Kathryn M. Hess

Large-scale natural gradient tracer test in sand and gravel, Cape Cod, Massachusetts: 1. Experimental design and observed tracer movement

A large-scale natural gradient tracer experiment was conducted on Cape Cod, Massachusetts, to examine the transport and dispersion of solutes in a sand and gravel aquifer. The nonreactive tracer, bromide, and the reactive tracers, lithium and molybdate, were injected as a pulse in July 1985 and monitored in three dimensions as they moved as far as 280 m down-gradient through an array of multilevel samplers. The bromide cloud moved horizontally at a rate of 0.42 m per day. It also moved downward about 4 m because of density-induced sinking early in the test and accretion of areal recharge from precipitation. After 200 m of transport, the bromide cloud had spread more than 80 m in the direction of flow, but was only 14 m wide and 4-6 m thick. The lithium and molybdate clouds followed the same path as the bromide cloud, but their rates of movement were retarded about 50% relative to bromide movement because of sorption onto the sediments.

Large‐scale natural gradient tracer test in sand and gravel, Cape Cod, Massachusetts: 3. Hydraulic conductivity variability and calculated macrodispersivities

Hydraulic conductivity (K) variability in a sand and gravel aquifer on Cape Cod, Massachusetts, was measured and subsequently used in stochastic transport theories to estimate macrodispersivities. Nearly 1500 K measurements were obtained by borehole flowmeter tests and permeameter analyses of cores. The geometric mean for the flowmeter tests (0.11 cm/s) is similar to that estimated from other field tests. The mean for the permeameter tests (0.035 cm/s) is significantly lower, possibly because of compaction of the cores. The variance for the flowmeter (0.24) is also greater than that for the permeameter (0.14). Geostatistical analyses applying negative exponential models with and without nuggets reveal similar spatial correlation structures for the two data sets. Estimated correlation scales range from 2.9 to 8 m in the horizontal and from 0.18 to 0.38 m in the vertical. Estimates of asymptotic longitudinal dispersivity (b.35–0.78 m) are similar in magnitude to that observed in the natural gradient tracer test (0.96 m) previously conducted at this site.

Simulation of fluid distributions observed at a crude oil spill site incorporating hysteresis, oil entrapment, and spatial variability of hydraulic properties

Subsurface oil, water, and air saturation distributions were determined using 146 samples collected from seven boreholes along a 120-m transect at a crude oil spill site near Bemidji, Minnesota. The field data, collected 10 years after the spill, show a clearly defined oil body that has an oil saturation distribution that appears to be influenced by sediment heterogeneities and water table fluctuations. The center of the oil body has depressed the water-saturated zone boundary and the oil appears to have migrated laterally within the capillary fringe. A multiphase cross-sectional flow model was developed and used to simulate the movement of oil and water at the spill site. Comparisons between observed and simulated oil saturation distributions serve as an indicator of the appropriateness of using such models to predict the actual spread of organic immiscible liquids at spill sites. Sediment hydraulic properties used in the model were estimated from particle size data. The general large-scale features of the observed oil body were reproduced only when hysteresis with oil entrapment and representations of observed spatial variability of hydraulic properties were incorporated into the model. The small-scale details of the observed subsurface oil distribution were not reproduced in the simulations. The discrepancy between observed and simulated oil distributions reflects the considerable uncertainty in model parameter estimates and boundary conditions, three-phase capillary pressure-saturation-relative permeability functions, representations of spatial variability of hydraulic properties, and hydrodynamics of the groundwater flow system at the study site.

Multispecies reactive tracer test in an aquifer with spatially variable chemical conditions

A field investigation of multispecies reactive transport was conducted in a well-characterized, sand and gravel aquifer on Cape Cod, Massachusetts. The aquifer is characterized by regions of differing chemical conditions caused by the disposal of secondary sewage effluent. Ten thousand liters of groundwater with added tracers (Br, Cr(VI), and EDTA complexed with Pb, Zn, Cu, and Ni) were injected into the aquifer and distributions of the tracers were monitored for 15 months. Most of the tracers were transported more than 200 m; transport was quantified using spatial moments computed from the results of a series of synoptic samplings. Cr(VI) transport was retarded relative to Br; the retardation factor varied from 1.1 to 2.4 and was dependent on chemical conditions. At 314 days after the injection, dissolved Cr(VI) mass in the tracer cloud had decreased 85%, with the likely cause being reduction to Cr(III) in a suboxic region of the aquifer. Transport of the metal-EDTA complexes was affected by aqueous complexation, adsorption, and dissolution-precipitation reactions of Fe oxyhydroxide minerals in the aquifer sediments. Dissolved Pb-EDTA complexes disappeared from the tracer cloud within 85 days, probably due to metal exchange reactions with Fe and adsorbed Zn (present prior to the injection from contamination by the sewage effluent). About 30% of the Cu-EDTA complexes remained within the tracer cloud 314 days after injection, even though the thermodynamic stability of the Pb-EDTA complex is greater than Cu-EDTA. It is hypothesized that stronger adsorption of Pb2+ to the aquifer sediments causes the Pb-EDTA complex to disassociate to a greater degree than the Cu-EDTA complex. The mass of dissolved Zn-EDTA increased during the first 175 days of the tracer test to 140% of the mass injected, with the increase due to desorption of sewage-derived Zn. Dissolved Ni-EDTA mass remained nearly constant throughout the tracer test, apparently only participating in reversible adsorption reactions. The results of the field experiment provide a chemically complex data set that can be used in the testing of reactive transport models of flow coupled with chemical reactions.

Determination of subsurface fluid contents at a crude-oil spill site

Hess, K.M., Herkelrath, W.N. and Essaid, H.I., 1992. Determination of subsurface fluid contents at a crude-oil spill site. J. Contam. Hydrol., 10: 75-96. Measurement of the fluid-content distribution at sites contaminated by immiscible fluids, including crude oil, is needed to better understand the movement of these fluids in the subsurface and to provide data to calibrate and verify numerical models and geophysical methods. A laboratory method was used to quantify the fluid contents of 146 core sections retrieved from boreholes aligned along a 120-m longitudinal transect at a crude-oil spill site near Bemidji, Minnesota, U.S.A. The 47-mm-diameter, minimally disturbed cores spanned a 4-m vertical interval contaminated by oil. Cores were frozen on site in a dry ice-alcohol bath to prevent redistribution and loss of fluids while sectioning the cores. We gravimetrically determined oil and water contents using a two-step method: (1) samples were slurried and the oil was removed by absorption onto strips of hydrophobic porous polyethylene (PPE); and (2) the samples were oven-dried to remove the water. The resulting data show sharp vertical gradients in the water and oil contents and a clearly defined oil body. The subsurface distribution is complex and appears to be influenced by sediment heterogeneities and water-table fluctuations. The center of the oil body has depressed the water-saturated zone boundary, and the oil is migrating laterally within the capillary fringe. The oil contents are as high as 0.3 cm 3 cm 3, which indicates that oil is probably still mobile 10 years after the spill occurred. The thickness of oil measured in wells suggests that accumulated thickness in wells is a poor indicator of the actual distribution of oil in the subsurface. Several possible sources of error are identified with the field and laboratory methods. An error analysis indicates that adsorption of water and sediment into the PPE adds as much as 4% to the measured oil masses and that uncertainties in the calculated sample volume and the assumed oil density introduce an additional___ 3% error when the masses are converted to fluid contents.