Resources / References / 03
Fervo Energy: Big Tech’s Clean Energy Bet Is Unraveling
E² review · 2026-10-06
Claims ledger
Each record separates the claim’s type, its support and the check actually performed. “Reported fact” describes an attributable statement; it does not certify the underlying event. Read the rubric.
Symbol key
Symbols supplement the words; they do not replace them.
Reported source Assumption Calculation Measurement
Source checked Arithmetic reproduced Conflict Check still open
Support uses one to five filled marks, from conjectural to established within scope. This is an ordinal evidence scale, not a probability. A dash means the scale does not apply.
- C01
Historical EGS projects have faced thermal decline, water loss and induced seismicity.
- Type
- Reported fact
- Support
- Supported
- Review
- Source checked
The technical paper and company seismic material recognize these mechanisms. Historical failures identify real risks. They do not establish that a different well geometry, site or mitigation system must fail.
Evidence and revision conditions
- Fercho et al., Stanford Geothermal Workshop 2023 — pp. 14–15; heat-in-place assumptions and 5.1 MW model
- Fervo April 2026 seismic update — June 29, 2026; 36 yellow events, ground motion, monitoring changes and mitigation
What would change this assessment: A project-level comparison showing which historical conditions recur at Cape Station and which have changed.
- C02
Fervo raised approximately $2.2 B in its IPO and has more than 1 GW of binding PPAs.
- Type
- Reported fact
- Support
- Supported
- Review
- Source checked
Company disclosures support the financing and contracted-capacity attributions. A signed PPA is not proof of sustained engineering performance or profitable delivery. Neither figure was independently reconciled to executed contracts in this review.
Evidence and revision conditions
- Fervo second-quarter 2026 results — IPO gross proceeds, installed-cost targets and development timelines
- Fervo commercial-operation announcement — October 1, 2026; attributed September 30 COD and 33 MW net output; not independently metered here
What would change this assessment: The final capital reconciliation or complete executed PPA schedule contradicting these disclosures.
- C03
Fervo trades at 8× estimated 2029 sales, a 179% premium to peers.
- Type
- Calculated comparison
- Support
- Not applicable
- Review
- Not reproduced
The report compares forward sales estimates with market values. This review did not obtain a frozen price, fully diluted capitalization, peer universe or consistent forecast table. The premium cannot be replicated from its headline alone.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
What would change this assessment: Publish the timestamped price/share count and identical valuation definitions for every peer.
- C04
Fervo’s 2024 temperature chart conflicts with monthly Nevada submissions for the same production well.
- Type
- Inference
- Support
- Reported
- Review
- Unresolved interpretation
The Nevada workbook contains abrupt monthly temperature changes, including 346°F to 256°F in March 2024. The paper presents a steadier series. Monthly regulatory averages and selected operating readings do not necessarily measure the same quantity. Sensor, weighting, operating-state and well mappings must be reconciled before declaring the readings contradictory.
Evidence and revision conditions
- Nevada geothermal production workbook — GeothermalProduction permit 1506 and GeothermalInjection permit 1505; December 2023–July 2026; retrieved October 6
- Norbeck, Gradl and Latimer, September 2024 preprint — Version 1; Figs. 3 and 9–11; initial test data and estimated electrical output
What would change this assessment: Release the daily series, operating-state filters, sensor locations and a month-by-month bridge to permit 1506.
- C05
Fervo deliberately altered temperature data to hide underperformance.
- Type
- Inference
- Support
- Conjectural
- Review
- Not established
Morpheus proposes concealment as an explanation for discrepant presentations. No original data file, edit history or evidence of intent was authenticated. Operational filtering, measurement definitions and erroneous filings remain live alternatives. A mismatch warrants investigation without settling motive.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
- Nevada geothermal production workbook — GeothermalProduction permit 1506 and GeothermalInjection permit 1505; December 2023–July 2026; retrieved October 6
- Norbeck, Gradl and Latimer, September 2024 preprint — Version 1; Figs. 3 and 9–11; initial test data and estimated electrical output
Depends on C04.
What would change this assessment: An authenticated data trail showing unjustified alteration, with the measurement alternatives resolved.
- C06
Water injection increased about 33% in March 2024.
- Type
- Calculated output
- Support
- Not applicable
- Review
- Replication completed
Permit 1505 reports 31,972,320 gallons in February and 42,656,079 in March. The increase is 33.42%. This reproduces the monthly-volume comparison. March has more days; the corresponding daily-volume increase is smaller. It does not establish which operating decision caused the change.
Evidence and revision conditions
- Nevada geothermal production workbook — GeothermalProduction permit 1506 and GeothermalInjection permit 1505; December 2023–July 2026; retrieved October 6
What would change this assessment: Corrected regulator rows or a documented change in the reporting units or period.
- C07
The injection–production imbalance rose from about 21% to 57% in March 2024.
- Type
- Calculated output
- Support
- Not applicable
- Review
- Replication completed
Using the matched monthly permit rows gives 21.17% in February and 56.86% in March. These are operational balances, not measurements of permanent aquifer depletion. Reservoir storage, timing and plant boundaries remain relevant.
Evidence and revision conditions
- Nevada geothermal production workbook — GeothermalProduction permit 1506 and GeothermalInjection permit 1505; December 2023–July 2026; retrieved October 6
Depends on C06.
What would change this assessment: A corrected balance or evidence that the paired meters do not describe the same circulation system.
- C08
The reported monthly production temperature fell 90°F in March 2024.
- Type
- Reported measurement
- Support
- Supported
- Review
- Source checked
The workbook’s TempF entry falls from 346 to 256. The same 90°F difference occurs between May and June 2026. These authenticate operator-submitted monthly readings. Their physical explanation and consistency with daily operating data remain unresolved.
Evidence and revision conditions
- Nevada geothermal production workbook — GeothermalProduction permit 1506 and GeothermalInjection permit 1505; December 2023–July 2026; retrieved October 6
What would change this assessment: Corrected submissions or a sensor/operating log explaining what each monthly temperature represents.
- C09
Latimer’s statement that the pilot met all technical objectives was false.
- Type
- Inference
- Support
- Reported
- Review
- Unresolved interpretation
The 2023 modeling target and later averages differ materially. But “technical objectives” can refer to drilling, connection and demonstration rather than the modeled average output. Fervo describes the pilot as a learning platform. This review did not authenticate the full interview or a prospectively fixed objective list.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
- Fercho et al., Stanford Geothermal Workshop 2023 — pp. 14–15; heat-in-place assumptions and 5.1 MW model
- Fervo, two years of Project Red production — April 13, 2026; gross/net averages, pilot purpose, 70% recapture and models
What would change this assessment: The dated objective specification and full interview establishing the intended claim and its test.
- C10
November 2025 to May 2026 temperatures imply a 24°F annualized decline.
- Type
- Calculated output
- Support
- Not applicable
- Review
- Replication completed
The monthly entries fall from 344°F to 332°F across six calendar months: 12°F × 2 = 24°F/year. A regression over those seven monthly entries gives about 24.43°F/year. Both describe this interval. They do not establish a persistent reservoir depletion rate, forecast error or Cape Station economics.
Evidence and revision conditions
- Nevada geothermal production workbook — GeothermalProduction permit 1506 and GeothermalInjection permit 1505; December 2023–July 2026; retrieved October 6
What would change this assessment: A longer operating-adjusted series and an independently tested temperature-decline model.
- C11
The annualized decline is about 16× a 1.5°F/year viability benchmark.
- Type
- Calculated comparison
- Support
- Not applicable
- Review
- Arithmetic reproduced
24 divided by 1.5 is 16. The calculation matches the report’s comparator. Economic viability cannot be inferred from this multiplier until the comparator is justified for this well geometry, resource, power cycle and cost structure.
Evidence and revision conditions
- Nevada geothermal production workbook — GeothermalProduction permit 1506 and GeothermalInjection permit 1505; December 2023–July 2026; retrieved October 6
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
What would change this assessment: A validated site-specific economic threshold replacing the reported expert benchmark.
- C12
A decline of about 1.5°F/year is the economic limit for Fervo’s EGS designs.
- Type
- Inference
- Support
- Reported
- Review
- Not validated
The report cites expert views, including traditional hydrothermal experience. It does not supply a discounted production/economic model establishing a universal cutoff. Different starting temperatures, flow rates, well costs and useful lives can produce different limits.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
- Fervo, two years of Project Red production — April 13, 2026; gross/net averages, pilot purpose, 70% recapture and models
What would change this assessment: A reproducible model linking thermal trajectories to project cash flow, with sensitivity to resource and design.
- C13
Average gross pilot output is about 41% of the 2023 modeled 5.1 MW figure.
- Type
- Calculated comparison
- Support
- Not applicable
- Review
- Arithmetic reproduced
2.1 / 5.1 = 41.18%. The 5.1 MW figure is a heat-in-place calculation assuming 35% recovery, 15% conversion efficiency and a ten-year life. The 2.1 MW figure is a later reported operating average. Their ratio is not a measured capacity factor or a matched out-of-sample model validation.
Evidence and revision conditions
- Fercho et al., Stanford Geothermal Workshop 2023 — pp. 14–15; heat-in-place assumptions and 5.1 MW model
- Fervo, two years of Project Red production — April 13, 2026; gross/net averages, pilot purpose, 70% recapture and models
What would change this assessment: A prospectively defined operating forecast and calendar energy series using the same gross/net and outage boundaries.
- C14
Net pilot output below 50% of its stated capacity predicts difficulty meeting a roughly 70% PPA threshold.
- Type
- Inference
- Support
- Conjectural
- Review
- Unresolved interpretation
The report divides 1.4 MW net by a prospectus figure of 3 MW gross. Using 2.1 MW gross against 3 MW gross gives 70%; neither comparison establishes calendar capacity factor. The commercial design and contractual excused-energy rules also differ. This comparison cannot establish a delivery default.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
- Fervo, two years of Project Red production — April 13, 2026; gross/net averages, pilot purpose, 70% recapture and models
- Fervo final IPO prospectus — May 14, 2026; production disclosure and Risk Factors, transmission and PPA deadlines
What would change this assessment: Comparable net nameplate, calendar generation, commercial well data and the executed PPA’s full conditions.
- C15
Project Red recaptures about 70% of injected water under steady-state conditions.
- Type
- Reported measurement
- Support
- Supported
- Review
- Source checked
Fervo itself discloses approximately 70% recapture. That supports a substantial short-term circulation imbalance. Its proposed reservoir-storage explanation and eventual recovery remain hypotheses requiring a longer, closed-boundary balance.
Evidence and revision conditions
- Fervo, two years of Project Red production — April 13, 2026; gross/net averages, pilot purpose, 70% recapture and models
What would change this assessment: A verified long-term water balance showing recoverability, storage and exchange with surrounding formations.
- C16
Commercial multi-well configurations can approach 99% water recapture.
- Type
- Model assumption
- Support
- Not applicable
- Review
- Source checked
ResFrac models alternating injectors and producers, including pumped outer wells that can pull surrounding fluid into the system. This is an inspectable design mechanism with a commercial relationship to Fervo. It is neither an independent field demonstration nor a guarantee for Cape Station.
Evidence and revision conditions
- ResFrac, well configuration and EGS water loss — Model discussion; outer producers, pumping, configuration and limits; supplier with commercial relationship
- Fervo, two years of Project Red production — April 13, 2026; gross/net averages, pilot purpose, 70% recapture and models
What would change this assessment: Publish measured multi-well recovery, pressure and pumping power over a specified operating interval.
- C17
The June 2026 pilot injection–production imbalance was approximately 35%.
- Type
- Calculated output
- Support
- Not applicable
- Review
- Replication completed
June shows 27,931,680 gallons injected and 18,244,800 produced: 34.68% imbalance. July falls to 22.68%. The June value reproduces, but a selected high month does not establish a monotonic deterioration or permanent water loss.
Evidence and revision conditions
- Nevada geothermal production workbook — GeothermalProduction permit 1506 and GeothermalInjection permit 1505; December 2023–July 2026; retrieved October 6
What would change this assessment: Corrected meter submissions or a reconciled field-wide balance.
- C18
The cumulative pilot injection–production imbalance is about 193 million gallons.
- Type
- Calculated output
- Support
- Not applicable
- Review
- Replication completed
The signed monthly balances from December 2023 through July 2026 total 192,872,191 gallons. Summing only positive deficits gives 206,737,068. Early production exceeds injection; July 2024 has unusually small reported volumes. The net sum reproduces the report, subject to those unresolved reporting boundaries.
Evidence and revision conditions
- Nevada geothermal production workbook — GeothermalProduction permit 1506 and GeothermalInjection permit 1505; December 2023–July 2026; retrieved October 6
What would change this assessment: A corrected series, unit audit and full accounting of initial storage, external flow and meter coverage.
- C19
Persistently high imbalance proves reservoir pressurization has not worked.
- Type
- Inference
- Support
- Reported
- Review
- Unresolved interpretation
Persistent imbalance is evidence against rapid convergence under this pilot configuration. It does not settle eventual recapture, subsurface retention or a different multi-well geometry. The report needs a dated predicted trajectory to test the pressurization hypothesis rather than treating any current deficit as its refutation.
Evidence and revision conditions
- Nevada geothermal production workbook — GeothermalProduction permit 1506 and GeothermalInjection permit 1505; December 2023–July 2026; retrieved October 6
- ResFrac, well configuration and EGS water loss — Model discussion; outer producers, pumping, configuration and limits; supplier with commercial relationship
What would change this assessment: A frozen predicted recovery curve and observed balance over its stated horizon, including pressure measurements.
- C20
June 2026 water imbalance reached a ten-month high.
- Type
- Reported measurement
- Support
- Conjectural
- Review
- Contradicted as labeled
Within September 2025–June 2026, September’s 37.82% exceeds June’s 34.68%. The claimed ten-month maximum does not reproduce in the retrieved workbook. This discrepancy does not erase the underlying substantial June imbalance.
Evidence and revision conditions
- Nevada geothermal production workbook — GeothermalProduction permit 1506 and GeothermalInjection permit 1505; December 2023–July 2026; retrieved October 6
Depends on C17.
What would change this assessment: The report’s original source edition, a justified exclusion rule or corrected regulator rows producing its maximum.
- C21
Former employees report worse-than-expected water loss and weak Cape Station test results.
- Type
- Alleged leak
- Support
- Reported
- Review
- Not independently verified
The full interview excerpts are preserved as Morpheus attributions. Identities, recordings, access to measurements and dates were not authenticated. Multiple quotations from one report do not become independent corroboration in this review.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
What would change this assessment: Authenticated dated testimony, test records and a company response addressing the same wells and conditions.
- C22
Fervo’s Frisco presentation publishes only initial power data from a thirty-day test.
- Type
- Reported fact
- Support
- Supported
- Review
- Source checked
The 2024 preprint labels its flow and estimated electrical-power plots “initial” data from the 30-day test. Electrical power is inferred from thermal output and anticipated plant performance. This supports a disclosure limitation; it does not prove the unseen interval performed poorly.
Evidence and revision conditions
- Norbeck, Gradl and Latimer, September 2024 preprint — Version 1; Figs. 3 and 9–11; initial test data and estimated electrical output
- Fervo, Frisco production test announcement — September 2024; thirty-day test and initial power estimate
What would change this assessment: The complete test time series, conversion model, net pumping demand and water balance.
- C23
A Bearskin test was completed by October 2025 and its results were withheld.
- Type
- Inference
- Support
- Reported
- Review
- Unresolved interpretation
Morpheus infers timing from a credit agreement that this review did not retrieve in full. Fervo’s seismic update dates Bearskin stimulation to November–December 2025. Those may concern different test stages. The claimed chronology and intentional withholding need a document-by-document reconciliation.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
- Fervo April 2026 seismic update — June 29, 2026; 36 yellow events, ground motion, monitoring changes and mitigation
What would change this assessment: The executed credit milestones, test dates and complete Bearskin reports.
- C24
Cape Station’s water access depends on a non-consumptive designation.
- Type
- Reported fact
- Support
- Reported
- Review
- Not independently verified
Utah’s primary materials confirm that water and geothermal use require regulatory treatment. The specific application and approval cited by Morpheus were not authenticated: its document links returned 403. The exact allocation, conditions, depletion accounting and backup rights remain unverified here.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
- Utah geothermal power-production rules — General permitting and operating requirements; not the specific Cape Station water-right order
- Utah Area 71 water policy — Milford district appropriation restrictions; does not authenticate the specific non-consumptive grant
What would change this assessment: The approved water-right order and application, including monitoring, return-flow and enforcement conditions.
- C25
Leak-off cannot recharge the source aquifer on a commercially useful timescale.
- Type
- Inference
- Support
- Reported
- Review
- Not validated
A six-thousand-foot vertical separation makes hydraulic communication a material question. Expert quotations alone do not determine permeability, pressure gradients or travel times. Operational recapture, aquifer storage and legal non-consumptive classification are different propositions.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
- ResFrac, well configuration and EGS water loss — Model discussion; outer producers, pumping, configuration and limits; supplier with commercial relationship
Depends on C24.
What would change this assessment: A site-specific hydrogeologic model validated with pressure, tracer and long-term water-balance observations.
- C26
A $99/MWh PPA against $107/MWh levelized cost means loss on every MWh sold.
- Type
- Inference
- Support
- Conjectural
- Review
- Unresolved interpretation
Lazard’s $107/MWh illustrative new-geothermal point uses $7,000/kW capex and $160/kW-year fixed O&M. The report’s $7,000/kWh wording has the wrong capital-cost unit. The generic estimate is not Fervo’s audited project cash cost. Tax credits, financing, escalation and generation matter; levelized total cost also differs from marginal operating cost. The PPA price remains a Jefferies attribution.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
- Lazard LCOE+ 2026 — Printed pp. 7–8; $107 illustrative new-geothermal point, $7,000/kW and $160/kW-year assumptions, tax-credit sensitivity
What would change this assessment: A matched project cash-flow model and executed price schedule, with tax credits and financing stated explicitly.
- C27
Reported installed capex rose from $5,000/kW to $7,000/kW.
- Type
- Calculated comparison
- Support
- Not applicable
- Review
- Not reproduced
The report juxtaposes 2024 conference comments, a 2025 interview and 2026 financial disclosures. This review did not reconcile scope, phase, gross/net capacity, actual spend and estimates across all three. A changed estimate may reflect overruns, design changes or a changed denominator.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
- Fervo second-quarter 2026 results — IPO gross proceeds, installed-cost targets and development timelines
What would change this assessment: Comparable dated project cost schedules and capacity definitions, with a bridge between estimates and actual spend.
- C28
Reducing installed capex from $7,000/kW to $3,000/kW requires a 57% reduction.
- Type
- Calculated output
- Support
- Not applicable
- Review
- Arithmetic reproduced
(7,000 − 3,000) / 7,000 = 57.14%. The arithmetic is sound. The long-term figure remains a target; no demonstrated completion-cost learning curve or achieved installed cost establishes that reduction.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
- Fervo second-quarter 2026 results — IPO gross proceeds, installed-cost targets and development timelines
What would change this assessment: A completed project achieving the target on a comparable all-in cost and capacity basis.
- C29
The changed drilling-cost chart conceals rising costs.
- Type
- Inference
- Support
- Reported
- Review
- Not established
The report reproduces two chart editions and alleges changes to wells 14–15. Their numerical source tables, well IDs and full 2024 conference version were not obtained. Different chart boundaries or corrections must be ruled out before concluding concealment.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
- Fervo final IPO prospectus — May 14, 2026; production disclosure and Risk Factors, transmission and PPA deadlines
- Norbeck, Gradl and Latimer, September 2024 preprint — Version 1; Figs. 3 and 9–11; initial test data and estimated electrical output
What would change this assessment: Both original series, a well-ID mapping, scope definitions and a dated explanation of revisions.
- C30
Completions cost roughly $3,000/kW and are a hidden obstacle to the target.
- Type
- Alleged leak
- Support
- Reported
- Review
- Not independently verified
The quoted employee allocates $2,000/kW each to plant and drilling and $3,000/kW to fracturing. No invoices or cost schedule authenticate these amounts. The report’s surrounding prose once uses “per kilowatt hour” for this capital figure; that is a dimensional error, not a different cost estimate.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
What would change this assessment: An audited cost breakdown and like-for-like completed-well comparison.
- C31
Late-operation damages can exceed half a million dollars per week.
- Type
- Calculated output
- Support
- Not applicable
- Review
- Arithmetic reproduced
The PDF’s contract screenshot states $700 per MW per day for days 121–180. At 115 MW this gives $563,500 per seven days. The screenshot supports conditional arithmetic. The full executed contract, amendments, excused delay and applicability were not independently authenticated.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
- Reported NV Energy/Google PPA excerpt — Supplied report PDF pp. 46–47 and footnotes 12–14. Full contract link returned 403; conditional calculations only
What would change this assessment: The complete signed agreement and project-specific capacity, deadline and exception schedule.
- C32
872,140 MWh/year from 115 MW implies an 86.6% capacity factor.
- Type
- Calculated output
- Support
- Not applicable
- Review
- Arithmetic reproduced
872,140 / (115 × 8,760) = 86.57%. This reproduces the report’s contract illustration for a 365-day year. It does not verify delivered generation, and a leap-year or contract-specific period may differ.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
- Reported NV Energy/Google PPA excerpt — Supplied report PDF pp. 46–47 and footnotes 12–14. Full contract link returned 403; conditional calculations only
What would change this assessment: The executed supply schedule and the agreement’s year definition.
- C33
80% of the stated supply amount corresponds to roughly 69.3% of nameplate annual energy.
- Type
- Calculated output
- Support
- Not applicable
- Review
- Arithmetic reproduced
The conditional threshold is 697,712 MWh, or 69.26% of 115 MW × 8,760 hours. The report’s own footnote mentions excused product and a two-year measurement period. Actual default depends on those definitions, cure rights and amendments, not this ratio alone.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
- Reported NV Energy/Google PPA excerpt — Supplied report PDF pp. 46–47 and footnotes 12–14. Full contract link returned 403; conditional calculations only
Depends on C32.
What would change this assessment: The full default clause and actual net deliveries over the contract measurement period.
- C34
Failure to commence operation within six months can permit PPA termination.
- Type
- Reported fact
- Support
- Supported
- Review
- Source checked
The prospectus discloses counterparty termination rights if applicable commercial deadlines are missed by six months, with potential penalties. This verifies disclosed contract risk. It does not show that a current deadline was missed or that every agreement has identical exceptions.
Evidence and revision conditions
- Fervo final IPO prospectus — May 14, 2026; production disclosure and Risk Factors, transmission and PPA deadlines
What would change this assessment: Executed agreements and actual delivery dates contradicting the risk disclosure.
- C35
The prospectus discloses about 290 MW of transmission rights, with additional capacity still needed.
- Type
- Reported fact
- Support
- Supported
- Review
- Source checked
The May prospectus states 290 MW, enough for original SCE/CPA PPAs but insufficient for their expanded 384 MW. It also discloses risk from unavailable additional transmission and alternative delivery. The report’s implied contradiction should retain those qualifications.
Evidence and revision conditions
- Fervo final IPO prospectus — May 14, 2026; production disclosure and Risk Factors, transmission and PPA deadlines
What would change this assessment: The rights’ nodes, duration, deliverability status and a complete contractual/transmission reconciliation.
- C36
California delivery faces a retiring Utah path and contingent new transmission.
- Type
- Reported fact
- Support
- Supported
- Review
- Source checked
Fervo’s August post acknowledges that its Utah CAISO delivery point will retire in 2027 and remaining paths are limited. Its CAISO comment conditions a pre-2032 pathway on timely interregional planning. This substantiates transmission risk; it does not alone settle every existing PPA or alternative route.
Evidence and revision conditions
- Fervo delivery and transmission update — August 19, 2026; retiring Utah delivery point and alternative routes
- Fervo comments in CAISO transmission planning — Fervo comments; contingent interregional study and pre-2032 delivery; whole surrounding comment inspected
Depends on C35.
What would change this assessment: Binding replacement transmission and executed customer arrangements with a dated delivery schedule.
- C37
Guidance for California power delivery has moved to “2030+”.
- Type
- Reported fact
- Support
- Reported
- Review
- Not independently verified
The report quotes a presentation graphic. This review verified the transmission constraint but did not authenticate the complete updated presentation. Fervo’s Q2 release still describes Phase II development toward 2028. Plant completion and delivery into California need separate timelines.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
- Fervo delivery and transmission update — August 19, 2026; retiring Utah delivery point and alternative routes
- Fervo second-quarter 2026 results — IPO gross proceeds, installed-cost targets and development timelines
Depends on C36.
What would change this assessment: The dated presentation and a reconciliation distinguishing plant COD, CAISO delivery and customer amendments.
- C38
The published Cape catalogue contains 58 yellow-level events through April 2026.
- Type
- Calculated output
- Support
- Not applicable
- Review
- Replication completed
The retrieved company-hosted catalogue yields 58 events at magnitude ≥2 and <3. Of these, 36 fall in November 2025–April 2026, matching Fervo’s update. The catalogue is a company-origin measurement series, not independent verification of causal attribution or ground damage.
Evidence and revision conditions
- Cape Seismic Events public data layer — Company-hosted layer discovered from public map; 58 retrieved magnitude≥2 records; frozen snapshot and IDs in inputs.json
- Fervo April 2026 seismic update — June 29, 2026; 36 yellow events, ground motion, monitoring changes and mitigation
What would change this assessment: A revised frozen catalogue, explicit magnitude definitions or a different justified traffic-light classification.
- C39
March and April account for 29 yellow-level events, exceeding the earlier period.
- Type
- Calculated comparison
- Support
- Not applicable
- Review
- Unresolved conflict
Using magnitude ≥2 gives 30 events in March–April and 28 earlier. Using >2 gives 29 and 28, with 57 total. One April event is exactly 2.00. The report’s 58 total and 29 recent count appear to use inconsistent boundary conventions in the retrieved edition.
Evidence and revision conditions
- Cape Seismic Events public data layer — Company-hosted layer discovered from public map; 58 retrieved magnitude≥2 records; frozen snapshot and IDs in inputs.json
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
Depends on C38.
What would change this assessment: The original event snapshot and one explicit threshold rule applied to both totals.
- C40
The blog acknowledging 36 events was published weeks after Latimer’s July interview.
- Type
- Reported fact
- Support
- Conjectural
- Review
- Contradicted as labeled
Both company pages display June 29, 2026. That precedes the interview described as July, rather than following it. Modification dates may differ from original publication; the report needs evidence for its chronology. The 36-event disclosure itself is supported.
Evidence and revision conditions
- Fervo April 2026 seismic update — June 29, 2026; 36 yellow events, ground motion, monitoring changes and mitigation
- Cape Station copy of seismic update — June 29,2026 displayed publication; same origin as Fervo, not independent corroboration
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
What would change this assessment: An archived publication history establishing later first public disclosure, or a corrected interview date.
- C41
The August 26–September 25 USGS window contains 39 magnitude-2-or-higher events.
- Type
- Calculated output
- Support
- Not applicable
- Review
- Unresolved conflict
The linked search starts August 25 and reproduces 39 events in the stated box. Starting August 26, as the prose says, yields 37. One is magnitude 3.13, so all 39 cannot be called yellow under a 2–3 threshold. Regional detections are not all authenticated Fervo-triggered events.
Evidence and revision conditions
- USGS linked regional search — Exact linked bounds and August 25 start; compare August 26 prose start; public query retrieved October 6
What would change this assessment: The frozen original query and event list, with precise endpoints, traffic-light bands and attribution criteria.
- C42
USGS records a magnitude-3.1 event near Cape Station on September 12, 2026.
- Type
- Reported measurement
- Support
- Supported
- Review
- Source checked
The event detail records ML 3.13 at 22:52:29 UTC, depth 3.65 km, at 38.51417, −112.8995. Rounded magnitude and date match. The catalogue supports location and magnitude within its uncertainty; it does not certify which operation caused the event.
Evidence and revision conditions
- USGS event uu80156586 — Reviewed event GeoJSON: September 12,2026,22:52:29 UTC; ML3.13; location and depth
What would change this assessment: A revised authoritative event solution or operational/seismological evidence establishing attribution.
- C43
The September event likely triggered at least a 24-hour operational pause.
- Type
- Inference
- Support
- Reported
- Review
- Not independently verified
The published traffic-light graphic specifies a 24-hour pause above magnitude 3.0. The threshold implication is conditional on the project’s applicable event assessment and protocol. No operational shutdown log or regulator instruction was retrieved. “Should pause” is not evidence that it did pause.
Evidence and revision conditions
- Fervo induced-seismicity management — Traffic-light response; magnitude 3 threshold and 24-hour pause; not evidence of actual shutdown
- USGS event uu80156586 — Reviewed event GeoJSON: September 12,2026,22:52:29 UTC; ML3.13; location and depth
Depends on C42.
What would change this assessment: A dated operational log or regulator/company notice specifying the event classification and response.
- C44
Seismic events have damaged Fervo well casing and increased leak-off.
- Type
- Alleged leak
- Support
- Reported
- Review
- Not independently verified
Anonymous quotations describe a plausible mechanism, but no failure log, casing survey, repair invoice or event-to-damage analysis authenticates these incidents. Higher regional event counts alone cannot distinguish seismic damage from completion, pressure, corrosion or other causes.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
- Cape Seismic Events public data layer — Company-hosted layer discovered from public map; 58 retrieved magnitude≥2 records; frozen snapshot and IDs in inputs.json
What would change this assessment: Authenticated well-level failures matched to event times, alternative-cause analysis and maintenance records.
- C45
Fervo’s commercial buildout will fail and consume its remaining capital.
- Type
- Forecast
- Support
- Conjectural
- Review
- Not yet resolvable
The report identifies serious execution risks but does not produce a probability distribution or integrated liquidity model. Fervo reports initial commercial operation; sustained output, cost and later phases remain future tests. No deadline defines “incinerated” capital or commercial failure.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
- Fervo commercial-operation announcement — October 1, 2026; attributed September 30 COD and 33 MW net output; not independently metered here
- Fervo second-quarter 2026 results — IPO gross proceeds, installed-cost targets and development timelines
Depends on C05, C12, C24, C26, C29, C36, C44.
What would change this assessment: Fix a resolution horizon and thresholds for net output, delivered energy, loss, installed cost, liquidity and default; compare actual outcomes.
- C46
Fervo had “zero megawatts in production,” as an employee quoted in the report says.
- Type
- Alleged leak
- Support
- Reported
- Review
- Unresolved interpretation
The quotation’s date is not supplied. On October 1 Fervo announced September 30 commercial operation and 33 MW net output from its first GeoBlock. Morpheus acknowledges first power elsewhere. The quotation may describe an earlier interview; it cannot establish publication-day absence of production.
Evidence and revision conditions
- Morpheus original report — October 6, 2026; full prose and supplied PDF pp. 1–65
- Fervo commercial-operation announcement — October 1, 2026; attributed September 30 COD and 33 MW net output; not independently metered here
What would change this assessment: Date the interview and reconcile it with actual metered commercial operation, rather than treating a historical quote as current evidence.
Review history: October 6, 2026 — ledger and arithmetic audit. Original article unchanged; corrections remain open.