Technical Report NTB 98-01

The "Excavation Disturbed Zone Experiment (EDZ)" was conducted at the Grimsel Test Site (GTS) within the framework of the near-field programme in investigation Phase IV (1994 – 1996). It concentrated on investigating the hydraulic regime of the near-field of drilled tunnel sections under fully saturated conditions, with the aim of contributing to the development of methods for measuring and modelling axial water flow along tunnels and caverns. The studies focused on the mechanical and hydraulic properties of the rock mass in the direct vicinity of the tunnel wall. The so-called excavation disturbed zone (EDZ) is defined as the zone around the tunnel where excavation has altered the rock properties.

This report provides an overview of the results obtained during the EDZ experiment. The selected test location was a tunnel section in the heater test drift (WT) where mechanical stressing of the rock and some breakouts had been observed. Detailed geological mapping confirmed the suitability of the chosen test site and also provided background information for locating short test boreholes. During site preparation, in-situ stress measurements using a borehole slotter probe were performed in order to record the actual stress redistribution in the tunnel near-field induced by excavation of the tunnel for rock mechanical design calculations. A small stress increase and microfissures could be identified in the tunnel near-field, which suggested the potential existence of a plastic zone.

The stress measurements and the results of the geological mapping formed the basis for the rock mechanical modelling of the EDZ. The aim of the modelling was to obtain information on the development and geometry of the EDZ (understanding of the primary and secondary stress field). For this purpose, two different models were used:

• The regional 3D stress field modelling indicated that the topography has a significant influence on the primary stress field and not the fault zone systems that were included in the model. A good agreement between the measured and calculated stresses in the GTS was achieved by applying an additional far-field tectonic stress component.

• With the local 2D numerical disturbed zone modelling of the tunnel section itself, stress redistributions, possible plastifications and joint behaviour (closure, opening and shear displacements) in the near-field of the tunnel were investigated. The initial and boundary conditions were derived from the 3D model. All displacements of the rock matrix and the shear displacements of the discontinuities seem to be the result of the tunnel excavation. The displacement field and the geomechanical behaviour are strongly influenced by the discontinuities. Also, temporary plastifications and the subsequent stress redistributions are strongly linked to these discontinuities (material returns to the elastic range). In the different model cases, maximum shear deformations of 2 – 5 mm occur at the tunnel wall. The largest convergence (inward movement) of up to 10 mm occurs on the eastern tunnel wall.

Prior to the hydraulic test phase, the test location was decoupled from the normal GTS tunnel ventilation using partitioning walls, which allow this tunnel section to be isolated. In this way, complete saturation of the rock was achieved and single-phase conditions established. The status of rock saturation was checked by evaporation measurements. Afterwards, a surface sealing with epoxy resin was implemented to establish defined boundary conditions for the hydraulic testing and to avoid short-circuits in the test area with direct outflows into the tunnel during the injection tests. The EDZ was then investigated by 4 cored short radial boreholes (EDZ95.001 to EDZ95.004) arrayed perpendicularly to the tunnel axis. To optimise the hydraulic tests in these boreholes in terms of configuration and performance, hydraulic design calculations were carried out using two different modelling approaches (Equivalent Porous Medium [EPM] and Discrete Fracture Network [DFN]).

For the hydraulic testing, a Modular Minipacker System (MMPS) was developed which allows a wide range of test configurations to be realised in a small borehole with a diameter of 50 mm.

The hydraulic testing campaign included a series of Pulse Tests, Constant Head Tests, Constant Rate Tests and Pressure Recovery Tests in different borehole intervals to provide an overview of the distribution of the hydraulic characteristics / properties. Of the 18 tested intervals, 14 had hydraulic conductivities between 3 × 10^-12 and 3 × 10^-11 m/s. Based on the results of these screening tests, the remaining four intervals with relatively higher hydraulic conductivities (3 × 10^-7 m/s to 4 × 10^-10 m/s) were selected for more detailed characteri¬sation. The higher hydraulic conductivity of these 4 intervals appears to be related to a feature independent of EDZ origin (i.e. a pre-existing fracture). Overall, the hydraulic test data show a zone with roughly constant conductivity of 2 × 10^-12 – 3 × 10^-12 m/s beyond 2 m from the tunnel wall, and a zone with conductivities ≥ 8 × 10^-12 m/s (which is larger than the expected matrix conductivity in all zones) within 1 m of the tunnel wall. This suggests the presence of an EDZ around the tunnel. From rock mechanical modelling, the shape of the EDZ would probably be elliptical, but this could not be confirmed by the results of the hydraulic testing due to the small number of drilled and tested boreholes.
During the single-hole hydraulic tests, acoustic emissions were registered in two separate boreholes (EDZ95.005 and EDZ95.006) to monitor the test site especially during hydrotesting. The measurements indicated that no new fractures were created by the hydrotests performed at the test location.
Following active testing, the monitoring system was left in place to monitor the recovery and development of hydraulic head distribution over a longer period of time.

A methodology for estimating axial flow in the near-field of the tunnel after closure and resaturation of the excavation had been developed. Based on the discrete fracture network models (DFN) used for the design calculations and the characterisation of the EDZ, a (revised) conceptual model of the EDZ was elaborated. The effects of changes in stress and pore pressure occurring between the characterisation of the damaged zone and closure and resaturation of the excavation were considered. The results of modelling post-closure flow through the EDZ suggested bounding values for (post-closure) effective axial conductivity from 3 × 10^-11 to about 6 × 10^-8 m/s. The higher value corresponds to the situation where highly transmissive damaged zone features are extensive and well connected and the lower value to the situation where smallscale fracturing is dominant in the damaged zone. However, the effective axial conductivity of the rock around the tunnel at the GTS site will be controlled by the extent and connectivity of the high transmissivity features within the damaged zone.

In general, the experimental aims have been met and the equipment and methodology developed are suitable for determining the hydraulic properties of the EDZ.