Date of Award

January 2014

Document Type


Degree Name

Master of Science (MS)


Civil Engineering

First Advisor

Yeo Howe Lim


It is hypothesized that the applied sampling techniques, water quality analysis, and statistical analysis predict pollutant removal efficiencies of the project site. Current practices in urban stormwater runoff are the design of systems that limit the developed peak discharge to less than or equal to the peak discharge of the pre-developed conditions. This is many times accomplished with the installation of stormwater retention, detention, or attenuation facilities that store the generated runoff from the drainage area. These are commonly known as structural Best Management Practices (BMPs). The City of Grand Forks, ND (City) implements BMPs into the stormwater management plans for all new developments. Design of these facilities for water quality is volume based, and considerations for removal efficiency are not currently integrated. The City is interested in determining the pollutant removal efficiency of their current in-situ structural BMPs.

This research is used to develop a sampling plan and parameter list for potential future expansion of the project. To determine an accurate sampling plan and parameter list, a baseline study on one operational wet detention pond located within the City was completed to prove the hypothesis. Since this is a baseline study, water quality parameters included the analysis of total suspended solids, nutrients of various forms, heavy metals, bacteria, and other chemical properties used to assess the current quality of stormwater influent and effluent going through the system. The sample collection

includes both single grab samples for instantaneous water quality analysis and manual flow-weighted composite samples for analysis of event mean concentration (EMC). The EMC influent and effluent results are compared to determine intra-event removal efficiency and a statistical analysis is performed to determine if the sample sets are statistically significantly different between the influent and effluent concentrations. Acceptance of the hypothesis is proven for the nutrients, total phosphorus and nitrate as nitrogen, and conductivity. The average removal efficiency of the nutrients is 73 percent for total phosphorus and 40 percent for nitrate as nitrogen. Conductivity was determined to increase between the influent and effluent concentrations. Other analytes that exhibited removal efficiency, but were not proven to be statistically significant, were total suspended solids (TSS) at 76 percent removal, phosphate as orthophosphate at 71 percent removal, and bacteria as E. Coli at 83 percent removal. The remaining parameters of ammonia as nitrogen, nitrite as nitrogen, total copper, total lead, total zinc, chloride, pH and dissolved oxygen did not suggest effective removal trends throughout the BMP. Continued analysis of the site is required to better define the statistical difference between the influent and effluent concentrations for these parameters. The removal of the nitrite from the parameter list is suggested based on low to non-detection of the analyte throughout the monitored sampling events. TSS and conductivity were observed to be potential surrogates for total phosphorus and chloride, respectively.