IEA Greenhouse Gas R&D Programme 34 Well Bore Integrity Network Meeting 12 & 13 March 2007 Santa Fe, New Mexico “A Review of Injection Well Mechanical Integrity Testing Data and Implications for Geosequestration” Jonathan Koplos, Chi Ho Sham, and Shari Ring The Cadmus Group, Inc. 13 March 2007 ™ Santa Fe, New MexicoStudy Objective * Draw lessons for CO, injection and geosequestration from the Underground Injection Control (UIC) Program’s mechanical integrity tests (MITs) of well performance * Assess the availability and adequacy of data on mechanical integrity testing 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NMOutline * Introduction — Mechanical Integrity Testing (MIT) and CO, injection * Underground Injection Control (UIC) Program -- Class I and II wells * Sources and availability/adequacy of MIT information * MIT failure rates, types, consequences from existing studies * MIT failure rates from Texas UIC Class Il EOR wells * Conclusions and Discussion Points 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NMInjection Well Mechanical Integrity An injection well has Mechanical Integrity (MI) if there is: —no leakage in the casing, tubing, or packer (internal MI) —no fluid movement vertically along the outside of the wellbore (external MI) 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NMImportance of MI for CO, Injection ¢ CO, corrosivity can affect well tubing, packers, casing, and cement — “Leakage from the injection well...is one of the most significant potential failure modes for injection projects” (IPCC, 2005) — Existence of IEA Well Bore Integrity Network (!) * Corrosive CO, requires design, operation, and risk considerations regarding the mechanical integrity of CO, injection wells 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NMUIC Program ¢ Maintaining MI is the protective cornerstone of the UIC Program * UIC objective: prevent endangerment to Underground Sources of Drinking Water (USDWs) * Since the 1980s, the UIC Program has regulated injection of fluids into the subsurface 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NMUIC Class I and II Wells ¢ Regulations define 5 classes of injection wells * Class I and II wells are CO, injection analogues — Class I - long-term storage, high volumes injected * Class I Hazardous Waste Injection Wells (no-migration) + Florida Class I Municipal Wells (buoyant, treated water) — Class II - oil & gas production-related fluids + Enhanced Oil/Gas Recovery (EOR/EGR) wells (CO,) + Acid Gas Injection (AGI) wells (CO, + H,S) 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NM 7UIC Requirements * Injection wells are subject to requirements for siting, construction, operation, monitoring, MITs, and closure (plugging) * Class I well requirements are more stringent, especially for wells that inject hazardous waste * Class II well requirements can be more flexible and varied; SDWA Section 1425 requires that State programs be “effective preventing underground injection which endangers drinking water sources” * Class I and II wells are required to formally demonstrate MI prior to and throughout injection operations 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NM 8Class I and II Injection Well MIT and Monitoring Requirements Well Class Mec! ical Integrity Tests Part I (Internal M1) Part II (External MI) Other Tests Reporting Frequency Pressure test Temperature, noise Radioactive tracer, pressure fall-off Annually Continuous injection press. flow rate, volume, temp, Ciase Ma [real 7arel ftir : 7 Sess Hazardous after each Jog at least every 5 Ce eae + fluid chemistry + GW gorse Quarterly reporting Continuous 2 Ro Pressure test or Temperature, noise ee eee eee ines I | sierantre actif epeecra Pressure fal-offtest | HOW ae, volume and annulus Ne wtentone | gat nto 5 aie eee oe Hazardous | © ao eam = Quarterly reporting eam ad ceeay Hisinpe aie initially and | Adequate coment Areal sd cumalative vokine, Class |levery's sexe | records may bo wed Goamrvod weekly for ea iets Orta eae ented cel |e Del acaty cial a Seep |oeeie 13 March 2007, UIC/MITT TEA Well Bore Integrity Network Meeting ~ Santa Fe, NM. 9EPA Approved MITs Tests for internal MI (40 CFR § 146.8(b)): * annulus pressure or annulus monitoring test + radioactive tracer test + water-brine interface test + _ pressure test with liquid or gas + monitoring records showing the absence of signific between pressure and injection flow rate (certain C! it changes in the relationship s II wells only) Tests for external MI (40 CFR § 146.8(c)): * temperature log + noise log * oxygen-activation log indicating lack of fluid migration behind the casing + radioactive tracer survey indicating lack of fluid migration behind the casing + cement bond log showing gamma ray, transit time, collar locator and variable density log * cementing records (in lieu of any tests or logs) that demonstrate the presence of adequate cement to prevent migration of fluids into a USDW (Class II wells only) 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NM 10Sources of MIT Information — State data * There is no national database of UIC MIT data + Many States maintain electronic UIC databases * State databases may have data on 90 percent of Class II wells + State data are not nationally compiled, readily accessible, or comparable 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NM uStudies of MIT Failure Rates + AClass I Injection Well Survey (Phase II Report): Survey of Operations (UIPC, 1987) + Hazardous Waste-Controls Over Injection Well Disposal Operations (GAO, 1987) * Class I Well Failure Analysis: 1988-1992. (USEPA, 1993) * Class I Mechanical Integrity Failure Analysis: 1993-1998. (USEPA, 1999) + Analysis of the Rate of and Reasons for Injection Well Mechanical Integrity Test Failure. (Class II wells) (USEPA, 1993) 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NM 2What An MIT Failure Is and Is Not ¢ MIT failure is a measure of well performance — it does not indicate well failure — it does not indicate a release to a USDW 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NMClass I MIT Failures: Rates * Rate of internal MIT failures/active well have declined from: — 32% in 1988-1992 to — 23% - 28% in 1993-1998 (proportion of failed tests per active number of wells; not MIT failure rate or percentage of wells with MIT failures) * Less than 2% of active wells have external MIT failures * MIT failure rates for hazardous waste injection wells are 2 to 3 times that of non-hazardous waste wells, possibly due to more corrosive injectate 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NM irClass I MIT Failures: Types ¢ Types of internal MIT failures — from 1998-1992: 37% tubing, 21% packer, 17% casing — from 1993-1998: 37-42% tubing, 20-25% packer, 34-23% casing ¢ Hazardous waste injection wells had slightly higher rates of casing failures than non-hazardous waste wells 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NM 1sClass I MIT Failures: Consequences * No reported USDW impacts associated with wells with known internal or external MIT failures in the 1988 to 1998 study period — ground water observation/monitoring well systems for identifying fluid releases or migration are rare 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NM 16Class II MIT Failures: Rates * 1983-1992 study: — 10.5% of MITs performed failed (percentage of tests that resulted in failures; not percent of wells with MIT failures) — very incomplete information (data from 4 States) — covers period where “grandfathered” older wells were being phased out 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NM ”Class II MIT Failures: Types * Only half the records reviewed identified the type of failure ¢ MIT failure types were: 54% casing, 25% tubing, 19% packer 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NM 18Class II MIT Failures: Consequences * No reported USDW impacts associated with wells with any known internal or external MIT failures in the 1983-1992 period — for the wells with casing MIT failures ~25% were plugged within 60 days, suggesting that failures were serious and that well re-working was expensive and not worthwhile — ground water observation/monitoring well systems for identifying fluid releases or migration are rare 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NM 19CO,/EOR Wells in the US: Location, Depth, & General Geology State | # fields | # inj wells depth (ft) injection zone co il 262 6,000 ss KS 1 3 2,900 LS MI 3 5 5,400 - 5,500 | LS/dolo MS: 6 89 10,300 - 11,000 |SS NM 3 168 4,200 - 4,550 | dolo OK 5 211 6,200-9,400 |SS ™X 50 3,521 2,680 - 8,000 _| Mix, mostly LS/dolo UT 2 130 5,600 - 5,700 dolo/LS WY 215 1,150 -9,000 | Mix, mostly SS Total 80 4,604 1,150- 11,000 | LS/dolo & SS Summarized from “Oil and Gas Journal,” April 17, 2006; Biannual EOR Survey, pg. 48, Table C 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NM 20Texas UIC Data - Class II EOR Wells (1983-2005) Total # of Wells 13 March 2007 UICIMIT TEA Well Bore Integrity Network Meeting ~ Santa Fe, NM % Wells with Injection Type Years Total # of We with Raa MIT Failure pre-1995, 568 35 6.2% 1995-1999 594 9 1.5% acid gas 2000-2005 748 61 8.2% pre-1995 3,324 135 41% 1995-1999 3,432 46 1.3% = 2000-2005 3,978 298 7.5% pre-1995, 5,395 197 3.7% 1995-1999 5,703 57 1.0% fresh water 2000-2005 6,175 359 5.8% pre-1995 10,713 483 45% 1995-1999 12,715) 223 1.8% brackish water 2000-2005 14,488 731 5.0% 21Percent TX Class II EOR Wells with MIT Failures (1983-2005) 15% 12% H2S/Acid Gas co2 Fresh Water Brackish Water 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NMPercentage of Active TX EOR Wells with at least one MIT Failure: Trends over Time 9% 8% 1% 6% Bi H,S/Acid Gas 5% co, CO Fresh Water 4% 1 Brackish Water| 3% 2% e i 0% pre-1995, 1995-1999 2000-2005 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NM. 2BRegulatory and Operational/Economic Context + 1985 — Standardized and fully documented MIT procedures were implemented in TX (MIT failures up) + 1990s — The TX MIT program was being refined and expanded + 1995 -- 3" 5-year MIT cycle shows old wells not worth repairing? (MIT failures down?) + 1995 — Operators required to post bonds for inactive wells and bring inactive wells into compliance before getting new permits (MIT failures up?) + 2000 — Rule change limited the use of TCAM (tubing-casing-annulus monitoring), an alternative to full MIT (MIT failure up?) * 1995-1999 — below average oil prices (effect on well use/MIT failures?) + 2000-2005 - above average oil prices (effect on well use/MIT failures?) 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NM 4Conclusions & Discussions + Akey component of the success of the Class I and Class II injection well programs is maintaining and monitoring injection well mechanical integrity ¢ Injection well mechanical integrity is a concern for CO, injection and storage due to corrosivity of CO, (and potential impurities) on injection well materials especially given the very large volumes and very long time frames anticipated for CO, geosequestration 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NMConclusions & Discussions (cont.) ¢ How might information from injection well MIT assessments apply to assessments of abandoned wells within proposed CO, storage reservoirs? + What types of well construction and integrity data will be generated during the DOE Pilot Projects? * Do operational, geological, risk, or other factors suggest a management framework with MIT guidelines or standards? Well construction guidelines or standards? 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NMConclusions & Discussions (cont.) ¢ How might information from injection well MIT assessments apply to assessments of abandoned wells within proposed CO, storage reservoirs? + What types of well construction and integrity data will be generated during the DOE Pilot Projects? * Do operational, geological, risk, or other factors suggest a management framework with MIT guidelines or standards? Well construction guidelines or standards? 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NMConclusions & Discussions (cont.) ¢ How might information from injection well MIT assessments apply to assessments of abandoned wells within proposed CO, storage reservoirs? + What types of well construction and integrity data will be generated during the DOE Pilot Projects? * Do operational, geological, risk, or other factors suggest a management framework with MIT guidelines or standards? Well construction guidelines or standards? 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NMEPA has published the guidance for using The Class V Experimental Technology Well Classification for Pilot Geologic Sequestration Projects http://www.epa.gov/safewater/uic/wells_sequestration.html 13 Mareh 2007 UIC/MIT IEA Well Bore Integrity Network Meeting ~ Santa Fe, NM 29