Date of Award
Doctor of Philosophy (PhD)
G. Richard Whittecar
This dissertation develops a new paradigm in a water quality monitoring approach to parameterize spatiotemporal estuarine water quality with sustainable reliability, less cost and less time. A key underpinning of this paradigm of the spatiotemporal estuarine water quality parameterization is various water quality parameters' interrelationship with ambient water temperature as a common factor, their time dependent characteristics, and spatiotemporal characteristics of remote sensing. It has two core models to provide input data of water quality parameterization model in a system; the transfer function models of the physical system and an analytical temperature time series model. The objective of this dissertation is to provide an alternative tool for monitoring water quality and decision-making in estuaries with time and space, to identify system components contributing to physical water quality, and to demonstrate the feasibility, reproducibility and applicability of the proposed model. The spatiotemporal estuarine water quality parameterization model monitors chlorophyll concentration using remote sensing, transfer function models of dissolved oxygen (DO) and orthophosphate (PO4) and ambient water temperature in spring and fall in the James River Estuary Mesohaline segment in Virginia. The proposed model is applicable in the temperature range between 6°C and 23°C in spring and in the temperature range between 21°C and 32°C in fall. The optimal operational temperature range of the proposed model is between 19°C and 25°C based on the relative sensitivity analysis of DO transfer function model. The proposed models in two seasons are compared with the models that use different approaches such as a conventional approach and a previously proposed approach based on various criteria. The results show that the proposed models present the variability of chlorophyll concentration better over time and temperature than other approaches. The results also support that the transfer function models can be successfully applied to estimate chlorophyll instead of using monitored water quality data directly. The proposed models present difficulty to estimate extremely high concentrations of chlorophyll; however, they produce estimations comparable to observed chlorophyll concentrations that are less than the extreme outliers in each season. The mean chlorophyll concentration that is produced by the best proposed model is 7.937μg/L and the +/- 95% confidence intervals of the mean are 7.977μg/L and 7.897μg/L after eliminating the extreme outliers (371μg/L) in spring. The mean, 7.937μg/L, is compatible with the mean of the observed concentrations that are less than the extreme outliers, 7.572μg/L. The mean chlorophyll concentration that is produced by the best proposed model is 5.520μg/L, and the +/- 95% confidence intervals (C.I.) of the mean are 5.538μg/L and 5.502μg/L after eliminating the extreme outliers (22μg/L) in fall. The mean, 5.520μg/L, is compatible with the mean of the observed concentrations that are less than the extreme outliers, 6.117μg/L. This dissertation demonstrates the feasibility, reproducibility and applicability of the paradigm in spatiotemporal estuarine water quality parameterization using remote sensing data and field measured water quality data in estuaries. The spatiotemporal estuarine water quality parameterization model can enhance an existing water quality monitoring and assessment program in estuaries that are managed by municipal agencies and local water quality decision makers. The spatiotemporal estuarine water quality parameterization model can be employed as a tool to guide management, since a systematic process of estimating water quality targets is difficult in a complex estuary. Over time, the model provides appropriate, up-to-date guidance. Careful consideration is necessary when applying transfer function models and seasonal spatiotemporal estuarine water quality parameterization models to the different estuaries directly. Although the models appear feasible with significant potential, direct implementation of the model requires a site-specific quality assurance/quality control (QA/QC) check.
Yu, Kwisun P..
"Spatiotemporal Estuarine Water Quality Parameterization Using Remote Sensing and in-situ Characteristics"
(2009). Doctor of Philosophy (PhD), Dissertation, Civil/Environmental Engineering, Old Dominion University, DOI: 10.25777/zrf7-ya13