Since 1934, the Delaware and Raritan Canal has been used to transfer water from the Delaware River Basin to the Raritan River Basin. The water transported by the Delaware and Raritan Canal in New Jersey is used primarily for public supply after it has been treated at drinking-water treatment plants located in the Raritan River Basin. Recently (1999), the raw water taken from the canal during storms has required increased amounts of chemical treatments for removal of suspended solids, and the costs of removing the additional sludge or residuals generated during water treatment have increased. At present, action to control algae is unnecessary.

The water quality of the Delaware and Raritan Canal was studied for approximately 16.5 months from mid-January 1998 through May 1999 to determine whether changes in water quality along the length of the canal are associated with storms. Nine water-quality constituents, and field measured specific conductance and turbidity were statistically tested.

Instantaneous or grab samples of water were collected from the Delaware and Raritan Canal after five storms and during four nonstorm events. Median values of water-quality constituents in samples collected immediately after storms and during nonstorm conditions when statistically compared by sampling location were not significantly different. Therefore, the data were combined or aggregated to eliminate one of the two explanatory variables, either individual sampling sites or the two types of sampling events, in order to generate a sample population large enough to show statistically significant differences. After combining sampling events, only the median concentration of suspended organic carbon, and field measured specific conductance and turbidity, were significantly different among sampling sites. Median concentrations of total and filtered ammonia plus organic nitrogen, total phosphorous, turbidity, ultraviolet absorbance at 254 nanometers, and dissolved organic carbon in samples collected after storms were significantly greater than in samples collected during nonstorm conditions, when the sampling locations were aggregated in the statistical analysis. Methyl tert-butyl ether, the most frequently detected volatile organic compound (VOC), was detected in 55 of 80 samples. The highest concentration of methyl tert-butyl ether, 3.2 micrograms per liter, was measured in a sample collected during nonstorm conditions.

The median of the continuously monitored specific conductance during nonstorm conditions at Port Mercer, N.J., increased by approximately 3 to 4 ?S/cm (microsiemens per centimeter) (1.5 to 2 percent of the median specific conductance) relative to that at the nearest upstream site, at Lower Ferry Road. The land use in the influent basins for this reach of the Delaware and Raritan Canal is primarily urban. One possible source of water with high specific conductance is either domestic or industrial wastewater that continuously discharges into pipes, then empties into the canal. Another possible source is ground water from an area within this reach where the elevation of the water table is higher than that of the water surface of the Delaware and Raritan Canal. The median continuously monitored specific conductance measured during nonstorm conditions at the Route 18 Spillway site increased relative to that of the nearest upstream site, Ten Mile Lock, by approximately 3 to 4 ?S/cm. The mean net change in continuously monitored specific conductance for this reach during storms also increased. Land use in the two largest influent basins within this reach, the Borough of South Bound Brook and Als Brook, is predominantly urban.

The mean and median of continuously monitored turbidity varied along the length of the canal. In the reach between Raven Rock and Lower Ferry Road, the mean and median for continuously monitored turbidity during the study period increased by 7.2 and 6.2 NTU (nephelometric turbidity units), respectively. The mean of continuously monitored turbidity decreased downstream from Lower Ferry Road to Ten Mile Lock. Turbidity could increase locally downstream from influent streams or outfalls, but because the average velocity of water in the canal is low, particles that cause turbidity are not transported appreciable distances. In the reach between Ten Mile Lock and the Route 18 Spillway, the mean and median of the continuously monitored turbidity changed less than 0.5 NTU during the period of record. The small change in turbidity in this reach is not consistent with an average velocity for the reach; the average velocity in this reach was the lowest in all of the reaches studied. The expected decrease in turbidity due to settling of suspended solids is likely offset by turbid water entering the canal from influent streams or discharges from storm drains. Field observation of a sand bar immediately downstream from the confluence of Als Brook and the canal confirmed that the Als Brook drainage basin has contributed stormwatergenerated sediment to the canal that could reach the monitor located at the Route 18 Spillway and the raw water intakes for two drinking-water treatment plants.