Borehole-geophysical and hydraulic investigation of the fractured-rock aquifer near the University of Connecticut Landfill, Storrs, Connecticut, 2000 to 2001

Abstract

An integrated borehole-geophysical and hydraulic investigation was conducted at the former landfill area near the University of Connecticut in Storrs, Connecticut, where solvents and landfill leachate have contaminated a fractured-bedrock aquifer. Borehole-geophysical techniques and hydraulic methods were used to characterize the site bedrock lithology and structure, fractures, and hydraulic properties. The geophysical and hydraulic methods included conventional logs, borehole imaging, borehole radar, flowmeter under ambient- and stressed hydraulic conditions, and discrete-zone hydraulic testing, sampling, and monitoring. The conventional geophysical-logging methods included caliper, deviation, electromagnetic induction, gamma, specific conductance, and fluid temperature. The advanced methods included optical and acoustic imaging of the borehole wall, heat-pulse flowmeter, and directional radar reflection. Borehole-geophysical methods were used to further define conductive features identified with surface-geophysical methods in the first phase of the investigation. The results of the surface- and borehole-geophysical logging were evaluated in an iterative and integrated manner to develop a conceptual model of ground-water flow at the site. The rock type, foliation, and fractures at the site were characterized from high-resolution optical televiewer (OTV) images of rocks penetrated by the boreholes and were compared to drilling logs and conventional geophysical logs. The rocks are interpreted as fine- to mediumgrained quartz-feldspar-biotite-garnet gneiss and schist with local intrusions of quartz diorite and pegmatite and minor concentrations of sulfide mineralization similar to rocks described as the Bigelow Brook Formation on regional geologic maps. Layers containing high concentrations of sulfide minerals appear as high electrical conductivity zones on electromagnetic-induction and borehole-radar logs. Foliation in the rocks generally strikes to the southwest and northeast, and dips to the northwest and southeast consistent with previous investigations in this area. The orientation of foliation, however, varies locally and with depth in some of the boreholes. These results are consistent with geologic mapping that has identified small-scale folding. The orientations of the transmissive fractures identified in the six boreholes logged for this investigation are similar to the fracture orientations mapped in a previous investigation. Many of these fractures are oriented with a north-northwest strike and have a shallow dip to the west. Other transmissive fractures have a southwest strike and dip at shallow angles to the northwest, and some strike roughly east-west and dip to the north and south. Flowmeter logging was used to identify transmissive fractures and to estimate the hydraulic properties in the boreholes. Ambient down flow was measured in one borehole, and ambient up flow and down flow were measured in another borehole. The other four bedrock boreholes did not have measurable vertical flow. Under low-rate pumping conditions (0.25 to 0.5 gallons per minute), one to three inflow zones were identified in each well. Commonly, fractures that are active under ambient conditions contribute to the well under pumping conditions. The ambient conditions were incorporated into the determination of the relative proportions of transmissivity. Specific capacity and transmissivity were determined for these open-hole low-rate pumping tests. Quasi-steady-state water levels were reached in four of the boreholes, including MW201R, MW204R, MW302R, and W202-NE. When pumped at low-rate conditions for 0.5 to 4 hours, the specific capacity ranged from 0.03 to 0.18 gallons per minute per foot. The open-hole transmissivity estimates ranged from 4.9 to 30 feet squared per day (ft2/d). Open-hole transmissivity was determined for boreholes that did not reach quasi-steady-state conditions under low-rate pumping conditions. Transmissivity was estimated for MW201R, MW202R, and MW203R using non-equilibrium methods, pumping rate, and the transient drawdown data to estimate the open-hole transmissivity. Transmissivity in these boreholes ranged from 0.98 to 3.2 ft 2 /d. The transmissivity and head of individual fractures or zones of fractures were estimated from heat-pulse flowmeter data acquired under ambient and stressed conditions. In the absence of ambient flow, data from two profiles of heat-pulse flowmeter data under two different stressed conditions were used to estimate the transmissivity and head of individual fracture zones. Only two boreholes, MW302R and W202-NE, had sufficient data for these analyses. The estimated transmissivity of individual transmissive zones ranged from 1.2 to 9.2 ft 2 /d. The transmissivity values determined by this numerical simulation method were less than the open-hole estimations, which were 15 and 30 ft 2 /d. Transmissivity also was measured directly over discrete intervals of the borehole using a straddle-packer apparatus and constant-rate pumping tests. Pumping rates were less than or equal to 0.25 gallons per minute. These discretezone single-hole pumping tests were conducted over a short period of time, usually about 30 minutes to 1 hour in duration. Pumping continued until the test zone reached a steady-state water level or until it was determined that the zone could not yield water at the pumped rate. The estimated transmissivity of individual transmissive zones ranged from about 0.21 to 11 ft 2 /d. The zone at a depth of 197 feet in W202-NE was the only zone that had discrete-interval testing with a straddle packer and sufficient heat-pulse flowmeter data for modeling the flow and estimating transmissivity and head. The two methods produced similar results. The straddle-packer method estimated a transmissivity of 4.7 ft 2 /d, and the heat-pulse flowmeter modeling results estimated a transmissivity of 6.9 ft 2 /d. A comparison of the transmissivity estimates indicate estimates typically are within an order of magnitude. The heat-pulse flowmeter methods used in this investigation to determine transmissivity of the boreholes and the individual fractures measure only the upper two or three orders of magnitude of transmissivity. Hence, other fractures in these boreholes permit the movement of water; their transmissivities, however, are lower than the detection limits of the methods that were used for this investigation and very small compared to the transmissive fractures that were studied. The data collected in this investigation were used to design discrete-zone monitoring systems for four of the boreholes used for monitoring. The results of the investigation are useful for refining the conceptual site model of ground-water flow, and for providing critical information for interpreting the results of water-quality sampling.