Page 211 - Origin and Prediction of Abnormal Formation Pressures

186 E AMINZADEH, G.V. CHILINGAR AND J.O. ROBERTSON JR.

is 4.0 ppg higher than the pore pressure at some locations. At the same locations, the
difference between the fracture and pore pressure was 2.5 ppg for the medium-depth
water. The pore and fracture pressures differed by 1.5 ppg for the very deep water. The
results show that as the water depth increased the difference between the fracture and
pore pressures decreased.
In a second model study, Lee et al. (1997) lowered a sonic log to the fixed depth levels
of 2000 and 3000 m, to analyze pressure gradients. They replaced the sediments by
adding water columns of 50, 1000, and 2000 m. They also assumed that the sediments
from the water bottom to the top of the sonic log were homogeneous. The second model
differed from the first model by the position of the sonic log. The sonic log was located
exactly below the water bottom for the first model, whereas for the second model, the
log was located at a fixed depth. Using this model, they calculated the pressure gradients
for each water column. The pressure gradient, ~, was calculated from the following
equation:

~. - PP (7-21)
(Zo - Zw) + 0.465Zw
where, pp is the pore pressure and Zo and Zw are the pressure observation depth and
water depth, respectively.

Mapping reservoir fluid movement and dynamic changes of reservoir pressure using
time lapse (4-D seismic)

Time lapse (4-D) seismic data has been proven useful for accurate dynamic reservoir
characterization. As Tufa and Aminzadeh (1999) stated, to achieve accuracy and to
ensure that all available information at any given time is incorporated in the reservoir
model, reservoir characterization must be dynamic. To achieve this goal, one starts
with a simple model of the reservoir. As new well log, petrophysical, seismic, and
production data become available, the reservoir model must be updated to reflect the
changes in the reservoir, and for the model to be more detailed and representative. Both
static reservoir properties (such as porosity, permeability and facies type) and dynamic
reservoir properties (such as pressure, saturation of fluids, and temperature) must be
updated as more field data become available. Characterizing a reservoir by updating
both static and dynamic reservoir properties during the life of the field is referred to as
dynamic reservoir characterization.
Reservoir pressure is lowered by fluid production, whereas gas injection or water
flooding increases the reservoir pressure. Such changes affect bulk density and seismic
velocity of the reservoir layers which, in turn, affect the travel time and amplitude of
seismic waves propagating through the reservoir rocks. Usually the amplitude variations
are more apparent than the travel changes in 4-D seismic surveys; however, these
variations must be of sufficient size to represent a difference between the base seismic
survey and the follow-up surveys.
Forward modeling using laboratory data is generally utilized to estimate the expected
changes in seismic amplitudes. The effect of fluids on reservoir rock velocity is more
pronounced than that on the density. The introduction of gas into a liquid-filled rock