Technical Report NTB 93-47
Grimsel Test SiteDevelopments in Hydrotesting, Fluid Logging and Combined Salt/Heat Tracer Experiments in the BK Site (Phase III)
This report documents the scientific studies carried out between January 1991 and December 1993 during Phase 3 of the Fracture System Flow Test Project (BK) at the Grimsel Test Site. The work was carried out as part of a joint investigation program between Nagra and the Federal Institute for Geoscience and Natural Resources (BGR), Hannover, and was aimed at developing new methods and concepts for characterizing flow processes in fractured formations. Numerical modeling of fracture flow also formed a focal point of the studies.
In the past, conventional fluid logging techniques (temperature/conductivity logging) in vertical boreholes have been used in Nagra's site investigation programs with considerable success. In the BK subproject "Fluid logging in inclined boreholes", it was shown that these techniques are either inapplicable, or have only restricted application, in inclined boreholes. New methods were therefore developed where the measuring probes are fixed in the boreholes before testing begins (fluid logging with fixed probes). Pilot tests showed that, in principle, this approach is also suitable for use in inclined boreholes.
Hydraulic crosshole tests represent the most frequently used method for investigating water-bearing fracture networks. The Multiple Source Crosshole Test (MSCT) was developed within the framework of the BK project with the aim of achieving an improved spatial resolution of the structure of such fracture networks. The principle of the MSCT is to carry out numerous short crosshole tests at different locations in the network to be characterized. A spectral interpretation procedure was developed for the MSCT, whereby hydraulic crosshole tests with variable flow rates can be evaluated using a diagnostic type curve procedure. A synthetic experiment was also performed to investigate a possible means of characterizing simple fracture structures deterministically on the basis of spectral hydrotest interpretations.
The focal point of the Salt/Heat Tracer Test (SHT) was characterization of the transport properties of fracture systems. The test focused around dipole tracer tests with complementary flow-fields (i.e. inversion of the dipole flow-field) and complementary tracers (NaCI and heat), which were aimed at providing a more detailed description of the complex transport processes in a fracture network. One particular aspect to be investigated was whether any information on the structure of the pore spaces in the fracture system can be derived from the differing transport behavior of salt (advective-dispersive) and heat (diffusion-dominated). Modeling of the field experiments clearly showed the sensitivity, and hence the selectivity, of such a test concept as regards different hypotheses of the fracture structure.
Data on structural geology and the results of hydrogeological tests from earlier phases of the BK project were synthesized to provide a conceptual hydrodynamic model of the BK fracture system. This was done using computer-aided 3D visualization of the structural and hydrogeological data. Commercial graphics software was used to develop applications which fulfil the specific criteria of a structural geology/hydrogeological interpretation (e.g. spatial correlation of petrography, structural data, fluid logging results, etc.). The conceptual model then serves as a basis for formulating a numerical model which is used to simulate fracture flow in the BK fracture system.
The main objective of this work was not to provide a detailed hydraulic model of the flow-field at the test location, but rather to try out inverse modeling methods based on a discrete fracture zone model.
Besides the development of techniques which was achieved in Phase 3 of the Fracture System Flow Test Project, important information was also obtained on the hydrogeological characterization of the test location. For example, it was shown that the two lamprophyres to the north and south of the site almost completely decouple the BK fracture system from the surrounding test locations.
The hydraulic system of the BK is therefore restricted and has a free surface which is controlled by drainage into the drift and by boundary influxes. Despite the large number of fractures identified from core logging, the component which is hydraulically active is very small. Hydraulic testing has shown that the significance of the so-called S-fractures in the vicinity of the northern lamprophyre has been underestimated in the past.
The numerical model was used for inverse modeling of a hydraulic crosshole test. Despite the relatively simple structure of the model, which was selected because of restricted timescales, the fit with observed data was surprisingly good. Finally a sensitivity study was carried out; this highlighted possibilities for improving the underlying structural model for application in future modeling studies.