Genetic fecal source identification in urban streams impacted by municipal separate storm sewer system discharges

Adam Diedrich; Mano Sivaganesan; Jessica R. Willis; Amirreza Sharifi; Orin C. Shanks

doi:10.1371/journal.pone.0278548

Abstract

Municipal stormwater systems are designed to collect, transport, and discharge precipitation from a defined catchment area into local surface waters. However, these discharges may contain unsafe levels of fecal waste. Paired measurements of Escherichia coli, precipitation, three land use metrics determined by geographic information system (GIS) mapping, and host-associated genetic markers indicative of human (HF183/BacR287 and HumM2), ruminant (Rum2Bac), dog (DG3), and avian (GFD) fecal sources were assessed in 231 urban stream samples impacted by two or more municipal stormwater outfalls. Receiving water samples were collected twice per month (n = 24) and after rain events (n = 9) from seven headwaters of the Anacostia River in the District of Columbia (United States) exhibiting a gradient of impervious surface, residential, and park surface areas. Almost 50% of stream samples (n = 103) were impaired, exceeding the local E. coli single sample maximum assessment level (410 MPN/100 ml). Fecal scores (average log₁₀ copies per 100 ml) were determined to prioritize sites by pollution source and to evaluate potential links with land use, rainfall, and E. coli levels using a recently developed censored data analysis approach. Dog, ruminant, and avian fecal scores were almost always significantly increased after rain or when E. coli levels exceeded the local benchmark. Human fecal pollution trends showed the greatest variability with detections ranging from 9.1% to 96.7% across sites. Avian fecal scores exhibited the closest connection to land use, significantly increasing in catchments with larger residential areas after rain events (p = 0.038; R² = 0.62). Overall, results demonstrate that combining genetic fecal source identification methods with GIS mapping complements routine E. coli monitoring to improve management of urban streams impacted by stormwater outfalls.

Citation: Diedrich A, Sivaganesan M, Willis JR, Sharifi A, Shanks OC (2023) Genetic fecal source identification in urban streams impacted by municipal separate storm sewer system discharges. PLoS ONE 18(1): e0278548. https://doi.org/10.1371/journal.pone.0278548

Editor: Alison Parker, Cranfield University, UNITED KINGDOM

Received: August 30, 2022; Accepted: November 17, 2022; Published: January 26, 2023

This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.

Data Availability: Data will be available to the pubic at DOI: 10.23719/1528293.

Funding: The author(s) received no specific funding for this work.

Competing interests: The authors have declared that no competing interests exist.

Introduction

An estimated 55% of the world population lives in urban areas [1]. Urban areas are built environments composed of elements such as homes, buildings, green spaces, streets, and sidewalks. These areas typically include elaborately engineered drainage networks collecting and directing stormwater runoff into local streams, often requiring minimal rain accumulation. Discharge from these municipal separate stormwater systems may contain unsafe levels of fecal pollutants originating from various sources ranging from illicit sewer connections to wildlife waste. Numerous studies report that rivers and streams that flow through urban areas are often impacted by fecal pollution to a greater extent compared to respective upstream, less populated locations [2, 3]. In addition, there is a large body of research identifying stormwater runoff as a vector of human pathogens [4, 5] and as a key contributor of fecal pollution [6–8] in urban waters. Urban stream microbial water quality is typically monitored with general fecal indicator bacteria, such as Escherichia coli, which provides information on the estimated total level of fecal pollution in a sample as well as indication of potential public health risk [9]. However, these measurements do not provide information on specific fecal pollutant sources. This limitation is often more pronounced in urban areas due to dense human, pet, and wildlife populations living near each other in areas with a high proportion of impervious surfaces (i.e., streets, sidewalks, roof tops), facilitating the rapid movement of stormwater and microbial contaminants from the city landscape into nearby natural receiving waters.

In response, the research community has developed numerous quantitative fecal source identification methodologies to compliment routine general fecal indicator bacteria monitoring offering the option to characterize fecal source pollution trends from key animal groups and potential public health risk results in parallel [10–14]. Many of these methods have been successfully used to characterize fecal pollution sources in natural waters impacted by rainfall [7, 15–17], but mostly focusing on human and/or dog waste contamination. However, less is known about the potential influence of stormwater outfall likely containing fecal matter from urban wildlife. Many cities are home to ducks, geese, and other waterfowl representing potential sources of general fecal indicators [18]. The GFD avian-associated fecal source identification method is reported to detect a broad range of waterfowl and other bird species [19, 20]. In addition, ruminant-associated methods can be used to track wildlife when there is confidence that sampling sites of interest are not impacted by agricultural ruminants (i.e., cattle, goats, sheep), such as urban catchments containing headwaters where animal feeding operations are typically not found. The Rum2Bac method is reported to be highly specific [21–23] and shown to detect at least 14 ruminant sources including wildlife species such as white-tail deer, mule deer, among others [23]. Together these methods provide an opportunity to assess human, dog, and wildlife impacts on urban streams impacted by stormwater outfall discharges.

This study evaluates paired measurements of host-associated genetic markers, E. coli, and precipitation from seven headwater urban streams situated at the southern portion of the Anacostia River Watershed (District of Columbia, U.S.A.) over a 13-month period. Unlike most highly urbanized areas, the District of Columbia is composed of approximately 27.9 km² of park land, further complicating management due to the presence of indigenous avian and ruminant wildlife populations. In addition, each stream is impacted by two or more municipal stormwater outfalls and stream catchments that represent a gradient of land use based on geographic information system (GIS) measurements of impervious surface, residential, and park surface areas. A panel of five host-associated genetic markers designed to characterize human (HF183/BacR287 and HumM2), dog (DG3), avian (GFD), and ruminant (Rum2Bac) fecal sources are employed to characterize fecal pollution trends. Potential links between fecal source identification measurements, E. coli, precipitation, and land use are explored with a recently developed Bayesian qPCR censored data analysis approach [24] that has been successfully used to rank recreational water sites based on human fecal pollution levels [24, 25] and identify links between fecal indicator bacteria levels and host-associated genetic marker measurements [17, 26]. Findings indicate that human waste and urban wildlife both contribute to water quality in the District of Columbia and demonstrates that combining fecal source identification with GIS mapping and censored data analysis complements routine E. coli monitoring to improve water quality management in urban areas.

Materials and methods

Catchment and sampling site descriptions

The Anacostia River Watershed is a 456 km² area situated in the District of Columbia and Maryland (United States). Seven catchments from the southern portion of the watershed located in the District of Columbia city limits were selected for water quality testing including Fort Chaplin, Fort Dupont, Pope Branch, Fort Davis, Texas Run, Alger Park, and Fort Stanton (Fig 1). Catchments ranged in size from 0.13 (Alger Park) to 1.23 (Fort Dupont) km² with a gradient of land use practices ranging from park land to residential. Catchment boundaries were defined with the Spatial Analysis Hydrology tool using stream and elevation data from the National Hydrology Dataset (http://datagateway.nrcs.usda.gov) combined with local stormwater transit pipe data (Department of Energy and Environment, Washington, DC). There are no septic systems or combined sewer overflow discharge points in catchments, all human waste is managed via centralized sewage systems. Each catchment contains two (Fort Dupont and Texas Run) to seven (Fort Davis) stormwater discharge outlets (Fig 1). Sampling sites were strategically located at the catchment outlet point where a respective headwater stream intersects the catchment boundary (Fig 1).

Download:

Fig 1. Geographic information system (GIS) map showing size and location of seven catchments included in study.

Sampling locations are denoted with a red symbol. The rain gauge is indicated by a black shaded square. Catchments are colored based on total impervious surface area (km²). See legend for stream and catchment boundary information. The underlaying street map is from OpenStreetMap (Data available under the Open Database License: https://www.openstreetmap.org/copyright).

https://doi.org/10.1371/journal.pone.0278548.g001

Precipitation and land use characterization

The delivery and magnitude of fecal pollutants into receiving streams can be strongly influenced by local rainfall, especially in catchments with multiple stormwater drainage points [27, 28]. To measure precipitation, a reference gauge (WH24B Wireless Weather Station, Fine Offset Electronics, Ltd) was selected based on proximity to catchments and completeness of hourly measurements over the study period (Fig 1). Due to the nature and impact of precipitation events on urban catchments, cumulative precipitation 12h prior to a sampling event (mm) was used for all statistical analyses. ArcGIS ArcMap (version 10.3; ESRI, Redlands, CA) was employed to define catchment land use metrics including areas of impervious surface, park, and residential (km²) using the local management authority shape files (Department of Energy and Environment, Washington, DC).

Water sampling

A total of 231 water samples were collected across 33 sampling events over 57-weeks from November 14, 2019 to December 15, 2020. Routine sampling events occurred twice per month. To increase the number of samples impacted by recent storms, seven additional sampling events were conducted immediately after cumulative rainfall from a single storm exceeded 2.54 mm following a 72 h dry period based on United States Environmental Protection Agency recommended storm event sampling guidance [29]. For each sampling event (total count = 33), water was collected from all sites on the same day within a short period of time (< 6 h). Samples were collected in sterile 1L containers from surface water and immediately stored on ice during transport to the laboratory. In addition, a field blank sample with molecular grade water substituted for receiving water was included for each sampling event.

E. coli measurements

E. coli enumeration was performed within eight hours of sample collection with Colilert IDEXX defined substrate technologies (IDEXX Laboratories, Inc. Westbrook, ME). To assess water quality trends, a single sample maximum concentration of 410 most probable number (MPN) per 100 ml was used based on District of Columbia Department of Energy and Environment practices [30].

Reference DNA preparations

Reference DNA sources consisted of two plasmid constructs and salmon sperm DNA. Plasmid constructs included the National Institute of Standards and Technology Standard Reference Material^® 2917 (SRM 2917; Rockville, MD) for qPCR calibration model generation [31, 32] and an internal amplification control (IAC; Integrated DNA Technologies, Coralville, IA). SRM 2917 is a commercially available, ready to use, linearized plasmid consisting of five dilution preparations as follows: Level 1 (10.3 copies/2 μL), Level 2 (1.11·10² copies/2 μL), Level 3 (1.06·10³ copies/2 μL), Level 4 (1.06·10⁴ copies/2 μL), and Level 5 (1.04·10⁵ copies/2 μL). The IAC plasmid was linearized by ScaI-HF restriction digest (New England BioLabs, Beverly, MA), purified via QIAquick PCR Purification Kit (Qiagen, Valencia, CA), quantified with a Qubit dsDNA HS assay kit on a Qubit 3 Fluorometer (Thermo Fisher Scientific, Grand Island, NY), and diluted in 10 mM Tris 0.5 mM EDTA (pH 9.0) to generate 10² copies/2μL. A salmon sperm DNA working stock containing 10 μg/ml was prepared by diluting a commercially available 10 mg/ml solution (Sigma-Aldrich, St. Louis, MO). All reference DNA material preparations were stored in low-adhesion microtubes either at 4°C (SRM 2917) or -20°C (IAC).

Water filtration and DNA extraction

A total of 231 water samples were filtered for DNA extraction. For each sample, 100 ml was filtered through a 0.45 μm polycarbonate filter (Fisher Scientific, Pittsburg, PA). Filters were placed in a sterile 2 ml screw cap tube containing silica bead mill matrix (GeneRite, North Brunswick, NJ) and stored at -80°C (< 18 months) until DNA extraction. DNA extraction was performed with the DNA-EZ RW02 kit (GeneRite) as previously described [25]. Briefly, 600 μL of 0.02 μg/ml salmon sperm DNA (Sigma-Aldrich) was added to each bead mill tube followed by homogenization with a MP FastPrep-24 (MP Biomedicals, LLC Solon, OH) at 6.0 m/s for 30 s. Three method extraction blanks (MEB), with molecular grade water substituted for receiving water sample, were performed for each sample processing batch (12 samples/batch). DNA was eluted with 100 μL elution buffer into GeneMate Slick low-adhesion microcentrifuge tubes (ISC BioExpress, Kaysville, UT) and stored at 4°C prior to qPCR amplification (< 24 h).

qPCR amplification

Fecal source identification employed five host-associated qPCR assays targeting four animal groups; namely human (HF183/BacR287 and HumM2), ruminant (Rum2Bac), dog (DG3), and avian (GFD) fecal sources [19, 21, 33–35]. The Sketa22 qPCR assay was also included as a sample processing control (SPC) [36]. All reactions contained 1X TaqMan Environmental Master Mix (version 2.0; Thermo-Fisher Scientific), 0.2 mg/ml bovine serum albumin (Sigma-Aldrich), 1 μM each primer, and 80 nM 6-carboxyfluoroscein (FAM)-labeled probe, and 80 nM VIC-labeled probe (HF183/BacR287 and HumM2 assays only). All reactions contained either 2 μL of DNA water sample extract or SRM 2917 in a total reaction volume of 25 μL. HF183/BacR287 and HumM2 multiplex reactions also contained 10² copies of IAC template. Triplicate reactions were performed for SRM 2917 containing reactions and all Sketa22 testing while all fecal source identification experiments included six replicate reactions. Amplifications were conducted on a QuantStudio 6 Flex Real-Time PCR System (Thermo Fisher Scientific) with the following thermal cycle profile: 10 min at 95°C followed by 40 cycles of 15 s at 95°C and 1 min at 60°C. The threshold was manually set to either 0.03 (HF183/BacR287, DG3, Rum2Bac, and Sketa22) or 0.08 (HumM2 and GFD). Quantification cycle (C_q) values were exported for further analysis. To monitor for potential extraneous DNA contamination during qPCR amplification, six no-template controls (NTC) with molecular grade water substituted for template DNA were performed with each instrument run.

qPCR data acceptance metrics

The following data acceptance scheme was used to ensure high quality molecular data in this study. All fecal source identification qPCR assays were subject to calibration model acceptance criteria for linearity (R² ≥ 0.980) and amplification efficiency (0.90 to 1.10, E = 10^(-1/slope) -1) [37, 38]. HF183/BacR287 and HumM2 multiplex IAC protocols were used to monitor for amplification inhibition in each water sample [34, 35]. Any DNA extract with evidence of amplification inhibition was discarded from the study. Instrument run-specific IAC proficiency testing (HF183/BacR287 and HumM2 NTC VIC C_q standard deviations ≤ 1.16 or 1.05, respectively) were also conducted to confirm reliable application of amplification inhibition testing from one instrument run to another [39]. A SPC protocol was implemented to monitor for suitable DNA recovery for each receiving water sample and to ensure consistent DNA recovery from one sample batch to another. Samples with unacceptable DNA recovery were excluded from the study based on batch-specific criteria derived from repeated MEB spike recovery measurements. SPC proficiency was determined for each sample batch requiring a standard deviation in Sketa22 qPCR MEB repeated measures of ≤ 0.62 C_q [39].

Calculations and statistics

qPCR calibration model and performance.

‘Mixed’ calibration models were generated for each fecal source identification qPCR assay and instrument run combination using a Bayesian Markov Chain Monte Carlo approach [40]. The lower limit of quantification (LLOQ) was defined as the instrument run specific 95% credible interval upper-bound from repeated measures (n = 18) of SRM 2917 dilution level 1. Outliers were defined as the absolute value of a studentized residual > 3.

Fecal score model.

The fecal score approach [24], a qPCR censored-data method was used to rank sites by fecal pollution source and investigate potential linkages between host-associated genetic marker measurements, rainfall, and E. coli levels with the following modifications. A fecal score is a host-associated genetic marker weighted average concentration [log₁₀ copies per 100 ml with 95% Bayesian credible interval (BCI)] calculated from a defined set of sample results including non-detections (C_q = 40, ND), detections below the LLOQ (below detection, BD), and measurements within the range of quantification (ROQ). For each set of water samples, C_q measurements were classified into three groups: ROQ group, BD group, or ND group. For the BD group, a Poisson distribution with a mean parameter λ₁ (1 ≤ λ₁ ≤ UB) was assumed for the number of genetic marker copies Z₁, where the upper-bound (UB) was the number of copies corresponding to the respective qPCR assay LLOQ. For the ND group, the number of copies was assumed to be less than one and governed by a Poisson distribution with mean λ₂ (λ₂ < 1). For the ROQ group, a ‘mixed’ calibration model [40] was used to estimate the number of copies Z₃ (log₁₀ base). For a given sample Y from the i^th instrument run, posterior distributions of Z₁, Z₂ and Z₃ were used to estimate the concentration X (log₁₀ base). The Bayesian models used for Z₁, Z₂ and Z₃ are given below:

For C_q measurements in the BD group,

Z₁ ~ Poisson (λ₁) · I (1,UB_i)

UB_i =

log₁₀ (λ₁) ~ N (0, 10³) · I (0,).

For C_q measurements in the ND group,

Z₂ ~ Poisson (λ₂) · I (0, 1)

log₁₀ (λ₂) ~ N (0, 10³) · I (-2, 0).

For C_q measurements in the ROQ group,

log₁₀ (Z₃) =

C_qr~ N (, s²).

A weighted-average fecal score is then given in Eq 1, (Eq 1) where r_1, r₂ and r₃ are the number of C_q replicate measurements in each BD, ND and ROQ group, respectively, α_i is the intercept parameter for i^th instrument run and is the master slope parameter of the mixed model calibration curve, is the mean and s is the standard deviation of derived from all measurements in the ROQ group. In addition, I(a,b) restricts the Poisson and Normal distributions to a range from a to b and I(0,) restricts these distributions to positive values. For given samples Y₁, Y₂,…,Y_m, the posterior distribution of Fecal Score can be used to estimate the mean and the corresponding 95% BCI, where (Eq 2)

For each water sample, six replicates were tested to ensure an adequate number of observations for suitable fecal score estimation [24]. A sample group was eligible for fecal score determination if it had 1) at least three samples (3 samples x 6 replicates = 18 data points), and 2) at least one replicate per sample must be from the BD or ROQ group (data sets consisting of all ND were not eligible).

To rank sites, fecal scores were calculated by site for each fecal source identification genetic marker data set using all samples. To investigate the potential influence of E. coli and precipitation levels on fecal pollution source trends, water samples from each site were then organized into two groups prior to fecal score determination. For E. coli, groups included: 1) samples with E. coli < 410 MPN/100 ml and 2) samples with E. coli ≥ 410 MPN/100 ml. For precipitation, groups included: 1) 12 h cumulative precipitation < 2.54 mm, and 2) 12 h cumulative precipitation prior to sampling ≥ 2.54 mm.

Land use trend analysis.

Correlation analyses were performed between land use metrics [impervious surface, park, and residential areas (km²)] and site ranking fecal scores (all samples from a site) or fecal scores generated from E. coli and precipitation sample groupings. Correlation analyses were only determined for combinations yielding eligible fecal scores for all sites. A significant land use trend was defined as any correlation combination yielding an R² ≥ 0.6 with p ≤ 0.05. All statistics were conducted with SAS software (Cary, NC) or WinBugs (Version 1.4.3; https://www.mrc-bsu.cam.ac.uk/software/bugs/).

Results

Precipitation

Precipitation was determined for each site and sampling event using a rain gauge in close proximity to the study area (Fig 1). Precipitation measurements ranged from 0 mm to 57.4 mm (Fig 2) with 41.6% of samples designated as rain event samples (n = 96; ≥ 2.54 mm).

Download:

Fig 2. Precipitation measurements (n = 231) ranked from lowest to highest quantities (values increase from left to right).

The vertical dashed line denotes number of samples with precipitation level < 2.54 mm.

https://doi.org/10.1371/journal.pone.0278548.g002

Land use metrics

Impervious surface, park, and residential area land use metrics were determined by GIS to investigate potential links to fecal pollution source trends (Table 1). Site catchments ranged in size from 0.13 km² (Alger Park) to 1.23 km² (Fort Dupont). Impervious surface and park areas exhibited a strong positive correlation (R² = 0.92). A gradient of impervious surface area was observed spanning 0.04 km² (Alger Park) to 0.22 km² (Fort Chaplin) (Fig 1). In contrast, the largest park area was observed at Fort Dupont (1.23 km²) followed by Pope Branch (0.80 km²).

Download:

Table 1. Summary of land use measurements (km²) for each catchment.

https://doi.org/10.1371/journal.pone.0278548.t001

E. coli measurements

E. coli MPN/100 ml were determined from 231 water samples collected over a 57-week period from seven sampling sites in the Anacostia River Watershed (Fig 1). Log₁₀ E. coli MPN/100 ml summary statistics are shown in Table 2. Measurable levels of E. coli were present in all samples except one (Fort Chaplin, February 18, 2020) ranging from 0 to 4.74 log₁₀ MPN/100 ml across sites. A total of 103 samples (44.6%) exceeded the local single sample maximum value of 410 MPN/100 ml (Fig 3).

Download:

Fig 3. Samples that exceeded the local single sample maximum of E. coli 410 MPN/100 ml.

https://doi.org/10.1371/journal.pone.0278548.g003

Download:

Table 2. Log₁₀ E. coli (MPN/100 ml) measurements by site.

https://doi.org/10.1371/journal.pone.0278548.t002

qPCR quality controls and data acceptance

A series of quality controls were used to identify high-quality fecal source identification qPCR data. Calibration model performance parameters are shown in Table 3. Extraneous DNA control reactions indicated no evidence of contamination (n = 4,275 reactions). Amplification inhibition was not identified in any multiplex IAC HF183/BacR287 or HumM2 experiments. Instrument run-specific IAC proficiency testing yielded a 100% pass rate with NTC VIC C_q standard deviations ranging from 0.07 to 0.75 for HF183/BacR287 (acceptance criteria ≤ 1.16) and 0.10 to 0.33 for HumM2 (acceptance criteria ≤ 1.05) [39]. Five water sample DNA extracts failed SPC testing (2.2%; 5 of 231) exhibiting unsuitable matrix interference and were discarded from the study. All instances were from the same site (Texas Run) between February 25 to May 14, 2020. A SPC proficiency test was used to monitor for suitable DNA recovery for each extraction batch. Sketa22 MEB C_q standard deviations ranged from 0.03 to 0.58 across 39 batch preparations (acceptance criteria ≤ 0.62) [39].

Download:

Table 3. Calibration model performance metrics.

https://doi.org/10.1371/journal.pone.0278548.t003

Host-associated qPCR measurements

Host-associated genetic markers were quantified for eligible water samples (n = 226) using fecal source identification qPCR methods for human (HF183/BacR287 and HumM2), dog (DG3), avian (GFD), and ruminant (Rum2Bac). Estimated mean log₁₀ copies per reaction concentrations are shown in Fig 4. The number of samples where all replicate reactions yield a 40 C_q (non-detection) ranged from 79 (34.2%, GFD) to 158 (68.4%, HumM2).

Download:

Fig 4. Host-associated qPCR genetic marker estimated log₁₀ copies per reaction concentrations for HF183/BacR287, HumM2, DG3, GFD, and Rum2Bac.

Heatmap keys are shown for each qPCR assay data set indicating estimated log₁₀ copies per reaction color coding information. Sampling date and site combinations denoted by a white cell represents samples that failed the sample processing control and were discarded from the study. Cells shaded grey indicates samples where all sample replicate reactions yielded a C_q = 40 (non-detection).

https://doi.org/10.1371/journal.pone.0278548.g004

Host-associated genetic marker fecal scores

For each host-associated genetic marker, site ranking fecal scores (log₁₀ copies per 100 ml with 95% BCI) are shown in Fig 5. The Fort Chaplin site ranked in the top three for human, dog, and ruminant fecal sources. In contrast, Texas Run fecal scores were in the bottom two rankings for avian, human, and ruminant waste. The Fort Stanton site had the highest levels of ruminant fecal waste significantly higher than all other sites (95% BCI do not overlap).

Download:

Fig 5. Fecal scores (log₁₀ copies per 100 ml with 95% BCI) for site rankings with each host-associated genetic marker.

Values are ranked from highest (left) to lowest (right).

https://doi.org/10.1371/journal.pone.0278548.g005

Fecal scores for each site using E. coli (threshold: ≥ 410 MPN/100 ml) and precipitation (threshold: ≥ 2.54 mm accumulation 12h prior to sampling) sample group definitions are shown in Fig 6. Overall, there was a high level of agreement between E. coli and precipitation sample group fecal score trends. Avian, dog, and ruminant fecal scores were always higher after rain or when E. coli levels exceeded the local management benchmark. When E. coli was low (< 410 MPN/100 ml) or there was no rain, 25 of 140 combinations (17.9%) were not eligible for fecal score determination due to high levels of non-detections in sample groups. This trend was most pronounced with ruminant fecal scores where the Rum2Bac genetic marker was almost always undetectable in 71.4% (9 of 14) of site E. coli and precipitation sample group combinations. Human fecal scores exhibited the most variable trends. HF183/BacR287 and HumM2 were rarely detected at Pope Branch and Texas Run in the absence of rain and when E. coli levels were low. In contrast, human-associated fecal scores from Fort Chaplin, Fort Stanton, Fort Dupont, Alger Park, and Fort Davis indicate the presence of measurable levels of human waste regardless of E. coli or precipitation grouping. Finally, the Fort Stanton site indicated no significant difference in human-associated fecal scores (95% BCI overlap) for either sample group definition.

Download:

Fig 6. Host-associated genetic marker fecal scores (log₁₀ copies per 100 ml with 95% BCI) for each site using E. coli.

(Panel A; threshold: ≥ 410 MPN/100 ml) and precipitation (Panel B; threshold: ≥ 2.54 mm) sample group definitions. An ‘X’ denotes that a sample grouping was ineligible for fecal score determination.

https://doi.org/10.1371/journal.pone.0278548.g006

Land use trend analyses

Two strategies were used to investigate potential links between host-associated qPCR measurements and catchment land use metrics [park, residential, and impervious surface areas (km²)]. First, fecal scores generated from all samples at each site indicated no significant correlation to a land use metric (p ≥ 0.06; R² ≤ 0.375). Second, samples from each site were then organized into two groups based on E. coli (≥ 410 MPN/100 ml) or 12h cumulative precipitation (≥ 2.54 mm) threshold definitions. Only one combination yielded a significant correlation: avian-associated fecal scores (precipitation group = samples ≥ 2.54 mm) and residential area (km²) (p = 0.038; R² = 0.612). All human-, dog-, and ruminant-associated fecal scores exhibited no significant land use trends (p > 0.05).

Discussion

Municipal stormwater outfall discharges may contain fecal waste originating from a variety of sources making it challenging to manage, especially when monitoring is limited to general fecal indicators such as E. coli. This study investigates paired measurements of five host-associated genetic markers indicating human, ruminant, dog, and avian fecal pollution sources with E. coli and precipitation from seven headwater urban streams that receive discharge from multiple stormwater outfalls. Findings demonstrate the benefits of combining fecal source identification with E. coli monitoring, provide new insights on host-associated genetic method selection and performance in urban catchments impacted by stormwater outfalls, and further refines a recently reported censored data approach for interpreting qPCR measurements.

General fecal indicator pollution trends

General fecal indicator bacteria such as E. coli are routinely used to assess whether fecal pollution is present in urban streams impacted by stormwater discharge. Using a single sample maximum value (410 MPN/100 ml), 44.6% of receiving water samples exceeded the local water quality assessment benchmark demonstrating that these urban streams frequently harbor fecal pollution levels that compromise water quality. In addition, E. coli levels were 3.57 log₁₀ MPN/100 ml or higher across sites after rain events corroborating other research reporting that fecal pollution from urban landscape run-off and stormwater outfalls contribute substantially to local water impairment [41, 42]. These findings provide evidence that there are significant issues with stormwater in the study area but provide minimal information on potential pollutant sources. As a result, it remains unclear whether E. coli originate from human waste, resident pet owner practices, local wildlife, or some combination of sources demonstrating that paired measurements of host-associated genetic markers could enhance water quality management.

Human fecal pollution trends

Human fecal pollution is commonly identified in urban stormwater [43–45] and can represent a considerable public health risk when present at unsafe levels [46, 47]. In this study, human waste fecal pollution trends were characterized using recently nationally validated HF183/BacR287 and HumM2 qPCR procedures [34, 35, 39] and SRM 2917 [31, 32], revealing multiple trends. First, human fecal pollution was detected at all sites, but occurrence was highly variable ranging from 9.1% (Texas Run) to 96.7% (Fort Chaplin) of samples suggesting that mitigation strategies will likely vary from one site to another. Second, human fecal pollution average concentrations were significantly higher (fecal score 95% BCI do not overlap) in samples where E. coli levels exceeded the 410 MPN/100 ml benchmark or after rain events indicating a close link between human waste, precipitation, and reduced water quality, except the Fort Stanton site. At this site, fecal scores were not significantly different (fecal score 95% BCI overlap) regardless of sample groupings even though 42.4% of samples (n = 14) indicated no detection of HF183/BacR287 or HumM2 genetic markers. An inspection of this catchment revealed a stormwater outfall discharging gray water indicating that the fecal score trends observed at this site may be indicative of an illicit connection to a stormwater drainage system. Finally, human-associated genetic results exhibited good agreement between HF183/BacR287 and HumM2 paired measurements. Fort Chaplin and Fort Dupont site samples demonstrated the highest average levels of human fecal pollution based on site ranking fecal scores regardless of human-associated genetic marker. Both methods also identified Texas Run to have the lowest human waste concentration on average. Almost identical trends were observed when samples were grouped by E. coli or rainfall fecal score definitions. Any differences were likely attributed to the well documented lower sensitivity of the HumM2 qPCR assay [48, 49] leading to a higher occurrence of non-detections in this study (68.4% for HumM2 compared to 48.1% for HF183/BacR287). A higher percentage of non-detections reduces fecal score values, and in extreme cases, renders a sample grouping ineligible for fecal score determination. Due to the high level of agreement and reduced HumM2 sensitivity, future studies may consider using only the HF183/BacR287 qPCR assay (EPA Method 1696.1) to reduce expenses.

Dog fecal pollution trends

There is a growing body of evidence indicating that pet owner waste management, or the lack thereof, can directly impact local water quality [6, 17, 25]. Antibiotic resistant bacteria [50] as well as some human pathogens and parasites are shed in dog fecal matter [51–53] and dog waste has been reported in stormwater discharges [10, 54, 55]. In District of Columbia urban stream samples, dog-associated genetic marker fecal scores were always higher after rain or when E. coli levels exceeded the local water quality assessment benchmark. In addition, no significant correlations were observed between fecal scores and impervious surface, park, or residential land use metrics (p > 0.05) suggesting that dog waste management practices may help improve water quality regardless of location. Dog fecal pollution was most pronounced in the Fort Chaplin catchment yielding a significantly higher fecal score (95% BCI does not overlap) compared to all other study sites making this a candidate location to implement improved dog waste management practices such as installation and upkeep of dog waste stations combined with a public awareness campaign. It is also interesting to note that the ability of the DG3 assay to delineate between dog (Canis lupus familiaris) and coyote (Canis latrans) fecal waste remains unknown. As a result, it is possible that coyotes in the District of Columbia could be contributing to these trends. However, coyote populations in this area are reported to be small with sightings extremely rare [56]. Additional research is needed to confirm the specificity of the DG3 assay and characterize the local coyote population in the study area.

Wildlife fecal pollution trends

Fecal pollution originating from wildlife can be a difficult challenge for water quality and stormwater managers. The District of Columbia is a highly urbanized city consisting of approximately 20% park land with a large amount of river shoreline providing habitats that can sustain bird and deer populations year-round. These animals shed fecal indicators such as E. coli and may harbor public health relevant pathogens as well as antibiotic resistance bacteria [57–60]. There is limited information on the occurrence of wildlife waste in urban areas, especially for ruminants [54, 55]. Prior to this study, the influence of wildlife on urban receiving water quality in the District of Columbia remained largely unknown. Findings suggest that wildlife fecal waste exhibit clear links to rainfall, land use, and urban stream water quality impairment in the study area. Results indicate that both avian and ruminant sources were always significantly higher (95% BCI do not overlap) after it had recently rained or when E. coli levels exceeded the local management benchmark. However, ruminant waste was often not detected in the absence of rainfall or when E. coli levels were low (9 of 14 fecal score combinations, 64.3%). Avian fecal scores exhibited a different trend where waste was detected regardless of sample groupings suggesting that birds have consistent access to urban streams. In addition, avian fecal scores were the only pollutant source yielding a significant correlation to land use indicating that after rain events, levels rise as the amount of residential area increases in a catchment (p = 0.038; R² = 0.612) supporting the notion that microorganisms from bird waste can occur in roof-captured rainwater [61] and other residential surfaces. Together, wildlife fecal source identification findings suggest a catchment loading potential, in which waste builds up on surfaces between rain events and is washed off during a subsequent storm contributing to poor water quality in urban streams.

Implications for urban receiving water and stormwater management

Findings have multiple implications for the management of urban receiving waters impacted by stormwater discharges. First, experiments demonstrate the utility of fecal source identification to help locate E. coli sources and identify actionable outcomes that can reduce or eliminate specific pollution sources. For example, host-associated genetic marker findings suggest that the elimination of human waste sources alone may not reduce E. coli to an acceptable level due to the presence of dog, ruminant, and avian sources. However, prioritizing sites with the highest average concentration of human waste may be an effective strategy to minimize potential exposure to human pathogens reducing public health risks. Site prioritization based on non-human sources can also be helpful to identify where the implementation of pet owner and wildlife mitigation programs will make the largest impact. Second, no sample amplification inhibition and minimal DNA recovery bias (2.2%) suggests that qPCR methods can be a reliable tool for urban streams impacted by municipal stormwater discharges, even in systems with very high levels of fecal pollution after large storm events. Results also suggest that it may not be necessary to include both HF183/BacR287 and HumM2 in future studies, reducing the cost of human fecal source identification implementation. Third, human fecal pollution trends were characterized with nationally validated protocols (EPA Methods 1696.1 and 1697.1) in conjunction with a NIST certified reference material (SRM 2917) allowing for a more direct comparison of measurements presented here with future study data sets that use the same standardized methodology. Fourth, fecal source identification testing of chronically polluted urban streams resulted in data sets with non-detections ranging from 34.2% (GFD) to 68.4% (HumM2) preventing the use of traditional statistical analysis methods. The censored data fecal score approach allowed for the quantification of host-associated genetic marker measurements providing useful insights on the link between pollutant sources, precipitation, land use, and water quality, despite high frequencies of non-detections. Finally, the scope of this study was restricted to urban receiving waters, therefore, additional research is needed to characterize fecal sources directly from stormwater outfalls in the study area.

Acknowledgments

This study was conducted in partnership with the District of Columbia Department of Energy and Environment (DOEE). Information has been subjected to U.S. EPA and DOEE peer and administrative review and has been approved for external publication. Any opinions expressed in this paper are those of the authors and do not necessarily reflect the official positions and policies of the U.S. EPA or DOEE. Any mention of trade names or commercial products does not constitute endorsement or recommendation for use.

References

1. UN. (2019) World Urbanization Prospects: The 2018 Revision. New York: United Nations.
2. Paule-Mercado MA, Ventura JS, Memon SA, Jahng D, Kang JH, Lee CH. (2016) Monitoring and predicting the fecal indicator bacteria concentrations from agricultural, mixed land use and urban stormwater runoff. Science of the Total Environment 550: 1171–1181. pmid:26895037
- View Article
- PubMed/NCBI
- Google Scholar
3. Goto DK, Yan T. (2011) Effects of land uses on fecal indicator bacteria in the water and soil of a tropical watershed. Microbes and Environment 26: 254–260. pmid:21666390
- View Article
- PubMed/NCBI
- Google Scholar
4. Colford JM, Wade TJ, Schiff KC, Wright CC, Griffith JF, Sandu SK, et al. (2007) Water quality indicators and the risk of illness at beaches with nonpoint sources of fecal contamination. Epidemiology 18: 27–35. pmid:17149140
- View Article
- PubMed/NCBI
- Google Scholar
5. Calderon R, Mood E, Dufour A. (1991) Health effects of swimmers and nonpoint source of contaminated water. International Journal of Environmental Health Research 1: 21–31.
- View Article
- Google Scholar
6. Ervin JC, Van De Werfhorst L, Murray J, Holden P. (2014) Microbial source tracking in a coastal california watershed reveals canines as controllable sources of fecal contamination. Environmental Science and Technology 48: 9043–9052. pmid:25055204
- View Article
- PubMed/NCBI
- Google Scholar
7. Staley ZR, Boyd RJ, Shum P, Edge TA. (2018) Microbial source tracking using quantitative and digital PCR to identify sources of fecal contamination in stormwater, river water, and beach water in a Great Lakes area of concern. Applied and Environmental Microbiology 84: e01634. pmid:30097445
- View Article
- PubMed/NCBI
- Google Scholar
8. Shehane SD, Harwood VJ, Whitlock JE, Rose JB. (2005) The influence of rainfall on the incidence of microbial faecal indicators and the dominant sources of faecal pollution in a Florida river. Journal of Applied Microbiology 98: 1127–1136. pmid:15836482
- View Article
- PubMed/NCBI
- Google Scholar
9. USEPA. (2012) Recreational Water Quality Criteria. Office of Water.
- View Article
- Google Scholar
10. Feng S, Bootsma MJ, McLellan SL. (2018) Human-associated Lachnospiraceae genetic markers improve detection of fecal pollution sources in urban waters. Applied and Environmental Microbiology 84: e00309–00318. pmid:29728386
- View Article
- PubMed/NCBI
- Google Scholar
11. Kildare BJ, Leutenegger CM, McSwain BS, Bambic DG, Rajal VB, Wuertz S. (2007) 16S rRNA-based assays for quantitative detection of universal, human-, cow-, and dog-specific fecal Bacteroidales: a Bayesian approach. Water Research 41: 3701–3715.
- View Article
- Google Scholar
12. McQuaig S, Scott TM, Harwood VJ, Farrah SR, Lukaskik JO. (2009) Quantification of human polyomaviruses JC Virus and BK Virus by TaqMan quantitative PCR and comparison to other water quality indicators in water and fecal samples. Applied and Environmental Microbiology 75: 3379–3388. pmid:19346361
- View Article
- PubMed/NCBI
- Google Scholar
13. Stachler E, Kelty CA, Sivaganesan M, Li X, Bibby K, Shanks OC. (2017) Development of CrAssphage quantitative real-time PCR assays for human fecal pollution measurement. Environmental Science & Technology 51: 9146–9154.
- View Article
- Google Scholar
14. Mieszkin S, Furet JP, Corthier G, Gourmelon M. (2009) Estimation of pig fecal contamination in a river catchment by real-time PCR using two pig-specific Bacteroidales 16S rRNA genetic markers. Applied and Environmental Microbiology 75: 3045–3054. pmid:19329663
- View Article
- PubMed/NCBI
- Google Scholar
15. Garcia-Aljaro C, Balleste E, Muniesa M, Jofre J. (2017) Determination of crAssphage in water samples and applicability for tracking human faecal polltuion. Microbial Biotechnology 10: 1775–1780.
- View Article
- Google Scholar
16. Reischer GH, Haider JM, Sommer R, Stadler H, Keiblinger KM, Hornek R, et al. (2008) Quantitative microbial faecal source tracking with sampling guided by hydrological catchment dynamics. Environmental Microbiology 10: 2598–2608. pmid:18564182
- View Article
- PubMed/NCBI
- Google Scholar
17. Shrestha A, Kelty CA, Sivaganesan M, Shanks OC, Dorevich S. (2020) Fecal pollution source characterization at non-point source impacted beaches under dry and wet weather conditions. Water Research 182: 116014. pmid:32622131
- View Article
- PubMed/NCBI
- Google Scholar
18. Ervin JS, Russell TL, Layton BA, Yamahara KM, Wang D, Sassoubre LM, et al. (2013) Characterization of fecal concentrations in human and other animal sources by physical, culture-based, and quantitative real-time PCR methods. Water Research 47: 6873–6882. pmid:23871252
- View Article
- PubMed/NCBI
- Google Scholar
19. Green HC, Dick LK, Gilpin B, Samadpour M, Field KG. (2012) Genetic markers for rapid PCR-based identificationof gull, Canada goose, duck, and chicken fecal contamination in water. Applied and Environmental Microbiology 78: 503–510. pmid:22081573
- View Article
- PubMed/NCBI
- Google Scholar
20. Ahmed W, Harwood VJ, Nguyen N, Young S, Hamilton K, Toze S. (2016) Utility of Helicobacter spp. associated GFD markers for detecting avian fecal pollution in natural waters of two continents. Water Research 88: 613–622.
- View Article
- Google Scholar
21. Mieszkin S, Yala JF, Joubrel R, Gourmelon M. (2010) Phylogenetic analysis of Bacteroidales 16S rRNA gene sequences from human and animal effluents and assessment of rumaint faecal pollution by real-time PCR. Journal of Applied Microbiology 108: 974–984.
- View Article
- Google Scholar
22. Raith MR, Kelty CA, Griffith JF, Schriewer A, Wuertz S, Mieszkin S, et al. (2013) Comparison of PCR and quantitative real-time PCR methods for the characterization of ruminant and cattle fecal pollution sources. Water Research 47: 6921–6928. pmid:23871256
- View Article
- PubMed/NCBI
- Google Scholar
23. Shanks OC, White K, Kelty CA, Hayes S, Sivaganesan M, Jenkins M, et al. (2010) Performance assessment of cattle-associated PCR and quantitative real-time PCR assays targeting Bacteroidales Genes. Applied and Environmental Microbiology 76: 1359–1366.
- View Article
- Google Scholar
24. Cao Y, Sivaganesan M, Kelty CA, Wang D, Boehm AB, Griffith JF, et al. (2018) A human fecal contamination score for ranking recreational sites using the HF183/BacR287 quantitative real-time PCR method. Water Research 128: 148–156. pmid:29101858
- View Article
- PubMed/NCBI
- Google Scholar
25. Li X, Sivaganesan M, Kelty CA, Zimmer-Faust A, Clinton P, Reichman JR, et al. (2019) Large-scale implementation of standardized quantitative real-time PCR fecal source identification procedures in the Tillamook Bay Watershed. PLoS ONE 14: e0216827. pmid:31170166
- View Article
- PubMed/NCBI
- Google Scholar
26. Li X, Kelty CA, Sivaganesan M, Shanks OC. (2021) Variable fecal source prioritization in recreational waters routinely monitored with viral and bacterial general indicators. Water Research 192: 116845. pmid:33508720
- View Article
- PubMed/NCBI
- Google Scholar
27. Gentry RW, McCarthy JF, Layton AC, McKay L, Williams D, Koirala SR, et al. (2006) Escherichi coli loading at or near base flow in a mixed-use watershed. Journal of Environmental Quality 35: 2244–2249.
- View Article
- Google Scholar
28. Shanks OC, Nietch C, Simonich MT, Younger M, Reynolds D, Field KG. (2006) Basin-wide analysis of the dynamics of fecal contamination and fecal source identification in Tillamook Bay, Oregon. Applied and Environmental Microbiology 72: 5537–5546. pmid:16885307
- View Article
- PubMed/NCBI
- Google Scholar
29. USEPA. (1992) NPDES Storm Water Sampling Guidance Document. Washington, DC: United States Environmental Protection Agency. EPA 833-8-92-001.
30. DDOE. (2020) District Department of the Environment Notice of Final Rulemaking: Triennial Review of District of Columbia’s Water Quality Standards.
- View Article
- Google Scholar
31. Kralj J, Servetas S, Hunter M, Toman B, Jackson S. (2021) Certification of standard reference material 2917 plasmid DNA for fecal indicator detection and identification. NIST Special Publication: 1–41.
- View Article
- Google Scholar
32. Willis JR, Sivaganesan M, Haugland RA, Kralj J, Servetas S, Hunter ME, et al. (2022) Performance of NIST SRM 2917 with 13 recreational water quality monitoring qPCR assays. Water Research 212: 118114.
- View Article
- Google Scholar
33. Green HC, White KM, Kelty CA, Shanks OC. (2014) Development of rapid canine fecal source identification PCR-based assays. Environmental Science and Technology: 11453–11461. pmid:25203917
- View Article
- PubMed/NCBI
- Google Scholar
34. USEPA. (2021) Method 1697.1: Characterization of human fecal pollution in water by HumM2 TaqMan quantitative polymerase chain reaction (qPCR) assay. Washington DC: United States Environmental Protection Agency.
35. USEPA. (2021) Method 1696.1: Characterization of human fecal pollution in water by HF183/BacR287 TaqMan quantitative polymerase chain reaction (qPCR) assay. Washington DC: United States Environmental Protection Agency.
36. Haugland RA, Varma M, Kelty CA, Peed L, Sivaganesan M, Shanks OC. (2010) Evaluation of genetic markers from the 16S rRNA gene V2 region for use in quantitative detection of selected Bacteroidales species and human fecal waste by real-time PCR. Systematic and Applied Microbiology 33: 348–357.
- View Article
- Google Scholar
37. Bustin SA, Nolan T. (2006) Template handling, preparation, and quantification. In: Bustin SA, editor. A-Z of Quantitative PCR. La Jolla: International University Line. pp. 189.
38. LifeTechnologies. (2014) Real-Time PCR Handbook 3rd Edition. pp. 11.
39. Shanks OC, Kelty CA, Oshiro R, Haugland RA, Madi T, Brooks L, et al. (2016) Data acceptance criteria for standardized human-associated fecal source identificationq quantitative real-time PCR methods. Applied and Environmental Microbiology 82: 2773–2782.
- View Article
- Google Scholar
40. Sivaganesan M, Seifring S, Varma M, Haugland RA, Shanks OC. (2008) A Bayesian method for calculating real-time quantitative PCR calibration curves using absolute plasmid DNA standards. BMC Bioinformatics 9: 120. pmid:18298858
- View Article
- PubMed/NCBI
- Google Scholar
41. Hong Y, Soulignac F, Roguet A, Li C, Lemaire BJ, Martins RS, et al. (2021) Impact of Escherichia coli from stormwater drainage on recreational water quality: an integrated monitoring and modelling of urban catchment, pipes and lake. Environmental Science and Pollution Research 28: 2245–2259.
- View Article
- Google Scholar
42. Daly E, Kolotelo P, Schang C, Osborne CA, Coleman R, Deletic A, et al. (2013) Escherichia coli concentrations and loads in an urbanised catchment: the Yarra River, Australia. Journal of Hydrology 497: 51–61.
- View Article
- Google Scholar
43. Olds HT, Corsi SR, Dila DK, Halmo KM, Bootsma MJ, McLellan SL. (2018) High levels of sewage contamination released from urban areas after storm events: a quantitative survey with sewage specific bacterial indicators. PLoS Medicine 15: e1002614. pmid:30040843
- View Article
- PubMed/NCBI
- Google Scholar
44. McGinnis S, Spencer S, Firnstahl A, Stokdyk J, Borchardt M, McCarthy DT, et al. (2018) Human Bacteroides and total coliforms as indicators of recent combined sewer overflows and rain events in urban creeks. Science of the Total Environment 630: 967–976.
- View Article
- Google Scholar
45. Ahmed W, Lobos A, Senkbeil J, Peraud J, Gallard J, Harwood VJ. (2018) Evaluation of the novel crAssphage marker for sewage pollution tracking in storm drain outfalls in Tampa, Florida. Water Research 131: 142–150.
- View Article
- Google Scholar
46. Soller JA, Schoen ME, Bartrand T, Ravenscroft JE, Ashbolt NJ. (2010) Estimated human health risks from exposure to recreational waters impacted by human and non-human sources of faecal contamination. Water Research 44: 4674–4691. pmid:20656314
- View Article
- PubMed/NCBI
- Google Scholar
47. Soller JA, Schoen ME, Varghese A, Ichida AM, Boehm AB, Efitm S, et al. (2014) Human health risk implications of multiple sources of faecal indicator bacteria in a recreational water body. Water Research 66: 254–264. pmid:25222329
- View Article
- PubMed/NCBI
- Google Scholar
48. Shanks OC, Kelty CA, Sivaganesan M, Varma M, Haugland RA. (2009) Quantitative PCR for genetic markers of human fecal pollution. Applied and Environmental Microbiology 75: 5507–5513. pmid:19592537
- View Article
- PubMed/NCBI
- Google Scholar
49. Shanks OC, White K, Kelty CA, Sivaganesan M, Blannon J, Meckes M, et al. (2010) Performance of PCR-based assays targeting Bacteroidales genetic markers of human fecal pollution in sewage and fecal samples. Environmental Science and Technology 44: 6281–6288.
- View Article
- Google Scholar
50. Nam EK, Ko S, Chae JS, Hwang CY. (2013) Characterization and zoonotic potential of uropathogenic Escherichia coli isolated from dogs. Journal of Microbiology and Biotechnology 23: 422–429.
- View Article
- Google Scholar
51. Chaban B, Ngeleka M, Hill JE. (2010) Detection and quantification of 14 Campylobacter species in pet dogs reveals an increase in species richness in feces of diarrheic animals. BMC Microbiology 10: 73.
- View Article
- Google Scholar
52. Katagiri S, Oliveira-Sequeira TC. (2008) Prevalence of dog intestinal parasites and risk perception of zoonotic infection by dog owners in Sao Paulo State, Brazil. Zoonoses and Public Health 55: 406–413.
- View Article
- Google Scholar
53. Traub RJ, Robertson ID, Irwin P, Mencke N, Monis P, Thompson RC. (2003) Humans, dogs and parastic zoonoses—unravelling the relationships in a remote endemic community in northeast Indai using molecular tools. Parasitology Research 90: S156–157.
- View Article
- Google Scholar
54. Staley ZR, Grabuski J, Sverko E, Edge TA. (2016) Comparison of microbial and chemical source tracking markers to identify fecal contamination sources in the Humber River and associated stormwater outfalls. Applied and Environmental Microbiology:
- View Article
- Google Scholar
55. Steele JA, Blackwood AD, Griffith JF, Noble RT, Schiff KC. (2018) Quantification of pathogens and markers of fecal contamination during storm events along popular surfing beaches in San Diego, California. Water Research 136: 137–149. pmid:29501758
- View Article
- PubMed/NCBI
- Google Scholar
56. Draheim M, Hadidian J, DePietro D, Kuczera L. (2022) The district coyote project. Washington, D.C.
- View Article
- Google Scholar
57. Devane ML, Nicol C, Ball A, Klena JD, Scholes P, Hudson JA, et al. (2005) The occurrence of Campylobacter subtypes in environmental reservoirs and potential transmission routes. Journal of Applied Microbiology 98: 980–990.
- View Article
- Google Scholar
58. Fallacara DM, Monhan CM, Morishita TY, Bremer CA, Wack RF. (2004) Survey of parasites and bacterial pathogens from free-living waterfowl in zoological settings. Avian Diseases 48: 759–767. pmid:15666857
- View Article
- PubMed/NCBI
- Google Scholar
59. Fallacara DM, Monhan CM, Morishita TY, Wack RF. (2001) Fecal shedding and antimicrobial susceptibility of selected bacterial pathognes and a survey of intestinal parasites in free-living waterfowl. Avian Diseases 45: 128–135.
- View Article
- Google Scholar
60. Rogers SW, Shaffer CE, Langen TA, Jahne M, Welsh R. (2018) Antibiotic-resistance genes and pathogens shed by wild deer correlat with land application of residuals. EcoHealth 15: 409–425.
- View Article
- Google Scholar
61. Ahmed W, Brandes H, Gyawali P, SIdhu JPS, Toze S. (2014) Opportunistic pathogens in roof-captured rainwater samples, determined using quantitative PCR. Water Research 53: 361–369. pmid:24531256
- View Article
- PubMed/NCBI
- Google Scholar

[ref1] 1. UN. (2019) World Urbanization Prospects: The 2018 Revision. New York: United Nations.

[ref2] 2. Paule-Mercado MA, Ventura JS, Memon SA, Jahng D, Kang JH, Lee CH. (2016) Monitoring and predicting the fecal indicator bacteria concentrations from agricultural, mixed land use and urban stormwater runoff. Science of the Total Environment 550: 1171–1181. pmid:26895037
View Article
PubMed/NCBI
Google Scholar

[3] View Article

[4] PubMed/NCBI

[5] Google Scholar

[ref3] 3. Goto DK, Yan T. (2011) Effects of land uses on fecal indicator bacteria in the water and soil of a tropical watershed. Microbes and Environment 26: 254–260. pmid:21666390
View Article
PubMed/NCBI
Google Scholar

[7] View Article

[8] PubMed/NCBI

[9] Google Scholar

[ref4] 4. Colford JM, Wade TJ, Schiff KC, Wright CC, Griffith JF, Sandu SK, et al. (2007) Water quality indicators and the risk of illness at beaches with nonpoint sources of fecal contamination. Epidemiology 18: 27–35. pmid:17149140
View Article
PubMed/NCBI
Google Scholar

[11] View Article

[12] PubMed/NCBI

[13] Google Scholar

[ref5] 5. Calderon R, Mood E, Dufour A. (1991) Health effects of swimmers and nonpoint source of contaminated water. International Journal of Environmental Health Research 1: 21–31.
View Article
Google Scholar

[15] View Article

[16] Google Scholar

[ref6] 6. Ervin JC, Van De Werfhorst L, Murray J, Holden P. (2014) Microbial source tracking in a coastal california watershed reveals canines as controllable sources of fecal contamination. Environmental Science and Technology 48: 9043–9052. pmid:25055204
View Article
PubMed/NCBI
Google Scholar

[18] View Article

[19] PubMed/NCBI

[20] Google Scholar

[ref7] 7. Staley ZR, Boyd RJ, Shum P, Edge TA. (2018) Microbial source tracking using quantitative and digital PCR to identify sources of fecal contamination in stormwater, river water, and beach water in a Great Lakes area of concern. Applied and Environmental Microbiology 84: e01634. pmid:30097445
View Article
PubMed/NCBI
Google Scholar

[22] View Article

[23] PubMed/NCBI

[24] Google Scholar

[ref8] 8. Shehane SD, Harwood VJ, Whitlock JE, Rose JB. (2005) The influence of rainfall on the incidence of microbial faecal indicators and the dominant sources of faecal pollution in a Florida river. Journal of Applied Microbiology 98: 1127–1136. pmid:15836482
View Article
PubMed/NCBI
Google Scholar

[26] View Article

[27] PubMed/NCBI

[28] Google Scholar

[ref9] 9. USEPA. (2012) Recreational Water Quality Criteria. Office of Water.
View Article
Google Scholar

[30] View Article

[31] Google Scholar

[ref10] 10. Feng S, Bootsma MJ, McLellan SL. (2018) Human-associated Lachnospiraceae genetic markers improve detection of fecal pollution sources in urban waters. Applied and Environmental Microbiology 84: e00309–00318. pmid:29728386
View Article
PubMed/NCBI
Google Scholar

[33] View Article

[34] PubMed/NCBI

[35] Google Scholar

[ref11] 11. Kildare BJ, Leutenegger CM, McSwain BS, Bambic DG, Rajal VB, Wuertz S. (2007) 16S rRNA-based assays for quantitative detection of universal, human-, cow-, and dog-specific fecal Bacteroidales: a Bayesian approach. Water Research 41: 3701–3715.
View Article
Google Scholar

[37] View Article

[38] Google Scholar

[ref12] 12. McQuaig S, Scott TM, Harwood VJ, Farrah SR, Lukaskik JO. (2009) Quantification of human polyomaviruses JC Virus and BK Virus by TaqMan quantitative PCR and comparison to other water quality indicators in water and fecal samples. Applied and Environmental Microbiology 75: 3379–3388. pmid:19346361
View Article
PubMed/NCBI
Google Scholar

[40] View Article

[41] PubMed/NCBI

[42] Google Scholar

[ref13] 13. Stachler E, Kelty CA, Sivaganesan M, Li X, Bibby K, Shanks OC. (2017) Development of CrAssphage quantitative real-time PCR assays for human fecal pollution measurement. Environmental Science & Technology 51: 9146–9154.
View Article
Google Scholar

[44] View Article

[45] Google Scholar

[ref14] 14. Mieszkin S, Furet JP, Corthier G, Gourmelon M. (2009) Estimation of pig fecal contamination in a river catchment by real-time PCR using two pig-specific Bacteroidales 16S rRNA genetic markers. Applied and Environmental Microbiology 75: 3045–3054. pmid:19329663
View Article
PubMed/NCBI
Google Scholar

[47] View Article

[48] PubMed/NCBI

[49] Google Scholar

[ref15] 15. Garcia-Aljaro C, Balleste E, Muniesa M, Jofre J. (2017) Determination of crAssphage in water samples and applicability for tracking human faecal polltuion. Microbial Biotechnology 10: 1775–1780.
View Article
Google Scholar

[51] View Article

[52] Google Scholar

[ref16] 16. Reischer GH, Haider JM, Sommer R, Stadler H, Keiblinger KM, Hornek R, et al. (2008) Quantitative microbial faecal source tracking with sampling guided by hydrological catchment dynamics. Environmental Microbiology 10: 2598–2608. pmid:18564182
View Article
PubMed/NCBI
Google Scholar

[54] View Article

[55] PubMed/NCBI

[56] Google Scholar

[ref17] 17. Shrestha A, Kelty CA, Sivaganesan M, Shanks OC, Dorevich S. (2020) Fecal pollution source characterization at non-point source impacted beaches under dry and wet weather conditions. Water Research 182: 116014. pmid:32622131
View Article
PubMed/NCBI
Google Scholar

[58] View Article

[59] PubMed/NCBI

[60] Google Scholar

[ref18] 18. Ervin JS, Russell TL, Layton BA, Yamahara KM, Wang D, Sassoubre LM, et al. (2013) Characterization of fecal concentrations in human and other animal sources by physical, culture-based, and quantitative real-time PCR methods. Water Research 47: 6873–6882. pmid:23871252
View Article
PubMed/NCBI
Google Scholar

[62] View Article

[63] PubMed/NCBI

[64] Google Scholar

[ref19] 19. Green HC, Dick LK, Gilpin B, Samadpour M, Field KG. (2012) Genetic markers for rapid PCR-based identificationof gull, Canada goose, duck, and chicken fecal contamination in water. Applied and Environmental Microbiology 78: 503–510. pmid:22081573
View Article
PubMed/NCBI
Google Scholar

[66] View Article

[67] PubMed/NCBI

[68] Google Scholar

[ref20] 20. Ahmed W, Harwood VJ, Nguyen N, Young S, Hamilton K, Toze S. (2016) Utility of Helicobacter spp. associated GFD markers for detecting avian fecal pollution in natural waters of two continents. Water Research 88: 613–622.
View Article
Google Scholar

[70] View Article

[71] Google Scholar

[ref21] 21. Mieszkin S, Yala JF, Joubrel R, Gourmelon M. (2010) Phylogenetic analysis of Bacteroidales 16S rRNA gene sequences from human and animal effluents and assessment of rumaint faecal pollution by real-time PCR. Journal of Applied Microbiology 108: 974–984.
View Article
Google Scholar

[73] View Article

[74] Google Scholar

[ref22] 22. Raith MR, Kelty CA, Griffith JF, Schriewer A, Wuertz S, Mieszkin S, et al. (2013) Comparison of PCR and quantitative real-time PCR methods for the characterization of ruminant and cattle fecal pollution sources. Water Research 47: 6921–6928. pmid:23871256
View Article
PubMed/NCBI
Google Scholar

[76] View Article

[77] PubMed/NCBI

[78] Google Scholar

[ref23] 23. Shanks OC, White K, Kelty CA, Hayes S, Sivaganesan M, Jenkins M, et al. (2010) Performance assessment of cattle-associated PCR and quantitative real-time PCR assays targeting Bacteroidales Genes. Applied and Environmental Microbiology 76: 1359–1366.
View Article
Google Scholar

[80] View Article

[81] Google Scholar

[ref24] 24. Cao Y, Sivaganesan M, Kelty CA, Wang D, Boehm AB, Griffith JF, et al. (2018) A human fecal contamination score for ranking recreational sites using the HF183/BacR287 quantitative real-time PCR method. Water Research 128: 148–156. pmid:29101858
View Article
PubMed/NCBI
Google Scholar

[83] View Article

[84] PubMed/NCBI

[85] Google Scholar

[ref25] 25. Li X, Sivaganesan M, Kelty CA, Zimmer-Faust A, Clinton P, Reichman JR, et al. (2019) Large-scale implementation of standardized quantitative real-time PCR fecal source identification procedures in the Tillamook Bay Watershed. PLoS ONE 14: e0216827. pmid:31170166
View Article
PubMed/NCBI
Google Scholar

[87] View Article

[88] PubMed/NCBI

[89] Google Scholar

[ref26] 26. Li X, Kelty CA, Sivaganesan M, Shanks OC. (2021) Variable fecal source prioritization in recreational waters routinely monitored with viral and bacterial general indicators. Water Research 192: 116845. pmid:33508720
View Article
PubMed/NCBI
Google Scholar

[91] View Article

[92] PubMed/NCBI

[93] Google Scholar

[ref27] 27. Gentry RW, McCarthy JF, Layton AC, McKay L, Williams D, Koirala SR, et al. (2006) Escherichi coli loading at or near base flow in a mixed-use watershed. Journal of Environmental Quality 35: 2244–2249.
View Article
Google Scholar

[95] View Article

[96] Google Scholar

[ref28] 28. Shanks OC, Nietch C, Simonich MT, Younger M, Reynolds D, Field KG. (2006) Basin-wide analysis of the dynamics of fecal contamination and fecal source identification in Tillamook Bay, Oregon. Applied and Environmental Microbiology 72: 5537–5546. pmid:16885307
View Article
PubMed/NCBI
Google Scholar

[98] View Article

[99] PubMed/NCBI

[100] Google Scholar

[ref29] 29. USEPA. (1992) NPDES Storm Water Sampling Guidance Document. Washington, DC: United States Environmental Protection Agency. EPA 833-8-92-001.

[ref30] 30. DDOE. (2020) District Department of the Environment Notice of Final Rulemaking: Triennial Review of District of Columbia’s Water Quality Standards.
View Article
Google Scholar

[103] View Article

[104] Google Scholar

[ref31] 31. Kralj J, Servetas S, Hunter M, Toman B, Jackson S. (2021) Certification of standard reference material 2917 plasmid DNA for fecal indicator detection and identification. NIST Special Publication: 1–41.
View Article
Google Scholar

[106] View Article

[107] Google Scholar

[ref32] 32. Willis JR, Sivaganesan M, Haugland RA, Kralj J, Servetas S, Hunter ME, et al. (2022) Performance of NIST SRM 2917 with 13 recreational water quality monitoring qPCR assays. Water Research 212: 118114.
View Article
Google Scholar

[109] View Article

[110] Google Scholar

[ref33] 33. Green HC, White KM, Kelty CA, Shanks OC. (2014) Development of rapid canine fecal source identification PCR-based assays. Environmental Science and Technology: 11453–11461. pmid:25203917
View Article
PubMed/NCBI
Google Scholar

[112] View Article

[113] PubMed/NCBI

[114] Google Scholar

[ref34] 34. USEPA. (2021) Method 1697.1: Characterization of human fecal pollution in water by HumM2 TaqMan quantitative polymerase chain reaction (qPCR) assay. Washington DC: United States Environmental Protection Agency.

[ref35] 35. USEPA. (2021) Method 1696.1: Characterization of human fecal pollution in water by HF183/BacR287 TaqMan quantitative polymerase chain reaction (qPCR) assay. Washington DC: United States Environmental Protection Agency.

[ref36] 36. Haugland RA, Varma M, Kelty CA, Peed L, Sivaganesan M, Shanks OC. (2010) Evaluation of genetic markers from the 16S rRNA gene V2 region for use in quantitative detection of selected Bacteroidales species and human fecal waste by real-time PCR. Systematic and Applied Microbiology 33: 348–357.
View Article
Google Scholar

[118] View Article

[119] Google Scholar

[ref37] 37. Bustin SA, Nolan T. (2006) Template handling, preparation, and quantification. In: Bustin SA, editor. A-Z of Quantitative PCR. La Jolla: International University Line. pp. 189.

[ref38] 38. LifeTechnologies. (2014) Real-Time PCR Handbook 3rd Edition. pp. 11.

[ref39] 39. Shanks OC, Kelty CA, Oshiro R, Haugland RA, Madi T, Brooks L, et al. (2016) Data acceptance criteria for standardized human-associated fecal source identificationq quantitative real-time PCR methods. Applied and Environmental Microbiology 82: 2773–2782.
View Article
Google Scholar

[123] View Article

[124] Google Scholar

[ref40] 40. Sivaganesan M, Seifring S, Varma M, Haugland RA, Shanks OC. (2008) A Bayesian method for calculating real-time quantitative PCR calibration curves using absolute plasmid DNA standards. BMC Bioinformatics 9: 120. pmid:18298858
View Article
PubMed/NCBI
Google Scholar

[126] View Article

[127] PubMed/NCBI

[128] Google Scholar

[ref41] 41. Hong Y, Soulignac F, Roguet A, Li C, Lemaire BJ, Martins RS, et al. (2021) Impact of Escherichia coli from stormwater drainage on recreational water quality: an integrated monitoring and modelling of urban catchment, pipes and lake. Environmental Science and Pollution Research 28: 2245–2259.
View Article
Google Scholar

[130] View Article

[131] Google Scholar

[ref42] 42. Daly E, Kolotelo P, Schang C, Osborne CA, Coleman R, Deletic A, et al. (2013) Escherichia coli concentrations and loads in an urbanised catchment: the Yarra River, Australia. Journal of Hydrology 497: 51–61.
View Article
Google Scholar

[133] View Article

[134] Google Scholar

[ref43] 43. Olds HT, Corsi SR, Dila DK, Halmo KM, Bootsma MJ, McLellan SL. (2018) High levels of sewage contamination released from urban areas after storm events: a quantitative survey with sewage specific bacterial indicators. PLoS Medicine 15: e1002614. pmid:30040843
View Article
PubMed/NCBI
Google Scholar

[136] View Article

[137] PubMed/NCBI

[138] Google Scholar

[ref44] 44. McGinnis S, Spencer S, Firnstahl A, Stokdyk J, Borchardt M, McCarthy DT, et al. (2018) Human Bacteroides and total coliforms as indicators of recent combined sewer overflows and rain events in urban creeks. Science of the Total Environment 630: 967–976.
View Article
Google Scholar

[140] View Article

[141] Google Scholar

[ref45] 45. Ahmed W, Lobos A, Senkbeil J, Peraud J, Gallard J, Harwood VJ. (2018) Evaluation of the novel crAssphage marker for sewage pollution tracking in storm drain outfalls in Tampa, Florida. Water Research 131: 142–150.
View Article
Google Scholar

[143] View Article

[144] Google Scholar

[ref46] 46. Soller JA, Schoen ME, Bartrand T, Ravenscroft JE, Ashbolt NJ. (2010) Estimated human health risks from exposure to recreational waters impacted by human and non-human sources of faecal contamination. Water Research 44: 4674–4691. pmid:20656314
View Article
PubMed/NCBI
Google Scholar

[146] View Article

[147] PubMed/NCBI

[148] Google Scholar

[ref47] 47. Soller JA, Schoen ME, Varghese A, Ichida AM, Boehm AB, Efitm S, et al. (2014) Human health risk implications of multiple sources of faecal indicator bacteria in a recreational water body. Water Research 66: 254–264. pmid:25222329
View Article
PubMed/NCBI
Google Scholar

[150] View Article

[151] PubMed/NCBI

[152] Google Scholar

[ref48] 48. Shanks OC, Kelty CA, Sivaganesan M, Varma M, Haugland RA. (2009) Quantitative PCR for genetic markers of human fecal pollution. Applied and Environmental Microbiology 75: 5507–5513. pmid:19592537
View Article
PubMed/NCBI
Google Scholar

[154] View Article

[155] PubMed/NCBI

[156] Google Scholar

[ref49] 49. Shanks OC, White K, Kelty CA, Sivaganesan M, Blannon J, Meckes M, et al. (2010) Performance of PCR-based assays targeting Bacteroidales genetic markers of human fecal pollution in sewage and fecal samples. Environmental Science and Technology 44: 6281–6288.
View Article
Google Scholar

[158] View Article

[159] Google Scholar

[ref50] 50. Nam EK, Ko S, Chae JS, Hwang CY. (2013) Characterization and zoonotic potential of uropathogenic Escherichia coli isolated from dogs. Journal of Microbiology and Biotechnology 23: 422–429.
View Article
Google Scholar

[161] View Article

[162] Google Scholar

[ref51] 51. Chaban B, Ngeleka M, Hill JE. (2010) Detection and quantification of 14 Campylobacter species in pet dogs reveals an increase in species richness in feces of diarrheic animals. BMC Microbiology 10: 73.
View Article
Google Scholar

[164] View Article

[165] Google Scholar

[ref52] 52. Katagiri S, Oliveira-Sequeira TC. (2008) Prevalence of dog intestinal parasites and risk perception of zoonotic infection by dog owners in Sao Paulo State, Brazil. Zoonoses and Public Health 55: 406–413.
View Article
Google Scholar

[167] View Article

[168] Google Scholar

[ref53] 53. Traub RJ, Robertson ID, Irwin P, Mencke N, Monis P, Thompson RC. (2003) Humans, dogs and parastic zoonoses—unravelling the relationships in a remote endemic community in northeast Indai using molecular tools. Parasitology Research 90: S156–157.
View Article
Google Scholar

[170] View Article

[171] Google Scholar

[ref54] 54. Staley ZR, Grabuski J, Sverko E, Edge TA. (2016) Comparison of microbial and chemical source tracking markers to identify fecal contamination sources in the Humber River and associated stormwater outfalls. Applied and Environmental Microbiology:
View Article
Google Scholar

[173] View Article

[174] Google Scholar

[ref55] 55. Steele JA, Blackwood AD, Griffith JF, Noble RT, Schiff KC. (2018) Quantification of pathogens and markers of fecal contamination during storm events along popular surfing beaches in San Diego, California. Water Research 136: 137–149. pmid:29501758
View Article
PubMed/NCBI
Google Scholar

[176] View Article

[177] PubMed/NCBI

[178] Google Scholar

[ref56] 56. Draheim M, Hadidian J, DePietro D, Kuczera L. (2022) The district coyote project. Washington, D.C.
View Article
Google Scholar

[180] View Article

[181] Google Scholar

[ref57] 57. Devane ML, Nicol C, Ball A, Klena JD, Scholes P, Hudson JA, et al. (2005) The occurrence of Campylobacter subtypes in environmental reservoirs and potential transmission routes. Journal of Applied Microbiology 98: 980–990.
View Article
Google Scholar

[183] View Article

[184] Google Scholar

[ref58] 58. Fallacara DM, Monhan CM, Morishita TY, Bremer CA, Wack RF. (2004) Survey of parasites and bacterial pathogens from free-living waterfowl in zoological settings. Avian Diseases 48: 759–767. pmid:15666857
View Article
PubMed/NCBI
Google Scholar

[186] View Article

[187] PubMed/NCBI

[188] Google Scholar

[ref59] 59. Fallacara DM, Monhan CM, Morishita TY, Wack RF. (2001) Fecal shedding and antimicrobial susceptibility of selected bacterial pathognes and a survey of intestinal parasites in free-living waterfowl. Avian Diseases 45: 128–135.
View Article
Google Scholar

[190] View Article

[191] Google Scholar

[ref60] 60. Rogers SW, Shaffer CE, Langen TA, Jahne M, Welsh R. (2018) Antibiotic-resistance genes and pathogens shed by wild deer correlat with land application of residuals. EcoHealth 15: 409–425.
View Article
Google Scholar

[193] View Article

[194] Google Scholar

[ref61] 61. Ahmed W, Brandes H, Gyawali P, SIdhu JPS, Toze S. (2014) Opportunistic pathogens in roof-captured rainwater samples, determined using quantitative PCR. Water Research 53: 361–369. pmid:24531256
View Article
PubMed/NCBI
Google Scholar

[196] View Article

[197] PubMed/NCBI

[198] Google Scholar

Figures

Abstract

Introduction

Materials and methods

Catchment and sampling site descriptions

Precipitation and land use characterization

Water sampling

E. coli measurements

Reference DNA preparations

Water filtration and DNA extraction

qPCR amplification

qPCR data acceptance metrics

Calculations and statistics

qPCR calibration model and performance.

Fecal score model.

Land use trend analysis.

Results

Precipitation

Land use metrics

E. coli measurements

qPCR quality controls and data acceptance

Host-associated qPCR measurements

Host-associated genetic marker fecal scores

Land use trend analyses

Discussion

General fecal indicator pollution trends

Human fecal pollution trends

Dog fecal pollution trends

Wildlife fecal pollution trends

Implications for urban receiving water and stormwater management

Acknowledgments

References