The injection of cold water into a hydrocarbon reservoir containing
relatively warmer, more saline formation brine may generate
self-potential anomalies as a result of electrokinetic,
thermoelectric, and=or electrochemical effects. We have
numerically assessed the relative contributions of these effects
to the overall self-potential signal generated during oil production
in a simple hydrocarbon reservoir model. Our aim was to
determine if measurements of self-potential at a production well
can be used to detect the movement of water toward the well.
The coupling coefficients for the electrochemical and thermoelectric
potentials are uncertain, so we considered four different
models for them. We also investigated the effect of altering the
salinities of the formation and injected brines. We found that
the electrokinetic potential peaked at the location of the saturation
front (reaching values of 0.2 mV even for the most saline
brine considered). Moreover, the value at the production well
increased as the front approached the well, exceeding the noise
level ( 0.1 mV). Thermoelectric effects gave rise to larger
potentials in the reservoir (10 mV), but values at the well
were negligible ð Þ .0:1 mV until after water breakthrough
because of the lag in the temperature front relative to the saturation
front. Electrochemical potentials were smaller in magnitude
than thermoelectric potentials in the reservoir but were measurable
ð Þ > 0:1 mV at the well because the salinity front was
closely associated with the saturation front. When the formation
brine was less saline (1 mol=liter), electrokinetic effects dominated;
at higher salinities (5 mol=liter), electrochemical
effects were significant. We concluded that the measurement of
self-potential signals in a production well may be used to monitor
the movement of water in hydrocarbon reservoirs during
production, but further research is required to understand the
thermoelectric and electrochemical coupling coefficients in partially
saturated porous media.