Interpretation of a network-scale tracer experiment in fractured rock conducted using open wells.

The presence of fractures in bedrock allows for rapid aqueous contaminant transport through complex pathways and for diffusion of solutes between the fractures and the matrix. To better understand transport in these settings, tracer experiments are a commonly used tool. The need for expensive multi-level wells to obtain depth-specific concentrations, however, significantly limits the cost efficiency. The primary objective of this study is to develop a method whereby a discrete fracture network approach can be used to simulate the results of a divergent tracer experiment conducted using open observation boreholes in a well-characterized dolostone over distances of 55 m to 242 m. The experiment was conducted using a fluorescent tracer which allowed for continuous concentration measurement with depth in each open observation well. Two numerical models were employed in the interpretation of the experiment. The first was a 1-D finite difference model focused on flow and transport in the observation wells and the second was a 3-D control-volume finite element model capable of simulating the entire fracture network. Through fitting the experimental data to simulations, the most important fractures for transport in the system were identified. The number of fractures that participated in transport was few relative to the number of fractures observed in core and in constant head test results. Heterogeneous distribution of the fracture apertures was determined to be the likely cause of the highly tortuous transport observed at the site. This study demonstrates that tracer experiments conducted using open observation boreholes and a downhole fluorometer can improve our understanding of large-scale transport in fractured rock, especially when analysed with multiple models, and compared to other measured properties such as matrix porosity, hydraulic aperture, and fracture orientation.

The use of scaling parameters and the impact of non-uniqueness in the simulation of a large-scale solute transport experiment conducted in discrete fractures.

There are few studies analyzing tracer experiments conducted in rock fractures at large scales (>50 m) and fewer still which have succeeded in simulating experimental data without the use of parameters which increase/decrease with increasing scale of observation (i.e. scaling or effective parameters). To explore the use of effective parameters at large scales, a numerical model is used to examine the results of a divergent tracer experiment conducted in sparsely fractured rock where five observation points ranged over distances from 30 to 125 m. The experiment was conducted in what was first interpreted as a single horizontal fracture, traceable using geological and hydraulic evidence over that range in distance. Initial simulations were conducted using measured values of porosity, fracture aperture, and regional hydraulic gradient but were not successful in matching the field data at all measured locations. Multiple fits of equal quality at each observation point however were obtained with models using effective parameters (such as increasing porosity with scale) to represent heterogeneity. Additional fits were achieved by physically representing the same heterogeneity with different conceptual models, such as added horizontal and vertical fractures. This illustrates the non-uniqueness that arises in the interpretation of a divergent tracer experiment with multiple observation well distances. Using the models, predictions at distances beyond the measured domain were then generated. These showed that the choice of conceptual model significantly impacted simulated arrival times and concentrations at distances of only 75-175 m larger than the experimental scale. In particular, predictions made with effective parameters provided estimates that were less conservative (i.e. under predicting concentrations) than those made by directly adding heterogeneity to the numerical model.

Influence of piezometer construction on groundwater sampling in fractured rock.

A numerical model for groundwater flow and solute transport was employed to examine the influence of the screen and sandpack on the collection of a representative geochemical sample from a piezometer monitoring well installation in a discretely fractured bedrock aquifer. The optimization of screen and sandpack combinations was explored for the potential to reduce purging times and volumes in practice. Simulations accounted for the location of the fractures along the well screen, fracture aperture, screen length, and the pumping rate. The variability in the required purging times (t(99)-the time required to achieve 99% fractional contribution from the formation to pump discharge) can be explained by: (1) the relative hydraulic conductivities of the components of the system (fracture, sandpack, and screen), (2) the truncation of the flow field from the fracture to the screen by the upper and/or lower boundary of the sandpack of the flow field from another fracture, and (3) time-dependent drawdown. During pumping, only a portion of the sandpack may actually become hydraulically active. The optimal configuration (shortest purging time) is achieved when the ratios of the screen, sandpack, and fracture hydraulic conductivities are close to 1:1:1. More importantly, the role of the fracture hydraulic conductivity in the ratios is not as crucial to reducing t(99) as having the hydraulic conductivities of the screen and sandpack as similar as possible. This study provides a better understanding of well dynamics during pumping for the purpose of obtaining representative groundwater samples.

Fecal indicator bacteria variability in samples pumped from monitoring wells.

The detection of microbiological contamination in drinking water from groundwater wells is often made with a limited number of samples that are collected using traditional geochemical sampling protocols. The objective of this study is to examine the variability of fecal indicator bacteria, as observed using discrete samples, due to pumping. Two wells were instrumented as multilevel piezometers in a bedrock aquifer, and bacterial enumeration was conducted on a total of 166 samples (for total coliform, fecal coliform, Escherichia coli, and fecal streptococci) using standard membrane filtration methods. Five tests were conducted using pumping rates ranging from 0.3 to 17 L/min in a variety of purging scenarios, which included constant and variable (incremental increase and decrease) flow. The results clearly show a rapid and reproducible, 1 to 2 log-unit decrease in fecal indicator bacteria at the onset of pumping to stabilized, low-level concentrations prior to the removal of three to five well volumes. The pumping rate was not found to be correlated with the magnitude of observed bacterial counts. Based on the results, we suggest sampling protocols for fecal indicator bacteria that include multiple collections during the course of pumping, including early-time samples, and consider other techniques such as microscopic enumeration when assessing the source of bacteria from the well-aquifer system.

Interpretation of injection-withdrawal tracer experiments conducted between two wells in a large single fracture.

Tracer experiments conducted using a flow field established by injecting water into one borehole and withdrawing water from another are often used to establish connections and investigate dispersion in fractured rock. As a result of uncertainty in the uniqueness of existing models used for interpretation, this method has not been widely used to investigate more general transport processes including matrix diffusion or advective solute exchange between mobile and immobile zones of fluid. To explore the utility of the injection-withdrawal method as a general investigative tool and with the intent to resolve the transport processes in a discrete fracture, two tracer experiments were conducted using the injection-withdrawal configuration. The experiments were conducted in a fracture which has a large aperture (>500 microm) and horizontally pervades a dolostone formation. One experiment was conducted in the direction of the hydraulic gradient and the other in the direction opposite to the natural gradient. Two tracers having significantly different values of the free-water diffusion coefficient were used. To interpret the experiments, a hybrid numerical-analytical model was developed which accounts for the arcuate shape of the flow field, advection-dispersion in the fracture, diffusion into the matrix adjacent to the fracture, and the presence of natural flow in the fracture. The model was verified by comparison to a fully analytical solution and to a well-known finite-element model. Interpretation of the tracer experiments showed that when only one tracer, advection-dispersion, and matrix diffusion are considered, non-unique results were obtained. However, by using multiple tracers and by accounting for the presence of natural flow in the fracture, unique interpretations were obtained in which a single value of matrix porosity was estimated from the results of both experiments. The estimate of porosity agrees well with independent measurements of porosity obtained from core samples. This suggests that: (i) the injection-withdrawal method is a viable tool for the investigation of general transport processes provided all relevant experimental conditions are considered and multiple conservative tracers are used; and (ii) for the conditions of the experiments conducted in this study, the dominant mechanism for exchange of solute between the fracture and surrounding medium is matrix diffusion.

Humic acid enhanced remediation of an emplaced diesel source in groundwater. 1. Laboratory-based pilot scale test.

The enhanced solubility of petroleum-derived compounds in humic acid solutions is the basis for a new groundwater remediation technology. In this unique pilot-scale test, a stationary contaminant source consisting of diesel fuel was placed below the water table in a model sand aquifer (1.2 x 5.5 x 1.8-m deep) and flushed with water at a flow rate of 2 cm/h over 5 years. At 51 days, laboratory grade humic acid was added to the water and maintained at a level of approximately 0.8 g/l. The addition of humic acid had only a small impact on the aqueous transport of the BTEX components, which were rapidly dissolved from the diesel, but had a large effect on the flushing of PAHs, including methylated naphthalenes (MNs). Binding to aqueous humic acid enhanced the solubilization of MNs two- to tenfold. During aqueous transport, biodegradation of the BTEX and PAHs occurred, limiting the lateral and longitudinal extent of the diesel contaminant plume in the model aquifer. It appears that through enhanced solubilization, the overall biodegradation rate of the MNs was increased. As the various MNs were depleted from the diesel source, the MN plume shrank and then disappeared.