PROBLEM
As part of an effort to sustain a viable water-supply system for 8 million residents in New York City, and 1 million other residents in upstate New York that rely on City water, the New York City Department of Environmental Protection (NYCDEP) has requested a multi-disciplinary study by the U.S. Geological Survey (USGS) to determine the source(s) of water to surface-water sites (springs and expressions) in areas adjacent to the Catskill Aqueduct. In the area near the Delaware Aqueduct, which has been determined by the USGS to have a water-tunnel contribution to flooding issues during periods of high precipitation, the NYCDEP plans to repair the leaks with a grouting procedure. However, without monitoring of hydrologic conditions, there is no way to assess the efficacy of the grouting repairs.
BACKGROUND
New York City completed construction of the Ashokan Reservoir and Catskill Aqueduct in 1915 to meet the city’s need for a larger and more reliable water supply for its growing population (fig. 1). Work on the Ashokan Reservoir and Catskill Aqueduct was followed by the construction of the Schoharie Reservoir and Shandaken Tunnel in 1928; and finally, the Delaware Aqueduct system, in stages between 1944 and 1964 (New York City Department of Environmental Protection, 2016).
Depending on the location and engineering considerations, the aqueducts were constructed using either cut-and-cover or drill-and-blast methods. During aqueduct construction, fractured and karstic zones were identified in the Heldeberg limestone, Binnewater sandstone, and the High Falls shale that produced 200 to 2,200 gal/min along the Catskill Aqueduct, and over 9,000 gal/min along the Wawarsing section of the Delaware Aqueduct (fig 1; Berkey, 1911; Berkey and Fluhr, 1936; New York City Board of Water Supply, 1941).
Over time, flow paths have developed along the fractured and karstic zones, enlarging cavities, resulting in hydraulic connections between the pressurized water tunnel and surrounding bedrock that in some cases may have reached land surface. In recent years, leaks near Roseton, New York, along the Delaware Aqueduct, were estimated to be about 35 Mgal/d by the NYCDEP (James Canale, New York City Department of Environmental Protection, written commun., 2011; DiNapoli, 2007). Also recently, leaks in the Delaware Aqueduct have been determined to contribute to flooding issues during extended periods of higher-than-average precipitation amounts (Brown and others, 2012; Stumm and others, 2012).
The NYCDEP is planning to grout the smaller tunnel leaks in Wawarsing, and build a bypass tunnel in the areas of the larger leaks along the Delaware Aqueduct. This work is to be implemented in stages over the upcoming years.
As of December 2015, the USGS was monitoring a combined groundwater and surface-water network of 54 sites for the Delaware Aqueduct leak at Wawarsing:
- 2 surface water
- 29 unconsolidated aquifer wells (22 of which were drilled and installed by USGS)
- 23 bedrock wells (16 of which are domestic supply wells)
Until the water-tunnel shutdowns and repairs are to be implemented, the USGS will maintain a reduced network of 25 sites—17 bedrock and 8 unconsolidated aquifer wells in the Wawarsing study area. Well information such as aquifer and location will be used to determine which wells will be re-activated for monitoring during the tunnel-repair shutdowns.
OBJECTIVES
The purpose of this study is to: (1) characterize the hydrogeology and geochemistry of the bedrock and unconsolidated aquifer systems in the vicinity of the Catskill and Delaware aqueducts, (2) delineate the magnitude and extent of influence of the leaking aqueducts on the aquifer systems, (3) determine the source of springs in the vicinity of the aqueducts, and (4) provide data to evaluate the efficacy and effects of the planned repairs of the aqueducts.
APPROACH
The approach for this project focuses on the NYCDEP’s interest in determining the source(s) of water to surface-water sites (springs and other surface expressions of groundwater discharge) adjacent to the Delaware and Catskill Aqueducts, and will initially target the Wawarsing (Delaware Aqueduct) and High Falls (Catskill Aqueduct) study areas which were determined by the NYCDEP’s Autonomous Underwater Vehicle (AUV), and land inspections to be problematic.
The USGS will coordinate with the NYCDEP through quarterly or semi-annual meetings to address areas where hydrologic or geochemical data will be critical. As requested by the NYCDEP, the USGS will transfer raw provisional data to the NYCDEP at these meetings
WAWARSING
The USGS has decommissioned all pressure transducers from privately-owned overburden wells in the Wawarsing study area. The USGS will maintain the reduced hydrologic network in advance of, during, and after NYCDEP’s planned grouting or extended water-tunnel shutdown is to occur. During that time, the USGS will review all of the wells in the network to determine which wells will be critical to assessing the efficacy of the tunnel repairs. (fig. 2). Hydrologic data (water level and water temperature) collected following repair (grouting) of the leaks will be compared with that of the pre-repaired conditions to assess hydraulic connections before and after repairs. Monitoring of the updated network will resume about 1 to 2 months prior to the first aqueduct shutdown. One week prior to the shutdowns, the frequency of monitoring visits with recorder downloads will increase to once a week on the updated network. If the shutdowns last more than one month, the period of visits may be reduced to every two weeks. The USGS will decommission the entire Wawarsing monitoring network after the grouting repairs have been determined to be successful.
After grouting repairs are completed, water samples from the aqueduct intake and selected sites that were sampled between 2009 and 2014 will be resampled, and compared with previous sample results. The response time to previous tunnel shutdowns in Wawarsing ranged from 0.5 to 60 hours, samples will be collected during the shutdowns after the response time has elapsed. To help assess the efficacy of the grouting, the sources of the water in wells and surface expressions will be re-determined from water samples collected after the repair grouting. Water sources can be identified by unique chemical signatures and isotopic ratios. Sample analytes will include major and trace constituents, stable isotopes, dissolved gases, and age-dating tracers. The isotope ratios will include: 2H/H and 18O/16O of water, 34S/32S of SO42-, 13C/12C of dissolved inorganic carbon, and 87Sr/86Sr (Brown and others, 2012). Ten percent of the pertinent analytes will have replicate samples collected for quality assurance. All lab analyses will be performed by USGS laboratories except for 3H/3He noble gases, for which samples will be sent to the USGS contract lab, the Noble Gas Laboratory at the Lamont-Doherty Earth Observatory. The contract permits analysis of the full suite of noble gases and the associated analytical uncertainty in replicate samples, and the results include age interpretation by the USGS Reston Groundwater Dating Laboratory.
Advanced radio imaging tomography techniques (Korpisalo, 2011) in 2D and possibly 3D may be used to image fracture and leakage anomalies between boreholes and (or) surface expressions.
HIGH FALLS
The USGS will contact the owners of existing wells (Oral commun., NYCDEP) in the High Falls area to establish a monitoring network to collect hydrologic and water-quality data. Once the network has been established, and the need for supplemental wells determined, the USGS will select locations for a NYCDEP contracted observation-well drilling program within the study area. If NYCDEP cannot provide a driller, USGS may use the USGS Research-Drilling Program to install dedicated monitoring wells at an additional cost and under a different agreement than this agreement. Once installed, the observation-well network will be monitored for changes in groundwater elevation using vented pressure transducers that are programmed to measure and record water level, temperature, and conductivity (only at selected locations) at periodic (user-defined) intervals, typically 15 minutes. These data will be periodically calibrated using discrete manual water-level and temperature measurements during seasonal and (or) episodic field inspections. In addition, select monitoring wells will be equipped with real-time telemetry that will collect water-level elevation and temperature data every 15 minutes, transmit the data hourly to the USGS National Water Information System (NWIS) database, and be displayed in near-real time on NWISWeb (https://waterdata.usgs.gov/nwis). Synoptic hydrologic data will be collected before, during, and after the 10-week water-tunnel shutdowns planned for 2018, 2019, and 2020 to determine the amount of influence, if any, from the tunnels on water levels in wells and discharge from springs (described below) within the study area(s).
Water-quality samples will be collected from: (1) water influent to the Catskill Aqueduct from the Ashokan Reservoir and (2) available wells and springs during normal tunnel operations and during extended shutdowns or depressurizations (similar to the previous study on the Delaware Aqueduct (Brown and others 2012)). The water-quality samples will be analyzed for major and trace constituents, stable isotope ratios, dissolved gases, and age-dating tracers; which will be compared for trends during each event. The isotope ratios may include 2H/H and 18O/16O of water, 34S/32S of SO42-, 13C/12C of dissolved inorganic carbon, and 87Sr/86Sr. Age-dating tracers can provide information on the age of the water to help determine the source. Age dating tracers to be analyzed will include 3H and 3H-3He ratios, and SF6. Five percent of the samples will have replicate samples collected for quality assurance. All lab analyses will be performed by USGS laboratories except for 3H/3He noble gases, for which samples will be sent to the USGS contract lab, the Noble Gas Laboratory at the Lamont-Doherty Earth Observatory. The contract permits analysis of the full suite of noble gases and the associated analytical uncertainty in replicate samples, and the results include age interpretation by the USGS Reston Groundwater Dating Laboratory.
The USGS will log selected observation wells using established borehole-geophysical techniques to characterize the bedrock (lithologic changes, faults, fractures, karst or other transmissive features). Borehole-geophysical methods may include gamma, spontaneous-potential, single-point resistance, long-medium-short normal resistivity, electromagnetic induction, magnetic susceptibility, caliper, fluid (temperature, specific conductance, pH, dissolved oxygen, and redox), optical and acoustic televiewer, and flowmeter (Keys and McCary, 1971; Johnson and others 2002). Advanced radio imaging tomography techniques in 2D and possibly 3D may be used to image fracture and leakage anomalies between boreholes and (or) surface expressions. Surface-geophysical techniques may be used to help delineate hydrogeologic framework, including the extent of valley-fill sediments, as well as potentially detecting weaker zones (faults and fractures) that could serve as conduits for groundwater flow in the shallower bedrock and overburden deposits, which may connect with seepages at land surface.
If site conditions permit, the USGS will monitor flow from the springs using weirs or Parshall flumes instrumented with digital-data recorders. Up to two instrumented springs will also be equipped with real-time recorders and telemetry that will gather stage and temperature data and compute discharge every 15 minutes, transmit the discharge and temperature data hourly to the USGS National Water Information System (NWIS) database, and be displayed in near-real time on NWISWeb (https://waterdata.usgs.gov/nwis). A precipitation-measurement station may also be established to monitor local precipitation conditions during hydrologic events. Web-based cameras may also be installed to provide remote visual inspection of conditions at the sites equipped with real-time monitoring equipment.
References
Berkey, C.P., 1911, Geology of the New York City (Catskill) aqueduct: New York Education Department Bulletin no. 489, New York State Museum Bulletin 146, 283 p.
Berkey, C.P., and Fluhr, T.W., 1936, The geologic formations of the new Rondout-West Branch Tunnel of the Delaware Aqueduct with generalized geological sections: New York City Board of Water Supply, 59 p.
Brown, C.J., Eckhardt, D.A., Stumm, Frederick, and Chu, Anthony, 2012, Preliminary assessment of water chemistry related to groundwater flooding in Wawarsing, New York, 2009–11: U.S. Geological Survey Scientific Investigations Report 2012–5144, 36 p. (Also available at
http://pubs.usgs.gov/sir/2012/5144/.)
Fluhr, T.W., and Terenzio, V.G., 1984, Engineering geology of the New York City water supply system: New York State Geological Survey Open File Report 05.08.001, 183 p.
Johnson, C.D., Haeni, F.P., Lane, J.W., and White, E.A., 2002, Borehole-geophysical investigation of the University of Connecticut landfill, Storrs, Connecticut: U.S. Geological Survey, Water Resources Investigations Report 01-4033, 187 p.
Keys, W.S., and L.M. MacCary, 1971, Application of Borehole Geophysics to Water-Resources Investigations: U.S. Geological Survey Techniques of Water Resources Investigations, Book 2, Chapter E1, 126 p.
New York City Board of Water Supply, 1941, Proposed steel interliner construction, plate 1.
Stolarczyk, L.G., and Fry, R.C., 1986, Radio imaging method (RIM) or diagnostic imaging of anomalous geologic structures in coal seam waveguides, accessed February 16, 2017, at http://stolarglobal.com/wp-content/uploads/Downloads/Coal-Seam-Imaging-SME-Paper.pdf
Stumm, Frederick, Chu, Anthony, Noll, Michael, and Como, Michael, 2012, Preliminary analysis of the hydrologic effects of temporary shutdowns of the Rondout-west branch water tunnel on the groundwater-flow system in Wawarsing, New York, SIR 2012-5015, 48 p.
Project
Location by County
Ulster County, NY, Dutchess County, NY, Orange County, NY, Putnam County, NY, Westchester County, NY
