Geothermal Electric Power Generation in the United States

Histometrics industry primer, NAICS 2022 code 221116

This industry covers companies whose main business is running geothermal power plants, using heat from inside the Earth to make electricity. The primer is written for both public-market and private investors. Core industry figures are our own ingested U.S. Census statistics; other numbers carry inline citations to the Sources list at the end. North American Industry Classification System = NAICS.

---

1. Overview

Geothermal power plants tap naturally hot underground water and steam, or, in newer systems, engineered hot rock, to spin turbines and generate electricity around the clock. It is a tiny but distinctive corner of the U.S. power sector: about 0.4% of the nation's electricity [7] and roughly $1.1 billion of annual business revenue [2]. Yet it does something wind and solar cannot: it runs day and night, in any weather. That "clean firm" quality is why utilities, states, and data-center operators pay a premium for it.

For a public-market investor, the choices are narrow: one mature pure-play (Ormat Technologies), one newly public next-generation developer (Fervo Energy), and two giants where geothermal is a rounding error (Constellation Energy and Berkshire Hathaway). For private investors such as infrastructure funds, direct owner-operators, and public-private partnerships, the industry is a classic contracted-cash-flow business: buy an operating plant with a long sales contract, finance a plant expansion, purchase tax credits, or take venture-style risk on the emerging drilling technology.

---

2. What it is and how it's structured

In scope (221116): establishments whose primary activity is operating geothermal electric power generation facilities, producing electricity delivered to the transmission or distribution grid [1]. The economic unit is the establishment (a single operating location), not the whole corporate group, so one company with plants in three states counts as several establishments [1].

What it excludes. 221116 is one of eight sibling electricity-generation codes; the others are separate industries [1]:

It also excludes the wires business (221121 transmission, 221122 distribution) and, importantly for investors, three things people confuse with geothermal power:

• Geothermal heat pumps (ground-source heating and cooling for buildings) are installed under construction/HVAC codes (238220), not here. They move heat; they do not generate electricity.
• Geothermal district heating (piping hot water to buildings) is a thermal business, not power generation.
• Turbine and binary-cycle equipment manufacturing, which a vertically integrated player like Ormat also does, sits in manufacturing (333611), not utilities.

Ownership mix. Unlike nuclear or the old coal fleet, U.S. geothermal is overwhelmingly independent power, not utility rate-base. In 2024, independent power producers (IPPs, companies that own generation and sell it wholesale rather than to their own captive customers) generated 94.1% of geothermal electricity; traditional electric utilities generated just 5.9% [6]. But "independent" does not mean "merchant": nearly all output is sold under long-term contracts (see §5). The owners break down roughly as:

• Listed companies: Constellation Energy (via its Calpine acquisition) and Ormat Technologies own the largest fleets; Berkshire Hathaway holds the Salton Sea plants inside its energy arm.
• Private / infrastructure: independents such as Cyrq Energy, Terra-Gen, and EnergySource, plus infrastructure-fund-owned complexes like Coso (~135 MW) [23].
• Public power / municipal: the Northern California Power Agency (NCPA), a municipal joint-action agency, owns ~165 MW at The Geysers (Geothermal No. 1 at 110 MW and No. 2 at ~55 MW) [19]. Public-power utilities and California community-choice aggregators (CCAs, government-chartered local power buyers) are also major offtakers.
• Federal government: essentially no plant ownership, but the federal government is the dominant landlord. Most Western resources sit on land leased from the Bureau of Land Management (BLM); 51 plants operate on BLM land [8].
• Venture-backed: the enhanced-geothermal insurgents, Fervo (now public), Sage Geosystems, XGS Energy, funded by venture capital and, increasingly, data-center contracts.

---

3. How big it is

Two lenses matter, and they don't fully overlap.

Business statistics (our ground-truth Census figures). These count employer businesses classified in 221116:

A newer Census survey put 2023 revenue slightly higher, near $1.18 billion [3]. Either way, this is a small, capital-heavy, labor-light industry: revenue per employee is very high (roughly $730,000) because a handful of staff run a plant worth hundreds of millions of dollars [2][3]. Note the size standard: because the whole industry employs only ~1,565 people across 22 firms, almost every individual geothermal operator qualifies as a "small business" (250-employee threshold) for federal purposes even though its plants are enormously capital-intensive [4].

This is a highly concentrated industry. Our Census concentration data show the top four firms earn 83.6% of revenue (the four-firm concentration ratio, CR4), the top eight earn 97.7%, and the Herfindahl-Hirschman Index (HHI, a standard 0–10,000 concentration score) is 2,055, above the 1,800 line that federal merger guidelines call "highly concentrated" [5]. Energy-industry data tell the same story: Ormat and Calpine (now Constellation) together held about 69% of capacity and 61% of operating plants in 2024 [8].

Physical fleet (the more current gauge). For a plant-level view, use the energy data:

• Installed capacity: ~3,969 megawatts (MW) across 99 plants in 2024, up ~8% from 3,673 MW in 2020 (National Renewable Energy Laboratory, NREL); the U.S. Energy Information Administration (EIA) counts 4,002 MW across 163 generators on a slightly different basis [6][8]. The U.S. is the world's largest geothermal power market by installed capacity [8].
• Net generation: 15.4 terawatt-hours (TWh) in 2024, essentially flat-to-declining versus 16.4 TWh in 2023, as output at some legacy fields slowly declines even while capacity creeps up [6].
• Capacity factor, the share of maximum possible output a plant actually delivers, is geothermal's signature strength. The national fleet has historically run near 76% (2018), and well-managed portfolios such as Ormat's ran at ~84% in 2024, far above wind or solar, because geothermal is firm and dispatchable rather than weather-dependent [8][11].

Why the two lenses differ, and the undercount. The 48 business establishments understate the 99 physical plants because an establishment is a business location, not a power plant, and several owner types fall outside the employer-business census entirely: municipally owned plants (NCPA's), development companies with no operating payroll, and plants classified under a diversified parent's main industry. Revenue is only mildly undercounted: 15.4 TWh sold at roughly $80/MWh is about $1.2 billion of energy value, close to the Census receipts figure. The establishment and employment counts are the larger gap [2][3][6].

Geography. This is a Western industry. California and Nevada produced 93.5% of U.S. geothermal electricity in 2024 [6]:

Geothermal supplies about 5% of California's and roughly 9% of Nevada's in-state electricity [7].

---

4. The investable universe

Pure public plays are scarce. Only one company is a true geothermal pure-play (Ormat); a second (Fervo) is a pre-revenue technology bet that only went public in 2026; the other two listings bury geothermal inside vast portfolios.

Public companies (market data as of mid-July 2026; refresh before acting):

Two notes for public investors. First, Ormat trades like a growth-and-scarcity asset, not an income stock: at a mid-July 2026 price near $105, its trailing price-to-earnings (P/E) ratio was ~50× and its enterprise-value-to-EBITDA multiple ~18×, with a dividend yield under 0.5% [12]. Second, Fervo's ~$7.3 billion market value rests almost entirely on future projects: trailing revenue is near zero and free cash flow was roughly –$591 million [14]. There is no U.S.-listed geothermal-only exchange-traded fund (ETF); broad clean-energy funds hold Ormat as a minor weight [14].

Major private and other owners:

---

5. How the money works

Geothermal is capital-heavy, operating-light, and fuel-free: nearly all the cost is upfront drilling and the power block, and once built there is no fuel to buy. That shapes every part of the economics.

The revenue engine: long-term contracts. Most plants sell power under 15-to-30-year power purchase agreements (PPAs), long fixed-price contracts, with utilities, community-choice buyers, or corporate customers. Because the output is firm, it commands a premium over intermittent renewables. Recent conventional and next-generation PPAs have been signed around $70–$110 per megawatt-hour (MWh), including Fervo's Corsac Station at $107/MWh. Ormat reported new U.S. contracts above $100/MWh, up from roughly $60–$80/MWh in the prior five years [8][11]. Plants often earn extra value for capacity/reliability and for renewable energy certificates (RECs) on top of the energy price [11].

A special federal status. Under the Public Utility Regulatory Policies Act (PURPA), a small-power "qualifying facility" (QF) is normally capped at 80 MW, but geothermal QFs qualify at any size [20]. Utilities must buy their output at "avoided cost," a rate set by state public utility commissions (PUCs) [20]. This underwrote the 1980s buildout and still shapes contracting today.

The regulated-utility exception. Where a utility owns a geothermal plant directly (the ~6% minority), it earns under the standard rate-base formula: revenue requirement = operating costs + depreciation + taxes + (rate base × allowed return). The allowed return on equity (ROE) is set by the state commission. The California Public Utilities Commission (CPUC) set 2026–2028 authorized ROEs just under 10% for the state's big utilities [21]. A utility that merely buys geothermal under a PPA earns no equity return on the energy cost; it passes prudent purchased-power costs through to ratepayers [21].

The cost curve. Overnight capital cost runs from about $4,350 per kilowatt (kW) at the hottest, most productive hydrothermal sites to well above $9,000–$19,000/kW at cooler or deeper resources (NREL) [9]; a 50-MW expansion at an existing field runs around $3,963/kW (EIA) [10]. On a levelized basis, high-temperature "flash" plants come in around $63–$74/MWh and lower-temperature "binary" plants around $90–$110/MWh, above the cheapest wind and solar, but competitive once firmness is valued [8]. The dominant risk is front-loaded: exploration and drilling can be more than half of a new greenfield project's cost, and wells can come up dry before the project is financeable [10].

Tax credits are decisive for returns. Geothermal qualifies for the federal clean-electricity credits that replaced the old technology-specific ones under the Inflation Reduction Act (IRA): the Section 48E investment tax credit (ITC) and the Section 45Y production tax credit (PTC), a 30% base credit when wage and apprenticeship rules are met, plus adders for domestic content and "energy communities" [23]. A project picks one or the other. Crucially, when the 2025 federal budget law curtailed many clean-energy credits, geothermal was preserved. It keeps the full credit for projects that begin construction through roughly 2033–2034, phasing down thereafter [24]. Ormat has "safe-harbored" PTC eligibility for its projects through 2028 [11]. Credits can also be sold for cash to other investors under IRA transferability (Section 6418), and qualifying property gets accelerated (five-year) depreciation, both important levers for private and tax-equity investors [23].

What it looks like at a real operator (Ormat, FY2024): revenue $879.7 million, net income $123.7 million, adjusted EBITDA (earnings before interest, taxes, depreciation, and amortization) $550.5 million, a ~63% EBITDA margin, the hallmark of a fuel-free, contracted asset base, and a $0.12 quarterly dividend [11].

---

6. What drives demand

1. "Clean firm" mandates. California's clean-electricity law (Senate Bill 100, 100% clean power by 2045) and the CPUC's 2021 order requiring 1,000 MW of non-weather-dependent, non-battery, zero-emission capacity created a pull built specifically for geothermal [21]. A 2026 CPUC order added 6,000 MW of new clean/storage capacity, at least a quarter of it clean-firm or long-duration [21].
2. State renewable and clean-energy standards across the West and Hawaii, renewable portfolio standards (RPS) and clean energy standards (CES).
3. Data centers wanting 24/7 carbon-free power are the newest and most powerful driver. Google contracted 115 MW of enhanced geothermal via NV Energy with Fervo; Meta signed 150 MW with Sage Geosystems and 150 MW with XGS Energy [22]. These corporate deals are effectively financing the commercialization of new geothermal technology.
4. Baseload replacement as coal retires and buyers seek firm, non-emitting supply.
5. Mineral co-products. The Salton Sea brines carry lithium, and direct lithium extraction (DLE) could add a second revenue stream; Berkshire Hathaway Energy and Occidental formed a joint venture to test it [18]. Treat lithium as option value until proven at commercial scale, not as a rescue for weak power economics.

Since 2021 the industry has signed 27 PPAs totaling 1,661 MW, 61% of it next-generation technology, and drawn roughly $1.5 billion of private capital into next-generation developers [8]. But a signed PPA is not a built plant; it must still clear resource confirmation, permitting, interconnection, financing, and construction.

---

7. Regulation

Geothermal answers to federal energy regulators, federal land managers, and state utility commissions. It carries a heavier permitting load than solar or wind because it requires subsurface drilling, often on public land.

• Federal Energy Regulatory Commission (FERC): regulates wholesale power sales, market-based rate authority, and interconnection, and administers PURPA qualifying-facility status (geothermal qualifies at any size; projects of 1 MW or more generally file Form 556) [20].
• State public utility commissions (PUCs), including the CPUC in California and the Public Utilities Commission of Nevada, are often the decisive commercial regulators: they set avoided-cost rates, approve utility PPAs and cost recovery, set allowed ROEs, and issue the procurement mandates that create demand [21].
• Bureau of Land Management (BLM): the dominant landlord. Most Western geothermal is on federal land leased under the Geothermal Steam Act; 51 plants operate on BLM land, and public-land projects represent ~2,600 MW of capacity [8]. Royalties run about 1%–2.5% of gross proceeds in the first ten years and 2%–5% thereafter [26]. Leasing, review under the National Environmental Policy Act (NEPA), and drilling permits are a principal schedule risk.
• U.S. Environmental Protection Agency (EPA): regulates injection/disposal wells under the Class V Underground Injection Control (UIC) program (protecting drinking water), plus hydrogen sulfide (H₂S) emissions at steam fields and handling of brine constituents such as arsenic and boron [27].
• U.S. Geological Survey (USGS): runs the national resource assessment and leads on induced-seismicity guidance [28].
• Nuclear Regulatory Commission (NRC): not applicable. Geothermal plants are not nuclear facilities (that is the adjacent code, 221113).

The regulatory cost driver is mainly time: multi-year leasing and permitting, interconnection queues, and environmental review are the biggest non-drilling schedule risks.

---

8. Competitive dynamics and consolidation

The federal concentration data (HHI ~2,055; top four firms = 83.6% of revenue [5]) match the on-the-ground reality: a small oligopoly of incumbents at legacy fields, now consolidating further, with a wave of technology insurgents from below.

• Consolidation is accelerating. Constellation's ~$26.6 billion acquisition of Calpine (closed January 2026) folded the world's largest geothermal complex into a listed power giant chasing data-center load [15]. Ormat's January 2024 purchase of Enel Green Power North America's geothermal and solar assets, and its earlier ~$110 million acquisition of U.S. Geothermal (2018), further concentrated the field [11].
• Incumbent oligopoly at legacy fields: Constellation (The Geysers), Berkshire/BHE (Salton Sea), and Ormat (Nevada) hold high barriers: proven resources, existing wells, interconnection, and long-dated PPAs [8].
• Technology disruption from below. Enhanced geothermal systems (EGS) entrants such as Fervo, Sage, and XGS use oil-and-gas horizontal drilling and fracturing to engineer reservoirs outside the naturally permeable Western fields, potentially breaking the industry's geographic concentration. Since 2021, 12 next-generation PPAs totaling ~1,007 MW have been signed, including projects in Texas and New Mexico, east of the traditional resource base [8].
• Oil & gas crossover: shale-drilling expertise and balance sheets are migrating into EGS, a competitive threat and a capital tailwind at once.

---

9. Risks

• Resource and geologic risk. Exploration wells can fail; producing fields decline and need reinjection and make-up wells. Flat-to-declining national output despite rising capacity shows this in the aggregate [6].
• High, front-loaded capital and long lead times (5–10 years from exploration to operation), with financing risk before the resource is confirmed [10].
• Geographic concentration. ~93% of output in California and Nevada exposes the industry to two states' policy, drought, and seismic conditions [6].
• EGS execution risk. The ~1,007 MW next-generation pipeline is largely unbuilt; a successful pilot does not prove repeatable 100–500 MW performance or multi-decade reservoir life [8].
• Contract and counterparty risk. Legacy 1980s-era PPAs are rolling off and must be repriced; revenue can concentrate in a few buyers. Ormat's largest customers, the Southern California Public Power Authority, NV Energy, and Kenya Power, were 17.8%, 13.8%, and 11.9% of company revenue in 2025 [11].
• Policy and tax-credit risk. Returns lean on the ITC/PTC; geothermal was spared in 2025, but future changes and qualification errors (wage, domestic-content, construction-start, recapture) matter [24].
• Permitting, interconnection, and social license. Federal review and grid queues can strand a proven resource for years.
• Environmental and seismicity risk. Induced seismicity is a live concern for EGS after the Basel and Pohang precedents; H₂S emissions and brine/water management add cost [27][28].
• Interest-rate sensitivity. As a capital-intensive, yield-oriented asset class, valuations move with the cost of debt.
• Lithium overvaluation. Do not capitalize an unproven brine co-product before commercial recovery is demonstrated.

---

10. How to invest and the outlook

Public markets. Ormat (ORA) is the only clean way to own the theme: a contracted, high-margin operator, but priced as growth (P/E ~50×, yield <0.5%), not income [11][12]. Fervo (FRVO), public since its May 2026 initial public offering (80.5 million shares at $27, ~$2.2 billion raised), is the most direct next-generation bet but is a construction-and-technology story with near-zero current revenue and heavy cash burn [13][14]. Constellation (CEG) and Berkshire (BRK) offer only diluted exposure. There is no geothermal-only ETF.

Private markets. This is where most of the industry actually trades. Geothermal plants are textbook contracted infrastructure: long PPAs, high capacity factor, and inflation-escalated tariffs, which is why infrastructure funds and private equity (Energy Capital Partners was Calpine's former owner) are natural owners [15]. The lower-risk entries are operating-asset acquisitions (a proven field with a long PPA and a creditworthy buyer), brownfield repowering and expansion, and non-recourse project debt. Higher on the risk curve sit development equity (staged against drilling and permitting milestones) and early EGS venture equity. Two credit-based channels run alongside: tax-credit purchases (buying transferable 45Y/48E credits at a discount) and public-private structures with municipalities and public-power agencies (project companies, energy-service contracts, tax-exempt or taxable project debt). Data-center operators signing 15-year PPAs are effectively a private-capital channel that de-risks new development [22].

Near-term drivers to watch. Fervo's Cape Station (Utah), the world's largest EGS project (500 MW contracted), targets first power in late 2026 and ~100 MW by early 2027, with a 400 MW second phase in 2028 [13]. Whether EGS drilling costs fall toward the ~$3,000/kW and $45/MWh targets is the swing factor: if it works, geothermal graduates from a 0.4% niche to a meaningful clean-firm category; if it stalls, the industry stays a ~4 GW Western niche.

The long-run ambition, kept in perspective. The Department of Energy's GeoVision study modeled ~60 GW of U.S. geothermal by 2050, and later analysis stretched to ~90 GW, a large multiple of today's ~4 GW, while the Enhanced Geothermal Shot targets a 90% cost cut to $45/MWh by 2035 [25]. These are policy scenarios, not forecasts. No official growth rate exists for NAICS 221116 revenue, so we do not state one. The credible outlook is conditional: a small, firm, slow-growing incumbent business with a genuine technology option embedded in it, and unusually durable political support, since geothermal survived the 2025 credit rollback and carries bipartisan permitting and critical-minerals interest.

---

Sources

1. U.S. Census Bureau, "2022 NAICS: 221116 — Geothermal Electric Power Generation" (scope; adjacent codes). https://www.census.gov/naics/?details=221116&input=221116&year=2022
2. U.S. Census Bureau, "2022 Economic Census, EC2200BASIC — Geothermal Electric Power Generation," 2024 (receipts $1,144,273 thousand; 22 employer firms). Histometrics ingested ground-truth statistic. https://data.census.gov/table/ECNBASIC2022.EC2200BASIC
3. U.S. Census Bureau, 2023 County Business Patterns / Annual Integrated Economic Survey, NAICS 221116 (48 establishments; 1,565 employees; annual payroll $199,277 thousand; ~$1.18 billion revenue). Histometrics ingested ground-truth statistic. https://data.census.gov/table/CBP2023.CB2300CBP
4. U.S. Small Business Administration, "Table of Small Business Size Standards" (250-employee standard for 221116). Histometrics ingested ground-truth statistic. https://www.sba.gov/document/support-table-size-standards
5. U.S. Census Bureau, 2022 Economic Census — Concentration statistics, NAICS 221116 (CR4 83.6%; CR8 97.7%; CR50 100%; HHI 2,055.1). Histometrics ingested ground-truth statistic. https://data.census.gov/
6. U.S. Energy Information Administration, "Electric Power Annual," Tables 3.20 / 4.3, 2024 data (15.407 TWh net generation; 4,002 MW capacity; state and IPP/utility shares). https://www.eia.gov/electricity/annual/
7. U.S. Energy Information Administration, "Geothermal Explained: Use of Geothermal Energy," 2025 (0.4% of U.S. generation; state shares). https://www.eia.gov/energyexplained/geothermal/use-of-geothermal-energy.php
8. National Renewable Energy Laboratory (NREL) & Geothermal Rising, "2025 U.S. Geothermal Power Production and District Heating Market Report," 2026 (3,969 MW / 99 plants; CA/NV shares; Ormat + Calpine ~69% capacity; 51 BLM plants; 27 PPAs / 1,661 MW and 12 EGS PPAs / 1,007 MW since 2021; LCOE ranges; recent PPA prices). https://www.nrel.gov/geothermal/
9. NREL, "2024 Annual Technology Baseline: Geothermal" (overnight capital ~$4,350/kW hot to ~$19,000/kW low-temp, 2022$). https://atb.nrel.gov/electricity/2024/geothermal
10. U.S. Energy Information Administration, "Capital Cost and Performance Characteristics for Utility-Scale Electric Power Generating Technologies," AEO 2025 ($3,963/kW brownfield binary; exploration/drilling >50% of greenfield cost; fixed O&M ~$150.60/kW-yr). https://www.eia.gov/analysis/studies/powerplants/capitalcost/
11. Ormat Technologies, Form 10-K and FY2024/FY2025 results (FY2024 revenue $879.7M, net income $123.7M, adjusted EBITDA $550.5M; $0.12 quarterly dividend; ~84% capacity factor; PPA pricing; customer concentration; PTC safe-harbor through 2028). https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001296445
12. Stock Analysis, "Ormat Technologies Statistics," market data as of July 17, 2026 (market cap ~$6.4B; P/E ~50×; EV/EBITDA ~18×; dividend yield ~0.5%). https://stockanalysis.com/stocks/ora/statistics/
13. Fervo Energy, "First-Quarter 2026 Results" and IPO disclosures (May 2026 IPO: 80.5M shares at $27, ~$2.2B gross; 658 MW binding PPAs; Cape Station Phase I 100 MW late 2026 / Phase II 400 MW 2028; $421M non-recourse Phase I financing). https://ir.fervoenergy.com/
14. Stock Analysis, "Fervo Energy Statistics," market data as of July 17, 2026 (market cap ~$7.3B; trailing revenue ~$0.2M; free cash flow ~–$591M); First Trust QCLN holdings. https://stockanalysis.com/stocks/frvo/statistics/
15. Constellation Energy, "Constellation Completes Calpine Transaction," January 2026 (~$26.6B; ~55 GW combined portfolio). https://www.constellationenergy.com/newsroom.html
16. Calpine / The Geysers, "About The Geysers," 2025–2026 (13 plants; ~730 MW; world's largest geothermal complex). https://geysers.com/about/
17. BHE Renewables (Berkshire Hathaway Energy), "Geothermal Power" (10 Imperial Valley plants, ~345 MW). https://www.bherenewables.com/geothermal_power.aspx
18. Reuters, "Occidental Petroleum and Berkshire Hathaway Form JV to Extract Lithium," 2024. https://www.reuters.com/markets/commodities/occidental-petroleum-berkshire-hathaway-form-jv-to-extract-lithium-2024-06-04/
19. California Energy Commission, "NCPA Geothermal Project No. 1 and No. 2" (110 MW + ~55 MW). https://www.energy.ca.gov/
20. Federal Energy Regulatory Commission, "Qualifying Facilities" (geothermal QFs qualify at any size; Form 556; avoided cost). https://www.ferc.gov/qf
21. California Public Utilities Commission, procurement decisions and cost-of-capital orders, 2025–2026 (1,000 MW clean-firm order; 2026 6,000 MW procurement; 2026–2028 authorized ROEs). https://www.cpuc.ca.gov/
22. Meta / Canary Media / Reuters Events, data-center geothermal offtakes (Meta–Sage 150 MW; Meta–XGS 150 MW; Google–Fervo 115 MW via NV Energy). https://about.fb.com/news/2024/08/new-geothermal-energy-project-to-support-our-data-centers/
23. Internal Revenue Service, "Clean Electricity Production Credit (§45Y)" and "Investment Credit (§48E)," updated 2026; transferability (§6418); five-year cost recovery. https://www.irs.gov/credits-deductions/clean-electricity-production-credit
24. Canary Media / ThinkGeoEnergy, geothermal preserved in the 2025 federal budget law; §45Y/§48E begin-construction runway to ~2033–2034 then phasedown. https://www.canarymedia.com/articles/geothermal
25. U.S. Department of Energy, "GeoVision: Harnessing the Heat Beneath Our Feet" (60 GW by 2050; later ~90 GW) and "Enhanced Geothermal Shot" ($45/MWh by 2035). https://www.energy.gov/eere/geothermal/geovision
26. Bureau of Land Management, "Geothermal Energy Program," and 30 U.S.C. §1004 (royalty ranges 1%–2.5% then 2%–5%). https://www.blm.gov/programs/energy-and-minerals/geothermal-energy
27. U.S. Environmental Protection Agency, "Class V Underground Injection Control" (geothermal injection wells; brine and H₂S handling). https://www.epa.gov/uic
28. U.S. Geological Survey, "Induced Seismicity" (monitoring and response for injection-related seismicity). https://www.usgs.gov/programs/earthquake-hazards/induced-earthquakes