Evaluating Microbial Water Quality and Potential Sources of Fecal Contamination in the Musconetcong River Watershed in New Jersey, USA

Microbial pathogens and indicators have contributed to major part of water quality degradation in the United States. Located in the northwestern New Jersey, the Musconetcong River has been included in the New Jersey Impaired Waters List or the 303(d) List due to high concentrations of fecal indicator bacteria. Hence, a Total Maximum Daily Load plan was established to address microbial water quality issues in the watershed. The objectives of this study were to assess the current status of microbial water quality and to determine potential sources of fecal contamination in the Musconetcong River Watershed using microbial source tracking techniques. Fifteen sampling events in total were carried out at nine sites throughout the Musconetcong River Watershed in August 2016, July and August 2017. E. coli enumeration was performed to determine the possible presence of fecal contaminations. Microbial source tracking techniques, specifically Canada goose, cow, deer, horse, and human-specific molecular markers, were used for real-time polymerase chain reaction (qPCR) analysis in order to identify and quantify potential sources of fecal contamination. The results indicated that E. coli was found present at all nine study sites. Two of the nine sites violated the New Jersey Surface Water Quality Standards in August 2016, while all of the nine sites exceeded the standards in both July and August 2017. Water temperature, dissolved oxygen (DO), and specific conductance at the study sites ranged from 13.5 ̊C to 25.3 ̊C, from 7.7 mg/L to 13.0 mg/L, and 278.5 μS/cm to 1335.0 μS/cm, respectively, at the time of sample collection. E. coli counts How to cite this paper: Hsu, T.-T.D., Lee, L.H., Rossi, A., Yussof, A., Lawler, N. and Wu, M.Y. (2019) Evaluating Microbial Water Quality and Potential Sources of Fecal Contamination in the Musconetcong River Watershed in New Jersey, USA. Advances in Microbiology, 9, 385-397. https://doi.org/10.4236/aim.2019.94023 Received: March 7, 2019 Accepted: April 22, 2019 Published: April 25, 2019 Copyright © 2019 by author(s) and Scientific Research Publishing Inc. This work is licensed under the Creative Commons Attribution International License (CC BY 4.0). http://creativecommons.org/licenses/by/4.0/ Open Access


Introduction
Microbial contaminations have contributed to a major part of water quality impairment in the United States [1], accounting for various drinking and recreational waterborne outbreaks in the US [2]. Sources of contamination were mostly attributed to fecal wastes originated from septic systems, livestock, domestic animals, and wildlife [3]. Conventionally, microbial water quality has been often evaluated using fecal indicator bacteria (FIB), because they are easy to measure as opposed to quantify each and every pathogen. Moreover, the enumeration results showed a clear linkage to fecal contaminations and potential presence of pathogenic microbes [4]. Common FIBs include total coliform, fecal coliform, Escherichia coli (E. coli), and enterococcus [4]. The US Environmental Protection Agency (EPA) has developed microbial water quality criteria using FIBs; in freshwater ecosystems, E. coli is the most commonly used FIB [4] [5]. The Clean Water Act requires states and tribal nations to assess waterbodies for water quality impairment. Once a waterbody is found impaired, the waterbody is listed on the State's Impaired Waters List or the 303(d) List and a Total Maximum Daily Load (TMDL) plan is developed as a planning tool to address the water quality impairment issues [6].
Once an area is identified to be impaired due to high FIB, the next step is to identify the sources of such contamination, so mitigation efforts can focus on eliminating the sources directly. Microbial source tracking (MST) provides the opportunity to identify potential sources of fecal contamination. The results of MST can be applied to aid in the development of a TMDL plan to better mitigate the water quality deterioration [7]. MST approaches include both phenotypic and genotypic methods. Phenotypic typing requires observation of physical and  [12]. Historical documents showed that bacterial water quality in Musconetcong River frequently exceeded water quality standards, failing to support primary contact recreation [13] [14]. In 2003, a TMDL plan for fecal coliform were established to address water quality impairment in the Musconetcong River, requiring 93% reduction of fecal coliform at multiple locations along the River [15]. Since then, various restoration efforts have been undertaken to reduce fecal contamination including implementation of new riparian buffers, erosion control, sinkhole closure and green infrastructure [12]. The objectives of this study were to assess microbial water quality and to determine potential sources of fecal contamination post restoration implementation in the Musconetcong River Watershed. Results from this study would improve understanding the status of microbial water quality and potential sources of contamination in the Musconetcong River Watershed and assist stakeholders to better control the fecal contaminations.

Field Sampling
Nine study sites were selected throughout the Musconetcong River Watershed, New Jersey, based on past water quality monitoring results and local knowledge (Table 1 and Figure 1). Surface water grab samples were collected aseptically

Microbial Source Tracking
concentrations of DNA 6.0221 10 molecules mole Copy number basepairs 660 g mole 1 10 ng g .
For each species, ten-fold serial dilutions of target species genes with predetermined copy numbers were used to generate standard curves to quantify copy numbers of target species genes from total environmental DNA extract. Advances in Microbiology tion (CV) among the triplicate samples must be within 15%. A sample with copy numbers higher than the respective detection limit was given a calculated copy number (expressed as copy numbers per 100 mL of water samples). A sample with no detection signal or with copy numbers lower than respective detection limits was absent of fecal markers and a value of zero was assigned for further analysis. A sample with calculated copy number lower than the lowest concentration of standard was considered containing a minimal amount of fecal markers with higher uncertainty for quantitative analysis; therefore those results were excluded from quantitative analysis as suggested by Helsel [22].

Statistical Analysis
Statistical analysis was performed using R language (Version 3.4.4 [23]) using RStudiosoftware (Version 1.0.44 [24]). Shapiro-Wilk test was used to assess the normality of the data distribution [25]. If variables were found to not follow normal distributions, nonparametric Spearman's correlation was carried out to test associations between E. coli and other parameters.

Field Indicator Bacteria
The Musconetcong River and its tributaries were classified as  This study also recorded water temperature, dissolved oxygen and specific conductance. Water temperature, dissolved oxygen (DO), and specific conductance at the study sites ranged from 13.5˚C to 25.3˚C, from 7.7 mg/L to 13.0 mg/L, and 278.5 μS/cm to 1335.0 μS/cm, respectively, at the time of sample collection. Correlation analysis of the study results (Table 3) demonstrated E. coli counts were found negatively correlated with temperature (r = −0.41, p < 0.05), negatively correlated with specific conductance (r = −0.40, p < 0.05), but positively correlated with dissolved oxygen (r = 0.25, p < 0.05), suggesting higher FIBs at water with a higher dissolved oxygen content, a lower temperature and a lower specific conductance during the summer months at the study sites. Although negatively temperature-dependent E. coli survival patterns have been well reported in a variety of water sources, including river and streams [28], the results of the above correlation tests are likely to underestimate the complexity of the river ecosystem.

Microbial Source Tracking
Microbial contaminations can originate from various sources, such as leaking sewer lines, failing septic systems, domestic animals, livestock, wildlife, and stormwater runoff [3]. Arnone and Walling have summarized that a variety of fecal indicator bacteria, including E. coli, Enterococcus, or fecal coliforms, were  best correlated with swimming-associated gastrointestinal illness in many epidemiological studies worldwide, including sewage or stormwater [2]. For example, a failing wastewater treatment facility caused a waterborne outbreak in Lake Erie, USA, with 1450 gastroenteritis cases reported [29]. On the other hand, fecal materials can also come from wildlife and livestock sources (i.e. Canada geese, deer and cow) and could harbor true pathogens [27] [33]. Deer is another wildlife species that has become a pest and a major concern in the U.S. urban and sub-urban communities. In addition to the notorious Lyme disease transmitted to human through the bite of deer ticks, deer feces also harbor a major type of STEC, E. coli O157:H7, which was believed to cause a foodborne outbreak through the consumption of strawberry in 2011 [32]. STEC have also been isolated from cattle feces, which is recognized as the most important natural reservoir for STEC [31].
MST was conducted to identify and quantify potential sources of fecal contamination with an ultimate goal to reduce inputs of fecal pollution. MST was previously used to assess potential sources of fecal contamination in the Musconetcong River Watershed. Unfortunately, only presence or absence of human and bovine fecal markers were included without quantitative results [12]. This study examined and quantified human and bovine as well as three additional markers, Canada goose, deer, and horse, at 9 study sites (both main stem and tributaries) throughout the Musconetcong River Watershed with a goal to discriminate specific sources of fecal contamination at each location. These potential sources were selected with an intention to provide data driven recommendation to aid future water quality management efforts. Different sampling locations may have distinct adjacent land use practices and may require individually designed restoration plan instead of a cookie-cutter approach. Site-specific results would enable stakeholders to develop a site-specific restoration plan targeting specific source(s) of contamination.  Canada geese) were the major sources of fecal contamination among the five markers tested at the study sites; however, sporadic fecal contributions from cow and horse were also substantial.

Conclusion
This study documented a high frequency of bacterial water quality standard violations at the selected nine study sites within the Musconetcong River Watershed. However, this result only reflects temporary status of microbial water quality at the time of water sample collection since the study only encompassed five sampling events within one month in 2016 and ten within two months in 2017. Future study incorporating a more frequent and longer-term sampling scheme is recommended to further confirm the status of bacterial water quality.
Additionally, this study only included a pair of dry and wet weather events in each year for microbial source tracking to identify sources of fecal contamina- Advances in Microbiology tion. A more frequent and longer-term sampling scheme covering more dry and wet weather events would also help elucidate the impacts of precipitation on study results, especially under the current trend of extreme weather events, as deteriorated water quality is often related to wet weather [34]. Nevertheless, this study has laid the groundwork for examining the bacterial water quality and demonstrating the usefulness of microbial source tracking when determining specific source(s) of fecal contamination. The results of this study will enable environmental managers in the Musconetcong Watershed to identify the best management practices most suited to control the specific fecal contamination identified.