Date of Award
Winter 2024
Project Type
Dissertation
Program or Major
Applied Mathematics
Degree Name
Doctor of Philosophy
First Advisor
Anne Lightbody
Second Advisor
Gregory Chini
Third Advisor
Laura Dietz
Abstract
In addition to transporting flowing water, river systems carry dissolved nutrients and pollutants, which can experience chemical reactions, biological uptake, or retention in surface or hyporheic zones. This dissertation enhances our understanding of solute transport and uptake in river reaches through three main components. The first component involves the development of a novel curve-fitting model designed to capture the characteristics of pulse-release breakthrough curves (BTCs) from tracer studies with conservative solutes. The model, which was found to be applicable to a wide range of field studies, enables the rapid comparison of transport parameters among diverse field studies obtained under a range of experimental conditions. The second component utilizes a particle tracking model (PTM) to simulate the movement of solute parcels within an idealized river system following a pulse release. PTM results show that increased access of hyporheic zones in the subsurface is reflected in the BTC and can elevate overall solute uptake. The third component develops a Random Forest (RF) machine learning model to predict reach-scale BTC shapes and nutrient uptake within stream systems. While the RF model provides accurate forecasts when much information is known about stream transport and solute reactivity, its accuracy diminishes with fewer input parameters. By introducing a new BTC fitting model, providing detailed insights into transport dynamics, and offering predictive tools that reduce the need for extensive field sampling, this work advances our understanding of the complex dynamics of solute movement and uptake in river systems, providing insight into nutrient cycling, pollutant dispersion, and other critical processes in river ecosystems and aiding in stream management and restoration efforts.
Recommended Citation
Liu, Jiaying, "Improving Understanding of Fluvial Reach-Scale Solute Transport and Uptake Using a Particle Tracking Model and Random Forest Machine Learning Model" (2024). Doctoral Dissertations. 2878.
https://scholars.unh.edu/dissertation/2878