state of michigan michigan gas utilities … · 4 schlumberger technology corporation ... schedule...

STATE OF MICHIGAN

BEFORE THE MICHIGAN PUBLIC SERVICE COMMISSION

* * * * * In the matter of the application of ) MICHIGAN GAS UTILITIES CORPORATION ) for authority to increase retail natural gas rates ) Case No. U-17880 and for other relief. ) )

DIRECT TESTIMONY AND EXHIBITS OF

THOMAS E. DAVIS

FOR

MICHIGAN GAS UTILITIES CORPORATION

June 2015

1

STATE OF MICHIGAN

BEFORE THE MICHIGAN PUBLIC SERVICE COMMISSION

* * * * *

In the matter of the application of ) MICHIGAN GAS UTILITIES CORPORATION ) for authority to increase retail natural gas rates ) Case No. U-17880 and for other relief. ) )

QUALIFICATIONS OF

THOMAS E. DAVIS PART I

Q. Please state your name, business address and position. 1

A. My name is Thomas E. Davis. My business address is 230 CR 2800 N, Fisher, Illinois 2

61843. I am a Supervisory Petroleum Engineer for Integrys Business Support (“IBS”). I 3

am testifying on behalf of Michigan Gas Utilities Corporation (“MGUC”) in support of 4

MGUC’s application in this proceeding for authority to adjust its natural gas rates. 5

6

Q. Briefly describe your educational, professional and utility background. 7

A. I received a Bachelor of Science degree in geology from Purdue University in West 8

Lafayette, Indiana, in 1978; a Master of Science degree in geology from Montana State 9

University in Bozeman, Montana, in 1980; and a Master of Engineering degree in 10

petroleum engineering from Texas A & M University in College Station, Texas, in 2007. I 11

am a Licensed Professional Geologist in the State of Illinois and have worked as a 12

geologist and engineer in the Oil and Gas industry for 26 years. I have worked for IBS, 13

or its predecessor company, for fifteen years, holding various engineering and 14

supervisory positions in gas storage. 15

16

2

THOMAS E. DAVIS DIRECT TESTIMONY

PART II

Q. What is the purpose of your pre-filed direct testimony? 1

A. The purpose of my pre-filed direct testimony is to discuss a study of the viability of 2

MGUC gas storage fields that was performed by the PetroTechnical Services Division of 3

Schlumberger Technology Corporation (“Schlumberger”) and dated January 29, 2015. 4

5

Q. Are you sponsoring any exhibits in this proceeding? 6

A. Yes, I am. I am sponsoring the following schedules to Exhibit A-3: 7

1. Schedule C31, “Inventory Analysis of Underground Gas Storage Reservoirs in 8 Southeastern Michigan.” 9 10

2. Schedule C32, “Table of Working Gas, Recoverable Cushion Gas, and non-11 Recoverable Cushion Gas.” 12

13

Q. Were these schedules prepared by you or under your direction and supervision? 14

A. Yes, they were. 15

16

Q. Please describe Schedule C31. 17

A. Schedule C31 is a study performed by Schlumberger that evaluated the viability of gas 18

storage reservoirs owned and operated by MGUC. The study determined the amount of 19

gas in storage for each reservoir. This gas in storage was broken down into three 20

components: 21

1. Working Gas – The amount of gas that can be cycled each year for use by the 22 customers of MGUC, 23 24

2. Recoverable Cushion Gas – The additional amount of gas that can be recovered 25 when the reservoirs are abandoned, and 26

27 3. Non-Recoverable Cushion Gas – The amount of gas that cannot be recovered 28

when the reservoirs are abandoned. 29 30

3

Q. Why was Schedule C31 created? 1

A. Schedule C31 was created to comply with Paragraph 7.i of the Settlement Agreement in 2

Case No. U-17273. This paragraph states “in MGUC’s next general rate case filing, 3

MGUC will submit a study regarding the viability of the Partello/Anderson gas storage 4

reservoir.” 5

6

Q. Please explain Working Gas, Recoverable Cushion Gas, Non-Recoverable 7

Cushion Gas and Base Gas. 8

A. All natural gas in storage can first be classified into one of two categories. The first 9

category is Working Gas, which is the gas that is produced and replaced (or cycled) 10

annually. The second category is Cushion Gas, which is gas that is not produced and 11

replaced annually. Cushion gas is the amount of gas that must remain in the reservoir to 12

maintain the minimum pressure required to annually cycle the Working Gas at the 13

required rate of withdrawal. Cushion Gas can be further classified as Recoverable 14

Cushion Gas and Non-Recoverable Cushion Gas. Recoverable Cushion Gas is the 15

amount of Cushion Gas that can be economically recovered when the gas storage 16

reservoir is abandoned. Non-recoverable Cushion Gas is the amount of gas that 17

remains in the reservoir when the reservoir is abandoned. The terms “Cushion Gas” and 18

“Base Gas” are inter-changeable. 19

20

Q. What were the results of the Schlumberger study? 21

A. The study Schlumberger performed included data through October 31, 2014. MGUC 22

stores natural gas in seven reservoirs or fields. The conclusion of the Schlumberger 23

study is that the amount of Working Gas has increased in three reservoirs, and 24

decreased in four reservoirs. This results in a net decrease of 785 MMscf in Working 25

4

Gas, an increase of 99 MMscf in Recoverable Cushion Gas, and an increase of 686 1

MMscf in Non-Recoverable Cushion Gas. 2

3

I shall collectively refer to the Cortright, Lee, and Lee 3A reservoirs for which the 4

Schlumberger study notes an increase in Working Gas as “volumetric reservoirs”, and I 5

shall collectively refer to the Partello, Anderson, Mymachod, and MGU 1-24 reservoirs 6

for which the Schlumberger study notes a decrease in Working Gas as “aquifer 7

reservoirs”. 8

9

Q. What are the reasons for the increase in Working Gas, and the decrease in 10

Cushion Gas in the volumetric reservoirs, and what quantity of gas is being re-11

classified? 12

A. The three volumetric reservoirs are named the Lee 3, Lee 3A, and Cortright. Typically, 13

gas withdrawal from these three reservoirs ceases at a static wellhead pressure 14

(“SWHP”) of 600 psig. There is no evidence of significant water movement into these 15

fields, and more gas can be recovered by flowing or compressing additional natural gas 16

from the reservoirs. Operationally, MGUC believes it can operate the three volumetric 17

reservoirs down to a SWHP of 550 psig, which will allow an additional 124 MMscf of 18

natural gas to be cycled annually. This change in operation will increase Working Gas 19

and decrease Recoverable Cushion Gas by a total of 124 MMscf from the Lee 3, Lee 3A 20

and Cortright fields. 21

22

Q. What are the reasons for the decrease in Working Gas, and the increase in 23

Cushion Gas, in the aquifer reservoirs? 24

A. The four aquifer reservoirs are named Partello, Anderson, Mymachod, and MGU 1-24. 25

Previous studies calculated the Working Gas based on these fields being able to operate 26

5

as low as a SWHP of 450 psig. In 2007, 2008, and 2009, only minimum SWHPs of 600 1

psig could be reached. Water was produced when the SWHPs in the aquifer reservoirs 2

fell below 600 psig, requiring the Partello, Anderson, Mymachod, and MGU 1-24 to be 3

shut-in. Since that time, MGUC has operated these reservoirs to a minimum SWHP of 4

700 psig in order to avoid producing water. Water production effectively shuts down 5

withdrawal operations. As a result of ending withdrawal at a higher SWHP, the amount 6

of working gas is reduced. 7

8

Q. What were the reasons for the changes in the amount of Recoverable and Non-9

Recoverable Cushion Gas in the aquifer reservoirs? 10

A In 2008, the aquifer reservoirs were shut-in when water was produced. One well used 11

for observation, the F-1, dropped to a SWHP of 484 psig. The SWHP of 450 psig is 12

believed to be a conservative estimate of the lower pressure limit that these fields can be 13

produced to when abandoned. Prior to the most recent Schlumberger study, 14

Recoverable Cushion gas was calculated with a SWHP range of 450 to 250 psig. Based 15

on the most recent Schlumberger study, Recoverable Cushion gas is calculated with a 16

SWHP range of 700 – 450 psig. The calculation of Non-Recoverable Cushion gas is 17

what remains after Working Gas and Recoverable Cushion gas has been withdrawn. 18

19

Q. What is the amount of Working Gas, Recoverable Cushion Gas, and Non-20

Recoverable Cushion Gas present in each reservoir? 21

A. Exhibit A-3, Schedule C32 shows the amount of Working Gas, Recoverable Cushion 22

Gas, and Non-Recoverable Cushion Gas in each reservoir as of the end of the gas day 23

on October 31, 2014. 24

25

Q. Why is water production now a problem, when it was not in the past? 26

6

A. I cannot say for certain. One possibility is that the decrease in production of oil, gas, and 1

water from the oil and gas wells in the area is allowing the aquifer pressure to begin 2

recovering. 3

4

Q. Is the aquifer still increasing the pressure in the storage reservoir? 5

A. No, it is not. We see no evidence that the aquifer is increasing in pressure. The 6

Schlumberger study shows no change in reservoir performance over the last several 7

years. 8

9

Q. Has the MGUC indigenous storage field increased deliverability? 10

A. Yes, it has. MGUC held a technical conference on August 27, 2013 that included 11

representatives of the Residential Ratepayer Consortium, Attorney General, Staff and 12

MGUC to discuss increasing the withdrawal capability from MGUC storage fields to 13

assist in covering a Peak Day. MGUC determined that if it injected an additional 14

150,000 Mcf into the storage fields, the Company would be able to increase the Peak 15

Day withdrawal from 20,000 Mcf to 30,000 Mcf for between five to ten days during the 16

winter of 2013/2014. 17

18

Q. Did the MGUC Partello, Anderson, Mymachod, and MGU 1-24 gas storage 19

reservoirs operate as needed during the recent Polar Vortex events? 20

A. Yes, they did. Specifically, MGUC was able to safely withdraw over 20,000 Mcf on eight 21

days during 18 January 2014 and nine days during February 2014. By maintaining 22

natural gas in these four reservoirs, a higher daily rate of gas production or “peaking” 23

can be obtained than if all of the natural gas were stored in only the Cortright, Lee, and 24

Lee 3A reservoirs. 25

26

7

Q What was significant about the winter 2013/2014? 1

A Many have called the winter of 2013/2014 the Polar Vortex. MGUC experienced prices 2

in excess of $20 per Dth and even exceeding $34 on February 6, 2014; however, MGUC 3

was able to withdraw gas from storage at the Weighted Average Cost of Gas 4

(“WACOG”) storage price that was significantly less than what it would have paid in the 5

day market. 6

7

Q. Has the total MGUC indigenous storage capacity of 3.9 Bcf changed? 8

A. Yes, it has been reduced to 2.92 Bcf; however, MGUC indigenous storage is still able to 9

support forecasted sales in the Coldwater area for a portion of April 2015 and October 10

2015, and 100% of forecasted sales for November 2015, December 2015, January 11

2016, February 2016 and March 2016. 12

13

Q. How much does MGUC plan to cycle MGUC indigenous storage in 2015/2016? 14

A. MGUC plans to cycle 2 Bcf from MGUC indigenous storage. 15

16

Q. Do the MGUC Partello, Anderson, Mymachod, and MGU 1-24 gas storage 17

reservoirs continue to be both useful and viable? 18

A. Yes, they do. While certain reclassifications of Working Gas and Cushion Gas are 19

recommended, and while their operating envelope may be smaller than in the past, 20

these gas storage reservoirs continue to be both useful and viable. 21

22

Q. In light of the approvals granted in Case No. U-17682 and the impending merger of 23

Wisconsin Energy Corporation (“WEC”) and Integrys to be completed in 2015, do 24

you anticipate that the substance of your testimony or sponsored exhibits will 25

need to be modified to reflect the change in MGUC’s control from Integrys to 26

8

WEC? 1

A. No, I do not at this time. With the merger of Integrys into WEC and WEC taking control 2

of MGUC, with the exceptions of IBS changing its name to WEC Business Support LLC., 3

(“WBS”), we expect business as usual, with no significant immediate changes. 4

5

Q. Does this conclude your pre-filed direct testimony? 6

A. Yes, it does. 7

Inventory Analysis of Underground Gas Storage Reservoirs in Southeastern Michigan

Prepared For

Michigan Gas Utilities Company

Prepared By

PetroTechnical Services Division of Schlumberger Technology Corporation

Canonsburg, Pennsylvania

January 2015

Case No.: U-17880 Exhibit No.: A-3

Schedule No.: C31 Page: 1 of 68

Witness: Thomas E. Davis

Disclaimer Any interpretation, research, analysis, data, results, estimates, or recommendation furnished with the services or otherwise communicated by Schlumberger to customer at any time in connection with the services are opinions based on inferences from measurements, empirical relationships and/or assumptions, which inferences, empirical relationships and/or assumptions are not infallible, and with respect to which professionals in the industry may differ. Accordingly, Schlumberger cannot and does not warrant the accuracy, correctness or completeness of any such interpretation, research, analysis, data, results, estimates or recommendation. Customer acknowledges that it is accepting the services "as is," that Schlumberger makes no representation or warranty, express or implied, of any kind or description in respect thereto. Specifically, customer acknowledges that Schlumberger does not warrant that any interpretation, research, analysis, data, results, estimates, or recommendation is fit for a particular purpose, including but not limited to compliance with any government request or regulatory requirement. Customer further acknowledges that such services are delivered with the explicit understanding and agreement that any action taken based on the services received shall be at its own risk and responsibility and no claim shall be made against Schlumberger as a consequence thereof. Copyrights Copyright © 2011, Schlumberger. All rights reserved. Trademarks All companies or product names mentioned in this document are used for identification purposes only and may be trademarks of their respective owners.

Submitted for and on behalf of: Registered name: Schlumberger Technology Corporation Registered address: 5599 San Felipe Street, Houston, Texas 77056 Registered in Texas, USA: Texas Charter number: 17985000 Document Tracking and Revision Summary

Version: 1.0 Walter K. Sawyer (wks)

Creation Date: 01-Jan-15 Wks

Modification Date: 29-Jan-15 Wks




4600 J Barry Court Suite 200 Canonsburg, Pennsylvania 15317 USA Tel: +1-724-416-9700 Fax: +1-724-416-9705

29 January 2015

Mr. Ronald L. Herr Michigan Gas Utilities Corp. 899 S. Telegraph Road Monroe, Michigan 48161 Dear Mr. Herr:

RE: Inventory Analysis of Gas Storage Reservoirs

We thank you for the opportunity to evaluate and update our previous study of the Lee 3/3A, Cortright, Partello-Anderson gas storage fields. We have updated the monthly volumes and average pressure for each reservoir in each field from April 2011 through October 2014. We have also updated our inventory analysis of each reservoir. This report provides you with our latest evaluation of the performance of each reservoir.

Again, we appreciate the opportunity to provide Michigan Gas Utilities with our detailed technical analysis of these reservoirs.

Sincerely,

Walter K. Sawyer Principal Consultant




PetroTechnical Services Division of Schlumberger Technology Corporation 29 January 2015 page 2

1. Executive Summary

This report summarizes the inventory analyses of seven underground gas storage reservoirs located in southeastern Michigan. The work was conducted by Schlumberger PetroTechnical Services (PTS) for Michigan Gas Utilities Corporation (MGU). The seven reservoirs include: Lee 3, Lee 3A, Cortright, Partello, Anderson, Mymachod, and MGU 1-24. The objectives of this study were to (1) estimate the current base gas, working gas, and lost or non-effective gas for each reservoir and (2) to recommend any possible operational changes that could facilitate future inventory tracking and field management. Schlumberger PTS evaluated these reservoirs in 20071 and again in 20122. This is an update study to our 2012 study.

The data provided to us by MGU for this update study consisted of (1) daily injection and withdrawal for each well in each reservoir, (2) deadweight surface pressures for each well in each reservoir, and (3) limited bottomhole pressure data based on echometer measurements.

We calculated the total monthly injection and withdrawal (for each reservoir) and updated our monthly injection/withdrawal vs. time plots from our previous study. We also converted end of month surface pressures to bottomhole conditions and updated our previous plots of P/z vs. time plots for each reservoir.

For our inventory analysis, we updated the working-gas-inventory (WGI) from our previous study2 using calculated semi-annual volumes from April 2011 to October 2014. We averaged the individual well deadweight pressures for each well and then converted to an absolute bottomhole pressure for the inventory analysis. We compared these with the bottomhole pressures determined by MGU from fluid levels using wellhead pressures and echometer data. The comparison was very good, which validated our calculated BHP’s from the wellhead deadweight measurements.

The field discovery pressure and primary depletion data (P/z vs. cumulative production) for the Lee 3 and Lee 3A reservoirs were used to establish the original total-inventory-per-pound (TIPP) value for these reservoirs. This value is directly proportional to the initial reservoir gas pore volume and is necessary to quantitatively determine gas losses and/or changes in reservoir volume over time. We did not have primary depletion data for the other five reservoirs. Therefore, other methods were devised to estimate the original TIPP for these reservoirs.

Table 1-1 gives the working gas (WG) as of 31 October 2014, the total estimated base gas (BG) and the estimated lost or non-effective gas (NEG) for each field based on inventory models that we developed using data from the last eight years of storage operations. A base wellhead pressure of 600 psig was used to estimate the base gas for the Lee 3, Lee 3A and Cortright reservoirs. A base wellhead pressure of 450 psig was used for the other four reservoirs. Table 1-1 also gives our estimates of recoverable and non-recoverable base gas for each reservoir, assuming an abandonment pressure of 50 psig.





Table 1-1: Summary of Inventory Models – Scenario 1

Table 1-2 shows results with different wellhead pressures for top gas, base gas and abandonment. In Table 1-2, the top pressure has been increased slightly for all reservoirs and the base wellhead pressure has been lowered from 600 psig to 550 psig for the Lee 3, Lee 3A and Cortright reservoirs. The base wellhead pressure has been increased from 450 psig to 700 psig for the remaining four reservoirs. Also the abandonment pressure has been increased from 250 psig to 450 psig for the last four reservoirs. These revised values are believed to be prudent for future operations in these reservoirs.

ReservoirAverage SIWHP

10/31/2014

WorkingGas

10/31/2014

BaseWHP

Base Gas(BG)

Abandon-ment

Pressure

RecoverableBase Gas

Non-Recoverable

Base Gas

Non-Effective

Gas(NEG)

(psig) (MMscf) (psig) (MMscf) (psig) (MMscf) (MMscf) (MMscf)

Lee 3 1237 712 600 823 50 542 281 224

Lee 3A 1228 393 600 504 50 303 200 168

Cortright 1240 566 600 551 50 425 127 82

Partello 873 791 450 773 250 347 426 0

Anderson 843 479 450 562 250 227 335 56

Mymachod 854 140 450 164 250 65 100 20

MGU 1-24 895 114 450 160 250 48 112 54

TOTAL 3196 3536 1956 1580 605





Table 1-2: Summary of Inventory Models – Scenario 2

2. Introduction

Michigan Gas Utilities Corp. (MGU) owns and operates three gas storage fields in Southeastern Michigan: Lee 3/3A, Cortright, and Partello-Anderson. For the purposes of our analyses, the Lee 3/3A field was subdivided into the Lee 3 and Lee 3A reservoirs, and the Partello-Anderson Field was subdivided into the Partello, Anderson, Mymachod, and MGU 1-24 reservoirs.

2.1 Field History - Lee 3/3A

Initial production from the Lee 3/3A reservoirs began in the mid-1970’s and continued through the early 1990’s, when storage operations began. The estimated initial gas in place (IGIP) from primary production was 1.68 Bscf and 0.91 Bscf for the Lee 3 and Lee 3A reservoirs, respectively.

2.2 Field History – Cortright

The Cortright field was discovered in 1972 and initial production began shortly thereafter. Primary production continued until conversion of the field to storage in the mid-1970’s. Cumulative production through April 1974 was 0.706 Bscf, and the Initial-Gas-In-Place (IGIP) was estimated to be about 1.1 Bscf. Analysis of this reservoir is complicated by the fact that detailed primary production and shut-in pressure data over time are not available.

ReservoirTop

SIWHPWorking

GasBaseWHP

Base Gas(BG)

Abandon-ment

Pressure

RecoverableBase Gas

Non-Recoverable

Base Gas

Non-Effective

Gas(NEG)

(psig) (MMscf) (psig) (MMscf) (psig) (MMscf) (MMscf) (MMscf)

Lee 3 1380 931 550 770 50 489 281 224

Lee 3A 1380 522 550 474 50 274 200 168

Cortright 1350 709 550 510 50 383 126 82

Partello 900 384 700 1232 450 459 773 0

Anderson 900 251 700 862 450 300 562 56

Mymachod 900 72 700 250 450 86 164 20

MGU 1-24 900 53 700 223 450 63 160 54

TOTAL 2922 4320 2054 2266 604





2.3 Field History – Partello-Anderson Complex

Historical documents suggest that the Partello-Anderson complex consists of four reef structures which may be in pressure communication with each other. These individual structures (or reservoirs) are known as the Partello, Anderson, Mymachod, and MGU 1-24. A three-dimensional representation of these structures was presented in our first study of these reservoirs1.

There are three issues that complicate the analysis of these reservoirs. First, the primary production and pressure data are not available to establish the IGIP and initial P/z. Second, there is strong evidence of pressure communication between the reservoirs. Third, water influx has been a major problem, making it difficult or impossible to maintain the historical pressure drawdown during withdrawal cycles.

The Partello reservoir was discovered in 1959. For our original (2007) study, MGU provided the estimated total primary production from four wells (1.695 Bscf) that produced from 1964 to 1970. However, no pressure or production data over time was available. The field was converted to storage in 1971 and the initial injection began that same year.

The Anderson reservoir was discovered in July, 1971 and produced a reported 1.1 Bscf through 1973. Conversion to storage occurred in 1974 and it was soon found that this reservoir was in pressure communication with the Partello reservoir. Also, it was believed that a strong water drive was present.

The Mymachod reef was discovered in 1974 and lies between the Partello and Anderson reservoirs. Although discovered in 1974, primary production did not begin until 1980. Primary production occurred during 1980-1984 and also for a few months in 1987. Prior to the start of primary production, shut in wellhead pressures indicated that the Mymachod reservoir was in active pressure communication with the Partello and Anderson reservoirs.

Very little historical information can be found for the MGU 1-24 reservoir. Apparently it was drilled in 1981; however, no primary production or inventory/pressure data prior to 1992 is available for analysis.

3. Conclusions

3.1 Lee 3/3A Reservoirs

Primary production data analysis (available from our 2007 Report1) greatly facilitated the analysis of these reservoirs. Based on our review of the inventory data provided by MGU for the Lee3/3A reservoirs, we offer the following conclusions.

For the Lee 3 reservoir, we developed an Inventory Model [I = 0.224 + 0.0008333 (P/z)] which fits the last eight years of data from storage operations. Using a base wellhead pressure of 600 psig and the actual fall 2014 wellhead pressure at the end of injection (1237 psig) gives P/z values of 719 and 1573, respectively. For these input values, our inventory model gives the following: Top Gas = 1.535 Bscf, BG = 0.823, WG = 0.712 Bscf, and NEG = 0.224 Bscf (Table 1-1).





We believe there was a gradual gas loss in the Lee 3 reservoir from 1992 to about 2007. However, since 2007 the volume appears to have stabilized. For the Lee 3A reservoir, we developed an Inventory Model [I = 0.1682 + 0.0004691 (P/z)] which fits the last eight years of storage operations.

Using a base wellhead pressure of 600 psig and the actual fall 2014 wellhead pressure at the end of injection (1228 psig), our inventory model gives the following current values: Top Gas = 0.897 Bscf, BG = 0.504, WG = 0.393 Bscf, and NEG = 0.168 Bscf (Table 1-1).

We believe there was a gradual gas loss in the Lee 3A reservoir from 1992 to about 2005. However, since 2005 the volume appears to have stabilized.

3.2 Cortright Reservoir

For the Cortright reservoir, we developed an Inventory Model [I = 0.0821 + 0.0006347 (P/z)] which fits the last eight years of storage operations. Using a base wellhead pressure of 600 psig and the actual fall 2014 wellhead pressure at the end of injection (1240 psig) gives P/z values of 715 and 1554, respectively. Using these values in our inventory model gives the following: Top Gas = 1.117 Bscf, Base Gas = 0.551, Working Gas = 0.566 Bscf, and Non-Effective-Gas = 0.082 Bscf (Table 1-1).

The volume of the Cortright reservoir appears to have increased slightly from 1987 to 1999 but has been relatively stable since 1999.

3.3 Partello Reservoir

For the Partello reservoir, we developed an Inventory Model [I = 0.001454 (P/z)] which fits the last eight years of storage operations. Using a base wellhead pressure of 450 psig and the actual fall 2014 wellhead pressure at the end of injection (873 psig) gives P/z values of 531 and 1631, respectively. Using these values in our inventory model gives the following: Top Gas = 1.564 Bscf, Base Gas = 0.773, Working Gas = 0.791 Bscf, and Non-Effective-Gas = 0.0 Bscf (Table 1-1).

The Partello reservoir appears to have had a shrinking reservoir volume (apparently due to water influx) from 1992 to about 2008. However, since 2008 the volume seems to have stabilized.

3.4 Anderson Reservoir

For the Anderson reservoir, we developed an Inventory Model [I = 0.05636 + 0.0009555 (P/z)] which fits the last eight years of storage operations. Using a base wellhead pressure of 450 psig and the actual fall 2014 wellhead pressure at the end of injection (843 psig) gives P/z values of 529 and 1030, respectively. Using these values in our inventory model gives the following: Top Gas = 1.041 Bscf, Base Gas = 0.562, Working Gas = 0.479 Bscf, and Non-Effective-Gas = 0.056 Bscf (Table 1-1).





The Anderson reservoir appears to have had a shrinking reservoir volume (apparently due to water influx) from 1993 until about 2009. However, since 2009 the volume seems to have stabilized.

3.5 Mymachod Reservoir

For the Mymachod reservoir, we developed an Inventory Model [I = 0.0203 + 0.0002708 (P/z)] which fits the last eight years of storage operations.

Using a base wellhead pressure of 450 psig and the actual fall 2014 wellhead pressure at the end of injection (854 psig) gives P/z values of 531 and, respectively. Using these values in our inventory model gives the following: Top Gas = 0.305 Bscf, Base Gas = 0.164, Working Gas = 0.140 Bscf, and Non-Effective-Gas = 0.020 Bscf (Table 1-1).

The Mymachod reservoir appears to have had a shrinking reservoir volume (apparently due to water influx). However, our calculated TIPP has been relatively stable since about 2000.

3.6 MGU 1-24 Reservoir

For the MGU reservoir, we developed an Inventory Model [I = 0.0537 + 0.0001994 (P/z)] fits the last eight years of storage operations. Using a base wellhead pressure of 450 psig and the actual fall 2014 wellhead pressure at the end of injection (895 psig) gives P/z values of 531 and 1105, respectively. Using these values in our inventory model gives the following: Top Gas = 0.274 Bscf, Base Gas = 0.160, Working Gas = 0.114 Bscf, and Non-Effective-Gas = 0.050 Bscf (Table 1-1).

The MGU reservoir appears to have had a slightly increasing reservoir volume from 1992 until 2004. However, since 2004 the TIPP has been relatively stable.

4. Recommendations

If possible, recover the primary production and pressure data from the Cortright reservoir and the four reservoirs in the Partello-Anderson complex. If the primary recovery material balance curve (P/z vs. cumulative production) could be constructed, it would allow the determination of the original TIPP for these fields. This would provide a more accurate analysis of the actual lost (or non-effective) gas in these reservoirs.

5. Discussion

In this section, we discuss in detail our development of the Inventory Model for each reservoir. We also show a Material Balance Model using only withdrawal data. However, we believe the Inventory Models are the most rigorous and therefore used those models in our final analysis results presented above.

We first determined the average (deadweight) wellhead pressure at the end of each injection and withdrawal period for each reservoir. We converted these values to bottomhole pressure values, assuming a gas-filled wellbore. We then compared the limited BHP values provided by MGU that were obtained from Echometer measurements. We





found that the comparison was excellent (see Appendix A) which validated the use of our bottomhole pressures from the deadweight measurements.

Using semi-annual injection/withdrawal volumes and bottomhole pressures calculated from deadweight measurements, we developed an inventory model for the last eight years of storage operations for each reservoir. Our inventory models were used to estimate current base gas, working gas, and non-effective gas for each reservoir.

For clarity, we use the following definitions throughout this report:

Cushion Gas: The volume of gas, including native gas that must remain in the reservoir to maintain an adequate reservoir pressure and deliverability rate throughout the withdrawal season (also called Base Gas).

Working Gas (WG): The volume of gas in the reservoir above the design level of cushion gas (or base gas). It may or may not be produced during any particular withdrawal season.

Working Gas Inventory (WGI): The summation of all injections and withdrawals since storage operations began.

Inventory Adjustment (IA): The (constant) value added to the WGI to account for native and/or injected base gas.

Adjusted or Book Inventory (I): The estimated total gas-in-place in the reservoir (WGI + IA).

Base Gas (BG): Base Gas in this study is defined to be the indicated volume of gas remaining in the reservoir at a specified P/z determined from a graph of P/z vs. adjusted inventory. Recoverable base gas is then defined as the difference in adjusted inventory at the base P/z and the abandonment P/z.

Non-Effective or Lost Gas (NEG): The indicated volume of gas on the adjusted inventory curve at P/z = 0.

For the Lee 3 and Lee 3A reservoirs, we had the benefit of primary production data and thus could establish the correct initial TIPP (Mscf/Psia/z). This provided a basis to determine the inventory adjustment (IA) to add to MGU’s working gas inventory to obtain a reasonable value for the current adjusted inventory or gas-in-place (GIP).

For the remaining five reservoirs, we did not have primary production. For these reservoirs, we used the change in P/z for each injection and withdrawal cycle to calculate a gas-filled reservoir pore volume and a GIP. We then averaged these values over time and calculated an estimated P/z and then a total-inventory-per-pound value, which we called TIPPmb. We then determined the Inventory Adjustment (IA) to add to the WGI for these reservoirs. The IA value we used was the value required to give the total-inventory-per-pound value from material balance; i.e. the TIPPmb, at the end of the first injection. Using the adjusted inventory (WGI + IA) we prepared the standard inventory plots (P/z vs. Adjusted or Book Inventory).

5.1 Diagnostic Plots for Underground Gas Storage

The commonly used “inventory” or “pressure content” plot consists of the stabilized (P/z) on the y-axis and book inventory (Gb) on the x-axis. For a constant volume reservoir with no gas loss, this plot is essentially a material





balance plot which theoretically must extrapolate to zero book inventory at zero pressure (assuming the book inventory includes not only injections and withdrawals, but also the native gas that was present in the reservoir at the time that storage operations began). In reality, there are often gas losses which may be due to pipeline system leaks, metering errors, or actual losses in the reservoir. Any type of gas loss will shift the entire inventory curve to the right over time, falsely indicating more gas at any measured P/z. The intercept of this shifted inventory curve with the x-axis is commonly called lost gas or non-effective gas (NEG). It is possible to calculate NEG for each withdrawal and injection cycle. This provides a diagnostic plot which gives an indication of the change in lost or non-effective gas over time. NEG is mathematically defined as:

NEG = Gbf - (P/z)f * (Gbf – Gbs)/[(P/z)f – (P/z)s]

where Gb is book inventory and subscripts “f” and “s” represent fall and spring values, respectively.

It should be noted that NEG is simply the extrapolation of the line connecting the two points for each withdrawal cycle to P/z = 0. As such, this value depends on the value of the fall book inventory, Gbf, which, in turn consists of two components, the base gas and the working gas inventory (WGI). Assuming no measurement or accounting errors, WGI is a “ground truth” value; however the base gas is often an estimated value. Hence, any change in “assumed” base gas will affect the calculated NEG.

Two other commonly used diagnostic plots are the “total-inventory-per- pound” or TIPP and the “incremental-inventory-per-pound” or IIPP which are defined by

TIPP = Gb / (P/z)

IIPP = (Gbf - Gbs) / [(P/z)f - (P/z)s]

Graphs of the TIPP and IIPP over time may be used to infer reservoir performance as follows:

Constant volume: TIPP & IIPP constant and approximately equal (they are both the inverse slope of the P/z versus book inventory plot, which is constant for a fixed volume reservoir.)

Gas loss: TIPP increase with time; IIPP constant with time (inventory curve shifts right)

Increasing volume: TIPP & IIPP increase with time and approximately equal (inventory curve slope decreases)

Decreasing volume: TIPP & IIPP decrease with time and approximately equal (inventory curve slope increases)

Gas loss and increasing volume: TIPP & IIPP increase with time and TIPP > IIPP (inventory curve shifts to right and slope decreases)

Example inventory, diagnostic and material balance plots for the last four cases are presented in Appendix B, based on simulated data.





5.2 Lee 3 Reservoir Analysis

Primary production data was available for the Lee 3 reservoir during our 2007 study1. Figure 5-1 shows the material balance plot of the primary recovery from the mid 1970’s until storage operations began about 1992.

Figure 5 1: Material balance graph of primary production for Lee 3 Reservoir.

Figure 5-2 shows the monthly injection and withdrawal for the Lee 3 reservoir during storage operations from 1992 to 2014, and Figure 5-3 shows the corresponding semi-annual shut-in pressure history. (Note: in these and subsequent figures, the data through 3/31/2011 is from our 2012 report and the new data (April 2011 through October 2014) provided for this update study is shown in a different color.)

y = -1223.6x + 2058.5R² = 1

y = -918.84x + 1650.4R² = 0.9915

0

500

1,000

1,500

2,000

2,500

0.0 0.5 1.0 1.5 2.0

BH

P/Z

-p

sia

Cum Gas Production - BCF

Lee-3





Figure 5-2: Historical storage operations – Lee 3 Reservoir.

Figure 5-3: Bottomhole pressure/z during storage operations – Lee 3 Reservoir.

-400,000

-300,000

-200,000

-100,000

0

100,000

200,000

300,000

400,000

Mar-92 Mar-94 Mar-96 Mar-98 Mar-00 Mar-02 Mar-04 Mar-06 Mar-08 Mar-10 Mar-12 Mar-14

Inje

ctio

n o

r W

ith

dra

wal

, Msc

f

Date

Lee 3 Reservoir - Monthly Injection and Withdrawal

0

200

400

600

800

1000

1200

1400

1600

1800

2000

3/31/92 3/31/94 3/31/96 3/31/98 3/31/00 3/31/02 3/31/04 3/31/06 3/31/08 3/31/10 3/31/12 3/31/14

BH

P/z,

psi

a

Date

Lee 3 Reservoir - BHP/z vs. Time





Figure 5-3a shows the seasonal injection and withdrawal from March 2011 to October 2014. Note that there was very little injection and withdrawal during the 2011-2012 storage season.

Figure 5-3a: Seasonal storage operations (2011 to 2014) – Lee 3 Reservoir.

To begin our inventory analysis for the Lee 3 reservoir, we used the linear regression of the material balance data from the primary production. Specifically, we used the red points in Figure 5-1 which give an initial P/z of 2,058 psia and an IGIP of 1.682 Bscf. This results in a TIPP of 817.26 Mscf/psia which establishes the initial inventory line shown as the dark red line in Figure 5-4. The new data provided for this update study (2011 to 2014) is plotted in orange.

-1.000

-0.800

-0.600

-0.400

-0.200

0.000

0.200

0.400

0.600

0.800

Fall 2011 Spring 2012 Fall 2012 Spring 2013 Fall 2013 Spring 2014 Fall 2014

Inje

ctio

n (

+) o

r W

ith

dra

wal

(-)

, B

scf

Lee 3

Withdrawal Injection





Figure 5-4: P/z vs. adjusted inventory for Lee 3 Reservoir.

In Figure 5-4, each storage cycle basically follows the same slope as the original inventory curve. There is, however, a gradual shift to the right, indicating possible lost or non-effective gas. Figure 5-5 shows only the data from the past eight years and indicates a NEG value of about 0.22 Bscf.

0

500

1,000

1,500

2,000

2,500

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

BH

P/z,

psi

a

Adjusted Inventory, Bscf

Lee 3 Reservoir - BHP/z vs. Adjusted Inventory

TIPP = 1,682,300 / 2,058.5= 817.26 Mscf/psi

(from primary production)

Starting time for analysis is 11/30/1992 with measured P/z of 1747.8.Adding 0.498 Bscf to MGU's Working-Gas-Inventory on this date gives a GIP of 1.43 Bscf and the calculated TIPP from primary production.

spring 2010 (.888, 790.5)

spring 2008 (.738, 615.5)





Figure 5-5: P/z vs. adjusted inventory for Lee 3 Reservoir (last eight years of storage).

Figure 5-5 includes our Inventory Model [I = 0.224 + 0.0008333 (P/z)] which may be used to calculate WG and BG for any desired P/z values. This model was used to calculate the working gas and base gas given in Table 1-1.

Figure 5.5a shows the same data as in Figure 5.5, but shows the data for the last three years in orange.

y = 1200.1x - 269.2

0

500

1,000

1,500

2,000

2,500

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

BH

P/z,

psi

a


Lee 3 Reservoir - BHP/z vs. Adjusted Inventory - 2006 to 2014

Primary Production

I = 0.224 + 0.0008333 (P/z)

WGBGNEG

Inventory Adjustment = 498,000 Mscf





Figure 5-5a: P/z vs. adjusted inventory for Lee 3 Reservoir (last eight years of storage) showing recent data in orange.

Figure 5-6 is the standard plot showing the calculated NEG vs. time for the Lee 3 Reservoir and shows that the change is gradual from 1993 to about 2008. However, from 2006 to 2014, the calculated NEG is fairly constant, indicating essentially no lost gas during this period.

0

500

1,000

1,500

2,000

2,500

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

BH

P/z,

psi

a


Lee 3 Reservoir - BHP/z vs. Adjusted Inventory - 2006 to 2014

Primary Production






Figure 5-6: Lost or non-effective-gas (NEG) over time for Lee 3 Reservoir.

Figure 5-7 shows both the calculated TIPP and IIPP and further confirms a gradual loss from 1993 to 2008, as indicated by a relatively constant IIPP, but a gradually increasing TIPP with time. However, the fall TIPP values (which are generally considered more reliable) from 2008 to 2014 are essentially constant, indicating a stabilized storage operation.

-0.150

-0.100

-0.050

0.000

0.050

0.100

0.150

0.200

0.250

0.300

3/31/92 3/31/94 3/31/96 3/31/98 3/31/00 3/31/02 3/31/04 3/31/06 3/31/08 3/31/10 3/31/12 3/31/14

No

n-E

ffec

tive

Gas

, Bsc

f

Date

Lee 3 Reservoir - NEG vs. Time

Lost or non-effective gas has stabilized





Figure 5-7: Total and incremental inventory per pound – Lee 3 Reservoir.

Finally, using all historical withdrawal cycles (including the updated data from March 2011 to March 2014), we created a material balance curve for the Lee 3 Reservoir which is shown in Figure 5-8. The material balance curve from primary production is also shown for comparison.

Figure 5-8 gives a material balance regression equation for the Lee 3 Reservoir [ Gp = 1.412 – 0.0008811 (P/z) ] which could be used to estimate working gas and base gas, in lieu of the Inventory Model previously given. For example, a P/z of 700 psia gives a WG of 0.800 Bscf.

0

200

400

600

800

1000

1200

1400

1600

0

200

400

600

800

1000

1200

1400

1600

3/31/92 3/31/94 3/31/96 3/31/98 3/31/00 3/31/02 3/31/04 3/31/06 3/31/08 3/31/10 3/31/12 3/31/14

IIP

P, M

scf/

psi

TIP

P, M

scf/

psi

Date

Lee 3 Reservoir - TIPP and IIPP

TIPP IIPP TIPP Update IIPP Update





Figure 5-8: Material balance curves for Lee 3 Reservoir.

Figure 5-8 also identifies the data (green squares) for the three withdrawals since our 2012 report. These new data did not significantly change the regression curve for Gp.

5.3 Lee 3A Reservoir Analysis

Primary production data was available for the Lee 3A reservoir during our 2007 study1. Figure 5-9 shows the material balance plot of the primary recovery from the mid 1970’s until storage operations began about 1992.

y = -1134.9x + 1602.1R² = 0.9299

0

500

1000

1500

2000

2500

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

P/z,

psi

a

Annual withdrawal, Bscf

Lee 3 Reservoir - P/z vs. Withdrawal - 1993 to 2014

Working Gas

Gp = 1.412 - 0.0008811 (P/z)

Material balance curve fromprimary production





Figure 5-9: Material balance graph for Lee 3A Reservoir.

Figure 5-10 shows the monthly injection and withdrawal for the Lee 3A Reservoir during storage operations. Data from 1992 to 2011 are plotted in blue and the updated data from 2012 to 2014 are shown in red. Note that there was no injection in 2011 and very little withdrawal the winter of 2011-2012, compared to withdrawal volumes in previous winters. Figure 5-10a shows only the seasonal injection and withdrawal for 2011 to 2014.





Figure 5-10: Historical storage operations – Lee 3A Reservoir.

Figure 5-10a: Seasonal storage operations (2011 to 2014) – Lee 3A Reservoir.

-200,000

-150,000

-100,000

-50,000

0

50,000

100,000

150,000

200,000


Inje

ctio

n o

r W

ith

dra

wal

, Msc

f

Date

Lee 3A Reservoir - Monthly Injection and Withdrawal

-0.500

-0.400

-0.300

-0.200

-0.100

0.000

0.100

0.200

0.300

0.400

0.500


Inje

ctio

n (

+) o

r W

ith

dra

wal

(-)

, B

scf

Lee 3A

Withdrawal Injection





Figure 5-11 shows the corresponding semi-annual shut-in pressure history for the two time periods.

Figure 5-11: Pressure/z during storage operations – Lee 3A Reservoir.

To begin our inventory analysis for the Lee 3A Reservoir, we used the linear regression of the material balance data from the primary production. Specifically, we used end points in Figure 5-9 which gives an initial P/z of 2,000 psia and an IGIP of 0.910 Bscf. This gives a TIPP of 455.00 Mscf/psia which establishes the initial inventory line for the reservoir as shown by the orange line in Figure 5-12.





Figure 5-12: P/z vs. adjusted inventory for Lee 3A Reservoir.

We adjusted MGU’s Working Gas Inventory (WGI) by the amount required to give the same TIPP as obtained from primary production. An Inventory Adjustment of 0.255 Bscf was used for the Lee 3A Reservoir. When added to MGU’s WGI, this gives an adjusted inventory of 0.798 Bscf on 30 November 1992. This adjusted inventory, when divided by the measured P/z value on 30 November 1992, gives the same TIPP value as obtained from the primary production.

In Figure 5-12, we observe that each cycle basically follows the same slope as the original inventory curve. There is a gradual shift to the right, indicating possible lost or non-effective gas from 1992 to 2014. Figure 5-13 shows only the data from the past eight years and indicates a NEG value of about 0.17 Bscf.

0

500

1,000

1,500

2,000

2,500

0.0 0.2 0.4 0.6 0.8 1.0 1.2

BH

P/z,

psi

a


Lee 3A Reservoir - BHP/z vs. Adjusted Inventory

Inventory 1992-2011

Primary production

Inventory 2012-2014

TIPP = 910,000 / 2,000= 455 Mscf/psi

(from primary production)

Starting time for analysis is 11/30/1992 with measured P/z of 1754.3 . Adding 0.255 Bscf to MGU's Working-Gas-Inventory on this date gives a GIP of 0.798 Bscf and the calculated TIPP from primary production.

spring 2010 (0.539, 789)

spring 2014 (0.506, 699)





Figure 5-13: P/z vs. adjusted inventory for Lee 3A Reservoir (last eight years of storage).

Figure 5.13a shows the same data as in Figure 5.13, but shows the data for the last three years in red.

y = 2131.6x - 358.63

0

500

1,000

1,500

2,000

2,500

0.0 0.2 0.4 0.6 0.8 1.0 1.2

BH

P/z,

psi

a


Lee 3A Reservoir - BHP/z vs. Adjusted Inventory - 2006 to 2014

I = 0.1682 + 0.0004691 (P/z)


Primary Production

WGBGNEG





Figure 5-13a: P/z vs. adjusted inventory for Lee 3A Reservoir (last eight years of storage) with recent data shown in red.

Figure 5-13 includes an Inventory Model [I = 0.1682 + 0.0004691 (P/z)] that may be used to calculate WG and BG for any desired P/z values. This model was used to calculate the values given in Table 1-1.

Figure 5-14 is the standard plot showing the calculated NEG vs. time for each withdrawal from the Lee 3A Reservoir and shows a gradual change increase from 1993 to 2006. However, from 2006 to 2014, the NEG is fairly constant at about 0.17 to 0.18 Bscf, indicating essentially no lost gas during this period.

0

500

1,000

1,500

2,000

2,500

0.0 0.2 0.4 0.6 0.8 1.0 1.2

BH

P/z,

psi

a


Lee 3A Reservoir - BHP/z vs. Adjusted Inventory - 2006 to 2014


Primary Production

spring 2014 (0.506, 699)





Figure 5-14: Lost or non-effective-gas (NEG) over time for Lee 3A Reservoir.

From 1993 to about 2005, the TIPP is increasing and the IIPP is relatively constant (Figure 5-15) which confirms a gradual gas loss. However, from 2005 to 2014 both the TIPP and IIPP are essentially constant, indicating stable storage operations.

-0.150

-0.100

-0.050

0.000

0.050

0.100

0.150

0.200

0.250

3/31/92 3/31/94 3/31/96 3/31/98 3/31/00 3/31/02 3/31/04 3/31/06 3/31/08 3/31/10 3/31/12 3/31/14

No

n-E

ffe

ctiv

e G

as, B

scf

Date

Lee 3A Reservoir - NEG vs. Time

Average lost or non-effective gas ~0.17 to 0.18 Bscf





Figure 5-15: Total and incremental inventory per pound – Lee 3A Reservoir.

Using all historical withdrawal cycles, we created a material balance model which is given by the black regression line in Figure 5-16. Also shown in Figure 5-16 is the material balance curve from primary production.

0

100

200

300

400

500

600

700

800

900

1000

0

100

200

300

400

500

600

700

800

900

1000

3/31/92 3/31/94 3/31/96 3/31/98 3/31/00 3/31/02 3/31/04 3/31/06 3/31/08 3/31/10 3/31/12 3/31/14

IIP

P, M

scf/

psi

TIP

P, M

s cf/

psi

Date

Lee 3A Reservoir - TIPP and IIPP


Both TIPP and IIPP relatively constant

TIPP increasing and IIPP relatively constant





Figure 5-16: Material balance curves for Lee 3A Reservoir.

Figure 5-16a shows the points added for this update in green. These points did not significantly affect the regression equation for Gp. As pointed out earlier, (see Figure 5-10) there was essentially no gas injected into the Lee 3A Reservoir during the summer of 2011 and very little withdrawal during the 2011-2012 withdrawal season. This resulted in the two somewhat anomalous points at 12/31/2011 and 3/31/2012.

y = -1994.7x + 1586.9

0

500

1000

1500

2000

0 0.2 0.4 0.6 0.8 1

P/z,

psi

a


Lee 3A Reservoir - P/z vs. Withdrawal - 1993 to 2014

Working Gas

Gp = 0.7956 - 0.0005013 P/z






Figure 5-16a: Material balance curves for Lee 3A Reservoir with recent data (2011 to 2014) shown in green.

5.4 Cortright Reservoir Analysis

Figure 5-17 shows the monthly injection and withdrawal for the Cortright reservoir during storage operations from 1987 to 2014, and Figure 5-18 shows the corresponding semi-annual shut-in pressure history.

0

500

1000

1500

2000

0 0.2 0.4 0.6 0.8 1

P/z,

psi

a


Lee 3A Reservoir - P/z vs. Withdrawal - 1993 to 2014

1993 to 2011

Primary Production

2012 to 2014

3/31/2012

12/31/2011


3/31/2011

9/30/2012





Figure 5-17: Historical storage operations – Cortright Reservoir.

Figure 5-18: Pressure/z during storage operations – Cortright Reservoir.

-300000

-250000

-200000

-150000

-100000

-50000

0

50000

100000

150000

200000

250000

Mar-87 Mar-89 Mar-91 Mar-93 Mar-95 Mar-97 Mar-99 Mar-01 Mar-03 Mar-05 Mar-07 Mar-09 Mar-11 Mar-13 Mar-15

Inje

ctio

n o

r W

ith

dra

wal

, Msc

f

Date

Cortright Reservoir - Monthly Injection and Withdrawal

0

200

400

600

800

1000

1200

1400

1600

1800

2000

3/31/87 3/30/89 3/31/91 3/30/93 3/31/95 3/30/97 3/31/99 3/30/01 3/31/03 3/30/05 3/31/07 3/30/09 3/31/11 3/30/13 3/31/15

BH

P/z,

psi

a

Date

BHP/z vs. Time - Cortright Reservoir





Figure 5-18a shows the seasonal injection and withdrawal from 2011 to 2014. Note that there was very little injection and withdrawal during the 2011-2012 storage season.

Figure 5-18a: Seasonal storage operations (2011 to 2014) – Cortright Reservoir.

No primary production was available for the Cortright reservoir. In order to estimate the original inventory curve without the benefit of primary production data, we used the incremental withdrawals and injections together with the corresponding P/z changes. For each withdrawal and each injection we first calculated a pore volume (PV) and a GIP value. Then we used the average PV and GIP numbers to calculate a (P/z). Having both a GIP and (P/z), we then calculated a TIPP value for each injection and withdrawal.

For the Cortright reservoir, the calculated PV and GIP from the cycle data are shown in Figures 5-19 and 5-20, respectively.

-0.800

-0.600

-0.400

-0.200

0.000

0.200

0.400

0.600

0.800


Inje

ctio

n (

+) o

r W

ith

dra

wal

(-)

, B

scf

CortrightWithdrawal Injection





Figure 19: Calculated pore volume from each cycle – Cortright Reservoir.

Figure 5-20: Calculated gas-in-place from cycle data – Cortright Reservoir.

0

2,000,000

4,000,000

6,000,000

8,000,000

10,000,000

12,000,000

14,000,000

16,000,000

1/1/1987 1/1/1992 12/31/1996 12/31/2001 1/1/2007 1/1/2012 12/31/2016

Cal

cula

ted

Po

re V

olu

me

, cu

ft

Cortright Reservoir - Calculated PV for each cycle

0

200,000

400,000

600,000

800,000

1,000,000

1,200,000

1,400,000

1,600,000

1/1/1987 1/1/1992 12/31/1996 12/31/2001 1/1/2007 1/1/2012 12/31/2016

Cal

cula

ted

GIP

, Ms c

f

Cortright Reservoir - Calculated IGIP for each cycle





Using the average PV and GIP from the first eight injections and withdrawals, we calculated a P/z of 1691 psia, from which we estimated a TIPP of 677 Mscf/psia for the Cortright Reservoir. We then found the Inventory Adjustment (0.579 Bscf) required to give the same TIPP for the P/z measured at the end of injection in the fall of 1987. This resulted in a GIP of 1.141 Bscf and placed the fall 1987 point on the TIPP curve, as shown in Figure 5-21. (Note that the estimated IGIP for this reservoir is 1.1 Bscf, as given in our 2007 Report1.)

Figure 5-21: P/z vs adjusted inventory for Cortright Reservoir.

We observe in Figure 5-21 that most cycles do indeed follow the same slope as our estimated TIPP curve. Figure 5-22 shows only the 2006 to 2014 cycles and indicates a possible lost or non-effective gas of about 0.082 Bscf. Figure 5-22a shows the recent data (2011 to 2014) in red and indicates that the performance of the Cortright reservoir has been very stable over the past eight years.

0

500

1,000

1,500

2,000

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

BH

P/z,

psi

a


Cortright Reservoir - BHP/z vs. Adjusted Inventory

Estimated TIPP = 1,144,000 / 1,690 = 677 Mscf/psi/z

spring 2010 (0.655, 816)

Starting time for analysis is 11/30/1987 with measured P/z of 1687.1 . Adding 0.579 Bscf to MGU's Working-Gas-Inventory on this date gives a GIP of 1.115 Bscf and the estimated TIPP from cycle data.

spring 2014 ( 0.544, 714)





Figure 5-22: P/z vs. adjusted inventory model for Cortright reservoir (last eight years of storage).

Figure 5-22a: P/z vs. adjusted inventory for Cortright Reservoir (last eight years of storage) with recent data (2011 to 2014) shown in red.

y = 1575.2x - 128.89

0

500

1,000

1,500

2,000

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

BH

P/z,

psi

a


Cortright Reservoir - BHP/z vs. Adjusted Inventory - 2006 to 2014


I = 0.0818 + 0.0006348 (P/z)

Estimated Primary Production

0

500

1,000

1,500

2,000

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

BH

P/z,

psi

a


Cortright Reservoir - BHP/z vs. Adjusted Inventory - 2006 to 2014

2006 to 2011

Estimated Primary Production

2011 to 2014




PetroTechnical Services Division of Schlumberger Technology Corporation 29 January 2015 page 34 Figure 5-22 above includes an Inventory Model [I = 0.0818 + 0.0006348 (P/z)] that may be used to calculate WG and BG for any desired P/z values. This model was used to determine the values given in Table 1-1. Figure 5-23 shows the graph of calculated NEG over time and indicates very little change from 2000 to 2014 with an average of about 0.08 Bscf, consistent with the NEG from the inventory model in Figure 5-22.

Figure 5-23: Lost or non-effective-gas (NEG) over time for Cortright Reservoir. Figure 5-24 shows the TIPP and IIPP and also indicates a small gas loss, as indicated by the slightly increasing TIPP from 1989 to about 2001. However, the fall TIPP values are essentially constant from 2001 to 2014, indicating no gas loss during this period. Also, the IIPP has been relatively constant the past few years. These results are consistent with the fairly constant NEG since 2001 shown in Figure 5-23.

-0.100

-0.050

0.000

0.050

0.100

0.150

0.200

3/31/87 3/30/89 3/31/91 3/30/93 3/31/95 3/30/97 3/31/99 3/30/01 3/31/03 3/30/05 3/31/07 3/30/09 3/31/11 3/30/13 3/31/15

No

n-E

ffe

ctiv

e G

as, B

scf

Date

Cortright Reservoir - NEG vs. Time

non-effective gas ~ 0.080 Bscf





Figure 5-24: Total and incremental inventory per pound – Cortright Reservoir.

Finally, using all historical withdrawal cycles, we created a material balance curve and a regression equation (Figure 5-25) which could be used to estimate working gas and base gas, in lieu of the Inventory Model given in Figure 5-22.

Figure 5-25: Material balance curve for Cortright Reservoir.

0

100

200

300

400

500

600

700

800

900

1000

0

100

200

300

400

500

600

700

800

900

1000

3/31/87 3/30/89 3/31/91 3/30/93 3/31/95 3/30/97 3/31/99 3/30/01 3/31/03 3/30/05 3/31/07 3/30/09 3/31/11 3/30/13 3/31/15

IIP

P, M

scf/

psi

TIP

P, M

scf/

psi

Date

Cortright Reservoir - TIPP and IIPP


TIPP relatively constant

IIPP relatively constant

TIPP slightly increasing

y = -1485.2x + 1655.2

0

500

1000

1500

2000

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

P/z,

psi

a


Cortright Reservoir - P/z vs. Withdrawal - 1988 to 2014

BaseGasWorking Gas

Gp = 1.114 - 0.000673 (P/z)




PetroTechnical Services Division of Schlumberger Technology Corporation 29 January 2015 page 36 Figure 5-25a shows the recent withdrawal data in green to distinguish from the earlier data. In the spring of 2011 and 2012 there was little withdrawal and during the summer of 2011 there was little injection (compared to previous cycles). This resulted in the two green points that are labeled.

Figure 5-25a: Material balance curve for Cortright Reservoir showing 2011 to 2014 data in green.

5.5 Partello Reservoir Analysis Figure 5-26 shows the monthly injection and withdrawal for the Partello reservoir during storage operations from 1992 to 2014. Note that there was no injection during the summer of 2011 and no withdrawal during the winter of 2012-2013. Also there is no injection in 2013 until September. Figure 5-26a shows the seasonal injection and withdrawal for the 2011 to 2014 storage cycles.

0

500

1000

1500

2000

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

P/z,

psi

a


Cortright Reservoir - P/z vs. Withdrawal - 1988 to 2014

1988 to 2011

2011 to 2014

10/31/2011

2/29/2012





Figure 5-26: Historical storage operations – Partello Reservoir.

Figure 5-26a: Seasonal storage operations (2011 to 2014) – Partello Reservoir.

-300,000

-200,000

-100,000

0

100,000

200,000

300,000


Inje

ctio

n o

r W

ith

dra

wal

, Msc

f

Date

Partello Reservoir - Monthly Injection and Withdrawal

No injection summer 2011

No withdrawal winter 2012-2013and no injection until Sep. 2013

-0.400

-0.300

-0.200

-0.100

0.000

0.100

0.200

0.300

0.400


Inje

ctio

n (

+) o

r W

ith

dra

wal

(-)

, B

scf

PartelloWithdrawal Injection




PetroTechnical Services Division of Schlumberger Technology Corporation 29 January 2015 page 38 Figure 5-27 shows the corresponding semi-annual shut-in pressure history for the Partello Reservoir. Note that the pressure was showing a gradual increase from about 2002 to 2011, but has now stabilized.

Figure 5-27: Pressure/z during storage operations – Partello Reservoir. Figure 5-28 shows P/z and our estimated GIP for the Partello Reservoir for the past nine years. Both the pressure and GIP have stabilized due to changes in field operations since 2011. Specifically, the increase in P/z and GIP from 2009 to 2011 no longer exists. We now see both stable P/z and GIP from 2010 to 2014.

0

200

400

600

800

1000

1200

3/31/92 3/31/94 3/31/96 3/31/98 3/31/00 3/31/02 3/31/04 3/31/06 3/31/08 3/31/10 3/31/12 3/31/14

BH

P/z,

psi

a

Date

Partello Reservoir - BHP/z vs. Time

Stabilized P/z





Figure 5-28: P/z and GIP from 2005 to 2014 – Partello Reservoir.

No primary production was available for the Partello reservoir. In order to estimate the original inventory curve without the benefit of primary production data, we used the same procedure as used for the Cortright reservoir. For the Partello reservoir, the calculated PV and GIP from the cycle data are shown in Figures 5-29 and 5-30, respectively.

Figure 5-29: Calculated pore volume from each cycle – Partello Reservoir.

0.000

0.200

0.400

0.600

0.800

1.000

1.200

1.400

1.600

1.800

0

200

400

600

800

1000

1200

1/1/05 1/1/06 1/1/07 1/1/08 1/1/09 1/1/10 1/1/11 1/1/12 1/1/13 1/1/14 1/1/15

Cal

cula

ted

GIP

, Bsc

f

BH

P/z

, p

sia

Partello Reservoir - P/z and GIP vs. time (2005 to 2014)

P/z - 2005 to 2011 P/z - 2011 to 2014 GIP - 2005 to 2011 GIP - 2011 to 2014

Stabilized P/z

Stabilized GIP

0

5,000,000

10,000,000

15,000,000

20,000,000

25,000,000

30,000,000

35,000,000

40,000,000

1/1/1993 1/1/1995 1/1/1997 1/1/1999 1/1/2001 1/1/2003 1/1/2005 1/1/2007 1/1/2009 1/1/2011 1/1/2013 1/1/2015

Cal

cula

ted

Po

re V

olim

e,

cu f

t

Partello Reservoir - Calculated PV for each cycle





Figure 5-30: Calculated gas-in-place from cycle data – Partello Reservoir.

We observed that the IGIP value calculated for the 2008 injection (1.713 Bscf) agreed fairly well with the estimated primary production of 1.695 Bscf presented in our 2007 Report1. Using this IGIP of 1.713 Bscf and our calculated pore volume, we calculated a P/z of 1018 psia. This resulted in a TIPP of 1,683 Mscf/Psia which was used to construct the estimated initial inventory curve for the reservoir. To obtain a TIPP of 1,683 at the end of the first injection in November, 1992 required an addition of 1.582 Bscf to MGU’s Working Gas Inventory (WGI). This resulted in the inventory graph shown in Figure 5-31, with the 2011 to 2014 data shown in red.

0

500,000

1,000,000

1,500,000

2,000,000

2,500,000

1/1/1993 1/1/1995 1/1/1997 1/1/1999 1/1/2001 1/1/2003 1/1/2005 1/1/2007 1/1/2009 1/1/2011 1/1/2013 1/1/2015

Cal

cula

ted

GIP

, M

scf

Partello Reservoir - Calculated IGIP for each cycle

Calculated GIP at end of 2008 injection was 1.713 Bscf.





Figure 5-31: P/z vs. adjusted inventory for Partello Reservoir.

Since most of the data falls above the estimated inventory curve with TIPP = 1,683 Mscf/psia, we divided the data into two time categories: before 2006 and after 2006. The results are shown in Figures 5-32 and 5-33. In Figure 5-33 the recent data from 2011 to 2014 is again shown in red.

Figure 5-32: P/z vs. adjusted inventory (before 2006) – Partello Reservoir.

0

100

200

300

400

500

600

700

800

900

1,000

1,100

1,200

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

BH

P/z,

psi

a


Partello Reservoir - BHP/z vs. Adjusted Inventory

TIPP = 1,713,152 / 1018.08 = 1,683 Mscf/psia/z

9/30/2006 (1.435, 1062)

4/30/1999 (1.088, 815)

Adjusted Inventory = 1,582,000 Mscf

0

100

200

300

400

500

600

700

800

900

1,000

1,100

1,200

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8

BH

P/z,

psi

a


Partello Reservoir - BHP/z vs. Adjusted Inventory 1992 to 2006

Inventory Adjustment = 1,582,000 Mscf





Figure 5-33: P/z vs. adjusted inventory (after 2006) – Partello Reservoir.

The data before 2006 follows the estimated TIPP slope fairly well, whereas the data after 2006 seem to indicate a lower TIPP of about 1,454 Mscf/psia. This could be due to water influx gradually reducing the gas-filled pore volume. Thus, for current operations, we will use the inventory model obtained using the 2006 to 2014 data; i.e. [I = 0.001454 (P/z)]. Figure 5-34 shows the graph of calculated NEG over time from the spring of 1993 to the spring of 2014. Except for the spring 2007 and 2009 points (highlighted), the calculated NEG is fairly low and constant at about 0.05 Bscf. The 2007 and 2009 points are high and low, respectively, due to the fact that the change in pressure and GIP were disproportionate from the fall of 2006 to the spring of 2007 and again from the fall of 2008 to the spring of 2009. This can easily be seen back in Figure 5.28.

0

100

200

300

400

500

600

700

800

900

1,000

1,100

1,200

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8

BH

P/z,

psi

a


Partello Reservoir - BHP/z vs. Adjusted Inventory 2006 to 2014

I = 0.001454 (P/z)

TIPP = 1,713,152 / 1018.08 = 1,683 Mscf/psia/z






Figure 5-34: Lost or non-effective gas (NEG) over time for Partello Reservoir.

Figure 5-35 shows the TIPP and IIPP over time for the Partello Reservoir.

Figure 5-35: Total and incremental inventory per pound – Partello Reservoir.

-1.000

-0.800

-0.600

-0.400

-0.200

0.000

0.200

0.400

3/31/92 3/31/94 3/31/96 3/31/98 3/31/00 3/31/02 3/31/04 3/31/06 3/31/08 3/31/10 3/31/12 3/31/14

No

n-E

ffec

tive

Ga s

, Bsc

f

Date

Partello Reservoir - NEG vs. Time

Average ~ 0.05 Bscf

0

500

1000

1500

2000

2500

0

500

1000

1500

2000

2500

3/31/92 3/31/94 3/31/96 3/31/98 3/31/00 3/31/02 3/31/04 3/31/06 3/31/08 3/31/10 3/31/12 3/31/14 3/31/16

IIP

P, M

scf/

psi

TIP

P , M

scf/

psi

Date

Partello Reservoir - TIPP and IIPP





PetroTechnical Services Division of Schlumberger Technology Corporation 29 January 2015 page 44 As shown by the gold line in Figure 5-35, the TIPP is decreasing with time until 2007, consistent with our calculated decreasing pore volume (Figure 5-29). However, as shown by the black line, the TIPP is very stable from 2008 to 2014. There is too much noise in the IIPP data to use it for analysis. This is due to the problems discussed above in 2007 and 2009, and also due to the fact that no gas was injected during the summer of 2011 and no gas was withdrawn in the winter of 2012-2013 (Figure 5-26), resulting in the two zero points on the IIPP Update curve. Using all historical withdrawal cycles, we created a material balance curve for the Partello Reservoir which is shown in Figure 5-36. The regression equation could be used to estimate working gas and base gas, in lieu of the Inventory Model given in Figure 5-33. Figure 5-36 shows the recent data (2011 to 2014) in green.

Figure 5-36: Material balance curve for Partello Reservoir.

5.6 Anderson Reservoir Analysis Figure 5-37 shows the monthly injection and withdrawal for the Anderson reservoir during storage operations from 1992 to 2014 and Figure 5-37a shows the seasonal storage operations from 2011 to 2014.

y = -657.64x + 993.82

0

500

1000

1500

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

P/z,

psi

a


Partello Reservoir - P/z vs. Withdrawal - 1992 to 2014

Base GasWorking Gas

Gp = 1.511 - 0.001521 (P/z)





Figure 5-37: Historical storage operations – Anderson Reservoir.

Figure 5-37a: Seasonal storage operations (2011 to 2014) – Anderson Reservoir.

Figure 5-38 shows the monthly P/z history.

-200,000

-150,000

-100,000

-50,000

0

50,000

100,000

150,000

200,000

Mar-92 Mar-94 Mar-96 Mar-98 Mar-00 Mar-02 Mar-04 Mar-06 Mar-08 Mar-10 Mar-12 Mar-14 Mar-16

Inje

ctio

n o

r W

ith

dra

wal

, Msc

f

Date

Anderson Reservoir - Monthly Injection and Withdrawal

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4


Inje

ctio

n (

+) o

r W

ith

dra

wal

(-)

, B

scf

AndersonWithdrawal Injection





Figure 5-38: Pressure/z during storage operations – Anderson Reservoir.

No primary production was available for the Anderson reservoir. Therefore we used the same procedure as we developed for the Cortright reservoir to estimate the original inventory curve. For the Anderson reservoir, the calculated PV and GIP from the cycle data are shown in Figures 5-39 and 5-40, respectively. It is significant to note that the calculated PV and GIP have now stabilized.

Figure 5-39: Calculated pore volume from each cycle – Anderson Reservoir.

0

200

400

600

800

1000

1200

3/31/92 3/31/94 3/31/96 3/31/98 3/31/00 3/31/02 3/31/04 3/31/06 3/31/08 3/31/10 3/31/12 3/31/14 3/31/16

BH

P/z,

psi

a

Date

Anderson Reservoir - BHP/z vs. Time

-5,000,000

0

5,000,000

10,000,000

15,000,000

20,000,000

1/1/1993 1/1/1995 1/1/1997 1/1/1999 1/1/2001 1/1/2003 1/1/2005 1/1/2007 1/1/2009 1/1/2011 1/1/2013 1/1/2015

Cal

cula

ted

Po

re V

olim

e, c

u f

t

Anderson Reservoir - Calculated PV for each cycle

Stabilized Pore Volume





Figure 5-40: Calculated gas-in-place from each cycle – Anderson Reservoir.

We used the average PV and GIP calculated from the 1995 to 1999 cycles to estimate an initial P/z of 951 psia. This resulted in a TIPP of 1,172 Mscf/Psia which was used as the estimated initial inventory curve for the reservoir. In order to obtain a TIPP of 1,172 at the end of the first injection in November, 1992 required an addition of 1.085 Bscf to MGU’s Working Gas Inventory. This resulted in the inventory graph shown in Figure 5-41. The most recent data (2011 to 2014) is shown in red.

-200,000

0

200,000

400,000

600,000

800,000

1,000,000

1,200,000

1,400,000

1/1/1993 1/1/1995 1/1/1997 1/1/1999 1/1/2001 1/1/2003 1/1/2005 1/1/2007 1/1/2009 1/1/2011 1/1/2013 1/1/2015

Cal

cula

ted

GIP

, Msc

f

Anderson Reservoir - Calculated IGIP for each cycle

Stabilized GIP





Figure 5-41: P/z vs. adjusted inventory for Anderson Reservoir.

Figure 5-42 shows only the 2006 to 2014 data and indicates that the reservoir volume has decreased. Also shown is an inventory model [I = 0.05636 + 0.0009555 (P/z)] which we used to calculate the WG and BG values given in Table 1-1. Figure 5-42a shows the same data with the most recent data (2011 to 2014) in red.

Figure 5-42: P/z vs. adjusted inventory for Anderson Reservoir (2006 to 2014).

0

100

200

300

400

500

600

700

800

900

1,000

1,100

1,200

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

BH

P/z,

psi

a


Anderson Reservoir - BHP/z vs. Adjusted Inventory - 1992 to 2014

TIPP = 1,115,000 / 952= 1172 Mscf/psi

10/31/2000 (1.186, 1026)10/31/2004 (1.092, 1049)

3/31/2001 (0.752, 633.5)

4/30/1994 (0.792, 621)


y = 1046.6x - 58.991

0

100

200

300

400

500

600

700

800

900

1,000

1,100

1,200

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

BH

P/z,

psi

a



TIPP = 1,115,000 / 952= 1172 Mscf/psi

I = 0.05636 + 0.0009555 (P/z)






Figure 5-42a: P/z vs. adjusted inventory for Anderson Reservoir (2006 to 2014) with most recent data shown in red.

The declining pore volume indicated in Figure 5-42 is consistent with the declining pore volume shown in Figure 5-39, which is apparently due to water influx. Figure 5-43 shows the graph of the calculated NEG over time and indicates only a very small amount of lost gas from 1992 to 2009. Note that in 2008 the calculated NEG is about 0.06 Bscf, consistent with the intercept of the inventory model in Figure 5-42. We do not believe that the last three calculated NEG’s (spring of 2012, 2013 and 2014) are significant, due to the fact that there was very little gas injection during this period.

0

100

200

300

400

500

600

700

800

900

1,000

1,100

1,200

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

BH

P/z,

psi

a



TIPP = 1,115,000 / 952= 1172 Mscf/psi






Figure 5-43: Lost or non-negative gas (NEG) over time for Anderson Reservoir. Figure 5-44 shows the graph of the calculated TIPP and IIPP over time. Note that both have stabilized since about 2010.

Figure 5-44: Total and Incremental Inventory per Pound – Anderson Reservoir.

-0.100

0.000

0.100

0.200

0.300

0.400

0.500

0.600

0.700

0.800

0.900

1.000

3/31/92 3/31/94 3/31/96 3/31/98 3/31/00 3/31/02 3/31/04 3/31/06 3/31/08 3/31/10 3/31/12 3/31/14

No

n-E

ffec

tive

Gas

, Bsc

f

Date

Anderson Reservoir - NEG vs. Time

NEG ~ 0.06 Bscf in 2009

0

200

400

600

800

1000

1200

1400

1600

1800

2000

0

200

400

600

800

1000

1200

1400

1600

1800

2000

3/31/92 3/31/94 3/31/96 3/31/98 3/31/00 3/31/02 3/31/04 3/31/06 3/31/08 3/31/10 3/31/12 3/31/14

IIP

P, M

scf/

psi

TIP

P, M

scf/

psi

Date

Anderson Reservoir - TIPP and IIPP


Both TIPP and IIPP have stabilized since 2010




PetroTechnical Services Division of Schlumberger Technology Corporation 29 January 2015 page 51 Using all historical withdrawal cycles, we created a material balance curve for the Anderson Reservoir which is shown in Figure 5-45, with the most recent data shown in green. The regression equation could be used to estimate working gas and base gas, in lieu of the Inventory Model given in Figure 5-42.

Figure 5-45: Material balance curve for Anderson Reservoir.

5.7 Mymachod Reservoir Analysis Figure 5-46 shows the monthly injection and withdrawal for the Mymachod reservoir during storage operations from 1992 to 2014 and Figure 5-47 shows the corresponding semi-annual shut-in pressure history. As shown in Figure 5-47a, there has been very little injection or withdrawal since 2011, compared to previous years.

y = -920.35x + 974.33

0

500

1000

1500

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

P/z,

psi

a


Anderson Reservoir - P/z vs. Withdrawal - 1992 to 2014

Gp = 1.0586 - 0.001086(P/z)

Working Gas Working Gas





Figure 5-46: Historical storage operations – Mymachod Reservoir.

Figure 5-47: Pressure/z during storage operations – Mymachod Reservoir.

-60,000

-40,000

-20,000

0

20,000

40,000

60,000

80,000


Inje

ctio

n o

r W

ith

dra

wal

, Msc

f

Date

Mymachod Reservoir - Monthly Injection and Withdrawal

no injection or withdrawal from 1/31/2012 to 8/31/2013

0

200

400

600

800

1000

1200

3/31/92 3/31/94 3/31/96 3/31/98 3/31/00 3/31/02 3/31/04 3/31/06 3/31/08 3/31/10 3/31/12 3/31/14

BH

P/z,

psi

a

Date

Mymachod Reservoir - BHP/z vs. Time





Figure 5-47a: Seasonal storage operations (2011 to 2014) – Mymachod Reservoir.

No primary production was available for the Mymachod reservoir. Therefore we used the same procedure as we developed for the Cortright reservoir to estimate the original inventory curve. For the Mymachod reservoir, the calculated PV and GIP from the cycle data are shown in Figures 5-48 and 5-49, respectively.

-0.050

-0.040

-0.030

-0.020

-0.010

0.000

0.010

0.020

0.030

0.040


Inje

ctio

n (

+) o

r W

ith

dra

wal

(-)

, B

scf

MymachodWithdrawal Injection





Figure 5-48: Calculated pore volume from each cycle – Mymachod Reservoir.

Figure 5-49: Calculated gas-in-place from each cycle – Mymachod Reservoir.

-1,000,000

1,000,000

3,000,000

5,000,000

7,000,000

9,000,000

11,000,000

1/1/1993 1/1/1995 1/1/1997 1/1/1999 1/1/2001 1/1/2003 1/1/2005 1/1/2007 1/1/2009 1/1/2011 1/1/2013 1/1/2015

Cal

cula

ted

Po

re V

olim

e, c

u f

t

Mymachod Reservoir - Calculated PV for each cycle

0

100,000

200,000

300,000

400,000

500,000

600,000

700,000

1/1/1993 1/1/1995 1/1/1997 1/1/1999 1/1/2001 1/1/2003 1/1/2005 1/1/2007 1/1/2009 1/1/2011 1/1/2013 1/1/2015

Cal

cula

ted

GIP

, Msc

f

Mymachod Reservoir - Calculated IGIP for each cycle




PetroTechnical Services Division of Schlumberger Technology Corporation 29 January 2015 page 55 We used the average PV and GIP calculated from the 1996 to 2000 cycles to estimate a P/z of 969 psia. This resulted in a TIPP of 345.6 Mscf/Psia which was used as the estimated initial inventory curve for the reservoir. In order to obtain a TIPP of 345.6 at the end of the first injection in November, 1992 required an addition of 0.320 Bscf to MGU’s Working Gas Inventory. This resulted in the inventory graph shown in Figure 5-50. Note that there is a vertical segment; this occurs because the inventory and pressure data show wellhead pressures of 686 and 741 psig at 1/31/2012 and 4/30/2013, respectively, with no injection or withdrawal during this period.

Figure 5-50: P/z vs. adjusted inventory for Mymachod Reservoir.

There is a gradual increase in slope over time, which indicates a decrease in pore volume over time. Figure 5-51 shows only the 2006 to 2014 data. This change in slope is consistent with the decrease in calculated pore volume noted Figure 5-48, and is believed to be due to water influx. Figure 5-51 includes an Inventory Model [I = 0.0203 + 0.0002708 (P/z)] which may be used to calculate WG and BG for any desired P/z values.

0

100

200

300

400

500

600

700

800

900

1,000

1,100

1,200

0.0 0.1 0.2 0.3 0.4

BH

P/z,

psi

a


Mymachod Reservoir - BHP/z vs. Adjusted Inventory (1992 to 2014)

TIPP = 328300 / 949.96 = 345.6 Mscf/psia/z






Figure 5-51: P/z vs. adjusted inventory (2006 to 2014) – Mymachod Reservoir.

Figure 5-52 shows a graph of the calculated NEG over time and indicates only a very small amount of lost gas.

Figure 5-52: Lost or non-effective-gas (NEG) over time for Mymachod Reservoir.

y = 3693.4x - 74.92

0

100

200

300

400

500

600

700

800

900

1,000

1,100

1,200

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50

BH

P/z,

psi

a


Mymachod Reservoir - BHP/z vs. Adjusted Inventory - 2006 to 2014


I = 0.0203 + 0.0002708 (P/z)

-0.150

-0.100

-0.050

0.000

0.050

0.100

0.150

0.200

0.250

0.300

3/31/92 3/31/94 3/31/96 3/31/98 3/31/00 3/31/02 3/31/04 3/31/06 3/31/08 3/31/10 3/31/12 3/31/14 3/31/16

No

n-E

ffec

tive

Gas

, Bsc

f

Date

Mymachod Reservoir - NEG vs. Time




PetroTechnical Services Division of Schlumberger Technology Corporation 29 January 2015 page 57 Figure 5-53 shows the graph of the calculated TIPP and IIPP over time. Both have decreased slightly, which further supports the hypothesis that the Mymachod reservoir volume has decreased. However, since about 2000 the TIPP is relatively constant, indicating a fairly stable pore volume.

Figure 5-53: Total and Incremental Inventory per Pound – Mymachod Reservoir. Using all historical withdrawal cycles, we created a material balance curve for the Mymachod Reservoir which is shown in Figure 5-54. The regression equation could be used to estimate working gas and base gas, in lieu of the Inventory Model given in Figure 5-51.

0

100

200

300

400

500

600

700

800

900

1000

0

100

200

300

400

500

600

700

800

900

1000

3/31/92 3/31/94 3/31/96 3/31/98 3/31/00 3/31/02 3/31/04 3/31/06 3/31/08 3/31/10 3/31/12 3/31/14 3/31/16

IIP

P, M

scf/

psi

TIP

P, M

scf/

psi

Date

Mymachod Reservoir - TIPP and IIPP


TIPP relatively stable since 2000





Figure 5-54: Material balance curve for Mymachod Reservoir.

Figure 5-54 also shows the recent data (2011 to 2014) in green. 5.8 MGU 1-24 Reservoir Analysis Figure 5-55 shows the monthly injection and withdrawal for the MGU 1-24 reservoir during storage operations from 1992 to 2014 and Figure 5-56 shows the corresponding semi-annual shut-in pressure history. Figure 5-56a shows the seasonal injection and withdrawal since 2011.

y = -3189.1x + 958.7

0

500

1000

1500

0 0.1 0.2 0.3 0.4

P/z,

psi

a


Mymachod Reservoir - P/z vs. Withdrawal - 1992 to 2014

Gp = 0.3006 - 0.0003136*(P/z)

Working Gas Base Gas





Figure 5-55: Historical storage operations – MGU 1-24 Reservoir.

Figure 5-56: Pressure/z during storage operations – MGU 1-24 Reservoir.

-60,000

-40,000

-20,000

0

20,000

40,000

60,000

Mar-92 Mar-94 Mar-96 Mar-98 Mar-00 Mar-02 Mar-04 Mar-06 Mar-08 Mar-10 Mar-12 Mar-14 Mar-16

Inje

ctio

n o

r W

ith

dra

wal

, Msc

f

Date

MGU 1-24 Reservoir - Monthly Injection and Withdrawal

0

200

400

600

800

1000

1200

3/31/92 3/31/94 3/31/96 3/31/98 3/31/00 3/31/02 3/31/04 3/31/06 3/31/08 3/31/10 3/31/12 3/31/14 3/31/16

BH

P/z,

psi

a

Date

MGU 1-24 Reservoir - BHP/z vs. Time





Figure 5-56a: Seasonal Storage Operations (2011 to 2014) – MGU 1-24 Reservoir.

No primary production was available for the MGU 1-24 reservoir. Therefore we used the same procedure as we developed for the Cortright reservoir to estimate the original inventory curve. For the MGU 1-24 reservoir, the calculated PV and GIP from the cycle data are shown in Figures 5-57 and 5-58, respectively.





Figure 5-57: Calculated pore volume from each cycle – MGU 1-24 Reservoir.

Figure 5-58: Calculated gas-in-place from cycle data – MGU 1-24 Reservoir.

0

500,000

1,000,000

1,500,000

2,000,000

2,500,000

3,000,000

3,500,000

4,000,000

4,500,000

5,000,000

1/1/1993 1/1/1995 1/1/1997 1/1/1999 1/1/2001 1/1/2003 1/1/2005 1/1/2007 1/1/2009 1/1/2011 1/1/2013 1/1/2015

Cal

cula

ted

Po

re V

olim

e, c

u f

t

MGU 1-24 Reservoir - Calculated PV for each cycle

0

50,000

100,000

150,000

200,000

250,000

300,000

1/1/1993 1/1/1995 1/1/1997 1/1/1999 1/1/2001 1/1/2003 1/1/2005 1/1/2007 1/1/2009 1/1/2011 1/1/2013 1/1/2015

Cal

cula

ted

GIP

, Msc

f

MGU 1-24 Reservoir - Calculated IGIP for each cycle




PetroTechnical Services Division of Schlumberger Technology Corporation 29 January 2015 page 62 We used the average PV and GIP calculated from the 1997 to 2001 cycles to estimate a P/z of 955 psia. This resulted in a TIPP of 223 Mscf/Psia which was used as the estimated initial inventory curve for the reservoir. In order to obtain a TIPP of 223 at the end of the first injection in November, 1992 required an addition of 0.203 Bscf to MGU’s Working Gas Inventory. This resulted in the inventory graph shown in Figure 5-59.

Figure 5-59: P/z vs. adjusted inventory for MGU 1-24 Reservoir.

There is a significant shift to the right which might indicate lost or non-effective gas. Figure 5-60 shows only the 2006 to 2014 data and indicates that, in the past few years, the performance has been relatively stable. Figure 5-60 includes an Inventory Model [I = 0.0537 + 0.0001994 (p/z)] which may be used to calculate WG and BG for any desired P/z values.

0

100

200

300

400

500

600

700

800

900

1,000

1,100

1,200

0.0 0.1 0.2 0.3

BH

P/z,

psi

a


MGU 1-24 Reservoir - BHP/z vs. Adjusted Inventory (1992 to 2014)

TIPP = 212,603 / 955.03 = 222.61 Mscf/psia/z






Figure 5-60: P/z vs. adjusted inventory (2006 to 2014) – MGU 1-24 Reservoir.

Figure 5-60a shows the 2006 to 2014 data again, with the recent data (2011 to 2014) in red.

Figure 5-60a: P/z vs. adjusted inventory (2006 to 2014) – MGU 1-24 Reservoir.

y = 5013.8x - 269.07

0

100

200

300

400

500

600

700

800

900

1,000

1,100

1,200

0.00 0.05 0.10 0.15 0.20 0.25 0.30

BH

P/z,

psi

a


MGU 1-24 Reservoir - BHP/z vs. Adjusted Inventory - 2006 to 2014

I = 0.0537 + 0.0001994 (P/z)


0

100

200

300

400

500

600

700

800

900

1,000

1,100

1,200

0.00 0.05 0.10 0.15 0.20 0.25 0.30

BH

P/z,

psi

a


MGU 1-24 Reservoir - BHP/z vs. Adjusted Inventory - 2006 to 2014





PetroTechnical Services Division of Schlumberger Technology Corporation 29 January 2015 page 64 Figure 5-61 shows the graph of the calculated NEG over time and indicates only a small amount of lost gas from 1994 to 2006. The average value for the last five years is relatively constant at about 0.05 Bscf, consistent with the Inventory Model (Figure 5-60). Note: We do not believe the last three calculated NEG values (spring of 2011, 2013, and 2014) are meaningful, due to the fact that there was no injection during the summer of 2011, very little injection in 2012, and no withdrawal the winter of 2012-2013.

Figure 5-61: Lost or non-effective-gas (NEG) over time for MGU 1-24 Reservoir.

Figure 5-62 shows the graph of the calculated TIPP and IIPP over time. The TIPP increased slightly from 1993 to about 2004 while the IIPP remained relatively constant. This confirms the possibility of a small gas loss. However, the TIPP has been quite stable since 2005, indicating a stable storage operation.

-0.050

0.000

0.050

0.100

0.150

0.200

0.250

0.300

3/31/92 3/31/94 3/31/96 3/31/98 3/31/00 3/31/02 3/31/04 3/31/06 3/31/08 3/31/10 3/31/12 3/31/14 3/31/16

No

n-E

ffe

ctiv

e G

as, B

scf

Date

MGU 1-24 Reservoir - NEG vs. Time





Figure 5-62: Total and Incremental Inventory per Pound – MGU 1-24 Reservoir.

Using all historical withdrawal cycles, we created a material balance curve for the MGU 1-24 Reservoir which is shown in Figure 5-63. The regression equation could be used to estimate working gas and base gas, in lieu of the Inventory Model given in Figure 5-60. Figure 5.63 also shows the recent data (2011 to 2014) in green.

Figure 5-63: Material balance curve for MGU 1-24 Reservoir.

0

100

200

300

400

500

600

700

800

0

100

200

300

400

500

600

700

800

3/31/92 3/31/94 3/31/96 3/31/98 3/31/00 3/31/02 3/31/04 3/31/06 3/31/08 3/31/10 3/31/12 3/31/14 3/31/16

IIP

P, M

scf/

psi

TIP

P, M

scf/

psi

Date

MGU 1-24 Reservoir - TIPP and IIPP


TIPP increases slightly frm 1993 to about 2004; then is relatively stable

y = -4816.9x + 994.36

0

500

1000

1500

0 0.1 0.2 0.3

P/z,

psi

a


MGU 1-24 Reservoir - P/z vs. Withdrawal - 1993 to 2014

Gp = 0.2064 - .0002076*(P/z)

Working Gas Base Gas




PetroTechnical Services Division of Schlumberger Technology Corporation 29 January 2015 page 66 6. References

1. “Inventory Analysis of Select Underground Gas Storage Fields in Michigan,” Schlumberger DCS, March, 2007.

2. “Inventory Analysis of Underground Gas Storage Reservoirs in Southeastern Michigan.” Schlumberger DCS,

March, 2012.




Case No. U-17880Exhibit No.: A-3

Schedule: C32Page 1 of 1


ReservoirTotal

Cushion GasRecoverable Cushion Gas

Non-Recoverable Cushion Gas

Working Gas

Adjusted Inventory

Total Cushion Gas

RecoverableCushion Gas


Working Gas

Adjusted Inventory

Total Cushion Gas

Recoverable Cushion Gas


Working Gas

(Bscf) (Bscf) (Bscf) (Bscf) (Bscf) (Bscf) (Bscf) (Bscf) (Bscf) (Bscf) (Bscf) (Bscf) (Bscf) (Bscf)

Lee 3 0.823 0.542 0.281 0.712 1.535 0.770 0.489 0.281 0.765 1.535 -0.053 -0.053 0.000 0.053

Lee 3A 0.504 0.303 0.200 0.393 0.897 0.474 0.274 0.200 0.423 0.897 -0.030 -0.030 0.000 0.030

Cortright 0.551 0.425 0.126 0.566 1.117 0.510 0.383 0.126 0.607 1.117 -0.042 -0.042 0.000 0.042

Partello 0.773 0.347 0.426 0.791 1.564 1.232 0.459 0.773 0.332 1.564 0.459 0.112 0.347 -0.459

Anderson 0.562 0.227 0.335 0.479 1.041 0.862 0.300 0.562 0.179 1.041 0.300 0.074 0.227 -0.300

Mymachod 0.164 0.065 0.100 0.140 0.305 0.250 0.086 0.164 0.055 0.305 0.086 0.021 0.065 -0.086

MGU 1-24 0.160 0.048 0.112 0.114 0.274 0.223 0.063 0.160 0.051 0.274 0.063 0.015 0.048 -0.063

TOTAL 3.536 1.956 1.580 3.196 6.732 4.320 2.054 2.266 2.411 6.732 0.785 0.099 0.686 -0.785

Michigan Gas Utilities Corporation

AdjustmentsPrior Operating Criteria New Operating Criteria

Table of Working Gas, Recoverable Cushion Gas, and Non-Recoverable Cushion Gas

state of michigan michigan gas utilities … · 4 schlumberger technology corporation ... schedule...

Documents