EPA Region 9
UIC PRESSURE FALLOFF
REQUIREMENTS
Condensed version of the
EPA Region 6
UIC PRESSURE FALLOFF
TESTING GUIDELINE
Third Revision
August 8, 2002
-------
TABLE OF CONTENTS
1.0 Background
2.0 Purpose of Guideline
3.0 Timing of Falloff Tests and Report Submission
4.0 Falloff Test Report Requirements
5.0 Planning
General Operational Concerns
Site Specific Pretest Planning
6.0 Conducting the Falloff Test
7.0 Evaluation of the Falloff Test
1. Cartesian Plot
2. Log-log Plot
3. SemilogPlot
4. Anomalous Results
8.0 Technical References
APPENDIX
Pressure Gauge Usage and Selection
Usage
Selection
Test Design
General Operational Considerations
Wellbore and Reservoir Data Needed to Simulate or Analyze the Falloff Test
Design Calculations
Considerations for Offset Wells Completed in the Same Interval
Falloff Test Analysis
Cartesian Plot
Log-log Diagnostic Plot
Identification of Test Flow Regimes
Characteristics of Individual Test Flow Regimes
Wellbore Storage
Radial Flow
Spherical Flow
Linear Flow
Hydraulically Fractured Well
Naturally Fractured Rock
Layered Reservoir
Page 2 of27
-------
Semilog Plot Determination of the Appropriate Time Function for the Semilog Plot
Parameter Calculations and Considerations
Skin
Radius of Investigation
Effective Wellbore Radius
Reservoir Injection Pressure Corrected for Skin Effects
Determination of the Appropriate Fluid Viscosity
Reservoir Thickness
Use of Computer Software
Common Sense Check
Page3 of 27
-------
REQUIREMENTS
UIC PRESSURE FALLOFF TESTING GUIDELINE
Third Revision
August 8, 2002
1.0 Background
Region 9 has adopted the Region 6 UIC Pressure Falloff Testing Guideline requirements for
monitoring Class 1 Non Hazardous waste disposal wells. Under 40 CFR 146.13(d)(l), operators
are required annually to monitor the pressure buildup in the injection zone, including at a
minimum, a shut down of the well for a time sufficient to conduct a valid observation of the
pressure falloff curve.
All of the following parameters (Test, Period, Analysis) are critical for
evaluation of technical adequacy of UIC permits:
A falloff tCSt is a pressure transient test that consists of shutting in an injection well and
measuring the pressure falloff. The falloff period is a replay of the injection preceding it;
consequently, it is impacted by the magnitude, length, and rate fluctuations of the injection
period. Falloff testing analysis provides transmissibility, skin factor, and well flowing and
static pressures.
2.0 Purpose of Guideline
This guideline has been adopted by the Region 9 office of the Evironmental Protection Agency
(EPA) to assist operators in planning and conducting the falloff test and preparing the
annual monitoring report
Falloff tests provide reservoir pressure data and characterize both the injection interval reservoir
and the completion condition of the injection well. Both the reservoir parameters and pressure
data are necessary for UIC permit demonstrations. Additionally, a valid falloff test is a
monitoring requirement under 40 CFR Part 146 for all Class I injection wells.
The ultimate responsibility of conducting a valid falloff test is the task of the operator.
Operators should QA/QC the pressure data and test results to confirm that the results "make
sense" prior to submission of the report to the EPA for review.
Page 4 of27
-------
3.0 Timing of Falloff Tests and Report Submission
Falloff tests must be conducted annually. The time interval for each test should not be less
than 9 months or greater than 15 months from the previous test. This will ensure that the tests
will be performed at relatively even intervals.
The falloff testing report should be submitted no later than 60 days following the test. Failure
to submit a falloff test report will be considered a violation and may result in an enforcement
action. Any exceptions should be approved by EPA prior to conducting the test.
4.0 Falloff Test Report Requirements
In general, the report to EPA should provide:
(1) general information and an overview of the falloff test,
(2) an analysis of the pressure data obtained during the test,
(3) a summary of the test results, and
(4) a comparison of those results with previously used parameters.
Some of the following operator and well data will not change so once acquired, it can be copied
and submitted with each annual report. The falloff test report should include the following
information:
1. Company name and address
2 Test well name and location
3. The name and phone number of the facility contact person. The contractor contact may
be included if approved by the facility in addition to a facility contact person.
4. A photocopy of an openhole log (SP or Gamma Ray) through the injection interval
illustrating the type of formation and thickness of the injection interval. The entire log is
not necessary.
5. Well schematic showing the current wellbore configuration and completion information:
X Wellbore radius
X Completed interval depths
X Type of completion (perforated, screen and gravel packed, openhole)
6 Depth of fill depth and date tagged.
7. Offset well information:
X Distance between the test well and offset well(s) completed in the same interval
or involved in an interference test
X Simple illustration of locations of the injection and offset wells
8. Chronological listing of daily testing activities.
9 Electronic submission of the raw data (time, pressure, and temperature) from all
pressure gauges utilized on CD-ROM. A README file or the disk label should list all
files included and any necessary explanations of the data. A separate file containing any
Page 5 of27
-------
edited data used in the analysis can be submitted as an additional file.
10 Tabular summary of the injection rate or rates preceding the falloff test. At a
minimum, rate information for 48 hours prior to the falloff or for a time equal to twice the
time of the falloff test is recommended. If the rates varied and the rate information is
greater than 10 entries, the rate data should be submitted electronically as well as a hard
copy of the rates for the report. Including a rate vs time plot is also a good way to
illustrate the magnitude and number of rate changes prior to the falloff test.
11. Rate information from any offset wells completed in the same interval. At a
minimum, the injection rate data for the 48 hours preceding the falloff test should be
included in a tabular and electronic format. Adding a rate vs time plot is also helpful to
illustrate the rate changes.
12 Hard copy of the time and pressure data analyzed in the report
13. Pressure gauge information: (See Appendix, page A-l for more information on
pressure gauges)
X List all the gauges utilized to test the well
X Depth of each gauge
X Manufacturer and type of gauge. Include the full range of the gauge.
X Resolution and accuracy of the gauge as a % of full range.
X Calibration certificate and manufacturer's recommended frequency of calibration
14. General test information:
X Date of the test
X Time synchronization: A specific time and date should be synchronized to an
equivalent time in each pressure file submitted. Time synchronization should also
be provided for the rate(s) of the test well and any offset wells.
X Location of the shut-in valve (e.g., note if at the wellhead or number of feet from
the wellhead)
15 Reservoir parameters (determination):
X Formation fluid viscosity, Uf cp (direct measurement or correlation)
X Porosity, (p fraction (well log correlation or core data)
X Total compressibility, Ct psi"1 (correlations, core measurement, or well test)
X Formation volume factor, rvb/stb (correlations, usually assumed 1 for water)
X Initial formation reservoir pressure - See Appendix, page A-l
X Date reservoir pressure was last stabilized (injection history)
X Justified interval thickness, h ft - See Appendix, page A-15
16. Waste plume:
X Cumulative injection volume into the completed interval
X Calculated radial distance to the waste front, rwaste ft
X Average historical waste fluid viscosity, if used in the analysis, ^iwaste cp
Page 6 of27
-------
17 Injection period:
X Time of injection period
X Type of test fluid
X Type of pump used for the test (e.g., plant or pump truck)
X Type of rate meter used
X Final injection pressure and temperature
18 Falloff period:
X Total shut-in time, expressed in real time and At, elapsed time
X Final shut-in pressure and temperature
X Time well went on vacuum, if applicable
19 Pressure gradient:
X Gradient stops - for depth correction
20. Calculated test data: include all equations used and the parameter values assigned for
each variable within the report
X Radius of investigation, n ft
X Slope or slopes from the semilog plot
X Transmissibility, kh/[j, md-ft/cp
X Permeability (range based on values of h)
X Calculation of skin, s
X Calculation of skin pressure drop, APskin
X Discussion and justification of any reservoir or outer boundary models used to
simulate the test
X Explanation for any pressure or temperature anomaly if observed
21 Graphs:
X Cartesian plot: pressure and temperature vs. time
X Log-log diagnostic plot: pressure and semilog derivative curves. Radial flow
regime should be identified on the plot
X Semilog and expanded semilog plots: radial flow regime indicated and the
semilog straight line drawn
X Injection rate(s) vs time: test well and offset wells (not a circular or strip chart)
22. A copy of the latest radioactive tracer run and a brief discussion of the results.
5.0 Planning
The radial flow portion of the test is the basis for all pressure transient calculations.
Therefore the injectivity and falloff portions of the test should be designed not only to reach
radial flow, but to sustain a time frame sufficient for analysis of the radial flow period.
General Operational Concerns
X Adequate storage for the waste should be ensured for the duration of the test
Page 7 of27
-------
X Offset wells completed in the same formation as the test well should be shut-in, or at a
minimum, provisions should be made to maintain a constant injection rate prior to and
during the test
X Install a crown valve on the well prior to starting the test so the well does not have to be
shut-in to install a pressure gauge
X The location of the shut-in valve on the well should be at or near the wellhead to
minimize the wellbore storage period
X The condition of the well, junk in the hole, wellbore fill or the degree of wellbore damage
(as measured by skin) may impact the length of time the well must be shut-in for a valid
falloff test. This is especially critical for wells completed in relatively low
transmissibility reservoirs or wells that have large skin factors.
X Cleaning out the well and acidizing may reduce the wellbore storage period and therefore
the shut-in time of the well
X Accurate recordkeeping of injection rates is critical including a mechanism to
synchronize times reported for injection rate and pressure data. The elapsed time format
usually reported for pressure data does not allow an easy synchronization with real time
rate information. Time synchronization of the data is especially critical when the
analysis includes the consideration of injection from more than one well.
X Any unorthodox testing procedure, or any testing of a well with known or anticipated
problems, should be discussed with EPA staff prior to performing the test.
X If more than one well is completed into the same reservoir, operators are encouraged to
send at least two pulses to the test well by way of rate changes in the offset well
following the falloff test. These pulses will demonstrate communication between the
wells and, if maintained for sufficient duration, they can be analyzed as an interference
test to obtain interwell reservoir parameters.
Site Specific Pretest Planning
1. Determine the time needed to reach radial flow during the injectivity and falloff portions
of the test:
X Review previous welltests, if available
X Simulate the test using measured or estimated reservoir and well completion
parameters
X Calculate the time to the beginning of radial flow using the empirically-based
equations provided in the Appendix. The equations are different for the
injectivity and falloff portions of the test with the skin factor influencing the
falloff more than the injection period. (See Appendix, page A-4 for equations)
X Allow adequate time beyond the beginning of radial flow to observe radial flow
so that a well developed semilog straight line occurs. A good rule of thumb is 3
to 5 times the time to reach radial flow to provide adequate radial flow data for
analysis.
2. Adequate and consistent injection fluid should be available so that the injection rate into
the test well can be held constant prior to the falloff. This rate should be high enough to
Page 8 of27
-------
produce a measurable falloff at the test well given the resolution of the pressure gauge
selected. The viscosity of the fluid should be consistent. Any mobility issues (k/u)
should be identified and addressed in the analysis if necessary.
3. Bottomhole pressure measurements are required. (See Appendix, page A-2 for additional
information concerning pressure gauge selection.)
4. Use two pressure gauges during the test with one gauge serving as a backup, or for
verification in cases of questionable data quality. The two gauges do not need to be the
same type. (See Appendix, page A-l for additional information concerning pressure
gauges.)
6.0 Conducting the Falloff Test
1. Tag and record the depth to any fill in the test well
2. Simplify the pressure transients in the reservoir
X Maintain a constant injection rate in the test well prior to shut-in. This injection
rate should be high enough and maintained for a sufficient duration to produce a
measurable pressure transient that will result in a valid falloff test.
X Offset wells should be shut-in prior to and during the test. If shut-in is not
feasible, a constant injection rate should be recorded and maintained during the
test and then accounted for in the analysis.
X Do not shut-in two wells simultaneously or change the rate in an offset well
during the test.
3. The test well should be shut-in at the wellhead in order to minimize wellbore storage and
afterflow. (See Appendix, page A-3 for additional information.)
4. Maintain accurate rate records for the test well and any offset wells completed in the
same injection interval.
5. Measure and record the viscosity of the injectate periodically during the injectivity
portion of the test to confirm the consistency of the test fluid.
7.0 Evaluation of the Falloff Test
1. Prepare a Cartesian plot of the pressure and temperature versus real time or elapsed
time.
X Confirm pressure stabilization prior to shut-in of the test well
X Look for anomalous data, pressure drop at the end of the test, determine if
pressure drop is within the gauge resolution
2. Prepare a log-log diagnostic plot of the pressure and semilog derivative. Identify the
Page 9 of27
-------
flow regimes present in the welltest. (See Appendix, page A-6 for additional
information.)
X Use the appropriate time function depending on the length of the injection period
and variation in the injection rate preceding the falloff (See Appendix, page A-10
for details on time functions.)
X Mark the various flow regimes - particularly the radial flow period
X Include the derivative of other plots, if appropriate (e.g., square root of time for
linear flow)
X If there is no radial flow period, attempt to type curve match the data
3. Prepare a semilog plot.
X Use the appropriate time function depending on the length of injection period and
injection rate preceding the falloff
X Draw the semilog straight line through the radial flow portion of the plot and
obtain the slope of the line
X Calculate the transmissibility, kh/^i
X Calculate the skin factor, s, and skin pressure drop,
X Calculate the radius of investigation, r;
4. Explain any anomalous results.
8.0 Technical References
1. SPE Textbook Series No. 1, "Well Testing," 1982, W. John Lee
2. SPE Monograph 5, "Advances in Well Test Analysis," 1977, Robert Earlougher, Jr.
3. SPE Monograph 1, "Pressure Buildup and Flow Tests in Wells," 1967, C.S. Matthews
and D.G. Russell
4. "Well Test Interpretation In Bounded Reservoirs," Hart's Petroleum Engineer
International, Spivey, and Lee, November 1997
5. "Derivative of Pressure: Application to Bounded Reservoir Interpretation," SPE Paper
15861,Proano, Lilley, 1986
6. "Well Test Analysis," Sabet, 1991
7. "Pressure Transient Analysis," Stanislav and Kabir, 1990
8. "Well Testing: Interpretation Methods," Bourdarot, 1996
9. "A New Method To Account For Producing Time Effects When Drawdown Type Curves
Are Used To Analyze Pressure Buildup And Other Test Data," SPE Paper 9289,
Agarwal, 1980
10. "Modern Well Test Analysis - A Computer-Aided Approach," Roland N. Home, 1990
11. Exxon Monograph, "Well Testing in Heterogeneous Formations," Tatiana Streltsova,
1987
12. EPA Region 6 Falloff Guidelines
13. "Practical Pressure Gauge Specification Considerations In Practical Well Testing," SPE
Paper No. 22752, Veneruso, Ehlig-Economides, andPetitjean, 1991
Page 10 of 27
-------
14. "Guidelines Simplify Well Test Interpretation," Oil and Gas Journal, Ehlig-Economides,
Hegeman, and Vik, July 18, 1994
15. Oryx Energy Company, Practical Pressure Transient Testing, G. Lichtenberger and K.
Johnson, April 1990 (Internal document)
16. Pressure-Transient Test Design in Tight Gas Formations, SPE Paper 17088, W.J. Lee,
October 1987
17. "Radius-of-Drainage and Stabilization-Time Equations," Oil and Gas Journal, H.K. Van
Poollen, Sept 14, 1964
18. "Effects of Permeability Anisotropy and Layering On Well Test Interpretation," Hart's
Petroleum Engineer International, Spivey, Aly, and Lee, February 1998
19. "Three Key Elements Necessary for Successful Testing," Oil and Gas Journal, Ehlig-
Economides, Hegeman, Clark, July 25, 1994
20. "Introduction to Applied Well Test Interpretation," Hart's Petroleum Engineer
International, Spivey, and Lee, August 1997
21. "Recent Developments In Well Test Analysis," Hart's Petroleum Engineer International,
Stewart, August 1997
22. "Fundamentals of Type Curve Analysis," Hart's Petroleum Engineer International,
Spivey, and Lee, September 1997
23. "Identifying Flow Regimes In Pressure Transient Tests," Hart's Petroleum Engineer
International, Spivey and Lee, October 1997
24. "Selecting a Reservoir Model For Well Test Interpretation," Hart's Petroleum Engineer
International, Spivey, Ayers, Pursell,and Lee, December 1997
27. "Use of Pressure Derivative in Well-Test Interpretation," SPE Paper 12777, SPE
Formation Evaluation Journal, Bourdet, Ayoub, and Pirard, June 1989
28. "A New Set of Type Curves Simplifies Well Test Analysis," World Oil, Bourdet,
Whittle, Douglas, and Pirard, May 1983
Page 11 of 27
-------
APPENDIX
Pressure Gauge Usage and Selection
Usage
X EPA recommends that two gauges be used during the test with one gauge serving as a
backup.
X Downhole pressure measurements are less noisy and are required.
X A bottomhole surface readout gauge (SRO) allows tracking of pressures in real time.
Analysis of this data can be performed in the field to confirm that the well has reached
radial flow prior to ending the test.
X The derivative function plotted on the log-log plot amplifies noise in the data, so the use
of a good pressure recording device is critical for application of this curve.
X Mechanical gauges should be calibrated before and after each test using a dead weight
tester.
X Electronic gauges should also be calibrated according to the manufacturer's
recommendations. The manufacturer's recommended frequency of calibration, and a
copy of the gauge calibration certificate should be provided with the falloff testing report
demonstrating this practice has been followed.
Selection
X The pressures must remain within the range of the pressure gauge. The larger percent of
the gauge range utilized in the test, the better. Typical pressure gauge limits are 2000,
5000, and 10000 psi. Note that gauge accuracy and resolution are typically a function of
percent of the full gauge range.
X Electronic downhole gauges generally offer much better resolution and sensitivity than a
mechanical gauge but cost more. Additionally, the electronic gauge can generally run for
a longer period of time, be programmed to measure pressure more frequently at various
intervals for improved data density, and store data in digital form.
X Resolution of the pressure gauge must be sufficient to measure small pressure changes at
the end of the test.
Test Design
General Operational Considerations
X The injection period controls what is seen on the falloff since the falloff is replay of the
injection period. Therefore, the injection period must reach radial flow prior to shut-in of
the well in order for the falloff test to reach radial flow
X Ideally to determine the optimal lengths of the injection and falloff periods, the test
should be simulated using measured or estimated reservoir parameters. Alternatively,
injection and falloff period lengths can be estimated from empirical equations using
assumed reservoir and well parameters.
Page 12 of 27
-------
X The injection rate dictates the pressure buildup at the injection well. The pressure
buildup from injection must be sufficient so that the pressure change during radial flow,
usually occurring toward the end of the test, is large enough to measure with the pressure
gauge selected.
X Waste storage and other operational issues require preplanning and need to be addressed
prior to the test date. If brine must be brought in for the injection portion of the test,
operators should insure that the fluid injected has a consistent viscosity and that there is
adequate fluid available to obtain a valid falloff test. The use of the wastestream as the
injection fluid affords several distinct advantages:
1. Brine does not have to be purchased or stored prior to use.
2. Onsite waste storage tanks may be used.
3. Plant wastestreams are generally consistent, i.e., no viscosity variations
X Rate changes cause pressure transients in the reservoir. Constant rate injection in the
test well and any offset wells completed in the same reservoir are critical to simplify
the pressure transients in the reservoir. Any significant injection rate fluctuations at
the test well or offsets must be recorded and accounted for in the analysis using
superposition.
X Unless an injectivity test is to be conducted, shutting in the well for an extend period of
time prior to conducting the falloff test reduces the pressure buildup in the reservoir and
is not recommended.
X Prior to conducting a test, a crown valve should be installed on the wellhead to allow the
pressure gauge to be installed and lowered into the well without any interruption of the
injection rate.
X The wellbore schematic should be reviewed for possible obstructions located in the well
that may prevent the use or affect the setting depth of a downhole pressure gauge. The
fill depth in the well should also be reported. The fill depth may not only impact the
depth of the gauge, but usually prolongs the wellbore storage period and depending on
the type of fill, may limit the interval thickness by isolating some of the injection
intervals. A wellbore cleanout or stimulation may be needed prior to conducting the test
for the test to reach radial flow and obtain valid results.
X The location of the shut-in valve can impact the duration of the wellbore storage period.
The shut-in valve should be located near the wellhead. Afterflow into the wellbore
prolongs the wellbore storage period.
X The area geology should be reviewed prior to conducting the test to determine the
thickness and type of formation being tested along with any geological features such as
natural fractures, a fault, or a pinchout that should be anticipated to impact the test.
Wellbore and Reservoir Data Needed to Simulate or Analyze the Falloff Test
X Wellbore radius, rw - from wellbore schematic
Page 13 of 27
-------
X Net thickness, h - See Appendix, page A-15
X Porosity, q> - log or core data
X Viscosity of formation fluid, Uf - direct measurement or correlations
X Viscosity of waste, Uwaste - direct measurement or correlations
X Total system compressibility, ct - correlations, core measurement, or well test
X Permeability, k - previous welltests or core data
X Specific gravity of injection fluid, s.g. - direct measurement
X Injection rate, q - direct measurement
Design Calculations
When simulation software is unavailable the test periods can be estimated from empirical
equations. The following are set of steps to calculate the time to reach radial flow from
empirically-derived equations:
1. Estimate the wellbore storage coefficient, C (bbl/psi). There are two equations to
calculate the wellbore storage coefficient depending on if the well remains fluid filled
(positive surface pressure) or if the well goes on a vacuum (falling fluid level in the
well):
a. Well remains fluid filled:
w -c waste where, Vw is the total wellbore volume, bbls
Cwaste is the compressibility of the injectate, psi"1
b. Well goes on a vacuum:
c=J^
&c where, Vu is the wellbore volume per unit
length, bbls/ft
p is the injectate density, psi/ft
g and gc are gravitational constants
2. Calculate the time to reach radial flow for both the injection and falloff periods. Two
different empirically-derived equations are used to calculate the time to reach radial flow,
w, for the injectivity and falloff periods:
Injectivity period:
(200000 +12000s)-C
radial flow
k-h
jU
b. Falloff period:
170000-C-eou's ,
> —; hours
hours
JU
The wellbore storage coefficient is assumed to be the same for both the injectivity and
falloff periods. The skin factor, s, influences the falloff more than the injection period.
Use these equations with caution, as they tend to fall apart for a well with a large
Page 14 of 27
-------
permeability or a high skin factor. Also remember, the welltest should not only reach
radial flow, but also sustain radial flow for a timeframe sufficient for analysis of the
radial flow period. As a rule of thumb, a timeframe sufficient for analysis is 3 to 5 times
the time needed to reach radial flow.
3. As an alternative to steps 1 and 2, to look a specific distance "L" into the reservoir and
possibly confirm the absence or existence of a boundary, the following equation can be
used to estimate the time to reach that distance:
948 •^•//•ct -Lbomdaty
t boundary = £ k°UrS
where, LbOUndary = feet to boundary
tboundary = time to boundary, hrs
Again, this is the time to reach a distance "L" in the reservoir. Additional test time is
required to observe a fully developed boundary past the time needed to just reach the
boundary. As a rule of thumb, to see a fully developed boundary on a log-log plot, allow
at least 5 times the time to reach it. Additionally, for a boundary to show up on the
falloff, it must first be encountered during the injection period.
4. Calculate the expected slope of the semilog plot during radial flow to see if gauge
resolution will be adequate using the following equation:
162.6-g-B
MI — i
semAoe /, _ /,
where, q = the injection rate preceding the falloff test, bpd
B = formation volume factor for water, rvb/stb (usually assumed to be 1)
Considerations for Offset Wells Completed in the Same Interval
Rate fluctuations in offset wells create additional pressure transients in the reservoir and
complicate the analysis. Always try to simplify the pressure transients in the reservoir. Do not
simultaneously shut-in an offset well and the test well. The following items are key
considerations in dealing with the impact of offset wells on a falloff test:
X Shut-in all offset wells prior to the test
X If shutting in offset wells is not feasible, maintain a constant injection rate prior to and
during the test
X Obtain accurate injection records of offset injection prior to and during the test
X At least one of the real time points corresponding to an injection rate in an offset well
should be synchronized to a specific time relating to the test well
X Following the falloff test in the test well, send at least two pulses from the offset well
to the test well by fluctuating the rate in the offset well. The pressure pulses can
confirm communication between the wells and can be simulated in the analysis if
observed at the test well. The pulses can also be analyzed as an interference test using an
Ei type curve.
Page 15 of 27
-------
X If time permits, conduct an interference test to allow evaluation of the reservoir without
the wellbore effects observed during a falloff test.
Falloff Test Analysis
In performing a falloff test analysis, a series of plots and calculations should be prepared to
QA/QC the test, identify flow regimes, and determine well completion and reservoir parameters.
Individual plots, flow regime signatures, and calculations are discussed in the following
sections.
Cartesian Plot
X The pressure data prior to shut-in of the well should be reviewed on a Cartesian plot to
confirm pressure stabilization prior to the test. A well that has reached radial flow during
the injectivity portion of the test should have a consistent injection pressure.
X A Cartesian plot of the pressure and temperature versus real time or elapsed time should
be the first plot made from the falloff test data. Late time pressure data should be
expanded to determine the pressure drop occurring during this portion of the test. The
pressure changes should be compared to the pressure gauges used to confirm adequate
gauge resolution existed throughout the test. If the gauge resolution limit was reached,
this timeframe should be identified to determine if radial flow was reached prior to
reaching the resolution of the pressure gauge. Pressure data obtained after reaching the
resolution of the gauge should be treated as suspect and may need to be discounted in the
analysis.
X Falloff tests conducted in highly transmissive reservoirs may be more sensitive to the
temperature compensation mechanism of the gauge because the pressure buildup
response evaluated is smaller. Region 6 has observed cases in which large temperature
anomalies were not properly compensated for by the pressure gauge, resulting in
erroneous pressure data and an incorrect analysis. For this reason, the Cartesian plot of
the temperature data should be reviewed Any temperature anomalies should be noted
to determine if they correspond to pressure anomalies.
X Include the injection rate(s) of the test well 48 hours prior to shut-in on the Cartesian plot
to illustrate the consistency of the injection rate prior to shut-in and to determine the
appropriate time function to use on the log-log and semilog plots. (See Appendix, page
A10 for time function selection)
Page 16 of 27
-------
Log-log Diagnostic Plot
X
Plot the pressure and semilog derivative versus time on a log-log diagnostic plot. Use the
appropriate time function based on the rate history of the injection period preceding the
falloff. (See Appendix, page A-10 for time function selection) The log-log plot is used
to
the flow
present
An
log-log
shown
10000
lent
100
Example Log-log Plot
- Wellbore Storage
X
• —
*
j
\\
|
1
a
/I
\
\
?*
Period £
^
ri
W
&
J^1^
Transitic
Unit slope during
wellbore storage
J55-— *5
w*
^
\
>n period \
•
\
f
-.1111 1 !,_ :
i
,
k'-'
r •
'ressure
Data
i i ^^^^m
Sen
Dem
LJ
\
^^m
lilc
/at
^m
g
V(
M
P
9
• • •
res
:un
EEE
Radii
Flow
^u.
f—
BOX
Derivative flatten
f
tenn c
• IM
sure
ctio
il
5 = ^
5
n
identify
regimes
in the
welltest.
example
plot is
below:
0.001
0.01
0.1
Elapsed Time (hours) - Tp=24.0
Identification of Test Flow Regimes
X Flow regimes are mathematical relationships between pressure, rate, and time. Flow
regimes provide a visualization of what goes on in the reservoir. Individual flow regimes
have characteristic slopes and a sequencing order on the log-log plot.
X Various flow regimes will be present during the falloff test, however, not all flow
regimes are observed on every falloff test. The late time responses correlate to distances
further from the test well. The critical flow regime is radial flow from which all
analysis calculations are performed. During radial flow, the pressure responses
recorded are representative of the reservoir, not the wellbore.
X The derivative function amplifies reservoir signatures by calculating a running slope of a
designated plot. The derivative plot allows a more accurate determination of the radial
flow portion of the test, in comparison with the old method of simply proceeding ll/2 log
cycles from the end of the unit slope line of the pressure curve.
X The derivative is usually based on the semilog plot, but it can also be calculated based on
other plots such as a Cartesian plot, a square root of time plot, a quarter root of time plot,
and the 1/square root of time plot. Each of these plots are used to identify specific flow
regimes. If the flow regime characterized by a specialized plot is present then when the
derivative calculated from that plot is displayed on the log-log plot, it will appear as a
Page 17 of 27
-------
"flat spot" during the portion of the falloff corresponding to the flow regime.
X Typical flow regimes observed on the log-log plot and their semilog derivative patterns
are listed below:
Flow Regime Semilog Derivative Pattern
Wellbore Storage Unit slope
Radial Flow Flat plateau
Linear Flow Half slope
Bilinear Flow Quarter slope
Partial Penetration Negative half slope
Layering Derivative trough
Dual Porosity Derivative trough
Boundaries Upswing followed by plateau
Constant Pressure Sharp derivative plunge
Characteristics of Individual Test Flow Regimes
X Wellbore Storage:
1. Occurs during the early portion of the test and is caused by the well being shut-in
at the surface instead of the sandface
2. Measured pressure responses are governed by well conditions and are not
representative of reservoir behavior and are characterized by both the pressure
and semilog derivative curves overlying a unit slope on the log-log plot
3. Wellbore skin or a low permeability reservoir results in a slower transfer of fluid
from the well to the formation, extending the duration of the wellbore storage
period
4. A wellbore storage dominated test is unanalyzable
X Radial Flow:
1. The pressure responses are from the reservoir, not the wellbore
2. The critical flow regime from which key reservoir parameters and completion
conditions calculations are performed
3. Characterized by a flattening of the semilog plot derivative curve on the log-log
plot and a straight line on the semilog plot
X Spherical Flow:
1. Identifies partial penetration of the injection interval at the wellbore
2. Characterized by the semilog derivative trending along a negative half slope on
the log-log plot and a straight line on the I/square root of time plot
3. The log-log plot derivative of the pressure vs I/square root of time plot is flat
Page 18 of 27
-------
X Linear Flow:
1. May result from flow in a channel, parallel faults, or a highly conductive fracture
2. Characterized by a half slope on both the log-log plot pressure and semilog
derivative curves with the derivative curve approximately 1/3 of a log cycle lower
than the pressure curve and a straight line on the square root of time plot. 3.
The log-log plot derivative of the pressure vs square root of time plot is
flat
X Hydraulically Fractured Well:
1. Multiple flow regimes present including wellbore storage, fracture linear flow,
bilinear flow, pseudo-linear flow, formation linear flow, and pseudo-radial flow
2. Fracture linear flow is usually hidden by wellbore storage
3. Bilinear flow results from simultaneous linear flows in the fracture and from the
formation into the fracture, occurs in low conductivity fractures, and is
characterized by a quarter slope on both the pressure and semilog derivative
curves on the log-log plot and by a straight line on a pressure versus quarter root
of time plot
4. Formation linear flow is identified by a half slope on both the pressure and
semilog derivative curves on the log-log plot and by a straight line on a pressure
versus square root of time plot
5. Psuedo-radial flow is analogous to radial flow in an unfractured well and is
characterized by flattening of semilog derivative curve on the log-log plot and a
straight line on a semilog pressure plot
X Naturally Fractured Rock:
1. The fracture system will be observed first on the falloff test followed by the total
system consisting of the fractures and matrix.
2. The falloff analysis is complex. The characteristics of the semilog derivative
trough on the log-log plot indicate the level of communication between the
fractures and the matrix rock.
X Layered Reservoir:
1. Analysis of a layered system is complex because of the different flow regimes,
skin factors or boundaries that may be present in each layer.
2. The falloff test objective is to get a total tranmissibility from the whole reservoir
system
3. Typically described as commingled (2 intervals with vertical separation) or
crossflow (2 intervals with hydraulic vertical communication)
Semilog Plot
X The semilog plot is a plot of the pressure versus the log of time. There are typically four
different semilog plots used in pressure transient and falloff testing analysis. After
plotting the appropriate semilog plot, a straight line should be drawn through the points
located within the equivalent radial flow portion of the plot identified from the log-log
Page 19 of 27
-------
plot.
X Each plot uses a different time function depending on the length and variation of the
injection rate preceding the falloff. These plots can give different results for the same
test, so it is important that the appropriate plot with the correct time function is used for
the analysis. Determination of the appropriate time function is discussed below.
X The slope of the semilog straight line is then used to calculate the reservoir
transmissibility - kh/u, the completion condition of the well via the skin factor - s, and
also the radius of investigation - n of the test.
Determination of the Appropriate Time Function for the Semilog Plot
The following four different semilog plots are used in pressure transient analysis:
1. Miller Dyes Hutchinson (MDH) Plot
2. Horner Plot
3. Agarwal Equivalent Time Plot
4. Superposition Time Plot
These plots can give different results for the same test. Use of the appropriate plot with the
correct time function is critical for the analysis.
X The MDH plot is a semilog plot of pressure versus At, where At is the elapsed shut-in
time of the falloff.
1. The MDH plot only applies to wells that reach psuedo-steady state during
injection. Psuedo-steady state means the pressure response from the well has
encountered all the boundaries around the well.
2. The MDH plot is only applicable to injection wells with a very long injection
period at a constant rate. This plot is not recommended for use by EPA Region 6.
X The Horner plot is a semilog plot of pressure versus (tp+At)/At. The Horner plot is only
used for a falloff preceded by a single constant rate injection period.
1. The injection time, tp=Vp/q in hours, where Vp=injection volume since the last
pressure equalization and q is the injection rate prior to shut-in for the falloff test.
The injection volume is often taken as the cumulative injection since completion.
2. The Horner plot can result in significant analysis error if the injection rate varies
prior to the falloff.
X The Agarwal equivalent time plot is a semilog plot of the pressure versus Agarwal
equivalent time, Ate.
1. The Agarwal equivalent time function is similar to the Horner plot, but scales the
falloff to make it look like an injectivity test.
2. It is used when the injection period is a short, constant rate compared to the length
of the falloff period.
3. The Agarwal equivalent time is defined as: Ate=log(tp At)/(tp+At), where tp is
calculated the same as with the Horner plot.
Page 20 of 27
-------
X The superposition time function accounts for variable rate conditions preceding the
falloff.
1. It is the most rigorous of all the time functions and is usually calculated using
welltest software.
2. The use of the superposition time function requires the operator to accurately
track the rate history. As a rule of thumb, at a minimum, the rate history for twice
the length of the falloff test should be included in the analysis.
The determination of which time function is appropriate for the plotting the welltest on semilog
and log-log plots depends on available rate information, injection period length, and software:
1. If there is not a rate history other than a single rate and cumulative injection, use a Horner
time function
2. If the injection period is shorter than the falloff test and only a single rate is available, use
the Agarwal equivalent time function
3. If you have a variable rate history use superposition when possible. As an alternative to
superposition, use Agarwal equivalent time on the log-log plot to identify radial flow.
The semilog plot can be plotted in either Horner or Agarwal time if radial flow is
observed on the log-log plot.
Parameter Calculations and Considerations
X Transmissibility - The slope of the semilog straight line, m, is used to determine the
transmissibility (kh/u) parameter group from the following equation:
k-h _ 162.6-ff-B
ju m
where, q = injection rate, bpd (negative for injection)
B = formation volume factor, rvb/stb (Assumed to be 1 for formation
fluid)
m = slope of the semilog straight line through the radial flow portion of
the plot in psi/log cycle
k = permeability, md
h = thickness, ft (See Appendix, page A-15)
[j, = viscosity, cp
X The viscosity, [j,, is usually that of the formation fluid. However, if the waste plume size
is massive, the radial flow portion of the test may remain within the waste plume. (See
Appendix, page A-14)
1. The waste and formation fluid viscosity values usually are similar, however, if the
wastestream has a significant viscosity difference, the size of the waste plume and
distance to the radial flow period should be calculated.
2. The mobility, k/u, differences between the fluids may be observed on the
derivative curve.
Page 21 of 27
-------
X The permeability, k, can be obtained from the calculated transmissibility (kh/u) by
substituting the appropriate thickness, h, and viscosity, u, values.
Skin Factor
X In theory, wellbore skin is treated as an infinitesimally thin sheath surrounding the
wellbore, through which a pressure drop occurs due to either damage or stimulation.
Industrial injection wells deal with a variety of waste streams that alter the near wellbore
environment due to precipitation, fines migration, ion exchange, bacteriological
processes, and other mechanisms. It is reasonable to expect that this alteration often
exists as a zone surrounding the wellbore and not a skin. Therefore, at least in the case of
industrial injection wells, the assumption that skin exists as a thin sheath is not always
valid. This does not pose a serious problem to the correct interpretation of falloff testing
except in the case of a large zone of alteration, or in the calculation of the flowing
bottomhole pressure. Region 6 has seen instances in which large zones of alteration were
suspected of being present.
X The skin factor is the measurement of the completion condition of the well. The skin
factor is quantified by a positive value indicating a damaged completion and a negative
value indicating a stimulated completion.
1. The magnitude of the positive value indicating a damaged completion is dictated
by the transmissibility of the formation.
2. A negative value of -4 to -6 generally indicates a hydraulically fractured
completion, whereas a negative value of-1 to -3 is typical of an acid stimulation
in a sandstone reservoir.
3. The skin factor can be used to calculate the effective wellbore radius, rwa also
referred to the apparent wellbore radius. (See Appendix, page A-13)
4. The skin factor can also be used to correct the injection pressure for the effects of
wellbore damage to get the actual reservoir pressure from the measured pressure.
X The skin factor is calculated from the following equation:
s =1.1513
m
•—log
k-t.
+3.23
where, s = skin factor, dimensionless
Pihr = pressure intercept along the semilog straight line at a shut-in time of 1 hour,
psi
Pwf = measured injection pressure prior to shut-in, psi
[j, = appropriate viscosity at reservoir conditions, cp (See Appendix, page A-14)
m = slope of the semilog straight line, psi/cycle
k = permeability, md
q> = porosity, fraction
Ct = total compressibility, psi"1
rw = wellbore radius, feet
Page 22 of 27
-------
tp = injection time, hours
Note that the term tp/(tp +At), where At=l hr, appears in the log term. This term is
usually assumed to result in a negligible contribution and typically is taken as 1 for large
t. However, for relatively short injection periods, as in the case of a drill stem test (DST),
this term can be significant.
Radius of Investigation
X The radius of investigation, r;, is the distance the pressure transient has moved into a
formation following a rate change in a well.
X There are several equations that exist to calculate the radius of investigation. All the
equations are square root equations based on cylindrical geometry, but each has its own
coefficient that results in slightly different results, (See Oil and Gas Journal, Van Poollen,
1964).
X Use of the appropriate time is necessary to obtain a useful value of r;. For a falloff time
shorter than the injection period, use Agarwal equivalent time function, Ate, at the end of
the falloff as the length of the injection period preceding the shut-in to calculate r;.
X The following two equivalent equations for calculating r; were taken from SPE
Monograph 1, (Equation 11.2) and Well Testing by Lee (Equation 1.47), respectively:
,,= 0.00105-
V 0-V-Ct
Effective Wellbore Radius
X The effective wellbore radius relates the wellbore radius and skin factor to show the
effects of skin on wellbore size and consequently, injectivity.
X The effective wellbore radius is calculated from the following:
wa w
X A negative skin will result in a larger effective wellbore radius and therefore a lower
injection pressure.
Page 23 of 27
-------
Reservoir Injection Pressure Corrected for Skin Effects
X The pressure correction for wellbore skin effects, APskin, is calculated by the following:
APskin =0.868 -m-s
where, m = slope of the semilog straight line, psi/cycle
s = wellbore skin, dimensionless
X The adjusted injection pressure, P«fa is calculated by subtracting the APskin from the
measured injection pressure prior to shut-in, Pwf. This adjusted pressure is the calculated
reservoir pressure prior to shutting in the well, At=0, and is determined by the following:
X From the previous equations, it can be seen that the adjusted bottomhole pressure is
directly dependent on a single point, the last injection pressure recorded prior to shut-in.
Therefore, an accurate recording of this pressure prior to shut-in is important. Anything
that impacts the pressure response, e.g., rate change, near the shut-in of the well should
be avoided.
Determination of the Appropriate Fluid Viscosity
X If the wastestream and formation fluid have similar viscosities, this process is not
necessary.
X This is only needed in cases where the mobility ratios are extreme between the
wastestream, (k/u)w, and formation fluid, (k/u)f. Depending on when the test reaches
radial flow, these cases with extreme mobility differences could cause the derivative
curve to change and level to another value. Eliminating alternative geologic causes, such
as a sealing fault, multiple layers, dual porosity, etc., leads to the interpretation that this
change may represent the boundary of the two fluid banks.
X First assume that the pressure transients were propagating through the formation fluid
during the radial flow portion of the test, and then verify if this assumption is correct.
This is generally a good strategy except for a few facilities with exceptionally long
injection histories, and consequently, large waste plumes. The time for the pressure
transient to exit the waste front is calculated. This time is then identified on both the log-
log and semilog plots. The radial flow period is then compared to this time.
X The radial distance to the waste front can then be estimated volumetrically using the
following equation:
Page 24 of 27
-------
_ 10.13368- V
'waste plume \\ J
F -
where, Vwaste injected = cumulative waste injected into the completed interval, gal
rwaste plume = estimated distance to waste front, ft
h = interval thickness, ft
(p = porosity, fraction
X The time necessary for a pressure transient to exit the waste front can be calculated using
the following equation:
IZO. 16 • //^ • C t • V
wasteinjected
yr-k-h
where, tw= time to exit waste front, hrs
Vwaste injected = cumulative waste injected into the completed interval, gal
h = interval thickness, ft
k = permeability, md
Uw = viscosity of the historic waste plume at reservoir conditions, cp
Ct = total system compressibility, psi"1
X The time should be plotted on both the log-log and semilog plots to see if this time
corresponds to any changes in the derivative curve or semilog pressure plot. If the time
estimated to exit the waste front occurs before the start of radial flow, the assumption that
the pressure transients were propagating through the reservoir fluid during the radial flow
period was correct. Therefore, the viscosity of the reservoir fluid is the appropriate
viscosity to use in analyzing the well test. If not, the viscosity of the historic waste
plume should be used in the calculations. If the mobility ratio is extreme between the
wastestream and formation fluid, adequate information should be included in the report to
verify the appropriate fluid viscosity was utilized in the analysis.
Reservoir Thickness
X The thickness used for determination of the permeability should be justified by the
operator. The net thickness of the defined injection interval is not always appropriate.
X The permeability value is necessary for plume modeling, but the transmissibility value,
kh/u, can be used to calculate the pressure buildup in the reservoir without specifying
values for each parameter value of k, h, and \i.
X Selecting an interval thickness is dependent on several factors such as whether or not the
injection interval is composed of hydraulically isolated units or a single massive unit and
wellbore conditions such as the depth to wellbore fill. When hydraulically isolated sands
Page 25 of 27
-------
are present, it may be helpful to define the amount of injection entering each interval by
conducting a flow profile survey. Temperature logs can also be reviewed to evaluate the
intervals receiving fluid. Cross-sections may provide a quick look at the continuity of the
injection interval around the injection well.
X A copy of a SP/Gamma Ray well log over the injection interval, the depth to any fill, and
the log and interpretation of available flow profile surveys run should be submitted with
the falloff test to verify the reservoir thickness value assumed for the permeability
calculation.
Use of Computer Software
X To analyze falloff tests, operators are encouraged to use well testing software. Most
software has type curve matching capabilities. This feature allows the simulation of the
entire falloff test results to the acquired pressure data. This type of analysis is
particularly useful in the recognition of boundaries, or unusual reservoir characteristics,
such as dual porosity. It should be noted that type curve matching is not considered a
substitute, but is a compliment to the analysis.
X All data should be submitted on a CD-ROM with a label stating the name of the facility,
the well number(s), and the date of the test(s). The label or READ.Me file should
include the names of all the files contained on the CD, along with any necessary
explanations of the information. The parameter units format (hh:mm:ss, hours, etc.)
should be noted for the pressure file for synchronization to the submitted injection rate
information. The file containing the gauge data analyzed in the report should be
identified and consistent with the hard copy data included in the report. If the injection
rate information for any well included in the analysis is greater than 10 entries, it should
also be included electronically.
Common Sense Check
X After analyzing any test, always look at the results to see if they "make sense" based on
the type of formation tested, known geology, previous test results, etc. Operators are
ultimately responsible for conducting an analyzable test and the data submitted to the
regulatory agency.
X If boundary conditions are observed on the test, review cross-sections or structure maps
to confirm if the presence of a boundary is feasible. If so, the boundary should be
considered in the AOR pressure buildup evaluation for the well.
X Anomalous data responses may be observed on the falloff test analysis. These data
anomalies should be evaluated and explained. The analyst should investigate physical
causes in addition to potential reservoir responses. These may include those relating to
the well equipment, such as a leaking valve, or a channel, and those relating to the data
Page 26 of 27
-------
acquisition hardware such as a faulty gauge. An anomalous response can often be traced
to a brief, but significant rate change in either the test well or an offset well.
X Anomalous data trends have also been caused by such things as ambient temperature
changes in surface gauges or a faulty pressure gauge. Explanations for data trends may
be facilitated through an examination of the backup pressure gauge data, or the
temperature data. It is often helpful to qualitatively examine the pressure and/or
temperature channels from both gauges. The pressure data should overlay during the
falloff after being corrected for the difference in gauge depths. On occasion, abrupt
temperature changes can be seen to correspond to trends in the pressure data. Although
the source of the temperature changes may remain unexplainable, the apparent
correlation of the temperature anomaly to the pressure anomaly can be sufficient reason
to question the validity of the test and eliminate it from further analysis.
X The data that is obtained from pressure transient testing should be compared to permit
parameters. Test derived transmissibilities and static pressures can confirm compliance
with non-endangerment (Area Of Review) conditions.
Page 27 of 27
------- |