e-dokumen.com

These means of CO2 storage can occur in the followinggeological media (IPCC, 2005): oil and gas reservoirs deep saline aquifers, saturated with brackishwater or brine coal seams (sorption is the only potentiallypractical technique in coal seams and is not asignificant storage mechanism in the otherclasses of geological media) man-made undergroundcaverns, in CO2)cavities(i.e.,saltAny geological site for CO2 storage must posses thefollowing characteristics: capacity, for accepting the volumes of CO2 thatneed to be stored; injectivity, to allow introduction of CO2 into thesubsurface at the desired rates; and confining ability, to retain the CO2 for thedesired period of time (i.e., avoidance ofleakage).These characteristics are largely met by geologicalmedia in sedimentary basins, where oil and gas reservoirs,coal beds and salt beds and domes are found. Igneous andmetamorphic rocks are generally not suitable for CO2storage because they lack the permeability and porosityneeded for CO2 injection and storage, and/or because oftheir lack of confining properties due to their fracturednature. Volcanic areas and orogenic belts (mountains) arealso unsuitable for CO2 storage mainly because they lackcapacity and are unsafe.In the case of British Columbia, its landmass includesnine sedimentary basins and two geological troughs(Figure 1) that may have, to various degrees, potential forCO2 geological storage. A previous analysis (Bachu,2005) has identified the portion of the Western CanadaSedimentary Basin (WCSB) in northeastern BritishColumbia (Figure 1) as the most suitable sedimentarybasin in B.C. for CO2 geological storage and likely withthe largest capacity. Most of the “large” stationary CO2sources have emissions less than 500 kt CO2/yr, and veryfew sources emit more than 1 Mt CO2/yr (Figure 1). TheCO2 sources are distributed according to the majorindustrial and population centers along the Pacific coast,in the B.C. interior, and in northeastern B.C., where gasplants produce a stream of high purity CO2 and where apipeline system is more developed locally.The portion of the WCSB in northeastern B.C. is themost suited, and practically immediately accessible, basinfor CO2 geological storage in British Columbia. It meetsall the general suitability criteria for CO2 geologicalstorage (Bachu and Stewart, 2002; Bachu, 2003):- it is located in a tectonically stable region;- has regional-scale flow systems confined by thickaquitards; - has significant oil and gas reservoirs;- has coal beds; - has significant infrastructure in place;- there are CO2 sources in the area, including high-puritySummary of Activities 2006sources (gas plants); and- there is experience with acid-gas injection operations.Northeastern British Columbia has significant CO2storage capacity in oil and gas reservoirs, in deep salineaquifers, and possibly in coal beds if the technology willprove successful. The purpose of the work reported hereis to evaluate the potential and capacity for CO2geological storage in northeastern British Columbia.CAPACITY FOR CO2 STORAGE IN OILAND GAS RESERVOIRSWorldwide, the largest CO2-storage capacity is likelyin deep saline aquifers, while the smallest is in coal beds(IPCC, 2005), and this is most probably true ofnortheastern British Columbia. On the other hand, it isrecognized that, generally, CO2 storage in geologicalmedia will occur first in oil and gas reservoirs because ofthe following reasons (IPCC, 2005): 1) their geology andtrapping characteristics are better known as a result ofexploration for and production of hydrocarbons, 2) thereis already infrastructure in place (pipelines and wells),and 3) in the case of oil reservoirs that are suitable forCO2-flood enhanced oil recovery (EOR), additional oilproduction will lower the cost of CO2 storage, in somecases even realizing a profit, and will increase the stabilityand security of energy supplies. For this reason, thisassessment starts with examining the CO2 storagecapacity in oil and gas reservoirs.METHODOLOGYThe capacity for CO2 storage in hydrocarbonreservoirs in any particular region is given by the sum ofthe capacities of all reservoirs in that area, calculated onthe basis of reservoir properties, such as original oil or gasin place, recovery factor, temperature, pressure, rockvolume and porosity, as well as in situ CO2characteristics, such as phase behaviour and density. Thefundamental assumption being made in these calculationsis that the volume previously occupied by the producedhydrocarbons becomes, by and large, available for CO2storage. This assumption is generally valid for reservoirsthat are not in contact with an aquifer, or that are notflooded during secondary and tertiary oil recovery. Inreservoirs that are in contact with an underlying aquifer,formation water invades the reservoir as the pressuredeclines because of production. However, CO2 injectioncan reverse the aquifer influx, thus making pore spaceavailable for CO2. However, not all the previouslyhydrocarbon-saturated pore space will become availablefor CO2 because some residual water may be trapped inthe pore space due to capillarity, viscous fingering andgravity effects (Stevens et al., 2001).Another important assumption is that CO2 will beinjected into depleted oil and gas reservoirs until thereservoir pressure is brought back to the original, orvirgin, reservoir pressure. The results thus obtainedrepresent a conservative estimate because the pressure cangenerally be raised beyond the original reservoir pressureas long as it remains safely below the threshold rock-3

Figure 1. Location of sedimentary basins and major stationary CO2-sources in British Columbia.fracturing pressure. In this case, the CO2 storage capacitywould be higher due to CO2 compression. However, therisk of raising the storage pressure beyond the originalreservoir pressure requires a case-by-case reservoiranalysis that is not practical for basin-scale evaluations.Several capacity definitions are being introduced toclarify the meaning of various estimates and therelationships between them. The theoretical capacityassumes that all the pore space (volume) freed up by theproduction of all recoverable reserves will be replaced byCO2 at in situ conditions. The effective capacity is themore realistic estimate obtained after water invasion,displacement, gravity, heterogeneity and water-saturationeffects have been taken into account. Practical capacity isthe storage capacity after consideration of technologicallimitations, safety, CO2 sources and reservoirdistributions, and current infrastructure, regulatory andeconomic factors. In the end, all the issues and factorsrelating to CO2 capture, delivery and storage contribute to4a reduction in the real capacity for CO2 storage inhydrocarbon reservoirs. However, none of these capacityestimates is final, in the sense that these values evolve intime, most likely increasing as new oil and gasdiscoveries take place, or as better productiontechnologies are developed.Theoretical CO2-Storage CapacityOnly non-associated and associated gas reservoirs areconsidered in CO2-sequestration capacity calculationsbecause solution gas is taken into account in oil reservoirsthrough the oil shrinkage factor. Since reserves databasesindicate the volume of original gas in place (OGIP) atsurface conditions, the mass-capacity for CO2 storage in areservoir at in situ conditions, MCO2, is given by:MCO2 CO2r· Rf · (1 – FIG) · OGIP · [(Ps · Zr · Tr) / (Pr · Zs · Ts)](1),Q WKH DERYH HTXDWLRQ &22 is CO2 density, Rf is therecovery factor, FIG is the fraction of injected gas, P, Tand Z denote pressure, temperature and theBritish Columbia Resource Development and Geoscience Branch

compressibility factor, and the subscripts ‘r’ and ‘s’denote reservoir and surface conditions, respectively. TheCO2 density at reservoir conditions is calculated fromequations of state (e.g., Span and Wagner, 1996).The CO2 storage capacity of single-drive oilreservoirs is calculated similarly to gas reservoirs on thebasis of reservoir rock volume (area [A] times thickness [email protected] SRURVLW\ DQG RLO VDWXUDWLRQ – Sw), where Sw isthe water saturation. For reservoirs flooded with orinvaded by water, the volume available for CO2 storage isreduced by the volume of injected and/or invading water(Viw). If water is produced with oil, then the volumeavailable for CO2 storage is augmented by the volume ofproduced water (Vpw). The same mass balance applies inthe case of miscible flooding with solvent or gas. Thus:MCO2 CO2res· [Rf   K  – Sw) – Viw Vpw]  (2)The volumes of injected and/or produced water,solvent or gas can be calculated from production records.However, the pore volume invaded by water fromunderlying aquifers cannot be estimated without detailedmonitoring of the oil-water interface and detailedknowledge of reservoir characteristics.Effective CO2-Storage CapacityIn the case of reservoirs underlain by aquifers, thereservoir fluid (oil and/or gas) was originally inhydrodynamic equilibrium with the aquifer water. Ashydrocarbons are produced and the pressure in thereservoir declines, a pressure differential is created thatdrives aquifer water up into the reservoir. The amount andrate of water influx is controlled by: 1) reservoirpermeability and heterogeneity; 2) water expansion in theaquifer; 3) pore volume contraction due to the increase ineffective stress caused by the pressure drop in thereservoir; 4) expansion of hydrocarbon accumulationslinked to the common aquifer; and 5) artesian flow wherethe aquifer is recharged by surface water. Ashydrocarbons are produced, some portions of the reservoirmay be invaded by aquifer water, in addition to the initialwater saturation. If CO2 is then injected into the reservoir,the pore space invaded by water may not becomeavailable for CO2 storage, resulting in a net reduction ofreservoir capacity. The reduced storage volume mayeventually become available if the reservoir pressurecaused by CO2 injection is allowed to increased beyondthe original reservoir pressure, which may or may notalways be allowed or possible.Furthermore, thehysteresis effect caused by various mechanisms may alsoprevent complete withdrawal of invaded water, leading toa permanent loss of storage space.Analysis of the production history of close to 300 oiland gas pools in western Canada led to the establishmentof a set of criteria for determining if an oil or gas reservoirhas strong or weak aquifer support (Bachu and Shaw,2003, 2005; Bachu et al., 2004) on the basis of pressurehistory, water production, and cumulative water-gas ratio(WGR) or water-oil ratio (WOR). For oil reservoirs, thegas-oil ratio (GOR) was also included in the analysisbecause, typically, an oil pool with strong aquifer supporttends to have a slow pressure decline and flat GOR profileclose to solution GOR, and vice-versa. In addition, theproduction decline versus reservoir pressure was analyzedfor these pools. For gas pools, P/Z plots were used toidentify the presence of aquifer support, or lack thereof.The criteria and threshold values for identification of thestrength of underlying aquifers are presented in Table 1.The effect of the underlying aquifers was assessedusing the Petroleum Expert’s MBALTM (MaterialBALance) software for a limited number of oil and gaspools distributed across the Western Canada SedimentaryBasin that were considered to be reasonablyrepresentative for the range of conditions found in thebasin (Bachu and Shaw, 2003; Bachu et al., 2004).Injection of CO2 was assumed to start immediately afterreservoir depletion and to continue until the pool pressureexceeded the original pressure. Although the materialbalance reservoir model simulated by MBALTM is a tankmodel and does not account for reservoir geometry,drainage area and wells location, it is a very useful tool inmatching the production history by determining thepresence, type and size of an aquifer, and predictingreservoir pressure and performance for given productionand/or injection scenarios.TABLE 1. CRITERIA FOR ESTABLISHING THE STRENGTH AND EFFECT OF UNDERLYING AQUIFERSON THE CO2 STORAGE CAPACITY IN DEPLETED OIL AND GAS RESERVOIRS IN THE WESTERNCANADA SEDIMENTARY BASIN AND THE CORRESPONDING COEFFICIENT OF REDUCTION IN CO2STORAGE CAPACITYSummary of Activities 20065

Table 1 shows the reduction in CO2 storage capacityfor reservoirs with strong aquifer support. The storagecapacity of reservoirs with weak or no aquifer support isnot affected by the presence of the underlying aquifer.However, a very small effect needs to be considered inlight of the fact that water is a wetting phase, as opposedto oil and gas, which are non-wetting, hence it should beexpected that some irreducible water would be left behindin the pore space by the receding aquifer. To account forthis effect it is assumed that the theoretical CO2-storagecapacity in oil and gas reservoirs with weak aquifersupport is reduced by 3%.Notwithstanding the effect of an underlying aquifer,three factors, in particular, control the effectiveness of theCO2 storage process: CO2 mobility with respect to oil andwater; the density contrast between CO2 and reservoir oiland water, which leads to gravity segregation; andreservoir heterogeneity. Because of the very low CO2viscosity in liquid or supercritical phase, on the order of10-5 Pa s, the CO2/oil and CO2/water mobility ratios atreservoir conditions are on the order of 20 and higher. Asa result, viscous fingering will develop and the CO2 willtend to bypass the oil/water system in place in thereservoir, leading to a very unfavourable displacementprocess (Bondor, 1992).Depending on reservoir temperature and pressure, thedensity of supercritical or liquid CO2 may range betweenapproximately 200 and 800 kg/m3. The density difference(buoyancy) between the lighter CO2 and the reservoir oiland water leads to gravity override at the top of thereservoir, particularly if the reservoir is relativelyhomogeneous and has high permeability (Bondor, 1992;Stephenson et al., 1993; Doughty and Preuss, 2004). Thisnegatively affects the CO2 storage, and the oil recovery inthe case of EOR.If the reservoir is heterogeneous, the injected CO2will flow along the path of less resistance, namely throughregions of high permeability, bypassing regions of lesserpermeability. This has a negative effect for oil recoverybecause whole regions of the reservoir may be leftunswept by CO2 before it breaks at the production well,thereby reducing the economic benefit. On the other hand,reservoir heterogeneity may have a positive effectbecause it may counteract the buoyancy effect by slowingdown the rise of CO2 to the top of the reservoir andforcing it to spread laterally, resulting in better verticalsweep efficiency (Doughty and Preuss, 2004).The presence of water in the reservoir also has theeffect of reducing the CO2 storage capacity, as discussedpreviously. Water may be present because of initial watersaturation, because of water invasion as the reservoir isdepleted, or because it was introduced during secondaryand/or tertiary recovery. As a result of capillary forces,irreducible water (Swirr) will remain in the reservoir evenif the water is ‘pushed back’ by the injected CO2.All the processes and reservoir characteristics thatreduce the actual volume available for CO2 storage can beexpressed by capacity coefficients (C 1) in the form(Doughty and Preuss, 2004):MCO2eff Cm · Cb · Ch · Cw · Ca · MCO2res6(3)where MCO2eff is the effective reservoir capacity for CO2storage, and the subscripts m, b, h, w and a stand formobility, buoyancy, heterogeneity, water saturation, andaquifer strength, respectively, and refer to the phenomenadiscussed previously. These capacity coefficients likelyvary over a wide range, depending on reservoircharacteristics, and this explains the wide range ofincremental oil recovery (7 to 23% of OOIP) and CO2utilization (0.7 to 4.7 m3 CO2 / m3 recovered oil atreservoir conditions) observed for 25 CO2-flood EORoperations in Texas (Holt et al., 1995). Unfortunately,there are very few studies and methodologies forestimating the values of these capacity coefficients,mostly on the basis of numerical simulations, andgenerally there are no data or past experience for thespecific case of CO2 storage in depleted hydrocarbonreservoirs. The first four capacity coefficients can becaptured in a single ‘effective’ coefficient:Ceff Cm · Cb · Ch · Cw(4)which can be estimated on the basis of experience withCO2-flood EOR. A review of capacity coefficients forCO2 storage in aquifers suggests that Ceff 0.3.Conditions are more favourable in the case of oilreservoirs (for example the buoyancy contrast is muchreduced), and a value of Ceff 0.5 was considered in thisstudy. For gas reservoirs, Cm § EHFDXVH ILQJHULQJ effects are very small to negligible. Because CO2 densityis greater than that of methane at reservoir conditions, theCO2 injected in gas reservoirs will fill the reservoir fromits bottom. Thus, it can be assumed that Cb § DV ZHOO The effect of initial water saturation was alreadyimplicitly taken into account in the estimates oftheoretical ultimate CO2-storage capacity, such that Cw § 1 too. Although reservoir heterogeneity may reduce theCO2 storage capacity by leaving pockets of original gas inplace, Ch is probably high, approaching values close tounity. Thus, the reduction in CO2 storage capacity for gasreservoirs is much less by comparison with oil reservoirsand a value of Ceff 0.9 was used in this study.CO2 Storage Capacity in Enhanced Oil RecoveryCarbon dioxide can be used in tertiary enhanced oilrecovery in miscible floods, if high purity CO2 isavailable. To date, except for the Weyburn oil field inSaskatchewan, operated by Encana, and the Joffre VikingA oil reservoir in Alberta, currently operated by PennWest, no CO2-EOR operations were implemented inwestern Canada because of the high cost of CO2 captureand lack of pipeline infrastructure for CO2 delivery at thewell head. However, this situation may rapidly change ifincentives for CO2 geological storage are introduced and amarket for carbon credits is created. For example, theAlberta Government introduced in 2004 a Royalty CreditProgram to encourage the development of a CO2-EORindustry in the province, and, as a result, currently fourpilot operations started in Alberta in 2005. In a futurecarbon-constrained environment and sustained high oilprices, CO2 flooding will probably become the preferredEOR option, leading to both CO2 geological storage andadditional oil recovery. In fact, it is most likely that thisoption will be implemented before any other. Thus,identification of reservoirs suitable for CO2 flooding andBritish Columbia Resource Development and Geoscience Branch

estimation of their CO2 storage capacity becomesessential.Based on the experience gained in the United Stateswhere CO2-EOR is being practiced for more than 30 yearsat close to 70 oil fields in the Permian Basin of westTexas, a series of technical criteria were developed forassessing the suitability of oil reservoirs for CO2-EOR,reviewed and summarized in several publications (Taberet al., 1997; Kovscek, 2002; Shaw and Bachu, 2002). Inassessing the s

man-made underground cavities (i.e., salt caverns, in CO 2) Any geological site for CO 2 storage must posses the following characteristics: capacity, for accepting the volumes of CO 2 that need to be stored; injectivity, to allow introduction of CO 2 into the subsurface at the desired rates; and confining ability, to retain the CO