Turbidity and suspended sediment in the upper Esopus Creek watershed, Ulster County, New York

Abstract

Suspended-sediment concentrations (SSCs) and turbidity were measured for 2 to 3 years at 14 monitoring sites throughout the upper Esopus Creek watershed in the Catskill Mountains of New York State. The upper Esopus Creek watershed is part of the New York City water-supply system that supplies water to more than 9 million people every day. Turbidity, caused primarily by high concentrations of inorganic suspended particles, is a potential water-quality concern because it colors the water and can reduce the effectiveness of drinking-water disinfection. The purposes of this study were to quantify concentrations of suspended sediment and turbidity levels, to estimate suspended-sediment loads within the upper Esopus Creek watershed, and to investigate the relations between SSC and turbidity. Samples were collected at four locations along the main channel of Esopus Creek and at all of the principal tributaries. Samples were collected monthly and during storms and were analyzed for SSC and turbidity in the laboratory. Turbidity was also measured every 15 minutes at six of the sampling stations with in situ turbidity probes. The largest tributary, Stony Clove Creek, consistently produced higher SSCs and turbidity than any of the other Esopus Creek tributaries. The rest of the tributaries fell into two groups: those that produced moderate SSCs and turbidity and those that produced low SSCs and turbidity. Within those two groups the tributary that produced the highest SSCs and turbidity varied from year to year depending on the hydrologic conditions within each subwatershed. During the 3-year study, Stony Clove Creek accounted for an average of 40 percent of the annual suspended-sediment load measured at the upper Esopus Creek watershed outlet at Coldbrook, more than all of the other measured tributaries combined. The other tributaries to the upper Esopus Creek, taken together, accounted for an average of about 20 percent of the load at Coldbrook during 2010 and 2011, when most of the tributaries were sampled. Woodland Creek, the third largest tributary in the watershed, also accounted for a substantial amount of the load at Coldbrook, an average of 10 percent during the 3 years. Stony Clove Creek appeared to be a persistent source of sediment to Esopus Creek; it had the highest sediment yield (load per unit area) of all monitoring sites, including the outlet at Coldbrook. Discharge, SSC, and turbidity were strongly related at the Coldbrook site but not at every monitoring site. In general, relations between discharge and SSC and turbidity were strongest at sites with high SSCs, with the exception of Stony Clove Creek. Stony Clove Creek had high SSCs and turbidity regardless of discharge, and although concentrations and turbidity values generally increased with increasing discharge, the relation was not strong. Five of the six sites used to investigate the relations between SSC and laboratory turbidity had a coefficient of determination (r 2 ) greater than 0.7. Relations were not as strong between SSC and the turbidity measured by in situ probes because the period of record was shorter and therefore the sample sizes were smaller. Data from in situ turbidity probes were strongly related to turbidity data measured in the laboratory for all but one of the monitoring sites where the relation was strongly leveraged by one sample. Although the in situ turbidity probes appeared to provide a good surrogate for SSC and could allow more accurate calculations of suspended-sediment load than discrete suspended-sediment samples alone, more data would be required to define the regression models throughout the range in discharge, SSCs, and turbidity levels that occur at each monitoring site. Nonetheless, the in situ probes provided much greater detail about the relation between discharge and turbidity than did the grab samples and storm samples measured in the laboratory.