissertation

Degree Type

PhD

College

Davis College of Agriculture, Natural Resources and Design

Department

Wildlife and Fisheries Resources

Committee Chair

Kyle J Hartman

Committee Co-Chair

Patricia M Mazik

Committee Member

Raymond P Morgan II

Committee Member

J Todd Petty

Committee Member

Stuart A Welsh

Abstract

The Influence of Dams on Downstream Larval and Juvenile Fish and Benthic Macroinvertebrate Community Structure and Associated Physicochemical Variables R. Daniel Hanks The influence of dams on downstream biotic and abiotic components of aquatic ecosystems has been largely studied within the context of the River Continuum (RCC) and Serial Discontinuity Concepts (SDC). Few of these studies have sufficiently studied how these variables change along the longitudinal gradient below the impoundments in a systematic manner, comparing equal distances below both epilimnetic and hypolimnetic dams to a reference condition. This is especially true of early life stages of fish (i.e., larval and juvenile stages) and macroinvertebrate functional groups. Here, we systematically evaluated the effects of dams at 16 sites downstream of dams for their impact on physicochemical (instream habitat [e.g., substrate, flow, etc.] and water quality [i.e., DO, pH, conductivity, and temperature], and landcover [i.e., % forested land, % developed land, and % grassland]) and various metrics for larval and juvenile fish and benthic macroinvertebrates.;Effective capture of larval and juvenile fish was paramount for the evaluation of dam influences on larval and juvenile. Sampling larval fish at various life stages can be difficult in shallow, structurally and spatially diverse streams. We evaluated three commonly employed methods (light traps, drift nets, and spot-and-sweep) for sampling larval fish in these systems. We found the spot-and-sweep method captured a higher abundance of larvae than either drift nets or light traps during both daytime and nighttime hours. Additionally the spot-and-sweep method captured as many different taxa as drift nets and more than light traps. The coefficient of variation was lower for spot-and-sweep than for either drift nets or light traps for both taxa richness and larval abundance. Richness for daytime and nighttime spot-and-sweep sampling was equal. Mean richness was also equal between the two periods, and mean CPUE was not significantly different between periods. The coefficient of variation was lowest for daytime spot-and-sweep sampling, suggesting it was less variable than nighttime sampling. The spot-and-sweep method showed promise for determining taxa presence and relative abundance. Discrepancies in the ability of personnel while performing spot-and-sweep sampling was investigated and found to be insignificant. Of the three methods evaluated for sampling structurally complex and spatially heterogeneous streams the spot-and-sweep method was found to be the most effective. We investigated the effects of dams on downstream larval and juvenile fish. Generalized additive models indicated that there was a general increase in abundance, genus richness, and Shannon diversity associated with increasing distance from dams. Principal component analysis (PCA) indicated three influential PC's that were structured by landcover, habitat and water quality, and disturbance. Nonmetric multidimensional scaling (NMDS) indicated larval and juvenile fish communities were structured differently between epilimnetic and hypolimnetic releases and that habitat variables structuring those communities were more variable in epilimnetic releases than hypolimnetic releases. We systematically evaluated both the abiotic and biotic (i.e., benthic macroinvertebrates at the family level) along the stream continuum below impoundments with both epilimnetic and hypolimnetic releases and compared those findings to a reference stream. Generalized additive models (GAMs) identified six habitat variables (i.e., substrate coarseness, substrate diversity, pH, temperature, stream width, and stream depth) as significantly related to distance from dam. GAMs also indicated that abundance was not significantly related to distance from dam but both family level richness and Shannon diversity exhibited significant increases with increasing distance from dams. We evaluated patterns of changes in physicochemical and macroinvertebrate functional group components of aquatic systems along the longitudinal gradient below dams and compared changes in these variables to an undammed reference stream. Generalized additive models indicated that genus richness, functional richness, tolerance, dispersal, percent five dominant genera, EPT, and GLIMPSS were lower in dammed streams than in our reference stream. Genus and functional richness, percent 5 dominant genera, EPT, and GLIMPSS all increased as distance from dams increased while they remained relatively consistent within our reference stream. Tolerance and dispersal changed with distance from dams in dammed streams but showed little change in our reference stream. Percent composition of functional groups was different between dammed and reference streams; in dammed streams the percent composition changed with increasing distance from dams, but remained relatively stable in our reference stream. Genus and functional richness also exhibited two distinct gradients within the 5,100-m that we sampled below dams where a short, rapidly changing gradient existed immediately below dams to approximately 2,000-m, followed by a more gradual steadily increasing gradient that appeared to continue beyond our most distant sampling location below dams (i.e., 5,100-m). Important explanatory variables that varied in statistical significance between response variables but were commonly significant with distance from dams was substrate coarseness and percent forested land. Eighty five percent of our measured abiotic variables below dams had higher r values where curvilinear relationships were modeled as compared to linear relationships; whereas only 46% of the biotic variables had higher r values with curvilinear models. Nonmetric multidimensional scaling (NMDS) confirmed our GAM results indicating benthic macroinvertebrates below dams show structural changes along the stream continuum.;In all cases (larval and juvenile fish, family level aquatic macroinvertebrates, and genus level aquatic macroinvertebrate metrics) our findings generally agreed with the SDC but future studies should aim to sample in a spatially systematic manner, as this will improve our understanding of how dams influence abiotic and biotic components of aquatic systems. Additionally, our studies consistently indicated two gradients existed for most biotic measures. We believe further studies are required to understand the two recovery gradients that exist below dams and the extent of dam influences along the stream continuum.

Download

Share

COinS