Feed aggregator

Editors’ Vox is a blog from AGU’s Publications Department.

Explosive volcanic eruptions inject gases and ash into the atmosphere, posing major hazards for human health, infrastructure, and aviation. A new article in Reviews of Geophysics examines recent advances in estimating Eruption Source Parameters (ESPs), the key conditions at the volcanic vent that are a necessity for modeling the behavior of volcanic plumes. Here, we asked the authors to explain what ESPs are, what technologies are used to observe eruptions, and which scientific challenges and future research directions remain for improving volcanic plume monitoring and modeling.

In simple terms, what are Eruption Source Parameters?

Eruption Source Parameters (ESPs) describe the key conditions at the volcanic vent during an eruption.

Eruption Source Parameters (ESPs) describe the key conditions at the volcanic vent during an eruption, such as the mass eruption rate, exit velocity, temperature, and particle size distribution. These parameters define how material is injected into the atmosphere and are essential inputs for models that simulate plume rise and subsequent dispersal of volcanic gases and ash in the atmosphere. In simple terms, ESPs represent the boundary conditions that control the behavior of volcanic plumes. Because they cannot usually be measured during an eruption, they must be estimated from indirect observations and models, which introduces significant uncertainty.

Why is it important to understand how volcanic ash and gases disperse after an eruption?

Volcanic ash and gases can travel long distances and affect aviation safety, human health, infrastructure, and even climate. Fine ash particles are particularly hazardous for aircrafts, while ash fallout can disrupt communities and critical services on the ground. Gas emissions may also impact air quality and alter the atmospheric radiative budget. Understanding volcanic dispersion is therefore essential for forecasting the movement of volcanic clouds and issuing timely warnings. Reliable forecasts support risk mitigation strategies and enable more effective responses by civil protection agencies and aviation authorities.

What technologies are used to observe volcanic plumes?

Volcanic plumes are observed using a combination of satellite, ground-based, and, more rarely, airborne measurements. Satellite observations are crucial for tracking ash and gas clouds over large spatial scales and in near real time. Ground-based instruments, such as radar, cameras, and infrasound sensors, provide detailed information on plume dynamics close to the source. Increasingly, these observations are integrated with numerical models to infer eruption conditions. The combination of multiple data streams is essential for constraining ESPs and improving the reliability of plume simulations.

What are some of the recent advances in estimating Eruption Source Parameters?

Recent advances have focused on combining observations with numerical models to better constrain ESPs. Multi-sensor approaches, data inversion techniques, and improved plume models have significantly enhanced our ability to estimate eruption rates and plume dynamics. At the same time, high-resolution computational fluid dynamics (CFD) simulations provide deeper insights into the complex fluid dynamic processes governing plume behavior. However, these models are computationally expensive and unsuitable for real-time applications, highlighting the need for approaches that bridge the gap between physical realism and operational efficiency.

What strategies do you propose in your review to improve Eruption Source Parameters estimation?

A central contribution of this review is the proposal of a new class of operational models for volcanic plumes.

A central contribution of this review is the proposal of a new class of operational models for volcanic plumes. These models integrate the physical realism of high-fidelity CFD simulations with the efficiency of simplified models used in forecasting. In particular, the review highlights the potential of artificial intelligence and machine learning techniques to “learn” from CFD results and optimally calibrate the key variables controlling plume dynamics. This hybrid approach allows complex physical processes to be represented in a computationally efficient framework, making it suitable for real-time applications while retaining improved accuracy.

How does improved volcanic plume monitoring lead to more effective volcanic hazard assessment?

Improved monitoring leads to more accurate estimates of ESPs, which directly translate into better forecasts of plume rise and ash dispersion. This reduces uncertainty in hazard assessments and supports more reliable decision-making. For example, more accurate forecasts can help aviation authorities minimize disruptions while maintaining safety and enable civil protection agencies to issue targeted warnings. Ultimately, better integration of observations and models enhances the capacity to respond effectively during eruptions and to mitigate their societal and economic impacts.

What are the remaining questions or knowledge gaps where additional research is needed?

Further research is needed to improve the coupling between observations, physics-based models, and data-driven approaches.

Despite progress, significant challenges remain. ESPs are still difficult to constrain in real time, and uncertainties in both observations and models propagate into forecasts. The integration of diverse data sources is not yet fully optimized, and different estimation methods can yield inconsistent results. Further research is needed to improve the coupling between observations, physics-based models, and data-driven approaches. In particular, developing robust hybrid frameworks that combine CFD, simplified models, and machine learning represents a key direction for advancing both scientific understanding and operational forecasting.

—Antonio Costa (antonio.costa@ingv.it, 0000-0002-4987-6471), Istituto Nazionale di Geofisica e Vulcanologia, Italy

Editor’s Note: It is the policy of AGU Publications to invite the authors of articles published in Reviews of Geophysics to write a summary for Eos Editors’ Vox.

Citation: Costa, A. (2026), From volcanic vents to safer skies, Eos, 107, https://doi.org/10.1029/2026EO265022. Published on 27 May 2026. This article does not represent the opinion of AGU, Eos, or any of its affiliates. It is solely the opinion of the author(s). Text © 2026. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.

The volcano on Indonesia's Halmahera Island routinely ejects ash, volcanic gases, and volcanic bombs. In May 2026, the Global Volcanism Program reported nine actively erupting volcanoes in Indonesia—more than any other country at the time. Such activity is typical for the Southeast Asian archipelago, where eruptions have occurred at 55 volcanoes since the 1960s—the highest total for any country. Japan ranks second with eruptions at 40 volcanoes over that time period, followed by the United States with 39, according to Global Volcanism Program data.

A new study (Sun et al. 2026) shows that in six earthquakes in China between 2010 and 2022, landslides and rockfalls were responsible for at least half of the total fatalities.

It is well-established that landslides are a major cause of loss of life in earthquakes in mountainous areas. The seismology maxim that “it is not earthquakes that kill people, it’s collapsing buildings” does not apply in its pure form in mountains – landslides also kill large numbers of people.

An earthquake triggered landslide from the 2008 Wenchuan earthquake.

However, the actual number of people killed by landslides in earthquakes is poorly understood. This is largely due to the challenges of collecting reliable information in the aftermath of a major earthquake, when the focus is on rescue and recovery rather than data collection. For this reason, many studies of landslide fatalities do not include seismically-triggered events. This is true of my own work.

However, a study has just been published (Sun et al. 2026) in the journal Natural Hazards Review that starts to address this issue. The paper nominally examines fatalities from all causes from earthquakes in China from 2001 to 2022. However, the authors note that the data has low reliability until 2010, so I’ll focus on the period from 2010 to 2022. I also note that the authors use the term “geological hazards“, which is a little broader than landslides. I should note that the paper isa broad look at fatalities from earthquakes – there is a much richer range of analyses than I will cover here.

In the period from 2010 to 2022, Sun et al. (2026) identified 14 earthquakes in which geological hazards caused loss of life. In some cases, the impacts were substantial. Thus, the M=6.5 3 August 2014 earthquake at Ludian in Yunnan led to 134 fatalities and 40 people missing from geological hazards from a total of 728 fatalities (c.24 % of the total), whilst the 5 September 2022 M=6.8 earthquake at Luding in Sichuan led to 76 geological hazard fatalities and 25 missing from a total of 118 fatalities (c.86% of the total). In six of the 14 examples, geological hazards caused at least 50% of the fatalities.

Sun et al. (2026) highlight that “fatalities from geological hazards concentrate in geologically complex, mountainous provinces, i.e., Sichuan, Yunnan, Gansu, Guangxi, and Guizhou”. They note that even small events can trigger fatal landslides – for example, six people were killed in a rockfall triggered by a M=4.3 earthquake in Guizhou in 2010, whilst a M=2.8 aftershock from the Yanjin earthquake in 2006 triggered a rockfall that killed a person.

This is an incredibly useful study. It starts to shed light on the impact of landslides in large earthquakes. It is not the definitive study, and questions remain – not least, the pattern of landslide losses in very large earthquakes, like the 2010 Wenchuan event, in which landslides were ferocious. But it forms the basis for such investigations, starting to fill a major gaps in our understanding.

Reference

Sun, B. et al. 2026. Causes Analysis of Earthquake-Related Deaths in Mainland China 2001–2022. Natural Hazards Review, 27 [2]. https://doi-org.ntu.idm.oclc.org/10.1061/NHREFO.NHENG-2458

Return to The Landslide Blog homepage Text © 2026. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.

From the Kleines Nesthorn to Pizzo Cengalo, the Alps appear to be crumbling. Permafrost researcher Robert Kenner has penned a summary explaining the role that thawing permafrost and melting glaciers play—or don't play—in this process.

Antarctica's Ross Ice Shelf and the West Antarctic Ice Sheet may have been far smaller during one of Earth's most recent warm periods, according to a new study that traced the origin of ancient dust preserved in Antarctic ice. Previous modeling studies suggest that the melting of the West Antarctic Ice Sheet could raise global sea levels by between three and five meters.

As meltwater drains through and beneath a glacier, it can alter how the ice flows and whether it breaks apart. Meltwater can also cause feedback that leads to more ice loss. Understanding when and how glacial meltwater drains is therefore critical to predicting how fast glaciers will lose ice and how that loss will affect sea level.

The first small satellite of the CloudCT network has been integrated, tested, and prepared for launch from California in June 2026. This precursor mission will be followed, if successful, by the launch of 10 additional CloudCT satellites in 2027, helping to fill gaps in our understanding of clouds and their role in climate.

Most people think of ice as frozen and lifeless, but research at Umeå University shows the opposite. A new study published in the Proceedings of the National Academy of Sciences demonstrates that ice actively speeds up the breakdown of iron minerals and may release more iron than current environmental models account for. This is crucial for predicting how nutrient cycles, carbon storage, and water quality will change in polar and mountain regions as the planet warms.

Magma on the move can cause the ground around a volcano to heave in measurable ways. But surface deformation doesn’t always point to an impending eruption—new results show that the terrain around a volcano can also shift during episodes of heavy rainfall. Researchers studying Japan’s Mount Fuji spotted instances of centimeter-level ground deformation tied to intense precipitation. Fortunately, such events can be readily differentiated from deformation caused by magmatic activity, the team reported in Geology.

Keeping an Eye on Volcanoes

Volcanoes around the world, from Kīlauea in the United States to Calbuco in Chile, are outfitted with arrays of sensors. Mount Fuji is no exception—the region around the edifice is equipped with dozens of instruments to detect ground movement, infrasound, and other signs of potential volcanic unrest. All that monitoring is warranted: Shin-Fuji (“Younger Fuji”)—the youngest of Mount Fuji’s three overlapping volcanoes—is currently active.

Shuo Zheng, a hydrological geodesist at Hong Kong Polytechnic University in China, and his colleagues recently mined some of those Mount Fuji data. The team focused on Global Navigation Satellite System (GNSS) observations—otherwise known as GPS data—collected daily from 2017 to 2023.

Rain and Rise

Zheng and his collaborators found several instances in which the two GNSS stations located within 10 kilometers of the summit of Mount Fuji recorded clear signs of uplift. Those signals, reflecting changes of roughly 1–2 centimeters, far exceeded the sensors’ millimeter-level precision. And when the team correlated the timing of that uplift with rain gauge records, they found that the ground often tended to rise almost immediately during periods of heavy precipitation (defined as several tens of millimeters of rain falling per day).

“They can store and transmit groundwater, acting like aquifers.”

There’s likely a physical link behind that correlation, the researchers surmised. The explanation involves the so-called clinkers that cap each of Mount Fuji’s subterranean layers of lava. Clinkers are layers of small rocks that form when the surface of a lava flow rapidly cools, and these structures persist in the shallow subsurface of Mount Fuji. “They can store and transmit groundwater, acting like aquifers,” Zheng said.

Clinkers, or layers of small rocks that form from cooling lava, can store and transmit water. They may be responsible for the way Mount Fuji’s surface uplifts in response to heavy rainfall. Credit: U.S. Geological Survey

When water fills up the pore space within a clinker, there’s no place for the overlying ground to go but up. It therefore makes sense that GNSS stations located atop old lava layers would exhibit uplift in response to intense rainfall, the team concluded.

When Zheng and his collaborators analyzed data from the nine GNSS stations located between 25 and 40 kilometers from the summit, however, they found that the ground actually tended to subside during periods of heavy precipitation. “There are two different responses,” said Kosuke Heki, a geophysicist and geodesist at Hokkaido University in Japan and a member of the research team. That subsidence is a known effect, and it’s been observed in a variety of locales. The subsidence doesn’t dominate closer to the summit of Mount Fuji because of the presence of the clinker layers there, the team reasoned.

Long-Lasting Magma

“Uplift by rain easily terminates when it stops raining.”

The uplift that the team recorded close to the summit of Mount Fuji tended to last just a day or two; it disappeared when the rainfall ceased. That timing is key for differentiating precipitation-induced uplift from magma-induced uplift. “Uplift by rain easily terminates when it stops raining,” said Heki. “But magma has a much longer timescale. It continues for weeks or months.”

That difference is critical, said Luca Caricchi, a volcanologist at the Université de Genève who was not involved in the research. There’s long been the mindset that ground deformation means that an eruption is imminent, but these new findings show that a heaving volcano doesn’t always mean that magma is on the move, said Caricchi. If the deformation is short-lived, the explanation might just be precipitation, he said. “You don’t need to worry.”

Zheng and his colleagues have looked for a similar effect for other volcanoes in Japan. They didn’t find any conclusive trends when they analyzed a chain of island volcanoes south of Tokyo, however. Perhaps that’s because the clinker layers beneath those edifices are so close to the sea that water efficiently drains out of them, the team hypothesized.

—Katherine Kornei (@KatherineKornei), Science Writer

Citation: Kornei, K. (2026), Heavy rainfall inflates Mount Fuji, Eos, 107, https://doi.org/10.1029/2026EO260169. Published on 26 May 2026. Text © 2026. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.

Source: AGU Advances

As meltwater drains through and beneath a glacier, it can alter how the ice flows and whether it breaks apart. Meltwater can also cause feedbacks that lead to more ice loss. Understanding when and how glacial meltwater drains is therefore critical to predicting how fast glaciers will lose ice and how that loss will affect sea level.

Chudley et al. modeled how the rate of water flowing into a glacier relates to seasonal changes in the forces that squeeze and stretch ice—forces caused by gravity pulling the glacier downhill, by the ice sliding over subglacial water, and by how portions of the ice interact with the ocean.

The researchers focused on the Sermeq Kujalleq glacier (also known as Store Gletsjer or Store Glacier) in Greenland. In spring, meltwater can fill cracks, or crevasses, that run through the surface of this glacier. These crevasses sometimes go on to drain as the year progresses.

The researchers used satellite imagery from the Sentinel-2 mission to see how much water was present in crevasses between 2016 and 2022, focusing especially on 2019, when the Sentinel-2 satellites provided the best coverage of the glacier. They fed those data into a convolutional neural network to map water cover through the season and looked for a relationship between the mechanical forces acting on the ice and the formation and drainage of crevasse ponds.

The researchers found that the mechanical forces acting on ice are the dominant factor in determining when crevasse meltwater drains into a glacier. When seasonal changes cause ice to stretch, crevasses can drain suddenly, releasing the water they held.

The Greenland Ice Sheet sheds trillions of gallons of water each year, and knowing when to expect that water to drain through the ice sheet is key to understanding processes such as how the glacier slides across the bed and when meltwater emerges in the ocean. The study’s results likely also shed light on dynamic processes in other glaciers and ice sheets, the authors say, and should help inform representations of ice behavior in numerical models. (AGU Advances, https://doi.org/10.1029/2025AV002150, 2026)

—Saima May Sidik (@saimamay.bsky.social), Science Writer

Citation: Sidik, S. M. (2026), Stretching and squeezing release glacial meltwater, Eos, 107, https://doi.org/10.1029/2026EO260152. Published on 26 May 2026. Text © 2026. AGU. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.

Editors’ Highlights are summaries of recent papers by AGU’s journal editors. Source: Journal of Geophysical Research: Solid Earth

Highly porous rocks, such as sandstones, often deform in a surprising way: instead of breaking apart or sliding, they develop thin zones called deformation bands. In these bands, the grains are squeezed closer together, making the rock denser, and reducing how easily fluids such as water or oil can move through it. This behavior is important because it affects both the strength of rocks and their ability to store and transport fluids underground. However, these bands are difficult to model because they form suddenly from an initially uniform material and concentrate deformation into very narrow zones.

Wang et al. [2026] developed a computer modeling approach called a “phase‑field model” to study this process. Instead of drawing the bands in the initially homogeneous rock, the model allows them to appear naturally as the system evolves and minimizes its energy. The study shows how grain crushing and rearrangement allows the formation of localized deformation zones. The results also demonstrate that natural spatial variations in the rock, such as differences in grain size or porosity, strongly influence where bands initiate and how they grow. Additionally, the model captures how deformation changes from sliding (shear bands) to pure compaction as pressure increases. Overall, this work provides a realistic way to understand how localized deformation develops in rocks, with important implications for geology, engineering, and energy applications.

Citation: Wang, Y., Zhang, C., Braun, P., Kang, X., & Wu, W. (2026). How does heterogeneity control strain localization patterns in high-porosity rocks? Journal of Geophysical Research: Solid Earth, 131, e2025JB032494. https://doi.org/10.1029/2025JB032494

—François Renard, Associate Editor, JGR: Solid Earth

Text © 2026. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.

Editors’ Vox is a blog from AGU’s Publications Department.

After 18 years of data collection, quality control, processing, and archiving, the United States Magnetotelluric Array (USMTArray) data set was completed in 2024. A new article in Reviews of Geophysics introduces this unprecedented data set and a new high-resolution model of the Earth’s crust and upper mantle that was made possible because of it. Here, we asked the authors to give an overview of magnetotellurics, how the USMTArray was developed, and future directions for research.

In simple terms for a non-specialist, what is the science of magnetotellurics?

Magnetotellurics (MT) is a passive geophysical technique capable of imaging the subsurface from hundreds of meters to hundreds of kilometers depth using the Sun and global lightning as sources. The science behind MT is largely based on Faraday’s law of induction, where external magnetic field variations induce telluric (from the Latin word ‘tellus’ meaning Earth) currents in the conducting Earth. These magnetic field variations are constantly occurring and happen over a wide range of time scales ranging from milliseconds to hours. And they are tiny – typically on the order of 0.1% of Earth’s magnetic field amplitude and even during intense magnetic storms rarely exceed 1%. 

By measuring these magnetic variations, and the induced electric field variations at Earth’s surface, we can constrain the 3D distribution of conductivity in the Earth. MT is an elegant method – we exploit powerful and distant energy sources which we have no control over and can mathematically remove the stochastic source spectrum to recover reliable estimates of Earth impedance. Impedance can be thought of as the Earth filter – a complex, frequency dependent set of functions that encapsulates all the information about the 3D conductivity structure beneath our feet. Through numerical inversion of impedance data at an array of sites, we build up 3D models of electrical conductivity.

What are some of the applications of the magnetotelluric method?

MT is applied across a broad spectrum of the Earth and space sciences ranging from mineral and geothermal resource investigations, to fundamental geologic and tectonic studies, to imaging the magmatic plumbing systems of active volcanoes, and to hazard mapping centered upon geomagnetically induced currents and the risk they pose to power grids.

Studies using MT are performed on every continent and in all tectonic settings, on land and on the ocean floor, on the Antarctic ice sheet, and even on the Moon. Because of its ability to image the entire lithospheric column, MT studies have made important contributions to our understanding of continental assembly by revealing ancient orogens and rifts. Moreover, MT is uniquely able to constrain the stability of cratonic roots by mapping hydration of the mantle lithosphere. MT studies are key to understanding active tectonic processes, including constraining the water budget in subduction zones, imaging melt zones beneath orogenic plateaus, and mapping the extent of crustal extension – for example beneath the western U.S.

Installation of a USMTArray site in the arid southwestern United States. Sites are installed in remote areas far from infrastructure (powerlines and pipelines) which can interfere with magnetotelluric measurements. Credit: Lena Tokmakoff

With the rise of computational power and 3D modeling and inversion codes, MT is now routinely used to study complex 3D systems, such as active volcanoes, geothermal systems, and mineral deposits. The sensitivity of MT to minor conductive phases – be it partial melt, clay, or conductive minerals such as graphite and metallic sulfides – make it ideal for studying these types of systems. As a result, MT is commonly employed within the resource sector at both the district and deposit scale. Many of the world’s iconic volcanoes have also been imaged with MT, where they constrain the geometry of crustal melt reservoirs – especially their volume and melt fraction which is in turn linked to the eruptibility of a subsurface magma. These analyses are especially powerful because they are sensitive to a distinct physical parameter – resistivity – of Earth materials. MT therefore provides unique and complementary information about the subsurface across a wide range of scales and is a particularly invaluable tool when other methods yield non-unique interpretations. 

One somewhat unexpected application of MT has been to space weather hazards. It was recognized a little over a decade ago that MT impedances are key to estimating surface electric fields generated during intense geomagnetic storms that can impact electric power grids. Past storms have knocked out power to vast areas and damaged critical infrastructure such as transformers. The importance of MT data to scenario analysis, in which power grid components are ‘stress tested’ against past geomagnetic storms, cannot be overstated. Regional to national-scale geoelectric hazard maps, both in the U.S. and internationally, are also informed by MT data, as are real-time geoelectric hazard estimates.

What is the United States Magnetotelluric Array (USMTArray)?

The USMTArray was an ambitious program begun in 2006 under the NSF-funded EarthScope program and completed in June of 2024 under USGS funding. The USMTArray collected long-period MT soundings on a 70-km grid across the contiguous U.S. – totaling more than 1,800 stations – each collected with uniform instrumentation, acquisition parameters, data processing, archiving, and metadata. Funded throughout its 18-year lifetime by three different federal agencies (the NSF, NASA, and USGS working closely with the Incorporated Research Institutions for Seismology and Oregon State University), the data – time series, response functions and metadata – were released incrementally to the public without data embargo or usage restriction.

Map of USMTArray site locations illustrating how the survey rolled across the country over its nearly two-decade lifetime. Credit: Kelbert et al. [2026], Figure 1

In broad terms, how was the USMTArray developed?

The USMTArray had humble beginnings – being mentioned in early planning documents as having value in understanding subduction zones and characterizing volcanic systems. Funded by NSF in 2003, the MT component of EarthScope was modeled after the much larger seismic component, with a transportable array of instruments to march across the U.S. on a 70-km spaced grid and a backbone array of seven instruments to study deep mantle structure. The USMTArray started off small and before dedicated instruments were even available. In 2006, a pilot study collected the first 30 stations in eastern Oregon using borrowed instruments, while subsequent years expanded what became known as the ‘northwest’ footprint, a 331-sites array completed in 2011 encompassing the Yellowstone-Snake River Plain, the Northern Rocky Mountains, the Cascades magmatic arc, and the northern Basin and Range province. Subsequent footprints in the midcontinent and the eastern U.S. continued to expand coverage.

What were some of the challenges in developing the USMTArray?

The biggest challenge by far was money. Within the EarthScope program, the USMTArray was never funded at the level needed to cover the contiguous U.S. The MT component was instead carried out as a series of footprints in areas deemed most scientifically advantageous. This limitation, however, led to one of the big successes of the USMTArray – active community engagement. Siting workshops held in 2008 and 2013 brought together participants from academia, government, and industry to discuss and prioritize where the array would go next, while a community working group provided scientific and operational guidance throughout the life of the array. The success of the USMTArray was recognized early on by the community governance of the EarthScope facility activities, with the ‘full-48’ concept endorsed in 2009, leading to modest increases in funding and an acceleration of station completions. In 2018, by the end of NSF-sponsored activities, roughly 2/3 of the contiguous U.S. had been covered. Seeing the array to completion, however, required additional funding, a challenge met by NASA (2019-2020) and the USGS (2020-2024), in large part due to recognition of the importance of USMTArray data to space-weather hazards and supported through executive orders in 2016 and 2019.

Another notable challenge that we faced while developing the USMTArray operations was the absence of established data sharing practices within the magnetotelluric community. Indeed, the concept of FAIR data was only introduced in 2016. Back in 2006 when this program commenced, the concepts of open data and systematic data sharing were largely unfamiliar, and no widely adopted, sustainable data formats existed. Available data formats were lacking in flexibility, consistency, and self-descriptive metadata. As the project progressed, our team developed such formats and accompanying databases, which have now reached maturity and are helping to drive more sustainable MT data‑sharing practices internationally.

How has the development of the USMTArray advanced the scientific field?

The USMTArray, along with parallel advances in modeling capabilities and increased computational power, ushered in a jump to 3D MT and to interrogating the Earth at regional to national scales. National-scale conductivity models, such as those developed from the USMTArray, now join the ranks of other data sets like magnetic, gravity, and seismic, and are a new lens with which to view the architecture of the North American continent. Numerous contributions to continental architecture and assembly and to understanding active tectonic processes have come from the USMTArray.

Map of the United States underground electrical structure integrated over mid- to lower crustal depths, illustrating the resistive (dark) and conductive (hot) regions. The latter reflects ancient tectonic scars within the crust. Credit: Kelbert et al. [2026], Figure 17

The USMTArray also serves as a framework for more detailed studies, allowing Principal Investigators (PIs) to derisk future surveys and industry to investigate anomalous or unexpected structure. Studies of the Cascadia subduction zone and the adjacent magmatic arc and geothermal energy prospectivity studies in the Oregon Cascades and Great Basin have been built upon the USMTArray while new MT surveys along the eastern seaboard are collecting high-resolution MT data to improve space-weather hazard maps over areas identified as particularly at risk from the analysis of USMTArray data.

Beyond the data and models derived from them, the USMTArray has motivated methodological advances, led to an investment in MT instrumentation and open-source software for researchers within the NSF-supported National Geophysical Facility, and served as a model for other regional and continental scale MT experiments.

What are some of the future directions for research in continental scale magnetotellurics?

With completion of the USMTArray, and the 3D conductivity models derived from it, there are numerous avenues for future research. Most models of continental evolution, for example, were developed prior to the advent of this rich data set. Critically evaluating such models in light of this new data set is paramount, and initial studies are already forcing a reexamination of certain paradigms.  

Multi-disciplinary studies incorporating geochronology, geochemistry, and rapidly evolving seismic models is another promising area as is the coupling of geophysical models to geodynamic models to examine the evolution of newly imaged model structure. Similarly, advancements in integrated and joint inversion are promising directions to leverage the wealth of public data sets available at regional to continental scales.  

Geology doesn’t stop at national borders or the land-sea interface – additional opportunities exist for cross-border arrays and onshore/offshore MT studies. Investigation of subduction zone processes and rifted continental margins by their very nature demand an amphibious approach.

On the applied front, resource assessments increasingly are applied at national and even global scales and demand data support at these same scales. Mineral resource assessments, for example, in the U.S., Canada, and Australia are exploring machine learning approaches to map prospectivity for various deposit types and incorporate a range of geophysical data layers to do so. Similarly, geothermal assessments can benefit from the consistent and synoptic data coverage offered by USMTArray data and models.

Finally, on the space-weather hazards front, partnering with power-system engineers to investigate data scale and uncertainty shows promise in generating accurate hazard maps and in improving upon operational, near real-time geoelectric field models. For all these future research directions the USMTArray remains both a framework and a benchmark upon which to build.

—Paul A. Bedrosian (pbedrosian@usgs.gov; 0000-0002-6786-1038), U.S. Geological Survey, United States; Anna Kelbert (anna.kelbert@cfa.harvard.edu; 0000-0003-4395-398X), Center for Astrophysics | Harvard & Smithsonian, United States; Adam Schultz (adam.schultz@oregonstate.edu; 0000-0003-1663-1547), Oregon State University, United States; and Gary D. Egbert (gary.egbert@oregonstate.edu; 0000-0003-1276-8538), Oregon State University, United States

Editor’s Note: It is the policy of AGU Publications to invite the authors of articles published in Reviews of Geophysics to write a summary for Eos Editors’ Vox.

Citation: Bedrosian, P. A., A. Kelbert, A. Schultz, and G. D. Egbert (2026), Mapping the hidden electrical anatomy of a continent, Eos, 107, https://doi.org/10.1029/2026EO265021. Published on 26 May 2026. This article does not represent the opinion of AGU, Eos, or any of its affiliates. It is solely the opinion of the author(s). Text © 2026. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.

Researchers from Tokyo Metropolitan University have categorized a unique, previously unclassified type of heat wave in Japan, so-called "moist heat waves" which are accompanied by an approaching tropical cyclone. These heat waves are accompanied by high humidity and/or heavy precipitation, and their frequency has been rising over the last 30 years, accounting for approximately a quarter of the heat wave days surveyed. They may potentially trigger multiple natural hazards at once.

SummaryThe conductive and capacitive properties of rocks are influenced by the type and concentration of the electrolyte present in the pore water. Sodium (Na⁺) and potassium (K⁺) are common pore water cations in saturated sedimentary rocks. Their distinct physicochemical properties are expected to produce different frequency-dependent electrical dispersion when adsorbed onto mineral surfaces. We tested this expectation by using spectral induced polarization (SIP), a method sensitive to interfacial processes. Complex conductivity spectra (10–2 to 105 Hz) were measured on two clayey, opal-A-rich diatomite samples, saturated with either NaCl or KCl solutions. One sample was tested over a stepwise increase in molar concentration (5.4–53 mM), while the other was tested over a stepwise increase in bulk water conductivity (0.050–0.48 S/m). At equivalent molar concentration, the in-phase conductivity of a sample was ~20 per cent higher when KCl saturated than when NaCl saturated, reflecting the greater molar conductivity of K⁺. At matched bulk water conductivity, which required a ~20 per cent higher NaCl molarity than KCl molarity, in-phase conductivity was ~10 per cent higher when NaCl saturated. In both tests, the quadrature conductivity and normalized chargeability followed a lower trend in the KCl-saturated state than in the NaCl-saturated state. This relatively low polarization for the K+ saturated state can be attributed to a weaker hydration and more compact adsorption of K⁺ within the inner layer of the electrical double layer. Additionally, time-lapse monitoring of complex conductivity spectra indicates that chemical equilibration via diffusion is achieved within 72 hours for both electrolyte types. This relatively rapid ionic diffusion is consistent with estimates based on the intrinsic formation factor and probably reflects the high porosity of the diatomite (~0.7). These findings establish that pore-water cation identity (Na⁺ vs. K⁺) is a primary control on SIP-derived polarization parameters, and cation identity must therefore be incorporated into petrophysical models to avoid biased estimates of surface area, permeability, and hydrogeochemical state.

Publication date: 15 August 2026

Source: Earth and Planetary Science Letters, Volume 688

Author(s): Fabrizio Di Fiore, Alessio Pontesilli, Giacomo Pozzi, Gianmarco Buono, Laura Calabrò, Silvio Mollo, Lucia Pappalardo, Piergiorgio Scarlato, Claudia Romano, Jacopo Taddeucci, Alessandro Vona

Publication date: 15 August 2026

Source: Earth and Planetary Science Letters, Volume 688

Author(s): Pingchuan Tan, Jiabiao Li, Jie Chen, Zhiteng Yu, Chunyang Wang

Publication date: 15 August 2026

Source: Earth and Planetary Science Letters, Volume 688

Author(s): Hongxi Pang, Tao Xu, Wangbin Zhang, Shuangye Wu, Shugui Hou

Publication date: Available online 21 May 2026

Source: Earth and Planetary Science Letters

Author(s): Sher-Rine Kong, Willem G.M. van der Bilt, Pål Tore Mørkved, Lars Wörmer, Katarzyna Hasal, William J. D’Andrea

Publication date: 15 August 2026

Source: Earth and Planetary Science Letters, Volume 688

Author(s): Yuanyuan Hua, Dapeng Zhao, Yang Yu, Yi-Gang Xu, Xiao-Long Huang

Publication date: 15 August 2026

Source: Earth and Planetary Science Letters, Volume 688

Author(s): Xianjun Fang, Sean D. Willett, Rong Yang, Dirk Scherler, Negar Haghipour, Marcus Christl