|
Figure and Table Captions
Return to top.
Although historically  production in Mata
field was ruled by the concept that every sand is hydraulically isolated
from the others, analysis of the production, injection, and pressure of
all sands of L and S packages shows evidences of communication between
the reservoirs. Pressure measurements in the last few years indicate
common pressures for the sands at a same datum, even though they
presented distinct recovery factors. Besides the anomalous behavior of
the production and pressure of the field, material balances show
evidences of mass flux within and/or between the sands (of the same
units) or between the L and S sand packages.
In order to understand the communication
between the sands, several geological factors have been considered.
Structural juxtaposition of porous sections from the different
reservoirs across faults and mechanical failures in wells (with more
than 50 years of life) are proposed as the main ways of communication.
Exhaustive material balances and production
analysis were made in order to understand the communication, estimate
the amount of mass exchanged between the sands, and evaluate its impact
on the potential of the whole field.
Field Location and
Regional Geology
Mata Field is located in the Greater Oficina
trend in the southern flank of Maturin sub-basin, representing the
oriental part of the Eastern Venezuelan Basin (Figure
1). This foreland basin was formed starting from the oblique
convergence, in a dextral movement, between the Caribbean and the South
American plates (Lugo and Mann, 1995). The Greater Oficina trend is
characterized by extensional tectonics with regional dips trending N to
NE.
Geologic
Characterization: Specific Objectives
As part of reservoir characterization,
emphasis was focused on the necessity of understanding the communication
mechanisms between distinct sands. To complete the current geological
model, a 3D geological model was build in order to:
1. Determine the role of the geological
factors in the communication between the reservoirs (stratigraphic
characteristics, structural features, etc.).
2. Characterize in detail the sand-sand
communication zones across the faults:
-
Practical evaluation
of the spatial distribution of the permeable zones juxtaposed across
secondary faults, allowing possible communication pathways.
-
Dynamic
characterization of the packages of L-S sands, allowing the
integrated simulation of all sands of the packages of L-S sands.
-
Geology-engineer
analyses of a new development plan and reserves reevaluation.
Stratigraphic Characterization
Sequence stratigraphy, mainly based on well
logs and core analysis, was analyzed from basin scale to a
sand-reservoir scale. The system tracts were defined and classified (lowstand,
transgressive and highstand). (Picarelli et al., 2001). The L-S sands
are associated with incised valley fills in channelized deposits in a
lowstand systems tract. The incised valleys, generated by lowering of
sea level, can produce contact within sands whenever the sealing horizon
is thin. Such a “kink” in sand communication has been identified in the
L and S sands (Figure 2).
Structural Characterization
The structure is characterized by a system of
normal faults. Fault displacements result in the juxtaposition of
distinct reservoirs across faults (Badley et al., 1991). Based on the 3D
geologic model, 3D juxtaposition maps were made (“Allan Maps”) over the
entire fault planes to analyze the spatial distribution of the fault
displacements and of sand porosity. Assuming that the fault itself do
not act as a seal, the potential communication zones (leakage zones)
were identified by the simultaneous analysis of (i) the juxtaposition
between sequences and within the sequences, for each individual sand,
across the fault planes; and of (ii) the spatial (geostatistical)
distribution of the porosity on the same fault plane (Figure
3). These zones of potential communication possess lengthened form,
and they are distributed in an irregular way on the fault planes.
This geologic study demonstrates that,
besides the erosive events that can constitute communication mechanisms
between sands, the displacements caused by the secondary faults may also
induce communications between the sands in the field.
Production
Motivation and Technique
In Mata field there are nine main reservoirs
with a total OOIP of 200 MMSTB and cumulative production of 48 MMSTBO.
Figure 4 shows the initial pressure, the
current pressure, and the recovery factor of the L-S sand package. It
shows that with different recovery factors the actual pressure is almost
the same for all sands, even though some sands have secondary recovery
and others do not. Due to the communication between the L-S sand
package, the traditional material balance technique was not able to fit
the field history; therefore, the material balance equation was adjusted
to identify the influx/outflux (source term: M) necessary to adjust the
history--
M = F - N * ( Eo + m*Eg + Efw)
Another technique implemented was to identify
“the total gas available to be produced” (Figures
5 and 6), as
given here below:
The solution gas produced with the produced oil + the original gas cap +
the injected gas + the gas that comes out of solution (due to the
depressurization of the reservoir)) - the gas trapped by the critical
saturation.
The idea is that the total available gas
minus the produced gas is equal to the actual free gas. A comparison of
this volume with the volume of the actual gas cap, volumetrically
calculated with the actual gas/oil contact, possibly indicates gas flux
into or out of the sand package.
In order to apply the material balance, the
L-S sand package was divided in four groups, L12, L34, S12, and S34.
Figures 5 and 6
illustrate the influx/outflux calculated, for 2 sand packages (L12 and
L34), in order for there to be a fit of the sand group history and of
the produced gas with the available gas for production.
In the beginning there was a loss of mass
from the L12 sand package; between 1962 and 1975 the sand received mass
and stayed more or less stable after that (Figure
5). At the end of 2002 the sand had lost around 3 MMBbl (reservoir
condition).
The actual produced gas is 40 MMMSCF, and the
actual available gas is 53 MMMSCF (Figure 5);
therefore, there ought to be around 13 MMMSCF of free gas in the sand.
The actual mapped gas cap size shows 18 MMMSCF of free gas, with a
difference of -5 MMMSCF.
Sand
Package L34
Through all the production time, the sand
package has received an influx (Figure 6),
having received, to date, around 21 MMBbl (reservoir condition).
The total gas available to be produced is 21
MMMSCF and the sand produced 40 MMMSCF (Figure 6).
In other words, the sand produced 19 MMMSCF more than it would be
possible if it were an isolated sand. This excess of gas certainly came
from others sands. Currently, there is no identified gas cap in the
sand.
The two others sand packages were analyzed
the same way. The source term (M) of all four sand packages is shown in
Figure 7. A table of the actual gas cap
volume, the produced gas, and the available gas to be produced is shown
as Figure 8.
L-S Sand
Package
The assumption that the L-S sand package is a
hydraulic unit implies that it is a closed system, and consequently the
entire package should not lose or gain mass. Therefore, if inside the
package one sand loses mass, it is due to the fact that another one is
gaining it. Based on this concept, the summation of all influx should be
zero. The graphics of the influxes (with the summation) and the pressure
history are shown in Figure 7.
In conclusion, at the beginning the L sands
were in communication and the S sands were not. After 1977 all the sands
became in communication, as seen in both plots (Figure
7), where all the pressures are similar, at a common datum, and the
summation of the influx in all sands tends to be zero.
Although there is a huge error in the gas
volumes calculated by the gas balance when analyzed sand by sand, it is
clear that the total error is small. This means that there was a strong
gas movement from the lower sand package S to the upper sand package L.
This fact leads to a hypothesis that the communication occurred at the
top of the structure.
In the beginning only the L sands behaved as
a hydraulic unit; the geological model easily explains this fact. In
1977 almost all sands in the L-S sand package started to behave as a
hydraulic unit. The geological model cannot explain this fact, and
although some models have been studied, there is no proved reason for
this. Based on the theory that the communication occurred at the top of
the structure, the two most reasonable hypotheses are the breaking of
the seal of the main fault or mechanical failures in wells.
Conclusions
In Mata field a multidisciplinary approach
was the key to understand the communication between sands historically
exploited as individuals reservoirs. This approach was the support for
achieving legal authorization to exploit all L-S sand packages as a
hydraulic unit.
We thank Petrobras Energía de Venezuela for
allowing us to publish the results of this work.
Badley, M.E., Freeman, B.,
Roberts, A.M., Thatcher, J.S., Walsh, J., Watterson, J. and Yielding,
G., 1991, Fault Interpretation during seismic interpretation and
reservoir evaluation, in The integration of geology, geophysics,
petrophysics and petroleum engineering in reservoir delineation,
description, and management: AAPG Special Volume (Proceedings First
Archie Conference), p. 224-241.
Lugo, J., and Mann, P., 1995,
Jurassic-Eocene tectonic evolution of Maracaibo Basin, Venezuela, in
Petroleum basins of South America: AAPG Memoir 62, p. 699-725.
Picarelli, A. Holder, B.,
Argüello, J., Perón, R., Nieves, I., and Rojas, J., 2001,
High-resolution sequence stratigraphy and reservoir characterization
applied to mature fields: Example from Eastern Venezuela Basin:
SPE 69602 (March, 2001).
Return to top.
|
Print this page
Click
to article in PDF format.
