The 13 major south-shore streams in Nassau and Suffolk Counties, Long Island, New York with adequate long-term (1971-97) water-quality records, and 192 south-shore wells with sufficient water-quality data, were selected for analysis of geographic, seasonal, and long-term trends in nitrogen concentration. Annual total nitrogen loads transported to the South Shore Estuary Reserve (SSER) from 11 of these streams were calculated using long-term discharge records. Nitrogen loads from shallow and deep ground water also were calculated using simulated ground-water discharge of 1968-83 hydrologic conditions.

Long-term declines in stream discharge occurred in East Meadow Brook, Bellmore Creek and Massapequa Creek in response to extensive sewering in Nassau County. The smallest longterm annual discharge to the SSER was from the westernmost stream, Pines Brook, which is in an area in which the water table has been lowered by sewers since 1952. The three largest average annual discharges to the SSER were from the Connetquot River, Carlls River, and Carmans River in Suffolk County; the discharges from each of these streams were at least twice those of the other streams considered in this study.

Total nitrogen concentrations in streams show a geographic trend with a general eastward increase in median total nitrogen concentration in Nassau County and a decreasing trend from Massapequa Creek eastward into Suffolk County. Total nitrogen concentrations in streams generally are lowest during summer and highest in winter as a result of seasonal fluctuations in chemical reactions and biological activity. The greatest seasonal difference in median total nitrogen concentration was at Carlls River with values of 3.4 and 4.2 mg/L (milligrams per liter) as N during summer (April through September) and winter (October through March), respectively. Streams affected by the completion of sewer districts show long-term (1971-97) trends of decreasing total nitrogen concentration and streams showing an increase in total nitrogen concentration are in unsewered areas with increased urbanization.

Discharges from shallow ground water (upper glacial aquifer) and deep ground water (upper part of Magothy aquifer) were simulated from a ground-water-flow model calibrated to steadystate (1968-83) conditions. Simulated discharges from shallow-ground-water system in Nassau County were 10,700 Mgal/yr (million gallons per year) or 40,500,000 m³/yr (cubic meters per year), and those from Suffolk County were 52,300 Mgal/yr or 198,000,000 m³/yr. Discharges from deep-ground-water system in Nassau County were 4,900 Mgal/yr or 18,500,000 m³/yr, and those in Suffolk County were 12,700 Mgal/yr or 48,200,000 m³/yr.

Ground-water concentrations of nitrogen decrease with depth and from west to east. The shallow ground water median nitrogen concentration for each county was determined using 1,155 samples collected at 167 shallow wells (125 feet deep or less) within 1 mile of the shore. The deep ground water median nitrate concentration (nitrate represented almost all of the total nitrogen) for each county was determined using 112 samples collected at 25 deep wells (greater than 125 feet deep) within 1 mile of the shore. The median nitrogen concentration for the shallow and median nitrate concentration for the deep ground water in Nassau County were 3.85 and 0.15 mg/L as N, during 1952–97; the corresponding concentrations for Suffolk County were 1.74 and <0.10 (less than 0.10) mg/L as N, during 1952–97.

Nitrogen loads discharged from streams to the SSER for each year during 1972–97 were calculated as the annual total nitrogen concentration multiplied by the annual discharge. These values were calculated only for the seven streams for which sufficient data were available. The largest long-term (1972–97) average annual nitrogen load from Carlls River was 104 ton/yr or 94,300 kg/yr—about twice that of Connetquot River (54 ton/yr or 48,900 kg/yr) and over three times that of Carmans River (33 ton/yr or 29,900 kg/yr). The smallest annual mean nitrogen load was from Pines Brook, which has the lowest annual mean discharge of all streams analyzed.

The nitrogen load carried to the SSER by ground-water discharge in shallow-ground-water system in Nassau and Suffolk Counties was calculated as the simulated discharge for each county multiplied by the respective median nitrogen concentration, and loads from deep-ground-water system were calculated as the simulated discharge for each county multiplied by the respective median nitrate concentration. All discharges were obtained from the U.S. Geological Survey's Long Island ground-water-flow model. The resultant nitrogen loads discharged to the SSER from shallow ground water were 172 ton/yr (156,000 kg/yr) from Nassau County and 380 ton/yr (345,000 kg/yr) from Suffolk County; equaling 552 ton/yr entering the SSER. Those from deep ground water were 3 ton/yr (2,700 kg/yr) from Nassau County and <0.5 ton/yr (480 kg/yr) from Suffolk County; equaling about 3.5 ton/yr entering the SSER.

The sum of both stream loads and groundwater loads results in the total load to the SSER. The largest calculated total nitrogen load entering the SSER from both streams and ground water occurred in 1979 with a total load of 1,260 ton/yr (1,140,000 kg/yr). The smallest calculated nitrogen load entering the SSER occurred in 1995 with a total load of 725 ton/yr (658,000 kg/yr).