Groundwater provides 40–60% of California's water supply. However, domestic wellwater quality is not regulated in California due to its decentralized nature and high cost of testing. Many well water users rely on regional groundwater monitoring data (e.g. Groundwater Ambient Monitoring and Assessment) for information about their water quality but there remains a need for an integrated approach that combines place-based quantitative analysis with cancer risk assessment and community engagement. Our pilot study aims to build relationships and science literacy with private well water users through compensated community sampling of California well water and provision of quantitative reports including 22 inorganic constituents (e.g. lead, arsenic, nitrate) with primary and secondary MCLs. Additionally, the pilot study seeks to investigate the impacts of seasonality, land use, and climate change phenomena (e.g. wildfire, drought, floods) on well water quality. Participants (n=113) were recruited in collaboration with our community partner, the Environmental Justice Coalition for Water. Chemical analyses were performed at UC Davis and used to generate water quality reports and to assess potential public health risks and resources for remediation (see Figure A6 in Appendix). Of the 113 well samples tested, 14 (12.4%) had at least one constituent exceedance of Primary Maximum Contaminant Levels (MCLs), 74 (64.5%) had at least one constituent exceedance of Secondary MCLs, and all had at least one constituent exceedance of Public Health Goals (PHGs) and Notification Levels (NLs) (e.g. Lead (57/57), Uranium (28/57), Arsenic (12/57), Cadmium (11/57)). This study will help study participants assess the health risks associated with consumption of their well water and provides the infrastructure to scale up to a larger community-based groundwater quality monitoring program.

Climate change in California is driving an increase in the duration of the wildfire season as well as severity and size of wildfires. With the increased prevalence of mega wildfires, there is a need to understand the long-term effects of wildfire on the water quality of forested watersheds. After the 2021 Caldor Fire in the Sierra Nevada mountain range, a long-term, high-frequency surface water sampling campaign was implemented in the Cosumnes River Watershed alongside the mining of historical stream water chemistry and discharge data available through National Water Quality Monitoring Council and USGS databases. Here we develop a pipeline to integrate current measurements and historical water quality data to analyze in detail the variations in dissolved organic carbon exported from the watershed. For the purposes of this study, dissolved organic carbon was defined to be any organic matter present in stream water that could pass through a 0.45 µm PVDF filter. We hypothesized dissolved organic carbon content in the Cosumnes River would be dependent upon burn severity experienced and would reach a maximum in the first-year post fire. Our work demonstrated that storm events drive increases in DOC concentration and instantaneous loads post fire, with maximum concentrations observed in the first year. Storms in subsequent years continue to drive spikes in DOC concentrations and instantaneous loads but require higher discharges to achieve the levels observed in the first-year post fire. This analysis indicates the readily available organic matter supplied by years of drought followed by fire are depleted after the first year. Maximum DOC concentrations and instantaneous loads during high discharge storm periods are higher in the first-year post fire than in available pre-fire data. Atmospheric rivers are a strong driver of DOC transport and subsequent high concentrations. During the years of this study, snowmelt was not a driver of DOC transport. Our work suggests that the Cosumnes River Watershed is resilient in the face of wildfire, with dissolved organic carbon values returning to historical ranges in the second year after fire.

The formation of Fe(III) (oxyhydr)oxide precipitates from Fe(II) oxidation and/or Fe(III) hydrolysis is a key process that often governs the fate and bioavailability of contaminants and nutrients in natural and engineered systems. Electrocoagulation (EC) using Fe(0) electrodes is a low-cost water treatment technology that generates reactive Fe(III) precipitates in-situ via the electrolytic dissolution of an Fe(0) anode. EC-generated Fe(III) precipitates efficiently sorb contaminants, such as arsenic, and can then be separated from treated water by gravitational settling and/or filtration. The goal of this work is to determine the influence of electrolyte chemistry and EC operating parameters on the structure and reactivity of EC-generated Fe(III) precipitates.

This dissertation begins with an investigation of the structure and arsenic removal mechanism of Fe(III) precipitates generated by EC at a range of current densities in synthetic Bangladesh groundwater (SBGW). Shell-by-shell fits of the Fe K-edge extended X-ray absorption fine structure (EXAFS) spectra indicate that EC precipitates consist of nanoscale chains (polymers) of edge-sharing FeO6 octahedra. Shell-by-shell fits of As K-edge EXAFS spectra show that arsenic binds to EC precipitates in the binuclear, 2C corner-sharing geometry. When bound in this specific configuration, arsenic prevents the formation of FeO6 corner-sharing linkages. Phosphate and silicate oxyanions, abundant in SBGW, likely bind to EC precipitates in a similar configuration and contribute to the absence corner-sharing Fe-Fe linkages. The high extent of As(III) oxidation indicated by the As K-edge spectra is likely due to the reactive intermediates generated during EC treatment by the oxidation of Fe(II) by dissolved oxygen (i.e., Fenton-type reactions). Fe and As K-edge EXAFS spectra were found to be similar among samples generated at a large range of current density (0.02, 1.1, 5.0, 100 mA/cm2), suggesting this operating parameter does not play a major role in the structure of EC precipitates.

The next objective after characterizing EC precipitates generated in SBGW is to systematically investigate the individual and combined effects of major solutes (Ca2+, Mg2+, P, As(V), Si) on the structure and reactivity of EC-generated Fe(III) precipitates. Fe(II) oxidation in the presence of weakly adsorbing electrolytes (NaCl, CaCl2, MgCl2) produces pseudo-lepidocrocite (pseudo-Lp; γ-FeOOH), a poorly crystalline version of Lp with low sheet-stacking coherence. In the absence of bivalent cations, P and As(V) have similar uptake behavior, but different effects on the average Fe(III) precipitate structure: pseudo-Lp dominates in the presence of P, whereas a disordered ferrihydrite-like precipitate akin to hydrous ferric oxide (HFO) is the dominant phase that forms in the presence of As(V). Despite its lower affinity for Fe(III) precipitates, Si leads to Si-HFO under all conditions tested. The presence of 1 mM Ca2+ or Mg2+ enhances oxyanion uptake, destabilizes the colloidally stable oxyanion-bearing particle suspensions and, in some P and As(V) electrolytes, results in more crystalline precipitates. The trends in oxyanion uptake and Fe(III) precipitate structure in the presence of Ca2+/Mg2+ suggest a systematic decrease in the strength of bivalent cation:oxyanion interaction in the order of Ca2+>Mg2+ and P>As(V)>>Si.

In the third research chapter, the effect of the interaction between cations and oxyanions on the structure of EC-generated Fe(III) precipitates is examined in greater detail. The sequential formation of Fe(III) precipitates from Fe(II) oxidation in the presence of oxyanions leads to multiple mineral phases: short-ranged oxyanion-rich Fe(III) polymers with a large sorption capacity form at the onset of electrolysis followed by pseudo-Lp after oxyanions are depleted from the electrolyte. The larger As:Fe and P:Fe solids ratios in the presence of Ca2+ leads to less Fe required to deplete oxyanions from solution, and thus pseudo-Lp forms earlier in the electrolysis stage in Ca2+-containing electrolytes. Shell-by-shell fits of the As K-edge EXAFS spectra of all samples indicate no significant change in As(V) coordination in the presence or absence of Ca2+ and that As(V) binds primarily in the 2C binuclear, corner-sharing geometry. Structural models for the oxyanion-bearing Fe(III) polymers that form at the onset of electrolysis are proposed using constraints derived from batch sorption experiments, mobilization data, and molecular scale characterization techniques. According to these structural models, the role of Ca2+ in increasing the As:Fe or P:Fe solids ratios is two-fold: 1) "structural" Ca2+ participates in direct bonding with sorbed As(V) or P and increases oxyanion uptake via specific bonding interactions and 2) "electrostatic" Ca2+ associates with the solid in weaker, non-specific interactions that increase oxyanion uptake by minimizing electrostatic repulsion between oxyanions and the negatively charged Fe(III) precipitate surface.

The sustainable operation of EC systems in the field requires a rigorous understanding of the effects of EC operating parameters, such as current density, on EC system performance. Knowledge of the effects of individual and combined concentrations of major dissolved species on the local bonding environment, mineral phase, crystallinity, and mechanism of ion uptake of EC-generated Fe(III) precipitates will permit more accurate predictions of EC system performance across a diverse range of chemical matrices likely to be encountered in the field. This information also advances the understanding of contaminant and nutrient transport at natural redox boundaries such as hyporheic zones in soils and sediments.