Date of Graduation

12-2014

Document Type

Thesis

Degree Name

Master of Science in Biological Engineering (MS)

Degree Level

Graduate

Department

Biological and Agricultural Engineering

Advisor

Brian Haggard

Committee Member

Robert Morgan

Second Committee Member

Wen Zhang

Third Committee Member

Thomas Costello

Keywords

Beaver Lake, Concentration-Discharge, Equal Width Increment, Grab Sample, Loadest, Water Sampling

Abstract

Two primary methods of stream water sampling, the U.S. Geologic Survey (USGS) equal-width increment (EWI) and point samples (PS) from vertical centroid of flow (VCF) were compared at three river sites, the White River near Fayetteville, Richland Creek at Goshen, and War Eagle Creek near Hindsville. A little over three years of concentration data, which was paired with corresponding instantaneous discharge values (http://ar.water.usgs.gov/), was gathered separately at each site by the Arkansas Water Resource Center (AWRC) and the United States Geological Survey (USGS). The purpose of this study was to evaluate how concentration is related to discharge when water samples are collected by the two different sampling methods. The measured constituents included nitrate-nitrogen (NO3-N), dissolved orthophosphorus (soluble reactive phosphorus, SRP), total phosphorus (TP), total nitrogen (TN), and suspended sediment (TSS). A three step process was used to analyze the concentration-discharge relationships: (1) simple linear regression comparison, (2) LOESS residual t-test, and (3) split base and storm flow linear regression comparisons. In addition, an estimation of mean constituent loads and corresponding 95th confidence intervals were calculated using LOAD ESTimator (LOADEST, USGS). In general, PS samples provided results similar to the more rigorous and expensive EWI method. TSS and TN concentrations were significantly lower during storm flow at the White River and War Eagle Creek; however, SRP concentrations gathered by PS sampling method were greater during storm flow at the same two rivers. TP was significantly greater for the PS method during base flow at multiple sites, and combined with SRP results, was most likely due to seasonal variation not captured by the EWI method. Interestingly, no significant differences between methods were shown at Richland Creek for split flow regression comparison. NO3-N was not significantly different between sampling methods at any of the three sites. While both methods provide similar results under certain conditions, research goals and sampling method limitations must be full understood in order to obtain accurate measurements.

APPENDIX A.pdf (2615 kB)

Share

COinS