Development of sediment oxygen demand predictive model by characterization of oxygen mass transfer at sediment-water interface

Chen, Chieh Ying

Permalink

https://hdl.handle.net/2142/127309 Copy

Description

Title

Development of sediment oxygen demand predictive model by characterization of oxygen mass transfer at sediment-water interface

Author(s)

Chen, Chieh Ying

Issue Date

2024-07-25

Director of Research (if dissertation) or Advisor (if thesis)

Garcia, Marcelo H.

Doctoral Committee Chair(s)

Garcia, Marcelo H.

Committee Member(s)

• Druhan, Jennifer L.
• Tinoco, Rafael O.
• Valocchi, Albert J.
• Fytanidis, Dimitrios K.

Department of Study

Civil & Environmental Eng

Discipline

Civil Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Sediment Oxygen Demand
• Oxygen Mass Transfer
• Dissolved Oxygen
• Water Quality
• Sediment-Water Interface

Abstract

Sediment oxygen demand (SOD) describes a deficit flux of dissolved oxygen (DO) between the water column and benthic sediment. This critical metric of the DO level in aquatic ecosystems is impacted by both physical effects (flow and transport) and chemical effects (reactivity in the sediment). Shear-induced turbulence at the sediment-water interface (SWI) promotes DO penetration from the water column into the sediment while chemical consumption of DO within sediments induces a diffusive flux at SWI. This strong coupling suggests that models for SOD should encapsulate a range of flow and benthic chemical conditions in order to ensure aquatic ecosystem health and management. This dissertation applied several existing SOD models, including zero-order constant flux SOD model, sediment diagenesis model and Waterman's SOD model with sediment resuspension, to a case study in Bubbly Creek, Chicago, IL. Bubbly Creek is a tributary of South Branch Chicago River in Chicago Area Waterway System (CAWS) and it was used as the drainage for stockyards waste between 1865 and 1939. Nowadays, the benthic sediment of Bubbly Creek still contains abundant organic muck resorting from combined sewer overflows (CSOs), so it is a good candidate to examine the performance of SOD models. The results of DO modeling using different SOD models in Bubbly Creek showed that the accuracy improved with increasing complexity of parameters in SOD models. However, the efficiency and the data availability became obstacles when the model became more complex. Consequently, this dissertation aimed at developing an efficient SOD model considering both physical and chemical effects. For physical effects, a meta-analysis on existing dataset of hyporheic mass flux, enhanced with high Reynolds number cases from a validated computational fluid dynamics (CFD) model in OpenFOAM was performed, to identify key parameters influencing oxygen mass transfer at SWI. Unifying mass transfer models using roughness and permeability Reynolds number were developed and validated against existing models and literature data. The mass transfer model using roughness Reynolds number is easy to use and can provide an estimate of mass transfer coefficients for DO, particularly in scenarios where detailed bed characteristics such as permeability might not be readily available. The obtained DO mass transfer coefficient not only gives the potential of SOD flux indicating maximum DO penetration from water to sediment but also it is a critical input for chemical reaction modeling for oxygen consumption in sediments. For chemical effects, a validated mass balance reactive and transport model (i.e., CrunchFlow) was applied to simulate chemical reactions and quantify oxygen consumption in benthic sediment. Two numerical models, PROBE for simulating near-bed hydrodynamics in boundary layer flows and CrunchFlow for modeling multi-component chemical equilibria and kinetics in sediments, were coupled to investigate SOD dynamics under various hydrodynamic and chemical conditions in flow and sediment. The coupling framework compared simulated fluxes with measurement data and verified the mass balance of oxygen mass transfer at SWI and consumption in sediment. Finally, SOD predictive model was established by applying multiple linear regression (MLR) to prioritize the key parameters of reset SOD, which is highly affected by perturbations at SWI, and steady-state SOD, which leans to chemical effect for oxygen consumption. The developed SOD predictive model was further applied to an analytical analysis for DO concentration prediction under coevolution of SOD and reaeration fluxes for real-scale case study in Bubbly Creek. The developed SOD model is robust to the flow environments and sediment chemical conditions. It is designed to serve as a user-friendly tool that can provide essential data for predicting prompt or long-term SOD values, particularly in scenarios with limited flow and chemical information. The main contribution of this dissertation is to advance water quality modeling in large-scale engineering applications by providing a sophisticated, yet convenient SOD predictive model that can be used to estimate DO levels in flowing aquatic systems.

Graduation Semester

2024-12

Type of Resource

Thesis

Handle URL

https://hdl.handle.net/2142/127309

Copyright and License Information

Copyright 2024 Chieh Ying Chen

Owning Collections