OHMEGA/MONSTER

Seismic noise points to Yellowstone-scale magma reservoir beneath Tuscany

Thousands of cubic kilometres of hot, partly molten rock imaged 8 to 15 kilometres down under Larderello's geothermal fields match Yellowstone in scale but are cool, viscous and quiet — and whether the slow seismic signal really means that much liquid magma remains contested.

Italy

What happenedScientists using ambient-noise seismic imaging described on 14 April 2026 a vast slow shear-wave zone 8 to 15 km beneath Tuscany's Larderello geothermal fields, interpreted as thousands of cubic kilometres of magma and crystal mush on the scale of Yellowstone.

Why it mattersIt explains Tuscany's extreme geothermal heat and shows a huge magma body can hide without a caldera or unrest, while testing whether cheap noise-based surveys can reliably hunt geothermal heat and resources elsewhere.

Still openWhether the exceptionally slow signal truly requires 20-40% melt and a largely liquid core or can be largely explained by brines, gas-filled fractures and seismic anisotropy, and whether viscous chemistry permanently prevents eruption or only delays it.

Tuscan hills with geothermal steam rising at dawn
AI-generated illustration — not a photograph of this story

TITLE: The Yellowstone-sized magma body hiding under Tuscany — and why it isn't about to erupt STANDFIRST: Seismic background noise has imaged thousands of cubic kilometres of hot, partly molten rock 8 to 15 kilometres beneath Tuscany's oldest geothermal fields — a reservoir on the scale of Yellowstone's, but cool, viscous and quiet at the surface.

A vast underground magma reservoir mapped with seismic background noise — thousands of cubic kilometres of hot, partly molten rock 8 to 15 kilometres beneath southern Tuscany — was described on 14 April 2026 in Communications Earth & Environment. The zone is on the scale of the mid-crust reservoirs beneath Yellowstone, Toba and Taupo, yet it sits under a landscape with no Holocene eruption, no caldera and no current signs of unrest. Whether that exceptionally slow seismic signal really requires 20 to 40% melt and a largely liquid core, or can be largely explained by brines, gas-filled cracks and seismic anisotropy, is the central uncertainty.

Southern Tuscany, about 200 kilometres northwest of Rome, hosts the Larderello-Travale geothermal system — the world's oldest exploited geothermal field, in production since 1913 — and the dormant volcano of Mt. Amiata. The Tuscan Magmatic Province there has produced only small, scattered volcanism for millions of years, even as boreholes and hot springs hinted at intense heat below.

A Yellowstone-scale slow zone at 8 to 15 kilometres

The new three-dimensional shear-wave velocity model shows two prominent slow zones beneath Larderello-Travale and Amiata. Shear-wave velocity, or Vs, is the speed of side-to-side seismic waves; in solid continental crust it normally increases with depth. Here it does the opposite. At about 10 to 12 kilometres depth the core drops to about 1.2 to 1.25 kilometres per second, roughly 40% slower than the average for that depth.

The anomaly is not a narrow pipe. A sub-vertical low-Vs domain rises from the bottom of the model, conservatively cut at 15 kilometres, to about 8 kilometres beneath Larderello, then spreads sub-horizontally south to Amiata. The paper conservatively reports more than 5,000 cubic kilometres of magma and partial melt stored in the middle crust of the province; press summaries from the University of Geneva round that to about 6,000 cubic kilometres — about 2.4 billion Olympic swimming pools. For the Larderello anomaly alone the authors calculate a core of about 3,000 cubic kilometres with a liquid fraction above 80% surrounded by about 5,000 cubic kilometres of crystal mush at about 20% liquid. If core and mush are summed the total approaches 8,000 cubic kilometres, but the quoted inventory is the more conservative figure. The Amiata anomaly is described as at least twice as large as Larderello's, but the authors flag it as near the edge of resolution and in need of denser coverage. The scale is put in context in the paper's Table 1, where about 5,000 cubic kilometres for Larderello is compared directly to Long Valley caldera in California, and more broadly to Yellowstone, Toba and Taupo.

How background noise made the invisible visible

The image comes from ambient-noise tomography, a technique that turns continuous background vibrations — ocean waves, wind, traffic — recorded at many seismometers into a three-dimensional picture of how fast shear waves travel underground. Slow zones imply hot, fluid-rich or molten rock.

The team led by Matteo Lupi, associate professor in the Department of Earth Sciences at the University of Geneva, with colleagues from the Institute of Geosciences and Earth Resources of the National Research Council (CNR-IGG) in Florence and the National Institute of Geophysics and Volcanology (INGV), deployed 30 broadband seismometers from September 2020 to September 2021 as the TEMPEST network and combined them with more than 30 permanent stations of the Italian INGV network and the French and AlpArray networks, for more than 60 stations in total. For each station pair they cross-correlated months of noise, extracted Rayleigh-wave dispersion curves, built two-dimensional group-velocity maps with a regularized linear least-squares inversion, then ran a one-dimensional depth inversion with a transdimensional Bayesian method called BayHunter. Checkerboard tests, sensitivity kernels and Vs uncertainty estimates support reliability to 15 kilometres, but the model is Rayleigh-wave only.

That scale mismatch explains why the body stayed hidden. Earlier active seismic lines such as CROP-18, gravity highs, electromagnetic soundings and sparse passive tomography either covered too large an area too sparsely or too small an area too densely to image a mid-crustal feeder of this size. Hints existed — a teleseismic receiver-function dome with Vs as low as 1.6 kilometres per second at 9 to 12 kilometres, and a P-wave reduction of about 18% — but not a quantitative location and volume.

From slow waves to magma volume — and why the number is shaky

Ambient-noise tomography images Vs, not melt. Converting Vs to melt fraction and volume is the most assumption-dependent step.

The authors define three thresholds: minus 40% (Vs at or below 1.3 kilometres per second), minus 30% (1.3 to 1.8) and minus 20% (1.8 to 2.4). Using MageMin, a thermodynamic Gibbs-energy minimizer, and assuming a melt-saturated porous granite, they find 20 to 40% melt is required to explain the minus 40% core, with the core above 80% liquid and the outer mush around 20% liquid. That conversion is what turns a map of slow voxels into cubic kilometres.

The same Vs can be produced in other ways. Peer-reviewed studies of similar low-Vs anomalies show identical signals without 20 to 40% melt: at Ebeko volcano in the Kurils, velocity anomalies alone are described as inherently ambiguous and may reflect gas saturation, dry fracturing, elevated temperature or partial melt; at La Palma in the Canaries, high-attenuation zones with low Vs and low resistivity are interpreted as hydrothermal alteration and fractured fluid pathways, not melt; at Kuju in Japan a plume-shaped zone with Vp/Vs (the ratio of compressional-wave speed to shear-wave speed, a diagnostic that behaves differently for melt versus gas-filled cracks) below 1.4 is interpreted as a crack-dominated, thermally altered formation with gas from supercritical fluid transitions. Modeling that includes crystal orientation, melt shape and matrix anisotropy finds huge variations in the same seismic property at the same melt fraction, with nonlinear behaviour. In the Cascades arc, a joint Rayleigh-Love inversion finds isotropic low-Vs zones of 3.45 to 3.55 kilometres per second at 15 to 30 kilometres interpreted as only 2.5 to 6% melt, and notes the same Vs could be explained by sub-solidus crustal compositions. Positive radial anisotropy — where horizontally polarized shear waves travel faster than vertically polarized ones because of sub-horizontal sill layering — can bias a Rayleigh-only Vs low if not jointly inverted. Lupi and colleagues note that radial anisotropy is strongly positive in supervolcano plumbing for exactly this reason and that Love waves may be faster than Rayleigh waves, but they do not perform a joint inversion.

Intrinsic attenuation, measured as Qc-1 from the coda of Rayleigh waves, is in those studies more sensitive to fluids, fracture density and thermo-mechanical state than Vs alone. At Ebeko a shallow high-attenuation zone maps fracture-controlled fluid reservoirs while a deeper low-attenuation core suggests consolidated crystalline mush — the inverse of a simple low-Vs equals melt mapping. Without joint Love-Rayleigh, Vp/Vs, attenuation and resistivity, Tuscany's low Vs cannot by itself discriminate melt from fluid-filled fractured and altered rock. No independent peer commentary specifically rebutting the Tuscany interpretation has been published; the alternative is a steelman built from analogous peer-reviewed cases.

Why Tuscany's heat makes sense now

Whatever the exact melt fraction, the reservoir quantifies a heat source long postulated but not located. Larderello-Travale shows extreme values: heat flow locally up to 1,000 milliwatts per square metre compared with about 65 for average continental crust, geothermal gradient above 150 degrees Celsius per kilometre, temperatures above 500 degrees at about 3 kilometres, and in the Venelle-2 well 42.5 megapascals and 512 degrees at 2.8 kilometres — supercritical conditions for brines, with a sharp gradient increase at 2,750 metres where temperature was 380 degrees. The K-horizon, a strong reflector at 3 to 4 kilometres seen on the CROP-18A seismic profile and debated as a tectonic, rheological or fluid boundary, now appears as the roof enveloping the low-Vs cupola. Tourmaline indicating supercritical conditions has been found at that depth.

The boundaries of the slow zones coincide with surface hydrothermal upwellings at Larderello, Venturina, Rapolano and Bagni di Petriolo, and with a northwest-striking fast body that tracks the Tuscan metalliferous region. Regional fluid flow from the hot mid-crust helps explain widespread travertine deposits and abundant carbon dioxide from heated limestones. Tuscany already hosts 34 geothermal plants with 37 generating units totalling about 800 megawatts at Larderello alone — the Valle Secolo plant at 117.6 megawatts is the largest in Europe — and about 916 megawatts nationally producing about 5.7 terawatt-hours a year, roughly 31% of Tuscany's electricity.

Why big does not mean about to blow

Hazard requires more than volume: eruptibility, pathways to the surface, and unrest. Tuscany currently lacks all three.

There is no recognized caldera or cone at Larderello, the most recent eruption at Amiata was about 300,000 to 200,000 years ago with only about 10 cubic kilometres erupted, and no major Holocene eruptions are recorded in the province. INGV, which operates the permanent seismic network and monitors the geothermal fields under a formal agreement with the Tuscany Region, publishes integrated monitoring reports — nine from July 2020 to July 2026 — with real-time acquisition at its Bologna monitoring centre. Its comunicato with CNR-IGG on 14 April 2026 independently confirmed thousands of cubic kilometres of magmatic fluids at 8 to 15 kilometres but noted magmatism has been predominantly plutonic for millions of years with no surface eruption signals. Regional Sentinel-1 radar interferometry (InSAR, satellite radar that detects centimetre-scale ground inflation), processed for Tuscany by the University of Florence and civil-protection authorities at centimetre precision, shows no volcanic inflation anomaly at Larderello or Amiata, consistent with no current deformation. Independent ground-movement GNSS (Global Navigation Satellite System, the satellite positioning that measures millimetre-scale ground movement) and carbon-dioxide flux time series for 2020 to 2026 were not available for this article, so the assessment of no current unrest rests on the seismic network and satellite radar deformation data.

Geochemistry offers a mechanism for the quiet. Unlike the mantle-derived, less viscous magmas that fed VEI-7 (Volcanic Explosivity Index 7, the second-highest rank on the scale volcanologists use for eruption size — super-eruptions that eject hundreds of cubic kilometres) eruptions at Colli Albani and Campi Flegrei to the south, linked to a veined, metasomatized asthenospheric wedge above the eastward rollback of the Adriatic slab, the Tuscan magmas are peraluminous anatectic granitoids formed by melting metasedimentary basement. They are cool, silica-rich and highly viscous, forming a viscous barrier that hinders ascent and promotes top-down pluton formation by downward accretion. The authors stress that from a geophysical perspective the plumbing looks like a caldera — intense microseismicity, vigorous fluid flow, gravity anomaly, fumaroles, volcano-like gradient — but without an eruption record it is classified as a high-enthalpy geothermal system rather than a volcanic system. The risk framed is long-term evolution, not imminent eruption.

Viscosity is a brake, not a lock

High viscosity does not guarantee permanent non-eruption. Silicate-melt physics shows viscosity varies by orders of magnitude with temperature, dissolved water and crystal and bubble content, and high viscosity traps volatiles, increasing explosivity when overpressure and strain rate cross fragmentation thresholds.

Tuscan peraluminous magmas have erupted. At Campiglia Marittima, the Botro ai Marmi granite and San Vincenzo rhyolites erupted 5.4 to 4.4 million years ago as viscous lava flows, domes and dykes with about 1 cubic kilometre preserved, at 68 to 72 weight percent silica and aluminium saturation index 1.1 to 1.3, cordierite-bearing peraluminous compositions. Type-B hybrid rhyolites there demonstrate mingling of crustal and mantle melts after about 500,000 years of melt-present storage, with antecryst recycling showing crystal mush can be remobilized. Amiata itself is the proof that the same province can erupt. A 300,000-year quiescence does not prove a permanent lock; remobilization on ten-thousand to hundred-thousand-year timescales remains plausible if the mush is recharged, pressurized by volatiles or strained quickly enough.

What changes for hazard and for heat-hunting

The broader significance is twofold, and both depend on validation.

For hazard detection, Tuscany shows thousands of cubic kilometres can hide at mid-crustal depth with only subtle geothermal clues. Classic supervolcano identification relies on caldera morphology, uplift, eruptive deposits, gas and deformation; absence of those signals does not equal absence of magma. Systematic ambient-noise surveys in other quiet, high-heat-flow provinces could test how common such hidden storage is.

For resources, the University of Geneva and INGV frame the method as faster, cheaper and zero-impact compared with active seismic or drilling, using existing networks plus temporary deployments, and potentially useful to locate high-enthalpy geothermal and the lithium and rare-earth elements that concentrate in magmatic-hydrothermal systems for batteries and the energy transition. Tuscany is already a test case for supercritical drilling through the DESCRAMBLE project. But Vs alone maps heat source location and size, not prospectivity: attenuation and resistivity are needed to discriminate a hot, fluid-rich geothermal target from a mineralized zone, and the paper presents the approach as proof-of-concept for low-cost screening, not a proven finder of ore.

The image is therefore both concrete and provisional: a Yellowstone-scale slow zone is now mapped where none was resolved before, it explains why Tuscany is so hot and why the K-horizon sits where it does, and it reframes how to look for hidden magma elsewhere — while the headline volume and the word magma itself remain interpretations of slow waves, not direct counts of liquid.

Source recordSources / claims / limits

How this piece is framed: A Yellowstone-scale reservoir hiding in plain sight: what was actually imaged beneath Tuscany, why it hasn't erupted, and why the image itself is contested

Sources

Claims, and how far we tracked each down

  • [confirmed] A peer-reviewed paper 'High-enthalpy Larderello geothermal system, Italy, powered by thousands of cubic kilometres of mid-crustal magma' was published 14 April 2026 in Communications Earth & Environment (Vol 7, Art 269, DOI 10.1038/s43247-026-03334-0) by Lupi et al. · read in full (as of 2026-08-06)
  • [confirmed] Lead author is Matteo Lupi, associate professor in Department of Earth Sciences, University of Geneva (UNIGE), with co-authors from Institute of Geosciences and Earth Resources (CNR-IGG) and National Institute of Geophysics and Volcanology (INGV). · read in full (as of 2026-08-06)
  • [confirmed] Study used ambient-noise tomography (ANT) to build a 3-D shear-wave velocity (Vs) model of the upper 15 km of crust in southern Tuscany. · read in full (as of 2026-08-06)
  • [confirmed] Data came from >60 broadband seismometers: 30 stations deployed Sept 2020-Sept 2021 (TEMPEST network, code 1K) integrated with the permanent Italian INGV network (IV) plus Z3 and FR networks, archived at EIDA and IRIS-DMC. · read in full (as of 2026-08-06)
  • [confirmed] Model shows prominent low-Vs anomalies beneath Larderello-Travale and Amiata geothermal fields, with absolute Vs as low as ~1.2-1.25 km/s at ~10-12 km depth, corresponding to ~-40% relative anomaly versus 1-D average Vs profile. · read in full (as of 2026-08-06)
  • [confirmed] Low velocities are interpreted as magma/partial melt; authors define three anomaly thresholds: -40% (Vs 241.3 km/s), -30% (1.3-1.8 km/s), -20% (1.8-2.4 km/s). · read in full (as of 2026-08-06)
  • [confirmed] Using thermodynamic modeling with MageMin (Gibbs energy minimizer) assuming melt-saturated porous granite, 20-40% melt fractions are required to explain extremely low Vs, with detailed modeling indicating a core >80% liquid and outer crystal-mush ~20% liquid. · read in full (as of 2026-08-06)
  • [confirmed] Paper estimates more than 5,000 km�b3 of magma and partial melt stored in middle crust of Tuscan Magmatic Province; press summaries round to ~6,000 km�b3 (equivalent to 2.4 billion Olympic swimming pools). · read in full (as of 2026-08-06)
  • [confirmed] For the Larderello anomaly alone, authors calculate ~3,000 km�b3 of partial melt surrounded by ~5,000 km�b3 of crystal mush; the Amiata anomaly holds at least twice as large volume but is near model resolution limits and requires additional acquisition for robust assessment. · read in full (as of 2026-08-06)
  • [confirmed] Anomaly extends across Tuscany at depths 8-15 km, with a sub-vertical low-Vs domain rising to ~8 km beneath Larderello and extending sub-horizontally south to Amiata; resolution tests (checkerboard, kernels) support reliability to 15 km depth. · read in full (as of 2026-08-06)
  • [confirmed] Volume is comparable to mid-crust reservoirs beneath recognized supervolcanoes: Yellowstone, Toba, Taup4d, and Long Valley (paper Table 1 compares ~5,000 km�b3 Larderello to Long Valley). · read in full (as of 2026-08-06)
  • [confirmed] Tuscan Magmatic Province shows sparse Quaternary volcanism and no major Holocene eruptions; most recent eruption is Mt. Amiata ~300-200 ka with only ~10 km�b3 eruptive volume, and no recognized caldera or cone associated with Larderello. · read in full (as of 2026-08-06)
  • [confirmed] Despite Yellowstone-scale volume, system poses no immediate volcanic threat; there is no current surface deformation, anomalous gas emission, or unrest indicating imminent eruption. · read in full (as of 2026-08-06)
  • [confirmed] Reservoir fuels the high-enthalpy Larderello-Travale geothermal system, which exhibits extreme heat flow locally up to 1,000 mW/m�b2, geothermal gradient >150�b0C/km, temperatures >500�b0C at ~3 km depth, and supercritical fluids at 2.8 km (42.5 MPa, 512�b0C in Venelle-2 well). · read in full (as of 2026-08-06)
  • [confirmed] Reservoir fuels the high-enthalpy Larderello-Travale geothermal system, which exhibits extreme heat flow locally up to 1,000 mW/m8, geothermal gradient >1506C/km, temperatures >5006C at ~3 km depth, and supercritical fluids at 2.8 km (42.5 MPa, 5126C in Venelle-2 well). · read in full (as of 2026-08-06)
  • [confirmed] Paper deep-read verifies extreme geothermal parameters: heat flow up to 1,000 mW/m8, gradient >1506C/km, >5006C at ~3 km, and Venelle-2 well at 2.8 km with 42.5 MPa / 5126C (supercritical brines per Driesner & Heinrich), with sharp gradient increase at 2750 m (3806C); K-horizon at 34 km on CROP-18A envelops low-Vs cupola. · read in full (as of 2026-08-06)
  • [confirmed] MageMin modeling assumes melt-saturated porous granite with three Vs thresholds (-40% Vs241.3 km/s, -30% 1.31.8 km/s, -20% 1.82.4 km/s); 2040% melt required for extremely low Vs; Larderello core >80% liquid (~3,000 km9) surrounded by ~20% liquid mush (~5,000 km9); Amiata at least twice as large but flagged as near resolution limits requiring additional acquisition; Rayleigh-only model noted with radial anisotropy caveat. · read in full (as of 2026-08-06)
  • [confirmed] ANT workflow verified as pre-processing, cross-correlation, dispersion curves, 2-D group-velocity maps via regularized linear least-square inversion, and 1-D depth inversion via McMC transdimensional Bayesian inversion (BayHunter); resolution to 15 km supported by checkerboard (Fig S9), kernels (S11), Vs uncertainty (S16); absolute Vs as low as ~1.21.25 km/s (-40%) at 1012 km core. · read in full (as of 2026-08-06)
  • [confirmed] INGV-Regione Toscana integrated seismic monitoring system for Larderello-Travale and Monte Amiata operates under formal agreement with defined monitoring domains, homogeneous data-sharing protocols, real-time acquisition at INGV Bologna Centro di Monitoraggio del Sottosuolo, ILG-compliant monitoring (localization, PGV, ML), with 9 published technical reports from July 2020 to July 2026. · read in full (as of 2026-08-06)
  • [confirmed] INGV official comunicato (with CNR-IGG, 14 Apr 2026) independently confirms thousands of km9 of magmatic fluids at 815 km beneath Larderello and Amiata, comparable to Yellowstone/Toba/Taupo, via zero-impact ambient-noise tomography; notes Tuscany magmatism has been predominantly plutonic for millions of years with no surface eruption signals, consistent with no Holocene eruption / no caldera / no current unrest. · read in full (as of 2026-08-06)
  • [likely] Copernicus Sentinel-1 provides operational regional InSAR deformation monitoring for Tuscany at centimetre precision (University of Florence / Civil Protection / Regione Toscana / TRE ALTAMIRA), with regularly updated ground deformation maps routinely sent to geohazard authorities; public ESA summary reports no Larderello/Amiata volcanic inflation anomaly, consistent with 'no current surface deformation', though specific GNSS velocity and CO2 flux time series for 20202026 were not retrieved in this pass. · read in full (as of 2026-08-06)
  • [contested] Low shear-wave velocity anomalies of -30% to -40% (Vs ~1.2-1.3 km/s) are not uniquely diagnostic of partial melt; peer-reviewed attenuation and velocity studies show identical low-Vs / low-Vp/Vs signatures can be produced by aqueous/saline fluids, supercritical brines, gas-saturated fractured rock, and hydrothermal alteration without requiring 20-40% melt. Ebeko study states "velocity-based anomalies alone are inherently ambiguous and may reflect different physical processes, including gas saturation, dry fracturing, elevated temperature, or partial melt" and La Palma study shows high-attenuation zones with low Vs/low resistivity interpreted as hydrothermal alteration and fractured fluid pathways, not melt. Kuju study interprets plume-shaped Vp/Vs <1.4 as crack-dominant thermally altered formation + gas from supercritical fluid phase transition, with high Vp/Vs below seal as pressurized supercritical fluid. · read in full (as of 2026-08-06)
  • [likely] Rayleigh-wave-only Vs model as used in Tuscany is susceptible to radial anisotropy bias; positive radial anisotropy (Vsh > Vsv) from sub-horizontal sill complexes makes Love waves faster than Rayleigh waves and can bias isotropic Vs low if not jointly inverted. Cascades arc joint Rayleigh-Love inversion finds isotropic low-Vs zones of 3.45-3.55 km/s at 15-30 km depth interpreted as only 2.5-6% melt assuming near-equilibrium geometry, and notes same Vs could alternatively be explained by sub-solidus crustal compositions; positive anisotropy adjacent to low-Vs is interpreted as mostly crystallized sill complexes, not melt. Lupi et al. explicitly note "Radial anisotropy is strongly positively anomalous in the plumbing systems of supervolcanoes because of the interlayered sill structure. This may result in Love waves being faster than Rayleigh waves" but do not perform joint inversion. · read in full (as of 2026-08-06)
  • [likely] Thermodynamic conversion of Vs to melt fraction via MageMin assuming melt-saturated porous granite is highly non-unique; modeling incorporating crystallographic preferred orientation, melt shape, orientation and matrix anisotropy shows "nonlinear behavior" and "huge variations in the same seismic property even if the melt fraction remains the same." Authors warn "Interpretation of seismic data to infer melt percentages or extent of melting should always be underpinned by robust modeling... and examination of multiple seismic properties" and that "different geophysical methods yield different results" (e.g., Tibet bright spots interpreted as 10% melt vs 3-7% vs no widespread melt). This directly qualifies the 20-40% and >80% liquid core estimates and the 5,000-6,000 km3 inventory as order-of-magnitude, assumption-dependent. · read in full (as of 2026-08-06)
  • [likely] Intrinsic attenuation (Qc-1) is more sensitive to fluid content, fracture density and thermo-mechanical state than Vs alone; La Palma ANAT study concludes "intrinsic attenuation retrieved from the coda of Rayleigh waves is more sensitive to the presence of fluids than velocity" and Ebeko study shows shallow high-attenuation (>0.45) maps fracture-controlled fluid reservoirs while a 4-6 km low-attenuation core (<0.1) suggests consolidated crystalline mush  the inverse of a simple low-Vs = melt mapping. Without joint Qc-1, Vp/Vs, resistivity or attenuation constraints, Tuscany low-Vs cannot discriminate melt vs fluid-filled fractures. · read in full (as of 2026-08-06)
  • [confirmed] ANT workflow included pre-processing ambient noise, cross-correlation for station pairs, extraction of dispersion curves, 2-D group-velocity maps via regularized linear least-square inversion, and 1-D depth inversion via McMC transdimensional Bayesian inversion (BayHunter). · read in full (as of 2026-08-06)
  • [likely] Authors attribute lack of eruptivity to peraluminous anatectic granitoids formed by crustal anatexis of metasedimentary basement: low-temperature, highly viscous magmas that form a viscous barrier, hinder ascent, and promote top-down pluton formation via downward accretion. · read in full (as of 2026-08-06)
  • [likely] The K-horizon (3-4 km depth transition seen on CROP-18A seismic profile) is interpreted as a transition zone enveloping the low-velocity mid-crustal domain, possibly representing supercritical fluid accumulation, tectonic or rheological boundary. · read in full (as of 2026-08-06)
  • [confirmed] Ambient-noise tomography is presented as faster and lower-cost than active seismic for large-scale subsurface imaging, with potential to locate geothermal reservoirs and lithium/rare-earth element deposits associated with deep magmatic systems. · read in full (as of 2026-08-06)
  • [confirmed] Enel Green Power operates 34 geothermal plants with 37 generating units totaling ~800 MW in Tuscany; Valle Secolo plant (largest in Europe) has 117.6 MW operating capacity (260 MW sections), 854 million kWh/year average production, with ~40% condensate reinjection and real-time microseismic monitoring. · read in full (as of 2026-08-06)
  • [confirmed] Italy has 916 MW installed geothermal capacity producing ~5.7 TWh/year (1.61.8% of national electricity, ~5% of renewables), concentrated in Tuscany across 34 plants in 10 concessions providing ~31% of Tuscany's electricity; PNIEC targets 7.5 TWh by 2030. · read in full (as of 2026-08-06)
  • [likely] High viscosity of peraluminous, high-silica anatectic magmas does not by itself suppress eruption; silicate-melt physics shows viscosity is strongly modulated by temperature, dissolved H2O, and crystal/bubble content, and high viscosity traps volatiles, increasing explosivity and fragmentation potential. Zhang review details that rhyolitic melt viscosity varies by orders of magnitude with water content and that fragmentation criteria depend on bubble overpressure and strain rate, not viscosity alone. Viscosity as a "barrier" is therefore a conditional, not deterministic, control on eruptibility. · read in full (as of 2026-08-06)
  • [confirmed] Tuscan Magmatic Province peraluminous magmas have erupted despite high viscosity: Botro ai Marmi granite (SiO2 68-72 wt%, ASI 1.1-1.3, cordierite-bearing peraluminous monzo-syenogranite) and San Vincenzo Type-A/B rhyolites at Campiglia Marittima (5.4-4.4 Ma) erupted as viscous lava flows/domes (~1 km3 preserved) and dykes; Type-B hybrid rhyolite demonstrates remobilization and mingling of crustal and mantle melts after ~500 ka of melt-present storage. System remained in melt-present condition for ~500 ka per cycle with antecryst recycling, showing crystal mush can be remobilized. This demonstrates peraluminous anatectic magmas are eruptible and that a 300 kyr quiescence at Amiata does not prove permanent viscous lock. · read in full (as of 2026-08-06)
  • [unconfirmed] No independent peer commentary or published critique specifically addressing Lupi et al. 2026 Tuscany interpretation was retrieved in this pass; counter-perspective above is built from analogous peer-reviewed studies of similar low-Vs anomalies and melt-physics reviews, not direct rebuttal of the Tuscany paper. Gap remains for targeted volcanology/seismology commentary. · read in full (as of 2026-08-06)
  • [confirmed] Model is based on Rayleigh waves; authors note radial anisotropy (Love faster than Rayleigh) in supervolcano plumbing due to interlayered sills could affect velocity interpretation. · read in full (as of 2026-08-06)

Where we hit a limit / what to double-check