Prediction of some in situ tracer tests with sorbing tracers using independent data
Section snippets
Introduction and background
In many instances, we wish to predict the migration behaviour of solutes over times and distances longer than what can be measured by experiments. A robust method that is based on the most important processes and mechanisms would then be useful. Such a method would have to rely on independently measured data on rock properties and on the solute interaction mechanisms.
Many different processes and mechanisms influence transport of solutes with the flowing water in fractured rocks. The
Model concepts
The flow is conceived as taking place in a three-dimensional network of channels that intersect at irregular intervals. At the intersections, the waters mix and then redistribute to those channels that carry the water from the intersection to the next. Each channel has a volume and a flow-wetted surface (FWS). The FWS is the surface over which a sorbing solute must pass in order to enter into the micropores of the rock. The magnitude of the FWS that the water flow “sees” is obviously one key
The Äspö TRUE experiments
Converging flow tracer tests were performed in water-saturated, fractured crystalline rock. The distance from injection points to pumping point was about 5 m. The injection flowrate was about 600 times smaller than the pumping flowrate. Sorbing, as well as nonsorbing tracers, were injected as a mixture. Extensive hydraulic tests were performed in the fracture and also in the surrounding rock mass. Laboratory measurements were made to determine matrix diffusion, porosities and sorption
Data used
Matrix diffusion coefficients and porosities were obtained from laboratory measurement reported in Byegård et al. (1998). Sorption data were taken from the same source but were reevaluated for the sorbing tracers to account for the experiments that were performed over only 2 weeks and equilibrium was not reached during this time. The method is described in Neretnieks (2002).
Transmissivity data for the rock mass surrounding the experimental region were taken from the Sicada database. One hundred
Simulations of the sorbing tracer tests
Consider a spherical volume of rock into which water flows from the surface of the rock towards its centre. This is depicted in Fig. 1. In the sphere, there is a multitude of fractures. The number of fractures is obtained from the fracture frequency observed in the boreholes surrounding the experimental site. The water collected in the pumping section in the centre of the sphere has flown through the three-dimensional network of fractures in the rock.
For any tracer injection point at a given
Sensitivity analysis
The two main parameters that influence the residence time distribution of the sorbing tracers in our model are one group that contains the FWS, sorption and diffusion data, and one parameter that pertains to the variability of the flowrates in the channels. We chose the most strongly sorbing tracer, Cs, for the sensitivity analysis. The group of parameters that governs the matrix diffusion effects in the model isL and W are the length and width of the channels,
Discussion and conclusions
An attempt was made to predict the RTDs for some strongly sorbing tracers in a fractured crystalline rock. A three-dimensional channel network model with essentially only the following two processes that influence the RTD was used. Matrix diffusion from the water flowing in the channels over the FWS is the interaction mechanism between solute and rock, and a random variation of flowrates in the different channels gives the influence of the variations in flowrates.
Laboratory data and hydraulic
Acknowledgements
The Swedish Nuclear Fuel and Waste Management Company, SKB, has supported this work.
References (6)
- et al.
Analysis of a long-term pumping and tracer test using the channel network model
J. Contam. Hydrol.
(1998) A stochastic multi-channel model for solute transport—analysis of tracer tests in fractured rock
J. Contam. Hydrol.
(2002)- Byegård, J., Johansson, H., Skålberg, M., 1998. The interaction of sorbing and non-sorbing tracers with different Äspö...
Cited by (25)
Saline groundwaters counteract up-flow of contaminants- implications for radionuclide repositories?
2024, Journal of Contaminant HydrologyThe effect of stagnant water zones on retarding radionuclide stransport in fractured rocks: An extension to the Channel Network Model
2016, Journal of HydrologyCitation Excerpt :To simulate the tracer experiment, it is assumed that the flow in the rock surrounding Feature A takes place in a three dimensional network of channels. The transmissivities of the channels are assigned randomly following a lognormal distribution found from borehole measurements (Neretnieks and Moreno, 2003), as detailed in Appendix A. In the realizations detailed in Section 6, we assume that the properties of the channels in Feature A can be characterized by the same mean and standard deviation values of the transmissivities as for all the other channels in the rock mass.
Radionuclide migration through fractured rock for arbitrary-length decay chain: Analytical solution and global sensitivity analysis
2015, Journal of HydrologyCitation Excerpt :To model solute transport in such networks, it is necessary to account for different mechanisms acting on solute transport in each individual channel. These mechanisms include diffusion into the porous rock matrix, diffusion into the stagnant water zone, hydrodynamic dispersion, sorption, and chain decay (Neretnieks, 2002; Neretnieks and Moreno, 2003). Fractures in rock have uneven walls and variable apertures, and the walls may come into contact at certain points and not at others, leading to a randomly distributed conductance within each fracture.
Long term infiltration and tracer transport in fractured rocks: Field observations and model analyses
2011, Journal of HydrologyCitation Excerpt :If the complexity of flow path geometry is the major mechanism for a specific problem and fully captured in an interpretation model, the enhancement of effective matrix diffusion should disappear. This argument is supported by a recent study by Neretnieks and Moreno (2003). However, it is also important to note that it is difficult, if not impossible, to fully capture all the details of the mechanisms discussed above at a scale of practical interest.
Field-scale effective matrix diffusion coefficient for fractured rock: Results from literature survey
2007, Journal of Contaminant HydrologyAn interpretation of potential scale dependence of the effective matrix diffusion coefficient
2007, Journal of Contaminant Hydrology