Streamflow in the Esopus Creek watershed is altered by two major watershed management activities carried out by the New York City Department of Environmental Protection as part of its responsibility to maintain a water supply for New York City: (1) diversion of water from the Schoharie Creek watershed to the Esopus Creek through the Shandaken Tunnel, and (2) impoundment of the Esopus Creek by a dam that forms the Ashokan Reservoir and subsequent release through the Catskill Aqueduct. Stakeholders in the Catskill region are interested and concerned about the extent to which these watershed management activities have altered streamflow, especially low and high flows, in the Esopus Creek. To address these concerns, natural (in the absence of diversion and impoundment) daily discharge from October 1, 1931, to September 30, 2012, was estimated for the U.S. Geological Survey streamgages at Coldbrook (station number 01362500), downstream of the Shandaken Tunnel discharge, and at Mount Marion (01364500), downstream of the Ashokan Reservoir.

A multiple linear regression approach, using nearby discharge records from unimpounded streams as predictive variables, was applied to estimate natural discharge at the Coldbrook streamgage. Estimated values of natural daily discharge at the Coldbrook streamgage were lower than values of gaged daily discharge throughout the flow range at this site. At moderate- and low-flow conditions, gaged daily-discharge values were about two to three times greater than natural daily-discharge estimates, whereas the difference between the two records was less than 5 percent for the highest 1 percent of daily-discharge values. These results indicate that Shandaken Tunnel discharge has a minor effect on flooding in the Esopus Creek Basin. However, a difference of 5 percent is within the uncertainty of the regression-based natural discharge estimates for Coldbrook; thus, it cannot be stated with certainty that the Tunnel has on average any effect on flow for the highest 1 percent of daily discharge values.

Natural discharge at the Mount Marion streamgage was estimated by summing the natural discharge estimated for the Coldbrook streamgage and the discharge estimated for the intervening basin area through application of the New York Streamflow Estimation Tool, recently developed for estimating unaltered streamflow at ungaged locations in the State. Estimates of natural daily discharge at the Mount Marion streamgage were about three times greater than gaged daily discharge throughout the moderate- to low-flow range from October 1, 1970, to September 30, 2012, the period of record for full water years at this streamgage. The relative difference between the two discharge time series declined as flow increased beyond the moderate range, but gaged daily discharge was still 25 to 43 percent less than estimated natural daily discharge for the high-flow metrics calculated in this analysis, and the mean relative difference was 43 percent for the annual 1-day maximum discharge. Overall, these estimates of natural discharge reflect the absence of effects of the Shandaken Tunnel and Ashokan Reservoir on flows in the Esopus Creek over broad time frames. However, caution is warranted if one is attempting to apply the natural estimates at short time scales because the regression prediction intervals indicate that uncertainty at a daily time step ranges from about 40 to 80 percent.