Should we include a default in-situ stress for non-zero reference pressure in poromechanics?

[jwboth]

When considering nonzero reference pressure, what is actually the correct force balance on the interface in terms of physics (disregarding PorePy for a second)? When trying to consider a stress-free reference configuration under some constant reference fluid pressure (under zero gravity), and by definition zero displacement, without external forces, intuitively, the resulting fluid pressure would need to be equal to the reference pressure as well. This is in accordance to the Biot equations considering reference pressures.

However, now at the fracture, the projected matrix stress needs to equilibrate with the fracture stress. The fracture stress is equal to the contact force and pressure contribution of the fluid. In a stress-free state, the contact force is zero. So the pressure contribution must be zero as well. Following the same idea as in the matrix, the fracture stress needs to be understood in relation to reference configuration. This would suggest that the pressure contribution, as for Biot, equals the pressure difference to the reference pressure.

In PP, the pressure contribution fracture_pressure_stress however is the absolute pressure only. To me this seems wrong. The use of self.pressure() - self.reference_pressure() would be the consistent choice, I believe.

[jwboth]

A new take. Conceptually more consistent would be instead to introduce an in-situ stress as a new component to the (mechanical) stress, which resembles the reference pressure. Effectively, such would transfer to the traction forces through the interface momentum balance, and so also to the fluid pressure (yet only for Biot coefficient 1, depending on whether the in-situ stress is modeled using poromechanical concepts).