Brazilian tests are indirect tensile strength tests. A total of 9 tests were performed on plugs cored in the vertical orientation (3 from each seal peel).

Three CID (consolidated isotopically, drained) triaxial tests were performed on each of the three sandstone seal peels. Stress was increased/decreased at a rate of 3.6 MPa/hour and strain was increased/decreased at a rate of 3 mS/hr. The procedure for each CID test was as follows:

• Increase cell pressure to 1 MPa.
• Saturate sample with brine, without flushing out pore fluid originally present within the rock.
• Increase effective horizontal stress, σ_h', to 5, 10, or 15 MPa with drainage from the specimen.
• Increase backpressure to 1 MPa.
• Drained loading-unloading cycling:
  • The loading-unloading cycles for the tests conducted at σ_h' = 5 MPa were strain-controlled to ensure that the sample deformation remained well within the elastic domain. Unloading began at a maximum strain of 2 mS (axial strain) before unloading back to 1 mS (axial strain) and reloading.
  • For tests conducted at σ_h' = 10 and 15 MPa, the load-unload cycles were stress-controlled.
• For all tests, samples were then axially loaded to failure at a rate of 3 mS/hr, under drained conditions and maintaining a constant effective horizontal stress.

The CAUST protocol was an adapted version of the Dudley et al. (2016) constant pore pressure uniaxial strain tests (UST) protocol. Both radial and axial wave velocity measurements were made throughout.

For consolidation, first the isotropic effective stress was increased to the in situ horizontal stress, and then the effective axial stress was increased to the in situ vertical stress (whilst maintaining a constant effective horizontal stress). This simplified consolidation approach, as opposed to using a constant stress ratio, was chosen to avoid uncertainty due to non-linearity in sample stiffness and minimises the influence of any damage to the sample from core retrieval (Dudley (2016)).

Three uniaxial strain tests (UST) were performed on the sandstones, one for each of the seal peels. The procedure was as follows:

• Increase cell pressure to 1 MPa.
• Saturate sample with brine (without flushing out pore fluid originally present within the rock).
• Increase effective isotropic stress to the minimum effective in-situ stress (σ_h' = 10.9 MPa for both Johansen seal peels, σ_h' = 10.6 MPa for Cook).
• Grain stiffness and bulk stiffness measurement (as per Dudley (2016)):
  • Increase pore pressure and cell pressure simultaneously from 1 to 6 MPa to maintain constant effective stress for grain stiffness measurement.
  • 2x Cycle cell pressure – decrease and increase by 5 MPa (for bulk stiffness measurement).
  • 2x Cycle pore pressure – increase and decrease by 5 MPa (for bulk stiffness measurement).
• Increase effective axial stress to in-situ effective vertical stress (σ_v' = 26.6 MPa for Johansen, σ_v' = 25.8 MPa for Cook).
• UST test:
  • Uniaxial strain unload/reload of σ_v' of 10 MPa for calculation of constrained modulus.
  • Main uniaxial strain loading/unloading phase to maximum σ_v' of 70 MPa.

Three hydrostatic tests were conducted on samples from each of the sandstone seal peals. A higher rate of stress increase of 5 MPa/hr was used to optimise the test duration and allow for refilling of pumps during the main loading phase within working hours. The procedure for hydrostatic testing was as follows:

• Increase cell pressure to 1 MPa
• Saturate sample with brine (without flushing out pore fluid originally present within the rock).
• Increase effective isotropic stress to 18.75 MPa.
• Increase backpressure to 1 MPa
• Effective isotropic stress unload-loading cycle of 10 MPa.
• Effective isotropic stress loading to σ_h' = 59 MPa.

Brazilian tests are indirect tensile strength tests. A total of 7 tests were performed on plugs cored in both the horizontal (4) and vertical (3) orientation. Tests were performed using ASTM D3967-16.

Four permeability tests were conducted using the constant head method. The pressure gradient along the specimen was adjusted to maintain a measurable and stable flow. To measure the permeability both along and normal to bedding, two plugs were drilled horizontally and two vertically. The measurement procedure was as follows:

• Increase cell pressure to 15 MPa.
• Saturate the specimen with brine.
• Simultaneously increase backpressure to 15 MPa and cell pressure to 30 MPa so that the final effective isotropic stress is 15 MPa.
• Apply a pore pressure gradient between the top and bottom of the specimen:
  • 1 MPa and 4 MPa for the horizontal plugs*.
  • 8 MPa for the vertical plugs.

*For the horizontal plugs, several layer-parallel fractures were observed which may influence the test results.

CIU tests are isotropically consolidated tests with an undrained shearing phase. A total of 5 CIU tests were performed on the Intra-Drake shale (3 on vertical plugs and 2 horizontal plugs) at room temperature and using brine as the testing fluid. The procedure of the CIU tests was as follows:

• Mount the specimen in the triaxial cell and apply a vacuum of ~-0.1 MPa.

• Fill the cell with oil and apply a cell pressure of 0.5 MPa.

• Increase the cell pressure to the target value (5, 10, or 15 MPa).

• Release the vacuum, allowing salt water to enter the filter disks (swelling was not observed in any of the tests).

• Apply a back pressure (resulting in a pore pressure) of 30 MPa whilst increasing cell pressure to maintain constant effective stress.

• Wait until the volumetric strain has sufficiently stabilised.

• Close off the pore line tubing between the sample and the pore pressure pump (undrained conditions).

• Increase the axial load on the sample so that the samples deforms at a constant axial strain rate of ~0.09 mS/hr until failure. The calculation of the shear rate is dependent on the permeability to avoid a significant pore pressure build-up.

For the CIU test on the vertically-cored plug consolidated at 10 MPa effective isotropic stress, an additional drained loading/unloading cycle was performed to determine the drained Young's modulus, Poisson's ratio, and shear modulus.

For the CIU test on the horizontally-cored plug with a consolidation stress of 15 MPa effective isotropic stress, the consolidation phase was stopped at 10 MPa effective isotropic stress to repeat the same cycling procedure as for the vertical plug.

As for the sandstones, the CAUST (confined anisotropically uniaxial strain test) protocol for the shale was an adapted version of the Dudley et al. (2016) pore pressure depletion uniaxial strain tests (UST) protocol. The main difference with the Dudley et al. (2016) procedure was that, as for the sandstone CAUSTs, the consolidation stresses were met through first increasing the effective horizontal stress to the in situ minimum horizontal stress, before increasing the axial stress. Both radial and axial wave velocity (P-wave and S-wave) measurements were made throughout. The procedure of the CAUST tests was as follows:

• Increase cell pressure to 1 MPa.
• Saturate the sample with brine (without flushing out pore fluid originally present in the rock).
• Increase total isotropic stress to the minimum effective horizontal in-situ stress (σ_h' = 13.9 MPa).
• Apply a backpressure of 25 MPa whilst controlling cell pressure to maintain an effective isotropic stress of 13.9 MPa.
• Increase backpressure to 30 MPa at slower rate of 0.036 MPa/hr to perform a measurement of grain stiffness.*
• Increase the effective axial stress to the in-situ effective vertical stress (σ_v' = 24.5 MPa). The chosen rate of stress increase is low enough to keep shear stress below the expected in situ shear stress (5.3 MPa).
• UST pore pressure depletion from 30 to 10 MPa, controlling pore pressure to maintain a constant rate of strain of 0.036 MPa/hr (determined from the permeability measurement).