How Does Underground Energy Storage Work? Comparing CAES, UHS, UGS and UGSF
The subsurface already plays a key role in managing energy systems. In the United States, several trillion cubic feet of natural gas are held in geological formations to balance supply and demand. As solar and wind power continue to grow, this role could become even more significant.
Underground energy storage covers several applications, including compressed air storage (CAES), hydrogen storage (UHS), and natural gas storage (UGS). They all follow the same basic logic: using deep geological formations, such as salt caverns, depleted reservoirs, and in some cases aquifers, to hold large quantities of energy for periods ranging from a few hours to several seasons.
Each approach nevertheless meets different needs. Capacity, storage duration, efficiency, withdrawal rate, and geological constraints determine which solutions best suit each context.
Table 1: Comparison of the main underground energy storage technologies
| Application | Fluid or energy stored | Formations or structures used | Storage duration |
|---|---|---|---|
| CAES | Compressed air | Salt caverns, hard-rock caverns, aquifers, and engineered reservoirs | From a few hours to several years depending on design; mainly used for hourly to weekly balancing |
| UHS | Gaseous hydrogen | Salt caverns, depleted reservoirs, and saline aquifers | Long-term and seasonal storage |
| UGS | Natural gas | Depleted reservoirs, salt caverns, and aquifers | Mainly seasonal; more frequent cycling possible in salt caverns |
| UPSH in salt caverns | Pressurized hydraulic energy | Pairs of caverns developed in salt domes or salt beds | About 20 hours in the configuration presented |
Compressed Air Energy Storage (CAES): Turning the Subsurface into a Mechanical Battery
Compressed air energy storage is one of the few large-scale, commercially available storage technologies capable of delivering more than 100 MW from a single unit [1][2]. Used for several decades now, CAES also ranks among the most proven ways to store large amounts of electricity in mechanical form.
How It Works: Converting Electricity into Compressed Air
CAES, short for Compressed Air Energy Storage, uses surplus electricity to compress air at high pressure. That air is then injected into an underground cavern or another suitable reservoir. When the grid needs power, the air is withdrawn, expanded through a turbine, and used to drive a generator.
In a conventional gas turbine, air compression and expansion happen within the same cycle. CAES separates these two steps in time by holding the compressed air between charging and generating periods. This lets a facility draw power when it's plentiful and release it later, according to grid needs.
Several types of structures can be used to hold the air, including salt caverns, cavities carved into hard rock, certain aquifers, and above-ground vessels [2]. Large-capacity underground facilities often rely on salt caverns whose volume can reach several hundred thousand cubic meters [1]. CAES is mainly suited to balancing the grid over periods ranging from a few hours to several weeks. It can also support certain frequency regulation services, particularly tertiary reserves and, to a lesser extent, secondary ones [1]. A facility's lifespan is generally estimated at 30 to 40 years [1].
The Three CAES Variants: Diabatic, Adiabatic, and Isothermal
The efficiency of compressed air energy storage largely depends on how a facility manages heat. When a gas is compressed, its temperature rises; conversely, it cools during expansion. The heat generated during compression can therefore be released, kept for later use, or minimized from the outset. These different thermal strategies define three main CAES variants.
These distinctions directly affect each variant's efficiency and level of maturity:
Diabatic CAES: in this setup, the heat generated during compression is released into the environment rather than kept [2]. Before reaching the turbine, the compressed air must then be reheated, usually by burning natural gas. This is part of why its round-trip efficiency stays relatively modest, typically between 40 and 60% [1]. It remains, however, the most mature variant, with a technology readiness level of 9 (TRL 9) and several decades of commercial operation [1].
Adiabatic CAES (A-CAES): this approach aims to capture the heat produced during compression in a thermal storage system. That heat can then be reused to warm the air during expansion, reducing or eliminating the need for outside fuel [2]. This improved heat management makes efficiencies in the range of 70 to 75% achievable [2]. Adiabatic CAES, though, remains less mature than the diabatic setup. Reference [1] rates it at a technology readiness level of 5 (TRL 5), which still corresponds to a development and demonstration phase.
Isothermal CAES: this variant seeks to keep the air's temperature as stable as possible throughout compression and expansion [2]. By limiting thermal swings, it aims to cut the losses tied to heating and cooling the gas. A perfectly isothermal process, however, is hard to achieve under industrial conditions. The solutions being studied therefore rely on heat-exchange devices meant to bring real-world performance closer to this ideal behavior. Prototypes have been built, but the technology is not yet commercially available at scale [2].
Heat management also affects losses during the storage period itself. In an adiabatic system, the stored thermal energy can gradually dissipate, with losses estimated at 0.5 to 1% per day [1]. Diabatic CAES doesn't store this heat and so isn't exposed to the same kind of thermal loss over time. That doesn't mean its cycle is loss-free, though, since overall efficiency is still affected by the initial heat rejection, compression, expansion, and the fuel burned to reheat the air.
Underground Hydrogen Storage (UHS): Holding an Energy Carrier at Scale
Underground hydrogen storage is being studied as a way to hold large quantities of hydrogen over extended periods. Unlike surface tanks, whose capacity tends to be limited, certain geological formations can accommodate far larger volumes. This approach is especially useful when hydrogen production and consumption don't line up in time [3].
Why Store Hydrogen Underground?
UHS, short for Underground Hydrogen Storage, involves injecting gaseous hydrogen into a suitable underground formation, then recovering it once demand rises. The gas can then be used to power industrial processes, generate electricity, or meet other energy needs.
This approach is mainly considered for seasonal hydrogen storage and for managing volumes that would be hard to hold in surface equipment alone [3][6]. It could, for instance, allow part of the output from periods of high electricity availability to be accumulated and then drawn on weeks or months later.
Pilot projects have already examined how hydrogen behaves in different formations. As part of the Austrian Underground Sun Storage project, up to 82% of the hydrogen injected as part of a gas blend was recovered, with no significant rock-integrity issues observed during the trial [3]. This result demonstrates the technical feasibility of certain setups, but it can't be generalized to every reservoir, since performance varies with site characteristics.
The main benefits being studied include:
Potentially high storage capacity, suited to managing large hydrogen volumes.
Long-duration use, particularly for aligning variable production with seasonal needs.
A small surface footprint, since most of the storage volume sits underground.
The possibility of reusing existing infrastructure, particularly at former gas storage sites, subject to technical assessment and retrofit work [3].
The Three Types of Geological Formations for UHS
A geological formation can act as a reservoir when it has properties that allow it to hold pressurized gas and release it under controlled conditions. For geological hydrogen storage, three main categories are mainly under study: salt caverns, depleted reservoirs, and saline aquifers.
Salt caverns are among the most mature and favorable options for cyclic hydrogen storage. Salt has very low permeability and can gradually seal certain microfractures under pressure. Caverns can also withstand relatively frequent injection and withdrawal cycles [3]. In some configurations, estimated losses stay below 0.1% per year [10], though this figure can't be applied uniformly across all sites. Their main limitation lies in geographic distribution, since developing them requires sufficiently thick and suitable salt formations.
Depleted gas or oil reservoirs generally offer substantial capacity and can benefit from wells, pipelines, and geological data that already exist. Their behavior is also better understood than that of formations yet to be developed. Hydrogen can, however, interact with residual fluids, minerals, and microorganisms present in the reservoir. Part of the gas may also stay trapped in rock pores or mix with the cushion gas, which lowers the recoverable fraction and can increase purification needs [3].
Saline aquifers offer considerable volumetric potential and are found across many regions. That said, they remain less well characterized for hydrogen storage than salt caverns and depleted reservoirs. Using them requires detailed knowledge of the reservoir rock, the caprock, subsurface fluids, and possible biological reactions. In the most unfavorable scenarios studied for certain high-risk aquifers, losses tied to microbial activity could be substantial, though these figures represent upper bounds rather than typical values across all sites [10].
Technical Challenges of Underground Hydrogen Storage
Underground hydrogen storage depends on a geological formation's ability, along with its associated infrastructure, to hold the gas under pressure and then release it at adequate quality. This performance doesn't rely on available volume alone. It's also shaped by rock composition, the presence of water or residual gases, microbial activity, and how the reservoir behaves through injection and withdrawal cycles [3].
Several phenomena therefore need to be assessed:
The formation's mechanical integrity, since repeated pressure swings can gradually deform the cavern or alter the surrounding rock's behavior.
Geochemical reactions, which can occur between the hydrogen, minerals, brine, and other subsurface fluids.
Microbial activity, as some microorganisms can consume hydrogen and generate other gases, which may lower the recoverable amount or affect its purity.
Pore retention, especially in depleted reservoirs and aquifers, where part of the gas can stay trapped or blend with fluids already present.
Diffusion and potential leakage through the rock, the wells, or the caprock [3].
Mineral leaching, particularly in salt caverns containing impurities [3].
How much these mechanisms matter varies with a site's temperature, pressure, salinity, and mineralogy. Geological, geochemical, and microbiological characterization therefore helps estimate a reservoir's usable capacity and set suitable operating conditions. Monitoring pressure, gas composition, and well integrity then makes it possible to track its evolution across cycles.
Underground Natural Gas Storage (UGS): A Well-Established Energy Infrastructure
Underground natural gas storage, or UGS for Underground Gas Storage, has been used for decades to match supply with shifts in demand. As of April 2026, U.S. facilities held roughly 6.71 Tcf of natural gas, close to 190 billion cubic meters. About 2.21 Tcf of that was working gas available for withdrawal [4].
These volumes highlight how significant the existing infrastructure already is. It allows gas to be accumulated during low-consumption periods and then drawn on when demand rises, particularly in winter.
The Three Types of Formations for Natural Gas Storage
An underground gas storage facility, known as a UGSF in English, injects natural gas into a suitable geological formation to hold it under pressure. Two categories of gas need to be distinguished:
Working gas refers to the portion that can be injected and withdrawn over the course of operating cycles.
Cushion gas, also called base gas, stays in the reservoir to maintain adequate pressure and support the recovery of working gas.
Three main types of formations are used. They differ mainly in how much cushion gas they require, their withdrawal rate, and how many cycles they can support [5].
Table 3: Characteristics of the three types of formations for natural gas storage
| Formation | Development profile | Cushion gas required | Indicative cycles | Operational characteristics |
|---|---|---|---|---|
| Depleted reservoirs | Often eased by existing data and infrastructure | About 50% | 1 to 2 per year | High capacity, mainly seasonal use |
| Salt caverns | Specialized development, but strong deliverability | About 20 to 30% | Up to 10 to 12 per year | Fast injection and withdrawal, suited to frequent cycling |
| Aquifers | Characterization and development generally more complex | About 50 to 80% | 1 to 2 per year | Large potential capacity, but behavior less well known before development |
Depleted reservoirs are widely used because their geology has already been studied and some infrastructure can be repurposed. Salt caverns offer greater operational flexibility thanks to their strong withdrawal capacity and their ability to run multiple cycles per year. Aquifers can hold large volumes, but they typically require more characterization work and a higher proportion of cushion gas [5].
Current Capacity and Strategic Role
Underground natural gas storage mainly helps balance seasonal swings between supply and demand. It also supports supply security by making it possible to quickly draw on reserves when consumption rises or when a disruption hits transport networks [5][6].
These facilities also form an important knowledge base for developing underground hydrogen storage. Decades of natural gas operations have deepened understanding of reservoir behavior, the role of cushion gas, well integrity, and injection and withdrawal cycles [3].
Conclusion: Underground Solutions Suited to Different Needs
Underground energy storage brings together several approaches that serve distinct purposes. CAES returns electricity from compressed air, hydrogen storage focuses mainly on managing large volumes over extended periods, while natural gas storage already relies on widely deployed infrastructure. Salt caverns, depleted reservoirs, and aquifers each bring their own particular strengths in terms of capacity, flexibility, and frequency of use.
That said, no solution can be considered apart from its geological context. The nature of the formation, well integrity, the quality of the recovered fluid, and grid requirements all shape a project's feasibility and operating conditions. Reported performance figures should therefore be read in light of the site, the technology used, and its level of maturity.
As energy systems continue to evolve, the subsurface could play an even greater part in aligning production with demand, supporting seasonal supply, and integrating variable energy sources.
To stay informed on developments in this field and related topics in underground energy and natural resources, follow Squatex on LinkedIn.
References
[1] International Energy Agency Energy Storage Technology Collaboration Programme. "Technology: Compressed Air Energy Storage." IEA ES TCP, July 2024, https://iea-es.org/wp-content/uploads/public/FactSheet_mechanical_CAES.pdf.
[2] Rabi, Ayah Marwan, Jovana Radulovic, and James M. Buick. "Comprehensive Review of Compressed Air Energy Storage (CAES) Technologies." Thermo, vol. 3, no. 1, 2023, pp. 104–126. MDPI, https://doi.org/10.3390/thermo3010008.
[3] Doukeh, Rami, et al. "A Review of Hydrogen Storage in Geological Formations." Romanian Journal of Petroleum & Gas Technology, vol. 6, no. 77, 2025. https://doi.org/10.51865/jpgt.2025.01.22.
[4] U.S. Energy Information Administration. "U.S. Underground Natural Gas Storage by Storage Type (Million Cubic Feet)." EIA, Release Date 27 Feb. 2026, https://www.eia.gov/dnav/ng/ng_stor_type_s1_m.htm.
[5] Federal Energy Regulatory Commission. Current State of and Issues Concerning Underground Natural Gas Storage. Staff Report, 30 Sept. 2004, FERC, https://www.ferc.gov/sites/default/files/2020-05/UndergroundNaturalGasStorageReport.pdf.
[6] International Energy Agency Energy Storage Technology Collaboration Programme. Annual Report 2021. IEA ES TCP, 2022, https://iea-es.org/wp-content/uploads/public/ES_TCP_Annual_Report_2021.pdf.
