Mer de Glace from Signal Forbes
1890 | 2013 | 2021 | 2024
Strong acceleration of the demise of the largest French glacier over last few years! 🧊🔥
03.03.2026 12:08
👍 70
🔁 47
💬 2
📌 3
Mer de Glace from Signal Forbes
1890 | 2013 | 2021 | 2024
Strong acceleration of the demise of the largest French glacier over last few years! 🧊🔥
Ubehebe crater. A phreatomagmatic eruption, where magmas impinge upon wet sediment or groundwater causing rapid vaporization. Volume expansion during vaporization causes the water saturated overburden (roof) to explode. Sudden depressurization causes magma to then erupt catastrophically.
🌋Retour en images sur une semaine d'éruption au Piton de la Fournaise.
Après une semaine d'éruption, l'activité est focalisée sur un seul site éruptif sur le flanc sud-sud-est du volcan où un cône est en cours de formation par accumulation des projections de lave.
©OVPF-IPGP
A thread for the 15-year anniversary of the 2011 Christchurch earthquake. I won’t have a chance to post it on the day, so I’m doing it a couple days early. Note that the work I’m posting was carried out by many brilliant people. 1/
I've been thinking about this a lot. AI will make everyone except the 0.01% worse off.
🌋Prélèvements de lave ce matin (16/02/2026) par les équipes de l'OVPF-IPGP.
Malgré une baisse d'intensité par rapport aux premiers jours de l'éruption, l'activité se poursuit.
🌋 Les premières minutes de l'éruption au Piton de la Fournaise un peu après 10h heure locale (06h UTC) le 13 février 2026.
In response to the Dec. 6 Mw 7.0 earthquake in southwest Yukon, colleagues and I put together this short paper for the YGS's annual Yukon Exploration and Geology volume.
data.geology.gov.yk.ca/Reference/96...
⚒️🧪Fiery Mud Volcano Eruption!
The Daşmərdan (Dashmardan) mud volcano in Azerbaijan erupted ~16:50 on the 30th Jan 2026. The eruption lasted ~15 mins and was accompanied by a pillar of flame several hundred meters high. Daşmərdan had previous documented eruptions in 1866, 1954, 1976, 1986 and 2011.
SeisBench v0.11 is out now, the largest feature and performance update to SeisBench so far! In this thread, I'll highlight some of the key contributions. Thanks to everyone who helped putting together this release!
github.com/seisbench/se...
Alex Pretti, wearing a green sweater with his VA badge on.
This is Alex Pretti.
This morning he was murdered by ICE. Six ICE agents held him down and shot him at point blank range. Alex was a nurse and researcher at the VA.
Our thoughts are with his loved ones and we stand united in action calling for the abolishment of ICE.
#ScientistsAgainstICE
The first earthquake early warning system for high speed railways in Italy is now operational. nhess.copernicus.org/articles/26/...
Volcanic features offshore Mayotte and seismic/hydrophone stations used. The network includes four onshore broad-band (BB) stations (black), one accelerometer (red) and multiple ocean-bottom hydrophones (OBH, orange). Offshore stations combine deployments MAYOBS1 (Feb–May 2019) and MAYOBS2 (May–Oct 2019). An asterisk (e.g. MOC*) marks instruments redeployed across campaigns. P-T is Petite-Terre, east of Mayotte’s main island. Geomorphological boundaries and volcanic features (shaded areas in the bathymetry) follow N. Feuillet et al. (2021); V. Puzenat et al. (2022). Bathymetry is from MAYOBS campaigns (S. Jorry 2019).
Published in #GJI Geophysical Journal International: "Source parameters of very low frequency events during the Mayotte seismo-volcanic crisis (May 2018–January 2022)", Laurent et al. This is Fig. 1: please visit academic.oup.com/gji/article/... to read the paper. @royalastrosoc.bsky.social
🌋 Eruption en cours au Piton de la Fournaise
Retrouvez tous les communiqués de l'OVPF-IPGP ici👇
www.ipgp.fr/communiques-...
dont le dernier ici👇
www.ipgp.fr/volcanoweb/r...
Copyright photos : OVPF-IPGP
Can someone explain to me how selling to an unemployed population can be lucrative?
#AI #AGI "offers the prospect of computer systems carrying out profitable work without the associated cost of human labour – a hugely lucrative scenario..." www.theguardian.com/technology/2...
The portable computer processing Early-Est died. As I have no funding for time or hardware to run Early-Est here, I am waiting to see if an opportunity to easily set-up and run Early-Est on freely available or surplus hardware arises.
#Accaddeoggi: tra il 12 e il 13 gennaio 2011 all'Etna cominciò un'eruzione parossistica sommitale con e attività effusiva. Così iniziò una lunga sequenza di episodi parossistici e fontane di lava al Cratere di Sud-Est, che portarono alla costruzione del Nuovo Cratere di Sud-Est.
Fotog di B. Behncke
Wanna see how an earthquake works? Here's today's lab in my seismology class. An explanation follows.
🧪
We conclude DL based monitoring can rapidly produce more consistent picks & better catalogs than standard procedures, freeing analysts for improved quality control, analysis, & dissemination of information on seismic activity, especially during significant seismic sequences.
doi.org/10.1093/gji/...
Violin plots of the distribution of the weighted RMS of the earthquake locations for the three pick collections used in the tomographic inversion (BSI-R, PN-IN and PN-OR). Panel (a) shows the weighted RMS for relocation of the 2000 earthquakes used for the inversion whereas (b) shows the weighted RMS for relocation of all the studied events adopting the inverted velocity model. The thick solid line indicates median value and the dashed lines the lower and upper quartile.
The tomographic inversions for 3D, seismic P- and S-wave velocity models show better convergence and more organized earthquake relocations using the DL picks than with the routine picks.
Cross-sections of BSI-R (a–c), PN-OR (d–f) and PN-IN (g–i) NLL-SC relocations. Each cross-section shows seismicity within ±3 km of the section. Events colour is based of event origin time. Main shocks are shown as yellow stars.
Hypocentre separation compared to waveform coherence for the three relocated catalogues. Upper panel: Statistics of 3-D distance between final NLL-SSST hypocentres for event pairs from the 3 catalogues plotted as a function of binned NLL-coherence waveform coherence for each pair. Horizontal bars show median values, vertical bars show 10–90 per cent quantile range. The BSI-R and PN-IN results are offset horizontally for clarity. Lower panel: relative number of event pairs in each waveform coherence bin; the total number N of event pairs for each catalogue is indicated in the legend.
Relative to location with routine picks (BSI), locations using DL picks (PN) show improved organization and clustering, and, in a ground-truth test, smaller hypocenter separation for event pairs with more similar waveforms.
Map of the central Apennines area indicated by the blue rectangle in the inset. Seismic stations used in the analysis are shown as triangles coloured according to network codes. Purple dots are the 84 991 earthquakes detected and located by BSI in the time period 2016 August 15–2017 August 31. Yellow stars are the three main shocks of the central Apennines seismic sequence.
Histograms of P (a, c) and S (b, d) pick difference for PhaseNet Original (PN-OR) and PhaseNet INSTANCE (PN-IN) with respect to INGV manual phase picks (BSI). Panels (e) and (f) show the histograms of P and S pick time differences for PN-OR with respect to PN-IN. Medians and Means are represented as red dashed and blue solid lines, respectively; ±1 and ±2 standard deviations are in dashed green and purple lines, respectively.
New paper! We compare PhaseNet (PN) #deep-learning (DL) with manual (BSI) event picking for routine seismic monitoring by analyzing high-precision earthquake locations and tomographic inversion for the 2016 Central Italy #earthquake sequence. doi.org/10.1093/gji/... @carlo-giunchi.bsky.social
me too, since 1990's:
StarOffice->OpenOffice->LibreOffice (has been really fine since about 5-10 years)
Zotero more recently, but fabulous!
Just after midnight on 1 January 2026, a new eruption started from #Etna's eastern flank: a fissure opened at ~2100 m elevation, and a lava flow is descending through the desert-like Valle del Bove. Photos taken in the evening of 1 January from Etna's east and southeast flanks
A jet of incandescent lava fragments shoots from a vent that opened on the morning of 27 December 2025 on the upper eastern flank of the Voragine crater at the summit of Etna, on the evening of the 28th. This type of activity is known as "Strombolian", after the volcano Stromboli in the Aeolian Islands north of Sicily. The snow-covered peak at right is the Southeast Crater, which last erupted in August 2025. View from the town of Nicolosi, on the south flank of Etna.
The new vent that opened on the morning of 27 December 2027 on the upper east flank of the Voragine crater at the summit of #Etna, seen in Strombolian eruption on the evening of the 28th. This activity has been continuing over the past 24 hours without significant variations. Seen from south
"I've never seen a year quite as worrying as 2025 has been - not just because several major conflicts are raging but because it is becoming clear that one of them has geopolitical implications of unparalleled importance"
www.bbc.com/news/article...
Eruption column produced by the first paroxysmal eruptive episode at Etna's Northeast Crater, on the morning of 27 December 2025, recorded by a monitoring camera of the INGV-Osservatorio in Bronte, on the western flank of the volcano.
Eruption column produced by the second paroxysmal eruptive episode at Etna's Northeast Crater, in the afternoon of 27 December 2025, recorded by a monitoring camera of the INGV-Osservatorio at the Montagnola, high on the southern flank of the volcano.
Minor explosive activity from a vent on the east side of Etna's Voragine crater at sunset on 27 December 2025, recorded by a monitoring camera of the INGV-Osservatorio at the Montagnola, high on the southern flank of the volcano.
Not one, but TWO paroxysmal eruptive episodes at #Etna 's Northeast Crater on 27 December 2025, AND minor activity at the Voragine crater. Though the Northeast Crater has frequently shown minor activity in recent decades, its last paroxysmal episode was on 27-28 March 1998.
