Using Reliable Technology

for Estimating Oil and Gas Reserves

Rod Sidle Consultant

Industry Advisor Reserves Management
to Energy Navigator

Dallas SPE 1/18/2017

(The Usual) DISCLAIMER
2

Theopinions in this presentation are mine

alone and do not represent legal or
regulatory guidance by me or any
organization. Statements are for
education and contemplation as we all
seek to understand and comply with
reserve requirements.

Dallas SPE 1/18/2017

Presentation Outline
3

Background on Reliable Technology

How to validate/demonstrate
Recent literature on Reliable Technologies
Example: Seismic for water level
Example: SPEE Monograph 3
Comments from the SEC
Q &A

Dallas SPE 1/18/2017

Definition from the
4

Modernization of SEC Rules

SEC Final Rule (Dec 2008) states
(25) Reliable technology. Reliable technology
is a grouping of one or more technologies
(including computational methods) that has
been field tested and has been demonstrated
to provide reasonably certain results with
consistency and repeatability in the
formation being evaluated or in an analogous
formation.

(Note: Emphasis added) Dallas SPE 1/18/2017

SEC Comments
5

SEC Compliance and Disclosure

Interpretation:

An issuer has the burden of establishing and

documenting the technology (or set of
technologies) that provides reliable results,
consistent with the criteria set forth in Rule
4-10(a)(25) of Regulation S-X. This
information should be made available to the
Commission's staff upon request in support
of any reserves estimates that the staff may
be reviewing.

(Note: Emphasis added) Dallas SPE 1/18/2017

(from SPE 139494) 6

Demonstrating Reliable Technology:

What did the SEC intend?
The SEC will expect persuasive empirical
evidence that the application of the
technologies has led to correct conclusions
when applied in the same reservoir, in an
analogous reservoir in the same formation, or
in an analogous reservoir in another
formation. Persuasive empirical evidence
includes drilling results (e.g., demonstrated
economic producibility) for locations similar to
the proposed PUDs or verified success in
locating fluid contacts. In all cases, the
sample size should be sufficiently large to
establish that conclusions are statistically
significant. Dallas SPE 1/18/2017
(Note: Emphasis added)
(from SPE 129689) 7

Scientific Method: Steps

1. Define the question.

2. Research the question and formulate a
hypothesis (define the theoretical
science behind your R.T.).
3. Perform experiments; collect and
analyze the data (test your R.T.).
4. Interpret data; draw conclusions;
document results.
5. If necessary, revise hypothesis and
repeat steps 3 and 4.
Dallas SPE 1/18/2017
(from SPE 129689) 8

Scientific Method: Adapted to

Demonstrating a R.T.

1. Define how the R.T. will contribute to

reserve estimation (e.g., define OWC).
2. Research the science behind this
application; define when results are valid.
3. Test to validate the hypothesis and
demonstrate requirements for R.T. have
been met.
4. Document results including conditions
needed to achieve reliability (i.e., what
are the limits on successful application).
Dallas SPE 1/18/2017
(from SPE 129689) 9

Scientific Method for R.T. Some

further thoughts

Include all test data in documentation;

selective exclusion of data will cause
questions about consistency and
repeatability.
Keep the analysis/documentation updated
with new data as the R.T. is used. Does
this change your conclusions on limits,
application?

Dallas SPE 1/18/2017

Recent Literature on R.T.
10

Technology Target Application

Seismic Lowest Known HC (Proved Area)
Reservoir Characterization (for
STOOIP)
Define level of uncertainty (Ps)
Dynamic Reservoir Estimate Ultimate Recovery (RF)
Simulation Static Reservoir Model (for STOOIP)
Combined Geological Unconventional Proved Area and
Modeling and related PUD volume
Statistical Methods
Other Lowest Known HC (Proved Area)

see SPE 179991 Dallas SPE 1/18/2017

Steps 1 & 2: Question & 11

Research to Develop a Hypothesis

1. Can seismic reliably identify

HC/water contact?

2. The science behind the hypothesis:

a. Where no interfering effects
distort high-quality, 3-D seismic
data, the portion of the seismic
related to fluid content can be
isolated and analyzed.

Dallas SPE 1/18/2017

Steps 1 & 2: Question &
12

Research to Develop a Hypothesis

2. The science behind the hypothesis:

b. Conclusive interpretation of the
fluid contact between a
commercial HC reservoir and an
aquifer requires the additional
condition of distinctly different
seismic amplitudes for pay v.
residual HC saturation or aquifer.

Dallas SPE 1/18/2017

 13

Seismic AVO Zero Offset

Amplitude Variation with Offset
Reflective
Source / Receiver Source / Receiver event
directly
below source
Seismic data
captures only
the p-
waves which
move in the
direction of
the
propagated
wave
Dallas SPE 1/18/2017
Seismic AVO With Offset
14

Seismic reflective
signals (amplitudes)
are converted waves
with both a vertical
(p-wave) and
lateral (s-wave)
component

Shear (s) waves

have different
characteristics than
p-waves allowing
additional
information to be
extracted from the
seismic
Dallas SPE 1/18/2017
Ensure applicability are all
15

conditions right for success?

Quality control of data e.g., well logs

provide good, complete data which ties
to zero-offset seismic
Stratigraphy e.g., no stratigraphic
variations that would compromise the
fluid signal interpretation
Structural factors e.g., good fit to
structure of the apparent contact
and others which develop a checklist
Dallas SPE 1/18/2017
Checklist of qualifying conditions
Quality Control
1a. Log quality
1b. Wavelet correlation
1c. Zero offset tie
1d. AVO tie, updip and downdip
1e. Seismic quality
Stratigraphy
2a. Stratigraphic bias
2b. N/G, well vs HWC
Structure
3a. Fit 100 ft
3b. Seismic HWC
3c. Trap failure analysis
3d. Residual Rim
3e. Loop Interference
3f. Dip < 30o
3g. Illumination
Sensitivity
4a. Sand thickness
4b. Pay/residual separation
Calibration
5a. Error radius
5b. Probability sensitivity
16
Evidence
Complex
Show sonic, density, gamma ray, caliper technologies can
Comparison software output
Show tie be influenced by
Show tie
Seismic traverse
many factors
Amplitude map, well data
Amplitude map, well data Key to reliable
Structure map
interpretation is
Amplitude map, seismic traverse the isolation of
Pressure versus depth plot
Amplitude map target signal
Seismic traverse from the noise
Seismic traverse or dip map
Seismic traverse or illumination map of these other
Isopach map and tuning model
factors
Monte Carlo histograms
Show map
Show map
Dallas SPE 1/18/2017
Step 3: Test the hypothesis
17

Thefollowing two example cases (from

many used) show the successful
application of this method
Oneis a success with reliably identifying a
water contact
One is a success as it exposes that an
apparent seismic water contact should not
be trusted

Dallas SPE 1/18/2017

 18
Successful "positive" case
Bin

-250
-500
-750
-1000
-1250
-1500
-1750
-2000
-2250
-2500
-2750
-3000
-3250
-3500
-3750
-4000
iS
-4250
-4500
-4750
-5000
-5250

Dallas SPE 1/18n017

Successful positive case
19

Confirm zero-offset tie: seismic to

well log
Do seismic inversion (simulated
seismic)
Model ranges of unknown inputs: ,
So, etc.
Develop statistical distribution of
seismic data output from Monte Carlo
on input distributions

Dallas SPE 1/18/2017

 20
Successful "positive" case
Tuning Curves

100

" - i

Q)
75

.s 50

25 -- ------+- ---- --
Minimum thickness tor separating
oil from residual amplitudes .

-0.250 -0.188 -0 .125 -0 .062

stack Amplitude

Dallas SPE 1/18n017

 21
Successful "positive" case

0.1 2 5

u Residual Oil
c:
Q)
:::J
0"
Q)
I....
LL... 0.062
Q)

>co-
-& Note the excellenl separation
between oil and residual ampliludes

0.062 0.125 0.1 8 8

Modeled Stack Amplitudes

Dallas SPE 1/18n017
 22
Successful "negative" case
Bin
0.0
- 2. 4
-4.9
-7.3
-9.8
-12.2
-14.6
- 17. 1
-19 .5
-22.0
- 24. 4
- 26. 8
- 29. 3
-31.7
-34 . 1
-36.6
-39.0
- 41 . 5
-43.9
-46.3
- 48. 8
iS
-51.2
$
- 53. 7
- 56. 1
- 58. 5
- 61. 0
- 63.. 43 ,
-68
- 70.
65. 79
- 73. 2
- 75. 6
- 78. 0
- 80. 5
-82 . 9
- 85. 4
-87 . 8
- 90. 2
- 92. 7
-95 . 1
-97 . 6
-100.0

Dallas SPE 1/18n017

 23
Successful "negative" case
Note overlap
of amplitudes
lr- of the pay and

0.094
J II residual cases

u r1 Brine
c n II
Q)
:::J ll l
C"'
Q)
0.062
Residual .
LL... I Pay
Q)
> 1--l I
L.

0.031
Il-
r- L
1 1

0.000 lr rr l
1.00 2.00 3.00

Modeled Stack Amplitudes

Dallas SPE 1/18n017
Step 4: Document results
24

Include all test data in documentation

Develop process aids to ensure consistency

and repeatability as the method is used

Example: Checklist for qualifying conditions

Keep the analysis/documentation updated

with new data as the R.T. is used.

Dallas SPE 1/18/2017

Summary of Key Steps
25

Hypothesis
Establish scientific basis
Test results must validate
Document (and update)

Dallas SPE 1/18/2017

SPEE Monograph 3 as Reliable
26

Technology
Scientific Method steps are used
Hypothesis: Proved Area can be defined with
applied statistical EURs for new wells
Science: Geologic consistency of productivity and
statistically repeatable EUR variations
Testing: Use existing well EURs to show commercial
productivity from analogous wells follows a narrow,
high confidence distribution
Document: Both analysis and ongoing results to
continuously validate technology use
(Be ready for the SEC to ask.)

Dallas SPE 1/18/2017

Example: SEC Comment letter 27

In one letter, the SEC asks about Proved Area:

In part, your response 10 indicates a

significant portion of your proved
undeveloped locations are 2 or more offsets
removed from a producing well(s). Tell us the
statistics of your drilling history for such
similarly situated locations, including the
success rate by distance/location removed
from production.

(Note: Emphasis added) Dallas SPE 1/18/2017

 28

Thank you

Acknowledgements

Ryder Scott for use of the SEC Seeker

application to search SEC Comment Letters

Dallas SPE 1/18/2017

Example: SEC Comment letter 29

After a description of the technology was provided,

the SEC asked about recovery prediction results:
Please explain to us the revision history due to
performance for proved reserves in your Barnett
Shale play. Include a comparison between the median
values as of year-end 2009 with median initial values
for proved Estimated Ultimate Recovery, realized well
cost and estimated well cost; producing rate vs. time
plot and associated decline curve parameters,
producing rate vs. cumulative production plot and
estimated future production projection
representative of the two median EURs.

Dallas SPE 1/18/2017

(Note: Emphasis added)
SEC Guidance on Comment Letters
The staffs comments are in response to a companys
disclosure and other public information and are based on
the staffs understanding of that companys facts and
circumstancesThese letters set forth staff positions and
do not constitute an official expression of the SECs views.
The letters are limited to the specific facts of the filing in
question and do not apply to other filings.

Dallas SPE 1/18/2017

Deconvolution of SPEE Monograph 3
31

Aswritten, Mono 3 combines geological

modeling and statistical methods into one
reliable technology:
to expand PUD locations beyond one offset

Restate as a two technologies process:

1. Establishing proved area
2. Forecasting recovery from undrilled
locations

Dallas SPE 1/18/2017

1. Geological Modeling
32

Example:
Define Areaof
Consistent Geological
Conditions
Establishreasonable
certainty of economic
producibility

Dallas SPE 1/18/2017

2. Forecast Recovery Methods
33

Example:
Define
sub-area of analogous
production results

Apply analogy (could be a

statistical method as in the
figure) to appropriate
undrilled locations

Dallas SPE 1/18/2017

Other Forecast Recovery Methods 34

SPEE Monograph 3 methodology requires

spatially independent statistical analysis to
assign volumes to undrilled locations.
Thedeconvoluted technology only
requires valid analogy which then allows
spatially oriented methods to be
considered. Example: Attanasi et al (SPE
107659) uses a nearby neighbor weighting
approach for a Michigan Antrim example.

Dallas SPE 1/18/2017