Impacts of Land Use Disturbance on Fish and Aquatic Macroinvertebrate Assemblages in the Nolichucky River Watershed
Gotwald, Hayley S.  2016.  M.S. Thesis.

Abstract:
Southern Appalachian watersheds of the United States are negatively affected by pesticides and fertilizers used in row crop agriculture. The objective was to determine if the amount of row crops is connected to changes in aquatic biotic assemblages draining the Nolichucky River watershed in east Tennessee. The hypothesis was the amount of row crops will negatively correlate with indices of biotic integrity (IBI) metrics for fish and benthic macroinvertebrates indicating healthy aquatic communities. For 18 sample sites in 2014 and 2015, IBI metrics were calculated. Water quality and elevation measurements were made before conducting IBIs. To assess changes in and amounts of land use/land cover (LULC), maps from 1999 to 2014 were produced with Landsat satellite imagery. Pollutant estimates (sediment, phosphorus, and nitrogen) were calculated using the Soil & Water Assessment Tool (SWAT) model. The area of row crops increased since 1999 (39 km2 in 1999 to 71 km2 in 2014). A principal component analysis was performed on LULC measurements from different scales (local, reach and catchment), water quality data, and elevation to produce a reduced set of explanatory variables that were uncorrelated but could be associated with IBI metrics. A canonical correspondence analysis associated fish metrics with LULC types: Impervious surfaces, non-row crop fields, and forest (p = 0.04 for axis 1 eigenvalue, p = 0.05 for species-environment correlations). For the benthic macroinvertebrate metrics, nonmetric multidimensional scaling found metrics indicative of poor stream health (percentage of oligochaetes and chironomids, percentage of nutrient tolerant organisms) were strongly positively associated with increasing use of row crops, impervious surfaces (p ≤ 0.01), and pollutant estimates (p ≤ 0.004). A redundancy analysis found increasing pollutant estimates were associated with fish metrics indicative of poor stream health vi (percentages of hybrids, piscivores, diseased fish, and number of sunfish species) (p = 0.03). When watersheds of tributary streams are converted to impervious and non-row crop field LULC, they function biologically like the larger main stem river. Although fish and benthic macroinvertebrate metrics indicated the tributary and main stem Nolichucky sites were in relatively good condition, increases in land conversion can further degrade stream biotic integrity.