The URL can be used to link to this page

Home My WebLink About2014 CINGSACook Inlet Alatural>'Gas STOPAQF. May 15, 2015 State of Alaska Alaska Oil and Gas Conservation Commission 333 West 7`" Ave, Suite 100 Anchorage, AK 99501 Attn: Cathy Foerster — Chair of Commission 3000 Spenard Road PO Box 190989 Anchorage, AK 99519-0989 RECEIVED MAY 15 2015 AOGCC RE: Cook Inlet Natural Gas Storage Alaska Annual Material Balance and Storage Performance Report Dear Chairman Foerster: Cook Inlet Natural Gas Storage Alaska, LLC (CINGSA) was granted a Storage Injection Order on November 19, 2010 by the Alaska Oil and Gas Conservation Commission, allowing it to operate the Cannery Loop Sterling C Pool for underground natural gas storage service. Rule 8 of Storage Injection Order No. 9 (SIO 009) requires that CINGSA annually file with the Commission a report that includes material balance calculations of the gas production and injection volumes and a summary of well performance data to provide assurance of continued reservoir confinement of the gas storage volumes. Per Storage Injection Order No. 9.001, the Commission revised the due date for this Report to May 15 of each year. CINGSA has now completed three full years of operation. The enclosed report, in compliance with Rule 8 of SIO 009, documents gas storage operational activity during the past thirty-six months and includes monthly net injection/withdrawal volumes for the facility and total gas inventory at month-end. Any questions concerning the attached information may be directed to Richard Gentges at 989-464-3849. Sincerely, Jar Green esident Cook Inlet Natural Gas Storage Alaska, LLC Attachment Cook Inlet Natural Gas Storage Alaska, LLC 2014-2015 Storage Field Injection/Withdrawal Performance and Material Balance Report Executive Summary/Conclusion Cook Inlet Natural Gas Storage Alaska, LLC (CINGSA) filed an application with the Alaska Oil and Gas Conservation Commission (AOGCC) on July 27, 2010 for authority to operate the Cannery Loop Sterling C Pool to provide underground natural gas storage service. In that application, CINGSA requested authority to store a total of 18 Bcf of natural gas, including 7 Bcf of base gas and 11 Bcf of working gas. CINGSA estimated that this initial phase of development would result in a maximum average reservoir pressure of approximately 1520 psia based upon the original material balance analysis of the reservoir, all as more fully described in the application. By Order dated November 19, 2010, AOGCC issued Injection Order No. 9 (SIO 9) granting CINGSA the authorization sought in its application, and limited the maximum allowed reservoir pressure to 1700 psia. In April 2014, CINGSA subsequently submitted an application to the AOGCC requesting authority to increase the maximum reservoir pressure to the original discovery pressure of 2200 psia. By Order dated June 4, 2014, the AOGCC issued Injection Order No. 9a, granting CINGSA the authorization sought in its April 2014 application. Rule 8 of SIO 9a states that CINGSA must annually file with the AOGCC a report that includes material balance calculations of the gas production and injection volumes and a summary of well performance data to provide assurance of continued reservoir confinement of the gas storage volumes. This report is the third such annual report to be filed by CINGSA. The CINGSA facility was commissioned in April 2012, and has now completed three full years of operation. This report documents gas storage operational activity during the past twelve months and includes monthly net injection/withdrawal volumes for the facility and total gas inventory at month-end. A plot of the actual wellhead pressure versus total gas inventory performance of the field is contained in this report; the plot demonstrates that the pressure versus inventory performance is generally consistent with design expectations, although actual pressure has trended above design expectations. CINGSA believes the primary reason for this is related to an isolated pocket of native gas, believed to be at or near native pressure conditions, which CINGSA encountered when it perforated/completed the CLU S-1 well. This gas has since commingled with gas in the depleted main reservoir and provides pressure support to the storage operation. Based upon currently available data, the estimated volume of gas associated with the isolated pocket is approximately 14.5 Bcf. CINGSA believes it will be able to refine this estimate with additional field -wide shut-in data as such shut-ins occur. This report also documents the injection/withdrawal flow rate performance of each of the five wells. A field -wide withdrawal test was performed on March 9, and again on March 13, 2015. Results from these tests confirm that field deliverability remains unchanged from the first year of storage operations. While there is some evidence that CLU S-3 and S-5 have experienced some decline in deliverability performance, wells CLU S-1 and S-2 appear to have improved somewhat, which offsets the declines in the other two wells. There is no evidence that suggests a permanent decline in well deliverability could be related to a loss of well bore integrity in any of the five wells. Consistent with standard operations, two planned facility shut -downs were conducted during the past twelve months, each one week in duration. The first shut -down occurred during October 2014 and the second in April of this year. The purpose of these two shut -downs was to suspend injection/withdrawal operations so that each well could be shut-in for pressure monitoring and to allow reservoir pressure to stabilize. The well shut-in pressure data was analyzed via graphical material balance analysis. The results of that analysis confirm that all of the injected gas remains confined within the reservoir. Each well that penetrates the cap rock of the Sterling C Gas Storage Pool could conceivably be a leak path for injected storage gas. If a loss of well or reservoir integrity were to occur, it is likely that it would manifest itself via a rise in annular pressure of any well that penetrates the storage pool. This report includes a summary of shut-in pressures recorded on all of the annular spaces of each of the CINGSA storage wells and select annular spaces of all third party wells which penetrate the Sterling C Gas Storage Pool. This annular pressure data also indicates there is no evidence of any gas leakage from the Sterling C Gas Storage Pool. Accordingly, all operating data indicate that reservoir integrity remains intact, and although the reservoir may now be effectively larger than expected due to encountering additional native gas in the Sterling Clc interval of the CLU S-1 well, all of the injected gas remains within the greater reservoir and is accounted for at this time. 2014-2015 Storage Operations The 2014-2015 storage cycle covers the period from April 9, 2014, the final day of the 2014 spring semi-annual shut -down, through April 8, 2015. Total storage inventory at April 9, 2014 was 13,147,315 Mcf. Table 1 lists the remaining native gas -in-place as of April 1, 2012, net injection/withdrawal activity by month during the past 36 months, and the total gas -in-place at the end of each month since storage operations commenced. Please note that the figures listed in Table 1 have not been adjusted to include the 14.5 Bcf of additional native gas associated with the isolated pocket encountered by CLU S-1. To identify any anomalous behavior, the reservoir's pressure vs. gas -in-place (total inventory) relationship has been monitored on a real-time basis since the commencement of storage operations. This type of plot is used in the gas storage industry to monitor reservoir integrity. By tracking this data on a real-time basis it is possible to detect a material loss of reservoir integrity. CLU S-3 was shut-in for most of the summer of 2012 so that wellhead pressure could be recorded for this purpose; thereafter it was shut-in periodically to confirm the pressure versus inventory trend remained consistent. Figure 1 is a plot of the actual wellhead pressure readings from CLU S-3 versus total inventory from April 1, 2012 through April 9, 2015 (again, excluding the 14.5 Bcf of native gas in the isolated pocket). This plot also includes the expected wellhead pressure versus inventory response based on CINGSA's initial storage operation design and computer modeling studies of the reservoir. The actual shut-in pressure of CLU S-3 initially aligned with simulated pressure from the modeling studies. However, at inventory levels above approximately 11 Bcf, the shut-in wellhead pressure on CLU S-3 has been consistently higher than expected when compared to predicted shut-in pressure derived from initial computer modeling studies. This sort of pressure response is not atypical of newly commissioned gas storage reservoirs and is often indicative of pressure transients that result from relatively high storage injection rates over a relatively short period of time, and not necessarily indicative of a lack of storage integrity. However, in this instance, CINGSA now believes the higher than expected pressure is due to the isolated pocket of native gas that CINGSA encountered when it initially perforated/completed the C I c sand interval in the CLU S-1 well. This gas has since commingled with gas in the depleted section of the Cannery Loop Sterling C Pool, occupies a portion of its storage capacity, and provides pressure support to the storage operation. That said, the overall trend of the wellhead shut-in pressure of CLU S-3 versus total inventory plot indicates there currently is no evidence of gas loss associated with storage operations, nor any other loss of well or reservoir integrity. Well Deliverabilitv Performance The CINGSA facility is equipped with a robust station control automation and data acquisition (SCADA) system, which includes the capability to monitor and record the pressure and flow rate of each well on a real time basis. Monitoring well deliverability is an important element of storage integrity management because a decline in well deliverability may be symptomatic of a loss of well integrity. Throughout the injection and withdrawal seasons the deliverability of each well was monitored via the SCADA system so that individual well flow performance could be tracked against past performance and the results of prior back -pressure tests performed on each well. CINGSA conducted field -wide deliverability tests of the CINGSA facility on March 9 and March 13, 2015. The objectives of these tests were to 1) assess withdrawal capability with all wells open under free flow conditions and with compression running, and 2) compare actual withdrawal performance to customer contract obligations at the current storage inventory. The working gas inventory at the start of the test on March 9 was approximately 4.6 Bcf. The aggregate customer contract Maximum Daily Withdrawal Quantity (MDWQ) obligation at this inventory is 103 mmcf/d. Thus, a key objective of the tests was to validate the capability of the facility to achieve and maintain a flow rate of at least 103 mmcf/d. A secondary objective was to compare actual flow rates and flowing pressures to those expected based on individual well tests conducted in 2012 (after re -perforating all five wells). A comparison of the actual field withdrawal rate and pressure during the free-flow and compression portions of the tests aligned closely with the expected withdrawal capability curve that was prepared at the end of the 2012-2013 withdrawal season. The combined maximum withdrawal rate from all five wells during the test was 117 mmcf/d, with CLU S- 1 contributing 55%, CLU S-2 about 21%, CLU S-3 about 10%, CLU S-4 about 12% and CLU S-5 about 2%. The test results effectively confirmed the field is capable of meeting the aggregate MDWQ obligations of CINGSA's customers at a working gas inventory of approximately 4.6 Bcf. While there is some evidence that the deliverability capability of CLU S-1 and CLU S-2 have improved slightly since the 2012 tests, there is also evidence that CLU S-3 and CLU S-5 have experienced some decline. That said, overall field deliverability appears consistent with the withdrawal performance capability at the end of the 2012-2013 withdrawal season, and there is no evidence to suggest a decline in deliverability performance that is a result of a loss of wellbore integrity. Three items from the tests are noteworthy based on discussions with station personnel. First, a total of approximately 80 barrels of fluid were recovered during the two tests; station personnel estimate that approximately 30 barrels were recovered during the March 9 test, and another 30 barrels were recovered during the March 13 test (total of 60 barrels). The remaining 20 barrels were recovered during the days in between the tests. Secondly, station personnel reported that the surface facilities (compression, dehydration, separation, etc.) generally performed as designed, albeit with some minor operational issues at the dehydration unit. Lastly, sonic surveys were shot on all five wells after the test on March 9. Results from those surveys indicate that CLU S-3 and CLU S-5 both appear to have a water column in the wellbore which extends above the perforations. CLU S-3 produced at withdrawal rates adequate to lift water from the reservoir during the test. Withdrawal rates from CLU S-5 were not adequate to lift water. The available operating data does not pinpoint the source of the water. Free water was recovered from drain taps in the header after the tests were completed; thus, it's likely that some portion of the recovered water was lying in the well header prior to the test and was simply swept into the separator when the flow rate increased. 2014 Infection Operations and October 2014 Shut-in Pressure Test Customer demand resulted in continued withdrawals from the field immediately after the April 2014 shut-in test. The remainder of April 2014, as well as May and June, continued to see both injections and withdrawals. Steady injections did not begin until July 2014 and extended into the beginning of October when operations again switched over to withdrawals. Total net injections during the summer 2014 season amounted to approximately 1,346,187 Mcf. During much of this time, injection rates ranged from 15 to 30 mmscf/d. On the morning of October 25, 2014, all of the wells were shut-in for pressure monitoring and remained shut-in until October 31. Total gas inventory at October 31, 2014 was 14,493,502 Mcf, which included 7,493,502 Mcf of customer working gas plus 7,000,000 Mcf of CINGSA base gas. Table 2 lists the wellhead shut-in pressure for all five wells each day during the shut-in period. It also lists the day-to-day decline in pressure and the overall weighted average pressure of all five wells. On the final day of shut-in, wellhead pressures ranged from a high of 1,478 psig on CLU S-1 to a low of 1,438 psig on CLU S-3. The corresponding calculated bottom hole, or reservoir pressure, for these two wells is 1,677 psia and 1,631 psia, respectively. It is clear from reviewing this data that wellhead pressure had not fully stabilized during the week-long shut-in; shut-in pressure on all five wells declined continuously during the period. On the final day of shut-in, field average pressure was still declining at a rate of approximately I psi/day. Figure 2 is a plot of the shut-in wellhead pressure of each of the five wells and the weighted average wellhead pressure for all five wells. The overall average wellhead pressure on October 31 was 1,465 psig and the average reservoir pressure was 1,662 psia. 2014-15 Withdrawal Operations and Aaril 2015 Shut-in Pressure Test Storage withdrawals from the field commenced on November 1, and were largely continuous through the remainder of the month and through the end of March. Withdrawals from storage during the entire 2014-2015 winter period amounted to approximately 3,370,213 Mcf. Field Operations reported that approximately 830 barrels of water was produced during the withdrawal season. The field was shut-in for pressure stabilization and monitoring on the morning of April 1. As noted previously, the withdrawal performance of CLU S-3 and CLU S-5 were below past performance, particularly that of CLU S-5. Sonic surveys after the withdrawal test in early March indicated that CLU S-5 had a fluid column that extended several hundred feet above the perforations. Based on a review of individual well withdrawal data, it appears that CLU S-5 may have had a fluid column since the end of the prior withdrawal season; the well contributed only minimally to overall withdrawals during the 2014-15 withdrawal season. A bottom hole pressure survey was run in this well on April 16. The fluid column was approximately 880 feet above the perforations, which accounts for the well's low contribution to withdrawals this past season. A coiled tubing clean out of the well will be performed later this year when reservoir pressure is higher and can assist with the clean out process. Total gas inventory at April 1, was 11,132,248 Mcf, which included 4,132,248 Mef of customer working gas and 7,000,000 Mcf of CINGSA-owned base gas. Table 3 lists the wellhead shut-in pressure for all five wells each day during the shut-in period. It also lists the day-to-day change in pressure and the overall weighted average pressure of all five wells. On the final day of shut-in, wellhead pressures ranged from a high of 1,224 psig on CLU S-5 to a low of 1,126 psig on CLU S-1 (note that the shut-in wellhead pressure on CLU S-5 has been adjusted to account for the large fluid column in the wellbore during the shut-in period), and field average pressure was still increasing at a rate of approximately 1 psi/day. It is clear that wellhead pressure had not fully stabilized during the week-long shut- in; shut-in pressure on Wells CLU S-1, S-2, S-3, and S-4 increased continuously during the period. Wellhead pressure on CLU S-5 was flat throughout the shut-in period because the well had watered -off. This well remained shut-in until April 16, when a bottom hole pressure survey was run to detennine the actual fluid level in the well. The reservoir pressure on CLU S-5 on that date was used to calculate a wellhead pressure assuming a full gas column in the well. This corrected wellhead pressure was then used in the field average pressure calculation. Figure 3 is a plot of the shut-in wellhead pressure of each of the five wells and the weighted average wellhead pressure data for all five wells. The overall average wellhead pressure on April 8 was 1,160 psig and the average reservoir pressure was 1,316 psia. Table 4 provides a summary of the surface and reservoir pressure conditions and the total gas -in-place at the time the reservoir was discovered. It also lists the same data for the six shut-in periods since commencement of storage operations. Lastly, it lists the gas specific gravity, the percentage of nitrogen and carbon dioxide contained in the storage gas, reservoir datum depth, and reservoir temperature. Figure 4 is a plot of the average bottom hole pressure adjusted for compressibility (BHP/Z) versus gas -in-place at November 8, 2012, April 15, 2013, November 4, 2013, April 9, 2014, October 31, 2014, and April 8, 2015 compared to the original (discovery pressure) conditions. Linear regression analysis of these six data points indicates there is a very strong correlation between the six points; the regression coefficient (Rz) is 0.9499. Thus, similar to Figure 1, Figure 4 strongly supports the conclusion that reservoir integrity is intact. The key point to note is that the observed BHP/Z values for all six of the shut-in periods (November 2012, April 2013, November 2013, April 2014, October 2014, and April 2015) are above the original pressure -depletion line which provides very compelling evidence that integrity is intact and the reservoir and wells are not losing gas. Preliminary Estimate of Additional Native Gas Volume CINGSA believes it encountered an isolated pocket of native gas which was possibly still at native discovery pressure when CLU S-1 was initially perforated/ completed. Wellhead pressure on the CLU S-1 well rose to approximately 1,600 psi within a few days after completion, while wellhead pressure on the remaining four wells was approximately 400 psi, which was in line with expectations. The C I c sand interval is one of five recognized sand intervals that are common to nearly all of the wells that penetrate the Cannery Loop Sterling C Pool. This particular sand interval was also one of the perforated/completed intervals in the CLU -6 well - the sole producing well during primary depletion of the Cannery Loop Sterling C Pool. Following initial perforation/completion, a temperature log was subsequently run in CLU S- 1 in an effort to identify the nature and source of the higher pressure. The temperature log exhibited strong evidence of gas influx from the sand interval which correlates to the Sterling C I c sand interval. The higher than expected shut-in pressure and evidence of gas influx strongly suggest the Clc was indeed physically isolated from the other four sand sub- intervals within the Sterling C Pool. It is unknown whether the C 1 c sand interval was at native pressure (2200 psi) at the time CLU S-1 was completed, or if it had been only partially depleted. If fully isolated from the pressure -depleted section of the reservoir, completion of the C 1 c effectively adds to the remaining native gas in the reservoir. This additional gas also accounts for the weighted average reservoir pressure during each of the six field -wide shut-in pressure tests plotting above the original BHP/Z versus gas -in-place line. This previously isolated pocket of native gas provides pressure support to the storage operation and effectively functions as additional base gas. Two independent methods are being used by CINGSA to estimate the volume of incremental native gas encountered by the CLU S-1 well. The first method is based on a material balance analysis which was performed using the shut-in reservoir pressure data gathered during November 2012, April 2013, November 2013, April 2014, October 2014, and April 2015, together with observed shut-in pressures from CLU S-3 to estimate the magnitude of additional native gas encountered in the C I c sand interval of CLU S-1. The approach used to analyze this data was to treat the originally defined reservoir and the previously isolated C 1 c sand interval as two separate reservoirs that became connected during perforation/completion work on the CLU S-1 well. A simultaneous material balance calculation on each reservoir was made in which communication is allowed between reservoirs after completion of CLU S-1 in late January 2012. Gas was allowed to migrate between the reservoirs. The connection between the reservoirs was computed by defining a transfer coefficient which, when multiplied by the difference of pressure -squared between the two reservoirs, results in an estimated gas transfer rate. In other words, storage gas is injected and withdrawn from the original reservoir and is supplemented by gas moving from or to the C I c interval according to the pressures computed in each reservoir at any given time. The volume of the original reservoir was well defined from the primary production data as having an initial gas -in-place of 26.5 Bcf. The volume of gas associated with the C lc sand interval in CLU S-1 and the transfer coefficient was varied to match the observed pressure history using a day-by-day dual reservoir material balance calculation. Figure 5 summarizes the results of the material balance procedure for the C 1 c sand interval having 14.5 Bcf of original gas -in-place at initial reservoir pressure conditions. Figure 6 illustrates the daily transfer rate between the main reservoir and the isolated pocket and the estimated cumulative net transfer of gas since commencing storage operations. The initial transfer rate was 43 mmcf/d. Thereafter the transfer rate has been a function of the pressure difference between the two reservoirs. Various combinations of C 1 c sand volume and transfer coefficients were explored. A range of C lc sand volumes from 14 Bcf to 16 Bcf gave reasonable solutions and can be considered a reasonable range of uncertainty. Given the early stage of storage operations, the value of 14.5 Bcf is the most reasonable estimate at this time. As additional data is obtained, particularly after a significant withdrawal season, this value can be more confidently determined. The second method used by CINGSA to estimate the volume of incremental native gas encountered by CLU S-1 is a three-dimensional computer reservoir simulation model. The modeling effort utilized an existing reservoir description/geologic model which was updated after the drilling and completion of the five injection/withdrawal wells. Thus, the current model incorporates all available well control data and petrophysical data from electric line well logs. Seismic data was also used to characterize channel boundaries and differentiate possible reservoir versus non -reservoir rock. A history match was then run which spans the operating history of the reservoir, including the entire primary production period and extending through October 2014. A simulation input file was constructed with actual (observed) daily flow from each well, including the CLU -6 well during primary production. The objective was to achieve an acceptable match between the observed flowing and shut-in wellhead pressures and the pressure predicted by the reservoir model. Emphasis was placed on matching the observed pressures during primary depletion, and pressures from October 2012 and beyond (after all five storage wells had been re -perforated and after cleaning up during initial withdrawals). An acceptable match is considered to be when the difference between actual pressures versus predicted pressure is less than 50 psi. Several simulation runs were made using various assumptions concerning reservoir configuration—i.e., channel geometry versus a "layer cake" configuration, aquifer support versus no aquifer support. Initial efforts focused on modifying wellbore skin factors and adjustments to grid block transmissibility to achieve an acceptable match to observed pressures. These efforts were largely unsuccessful because they required what were considered extreme assumptions for skin factor values and/or transmissibility assumptions that did not honor the basic petrophysical data. It was discovered early in the modeling process that some form of external pressure support was necessary to achieve an acceptable history match. Several attempts to provide support via an analytical aquifer yielded unacceptably high rates of water production that did not match historical operating data. A reasonably acceptable history match was ultimately achieved only when additional pore volume outside of the channel boundaries (but within CINGSA's approved storage boundary) was incorporated into the model adjacent to CLU S-1. The match between observed pressure and production data and that computed by the reservoir model was very good on CLU S-2 and CLU S-4, and reasonably good on CLU S-1, but not quite as good on CLU S-3 and CLU S-5. The estimated volume of incremental gas that yielded the best history match was 18 Bcf. The modeling effort thus far has resulted in a reasonably acceptable match between actual observed pressures and pressures predicted by the model. As noted above, the current modeling effort includes operating history through October 2014. CINGSA intends to resume the modeling process in the very near future as it completes work related to re- processing seismic data from 1995 which may reveal more detail concerning the location of the isolated reservoir and its areal extent. Once the modeling effort is resumed, key objectives include achieving a better match between observed and simulated pressures on CLU S-3 and CLU S-5, and to a lesser extent CLU S-1. In addition, it may be possible to more fully characterize the volume of incremental gas associated with the CIc sand interval that was encountered when CLU S-1 was initially perforated/completed. Annulus Pressure Monitorina Prior to CINGSA commencing storage operations, all of the Marathon Alaska Production Company (now Hilcorp Alaska LLC) wells that penetrate the Sterling C Gas Storage Pool were subjected to AOGCC-mandated Mechanical Integrity Tests (MIT), and all of the wells successfully demonstrated integrity. Shortly after commencing storage operations, all of the CINGSA wells were also subjected to MITs, and they likewise demonstrated integrity. CINGSA monitors and records both the tubing/intermediate casing string annulus (7" x 9 5/8") and intermediate/surface casing string annulus (9 5/8" x 13 3/8") pressure of each of its wells on daily basis to identify any evidence of loss of well or reservoir integrity. In addition, Hilcorp monitors and records pressure on each of the annular spaces of its production wells which penetrate the Sterling C, as well as pressure on the tubing string in certain wells. Hilcorp provides a copy of this data to CINGSA monthly and CINGSA analyzes this pressure data for any evidence of a loss of well/reservoir integrity, in the same manner as it does for its own wells. All of these annulus pressure readings are submitted to the AOGCC monthly and are part of routine and ongoing surveillance to ensure the integrity of the storage operation. Figures 7-11 illustrate the historical tubing and annulus pressures on each of the CINGSA gas storage wells. The observed inner and outer annulus pressures on all of the CINGSA storage wells track the tubing pressure. The inner annulus (7" x 9 5/8") of all five wells is filled with brine and diesel, while the outer annulus (9/58" x 13 3/8") is filled with cement, largely to surface. Thus, a more pronounced pressure swing is observed on the inner annulus than the outer. In both cases, the pressure swing is due entirely to expansion of the 7" casing string which results from higher pressure and temperature when injections are occurring. The key point for all five wells is that the pressure of the tubing and annulus are never equal, which demonstrates wellbore integrity. Figures 12-22 illustrate similar data on each of the Hilcorp wells that penetrate the Sterling C gas storage pool. With the exception of CLU -6, all of the annulus and tubing pressure readings on the Hilcorp wells are very low (below 200 psi). The CLU -6 well was originally the sole production well associated with the Sterling C Pool. The Sterling C Pool was plugged prior to CINGSA commencing storage operations and the plug was pressure tested to AOGCC specifications at that time. Shortly thereafter, the well was recompleted in the upper (shallower) Sterling Sands. Thus, pressure on CLU -6 was significantly higher than the other Hilcorp wells because it was re -completed in the upper Sterling Sands and its tubing pressure is reflective of native (discovery pressure) conditions associated with this strata. Since its recompletion, pressure on the CLU -6 has declined to near zero in early 2013 and it is clear the well is incapable of producing in its current state. Since pressure on this well is now well below any of the CINGSA wells and is not tracking the operating pressure of the CINGSA wells, there is no evidence of a loss of integrity. For the remaining Hilcorp wells, all of the pressure readings are well below tubing pressure of any of the CINGSA wells and do not track the CINGSA well tubing pressure trends, which again demonstrates isolation/integrity. Thus, based on a thorough review of the annular pressure data for all wells, there is no evidence of any loss of integrity of any of the CINGSA injection/withdrawal wells or any of the Hilcorp wells which penetrate the Sterling C Pool. This data lends additional support to the conclusion that reservoir integrity is intact and all of the storage gas remains within the reservoir, and is thus accounted for. Summary and Conclusion CINGSA commenced storage operations on April 1, 2012 and has now completed three full years of storage operations. All of the operating data associated with the CINGSA facility indicate that reservoir integrity is intact. The observed pressure vs. inventory trend is consistent with modeling studies of the reservoir prior to placing the facility in service, although wellhead shut-in pressure on CLU S-3 has trended above the stabilized pressure line developed from initial computer modeling studies of the reservoir Overall field deliverability appears unchanged from the 2012-2013 initial storage cycle. Field -wide withdrawal tests were conducted on March 9 and March 13, 2015 and a maximum stabilized withdrawal rate of 117 mmscf/d was achieved during the tests. Results of both tests confirm the facility is capable of meeting the aggregate contract MDWQ obligation of 103 mmcf/d at a working gas inventory level of 4.6 Bcf. While CLU S-5 and CLU S-3 appear to have experienced some performance decline during the 2014-15 withdrawal period, it appears the cause of that decline is fluid in the wellbore, and possibly sand in the case of CLU S-5. During the bottom hole pressure survey of CLU S-5, the tools tagged what appears to be sand fill at a measured depth of 9280 feet. If that is in fact the case, over 80 percent of the perforations may be covered with sand fill. A foam clean out of CLU S-5 will likely be scheduled for later this year. CLU S-3 will be placed on injection to see if the fluid can simply be displaced back into the storage formation. The injection/withdrawal capability of wells CLU S-1 and CLU S-2 appear to have improved somewhat over time and effectively offset the decline in performance of wells CLU S-3 and CLU S-5. There is no evidence of a change in deliverability in any of the CINGSA storage wells that may indicate a loss of well integrity. There is evidence indicating that initial completion work on CLU S-1 encountered an isolated pocket of native gas within the Sterling C I c sand interval. This gas has since commingled with gas in the main (depleted) portion of the reservoir, effectively adding to the remaining native gas reserves and providing pressure support to the storage operation. This additional gas is functioning as base gas and accounts for the higher than expected shut-in wellhead pressure readings on CLU S-3 and the field -wide shut-in pressures observed during each of the six shut-in periods. Two methods were used to estimate the volume of incremental native gas encountered by CLU S-1. The two methods yielded volumes that range from 14 to 18 Bcf. The range of this estimate will very likely narrow with additional field -wide shut-in tests. That said, field weighted -average shut-in pressure during the November 2012, April 2013, November 2013, April 2014, October 2014, and April 2015 shutdowns exhibit a very strong linear correlation (Rz = 0.9499). Thus, the results of these six shut-in pressure tests support the conclusion that no loss of gas from the reservoir is occurring, and that all of the injected gas remains within the storage reservoir. Finally, annulus pressure readings on all of the CINGSA wells and the Hilcorp production wells which penetrate the Sterling C Gas Storage Pool demonstrate the confinement of gas to the storage reservoir. No anomalous pressure increases have been observed on any of the annular spaces associated with the CINGSA or Hilcorp wells, nor are any of these same wells exhibiting annular pressure readings that match the injection tubing pressure on any of the CINGSA wells. Thus, there is no evidence at this time of any loss of integrity based on annulus pressure readings. Accordingly, all operating data indicate that reservoir integrity remains intact, and although the reservoir may now be effectively larger than expected due to encountering additional native gas in the Sterling C I c interval of the CLU S-1 well, all of the injected gas remains with the greater reservoir and is accounted for at this time. Table 1 -Monthly Injection and Withdrawal Activity Cook Inlet Natural Gas Storage Alaska Storage Monthly Storage Activity (all volumes reported are at month end unless noted otherwise) Month Infections -Mcf Withdrawals - MA Compressor Fuel&lasses Total gas in Storage -Mtf Mar -12 0 0 3,556,165 Apr -12 146,132 394 2,289 3,699,614 May -12 1,238,733 11163 11,540 4,925,644 Jun -12 1,245,041 1,048 16,769 6,152,868 Jul -12 986,472 714 12,529 7,126,097 Aug -12 1,245,260 93 14,038 8,357,226 Sep -12 1,300,153 982 13,221 9,643,176 Oct -12 1,624,167 691 15,285 11,251,367 Nov -12 165,866 72,417 4,895 11,339,921 Dec -12 379,205 470,886 5,839 11,242,401 Jan -13 496,560 209,334 7,976 11,521,651 Feb -13 1,765,296 858 19,372 13,266,717 Mar -13 667,603 554,597 7,594 13,372,129 Apr -13 438,717 254,734 6,315 13,549,797 May -13 509,694 12,769 7,680 14,039,042 Jun -13 615,458 1,274 11,185 14,642,041 Jul -13 468,599 822 12,118 15,097,700 Aug -13 499,748 3,392 11,766 15,582,290 Sep -13 306,323 16,743 9,074 15,862,796 Oct -13 530,289 27,585 10,287 16,355,213 Nov -13 9,608 902,874 214 15,461,733 Det -13 5 1,156,534 61 14,305,143 Jan -14 261,325 127,655 7,352 14,431,461 Feb -14 4,143 517,894 534 13,917,186 Mar -14 1 766,800 - 13,150,387 Apr -14 97,548 190,563 3,671 13,053,701 May -14 64,435 388,647 1,597 12,727,892 Jun -14 509A45 502,790 7,444 12,727,103 Jul -14 687,386 108,786 11,165 13,294,538 Aug -24 728,130 219 12,423 14,010,026 Sep -24 537,858 4,705 11,712 14,531,467 OR -14 155,673 189,157 4A77 14,493,506 Nov -14 66,645 291,368 2,126 14,266,657 Dec -14 32,716 380,170 1,897 13,917,306 Jan -15 - 1,104,457 76 12,812,773 Feb -15 - 971,590 288 11,840,895 Mar -15 11,253 719,045 855 11,132,248 Table 2 - November 2014 Wellhead Shut-in Pressure Data Wellhead shut -In ermwnloslel and Dates a.. s- IYeIILEme me.. Mwu-9.n IDB56Gta 10.1.19 1471.1 le/Et/J019 1MlY lafWIDt9 un.a leQ9♦m19 u]As ]p14M.2 la]9.z 1V.11/ID19 un.s ttusa W 4xU57.M lan9 =9 lovas wn.1 lm<a 1na1 un.] ..M4 sa 17..11 14MA 14n.x .4 u39.3 ta39.z 1i39.o wsa 97n11 11]3.1 11>x.s ..a 1a W.2 Nw, 1165.1 s5 1169.1 1147.6 .1 1W.s 1.. 1161.1 ]9Wx .1 11593 lases 3jyj5 333.111 eWYtl Aw. elae�we9� 11129 11]0.5 ]16&9 1K]A IYd.S IMS9 1K5.] w•a.ap P9eL . 2 .z.14 Paa3.mDm1 Pav4n. v wV1 a.1 PavS.m0av9 a.3 Oaten.➢9rS 07 GErLa.W1§ a.a waLLtlsO( s.I GaY2ia.GW .1s 090WaU -LI 41 A6 Paysss.anl4 L9 Paxa n, a33 GaY)_w a7 <w&x .La 69 Ls La a.z 1 cwsa -1.3-1.3ae ov oz a z . a.x cwsa -3 d1 as -1 11v ae cwss s a3 -Q.,s a.6 n1 a9 a.e We Mia2o�'�bxMan Pa Eaalwadll MWeI Table 3 - April 2015 Wellhead Shut-in Pressure Data ahmammM„ ISNraeeemeAM- WeStbma Ip_-, .n US1 1124.3 yIA3 vIMB 11166 11183 3LIM,i N$Qpj 11295 9ILD., 1131.4 1I.e. 122.1 41.1.7 133.1 4TOM U13 2976 IMM 11364 11381 11791 193.4 113$.1 11333 USd 1166.5 118.9 11]5.5 11N5 1]83.3 11825 1188.5 W011 U56 9]011 3.3 1114.9 11183 11713 0.5 1171.3 1171.8 MIS U53 111,1 1$$3.9 13339 1$339 13$39 .21.9 V$3.9 11139 tY139 3j,jjj 333.171 14,493.502 4/8/2015 1159.6 1315.8 0.877 1500.3 ,ow am. amp W.) 1149.5 1151.1 119.5 IISS3 1152.3 115&1 11H.6 NUE: W RM reMehu tare MJ wel1med preuuee ne de, 6-1 lluq In the uellWee. Useh4atd,hutlnpensurtDnly. Pa192. D.a1 3.3 We'rFtMBweawate Paint 18 Wv.4n.6y3 Le D Dneaa4 pas n.pn4 19 D,n6 Wxfin.mr5 IS Wy2n.a 9H PavS 2.5 3.5 97x31., �P1r41.8 1.9 un 1.H loaeaa.wrAenael Wv£ 2.2 33 PaL9.L 1.3 PaL L7 11 1.1 1.1 1.6 1.2 0.1 L, 64 26 4 tE 3.6 3 3.3 3.1 16,339.046 2:8 0.5 0 0 D D 0 0 0 D Table 4 - Shut-in Reservoir Pressure History and Gas- in -Place Summary Wenxeaa Presse mug. Bottom Rala pressure Plla 10/28/2000 1950 2206 msrnvervl Butternut, cnnenbns Z -Factor Blapa -¢am 0 0.8465 2606 Toul Gash, plare- maps6 0 26,500 Qelt WeleRtea Aw. WelTeaa Presmre Bale, Sa mated Bmwm Vreatere - ,ale Ro e Stoma, oaeraNna B - Fen. canoe m Byp(j.: iai Sebleasin Vlue-mmsd 11/0/2012 1269.9 1434.9 0.8719 16453 11,223.715 4/15/2023 1344.4 1522.35 0.8668 1756.3 13,106.807 11/4/2013 1580.7 1798.1 0.8508 2113.4 16,339.046 4/0/2014 1320.6 1497.7 08662 1729.0 13,147.315 10/31/2014 1465.1 1662.3 0.858 1937.4 14,493.502 4/8/2015 1159.6 1315.8 0.877 1500.3 11,123.289 Gas Gravity: D.56 N2 Cont, 0.3% CO2 Con, 0.3% Reservoir Temp (de& F): 105 Datum Depth (ft.): 4950 Figure 1— CLU S-3 Wellhead Pressure versus Inventory 2000.0 1800.0 1800.0 1400.0 a v w 1200.0 m a 1000.0 u 800.0 N 800.0 400.0 200.0 0.0 CINGSA I Wellhead Pressure vs. Inventory Hysteresis (Original Reservoir Only) 5,000 10,000 15,000 20,000 25,000 Total Field Inventory, mmscl —InNal Cycle Deelgn —Second UYce Design �$Minted Wellhead Pressure Design • ACLal Shut-in Pressure vs. 1maMwy-CLOS-3 Piessun • Fal 2012 WASIWHP • Spring 2013 WASI WHP • F612013WASIWHP o Spdng 2014 WASIWHP • Fa12014WASIWHP • SpMg 2015 WASIMP air 5,000 10,000 15,000 20,000 25,000 Total Field Inventory, mmscl Figure 2 — November 2014 Wellhead Shut-in Pressures CINGSP Gall 2014 Wellhead ShU 4n Pressures IIS0.0 1460.0. t4r0u a �� cwudonrl C �auvx.v+ IeGOn x nuu�ro..s S ti—Pi h Wx ,.M 4vg. Pres.. 14300. _ IW25 IW26 10/27 10128 1Wd0 10/30 20/31 ShOtrin M. Figure 3— April 2015 Wellhead Shut-in Pressures CINGSA Spring 2015 Wellhead Shut -In Preuures 11100 1110.0 - vm0 uu0.0 o119 0 — �_y cluelnro•1 —r.ausv.g.] 11800 au _ Slnyee P. 11100 .�OW".., 311600 _ �. + field Welgh140 Ae,. Puss. 15 1500 114U.0 - - 1130.0 1120.0 - 11100 4/2 4/3 4/4 4/5 4/6 4/2 4/8 Shemin 04fe Figure 4 — Material Balance Plot Cannery Loop Sterling C Gas Storage Pool -Material Balance Plot November2012 April 2015 ljCb oixwery XHV/E- ]606PW 2.50D j ro112013 BHP/2-2113.Apg4 _N B L.WO \D(.X8/\4.3pSli ti- ` m LOId tlHP [-1/tY.0 is � ra112016NIP/E-19lE.A DSu Fall EOli BNPR- 16A5)pu R 1.V10 p a` 0 Spring MIS BHP/)-ISpt.3psu i E -+plKOvlvy PHP(2vS GM in-flf[e O mI,fNXI j ♦ Na IDt/XNv//v5,fawYn lNxr / ♦ SpinB 2013 IOWA vl. Gas in Rxe / • 'A 20130HP/Z n, Cul hl Rxe A Spring 2014 BIIP/L vs. Gas lD Pla[e 5Lb j/ fan M14 in Wc• Sprilg7m5HXP/2u G,rr,Pd[r 0a 0 5,000 111=, 15" 2a" 25," Moos Total Gas -i. Pl.. MW Figure 5 - Historical and Computed Pressures vs. Rate Figure 5 - Historical and Computed Pressures vs. Rate 2300 2100 Iwu> 1100 4'000 uala HIM loc 500 ;m I'MIXI I'M ♦\�0 `\,y o\,y1 al~M1 ��l \�,� °j\ti, v�y �Y, 6y�e 9�ve �> ��e yyh ��.sn ow oaly IN/wOM RXr rr.YIjo 'kWSHIP Pva "(ahs" P59 0"!IMSIPHP Arg psla" Figure 6 - Estimated Gas Transfer to/from Original Reservoir ta.w tlU IV Fom Figure 6 - Estimated Gas Transfer to/from Original Reservoir w woo rxn 9" LEE N`ria E v 148 iUON 0 of �, bt'� y(� a'1' .4\� b`L Nf4 'h�ao b`•' N�$ Vit\• �'yP 14 Date —Dalry irpWdtl nate mmr.7a 111" rIM, einx9c Net Uasiransferted,wtcf 5 5 6 z Figure 7 — Annulus Pressure of CLU Storaee —1 Plot ofhbing and Muulus Pressure vs Time -CLU SI 2000 1600 1400 m 1200 1m0 L v a 6m 6O 400 200 0 a \�h � ^A`\¢+m,\�M1 \4 �0�1 \p9�3 p"o "o pJ�� 4�^a J�3Op OpJ�v\�a 4vn its APs RPh v\�h D°'\ p0\ — Figure 8 — Annulus Pressure of CLU Storage — 2 Plot of Tubing and Annulus Pressure .s Timc -CLU S-2 20M —95/8 Ann ulue 1AM —133/84nnmos —r mur 1800 \4M — S 1200 n P IMO 6 800 600 400 200 0 O O O^ ♦ O O On �Q Off' ♦ 'UO O'110 O1' �' Figure 9 — Annulus Pressure of CLU Storage — 3 Plot of Tubing and Annulus Pressure vs Time - CLU S-3 2000 —95/n Pnnulu> 1 1800 33/MMmmm —iuGne IBM rano _m 1200 1000 N 800 400 0 1�M1 115 pol " MP3 1'9 o" �p'N" �p�i`n ��p,�Nl �,pJ'P 'OM 4,M o �n 00fl ndn� "o d O"" do no "o 10 6✓ a dt 0^ on "o don Fieure 10 — Annulus Pressure of CLU Storaee — 4 Plot ofrobing and Annulus Pressure rs Time -CLU S4 2000 —"18G nulnv 1800 —13 Mk,m We —mune 1600 1400 ,o 1200 n 1000 i 800 600 400 200 - - u Op -p�\�0' �,4 'INS "PO' 9NA ��p,,5 �q'\NA ��p\fin' \0 \NS O�W 0 ��p4�W �p�'0 'es" ,"Is P.,9 ��p,\�O 049 C, s.\. `'o ONO ow. O do, �� "0 O'lQ Oe d+ dT 'es op de "s t�' Fieure 11 — Annulus Pressure of CLU Storaee — 5 Fieure 12 — Annulus Pressure of Marathon CLU RD -1 CLU 1RD Annulus Pressure History 90 80 41/2 x 7 m 'a 70 �7x 95/8 w 60 a 50 a 40 V 0 30 ^ 20 10 0 11 .1ti ,1ti ,1'' 1'' .1p 1A 15 15 SeQ < SeQ" fat P�� Feb" Pig Month/Year Fieure 13 — Annulus Pressure of Marathon CLU 3 CLU 3 Annulus Pressure History 600 m 500 —3 1/2 x 9 5/8 N Q v 400 a 300 a V t 200 Vf 100 0 11 SeP" �a< SeP �a< SeQ" mat P�� Fee" Pig Month/Year Figure 14 — Annulus Pressure of Marathon CLU 4 CLU 4 Annulus Pressure History 12 A 10 +31/2x135/8 N n �-13 5/8 z 20 v g N d a` 6 a V A i 4 N 2 0 SeQ11 �a11'L Sep,1'L C�a1N$ Sep,13 �a<"10. PJg10. �e�15 PJ�15 Month/Year Figure 15 — Annulus Pressure of Marathon CLU 5 CLU 5 Annulus Pressure History 250 00 200 31/2 x 9 5/8 'N � 9 5/8 x 13 3/8 n y 150 N w 100 a` v m 50 3 N -50 11 ,1ti 1ti ,13 13 ,10. 10. 1y 15 Noy""Noy" �aV Off' Pppd' Month/Year Figure 16 — Annulus Pressure of Marathon CLU 6 CLU 6 Annulus Pressure History 2000 1800 —41/2 tbg •a 1600 41/2x7 y 1400 N 1200 a a` 1000 a w 800 600 400 200 0 11 ,1ti 1ti 1''13 ;1a 1h 15 15 Noy" �xaV ao�" �xaV �o�" 00pct pp�" pct' Month/Year Figure 17 — Annulus Pressure of Marathon CLU 7 CLU 7 Annulus Pressure History 70 60 .3 1/2 x 9 5/8 N Q 50 �95/8x 133/8 d 3 w 40 6 u 30 m 't 'n 20 10 0 \'04,11 Cray"12 ,1ti day 19 CtOJ"13 0a" 1A Oc�1G Ppt.15 O 1y Month/Year Figure 18 —Annulus Pressure of Marathon CLU 8 CLU 8 Annulus Pressure History 120 31/2 x 9 5/8 ,a 100 .a 9 5/8 x 13 3/8 m 80 N a 60 a V m 40 N 20 0 11 ,1ti 1ti .13 1'� .1b 1a .15 15 Noy" OSA �o�.0" �o�" �aV Oct ppt Oct' Month/Year Fi¢ure 19 — Annulus Pressure of Marathon CLU 9 CLU 9 Annulus Pressure History 180 160 m 140 a v 120 '^ 100 OV; -31/2x95/8 80 9 5/8 x 13 3/8 m 60 u' 40 20 0 0, 'S!, PQt;yS o`�15 Month/Year Figure 20 — Annulus Pressure of Marathon CLU 10 CLU 10 Annulus Pressure History --r-31/2x95/8 60 - 9 5/8x 13 3/8 m 50 .N a a 40 N d a` 30 a w 3 20 N 10 ji 0 C`04' OZW"1L �0�"1� �aV"1� Noy 13 �xaV,16 O�,1a Pps,1`' 0 15 Month/Year Fieure 21 — Annulus Pressure of Marathon CLU 11 CLU 11 Annulus Pressure History 100 90 °p 80 70 a � 60 N C v 50 a 40 a u f0 30 —31/2 z95/B 20 95/8 x 133/8 10 0 �0�.11 �aV.1ti �04.1�' �aV �'' to�,13 �aV 10, O,1a Ppm 15 0.15 Month/Year Fieure 22 — Annulus Pressure of Marathon CLU 12 CLU 12 Annulus Pressure History 30 m jinside N a u 20 N d a` y u 10 ut 0 �o� S, �aV 1ti �o� 1ti �aV 13 Noy 13 �aV 1A Occ D, Pp� 15 Oct 15 Month/Year