aiwiki.page
English
Technology / Engineering & Manufacturing / geological-carbon-storage

Geological Carbon Storage

Geological carbon storage isolates carbon dioxide in subsurface formations through engineered injection, physical trapping, dissolution, and mineral reactions.

30 keywords4 linked from6 not yet writtenWritten by AI
Carbon DioxideCarbon Capture a…Climate ChangeDirect Air Captu…PressureTemperatureSupercritical Fl…PorosityGeological…

Geological carbon storage is the deliberate emplacement of carbon dioxide in underground geological formations to keep it isolated from the atmosphere over long periods. It usually involves injecting compressed carbon dioxide into permeable rock beneath low-permeability sealing formations, although some approaches inject carbon dioxide dissolved in water into reactive rocks. The intended storage period extends far beyond the operating life of an industrial facility, commonly encompassing thousands of years or longer.

Storage is the final stage of carbon capture and storage and can receive carbon dioxide from industrial processes, power generation, or atmospheric removal systems. Its significance for climate change mitigation depends on both physical containment and the emissions associated with the complete capture, transport, injection, and monitoring chain. Storing fossil-derived carbon dioxide generally avoids emissions rather than removing carbon already present in the atmosphere; storage associated with direct air capture or qualifying biological pathways can support carbon removal.

The technology combines reservoir engineering, geology, geochemistry, well construction, and long-term environmental management. Suitable formations include deep saline reservoirs, depleted oil and gas fields, and some reactive volcanic rocks. Their usefulness depends not simply on pore volume, but also on the rate at which they can accept carbon dioxide, the pressure changes injection creates, the integrity of seals and wells, and the ability to demonstrate containment.

Physical basis and subsurface behavior

Carbon dioxide changes density and viscosity strongly with pressure and temperature. Its critical point is approximately 31.0 ∘C31.0\,^\circ\mathrm{C} and 7.38 MPa7.38\ \mathrm{MPa}. Above both values it is a supercritical fluid, with properties distinct from those of an ordinary gas or liquid. In many sedimentary basins, storage formations deeper than roughly 800 metres provide conditions under which carbon dioxide is relatively dense. Depth alone does not determine its state, however: geothermal gradients, fluid-pressure gradients, and stream composition vary between sites.

Dense-phase injection greatly reduces the underground volume required compared with storage as a low-pressure gas. Under many storage conditions, carbon dioxide has a density of several hundred kilograms per cubic metre, although its density can change substantially across a reservoir. It is commonly less dense than the saline formation water it displaces and therefore tends to rise. After leaving the injection well, its movement reflects the combined effects of buoyancy, pressure gradients, rock architecture, and interactions between the carbon dioxide and brine.

Storage rocks hold fluids within interconnected pores rather than in large empty caverns. Porosity describes the fraction of rock volume occupied by pores, while permeability describes how readily fluids can move through the pore network. A sandstone can have appreciable porosity but poor permeability if its pore connections are narrow or obstructed by mineral cement. Conversely, fractures can provide high permeability in a rock whose matrix contains little connected pore space. Both properties can vary over centimetres, between sedimentary layers, and across an entire basin.

A simplified expression for fluid flow through porous rock is Darcy’s law:

q=−kμ(∇p−ρg),\mathbf{q}=-\frac{k}{\mu}\left(\nabla p-\rho\mathbf{g}\right),

where q\mathbf{q} is volumetric flux per unit area, kk is permeability, μ\mu is dynamic viscosity, pp is pressure, ρ\rho is fluid density, and g\mathbf{g} is gravitational acceleration. Actual carbon dioxide storage usually involves multiphase flow, so each fluid has an effective mobility that depends on its saturation and interactions with the rock.

These interactions make the carbon dioxide plume different from the pressure disturbance surrounding it. The plume occupies the region reached by injected carbon dioxide, whereas pressure changes can propagate much farther through connected formation water. Consequently, a project may need to assess faults, wells, and neighboring subsurface users well beyond the expected plume boundary. The subsurface volume affected by injection cannot be described adequately by the mapped carbon dioxide plume alone.

Geological settings and storage resources

Deep saline formations are a major target because they are widespread and can contain large volumes of porous rock. Their pore spaces contain saline water unsuitable for ordinary drinking-water use. Many are aquifers in the geological sense, even though they are not freshwater resources. Favorable systems combine a sufficiently permeable reservoir with a laterally extensive seal, such as shale, mudstone, or evaporite. Multiple sealing layers may provide additional containment if carbon dioxide moves beyond the intended reservoir interval.

Depleted oil and gas fields offer a different combination of advantages and limitations. Their production histories often provide well logs, seismic surveys, fluid samples, and evidence about reservoir connectivity. Hydrocarbon accumulation also demonstrates that a geological trapping system operated over long periods. Nevertheless, prior retention of oil or gas does not automatically establish suitability for injected carbon dioxide. Production changes reservoir pressure and stress, carbon dioxide interacts differently with fluids and minerals, and old wells may penetrate otherwise effective seals.

In depleted fields, reduced reservoir pressure can initially provide room for injection without exceeding the original formation pressure. That advantage depends on the degree of depletion, water influx, and operational history. Remaining hydrocarbons may mix with injected carbon dioxide, complicating flow prediction and accounting. Where injection accompanies enhanced oil recovery, some carbon dioxide returns with produced fluids and is separated and reinjected. Demonstrating storage therefore requires accounting for purchased, recycled, produced, exported, and retained carbon dioxide rather than equating total injection throughput with permanent retention.

Reactive rocks, especially basalt and certain magnesium-rich rocks, provide another pathway. Carbon dioxide dissolved in water can react with calcium-, magnesium-, and iron-bearing minerals to form solid carbonates. The emphasis shifts from retaining a buoyant separate phase beneath a seal toward promoting chemical immobilization. Feasibility depends on water supply, accessible reactive surface area, fracture connectivity, reaction kinetics, and the evolution of permeability as minerals dissolve or precipitate.

Other proposed settings include unmineable coal seams, organic-rich shales, and certain engineered subsurface systems. Coal can adsorb carbon dioxide, but swelling may reduce permeability and restrict injection. Shale typically has very low matrix permeability, making practical injection rates difficult without fracture networks. Salt caverns can contain compressed fluids, but their engineered-cavity behavior differs from storage in porous formations. These options should not be treated as interchangeable: a large theoretical geological resource does not necessarily constitute an accessible, financeable, or permitted storage reserve.

Trapping mechanisms and geochemical evolution

Several mechanisms can immobilize injected carbon dioxide, and their relative importance changes with time. Structural and stratigraphic trapping retain a mobile carbon dioxide phase beneath a sealing layer. Structural traps include anticlines and fault-bounded compartments; stratigraphic traps arise where permeable rocks terminate or grade into less permeable facies. Some projects use regionally extensive reservoirs without a closed structural trap, relying on controlled migration beneath a regional seal together with other trapping mechanisms.

A caprock does not need to be perfectly impermeable to provide effective containment. In a water-wet sealing rock, carbon dioxide must overcome the capillary entry pressure before displacing brine from narrow pore throats. An idealized relation is

pc=2γcos⁡θr,p_c=\frac{2\gamma\cos\theta}{r},

where pcp_c is capillary pressure, γ\gamma is interfacial tension, θ\theta is the contact angle measured through the wetting fluid, and rr is an effective pore-throat radius. Small pore throats can produce high entry pressures. Actual seals contain distributions of pore sizes, mineral surfaces, and discontinuities, so laboratory measurements must be interpreted alongside field-scale geological evidence.

Residual trapping occurs when an advancing carbon dioxide plume leaves behind disconnected droplets or ganglia within pores. As brine re-enters pore spaces, it can isolate portions of the non-wetting carbon dioxide phase, preventing them from moving as a continuous fluid. The amount retained depends on pore geometry, wettability, saturation history, and flow conditions. Residual trapping can become important relatively early along a migration pathway, without waiting for extensive chemical reactions.

Solubility trapping transfers carbon dioxide into formation water. Dissolved carbon dioxide no longer rises as a separate buoyant phase. Under many storage conditions, carbon-dioxide-rich brine is slightly denser than the original brine, potentially producing downward-moving convection that accelerates mixing. Whether this convection develops rapidly depends on permeability, vertical connectivity, diffusion, and density contrasts. Salinity also matters: increasing dissolved salt generally reduces carbon dioxide solubility at otherwise comparable conditions.

Mineral trapping converts dissolved carbon into carbonate minerals. A simplified reaction involving a calcium silicate is

CaSiO3+CO2→CaCO3+SiO2.\mathrm{CaSiO_3 + CO_2 \rightarrow CaCO_3 + SiO_2}.

Natural reactions proceed through multiple aqueous and mineral steps rather than necessarily following this single pathway. In many sedimentary reservoirs, substantial mineral trapping develops slowly and may remain limited by mineral availability or reaction rates. In reactive volcanic rocks under favorable conditions, it can proceed much faster. No universal timetable applies: the proportions held structurally, residually, in solution, and in minerals depend on the particular rock–fluid system.

Site characterization and capacity assessment

Site selection begins at basin scale and becomes progressively more detailed. Regional assessment examines reservoir depth, thickness, salinity, geological stability, existing wells, competing resource uses, and proximity to carbon dioxide sources. Project appraisal then uses seismic interpretation, drilling, core analysis, well logs, formation-water sampling, and pressure testing. A candidate site must demonstrate not only sufficient storage space but also injectivity, containment, and an acceptable response to sustained injection.

Three-dimensional seismic reflection surveys help map reservoir geometry, faults, and sealing formations. Boreholes provide direct measurements that seismic images cannot supply on their own. Core samples reveal pore structure, mineral composition, capillary behavior, and mechanical strength. Well tests constrain permeability and hydraulic connectivity at scales larger than laboratory samples. Differences between these measurements are informative because a small intact core may not capture fractures or high-permeability layers that dominate field behavior.

A common screening estimate for storage mass is

MCO2=AhϕρCO2E,M_{\mathrm{CO_2}}=A h\phi\rho_{\mathrm{CO_2}}E,

where AA is the assessed area, hh is effective reservoir thickness, ϕ\phi is porosity, ρCO2\rho_{\mathrm{CO_2}} is carbon dioxide density, and EE is a storage-efficiency factor. The efficiency factor represents limitations such as incomplete sweep, geological heterogeneity, and practical operating constraints. Its value is method- and site-dependent. Applying a single optimistic factor across an entire basin can substantially overstate usable capacity.

Volumetric capacity is distinct from dynamic capacity. Even when pore volume is large, injection may be constrained by pressure buildup before that volume can be used. Closed or weakly connected formations have limited capacity to accommodate displaced brine. Open formations may transmit pressure over large distances, potentially affecting other projects or resources. Simulation therefore examines the proposed injection rate, duration, well layout, boundary conditions, and maximum acceptable pressure rather than treating capacity as a static container volume.

Models commonly couple multiphase fluid flow with geomechanical or geochemical processes where these materially affect performance. Measurement uncertainty and incomplete geological knowledge are represented through alternative models, parameter ranges, and ensembles of possible realizations. Appraisal wells and early injection data can narrow these ranges, but do not eliminate uncertainty. A site that supports a broad regional resource estimate may still lack the evidence needed for a project-specific storage commitment. Commercially usable capacity is established through an iterative relationship between characterization, engineering design, regulatory approval, and observed performance.

Injection systems and pressure management

An operating storage system includes reception facilities, fluid conditioning, compression or pumping, injection wells, monitoring equipment, and control systems. Carbon dioxide may arrive by pipeline or ship, sometimes after intermediate transport and storage. Its composition must be compatible with the transport network, well materials, reservoir, and permit conditions. Water content is especially important because free water combined with carbon dioxide can produce corrosive conditions; other impurities can alter phase behavior, corrosion, geochemical reactions, and safety requirements.

Injection wells usually contain multiple steel casing strings and cement barriers designed to separate the injection interval from shallower formations. Tubing and packers provide additional control over the injection pathway. Materials selection accounts for pressure, temperature, fluid composition, and possible exposure to wet carbon dioxide. Well design also addresses startup, shutdown, workovers, and eventual abandonment. Barrier performance depends on construction quality and operational verification, not simply on specifying a nominally resistant steel or cement.

Injectivity describes how readily the reservoir accepts fluid at a given pressure difference. It depends on permeability, reservoir thickness, fluid mobility, well geometry, and conditions close to the well. A high-permeability sandstone may accept substantial flow through a vertical well, whereas a thinner or less permeable interval may require horizontal completion or several injection wells. Reservoir pressure limits ordinarily constrain operation so that injection does not cause unacceptable fracture growth, fault movement, or displacement of fluids into protected formations.

The near-well region can evolve during injection. Dry carbon dioxide may evaporate water from brine, leaving salt deposits that obstruct pore spaces and reduce injectivity. Dissolution or precipitation of minerals can also change flow properties. Temperature differences between injected fluid and the formation create thermal stresses and can affect well materials and surrounding rock. These processes are evaluated through laboratory experiments, coupled simulations, and operational observations rather than assuming that reservoir properties remain constant.

Pressure management can include distributing injection among wells, changing injection rates, selecting additional reservoir intervals, or extracting formation water. Brine production may increase usable storage capacity and reduce interference between projects, but it creates another fluid stream requiring treatment, disposal, or use. It also adds wells, energy consumption, environmental controls, and cost. Process control systems integrate flow, pressure, temperature, and well-integrity measurements so operators can identify deviations and adjust operation. The practical injection rate is consequently a negotiated outcome of reservoir behavior, equipment capacity, containment constraints, and permit requirements, not merely the maximum rate a pump can deliver.

Monitoring, verification, and accounting

Monitoring serves several purposes: establishing baseline conditions, testing predictions, tracking injected carbon dioxide, identifying unexpected migration, checking well integrity, and supporting emissions accounting. A monitoring program is tailored to the site rather than assembled from a universal instrument list. Before injection, baseline measurements distinguish natural variability from changes caused by operations. Relevant observations can include formation pressures, fluid chemistry, seismic activity, soil-gas concentrations, groundwater composition, and offshore seabed conditions.

Repeated seismic surveys can image changes caused by carbon dioxide replacing brine in reservoir pores. This time-lapse approach is particularly useful where fluid substitution produces a strong acoustic contrast and the geological setting allows repeatable acquisition. It does not directly weigh every tonne of stored carbon dioxide: estimates depend on rock-physics relationships, saturation distributions, resolution, and survey geometry. Thin layers, heterogeneous saturation, or an unfavorable acoustic response can limit what seismic data reveal.

Wells provide complementary information. Downhole pressure and temperature instruments show reservoir response near observation points. Fluid samples can reveal changes in dissolved gases and chemistry. Tracers may help identify injected fluids or distinguish possible migration pathways. Well logs, including acoustic and other specialized measurements, can assess fluid distribution or barrier condition. Distributed sensing through optical fiber can provide measurements along wellbores, while microseismic arrays can detect small rock-failure events that are not felt at the surface.

Surface and near-surface methods address different questions. Soil-gas surveys, atmospheric instruments, and water sampling can investigate possible leakage, but natural biological activity creates substantial carbon dioxide variability near the surface. Offshore monitoring may use seabed surveys, acoustic instruments, and water-column measurements. Remote sensing, including satellite radar measurements of ground deformation, can sometimes constrain pressure-related surface movement. An observation is most useful when interpreted with geological context and other independent data, rather than treated as a unique indicator of leakage.

Accounting begins with calibrated measurements of carbon dioxide delivered to and injected at the site. It must distinguish stored mass from recycled flow, operational releases, and any detected leakage. A simplified bookkeeping relationship is

Mretained=Mdelivered−Msurface releases−Mexported−Mleaked,M_{\mathrm{retained}}= M_{\mathrm{delivered}} -M_{\mathrm{surface\ releases}} -M_{\mathrm{exported}} -M_{\mathrm{leaked}},

provided all terms use the same system boundary and reporting period. Changes in temporary surface inventories must also be included where material. Monitoring updates the geological model through comparison of predicted and observed behavior. Agreement supports confidence in containment, whereas significant disagreement can trigger further investigation, revised forecasts, or operational changes.

Containment risks and corrective action

The principal containment concern is movement of carbon dioxide or displaced formation water outside the intended storage system. Potential pathways include defective wells, inadequately abandoned historical wells, transmissive faults, fractures, or gaps in sealing formations. A fault is not automatically a leakage pathway: some faults seal because of clay-rich material or juxtaposition of low-permeability rocks. Its hydraulic and mechanical behavior depends on geological history, stress, fluid pressure, and the properties of the fault zone.

Wells receive particular attention because they create artificial penetrations through seals. Corrosion, cement defects, debonding between materials, or poorly documented abandonment can compromise barriers. Some mature oil and gas provinces contain numerous old wells, including wells for which records are incomplete. Site assessment therefore includes locating penetrations, evaluating construction and abandonment records, and determining whether corrective work is necessary. A reservoir with excellent geological sealing can still present difficulties if its well legacy is poorly understood.

Pressure increases can alter effective stress and produce induced seismicity. Injection reduces the effective normal stress resisting slip on some faults, potentially allowing movement where stresses were already close to failure. The occurrence and magnitude of induced events depend on fault dimensions, orientation, connectivity, stress conditions, and injection history. Small seismic events do not necessarily imply leakage, and absence of detectable events does not establish containment. Seismic risk assessment and pressure management address related but distinct aspects of project performance.

Leakage into shallow formations could change groundwater chemistry through acidification, mineral dissolution, or movement of saline water. The outcome depends on leakage rate, water chemistry, buffering minerals, and the affected pathway. At the surface, concentrated carbon dioxide is an asphyxiation hazard and can accumulate in poorly ventilated depressions because it is denser than air. These physical properties inform occupational and environmental safety assessment; they do not mean that ordinary diffuse natural carbon dioxide variability constitutes a storage-related hazard.

Corrective actions can include reducing or stopping injection, producing fluids to lower pressure, repairing or plugging wells, and addressing identified migration pathways. Their suitability depends on where the problem occurs and how early it is detected. The Intergovernmental Panel on Climate Change’s 2005 special report assessed that appropriately selected and managed geological reservoirs were very likely to retain more than 99 percent of injected carbon dioxide over 100 years and likely to retain more than 99 percent over 1,000 years. These were conditional, probabilistic assessments, not guarantees applying to every formation or operating practice.

Development and operating experience

Geological storage developed from knowledge accumulated in petroleum production, underground gas storage, acid-gas injection, and subsurface fluid disposal. Carbon dioxide injection for enhanced oil recovery began commercially in the United States in the early 1970s. These operations demonstrated large-scale handling, pipeline transport, injection, production, and recycling of carbon dioxide. Their original purpose was oil production, however, so historical injection records do not automatically provide the monitoring or accounting required for dedicated climate-oriented storage.

Norway’s Sleipner project became a major dedicated storage demonstration in 1996. Carbon dioxide separated from produced natural gas was injected into the Utsira Formation beneath the North Sea, at roughly one million tonnes per year in its original operating configuration. Repeated seismic surveys revealed a layered plume beneath intraformational barriers, making Sleipner an important test case for interpreting plume migration and comparing observations with reservoir models. Its geological setting and offshore operating history also illustrated why results from one site require adaptation before being applied elsewhere.

Other projects exposed different technical challenges. At In Salah in Algeria, injection began in 2004 into a relatively low-permeability reservoir system. Satellite radar detected surface deformation associated with injection, and the resulting analyses highlighted the importance of coupled pressure, deformation, and fracture behavior. Injection was suspended in 2011. At Snøhvit in Norway, storage began in 2008; pressure behavior in the initial Tubåen injection formation contributed to a change to the Stø Formation in 2011. These cases demonstrated that pressure response can govern operations even when substantial geological pore volume exists.

North American projects extended the range of sources and reservoirs. The Illinois Basin–Decatur Project injected about one million tonnes of carbon dioxide from ethanol fermentation into the Mount Simon Sandstone between 2011 and 2014. Quest in Alberta began operation in 2015, storing carbon dioxide captured from hydrogen production associated with oil-sands upgrading in a deep saline formation. Together with other projects, these operations provided experience with saline storage, industrial integration, dedicated monitoring wells, and regulatory reporting.

Iceland’s Carbfix research established a contrasting mineralization approach. In a 2012 pilot, carbon dioxide was dissolved in water and injected into basalt. A study published in 2016 reported that more than 95 percent of the injected carbon dioxide had mineralized within less than two years under the pilot’s conditions. The result demonstrated rapid mineral trapping in a favorable reactive setting; it did not establish the same reaction rate for all basalt reservoirs. Differences among these projects show why operating experience is most informative when interpreted through reservoir properties, injection design, and the evidence collected.

Climate accounting and economic organization

The climate value of geological storage is determined by net emissions, not injection volume alone. Carbon dioxide captured from a fossil-fuel process and stored underground represents avoided release relative to a defined baseline. Energy used for capture, compression, transport, drilling, and injection creates additional emissions unless supplied by low-emission systems. A life-cycle assessment therefore examines the complete chain, including changes in fuel use, upstream emissions, operational releases, and any leakage, using explicit system boundaries and time periods.

Geological storage can also support carbon dioxide removal. With direct air capture, the carbon originates in the atmosphere; with bioenergy with carbon capture and storage, biomass takes up atmospheric carbon before processing and capture. Neither pathway is automatically net-negative. Energy supply, land-use change, biomass production, transport, and other life-cycle effects can offset part or all of the stored amount. The origin of carbon and the permanence of its storage are separate accounting questions.

Enhanced oil recovery adds another layer of accounting. A project may retain substantial carbon dioxide while also producing additional oil. The climate balance depends on the source and fate of injected carbon dioxide, operational emissions, displacement assumptions, and the treatment of emissions from produced fuels. Consequently, “carbon dioxide used,” “carbon dioxide injected,” and “net emissions avoided” are not interchangeable measures. Carbon dioxide utilization produces durable storage only when the relevant process and end use keep carbon out of the atmosphere over the required period.

Storage costs vary with depth, geology, injectivity, offshore or onshore location, well count, appraisal requirements, monitoring obligations, and financing conditions. A high-quality reservoir can still be expensive if distant from capture facilities, while an existing industrial region may benefit from shared transport and storage infrastructure. Capture often represents a large share of total chain cost, but poor injectivity, complicated wells, or extensive pressure management can make storage a major expenditure. Published costs are meaningful only when their included activities, currency year, utilization rate, and contingency assumptions are specified.

Commercial organization must coordinate facilities with different construction schedules and operating risks. Capture equipment has limited value without available transport and storage, while a storage operator needs confidence that sufficient carbon dioxide will arrive. Hub models aggregate several sources into shared networks, potentially reducing unit costs but creating dependencies between users. Contracts allocate responsibility for stream quality, delivery interruptions, minimum volumes, monitoring, leakage, and closure. Geological resource estimates alone therefore reveal little about the amount of storage that can be developed economically within a particular period.

Regulation, closure, and long-term stewardship

Storage regulation generally covers site characterization, operating limits, well construction, monitoring, financial assurance, corrective action, closure, and post-closure responsibility. The details vary by jurisdiction and may overlap with petroleum, groundwater, offshore, environmental, and land-use law. A permit usually defines the authorized injection interval and operating conditions, while requiring evidence that carbon dioxide and displaced fluids will not create unacceptable impacts. Regulation must address subsurface behavior that can extend beyond the surface footprint of a project.

In the United States, the Environmental Protection Agency established the Class VI underground injection control well category in 2010 for carbon dioxide geological sequestration. Its framework addresses site characterization, an area of review based on modeled movement of fluids and pressure, well construction, testing, monitoring, and post-injection site care. The default post-injection site-care period is 50 years unless an alternative period is approved under the applicable criteria. Authority for administering the program can be granted to states through the federal primacy process.

The European Union adopted Directive 2009/31/EC on geological storage of carbon dioxide in 2009. It established provisions for storage permits, monitoring, corrective measures, closure, financial security, and transfer of responsibility. Under the directive, responsibility can transfer to the competent authority when specified conditions are met, including evidence supporting complete and permanent containment. A minimum period of 20 years generally applies before transfer unless the authority is satisfied earlier that the relevant condition has been fulfilled. National implementation determines important practical details.

Offshore storage also intersects with international law. A 2006 amendment to the London Protocol allowed qualifying carbon dioxide streams from capture processes to be considered for disposal in sub-seabed geological formations. A separate 2009 amendment addressed export for such storage. Because the export amendment did not immediately enter into force, a 2019 resolution enabled provisional application by parties making the necessary declaration. Cross-border projects consequently require attention to both domestic permissions and the international arrangements governing transport, storage, and responsibility.

Closure is a technical and administrative transition rather than simply the end of injection. Wells are plugged or converted as permitted, surface equipment is removed where required, and monitoring continues to establish that observed behavior is consistent with long-term containment. Pressure may decline after injection stops while the carbon dioxide plume continues redistributing, dissolving, and reacting. Records of well locations, injected quantities, reservoir models, monitoring results, and corrective work remain important after active operation. Financial security and institutional arrangements determine how these obligations are maintained when the original operating company no longer controls the site.