Contamination of groundwater resources with naturally occurring arsenic (As) is a major health concern globally that affects millions of people. Experimental studies and geochemical modeling can be useful tools for understanding the mechanisms controlling speciation and partitioning of As, predicting the mobility and transfer of As, assessing the risks, and developing effective in-situ remediation schemes for affected areas.

The wide variety of experimental details used in the published adsorption studies, and limited number of studies conducted for adsorption of As on natural samples underscores the need for further research in this area. Application of currently available empirical models, and mechanistic surface complexation modeling (SCM) approaches for describing the adsorption of As on natural sediments is also challenging due to the complexity of the heterogeneous solid phases of porous media.

In Chapter 2, we compare different kinetic and equilibrium adsorption models for describing the adsorption data from batch experiments conducted on natural sediment samples collected from a contaminated site in New England, USA. We evaluate the capability and sensitivity of commonly used empirical models for modeling the concentrations of dissolved As as well as the relations between the derived model parameters and chemical/physical characteristics of sediments.

In Chapter 3, four different SCMs are developed to model the data from batch experiments conducted on natural sediment and groundwater samples. The SCMs used in this study vary in terms of 1) method for estimating the total number of sorption sites, 2) method for describing the surface site heterogeneity, and 3) method for applying electrostatic and pH-dependent correction factors to surface reaction constants. we utilize the easily-accessible modeling tools PHREEQC and FITEQL to predict and simulate the experimental results.

Chapter 4 focuses on batch and column experiments conducted to study the effectiveness of applying chemical amendments with the purpose of facilitating and enhancing natural attenuation of As in contaminated subsurface environments. The results will provide a qualitative basis for evaluating the suitability and efficiency of applying these methods in actual contaminated sites.

The goal of this study was to characterize the soil constitution and the presence of antibiotic resistance gene (ARG) erm(F) in soils across California to investigate the relationship between the levels of anthropogenic influence and ARG erm(F) levels locally and regionally. Soil samples were collected from 26 parks in four cities, San Diego, Los Angeles, Bakersfield, Fresno, where six publicly accessible parks sites were selected in each, and two pristine sites, Yosemite National Park and Mount Baldy. DNA was extracted, normalized, and tested for the presence and quantity of genes 16S rRNA and erm(F) using quantitative polymerase chain reaction (qPCR). Erm(F) quantities were determined via two analysis approaches—on a per gram of soil basis, and relative to 16S gene abundance. Locally among 26 parks, the five quantifiably detected erm(F) sites all occurred in high-level anthropogenic influenced sites. Regionally among four cities, urban and agricultural cities shared the similar soil constitution respectively and urban cities had higher erm(F) gene levels than agricultural ones. Population and population density in each city is positively correlated with erm(F) gene levels whichever analysis approach was applied.

Fecal contamination of coastal waters is known to degrade the environment and poses a health risk to recreational beach users. Fecal indicator bacteria (FIB) are used around the world to assess water quality and characterize fecal contamination. Elevated levels of FIB have been linked to health risks in epidemiological studies. However, some limitations exist with this indicator. FIB cannot be used to identify the specific sources as they originate from both human and animal sources. FIB may also persist and regrow in the environment. In order to effectively remediate the cause of pollution and characterize the hazards at chronically impaired beaches it is necessary to measure indicators that can provide information about the sources of the general fecal pollution. Tracking pollution sources at impaired beaches is critical to ensuring the health of coastal watersheds and reducing the incidence of swimming related illness. Molecular methods have gained popularity to identify and detect sources of fecal contamination using host-associated markers.

The work presented here addresses areas warranting further research in the state of the science of water quality monitoring. In Chapter 2, we demonstrate that host-associated markers exhibit similar limits of detection in different water types and are robust in environmental field applications. Additionally, we provide a cost-benefit analysis and provide water quality managers with information supporting the inclusion of molecular methods in current monitoring practices. This body of work also presents novel methods for rapid and viability-based detection of recent fecal contamination with propidium monoazide (PMA-qPCR) and a field portable method covalently-linked IMS/ATP technique (Cov-IMS/ATP). In Chapters 3 and 4, we present results on optimization and specificity of the Cov-IMS/ATP. We evaluated the performance of Cov-IMS/ATP at three different watersheds for rapid quantification of enterococci, and show this method to be a robust tool in assessing water quality at complex sites. This work also addresses drawbacks of traditional qPCR to quantify viable fecal contamination. We validate the PMA-qPCR method and demonstrate its performance in detecting recent fecal contamination in environmental waters. Use of these methods demonstrates a new framework that can enhance current microbial source tracking studies and water quality monitoring.

The rising level of antibiotic resistance worldwide is a critical public health challenge. Antibiotic resistance genes (ARGs) and mobile genetic elements allow for bacteria to confer resistance to antibiotics in as little as a year. Antibiotics utilized for growth promotion in confined animal feeding operations (CAFOs) generates manure that is an anthropogenic source of ARGs into the environment. In this study, 10 potting soils, 7 garden soils, 4 food amendments, 4 lawn amendments, 6 manure-based soils, 5 natural soils, 3 community soils, and 3 compost soils were surveyed for antibiotic resistance gene ermF, mobile genetic element intI1, and proxy for bacterial content 16S rRNA. The ARGs were quantified through quantitative polymerase chain reaction (qPCR) to per gram of fertilizer and per gene copies of 16S rRNA. Most soils contained detectable levels of ermF and intI1, ranging from 9.66 x 10-7 to 8.31 x 10-2 gene copies ermF/copies of 16S rRNA and 8.92 x 10-7 to 6.21 x 10-2 gene copies intI1/copies of 16S rRNA. Natural soils were significantly lower in ermF and intI1 than the other soil types. Natural soils ranged from 10-7 to 10-5 gene copies per copies of 16S rRNA while the other types ranged from 10-7 to 10-2 gene copies per copies of 16S rRNA. There was no strong correlation between intI1 and ermF.

Overall, recreational beach water quality remains an issue of concern in Southern California and across the globe. Many factors come into play when determining water quality, including physical issues such as the myriad sources that contribute pollution to the site and financial and political issues that control the way water quality is monitored and determined. Current national regulations require the monitoring of fecal indicator bacteria in order to determine recreational water quality. However, it is also important to identify biological and geographical sources of pollution to consistently impaired locations. A commonly applied approach to meet the goals of source identification is to sample sites that have been high in FIB for further study. A tiered approach such as this, however, assumes a correlation between FIB and the sources of interest in the watershed. The research described in this dissertation tests this assumption in two Southern California watersheds, Santa Monica Canyon and Ventura Harbor. In both cases, a tiered approach to sampling using FIB as a first tier to guide sampling would have failed to identify sources of human fecal pollution (as identified by the presence of the human-associated Bacteroides marker HF183).

Every watershed is a distinct environment that has different potential sources of bacteria and many factors contributing to the persistence of the bacteria. Rather than attempting to apply an indicator that has worked as a first tier in other watersheds, it would be better to have as a first tier an in-depth study of the watershed using historical data or local experts to provide information on the most likely sources of pollution in the watershed. Using this information it would be possible to design a study using FIB and one or more source-associated parameters to identify specific sources of pollution in the watershed. In addition, sampling FIB and other parameters such as HF183 allow the application of other microbial source tracking tools including indicator ratios and detection frequencies. Source identification studies do not necessarily have to be long-term to identify consistent sources of pollution. For example, within the first four months of sampling at Ventura, the increased frequency of detection of HF183 at the Marina Dock sample location was apparent, and a dry weather influx of HF183 was seen in the Keys channels.

In addition to the many sources of FIB to the environment such as storm drains, leaking sewers, and wildlife, there are important environmental reservoirs such as sand and seaweed that can foster FIB growth and persistence in the environment. As such, it is important to understand the effect of different factors on the ability of bacteria to survive and persist in these reservoirs. Microcosm experiments conducted during the course of this dissertation research found that in dry beach sand (0.1% moisture), the addition of moisture was detrimental to the survival of the indicators studied (General Bacteroidales, E. coli, and enterococci). While increased moisture was not always detrimental to bacterial survival, these results point to the ability of bacteria to persist for long periods of time in beach environments under in-situ conditions (including dry sand). These findings point to the importance of understanding the behavior of indicator bacteria populations that have evolved to survive in environmental conditions so that their potential impact on overlying or adjacent water quality can be better understood.

In summation, results from this research point to the importance of selecting indicators and sample locations that are most relevant to watershed concerns rather than using a first tier such as FIB to preferentially select sites for further analysis. Measuring a marker for human fecal pollution in both watershed studies provided useful information for potential human inputs that would have been missed if sites were chosen based on high FIB levels. In addition it is very important to understand the contribution of different reservoirs, such as sand, in the study area to the observed microbial pollution. Overall, these results point to the need for further examination of the ability of bacteria to survive under various environmental conditions in both water and sand, using both environmental microbial populations and populations from likely sources such as human sewage.

The presence of antibiotic resistant genes (ARGs) and antibiotic resistant bacteria (ARB) in aquatic environments is cause for concern. Research studies have linked animal and human waste as a source for ARGs and ARB found in stormwater runoff, agricultural runoff, wastewater treatment plants, and hospital effluents. Several of these pathways connect to the ocean, potentially creating another reservoir for ARGs. Coastal environments are relied on for recreational uses, food, and more, but the relationship between pathogens present in coastal environments and human health have not been fully established. Thus, quantifying the amount of ARGs present in ocean water, particularly in surface coastal waters where humans are more likely to encounter any harmful pathogens, is pivotal to understanding and mitigating the threat of antibiotic resistance in aquatic environments. This thesis is divided into two chapters. Chapter one is a systematic review for quantitative resistance data available for coastal surface waters. Chapter two investigates ARGs in ocean water at two California beaches.

Many PCR-based methods for microbial source tracking (MST) have been developed and validated within research laboratories, but require inter-laboratory validation before implementation. As part of the State of California Source Identification Protocol Project (SIPP), a blinded set of challenge filters were analyzed by three to five laboratories with a suite of PCR-based methods utilizing standardized methods. Repeatability (within lab agreement) and reproducibility (between lab agreement) of results were assessed by multiple metrics and compared to previously observed values for other environmental methods. Repeatability and reproducibility were found to be generally comparable to previously observed values for other methods (Median CV .001-.033 and .020-.106, respectively). Variance component analysis showed contribution of laboratory to total variability to be larger but of similar magnitude to inherent intra-laboratory variability among laboratories using standardized methods. Results among laboratories using non-standardized protocols for the same methods were also observed to have >2 log differences at times. These findings verify the repeatability and reproducibility of these MST methods and highlight the need for standardization of protocols and consumables prior to implementation of larger scale microbial source tracking studies involving multiple laboratories.

Microbial contamination of coastal waters is a global environmental and public health concern. However, monitoring and tracking of microbial contaminants may require technical expertise and may be costly and labor-intensive. Thus, to address this environmental issue in both developed and developing countries, current methods need to be cross-validated, and more accessible methods need to be proposed.

Therefore, this dissertation investigates the sources, fate, and transport of fecal indicator and antibiotic resistant bacteria in coastal watersheds. This dissertation utilizes culture-, molecular-, and satellite-based techniques to provide cross comparison. Additionally, this dissertation implements these techniques into course-based research experiences (CREs) for graduate students to assess whether CREs influence graduate students’ confidence and interest in research.

In chapter 1, a team of collaborators from University of California, Los Angeles (UCLA), the University of Pennsylvania, the Jet Propulsion Laboratory, and Health the Bay investigate oceanic responses to the 2018 Woolsey Fire using in-situ data of fecal indicator bacteria and satellite-derived data of turbidity. Chapter 2 examines commercially available garden products as sources of antibiotic resistance genes. In Chapter 3, researchers from UCLA and the Autonomous University of Baja California evaluate the effects of reclaimed water irrigation on antibiotic resistance gene levels in a coastal agricultural region in Mexico. Further, chapter 4 and 5 implement traditional and novel culture- and molecular-based methods to quantify antibiotic resistant bacteria and antibiotic resistance genes to examine the microbial burden hospital sewage discharges to the sewershed and to assess impacts of urbanization by comparing an urbanized watershed to an adjacent natural watershed. Lastly, chapter 6 incorporates these methods into a CRE for graduate students. Ultimately, this dissertation showcases the utility of equitable research methodologies in the field and the classroom.

With increased antibiotics usage over recent decades, there has been an emerging concern about the accelerated development of antibiotics resistance in the environment. This development poses several public health concerns, such as, the higher frequency of multiple drug resistant bacteria both in the environment and in clinical settings. Exposure to metals can increase antibiotic resistance through co-selective pressure; however, this non-traditional route of antibiotic resistance acquisition has not been extensively studied. A high-throughput culture based method was developed to assess the co-selective pressure of arsenic pre-exposure to oxytetracycline resistance in planktonic cultures of Escherichia coli O55. A glucose minimal salts (GMS) media was used for the study. As2O3 and oxytetracycline hydrochloride solutions were made ranging from 1-100 ppb and 10-40 ppm, respectively. Lastly, the geochemical equilibrium model PHREEQC was used to determine the bioavailable concentrations of the metals and antibiotics in the defined GMS media. The co-selective relationship was investigated by comparing the net growth rates of the cultures pre-exposed to varying levels of arsenic to the unexposed arsenic cultures at various concentrations of oxytetracycline.

Antibiotic resistance genes (ARGs) and antibiotic resistant bacteria (ARB) are an emerging threat to public health and are identified as contaminants in aquatic environments. This study investigated the effect of chlorine on the inactivation and reactivation of Escherichia coli K12 and its ampicillin resistance gene. Inactivation of K12 was achieved with exposure to chlorine for 10 minutes, which were then allowed to reactivate in the dark for 24 hours. Bacterial populations were enumerated using heterotrophic plate counts containing concentrations of ampicillin, which caused further inactivation. The results showed that nearly 100% of the bacterial population was inactivated by ampicillin regardless of chlorination. Upon reactivation, a sharp decrease in bacterial concentration is observed between the control (0 mg/L Ampicillin) and any ampicillin concentration in unchlorinated samples. However, after chlorination, there is a much less decrease, suggesting that bacteria that survived chlorination were more resistant to ampicillin. Both inhibitory concentrations (MIC50 and MIC90) were greater than 16 mg/L Amp for at 1 mg/L Cl2. At higher chlorine concentrations, the MIC90 shifted from 0-4 mg/L to 4-8 mg/L ampicillin.