Files in this item



application/pdfKANDEL-THESIS-2018.pdf (4MB)
(no description provided)PDF


Title:Comparative analysis of nitrate load estimation techniques in watersheds with different land-use characteristics
Author(s):Kandel, Rajesh
Advisor(s):Bhattarai, Rabin
Department / Program:Engineering Administration
Discipline:Agricultural & Biological Engr
Degree Granting Institution:University of Illinois at Urbana-Champaign
Linear Interpolation
nutrient load
nutrient concentration
Abstract:Water quality issues stemming from high nutrient concentration in water bodies are a commonplace in the United States (US) and all over the world. Nutrient loads in surface water can be estimated by monitoring their concentrations in the water bodies. The challenge is to determine the optimal data sampling frequency and the best method for the estimation of nutrient concentrations that minimizes the uncertainty caused by data and modeling errors in the estimated data. For that purpose, researchers have used various regression models such as LOADEST, WRTDS, linear interpolation, and error minimization techniques over the years. However, there is no consensus in the scientific community on which method or sampling frequency is best suited to different scenarios of water quality data. This analysis provides further scientific insight by comparing the performances of different estimation techniques on long-term nutrient concentration datasets for three different watersheds in Ohio with different land-use characteristics. This study compared the accuracy of LOADEST, WRTDS, and linear interpolation methods in predicting daily nitrate concentration at four different sampling frequencies using long-term data sets for Maumee River, Cuyahoga River, and Grand River. The land-use patterns in Maumee, Cuyahoga, and Grand River watersheds were agricultural, urban, and forest dominated with mixed land-use respectively. The relationship between the yearly discharge and nitrate concentration were notably different for each watershed. A higher data sampling frequency gave better results for most of the scenarios in the study. The bias for WRTDS method was usually lower than for linear interpolation and LOADEST and the results for WRTDS were consistent for all four data frequencies in most cases. The RMSE and R2 values for LOADEST and linear interpolation models were comparable and better than for WRTDS for all three river datasets. The tradeoff between RMSE and PBIAS was particularly obvious in the Maumee results. The Linear Interpolation method was the best-suited method for Cuyahoga and Grand River datasets which are dominated by urban and forest land respectively.
Issue Date:2018-12-13
Rights Information:Copyright 2018 Rajesh Kandel
Date Available in IDEALS:2019-02-06
Date Deposited:2018-12

This item appears in the following Collection(s)

Item Statistics