Temperature Sensitivity of Wave Velocities in Shales: Application for 4D Time-Lapse Seismic

Manafov, Rafig¹, Rune M. Holt², Erling Fjær³ (1) Norwegian University of Science and Technology, Trondheim, Norway (2) and NTNU, Trondheim, Norway (3) SINTEF Petroleum Research, and NTNU, Trondheim, Norway

Most of the laboratory experiments on shales samples are done at room temperature which is much lower than those measured in situ conditions. There is also a temperature change during drilling of oil and gas wells through overburden which can affect sonic log velocities and can cause possible wellbore stability problems. Investigation of the temperature sensitivity of seismic velocities in shales and the key factors controlling them are the main objectives of this work. The ultrasonic temperature experiments were performed on three field cores and one outcrop shale sample under different stress and pore pressure conditions. The results show that both P- and S-wave velocities in shales are temperature sensitive and decrease with increasing temperature. The P-wave velocity increase during cooling from 100 to 25ºC can be as much as 9-10% (compared to the initial value) in the direction normal to bedding, and 8-9% in the direction along the bedding. The S-wave velocity increase is even higher (16-17%). The investigation of key factors controlling the temperature sensitivity of seismic velocities show that low porosity shales are more temperature sensitive than high porosity shales and temperature sensitivity decreases exponentially with increasing porosity. For highly porous shales the pore water velocity is largely controlling the temperature sensitivity while for low porosity shales other – currently unknown – factors are dominating. To apply our experimental results for a field case, a simple temperature change modelling was performed for one segment of a North Sea field where cold water was injected for the last two years. The modelling shows that cold water injection cools down parts of both the reservoir and the overburden. The effect should be visible on highly repeatable 4D time-lapse Seismic, and there is actually some evidence of seismic response to the water injection at this field.