Assessing Microplastic Contamination in Urban Rainwater Harvesting Systems

Thesis Abstract

Urban rainwater harvesting is expanding as a climate-resilient water supply, yet microplastic contamination in stored and distributed rainwater poses potential risks to human and ecosystem health, undermining the sustainability of green infrastructure. This study addresses the gap by empirically assessing the abundance, composition, and sources of microplastics within urban rainwater harvesting systems (RHS) across different city typologies, storage configurations, and filtration regimes. The aim is to quantify microplastic loads, identify dominant polymer types and particle sizes, trace potential entry pathways, and evaluate treatment effectiveness. Specific objectives include (1) determining microplastic concentrations in rainfall, catchment runoff, storage tanks, and rooftop filters; (2) characterizing polymer composition and particle morphology using micro-Fourier-transform infrared spectroscopy (?-FTIR) and scanning electron microscopy (SEM); (3) assessing spatial and temporal variation in microplastic loads across four urban districts with contrasting land-use and traffic densities; (4) evaluating the effectiveness of common RHS components (first-flush devices, mesh screens, and activated carbon or biological filters) in reducing microplastic carryover; (5) modeling predictors of microplastic concentration using multiple linear regression and generalized additive models (GAMs); and (6) discussing policy and design implications guided by the Theory of Planned Behavior and the Source–Pathway–Receptor framework. A mixed-methods approach combines quantitative environmental sampling with qualitative system assessments. The population comprises urban households and municipal RHS installations across four districts within a metropolitan area. A stratified random sampling design selects 40 households and 12 RHS facilities, ensuring representation of roof materials, storage volumes, and treatment trains. Water samples are collected seasonally over one year from rain events, catchments, tanks, and outlet points, with a target total sample volume of 600 liters per site per season. Laboratory analysis employs ?-FTIR for polymer identification and particle sizing, Raman spectroscopy for confirmatory analysis, and SEM for morphological characterization; polymer abbreviations include polyethylene (PE), polypropylene (PP), polystyrene (PS), and polyethylene terephthalate (PET). Quality control procedures include procedural blanks, triplicate analyses, and recovery tests using spiked samples. Data analysis proceeds in three stages descriptive statistics to report concentrations and mass loads, inferential statistics to examine differences by district, roof type, and treatment stage, and multivariate modeling to identify predictors of microplastic concentration. Regression analyses (multiple linear regression) estimate the influence of roof material, rainfall intensity, tank age, and filter type on microplastic loads, while GAMs capture nonlinear relationships with seasonal effects. A structural equation model (SEM) tests the hypothesized pathways from urban activities to microplastic presence in RHS, integrating the Source–Pathway–Receptor framework with Theory of Planned Behavior constructs to interpret user-implemented mitigation practices. Key expected findings include (i) microplastic concentrations increasing downstream from catchment to storage and outlet, with median loads ranging from 1.2×10^3 to 4.9×10^3 particles per liter in storage tanks during peak rainfall; (ii) predominance of microplastics in the 20–100 ?m range, with PE and PP as the dominant polymers, and irregular elongated morphologies indicating weathered fragments; (iii) significant effects of roof material (e.g., asphalt shingles vs. metal roofs) and first-flush device integrity on downstream microplastic loads; (iv) a substantial reduction in microplastic carryover associated with effective filtration regimes (mesh >100 ?m and activated carbon filters showing lower counts compared with basic screens); (v) nonlinear associations between rainfall intensity and microplastic concentration, modulated by storage age and maintenance status; and (vi) SEM results linking drivers identified in the SEM to behavioral determinants from the TPB, suggesting that user awareness and maintenance motivation are critical for system-level attenuation. The study contributes to knowledge by providing robust field-based evidence on microplastic transport in urban RHS, delineating the effectiveness of common mitigation strategies, and integrating environmental measurements with behavioral theory to inform design standards and policy guidelines. It offers practical recommendations for RHS designers and city planners, including standardized filtration specifications, routine maintenance protocols, and educational interventions to strengthen pro-environmental handling of RHS. The main conclusion is that while current RHS configurations can reduce microplastic entry to varying degrees, optimizing roof-site practices and implementing tiered filtration informed by empirical concentrations are essential to minimizing human and ecological exposure, with future research recommended to explore long-term health risk assessments and standardized monitoring frameworks across diverse climatic regions.