Abstract

Fresh ground water on the North Fork of Long Island is contained within a series of hydraulically isolated lenses that decline in thickness eastward. In the recent past, several SCWA public supply wells on the North Fork have been impacted by elevated chloride concentrations. In an effort to determine the source of the high chlorides, the Suffolk County Water Authority (SCWA) studied the impacts of stormwater runoff at two well fields in Peconic- Mill Lane and Ackerly Pond Road- using samples obtained from vertical profile wells. Since the relative concentration of chloride is quite high in road salt, the bromide–chloride (Br/Cl) molar ratio in groundwater affected by road salting is much lower than the Br/Cl molar ratio in groundwater impacted by lateral or vertical saltwater intrusion. The Br/Cl molar ratios of selected well samples were analyzed to determine if the chloride increases observed in the public supply wells were due to upconing from overpumping or from road salting.

Results indicate that road salting has impacted both well fields. This is especially true at the Ackerly Pond Road well field, since the site is located in close proximity to a recharge basin which retains stormwater from a section of a major highway. Vertical profiles at Mill lane detected similarly abnormal Br/Cl molar ratios opposite the screen setting of the public supply wells, also indicating impacts from road salting. It is hoped that future changes in the operation of both well fields, as well as closer monitoring of water quality trends, will allow the SCWA to continue to use both well fields despite these water quality impacts.

Introduction

Ground water is the sole source of drinking water for residents of Suffolk County. Concerns as to the quality of drinking water are particularly acute on the North Fork, where freshwater is only available at shallow depths. Unfortunately, the groundwater quality of the shallow aquifer system has been deteriorating as a result of residential expansion, agricultural practices, and stormwater runoff. Denitrification and carbon adsorption systems have been constructed at several well fields to address water quality concerns. A contaminant of particular interest is road-deicing salt (NaCl), and the role it plays in elevating chloride levels within the upper glacial aquifer. This process is of concern at present because high concentrations have been detected in stormwater and in shallow ground water. The leaching of salts, particularly sodium chloride to groundwater is the primary public concern since drinking water contaminated by sodium can pose a health risk to individuals with sodium restricted diets (Koppleman, 1984). Nationally, state and local governments spend about $10 million each year to prevent and remediate problems of road salt contamination (S.R, 1991) In the recent past several Suffolk County Water Authority (SCWA) public supply wells on the North Fork of Long Island became impacted by elevated chloride concentrations, potentially due to road salt contamination, or salt-water intrusion. Seasonal variations in the thickness of the freshwater lens, lowering of water levels due to public supply and agricultural pumpage, and road salt contamination contribute to the vulnerability of the aquifer system to elevated chlorides. Hydrogeologic and geochemical knowledge of the watershed environment surrounding major roads is an important factor in evaluating the effects of highway runoff on local waters (Bank, 1996). In 2000, the Suffolk County Water Authority studied the effects of stormwater runoff on the water quality at the Mill Lane and Ackerly Pond Road wellfields in Peconic. This report will provide (1) a brief synopsis of the deteriorating water quality at Mill Lane and Ackerly Pond Road (2) the sampling and hydrogeologic techniques used to acquire the data and (3) the delineation between freshwater and water impacted by road runoff.

Background

Analyses of the chloride-bromide concentrations are an effective means to determine the source of elevated chloride concentrations as either seawater or roadsalt. In 2000, the Suffolk County Water Authority began an in depth study on the effects of stormwater runoff on the water quality at the Ackerly Pond Road, Southold, and Mill Lane, Peconic wellfields (fig. 1). Because these wellfields consist of shallow wells that are in close proximity to water bodies and a major road (County Route 48), SCWA was concerned with the vulnerability of these wells to contamination. The wellfields lie within the SCWA’s Southold Low water supply system, which extends from Mattituck Creek to Dam Pond on the North Fork of Long Island. The Hamlets serviced include, East Mattituck, Cutchogue, Peconic, Southold, Sterling, and East Marion. The locations of these sites are approximately mid-way between Peconic Bay and the Long Island Sound.

Figure 1

Hydrogeologic Setting

Earlier research by the USGS classified the North Fork aquifer system as part of the Freshwater Lens Setting. This is due in part to the freshwater lenses being bound laterally and below by seawater. As a result, the lenses are isolated from the rest of the Long Island fresh groundwater system and have no adjacent freshwater to provide recharge. The North Fork has a series of freshwater lenses that usually decline in thickness eastward. The unconfined part of the freshwater lenses is the upper glacial aquifer, and the horizontal hydraulic conductivity ranges from 40 to 750 ft per day. Drilling done within the Ackerly Pond Road wellfield has shown the unconfined part of the aquifer to be underlain by fine to coarse-grained sand and gravel deposits. Geologic logs from production well PW-1 indicated that a clay layer is present at a depth of approximately 100 feet below grade (approximately 80 feet below msl). Research done by the USGS has indicated that this clay boundary is present throughout the Southold region at similar elevations. Depth to water within the wellfield is approximately 20 feet below grade, therefore, there are approximately 80 feet of saturated thickness above the lower clay unit. The Ackerly Pond Road Well Field is located near a groundwater flow divide, therefore the direction of groundwater flow in the well field is not known with certainty (LB&G, 1999). Average annual precipitation for the area is approximately 45 inches(Simmons, 1986). During the primary recharge period (October 15 through May 15), the normal precipitation for the Southold area is approximately 28.7 inches. Studies by the Cornell University Station (located northwest of the study area) indicate that 75 – 90 percent of the rainfall during the primary recharge period actually recharges the groundwater system. Movement of the freshwater/saltwater interface through pumping results in a zone of diffusion and disproportionate withdrawals from public supply or agriculture can result in saltwater upconing, resulting in contamination of the freshwater supply(Cartwright, 1997). Other sources of groundwater contamination in this hydrogeologic setting are surface contaminants, such as pesticides and fertilizers used on farms.

Br/Cl Molar Ratio Applications

Since the majority of precipitation on Long Island originates from seawater and has an average molar Br/Cl ratio of 1.54 nM/uM (nanomol/micromole), the Br/Cl ratio in precipitation will be similar to that of ocean water (Schoonen, 1995). As a result, molar Br/Cl ratio applications have been extremely useful in water quality analysis on Long Island. Analyzing the deviations in the Br/Cl ratio can be used to determine source of contamination, because certain contaminants will result in a Br/Cl ratio above or below that of normal precipitation. Road salt, which in Suffolk County consists of NaCl (halite), is an ever-increasing problem affecting shallow aquifers and is often mistaken for seawater encroachment. Analyzing molar Br/Cl ratios allow for differentiation between the two contamination sources. Since seawater has an average molar Br/Cl ratio that is relatively constant, a wellfield impacted by saltwater encroachment will still show the Br/Cl ratio equal to or minimally higher than that for seawater (Schoonen, 1995). By contrast halite will result in a molar Br/Cl ratio orders of magnitude lower than seawater, due to the exclusion of Br from the NaCl crystal lattice (Schoonen 1995). Since some pesticides contain Br, analyzing the molar Br/Cl ratio in rural areas may be useful in determining pesticide contamination.

Research Strategy

Since groundwater quality varies throughout the North Fork, the first objective was to accurately assess water quality at the Ackerly Pond Road and Mill Lane, well fields Peconic. These wells, which are located in the Town of Southold region (Fig. 1) were sampled in order to determine 1) the quality of ambient local water, and 2) the effects of nonpoint contaminant sources and land use (Brown, 1998). In addition, historical pumpage vs. chloride data of the Mill Lane and Ackerly Pond production wells were analyzed to allow for temporal analysis of pumpage induced fluctuations in chloride levels (fig2). Vertical profiles collected from both January and July, 2000 drilling programs allowed for both spatial and temporal analysis of Na+, Br, and Br/Cl ratios During the July 2000 drilling program, the initial undisturbed water quality of the boreholes at Mill Lane was analyzed using vertical profiles. The temporary boreholes were drilled with a Failing F-10 drill rig with hollow stem augers. The cutter head of the lead auger was sealed with a plastic plug to prevent leakage into the augers. Subsequent auger sections were sealed at the joints by O-rings to prevent groundwater infiltration. The augers were advanced through the unsaturated zone into the water table. This methodology allows the auger string to act as a well casing and prevent a collapse of the borehole. Drilling stopped at 170 feet due in part to the thick clay layer encountered at that depth. A 10-foot section of 2-inch diameter slotted PVC screen was lowered into the borehole and used to collect samples. Water samples were collected in 10-foot increments beneath the water table as the augers were raised. The profile samples were purged using a stainless steel pump and teflon lined tubing until pH, water temperature and specific conductance stabilized before sampling. Decontamination of pumping equipment was conducted prior to sampling each borehole to prevent cross contamination. The same methodology was used to collect samples at Ackerly Pond during the March 2002 drilling program. Due to the differences in the depths of the clay layer between Mill Lane and Ackerly Pond Road , the vertical profile was drilled to depth of only 90 feet. Samples were stored in HDPE (High-Density Polyethylene) containers and analyzed the same day. Br, Cl and N0₃concentrations weredetermined using ion chromatography. A DX 500 system with AS4S-SC column and a standard sodium carbonate sodium bicarbonate eluant were used. Conductivity was used to determine dilution of samples. In addition to the vertical profile samples, the Br content of road salt was also determined.

Figure 2 Pumpage VS. Chloride (bar represents chloride data in mg/l)

Sample Results

The molar Br/Cl data analyzed from the sites were summarized in table 1 and figure 4b. The tables and vertical profile layouts created on excel aided in the visualization of the numerical data from the vertical profiles

Mill Lane

The sample results from profile S-116843 in July 2000(table 2) indicated that at a depth of 90 feet, chloride concentrations reached a high of 116 mg/L. This abnormally high concentration was also present at a depth of 100 feet (115.4 mg/L), but subsided at a depth of 150 feet (50.2 mg/L). The sample data from profile S-116844 in July 2000 did not detect any unusual chloride concentrations. Chlorides within the profile had an average concentration of 27 mg/L. Samples collected approximately seven months earlier showed abnormal chloride concentrations at a depth of 80 feet (122.7 mg/L) and 90 feet (132.9 mg/L). The sample data collected from profile S-116845 had an average chloride concentration of 37 mg/L. Average chloride concentrations for production well no. 1 were 68 mg/L. during the month of December 2000 and increased to 73 mg/L during the month of January 2001. Chloride concentrations for production well no. 2 averaged 40 mg/L for the month of December 2000 and 44 mg/L during the month of January 2001.

Fig 3

Table 2 Mill Lane Vertical Profile Results (chlorides in mg/l)

The Br/Cl molar ratios of the Mill Lane profiles are summarized in Table 1.

The vertical profiles show a distinct difference in Br/Cl ratios at depths where the unusual chloride concentrations were detected. The sample results from profile S-116843 indicated at a depth of 90 feet, the Br/Cl ratio is .3 nM/uM (table 1). This unusually low ratio was also present at a depth of 100 feet (.35 nM/uM), but improved at a depth of 120 feet (1.8 nM/uM). Although the sample data from profile S-116844 did not detect any unusual chloride concentrations, the Br/Cl ratio was low at a depth of 80 feet (.79 nM/uM) to 90 feet (.69 nM/uM). Samples collected approximately seven months earlier showed abnormal Br/Cl ratios at a depth of 80 feet (.40 nM/uM) and 90 feet (.30 nM/uM). The sample data collected from profile S-116845 indicated low Br/Cl ratios at depth of 100 feet (.4 nM/uM) and 120 feet (.35 nM/uM).

Ackerly Pond

The sample results from profile 1 (fig. 4a & 4b) indicated that at a depth of 60 feet, chloride concentrations reached a high of 147.2 mg/l. The corresponding molar Br/Cl ratio at this depth was .45 nM/uM (fig. 4a). This abnormally high chloride concentration was present at most depths, but subsided at a depth of 150 feet (62.2 mg/l). The sample data collected from profile 2 had an average concentration of 32.3 mg/l. the corresponding Br/Cl ratio had an average concentration of 1.24 nM/uM. The sample data collected from profile 3 detected chloride concentrations as high as 355.7 mg/l at 90 feet. The molar Br/Cl ratio at that depth was .14 nM/uM.

Figure 4a. Bromide and Chloride concentrations from Ackerly Pond Road vertical profiles

Figure 4b. Bromide and Chloride molar ratios from Ackerly Pond Road vertical profiles

Discussion

Figure 2 shows the chloride concentration vs. pumpage for Ackerly Pond Road Production Wells 1 and 2 over time. As shown, chloride concentrations at PW-1 have trended upward since may 2000. Two notable spikes in chloride concentrations were observed at PW-1, one in December 2000 and the other at the end of July 2001 through August 2001. The chloride spikes correspond with a reduction in pumpage. Therefore it is reasonable to conclude that these increases in chloride concentrations are representative of a source other than seawater upconing. Low Br/Cl molar ratios are detected in wells at close proximity to the recharge basin, which is the likely source of the chloride concentrations. The likely cause for the moderate increase in chloride concentrations in PW-2 is the drastic reduction in pumpage at PW-1 during the winter of 2000, allowing for a stronger gradient created by the operation of PW-2 to induce additional chlorides to the southern portion of the well field. Road salt contamination may also be a problem afflicting the wellfield at Mill lane Peconic. The vertical profiles indicate that, at depths ranging from 80 –120 feet the molar Br/Cl ratio is consistent with road salt (Tables 1 and 2). This depth corresponds to the screen setting of the production well. Pumping of the production well would be expected to draw in contaminants (such as road salt) from the surrounding area and the data collected supports the idea of surface contaminants affecting the wellfield. The results of this study shows that Br/Cl ratios are an effective means of identifying road salt contamination within wellfields and differentiating between road salt and natural seawater.

References

Bank, F.G., 1996, “Transport planning- the watershed connection”, Proceedings of Watershed ’96, a National Conference on Watershed Management and Protection: Baltimore, Md., U.S. Environmental Protection Agency and the Water Environment Federation.

Brown, Craig J. and Scorca, Michael P. (in press) “Effects Of Road Salting On Stormwater and Groundwater Quality at the East Meadow Brook Headwaters Area, Nassau County, Long Island, New York” ”, Geology of Long Island and Metropolitan New York, Programs with Abstracts, April 22, 1995, Long Island Geologists, Stony Brook, New York.

Cartwright, Richard A., 1997, “Hydrogeologic-Setting classification for Suffolk County, Long Island, New York (with results of selected aquifer-test analyses)”, United States Geological Survey, Water Resources Report 96-457.

Koppleman, Lee A. “Non Point Source Management Handbook”, Long Island Regional Planning Board 1984

Leggette, Brashears & Graham, Inc., 1999 Letter Report to Suffolk County Water Authority “Middle Road (North Road C.R. 48) Well Field Town of Southold, New York”

NOAA, “Climatological Data Annual Summary, New York, 1997”, Volume 109, Number 13.

Schoonen, Martin A.A., Devoe, Vicki and Brown, Craig J. (in press) “Bromide in Long Island Groundwaters and Surface Waters”, Geology of Long Island and Metropolitan New York, Programs with Abstracts, April 22, 1995, Long Island Geologists, Stony Brook, New York.

Schubert, Christopher, E. 1,999, “Ground-Water Flow Paths and Traveltime to Three Small Embayments within the Peconic Estuary, Eastern Suffolk County, New York”, United States Geological Survey.

Simmons, Dale, 1986, “Geohydrology and Ground-Water Quality on Shelter Island, Suffolk County, New York”, 1983-84, United States Geological Survey, Water Resources Report 85-4165.

Stumm, Frederick and Lange, Andrew D (in press) “Hydrogeologic Framework and Extent of Saltwater Intrusion On The Manhasset Neck Peninsula, Long Island, New York, by use of Hydrogeologic and Geophysical Methods”, Geology of Long Island and Metropolitan New York, Programs with Abstracts, April 20, 1996, Long Island Geologists, Stony Brook, New York.

“Special Report 235: Highway Deicing: Comparing Salt and Calcium Magnesium Acetate.” TRB, National Research Council, Washington D.C., 1991

Suffolk County Department of Health Services, 1997, “Water Table Contours and Locations of Observation Wells in Suffolk County, New York, March 1997”.

Table 1 Br/Cl Molar Ratio of Vertical Profiles