The danger zones are getting sharper: seismic imaging of Campi Flegrei, the Himalaya, Cascadia, and Venezuela

A volcano's plumbing, mapped from deep quakes

The crust beneath Campi Flegrei just got mapped with unusual precision. The authors of an ESS Open Archive paper used deep-focus earthquakes from the Ionian slab, whose steep P-wave incidence reduces high-frequency attenuation in receiver-function analysis, to image structure beneath the seismic network. They report a prominent crustal interface at 7.7±0.4 km depth, associated with a low-velocity, high-density layer, and place the Moho discontinuity at 21.7±0.9 km. m:…4739

The method is half the news here. Receiver functions live or die on arrival sharpness, and deep-focus quakes supply steep arrivals that cut the high-frequency attenuation that usually blurs the picture. That is the authors' claim, not ours; our read is that sharper crustal geometry is exactly the input hazard modelers need for the volcanic system. m:…4739

The Himalaya may be loading twice as fast as the old math said

A second paper reworks the bookkeeping on the Main Himalayan Thrust. Using nonstationary viscoelastic models rather than stationary ones, the authors estimate locked-zone moment accumulation rates roughly twice those inferred from the stationary model, and show a rapid transition to full creep outside the locked zone. In their models, increasing lower-crustal and mantle viscosity shifts the inferred locked zone farther downdip. m:…7427

Twice the accumulation rate means the seismic budget refills roughly twice as fast, which — reasoning from the doubling figure, not predicting any event — compresses the timeline over which a given amount of slip builds up. But the result comes with the model's own error bars: the locked zone's position moves with the viscosity you assume, so the hazard picture depends on a parameter no one measures directly. m:…7427

The models also show a rapid transition to full creep outside the locked zone — beyond that boundary the fault slides along with plate motion rather than storing it. Our read: that sharp a boundary is what concentrates the stored slip, and it is the transition's position that the viscosity assumption moves around, which is why the downdip shift matters for the hazard map. m:…7427

Cascadia's slab sheds its water at depth

Controlled-source seismic imaging of the Cascadia margin pinpoints where the subducting Juan de Fuca plate gives up its pore water: at 12–15 km depth into the overlying accretionary wedge, beneath the Siletzia block. The paper's authors describe the plate's upper crust as porous and hydrated, in contrast to a dry lower crust and uppermost mantle. Seismic velocity boundaries in the accretionary prism line up with kinematic discontinuities, including Siletzia's static backstop. m:…2891

Our read: water in and out of the slab is a control knob on megathrust friction, so knowing where the plate dewaters — and where it stays dry — tightens the mechanical picture of the shallow subduction zone. This is the paper's mechanical contribution: a hydration map with depth, not a forecast. m:…2891

Seismic velocity boundaries in the accretionary prism align with kinematic discontinuities, including the static backstop of Siletzia, the authors report — the deep velocity structure and the shallow deformation agree with each other. When independent lines of evidence converge on the same architecture, the dewatering-depth picture gets harder to dismiss. m:…2891

Venezuela's big quake was one event, not two — in a preprint

In a preprint, a re-characterization of the 2026 Venezuela earthquake argues it was a single Mw 7.6 event rather than a doublet of Mw 7.2 and 7.5. The authors reconstruct episodes of rupture acceleration and deceleration that let the event break through the rupture area of the 1967 Caracas earthquake, and estimate roughly 26% of the seismic moment was released beneath the coastal cities of Playa Grande and Caraballeda. m:…6634

Doublet versus single event is not taxonomy for its own sake: damage models, aftershock expectations, and stress-transfer maps all change with the rupture story, and the coastal concentration of moment release helps explain where shaking hit hardest. But this is pre-peer-review, so treat the single-event call as a candidate interpretation awaiting independent verification. m:…6634

The rupture was not smooth: episodes of acceleration and deceleration let it punch through the older rupture boundary — a fitful but sustained propagation, in the authors' reconstruction. In our reading, variable rupture speed is one reason the same magnitude can produce very different shaking footprints — which is why re-characterizations like this one change more than the catalog entry. m:…6634