A Mohr-Coulomb failure criterion is applied to estimate fluid pressures that may cause fault reactivation during the depletion of hydrocarbon reservoirs. The estimates incorporate the decline in total minimum horizontal stress that accompanies fluid pressure depletion in hydrocarbon reservoirs. Such pore pressure/stress coupling must be incorporated in predictions of depletion-induced failure because it significantly influences the fluid pressures at which faulting occurs. A new algorithm for failure incorporating the coupled decrease in pore pressure and stress is derived to calculate the fluid pressures that can cause slip on normal faults during ongoing production. The algorithm is applied to the Ekofisk reservoir, Norway, using various friction coefficients for chalk and incorporating the observation that the minimum horizontal stress decreased at 80% the rate of pore pressure depletion in the field. A friction coefficient of 0.6 yields realistic results when modelling the depletion period 1975 to 1990. A fluid pressure decrease from the initial 45 MPa to 38 MPa is required to activate optimally oriented faults with dip angles of approximately 60°. This fluid pressure level (38 MPa) was attained in 1978–1980 and marks the onset of significant subsidence in the Ekofisk field. Ongoing fluid pressure depletion from 38 MPa to the present level of approximately 25 MPa is sufficient for sliding on faults with dip angles of 48° to 73°. Preexisting fractures in the Ekofisk reservoir fall in this range, as they exhibit predominantly steep dip angles (65°). Slip events recorded during seismic monitoring that was conducted in 1994, are likely to represent the reactivation of such steeply dipping faults and possibly the formation of new fractures. The modelling technique presented for predicting induced reservoir and fault failure is an essential requirement for the long-term planning of hydrocarbon field depletion.


The production of hydrocarbons from reservoir rocks can induce brittle failure in reservoirs and their immediate vicinity1–3. Induced failure in reservoirs may increase existing fracture permeability and may thus be beneficial for hydrocarbon recovery. However, the withdrawal of pore fluid and thus the depletion of pore fluid pressures can also cause well bore casing failure and hence reduce the stability of the well bore4. In fact, the reactivation of preexisting faults can shear off wells and damage surface constructions such as oil, gas and water pipelines1. The risk of damage can be managed by predicting pore fluid pressures at which brittle failure is likely to occur.

Hydrocarbon withdrawal over decades of production in oil fields can cause fluid pressure depletion that is in turn associated with a decrease of the minimum horizontal stress3,5,6. Such pore pressure/stress coupling, that was observed in the Ekofisk field, Norwegian North Sea, is shown in Fig. 1. Some of the effects of pore pressure/stress coupling on reservoir integrity in normal fault stress regimes have been outlined by Hillis7,8. A decrease in the minimum horizontal stress, which is the minimum principal stress in a normal fault stress regime, leads to an increase in the differential stress (s1 - s3). Unlike the total horizontal stress in the earth's crust that is strongly controlled by the elastic properties of rocks, the total vertical stress is given by the weight of the overburden load. The vertical stress, which is the maximum principal stress in a normal fault regime, is thus largely unaffected by pore pressure depletion. Although fluid pressure depletion without changes in (s1 - s3) decreases the likelihood of induced failure, the increase in (s1 - s3) that is associated with depletion promotes failure in a normal fault stress regime (Fig. 2). Thus, predictions of induced failure require identification of the stress and fluid pressure path of a hydrocarbon field during production, as well as knowledge of the rock and fault strengths.

This content is only available via PDF.
You can access this article if you purchase or spend a download.