Document Type

Thesis - Open Access

Award Date


Degree Name

Master of Science (MS)


Agricultural and Biosystems Engineering

First Advisor

Rachel McDaniel


attachment, baseflow, E. coli, fecal indicator bacteria, shear stress, storm event


Fecal indicator bacteria, including E. coli, are the leading cause of water quality impairments within assessed waters in the United States. The source of E. coli includes WWTP, leaking sewers, animal manure, wildlife, livestock, and stream bed sediment. Storm events contribute to bacteria loading within waters through wash-in of land sources of bacteria and resuspension of bacteria within sediments. Bacteria introduced into the water column are either attached to particles or are unattached (or free-living). The goal of this study was to examine the attachment of E. coli to different particle sizes, including their impact on contributing to water quality impairments during storm events. A series of storm events and baseflow conditions were monitored within an impaired stream (Skunk Creek) located in eastern South Dakota. Samples were taken during storm events over a 5- hour duration via autosampler while baseflow samples were taken via grab sampling. In addition, flow and water quality parameters (i.e. turbidity and temperature) were monitored, and the bed shear stress was estimated. These variables were used in a correlation analysis to determine their relationship with E. coli, including the prediction of E. coli within the water column during storm events. Unattached E. coli dominated total E. coli concentration across both storm and baseflow events (i.e. at least 75% of total E. coli concentrations). The water quality standard during baseflow conditions was satisfactory while storm events consistently exceeded the standard. Total, settleable and free-living E. coli concentrations ranged from 7 × l02 to 22 × l03 CFU 100 mL-1 , 4 × l01 to 66 × l02 CFU 100 mL-1, and 5× l02 to 15 × l03 CFU 100 mL-1, respectively. The high levels and exceedance rate of free-living E. coli mean that sedimentation of the settleable fraction of E. coli would not be adequate to reduce bacteria to within the microbiological water quality standard. Many instream water quality models assume that the total bacteria concentration within the water column can be predicted by modeling bacteria as free-living; this assumption was tested by assessing the statistical difference between total and unattached bacteria. The findings revealed that free-living E. coli concentrations were equal to total E. coli concentrations 5 out of 8 times (63%), meaning that over one third of events would not be accurately modelled with only unattached bacteria. Thus, increased understanding of attachment and incorporation of bacteria partitioning between attached and unattached (free-living) into water quality models could improve model performance and predictive capabilities. The correlation analysis revealed a weak (p > 0.05) relationship between flow, temperature, turbidity, shear stress and E. coli fractions. Regression models developed to predict total E. coli and those attached to different particle fractions during storm events performed poorly (R2 = 0.09-0.22). The results presented in this study will further the understanding of fate and transport of bacteria within water as well as provide information that can be incorporated into the development of microbial water quality models.


Includes bibliographical references



Number of Pages



South Dakota State University


In Copyright - Educational Use Permitted