The bluehead chub (Nocomis leptocephalus) is a keystone species, an ecosystem engi- neer that constructs mounded gravel nests for spawning. Chubs provide benefits for other spawning fishes, predators, and benthic organisms through their nest construction. This study seeks to apply sediment transport models to find incipient motion criteria and erosion susceptibility of chubs nests. Field water flow velocities were measured with an acoustic Doppler velocimeter (ADV) in Tom's Creek, Blacksburg, Virginia, USA. ADVs are often used to collect in-situ turbulent velocity data. In almost all applications of ADVs, erroneous spikes are recorded during collection, which can significantly distort turbulence statistics de- rived from velocity fluctuations. In this study, a bivariate kernel density estimation despiking algorithm is compared to a novel univariate simplification developed as part of this work. Despiking methods are evaluated using field ADV and direct numerical simulation (DNS) data of a turbulent boundary layer. Visual assessment of despiked velocity time series and power spectra and corresponding changes in statistical moments, as well as response to arti- ficial spiking of DNS data, yield valid performance of the univariate method. After despiking chub nest data, multiple methods of finding bed shear stress from velocity vertical profiles are evaluated. Bed shear stress is found over the profile of 26 field nests. The ambient to peak flow stress amplification due to a nest's bed protrusion is found to be a proportion of τ = 1.66τ to determine a critical ambient Shields parameter of approximately τ∗ = 0.03 pa c,a for nests. / Master of Science / The bluehead chub is a keystone species, an ecosystem engineer that constructs mounded gravel nests for spawning. These nests benefit numerous other species within their habitat. The possibility of nest erosion is characterized in this study using existing sediment transport principles. Nest flow characteristics are measured using an acoustic Doppler velocimeter (ADV). ADVs are often the instrument of choice in measuring water velocity. The high resolution of these devices can capture turbulent flows well, however data collection often results in inclusion of erroneous spikes in velocity. These spikes represent points deemed impossible due to their sudden change in velocity magnitude. Spikes do not have a large effect on average velocities of ADV data, but can influence turbulence statistics that describe the turbulent fluctuations in flow velocities. To remove spikes, multiple methods incorporate different outlier detection principles. This study evaluates a popular method that employs a two-dimensional kernel density estimation (KDE) algorithm. A recent study suggested the possible simplification of this method to use a one-dimensional kernel density estimation instead. Both the 1D and 2D methods are evaluated in this study in how they filter ADV data and whether it results in a clean, improved velocity time series that would be expected in turbulent flows. A novel 1D KDE method was also developed as part of this study. The new method is found to produce the most reliable filtering. Despiked ADV data is used to characterize the hydraulics over bluehead chub nests sampled in Tom's Creek, Blacksburg, Virginia, USA. Hydraulics are evaluated to see if ambient flow upstream of a nest can characterize the flow characteristics over the peak of the nest. Shear stress amplification over the nest is used to find a critical threshold for nest erosion based on ambient flows. Stress amplification in the ambient flow to the peak over the nest is found to be a simple proportion, and amplification factor of 1.66. This means we expect shear stress over the peak of a chub nest to be 1.66 times greater than the ambient bed shear stress upstream of the nest. This amplification factor can be used with existing methods to calculate critical non- dimensional shear stress values, also known as the critical Shields stress. After accounting for the amplification factor of 1.66, a critical Shields of approximately 0.03 is found.