Masters Theses

Date of Award

8-2006

Degree Type

Thesis

Degree Name

Master of Science

Major

Environmental Engineering

Major Professor

R. Bruce Robinsonq

Committee Members

Randall W. Gentry, John Steven Schwartz

Abstract

The purpose of this study was to combine fish, water quality, and watershed data- bases in order to determine what relationships exist between trout biomass and base flow water quality in the Great Smoky Mountains National Park (GRSM). Quarterly base flow water quality data collected from 1993 to 2003 at 31 stream sites in the GRSM were used in step-wise multiple linear regression models to analyze brook and rainbow trout biomass (kg/ha). Stream samples were analyzed for pH, acid neutralizing capacity (ANC), conductivity, major cations, and major anions. The potential predictor variables included seasonality, basin characteristics, USGS stream flow data as surrogate hydrologic data, precipitation data, e.g. cumulative inches of rain on preceding days, and water quality data. Each of the predictor variables were found to be statistically significant (p<0.05) influencing factors to trout biomass, particularly elevation, basin area, sulfate concentration, maximum stream flow, conductivity, ANC and percent anakeesta geology.

Final correlation analysis, where zeros were assumed for biomass when there were no trout present, revealed that pH, ANC, conductivity, and sulfate are important predictors of trout biomass. Brook trout biomass was not significantly correlated with median pH or log (ANC). However, rainbow trout young-of-year and adult biomass had correlation coefficients of 0.514 and 0.504 respectively with median pH. Furthermore, rainbow trout young-of-year and adult biomass had correlation coefficients of 0.635 and 0.544 respectively with log (ANC). Brook trout young-of-year and adult biomass had correlation coefficients of –0.237 and –0.239 respectively with log (conductivity). Rainbow trout young-of-year and adult biomass had correlation coefficients of 0.613 and 0.368 respectively with log (conductivity). Lastly, both brook and adult rainbow trout biomass had significant negative correlations with sulfate concentrations. Brook trout young-of-year and adult biomass had correlation coefficients of –0.346 and –0.303 respectively with sulfate. Rainbow trout adult biomass had a correlation coefficient of -0.190. Rainbow trout young-of-year biomass was not significantly correlated with sulfate concentrations.

Modeling revealed that brook trout biomass is most strongly related to elevation while rainbow trout biomass is related more to basin area. Elevation is positively correlated with brook trout biomass and accounts for 31% and 40% of the variability in brook trout young-of-year and adult biomass respectively. Similarly, basin area is positively correlated with rainbow trout biomass and accounts for 68% and 40% of the variability in rainbow trout young-of-year and adult biomass respectively. It is thought that basin area and elevation are possible surrogates for stream size. Elevation and basin area have a correlation coefficient of –0.4794, meaning that large basin areas occur at lower elevations and higher elevations have smaller basin areas. Results also showed that young-of-year trout were negatively affected by increases in maximum stream flow and cumulative precipitation for the previous 90 days. Also results for adult trout biomass show a negative relationship with biomass of a competing trout species. Both trout species were negatively affected by increases in sulfate concentrations and percent anakeesta geology. The overall models for biomass produced r-squared values of 0.54 and 0.63 for brook trout young-of-year and adult respectively and values of 0.73 and 0.49 for rainbow trout young-of-year and adult respectively.

Based on the regression results, trout biomass decreases with increases in sulfate concentrations and percent anakeesta. Both of these variables can be linked to acidic stream conditions. Sites along Shutts Prong, Porters Creek, and Walker Camp Prong are at risk because total trout biomass at these locations is either zero or very small and they have greater than 90% anakeesta geology and/or elevated sulfate concentrations.

Brook trout biomass showed a significant positive relationship with elevation. This was expected since results from Baldigo and Lawrence (2001) showed that brook trout seem to be present in streams at high elevations with cold water, steep gradients, small channels, and fast water velocities. Furthermore, brook trout are most often found in small lakes and streams at high elevations which are most susceptible to acidic deposition (Turner et al. 1992). In the GRSM, brook trout are relegated to high elevation streams by historic logging and competition from exotic rainbow trout. Apparently due to competition, brook trout biomass was adversely affected by increases in rainbow trout biomass. Rainbow trout are larger than brook and therefore able to out-compete them for feeding territories. Larson and Moore (1985) found that in GRSM stream segments with similar physical characteristics rainbow trout have biomasses about 1.8 times greater than that of brook trout. Rainbow trout biomass showed a significant positive relationship with basin area, which is consistent with King’s (1943) observation that rainbow trout are relatively larger and more active and therefore choose larger streams, other conditions being the same, than brook trout. Rainbow trout biomass also increases with increasing stream conductivity and ANC. Conductivity is related to the amount of ions (including calcium, magnesium, sodium, and potassium) found in a stream’s water which should benefit fish. The negative relationships between young-of-year biomass and hydrologic conditions can be explained by young-of-year trout being washed away by high flows and/or high flows often corresponding to decreases in stream pH and ANC (Latterell et al. 1998; Driscoll et al. 2001). Final correlation results also indicate that both brook and rainbow trout biomass decrease as the cumulative precipitation for the previous 90 and 180 days increases. This is consistent with results from Barnett (2003) that water quality conditions in the GRSM were adversely affected by increased stream flows, acid deposition and precipitation. Furthermore, negative correlations between sulfate concentrations and biomass give possible evidence to the problem of acid deposition. However, sulfate is also strongly correlated with anakeesta geology. Since this analysis does not consider storm event water quality, it is difficult to separate out the affects of acid deposition versus acidic geology in the GRSM.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Included in

Engineering Commons

Share

COinS