Back to search

University of British Columbia

Development of machine learning methods for improved fluorescence-based monitoring of environmental contaminants in surface waters

Abstract

dc:description

This research examined the use of fluorescence spectroscopy as a rapid contaminant detection method. Identifying and quantifying environmental contaminants in surface waters has always been challenging after transient events such as spills or storm events. Previous research has shown fluorescence spectroscopy to be a promising tool for continuous monitoring of water quality due to its real-time and sensitive detection capabilities. However, persistent challenges with fluorescence approaches remain and limit its use in contaminant monitoring. In this study, research was carried out to identify possible data analysis techniques that improve the separation of overlapping fluorescence signals from background noise to accurately identify and quantify contaminants of interest. Algorithm-based dimensionality reduction methods, including parallel factor analysis (PARAFAC), principal component analysis (PCA), and autoencoder neural networks, were compared to a basic peak-picking method. It was observed that contaminant detection was most accurate using dimensionality reduction methods that maximally explain dataset variance, such as PCA, although at the cost of component interpretability. Due to the interference from background organic matter and other environmental impacts on fluorescence signals, fluorescence-based contaminant detection models are highly source-specific and require significant effort and resources for calibration. The novel application of data processing techniques was investigated to enable the transfer of fluorescence detection models from one water source to another. A contaminant detection model from a relatively consistent and low organic background source was successfully transferred to a second source with higher organic concentration and distinct water quality. Only a few additional fluorescence spectra of the background water quality and contaminants of interest were required to successfully transfer the model without the need for labelled samples in the new source. Notable differences in peak location and spectral shape of identical compounds were found in source-specific models between the two water sources, implying variability in fluorescence signals resulting from environmental conditions. Despite the impact of environmental conditions, features identified by principal component analysis (PCA) or an autoencoder produced sensitive transferred models capable of addressing the spatial and temporal variability in source waters.

Degree

thesis:*
Name thesis:degree_name
Master of Applied Science - MASc
Level thesis:degree_level
master's
Discipline thesis:degree_discipline
Civil Engineering
Grantor dc:publisher
University of British Columbia
Year dc:date
2021

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Li, Ziyu
Contributors dc:contributor
  • Peleato, Nicolas M.

Rights

dc:rights
Statement dc:rights
  • Attribution-NonCommercial-NoDerivatives 4.0 International
Language dc:language
eng

Identifiers

dc:identifier.*
Handle dc:identifier
http://hdl.handle.net/2429/79739

Chain of custody

source
Harvested from
University of British Columbia
Base URL
circle.library.ubc.ca/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Li, Ziyu. Development of machine learning methods for improved fluorescence-based monitoring of environmental contaminants in surface waters. master's thesis, University of British Columbia, 2021. http://hdl.handle.net/2429/79739