Executive Summary

The U.S. Geological Survey worked collaboratively with the New York State Department of Environmental Conservation to assess the potential sources of fecal contamination entering seven estuarine embayments across Long Island, New York, from June 2018 to July 2019. Water samples are routinely collected by the New York State Department of Environmental Conservation in Long Island embayments and analyzed for fecal coliform bacteria, an indicator of fecal contamination, to determine the closure of shellfish beds for harvest and consumption. Concentrations of fecal coliform signify the potential for pathogenic (disease-causing) bacteria to be present.

Indicator bacteria alone cannot determine the biological or geographical sources of contamination. Microbial source tracking is a method used to determine these sources of contamination. Microbial source tracking laboratory techniques can ascertain whether genetic material obtained from Bacteroides or Helicobacter bacteria in water or sediment is consistent with humans, canines (dogs), ruminants (deer, sheep), or waterfowl. Bacteroides and Helicobacter are genera of bacteria found in the gut of most warm-blooded animals. These techniques can also quantify the concentration of genetic markers found.

Information such as sample location, weather, and season, surrounding land use, and additional water-quality data help determine the geographical source and conveyance of land-based water to the embayment. The presence of genetic material and fecal coliform bacteria in samples collected at the same time is important to show that the fecal coliform bacteria are likely from the host source detected. It is possible for waters with waste infiltration to have genetic material present but no fecal indicators, such as disinfected water from a sewage treatment plant or from groundwater that has passed through a sandy aquifer.

Water samples were collected in the summer and winter seasons. In each of the two seasons, one sample was collected in dry weather, and one, after substantial rainfall. Groundwater and sediment samples were also collected throughout the study. Surface-water source sites sampled for Port Jefferson Harbor and Setauket Harbor include the culvert north of State Route 25A, the culvert north of Shore Road, the Old Mill Creek culvert, the outfall near Danfords Marina, and the Port Jefferson sewage treatment plant. The Mill Pond culvert was sampled as a surface water source site for Conscience Bay.

Human, canine, and waterfowl fecal contributions to the Port Jefferson Harbor complex were frequently detected across the landscape. Among the potential sources of fecal contaminated water contributing to the Port Jefferson Harbor complex—groundwater, pond and wetland drainage, stormwater, marinas (boats), and the Port Jefferson sewage treatment plant—pond and wetland drainage and stormwater from municipal separate storm sewer system conveyances were found to be the most likely transport mechanism of fecal contamination.

The most substantial source of fecal contamination to the Port Jefferson Harbor complex was found to be from sites draining ponds and wetlands, particularly during the summer season with higher concentrations of fecal coliform bacteria in wet weather. These sites include the culvert north of Shore Road, the culvert north of State Route 25A, Old Mill Creek, and Mill Pond to Conscience Bay. Discharge from the single stormwater outfall sample (outfall at Danfords Marina) was also a substantial contributor of fecal coliform bacteria. Contributions of fecal contamination did not appear to be attributable to the Port Jefferson sewage treatment plant or to boats and marinas in Port Jefferson Harbor. While the microbial source tracking results indicated influence from septic systems, it has not been shown that groundwater contributes to fecal coliform bacteria loads. One of the two groundwater samples along the shoreline of Port Jefferson Harbor where microbial source tracking analyses were conducted showed evidence of human and waterfowl waste that did not have corresponding fecal coliform bacteria detections, indicating limited transport of bacteria in the subsurface groundwater through the sandy aquifer material, which is consistent with results throughout Long Island. Sediment sampled along with the groundwater sample at the Public Beach and Dog Park is unlikely to act as a source during resuspension events, such as an ebbing (outgoing) tide, storm-driven waves (particularly at low tide), and boat wakes.

A classification scheme was developed that considered conditions such as high fecal coliform bacteria concentrations (particularly during dry weather samples) and contributions from human and canine waste based on detection of microbial source tracking markers. The class of sites were assessed to assist stakeholders and resource managers in prioritizing which source sites are consistently contributing fecal coliform bacteria to the Port Jefferson Harbor complex throughout the year. The culvert north of State Route 25A, the culvert north of Shore Road, the Old Mill Creek culvert, the outfall near Danfords Marina (only one sample collected), and the Mill Pond culvert to Conscience Bay were classified as locations that contribute substantial fecal contamination to the Port Jefferson Harbor complex.

Introduction

To better understand fecal contamination sources in coastal embayments on Long Island, microbial source tracking (MST) and fecal coliform bacteria (FC) data were collected from seven embayments from June 2018 to July 2019 (U.S. Geological Survey, 2020a). The objective of this research is to identify the different pathogen sources, both the host organism (such as human, mammals, or birds) and geographic origin (such as urban and storm runoff or submarine groundwater discharge), to embayments with fecal contamination. The embayments selected represent a mix of sewered and unsewered areas, differing levels of impervious land cover, varying population density, and a variety of land-use types so as to transfer what is learned to other areas beyond the seven studied. In total, 353 samples were collected in the seven embayments on Long Island. These were primarily surface-water samples but also included 17 groundwater, 11 sediment, 2 sewage influent, and 8 fecal samples.

The presence of fecal contamination in the Port Jefferson Harbor complex and resulting hazards posed to marine resources and human health are of increasing concern. Recreational uses of the harbor, including shellfishing and public bathing, are frequently restricted due to high counts of FC, which are used as surrogate indicators for when pathogens may be present. Pathogen loads can be influenced by watershed land use, proximity to point sources such as municipal separate storm sewer systems, as well as tidal exchange and circulation transporting pathogens from outside the study area (Gao and others, 2015), sediment resuspension (O’Mullan and others, 2019), and human recreation (including boating). Sewage treatment plants can also act as point sources (through effluent discharge) for pathogens when standard operations and procedures for disinfection are compromised. Nonpoint sources may include shallow groundwater discharge because many of the areas surrounding the Port Jefferson Harbor complex still rely on onsite wastewater disposal systems, such as cesspools and septic systems. Through routine FC monitoring, the New York State Department of Environmental Conservation (NYSDEC) has identified fecal contamination as a concern in the Port Jefferson Harbor complex. Microbial source tracking techniques are necessary to determine the dominant source (human, canine, ruminant, waterfowl) and geographic origin of FC, using land use, topography, and hydrology to establish a framework. The success of future pathogen-control measures may be measured by the decline in FC concentrations.

Site Description

The Port Jefferson Harbor complex is an interconnected system of embayments that include Port Jefferson Harbor, Setauket Harbor, Little Bay, the Narrows, and Conscience Bay (fig. 1). These embayments are connected to Long Island Sound by a single inlet on the northern end of Port Jefferson Harbor in the town of Brookhaven, Suffolk County, New York. Port Jefferson Harbor is bordered by the villages of Belle Terre to the east, Port Jefferson to the south, and Poquott to the west. Setauket Harbor is bordered by Poquott to the northeast and Setauket-East Setauket to the south and west. Conscience Bay is bordered by Setauket-East Setauket to the east and the village of Old Field to the west. Strongs Neck is a dividing peninsula within the Port Jefferson Harbor complex and is west of Port Jefferson Harbor, northwest of Setauket Harbor, and east of Conscience Bay.

Much of the shoreline is unsewered and served by cesspools or septic tanks. Port Jefferson village at the southern end of Port Jefferson Harbor is served by the Port Jefferson sewage treatment plant. The State University of New York at Stony Brook is served by the Stony Brook sewage treatment plant and shares an outfall with the Port Jefferson sewage treatment plant at the southern end of Port Jefferson Harbor. For regulatory (shellfish) purposes, the NYSDEC distinguishes between Port Jefferson Harbor North, Port Jefferson Harbor South, and Conscience Bay (fig. 1).

Land use in the surrounding area is primarily low- to medium-density residential, with commercial and industrial uses on the southern ends of Port Jefferson Harbor and Setauket Harbor. There are low-density residential areas on the eastern side of Port Jefferson Harbor and on the western and southern sides of Conscience Bay. Conscience Bay is connected to the northern side of Port Jefferson Harbor by the Narrows, which is a narrow channel to the north of Strongs Neck. Conscience Bay differs from Port Jefferson Harbor in that it is shallow and surrounded primarily by low-density residential land use. Port Jefferson Harbor has deep navigation channels to support industrial and commercial facilities, including a ferry service operating between Port Jefferson Harbor and Bridgeport, Connecticut. Potential contributions of contaminated water to the harbor include storm runoff, groundwater inputs, and drainage of coastal ponds, creeks, and wetlands. Illicit boat discharges and effluent from the Port Jefferson and Stony Brook University sewage treatment plants are also possible contributors of fecal contamination to the harbor. Industrial activities on the western side of Port Jefferson Harbor directly discharge water used for cooling. Moderately to steeply sloping bluffs along the eastern edge of Port Jefferson Harbor are typical of the northern shore of Long Island and allow for rapid delivery of stormwater through direct runoff or storm drains. The sites where surface-water samples were collected (U.S. Geological Survey, 2020a) for the Port Jefferson Harbor complex (fig. 2; tables 1.1 and 2.1) are as follows:

Approach and Methods

Sampling.—Samples were collected and analyzed in accordance with methods detailed in Tagliaferri and others (2021). Groundwater and sediment samples in Port Jefferson Harbor were collected in November 2018. No groundwater or sediment samples were collected in Conscience Bay during this sample collection.

On the basis of the methodology used for this study, wet conditions were defined as more than 0.25 inch of precipitation in 24 hours or 0.50 inch in 48 hours, and dry conditions were defined as less than 0.25 inch of precipitation in 72 hours. Rainfall totaled 1.21 inches on July 17, 2018, and 0.11 inch on July 18, 2018, designating samples collected in Port Jefferson Harbor and Setauket Harbor on July 18, 2018, as summer wet samples (more than 0.25 inch within 24 hours). Winter wet samples were also collected in Port Jefferson Harbor and Setauket Harbor on April 1, 2019, with 0.26 inch of precipitation falling on March 31, 2019. Rainfall totaled 3.57 inches on September 25, 2018, and 0.39 inch September 26, 2018, designating samples collected in Conscience Bay on September 26, 2018, as summer wet samples. Samples were collected in Conscience Bay on April 9, 2019, with 0.13 inch of precipitation falling on April 8, 2019, and 0.04 inch falling on April 9, 2019, totaling 0.17 inch of rain in 24 hours. Although these measurements fall outside the wet sample criterion of 0.25 inch in 24 hours, samples collected on April 9, 2019, were considered to be winter wet samples. Rainfall values were obtained from the Stony Brook mesonet weather gages identified as “South P Lot” and “Health Science Center” weather stations (Stony Brook University Weather, 2021).

From June 2018 to April 2019, 13 sites were sampled within Port Jefferson Harbor and Setauket Harbor, yielding 43 surface-water samples (including 1 replicate and 1 blank sample), 2 groundwater samples, and 1 sediment sample collected concurrently with the groundwater sample at the Public Beach and Dog Park (fig. 2). Of the surface-water sites, five were considered source sites (waters flowing directly into the Port Jefferson Harbor complex from the land through culverts and groundwater discharge), and six were considered receptor sites (the harbor itself). Twenty surface-water samples were collected under wet conditions, and 18, under dry conditions; 20 surface-water samples were collected in the summer, and 18, in the winter.

From September 2018 to May 2019, 5 sites were sampled within Conscience Bay, yielding 25 surface-water samples (including 3 replicates and 2 blank samples). No groundwater or sediment sample was collected around Conscience Bay. One of the sample sites was considered a source site, and four were considered receptor sites. Ten samples were collected under wet conditions, and 10, under dry conditions; 10 samples were collected in the summer, and 10, in the winter.

Laboratory analysis.—Laboratory methods used for this study are detailed in Fisher and others (2020). Samples for FC were analyzed at the Suffolk County Department of Health Services Public Environmental Health Laboratory (PEHL) and the NYSDEC Marine Laboratory. Additional information was either observed or compiled to facilitate interpretation of data. Precipitation measurements were collected from the Stony Brook University weather stations designated as South P Lot and Health Science Center that are approximately 4.5 and 3.5 miles from the center of Conscience Bay, respectively (Stony Brook University Weather, 2021). Quality assurance for samples collected at the Stony Brook University weather stations are not available; however, the data from the samples collected at Stony Brook University weather stations were not used in any quantitative analyses.

Approximate tide stage based on tidal predictions available (National Oceanic and Atmospheric Administration, undated) was documented along with the sample information in the National Water Information System (NWIS; U.S. Geological Survey, 2020a). Land-cover and sewer district geographic information system coverages (Nassau County, 2015; Suffolk County, 2020; U.S. Geological Survey, 2020b) were used for data interpretation. These data support the MST results to increase confidence in the geographical source of the water and likely transport mechanisms of fecal contamination to Port Jefferson Harbor, Setauket Harbor, and Conscience Bay.

Results

Results of MST and FC were assessed based on concentrations, presence or absence, and relative abundance of host markers with respect to surface-water source and receptor sites. Differences in MST markers and FC concentrations were also compared seasonally and conditionally (that is, wet or dry), with dry weather discharges with high FC concentrations from municipal separate storm sewer system source sites being of particular concern.

Host Specific Markers

Reporting limits for the three MST markers (human, canine, and waterfowl) predominantly detected in Port Jefferson Harbor, Setauket Harbor, and Conscience Bay varied slightly because of analytical processes, such as required dilutions because of matrix interferences. Understanding occurrence of the predominant MST markers in the Port Jefferson Harbor complex can help prioritize locations for additional investigation and infrastructure improvement and provide a baseline for host contributions when prioritizing sites of concern for assessing FC and after total maximum daily load implementation.

HF183

In Port Jefferson Harbor and Setauket Harbor, the human marker was detected at least once in surface-water samples at each receptor site (with the exception of Mid Setauket Harbor where only one sample was collected during the study), which is indicative of continuous inputs from potential inland sources (including illicit connections to the stormwater drain system and the Port Jefferson sewage treatment plant) or sources from within the receptor (including boats and marinas). Concentrations of the human marker were of the same order of magnitude (hundreds of copies per 100 milliliters [mL]; table 1.2) at receptor sites (fig. 3). These detections within the harbor were close to the reporting limit and not associated with high FC concentrations, thereby making it unlikely that the source was from boats and marinas. The Culvert North of State Route 25A and sewage treatment plant outfall sites were the only source sites where the human marker was detected in all four samples; however, all sewage treatment plant outfall samples and winter samples at Culvert North of State Route 25A were associated with low to nondetectable FC concentrations. Samples from 13 of 17 surface-water source sites in Port Jefferson Harbor and Setauket Harbor had detections of HF183 above the reporting limit. The concentration of HF183 in samples collected from the same site were on the same order of magnitude in dry and wet weather (between hundreds and thousands of copies per 100 mL; table 1.2), except summer, dry weather samples collected at the source sites Culvert North of State Route 25A (46,000 copies per 100 milliliters [copies/100 mL]) and Old Mill Creek Culvert (19,000 copies/100 mL). FC counts were generally higher in all source sites during the summer as opposed to the winter, with median FC concentrations at 800 most probable number per 100 milliliters (MPN/100 mL) in the summer and 20 MPN/100 mL in the winter.

Conscience Bay had 5 positive HF183 detections of 20 total samples, all of which were at receptor sites. Two positive HF183 detections were in winter, dry weather samples, and the remaining three were in summer wet weather samples (fig. 4). Conscience Bay Northwest was the only receptor site within Conscience Bay with no positive HF183 detections. The median HF183 concentration for all positive detections was 460 copies/100 mL. All samples collected during winter in wet weather and summer during dry weather had concentrations below the HF183 reporting limit. HF183 concentrations for the two detections in winter, dry weather samples were 360 and 1,200 copies/100 mL, and for the three summer, wet weather samples, 390, 470, and 460 copies/100 mL (table 2.2; U.S. Geological Survey, 2020a).

BacCan

At Port Jefferson Harbor and Setauket Harbor, the canine-associated Bacteroides (BacCan) marker was detected at 11 of 17 total source samples and 1 of 21 total receptor samples (table 1). Positive BacCan samples at the source sites were very similar between the winter and summer seasons, with one more positive detection in wet conditions than in dry (fig. 5). Conscience Bay Narrows is the only receptor site with a positive BacCan detection in Port Jefferson Harbor (2,000 copies/100 mL; table 1.2). Concentrations of BacCan were highest at the Culvert North of Shore Road (24,000 copies/100 mL) site in the winter, wet weather sample followed by the Culvert North of State Route 25A (21,000 copies/100 mL) site in the summer, wet weather sample. All four samples taken at the Culvert North of State Route 25A and Culvert North of Shore Road sites yielded concentrations of BacCan above the reporting limit. BacCan was detected in the sediment sample (200 copies/100 mL; table 1.2) but not in the groundwater at the Public Beach and Dog Park site on the southeastern side of Port Jefferson Harbor.

Conscience Bay had seven positive BacCan detections, one detection at each of the five sampling sites plus one additional positive detection at the Conscience Bay at Mouth of Tidal Creek and Mill Pond at Culvert to Conscience Bay sites. Five of seven positive detections (one at each site) were in the summer, wet weather samples (fig. 6). BacCan concentrations were highest at the Mill Pond at Culvert to Conscience Bay site, the single source site in Conscience Bay, during the summer, dry weather sample (28,000 copies/100 mL; table 2.2); this was the highest concentration detected in the Port Jefferson Harbor complex, with the exception of the summer wet weather replicate sample (110,000 copies/100 mL). In Conscience Bay, the only winter detection of BacCan was in dry weather at the Conscience Bay at Mouth of Tidal Creek site and was the only winter sample associated with a high FC concentration. Every positive BacCan detection corresponded to a high FC concentration in Conscience Bay but not in Port Jefferson Harbor and Setauket Harbor.

GFD

The waterfowl-associated Helicobacter (GFD) marker was detected in 7 of 17 surface water samples taken at source sites and 1 of 21 samples for receptor sites at Port Jefferson Harbor and Setauket Harbor (table 1). All samples collected at the Culvert North of Shore Road site were positive for GFD, and the summer, dry weather sample at this site had the highest concentration of GFD in the Port Jefferson Harbor complex (8,300 copies/100 mL; table 1.2). All GFD detections at surface-water source sites in Port Jefferson Harbor and Setauket Harbor were of the same order of magnitude (thousands of copies per 100 mL). Concentrations in Port Jefferson Harbor and Setauket Harbor were closer to the reporting limit (hundreds of copies per 100 mL) in the single receptor site with a positive detection (Mouth of Setauket Harbor) and in both the groundwater and sediment samples at the Public Beach and Dog Park site (fig. 7).

Conscience Bay had 8 positive GFD detections of 20 total samples. Seven of eight positive samples were at receptor sites, three at the Mouth of Tidal Creek site, three at the Conscience Bay Northeast site, and one at the Mouth of the Narrows site (fig. 8). This is in contrast to Port Jefferson Harbor and Setauket Harbor where seven out of eight detections of GFD occurred in source samples. There were no positive GFD detections at the Conscience Bay Northwest site. The one positive GFD detection at a source site (Mill Pond at Culvert to Conscience Bay) was during the summer, dry weather sample. All detections were of the same order of magnitude (between hundreds and thousands of copies per 100 mL; table 2.2).

Classification of Source Sites

The sites in the Port Jefferson Harbor complex were assessed to better understand geographical sources and transport mechanisms to surface-water receptor sites and the relationship between FC and MST marker presence and abundance (table 2). Source sites were assigned a numeric classification as described in Tagliaferri and others (2021) to aid managers with potential remediation efforts (table 3). Class 1 sites have the highest levels of contamination and class 5 sites have the least. Sites designated as sources to the Port Jefferson Harbor complex are further characterized below.

Summary

The U.S. Geological Survey worked collaboratively with the New York State Department of Environmental Conservation to assess the potential sources of fecal contamination entering seven estuarine embayments across Long Island, New York, from June 2018 to July 2019. Water samples are routinely collected by the New York State Department of Environmental Conservation in Long Island embayments and analyzed for fecal coliform bacteria, an indicator of fecal contamination, to determine the closure of shellfish beds for harvest and consumption. Concentrations of fecal coliform signify the potential for pathogenic (disease-causing) bacteria to be present.

Host sources of fecal contamination in the Port Jefferson Harbor complex were assessed based on bacterial Bacteroides and Helicobacter microbial source tracking markers. Overall, human contributions were ubiquitous in receptor samples; canine and waterfowl markers were prevalent in surface water source samples in Port Jefferson Harbor and Setauket Harbor and in receptor samples in Conscience Bay. By evaluating the microbial source tracking markers alongside fecal coliform bacteria data and land-use information, geographical sources of fecal contamination discharging from the culverts and outfalls sampled as part of this study were better differentiated.

Among the potential sources of fecal contaminated water contributing to the Port Jefferson Harbor complex—groundwater, pond and wetland drainage, stormwater, marinas (boats), and the Port Jefferson sewage treatment plant—pond and wetland drainage and stormwater, direct and from stormwater conveyances, were the most likely transport mechanism of fecal contamination into the Port Jefferson Harbor complex. The most substantial source of fecal contamination to the Port Jefferson Harbor complex was found to be discharge from sites draining ponds and wetlands, particularly during the summer months when wildlife is more concentrated and tourism peaks. Fecal coliform bacteria at these site types are increased by stormwater runoff, which is another substantial source. Sites with high fecal coliform bacteria in the summer that is exacerbated by stormwater include the Culvert North of Shore Road, the Culvert North of 25A, Old Mill Creek and Mill Pond to Conscience Bay sites. Stormwater contributions of high fecal coliform bacteria can also be seen at the Outfall at Danfords Marina site.

Of the sites sampled for this study, four had persistently high fecal coliform bacteria concentrations in the summer, two of which had a consistent presence of human (HF183) and canine (BacCan) microbial source tracking markers (Old Mill Creek and Culvert North of State Route 25A). The Culvert North of State Route 25A, the Culvert North of Shore Road, the Old Mill Creek Culvert, and the Mill Pond Culvert to Conscience Bay sites were identified as sites that contributed a substantial source of fecal contamination to the Port Jefferson Harbor complex. Low concentrations of HF183 within Port Jefferson Harbor coupled with low to no detections of fecal coliform bacteria suggest that boats are not substantially contributing to fecal coliform bacteria. The absence of fecal coliform bacteria in groundwater samples that tested positive for the human microbial source tracking marker indicates that water from septic systems influences the harbor; however, it may not contribute to the elevated fecal coliform bacteria often detected. Further, the sandy sediment alongside Port Jefferson Harbor is unlikely to contribute fecal coliform bacteria when resuspended in the water column through tidal shifts or boat activity.