Geophysical and hydrologic analysis of an earthen dam site in southern Westchester County, New York

Abstract

Ninety percent of the drinking water for New York City passes through the Hillview Reservoir facility in the City of Yonkers, Westchester County, New York. In the past, several seeps located downslope from the reservoir have flowed out from the side of the steepest slope at the southern end of the earthen embankment. One seep that has been flowing continuously was discovered during an inspection of the embankment in 1999. Efforts were made in 2001 to locate the potential sources of the continuous flowing seep. In 2005, the U.S. Geological Survey, in cooperation with the New York City Department of Environmental Protection, began a cooperative study to investigate the relevant hydrogeologic framework to characterize the local groundwater-flow system and to determine possible sources of the seeps. The two agencies used hydrologic and surface geophysical techniques to assess the earthen embankment of the Hillview Reservoir. Between April 1, 2005 and March 1, 2008, water levels were measured manually each month at 46 wells surrounding the reservoir, and flow was measured monthly at three of the five seeps on the embankment. Water levels were measured hourly in the East Basin of the reservoir, at 24 of 46 wells, and discharge was measured hourly at two of the five seeps. Slug tests were performed at 16 wells to determine the hydraulic conductivity of the geologic material surrounding the screened zone. Estimated hydraulic conductivities for 25 wells on the southern embankment ranged from 0.0063 to 1.2 feet per day and averaged 0.17 foot per day. The two-dimensional resistivity surveys indicate a subsurface mound of electrically conductive material (low-resistivity zone) beneath the terrace area (top of dam) surrounding the reservoir with a distinct elevation increase closer to the crest. Two-dimensional shear wave velocity surveys indicate a similar structure of the high shear wave velocity materials (high-velocity zone), increasing in elevation toward the crest and decreasing toward the reservoir and toward the northern part of the study area. Water-quality samples collected from 12 wells, downtake chamber 1 of the reservoir, and two seeps detected the presence of arsenic, toluene, and two trihalomethanes. Water-quality samples collected at the two seeps detected fluoride, indicating a connection with reservoir water. Shallow wells on the southern embankment exhibited the largest seasonal water-level fluctuations ranging between 6 feet and 12 feet. The embankment is constructed from reworked low-permeability glacial deposits at the site. Water-level responses in observation wells within the embankment indicate that there is a shallow (approximately the upper 45 feet of the embankment) and a deep water-bearing unit within the embankment with a large downward vertical gradient between the shallow and deep water-bearing units. Precipitation strongly affected water levels in shallow wells, whereas the basin appears to be the main control on water levels in the deep wells. Seeps on the embankment slope appear to be caused by above-average precipitation that increases water levels in the shallow water-bearing unit, but does not easily recharge the deep water-bearing unit. Based on the data that have been analyzed, source water to the seeps appears to be primarily groundwater and, to a lesser extent, water from the East Basin of the reservoir.