First Advisor

David Jay

Term of Graduation

Winter 2021

Date of Publication

3-10-2021

Document Type

Dissertation

Degree Name

Doctor of Philosophy (Ph.D.) in Civil & Environmental Engineering

Department

Civil and Environmental Engineering

Language

English

Physical Description

1 online resource (xi, 150 pages)

Abstract

The purpose of this work is to investigate via data analysis and numerical modeling the SPM (suspended particulate matter) dynamics of a heavily contaminated partially urban estuary, the Lower Passaic River estuary (LPR), NJ. Accordingly, I investigate the quantity and mechanics of variation of fine and coarse SPM in the LPR via data analysis. Data analysis focuses on the parameters that affect SPM dynamics at six moored stations occupied during the Fall and Spring seasons, from near the estuary mouth to tidal freshwater. A 3D hydrodynamic model (Delft3D-FM) is used to analyze the effects of estuary topography on the dynamic distribution of bed shear stress, τb, and to interpret the observations. Moored data from a station seaward of the LPR are used to for model calibration. This work will address three primary issues. The first is to determine bulk settling velocity (Wsb) values and the factors that affect Wsb along the estuarine salinity gradient. The second is to determine the quantity of fine and coarse SPM throughout the water column distributed in Rouse-like and Modified-Rouse profiles, and to (a): investigate the dynamical importance of advection in influencing SPM profile structure for fine and coarse SPM, and (b) determine how the SPM concentration varies with particle size, river flow, and tidal range. These two issues are analyzed using acoustic Doppler current profiler (ADCP) data. An ADCP provides simultaneous profiles of velocity and acoustic backscatter (ABS); the ABS signal can be converted to SPM concentration using appropriate calibration data. Finally, Delft3D-FM was set up on a grid of a generic, convergent estuary similar to the LPR. This grid was used to investigate how oceanographic factors (e.g., channel curvature and tidal range to depth ratio), natural and man-made roughness elements (e.g., grains, meanders, and bridge pilings), and external forcing by river inflow influence the distribution of bed shear stress in a stratified estuary similar to the LPR. To investigate the behavior of bulk settling velocity Wsb (the first question), friction velocity (u*) estimated from the ADCP velocity profile taking into consideration the effect of density stratification due to salinity intrusion. A log-linear velocity equation used when the water column stratified, and a logarithmic velocity profile used to estimate shear velocities,u* for unstratified conditions. Suspended sediment concentration, SSC, was estimated from ADCP acoustic backscatter (ABS) and calibrated against gravimetric SSC samples. Time series of profiles of flow velocity and SSC, and shear velocities used to calculate time series of Wsb via a least-squares analysis that fit a theoretical SSC profile to the ADCP-derived SSC values. Analysis of the resulting time and space distributions of Wsb shows that the mean Wsb decreases landward. In addition, Wsb mainly correlated with Simpson Number (Si, defined in Section 4) in brackish waters, while it primarily correlated with flow velocity in tidal freshwater. Greater diurnal tidal range, TR, and river flow, QR, were secondary factors throughout the system. Investigating the second question (the different behaviors of fine and coarse material) involves making use of defined settling velocity values, the Wsi, to fit observed SPM profiles. These following values were chosen: 0.05 mm/s to represent the fines (wash load to medium silt) at all stations, and 10 mm/s for River Mile (RM) 1.4 and 4.2 and 7 mm/s for RM 6.7, 10.2, and 13.5 to represent the coarser load (fine sand above salinity intrusion and aggregate in the salinity intruded part of the system). A Rouse profile is then assumed for each of the two SSC components, and a non-negative least square regression is applied to calculate the profiles of fine and coarse components in terms of a reference concentration for each component at the base of the profile. The results show a significant ability to describe observed SSC profiles, especially when the profiles are Rouse-like. For other periods, the results showed a good match to the observed SSC profiles when modified Rouse profiles have used that account for the effects of advection on the SSC profiles during periods of strong currents. Also, QR, TR, and horizontal advection are the dominant hydrodynamic factors controlling the variability of fine and coarse SSC, though settling-resuspension processes (not quantified here) are also likely important. The percentage of coarse suspended particles near the estuary mouth is greater than in low-salinity areas and freshwater by ~60% in Fall and ~80% in Spring. This is likely related to aggregation of fines in the moderate salinity waters near the LPR mouth. Furthermore, SSC responded directly to change in velocity; thus, the variation of fine and coarse particles is largely in phase with velocity. The third question, the question of the effects of channel topography and oceanographic factors like stratification and δpx on shear bed stress, will be addressed using a 3D (three-dimensional) grid with the hydrodynamic model Delft3D-FM. The model runs will represent plausible projections of the effect of the roughness elements (from grains roughness, meanders, and bridge pilings) together with tidal range to depth ratio, vertical density gradients, and river flow on the distribution of bed shear stress. The LPR is an urban estuary with many bridges -- 25 below the head of the tide. Not surprisingly, model results have shown a significant influence of these bridge piers (acting as large roughness elements) on τb, stratification and salinity intrusion. Model results show that τb is highest around the bridge's piers and outer sides of the curvatures. Modeled τb is higher upstream near the head of the tide for high flows than low flows, and with rough bed (Chezy 50-30) than the smoother bed (Chezy 70-50). Moreover, more erosion (as inferred from τb distributions) took place on spring-tide ebbs during high flow periods, but on spring-tide floods during low flow periods. Modeled salinity contours move farther landward without bridge piers and lower bed roughness (higher Chezy number) due to reduced vertical mixing. Also, vertical salinity stratification is affected by bridge piers and river flow. The modeled occurrence of stable stratification was reduced during low-flows in the LPR model with piers, while stable stratification occurred prominently near the estuary with/without piers and with high flow. Unstable stratification occurred farther landward direction.

Rights

In Copyright. URI: http://rightsstatements.org/vocab/InC/1.0/ This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).

Persistent Identifier

https://archives.pdx.edu/ds/psu/35134

Share

COinS