Water Quality Analysis of the St. Lucie Estuary: A Statistical and Remote Sensing Approach

Authors

Angelica MoncadaFollow

Document Type

Dissertation

Degree

Doctor of Philosophy (PhD)

Major/Program

Earth Systems Science

First Advisor's Name

Assefa M Melesse

First Advisor's Committee Title

Committee chair

Second Advisor's Name

Daniel Gann

Second Advisor's Committee Title

Committee member

Third Advisor's Name

René M. Price

Third Advisor's Committee Title

Committee member

Fourth Advisor's Name

Leonard J. Scinto

Fourth Advisor's Committee Title

Committee member

Keywords

water quality, pollution, remote sensing, CDOM, estuary, St. Lucie Estuary

Date of Defense

3-28-2023

Abstract

Nutrient enrichment has triggered substantial ecological changes in coastal waterbodies. The St. Lucie Estuary in Florida, USA, is one of many with reported increases in toxic algal bloom events, low oxygen, and declines in fish populations. Extensive hydrologic modifications and increased population have created a hydrologically and ecologically complex system. As such, water quality monitoring and mitigation efforts are of utmost importance. Long-term water quality data and remote sensing technologies can improve our understanding of water quality fluxes in complex systems. The present study used two robust datasets, including real-time sensor data, WorldView-2 (WV-2) and Landsat-8 images, to assess the variation of water quality constituents and their effects on surface water reflectance (R_rs). Principal component analysis and nonparametric trend tests were applied to a 20-year dataset (1999 – 2019) of ten water quality variables covering ten sampling sites. Nutrients were negatively correlated with pH and dissolved oxygen (DO). Phosphorus and nitrogen had moderately decreasing trends at the North Fork, while DO and pH had moderately increasing trends. Inflows from Lake Okeechobee were characterized by high turbidity, while higher phosphorus and color in platinum-cobalt units (PCU) concentrations characterized inflows from other tributaries. Field spectra and simultaneous water quality samples were collected from various sites to assess their relationships. The field-simulated WV-2 coastal blue-to-green and coastal blue-to-red band ratios captured differences in chlorophyll-a and turbidity in the North and South Forks of the estuary. Significant limitations with WV-2 images impacted results, including negative values, insufficient images, and unsuitable atmospheric correction models. However, correlations between R_rs and water quality constituents showed that color PCU had a strong inverse correlation (p < 0.001) with WV-2 coastal blue band (R = -0.83), WV-2 blue band (R = -0.73), and Landsat-8 green/red band ratio (R = -0.76). Salinity was positively correlated with the Landsat-8 green/red band ratio. A forward stepwise multiple regression analysis using Landsat-8 bands revealed the best-fitted linear model (R² = 0.67) for fluorescent dissolved organic matter in the estuary was developed using 18 Landsat-8 images with the green-to-red, coastal aerosol-to-red, and blue-to-NIR bands. Larger in situ-satellite matchups and exploration of nonlinear relationships could potentially improve the prediction models.

Identifier

FIDC011020

ORCID

0000-0002-6473-6768

Previously Published In

Moncada AM, Melesse AM, Vithanage J, Price RM. Long-Term Assessment of Surface Water Quality in a Highly Managed Estuary Basin. International Journal of Environmental Research and Public Health. 2021; 18(17):9417. https://doi.org/10.3390/ijerph18179417

Recommended Citation

Moncada, Angelica, "Water Quality Analysis of the St. Lucie Estuary: A Statistical and Remote Sensing Approach" (2023). FIU Electronic Theses and Dissertations. 5303.
https://digitalcommons.fiu.edu/etd/5303