In situ stress and pore pressure are key parameters governing rock deformation, yet direct measurements of these quantities are rare. During Integrated Ocean Drilling Program (IODP) Expedition #319, we drilled through a forearc basin at the Nankai subduction zone and into the underlying accretionary prism. We used the Modular Formation Dynamics Tester tool (MDT) for the first time in IODP to measure in situ minimum stress, pore pressure, and permeability at 11 depths between 729.9 and 1533.9 mbsf. Leak‐off testing at 708.6 mbsf conducted as part of drilling operations provided a second measurement of minimum stress. The MDT campaign included nine single‐probe (SP) tests to measure permeability and in situ pore pressure and two dual‐packer (DP) tests to measure minimum principal stress. Permeabilities defined from the SP tests range from 6.53 × 10⁻¹⁷ to 4.23 × 10⁻¹⁴ m². Pore fluid pressures are near hydrostatic throughout the section despite rapid sedimentation. This is consistent with the measured hydraulic diffusivity of the sediments and suggests that the forearc basin should not trap overpressures within the upper plate of the subduction zone. Minimum principal stresses are consistently lower than the vertical stress. We estimate the maximum horizontal stress from wellbore failures at the leak‐off test and shallow MDT DP test depths. The results indicate a normal or strike‐slip stress regime, consistent with the observation of abundant active normal faults in the seaward‐most part of the basin, and a general decrease in fault activity in the vicinity of Site C0009.

Summary: Hydraulic fracturing can produce a main fracture and increase flow efficiency. It can also result in the invasion of fracturing fluid, which can produce clay swelling. The invasion can block the pores and throats in the porous media and damage matrix permeability, while it can also induce microcracks under certain conditions. This study developed a numerical model to evaluate the permeability change induced by the invasion. Both pore-throat clogging and microcracks are integrated into the model. The numerical model coupled a flow model based on the pore-scale network method (PNM), a solid model, and a permeability evaluation model. The solid model and permeability evaluation model are based on our microcrack performance model reported previously. We validated the flow model by analytical results and the coupled model by previous experiments of the soaking test. Simulation results show that the invasion can decrease the permeability for the clogging effect and increase the permeability for the microcrack generation. The generation of microcracks can greatly increase the permeability even with the clogging effect for clay swelling. Our simulation results indicated that rock properties can affect the effect of clogging and microcracks on permeability. This study provides an approach for the selection of fracturing fluid and the design of fracturing fluid flowback. This is crucial for the understanding of the invasion on permeability and can serve as a guide for fracturing operations.

In the field, locally applied fluid pressure can initiate fault slip, which may expand unstably over a wide area under certain conditions, generating elastic waves and inducing earthquakes. Therefore, examining the unstable expansion of the initial slip is important. However, reproducing this process in laboratory experiments, such as triaxial loading tests on small cylindrical specimens with inclined faults, is challenging. To achieve this, we prepared a sub-meter-scale cubic specimen, which was separated into two triangular prisms by a model fault. The specimen was subjected to biaxial compression of different magnitudes. A 2D array of strain gauges was embedded beneath the fault plane to measure the changes in shear strain with fault slip driven by fluid injection. Based on the experimental results, we discussed the features of the injection-induced fault slips that lead to earthquakes. The strain increased locally around the edge of the fault slipping area by approximately 10 με, which is equivalent to a shear stress of ~ 0.1 MPa. The fault slipping area first expanded gradually and then unstably beyond the fluid invasion area approximately 3 s after the slip was initiated. The unstable expansion of initial slips was suppressed by reducing the initial shear stress on the fault by 0.3 MPa. In this case, the initial shear stress was possibly too small for additional stress to accumulate at the edge of the fault-slipping area to overcome the static frictional strength of the fault. Our experimental study explicitly confirms that injection-induced aseismic slip can outpace the fluid migration front, finally leading up to a runaway rupture. It also confirms that the level of initial shear stress is important in controlling the rupture size under fluid injection. Graphical Abstract

Summary: With the development of unconventional resources, such as oil sands and methane hydrate reservoirs, the importance of the mechanical performance model for underground unconsolidated rocks has increased significantly. The commonly used numerical approach for unconsolidated rocks is the discrete-element method (DEM). However, the extensive calculations required by the DEM make it inadequate for simulating unconsolidated rock behavior on a field scale. An alternative is the continuum approach, used to simulate the behavior of unconsolidated rocks on the field scale. In previous continuum approaches, unconsolidated rocks have been modeled as a visco-plastic fluid (i.e., Bingham fluid). The continuum approach based on visco-plastic fluid uses pressure (scalar) to describe the stress state of the particles. However, this approach does not account for the difference between the maximum and minimum principal stresses of the in-situ stress field when simulating the mechanical performance of unconsolidated rocks. Here, we developed an improved continuum approach for unconsolidated rocks and used the finite-element method as a numerical approach. Our improved model can consider the difference between the maximum and minimum principal stresses of the in-situ stress field and the pore pressure of the unconsolidated formation. We validated our numerical model with the angle of repose test, a benchmark problem for unconsolidated rocks. The validation results confirm the accuracy of our unconsolidated model. For the coupled model between the unconsolidated model and the flow model, we used an analytical solution to verify its reliability. Unconsolidated rock performances in an unconsolidated reservoir with fluid injection have been investigated based on our coupled model. The simulation results show that injection can activate the movement of unconsolidated rock particles, leading to changes in the distribution of effective stress and permeability. Our model has the potential to address large-scale unconsolidated rock issues and contribute to energy extraction.