Graduation Year

2023

Document Type

Thesis

Degree

M.S.

Degree Name

Master of Science (M.S.)

Degree Granting Department

School of Geosciences

Major Professor

Barnali Dixon, Ph.D.

Committee Member

Joseph Smoak, Ph.D.

Committee Member

Jeremy Decker, Ph.D.

Keywords

Multi-Criteria Decision Making (MCDM), Principal Component Analysis (PCA), Python, Weighted Linear Combination (WLC)

Abstract

While many aspects of climate change are well-studied, the vulnerability of Onsite Sewage Treatment and Disposal Systems (OSTDS) to sea-level rise (SLR) has not been thoroughly examined despite the potentially significant environmental impacts, including nutrient and pathogen loading. Research gaps include the need to determine the suitability of models to predict groundwater inundation due to SLR and the consequent impact on OSTDS. As such, this study aims to compare Bathtub/hydrostatic and physics-based/MODFLOW/3D numerical modeling methods of simulating groundwater inundation from SLR to identify OSTDS vulnerability to failure due to a reduction in the height of the vadose zone (saturation or flooding of the drain field/treatment failures) for Broward County, Florida.

The primary goal of this project is to evaluate OSTDS vulnerability under current and future inundation criteria using two models (Bathtub and MODFLOW) integrated with Multi-Criteria Decision Making (MCDM). The secondary objective is to evaluate the predictability and uncertainty of SLR models (Bathtub and MODFLOW) and their impacts on identifying vulnerable OSTDS. A GIS-based MCDM technique is used to create three OSTDS vulnerability assessment models referred to as MCDM-Bathtub-Current, MCDM-Bathtub-Future, and MCDM-MODFLOW-Future where two different methods of SLR predictions and two SLR scenarios are used. Model inputs were selected and processed using analytical hierarchy process (AHP) and weighted linear combination (WLC).

This study integrates census data and Multi-Criteria Decision-Making (MCDM) model results to assess tract-level disparate socioeconomic vulnerabilities to Onsite Sewage Treatment and Disposal Systems (OSTDS) vulnerable to failure in Broward County, Florida (USA). Physical processes (i.e., Sea-Level Rise) govern OSTDS failures. Failing OSTDS pose human-health and environmental impacts directly and indirectly through nutrient and pathogen loading, requiring a human dimension of analysis. Within the context of human-health, OSTDS failure will have disparate impacts based on socioeconomic vulnerability.

The objective is to determine OSTDS Vulnerability Zones (OVZ) using the MCDM results, where OVZs overlaid with census data identify population and their characteristics (elderly, child, unemployed, limited English, poverty, and uninsured) residing within OVZs. This study will identify the ratio of people within and outside OVZs in the context of: The ratio of land area, income-based differences, and the ratio of disadvantaged populations. A hotspot analysis will provide further insights by identifying local clusters of disadvantaged groups.

This study also uses the three MCDMs to identify three OVZs; these OVZs identified vulnerable OSTDS that were then used as input in an ArcGIS-based Nitrate Load Estimation Toolkit (ArcNLET). ArcNLET estimated particle paths for each set of OSTDS identified and were used to track OSTDS effluent to waterbodies. Spatial co-occurrence was used to determine if these OSTDS may be contributing to high levels of Nitrogen or Fecal Coliform, and to compare the water table surface generated via MODFLOW and a bathtub equivalent to identify similarities.

Results show the SLR-Bathtub-Future model was able to reproduce SLR-MODFLOW-Future model’s inundation surface at the 3-foot level (0.9 m) and 2-foot level (0.6 m) with an accuracy of 89% and 86%, respectively. Of the 56,080 OSTDS found in Broward County. A total of 1,961 OSTDS were identified as highly vulnerable by all models and 577 of these 1,961 were identified as common across all three models, and 4,352 were identified as moderately highly vulnerable from all three models. These OSTDS need to be prioritized to minimize failure.

Statistical results indicated that disadvantaged groups residing in OVZs are characterized by Poverty, limited English, unemployed, and uninsured. These individuals tend to lack adequate resources to cope with OSTDS failure. For example, populations with low income can experience disproportionate fiscal burden associated with modifications/repairs for OSTDS if they are homeowners. Therefore, the use of tract-level information can help comprehensive impact assessment of OSTDS failure and help policy development to provide targeted assistance and develop action plans to ameliorate OSTDS failures and minimize OSTDS impacts amongst these individuals.

Results of the analysis show that Bathtub-Future predictability differs greatly from MODFLOW-Future with a percentage error of -68.7%. OSTDS are contributing to the exceedance of the TMDL at certain waterbodies, especially South New River Canal (C-11). Furthermore, it has been observed that fecal coliform numbers and nitrate numbers in the area south of South New River Canal (C-11) are much higher than expected levels, where OSTDS identified as vulnerable from the OVZs are present.

Download

Share

COinS