816003
816003
816003
This report was prepared as an account of work sponsored by an agency of the United States
Government. Neither the United States Government nor any agency thereof, nor any of their employees,
makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy,
completeness, or usefulness of any information, apparatus or product, or represents that its use would not
infringe privately owned rights. Reference to any specific commercial product, process or service by trade
name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement,
recommendation, or favoring by the United States Government or any agency thereof. The views and opinions
of authors expressed herein do not necessarily state or reflect those of the United States Government of any
agency thereof.
OBJECTIVES
The Class 2 Project at West Welch was designed to demonstrate the use of advanced technologies
to enhance the economics of improved oil recovery (IOR) projects in lower quality Shallow Shelf Carbonate
(SSC) reservoirs, resulting in recovery of additional oil that would otherwise be left in the reservoir at project
abandonment. Accurate reservoir description is critical to the effective evaluation and efficient design of IOR
projects in the heterogeneous SSC reservoirs. Therefore, the majority of Budget Period 1 was devoted to
reservoir characterization. Technologies being demonstrated include:
1.Advanced petrophysics
2.Three-dimensional (3-D) seismic
3.Crosswell bore tomography
4.Advanced reservoir simulation
5.Carbon dioxide (CO2) stimulation treatments
6.Hydraulic fracturing design and monitoring
7.Mobility control agents
West Welch Unit is one of four large waterflood units in the Welch Field in the northwestern portion
of Dawson County, Texas. The Welch Field was discovered in the early 1940's and produces oil under a
solution gas drive mechanism from the San Andres formation at approximately 4800 ft. The field has been
under waterflood for 30 years and a significant portion has been infill-drilled on 20-ac density. A 1982-86 pilot
CO2 injection project in the offsetting South Welch Unit yielded positive results. Recent installation of a CO2
pipeline near the field allowed the phased development of a miscible CO2 injection project at the South Welch
Unit.
The reservoir quality at the West Welch Unit is poorer than other San Andres reservoirs due to its
relative position to sea level during deposition. Because of the proximity of a CO2 source and the CO2
operating experience that would be available from the South Welch Unit, West Welch Unit is an ideal location
for demonstrating methods for enhancing economics of IOR projects in lower quality SSC reservoirs. This
Class 2 project concentrates on the efficient design of a miscible CO2 project based on detailed reservoir
characterization from advanced petrophysics, 3-D seismic interpretations and crosswell tomography
interpretations.
During the quarter the 3-D depiction of the CO2-invaded area was quantified in terms of relative
saturation. Also, the second interwell seismic monitor survey was successfully acquired on the southern
pattern. A second completion attempt was conducted on the horizontal lateral drilled in 11/00.
Beginning on January 29, 2001, the second round of interwell seismic monitor surveys was acquired
on the same six survey lines (south pattern) used for the first monitor survey conducted in December 1999
(Figure 1). Field work was completed in February and data processing and interpretation continued through
the end of the first quarter. Preliminary results indicate presence of CO2 on all six survey lines as of the date
of the second monitor surveys.
During the first quarter of 2001, Advanced Reservoir Technologies was successful in developing a
technique for estimating saturation of the injected CO2 based on corresponding changes in compressional
and shear wave velocities computed from interwell seismic data. The technique involves using a combination
of the Biot-Gassman equations together with Woods equation, which allows an estimate of compressibility for
a mixture of two fluids based on compressibility and relative saturations of each. Currently there is not a
method available for calibrating CO2 saturations in the Welch study area so the results can only be scaled
as estimated relative saturation based on the highest indicated saturation being designated as 100 %. If a
method can be developed for calibrating the results then absolute values of saturation can be estimated. Such
saturations estimates could then be compared to time equivalent CO2 saturations generated by the simulator
to allow further calibration of the reservoir model. The relative saturation estimates, however provide useful
information in understanding current performance in the CO2 focus area. The usefulness of this information
should increase greatly when the processing and interpretation of the second interwell monitor survey results
are completed.
NUMERICAL SIMULATION
The modeling requirements have been developed and it is anticipated that the actual simulation will
be conducted during the third quarter of 2001. Several future scenarios will be modeled, including various slug
sizes, different WAG schedules and stopping CO2 injection in September 2001 when the project terminates.
HORIZONTAL DRILLING
In the last quarter of 2000, producer No. 4853, located at the south end of the DOE, project area was
re-entered and a horizontal lateral drilled due north 3500 ft (Figure 2). The strategy behind drilling the lateral
and details on the drilling operations were reported in the 4thQ 2000 report along with preliminary information
on the initial completion attempt using Halliburton’s newly developed Surgi-frac technique. Table 1 is the daily
activity log covering the drilling and completion operations. The proposed and actual (surveyed) wellbore path
is shown by Figure 3. The wellbore path penetrated the main pay interval- M-3- for most of its lenght (Figure
4). Since the lateral was going updip to the north it was necessary to incline the second half of the lateral to
stay in the main pay zone and out of the water bearing lower zone.
Six intervals along the lateral (8385', 7850', 7250', 6520', 5900' and 5300') were chosen for Surgi-frac
completion and spaced so no completion would be in line with either of the two rows of injectors that had been
intersected. The Surgi-frac technique pumps sand slurry down the tubing and out jets orientated to the plane
of minimum stress, allowing the hydraulic horsepower to be focused at one point. The formation is notched
by erosion and at the point of impact the kinetic energy of the jetted stream is converted to pressure. When
this “stagnation” pressure is slightly greater than the ambient pressure a fracture is initiated.. The fracture is
propagated and propped by continuing to pump the sand slurry down the tubing. The jetting action creates
an area of low pressure at the mouth of the fracture due to the Bernoulli effect. Fluid is pumped down the back
side to maintain pressure in the wellbore. In theory all of the fluid and sand is drawn into fracture by the low
pressure zone and there is no leak off into other fractures.
A summary of the initial Surgi-frac completion attempt on November 29, 2000 is given on Table 2. The
first of six planned treatment intervals was near the toe of the lateral at 8385 ft (MD). After notching and
initiating the fracture, 19,285 lbs of proppant and 19,714 gals of gelled water were pumped away at 18 BPM.
The formation broke down at a bottom hole pressure (BHP) of 4250 psi and treated around 4000 psi
(Figure 5). The maximum tubing pressure was 8500 psi, but tubing pressure is not a good indicator of bottom
hole performance due to the large pressure loss due to friction and the pressure drop across the jets. The
treatment at this interval appears to have successfully initiated and propagated a fracture as indicated by the
BHP, which built nearly straight up at the start and then broke over sharply as the formation broke down.
Unfortunately toward the end of the treatment one of the jets washed out as indicated by the drop-off
in tubing pressure. The surgi-frac tool was repositioned and the second and third intervals were treated. The
tubing pressure was much lower at the same pump rate (18 BPM) due to the elimination of pressure drop
across the jets and the BHP did not show the sharp breakover caused by the formation parting. Only shallow
erosion of the wellbore probably occurred at the second and third intervals and the sand slurry flowed toward
the initial fracture in the toe of the lateral at 8385 ft (MD). Sand fill occurred from 8051 ft to 8410 ft (MDTD)
The proppant had been tagged with radioactive material, but a tracer log run 12/07/00 failed to give any
definitive interpretation (Figure 6). The well was placed on rod pump 12/25/00 and achieved a maximum oil
rate of 41 BOPD and 610 BWPD with 1206 MCF of gas (CO2). The well test history is shown on Table 3. The
CO2 breakthrough had occurred immediately upon producing the well. In the second half of January it was
necessary to cut back on CO2 injection to reduce the gas volume being produced in 4853. A severe scale
problem was discovered in late January when pulling the production tubing.
The second Surgi-frac completion attempt was conducted 1/25/01 starting with the second interval
from the toe at 7246 ft (MD). As shown on Table 4, the five remaining intervals were treated with an average
of 15,900 gals of gelled water and 22,200 lbs of proppant at rates varying from 14.8 to 17.5 BPM. The BHP
performance (Figure 7) shows that the first two intervals broke back sharply, indicating that a fracture had
been initiated and propagated. For the remaining three intervals, the BHP broke over sharply to a fairly
constant treating pressure, but did not break back to a lower pressure as often occurs when the formation is
parted. While the tracer log (Figure 8) again failed to give any definitive answer as to the placement of
proppant, a majority of the tracer activity was at the end of the lateral, implying that most of the stimulation
went toward the toe. However pressure performance suggest that the second and third intervals were treated
and it is possible that the other three were also.
Sand and scale were reversed out from 7759 ft to TD at 8410 ft (MD). After treating the well for scale,
a Reda was run in the vertical hole and the well placed on production 2/8/01. The first reported test (Table 4)
was 0 BOPD and 1271 BWPD with 131 MCF of gas on 2/15/01. The well has produced very little oil through
March 2001. Due to the scale problem and the breakthrough of CO2, it is impossible to judge the effectiveness
of the second Surgi-frac completion attempt based on well tests.
Through March 2001 a total of 4.4 BCF of CO2 had been injected into the project area since initiation
of injection in October 1997. Overall oil and gas production was lower than the previous quarter due to
restrictions in C02 injection for four of the last six months that started in November 2000 with the four injectors
that offset the horizontal lateral. In this quarter injection was restricted during the acquisition of the second
Monthly performance for the focus area during the first quarter of 2001 is shown on Table 5. Injector
4805 was returned to CO2 injection in March 2001 after having been wagged to water in January 2000.
Because of CO2 breakthrough into the 4853 lateral, the four offset injectors were kept at a restricted input rate
during the quarter. Injectors 4805 and 4808 injected at their maximum pressures. The gas breakthrough into
the 4853 lateral most likely is coming from a gas saturated area near the toe of the lateral as opposed to a
direct channel with an injector. As discussed in the Horizontal Lateral Section, there is some evidence that
most of the Surgi-frac treatments went into the initial fracture interval near the toe of the lateral, possibly
creating an extended frac wing just north of the 4809 and 4811injectors. The volume of gas being produced
by 4853 (about 800 mcfpd) is about equal to the individual average injection rate for any of the four injectors
offsetting the lateral. A direct channel between an injector and the lateral would be evidenced by a significant
change in the injection pressure as the lateral was shut in or opened up. This was the case with injector 4810
which was in direct communication with producer 4843 at one time, but has not occurred in regards to the
4853 horizontal lateral.
It was necessary to shut in 4853 because the volume of produced gas was exceeding the capacity
of the gas plant. However 4853 was scheduled to be turned back on in early April 2001. The offsetting
injectors will be maintained at their current choke settings and the pressures and rates closely monitored for
change when 4853 is turned back on. The 4853 lateral has also communicated with one of the two producers
near to the lateral well path and No. 4829 was shut in during the quarter due to high gas (CO2) production.
The strong oil responses in 4844 and 4847 are still occurring. Both responding wells are directly offset
by north and south injectors, although one of 4844 offsetting injectors was wagged to water in January 2000.
There appears to be oil response in 4850 and possible oil response in 4841. These two wells are direct offsets
to active CO2 injectors. Definite gas (CO2) breakthrough has occurred prior to this quarter in Well Nos. 4841,
4843, 4844, 4850 and 4854. These producers are all direct north or south offsets to active CO2 injectors.
There also is evidence of limited gas breakthrough in Well Nos 4827, 4828, 4829 and 4842. However, the
limitations on well test accuracy and the impact of changes in the injection scheme are such that a well’s
performance needs to be observed over several quarters before making a final judgement. Well 4843 is a
good example of this. For the last couple of quarters it was reported as having an oil response to CO2
injection, but that no longer appears to be the case. Figure 9 shows well response in terms of oil response and
gas breakthrough.
As discussed in the last quarterly report, determining when response occurs is not an exact science
in a reservoir with the injection and withdrawal history of West Welch. The classic response of a producer to
CO2 injection is an initial lowering of water production followed by an increase in gas (mainly CO2) as the
miscible front encroaches into the well’s drainage area. Oil rate increases as the front approaches the
wellbore. Several variations to this pattern have occurred as wells are influenced not only by changes in CO2
injection rates, i.e. pinching back or wagging, but also the changes in what was the established waterflood
injection pattern and resulting reservoir pressure distribution. Although our judgement has changed on some
wells based on long term performance, there is no new oil response or gas breakthrough that occurred in this
quarter.
The low hydrocarbon pore volume processing rate of the CO2 remains the main concern in light of
the September 2001 termination date of the project. Only a few of the pattens are approaching a processed
reservoir volume where similar projects started to experience response. The restrictions on CO2 injection
since November 2000 have only worsened the problem. The return of the 4805 injector to CO2 input will help
as well as the injectors offsetting the 4853 lateral being restored to their maximum input rate.
TECHNOLOGY TRANSFER
Treatment Summary
Initial Attempt 11/29/00
OXY-WWU 4853
Notes: 1 Proppant for initial stage at each interval was 20/40 Sintered Bauxite
2 Proppant for subsequent stages was 20/40 resin-coated sand
3 Fluid is cross link gelled water (Deltafrac)
Treatment Summary
Second Attempt 1/25/01
OXY-WWU 4853
Notes: 1 Proppant for initial stage at each interval was 20/40 Sintered Bauxite
2 Proppant for subsequent stages was 20/40 resin-coated sand
3 Fluid is cross link gelled water (Deltafrac)
# of Injectors on water 1 1 0
Production
Base oil production (bopd) 131 130 129 130
Actual oil production (bopd) 178 180 186 182
Incremental oil production (bopd) 47 50 57 52
Cum % OOIP
Base WOR 13 13 13
WOR 5.3 5.2 5.1 5.2