Feed aggregator

A new study found that land use (agricultural or forested) and the season (wet or dry) significantly impact groundwater quality, but in different ways.

As a glaciologist who thinks about ice a lot, rewatching the movie Frozen umpteen times with my six-year-old daughter provides ample opportunity for my imagination to run wild. The movie is set in the fictional kingdom of Arendelle, which is modeled on a fjord landscape, complete with a large glacier at the head of Arenfjord. Ice unsurprisingly plays a very prominent role in the story. Yet this glacier receives very little attention.

On the first day of the One Oceans Science Congress (OOSC) in France, it was clear this was not a normal science conference.

With flooding now an ever-present danger for communities in Australia and around the world, Australian researchers have demonstrated the effectiveness of world-first real-time water level and rainfall sensing technology using existing mobile phone networks.

Canada is home to more than 400,000 nonproducing oil and gas wells. These abandoned facilities still emit methane, which can contaminate water supplies and pollute the atmosphere with a greenhouse gas more potent than carbon dioxide. The scope of these emissions may be greater than previously understood, according to a new study.

“There’s a range of engineering, geological, and policy-related factors that are all playing a role in what emissions rates are observed.”

In 2023, nonproducing wells may have leaked 230 kilotons of methane, about 7 times more than the official estimates published in the government’s annual National Inventory Report (NIR). The NIR, compiled by Environment and Climate Change Canada (ECCC), informs the country’s greenhouse gas mitigation efforts and is submitted as part of Canada’s reporting obligations to the United Nations Framework Convention on Climate Change.

Methane estimates are calculated by multiplying the total number of nonproducing wells by emissions factors determined by well characteristics, such as the type of well (oil, gas, or unknown), depth, and whether it is plugged with concrete. These emissions factors offer only a rough idea of methane leakage, however.

“It’s really hard to predict emissions,” said Mary Kang, a study coauthor and associate professor of civil engineering at McGill University in Montreal. “There’s a range of engineering, geological, and policy-related factors that are all playing a role in what emissions rates are observed.”

Surprising Discoveries

To address this ambiguity, Kang and her colleagues measured methane flow rates at 494 nonproducing wells throughout Canada between 2018 and 2023 to define new emissions factors. While these sites account for only a fraction of the country’s abandoned wells, making uncertainty inevitable, the authors describe their data as the largest set of direct methane emissions figures collected through consistent methods.

They reported that the amount of methane leaked from the nonproducing wells was 1.5–16 times greater than NIR estimates.

Most of the departure from the NIR figures was driven by leaks from surface casing vents, narrow slits that ring the outermost steel layer surrounding the wellbore itself. Kang explained that emissions from surface casing vents indicate issues with a mine’s structural integrity and are trickier to manage than wellhead leaks, which may require only minor adjustments at the surface.

“The geology doesn’t care if you’re in one province or another….So what’s going on?”

The researchers analyzed their measurements to gauge how different well attributes contribute to methane flow rates. Whether a well is more prone to leakage than others, they found, isn’t determined by a single emissions factor such as its age or operating company.

Still, Kang was surprised to discover how much flow rates varied by province, even between wells operated by the same company in similar locations. The highest rates were observed in Alberta, where 74% of Canada’s known nonproducing wells are located.

“The geology doesn’t care if you’re in one province or another,” she said. “It’s the same formation. So what’s going on?”

Kang noted that each province and territory has its own emissions regulations, and policy factors might explain the differences in methane flow rates, though other geological differences such as seismic activity could also be at play.

Continuous Improvement

Complicating any study of methane emissions from nonproducing wells is the large number of sites abandoned before contemporary recordkeeping practices were established, said Maurice Dusseault, professor emeritus of engineering geology at the University of Waterloo in Ontario, who was not involved in the research.

A history of well abandonment practices in Ontario illustrates how hard it is to identify older wells throughout Canada. The first oil well in Ontario was drilled in 1858, but records were not mandatory in the province for another 60 years. Surface casings were often removed when a well closed so that the steel could be reused in other mines. This means some legacy wells cannot be detected with conventional magnetic techniques.

Still, Dusseault praised the researchers for their rigorous pursuit of better emissions estimates.

Kang and her colleagues returned to the field this year and last year, measuring methane flow at additional known well sites and revisiting previous sites to observe how leakage changes over time.

Meanwhile, their work is already affecting how the country approaches methane emissions. “Continuous improvement is a key principle of Canada’s NIR,” wrote ECCC spokesperson Cecelia Parsons in an email, noting that the improvement plan in the 2025 NIR draws from the new research.

—Lauren Schneider (@laur_insider), Science Writer

Citation: Schneider, L. (2025), Nonproducing oil wells may be emitting 7 times more methane than we thought, Eos, 106, https://doi.org/10.1029/2025EO250225. Published on 16 June 2025. Text © 2025. 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.

During explosive eruptions, tephra particles are injected into the atmosphere and undergo different fates: while larger particles settle close to the volcano, smaller ones remain suspended, forming volcanic clouds. In all cases, tephra poses significant hazards to human activities both near the volcano and hundreds of kilometers away.

A new article in Reviews of Geophysics explores our current understanding of tephra plumes and clouds, including their generation, characteristics, and monitoring strategies. Here, we asked the lead author to give an overview of tephra plumes, recent advances in modeling them, and what questions remain.

How does tephra form and spread?

Tephra forms through a process called fragmentation within volcanic conduits and is then expelled into the atmosphere by volcanic plumes. These fragments are classified based on their size: blocks and bombs (greater than 16 mm in diameter), lapilli (ranging from 2 mm to 16 mm), and ash (less than 2 mm in diameter). Once airborne, larger tephra particles typically settle near the volcanic vent, while finer particles (ash) can be carried by the wind over vast distances, forming what are known as volcanic clouds.

What kinds of hazards does tephra pose both in the air and on the ground?

Volcanic clouds pose a significant threat to aviation safety. When aircraft encounter these clouds, tephra particles can be ingested by jet engines, leading to performance degradation and, in severe cases, catastrophic engine failure. In addition, airborne tephra poses serious risks to public health. Studies on populations exposed to volcanic ash have documented increases in both acute and chronic respiratory conditions. The most dangerous particles are those smaller than 4 micrometers in diameter as they can penetrate deep into the lungs’ alveolar region, potentially triggering toxic reactions.

Tephra fallout can cause extensive damage to critical infrastructure, leading to substantial economic losses across multiple sectors.

On the ground, tephra fallout can cause extensive damage to critical infrastructure, leading to substantial economic losses across multiple sectors. These include energy systems, water and wastewater services, transportation networks (aviation, land, and maritime), food and agriculture, manufacturing, and communications. Rural communities, particularly those reliant on agriculture and livestock, are especially vulnerable. Tephra fallout can disrupt livelihoods not only in the immediate aftermath of an eruption but also over the long term. This is because tephra deposits can be remobilized by wind, generating ash storms that resemble the effects of the original eruption. These recurring events can persist for years, hindering economic recovery and prolonging hardship for affected communities.

What factors influence how far tephra spreads?

Tephra can be dispersed over vast distances, and in some cases, it may even travel around the globe. The extent of tephra dispersal is influenced by several factors, including the magnitude of the eruption, the size of the tephra particles (with smaller fragments remaining suspended in the atmosphere for longer periods), the volcano’s geographic location, and atmospheric conditions (particularly wind strength and direction). For example, the 1991 eruption of Mount Pinatubo in the Philippines, one of the major eruptions of the 20th century, injected massive amounts of volcanic ash into the stratosphere, which were carried by high-altitude winds and circled the globe in just 22 days.

Even relatively small eruptions can have major impacts when atmospheric and geographic conditions are unfavorable.

Similarly, the 2010 eruption of Eyjafjallajökull in Iceland, although moderate in size, caused significant disruption to European air travel. The fine-grained ash particles were carried thousands of kilometers by the jet stream, grounding flights across much of Europe for several days. This highlights how even relatively small eruptions can have major impacts when atmospheric and geographic conditions are unfavorable.

How do scientists monitor tephra plumes and clouds?

Scientists monitor tephra plumes and volcanic ash clouds using a combination of ground-based instruments and satellite observations. These data are essential for characterizing key aspects of volcanic activity, including plume extent, eruption column height, umbrella cloud spread, ash cloud altitude and thickness, tephra particle properties (such as size, shape, and settling velocity), mass eruption rate, sedimentation rate, and eruption duration. Ground-based tools include visible and thermal cameras, lidar, radar, infrasound microphones, and lightning detection antennas, each optimized for specific types of observations and deployed at varying distances from the volcanic vent.

Satellite sensors support global monitoring efforts through both active and passive remote sensing across a wide range of wavelengths, from ultraviolet to microwave. Modern ash cloud detection relies heavily on geostationary satellites, which provide high-temporal-resolution imagery (every 1 to 10 minutes), ideal for continuous real-time observation. However, these systems have trade-offs, such as coarse spatial resolution (approximately 4 km² at nadir) and limited coverage at high latitudes due to their equatorial orbital positioning.

Visible (a and c) and thermal (b and d) images of Mount Etna (Italy) plumes, acquired by the monitoring network of the Italian Institute of Geophysics and Volcanology, Osservatorio Etneo (INGV-OE). Courtesy of INGV-OE. Credit: Pardini et al. [2024], Figure 9

What are some recent advances in modeling tephra dispersal?

The movement of volcanic clouds and the deposition of tephra on the ground can be simulated using specialized numerical tools known as Tephra Transport and Dispersal Models (TTDMs). These models first emerged in the 1980s and have since undergone significant advancements in model physics, numerical solvers, and computational efficiency.

TTDMs require two main types of input data: meteorological information (such as wind speed, temperature, and pressure) and volcanic source parameters, which define what is emitted, how much is emitted, how particles are injected into the atmosphere (including their height and distribution), and the duration of the emission (start and end times). These models produce outputs that describe both the distribution of tephra suspended in the atmosphere and the patterns of tephra deposition on the ground. Modern TTDMs are capable of simulating complex atmospheric processes affecting tephra transport, such as particle aggregation and wet deposition (removal of ash particles by precipitation).

A recent development in the field is the emergence of in-line modeling approaches, which couple TTDMs directly with numerical weather prediction (NWP) models. In this integrated setup, the atmospheric evolution and tephra transport are computed simultaneously, eliminating the need to interpolate meteorological data between separate models. This approach improves the accuracy of tephra dispersal simulations, particularly under rapidly changing weather conditions. However, it comes at the cost of increased computational demand and is currently used primarily in research settings rather than for operational forecasting.

How have models contributed to improved forecasting and risk mitigation?

TTDMs play a crucial role in volcanic risk mitigation by providing forecasts of volcanic cloud movement and tephra deposition during eruptions. These models are especially valuable for early warning systems, enabling timely decisions to protect public health, aviation safety, and critical infrastructure.

One of the key operational users of TTDMs are the Volcanic Ash Advisory Centers (VAACs), which are a network of nine specialized agencies distributed globally under the mandate of the International Civil Aviation Organization (ICAO). VAACs are responsible for monitoring volcanic ash clouds and issuing advisories to aviation authorities. To do so, they routinely run TTDMs to predict the spatial and temporal extent of ash clouds, helping to prevent aircraft encounters with hazardous volcanic plumes. In addition, TTDMs are used by national meteorological and civil protection agencies to forecast and manage the ground-level impacts of tephra fallout. For example, the Japan Meteorological Agency (JMA) issues real-time forecasts of tephra dispersal and deposition following eruptions to support public safety measures. Similar practices are implemented in other volcanically active countries, such as Iceland and Italy.

Example output from the TTDM Ash3d, used at the Alaska Volcano Observatory to forecast the movement of volcanic clouds during periods of unrest. The example shown simulates a hypothetical volcanic cloud from Shishaldin volcano on 7 August 2024, using eruption source parameters that are considered realistic for that volcano. Results are publicly available at the Alaska Volcano Observatory website. Credit: Pardini et al. [2024], Figure 18

What are some of the remaining questions where additional modeling, data, or research efforts are needed?

In recent decades, there has been significant progress in our conceptual understanding of the processes that drive tephra plumes and the behavior of volcanic clouds. However, the inherent variability of explosive eruptions (ranging in style, location, and unique characteristics) continues to pose major challenges for both comprehensive understanding and effective monitoring.

Improving observational capabilities represents a critical frontier in volcanology.

One persistent difficulty lies in connecting model predictions with real-world observations. Large eruption plumes are rare, and even smaller events are difficult to characterize due to the limitations of current satellite systems, ground-based instruments, and visual data. As a result, improving observational capabilities represents a critical frontier in volcanology. Integrating these improved observations into modeling frameworks is essential, also to better understand underexplored processes such as particle aggregation and in-plume phase-change of water. 

The emerging potential of artificial intelligence in the detection and forecasting of tephra is increasingly recognized, although its current application remains limited, primarily to a few ash retrieval algorithms. In contrast, the use of large synthetic datasets generated by TTDMs to train data-driven models remains largely unexplored, despite the encouraging results achieved in other atmospheric dispersion contexts, where machine learning models have demonstrated strong generalization capabilities even under previously unseen conditions not represented in the training data.

—Federica Pardini (federica.pardini@ingv.it; 0000-0001-6049-5920), Istituto Nazionale di Geofisica e Vulcanologia (INGV), Sezione di Pisa, Pisa, 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: Pardini, F. (2025), Inside volcanic clouds: where tephra goes and why it matters, Eos, 106, https://doi.org/10.1029/2025EO255020. Published on 16 June 2025. 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 © 2025. 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.

Author(s): S. Morris, D. O. Gericke, S. Fritzsche, J. Machado, J. P. Santos, and M. Afshari

Recombination effects are shown to modify the charge-state distribution achieved in the acceleration of heavy ions during high-intensity laser-solid interactions. A recombination package has been developed for particle-in-cell codes, which includes dielectronic, radiative, and three-body recombinati…

[Phys. Rev. E 111, 065209] Published Mon Jun 16, 2025

Between 4 and 6 August 2023, intense rainfall triggered at least 346 landslides in the area around Ho Bon commune in Mu Cang Chai district, Yen Bai province in Vietnam

I have written frequently on this blog about clusters of rainfall-induced landslides. Another really interesting example has been highlighted in a paper (Toan et al. 2025) in the journal Landslides. Between 4 and 6 August 2023, intense rainfall triggered at least 346 landslides in the area around Ho Bon commune in Mu Cang Chai district, Yen Bai province in Vietnam.

Frustratingly, the paper does not give a lat / long location for this event (it continues to amaze me that this is not mandatory), but I believe the location is: [21.87657, 103.91738]. The exact nature of the rainfall event that triggered these landslides is uncertain as the loacl rain gauge was destroyed during the event. However, in the hour before the loss of data (6-7 pm on 5 August 2023), the rain gauge recorded 62.6 mm of precipitation.

The two Planet Labs images below show the outcome. The first was collected on 22 May 2023, before this event:-

Satellite image of the area that was affected by the August 2023 landslides around Ho Bon commune in Vietnam. Image copyright Planet Labs, used with permission. Image dated 22 May 2023.

The marker is situated on Ho Bon commune. And this is the aftermath of the event:-

Satellite image of the aftermath of the August 2023 landslides around Ho Bon commune in Vietnam. Image copyright Planet Labs, used with permission. Image dated 25 December 2023.

Image compare showing the landslides around Ho Bon commune in Vietnam. Images copyright Planet Labs.

This image highlights the landslides to the south of Ho Bon commune:-

Satellite image of the aftermath of the August 2023 landslides to the south of Ho Bon commune in Vietnam. Image copyright Planet Labs, used with permission. Image dated 25 December 2023.

Toan et al. (2025) do not claim that their mapping is comprehensive – and I think this is right as there appears to be more failures in the imagery than they have described. They note that the majority of the landslides that they mapped were debris flows, but I would probably characterise most of the failures in the imagery as shallow, disrupted landslides. They note that areas without forest cover were most seriously affected by landslides.

A really interesting aspect of this event is the number of failures that originate from the ridge line – this is commonly the case for earthquake initiated failures, but not for those triggered by rainfall. But Toan et al. (2025) note that this area has water traps on the ridgelines to feed water for rice field irrigation, so it is likely that these increased the rate of saturation, triggering failure.

In some locations, multiple shallow landslides combined to generate channelised debris flows. Parts of Ho Bon commune itself were damaged by such an event.

In total, Toan et al. (2025) document 88 damaged or destroyed houses, and extensive damage to the main road (NH32) through the area. They do not document any fatalities.

References

Planet Team 2025. Planet Application Program Interface: In Space for Life on Earth. San Francisco, CA. https://www.planet.com/

Toan, D.T., Duc, D.M., Quynh, D.T. et al. 2025. Extreme-rainfall-induced series of landslides and large flow-like landslides in Ho Bon commune, Mu Cang Chai district, Yen Bai province, Vietnam, in August 2023. Landslides. https://doi.org/10.1007/s10346-025-02544-5

Return to The Landslide Blog homepage Text © 2023. 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.

SummaryInduced polarization is a geophysical method that can be applied to determine the water content and CEC (Cation Exchange Capacity) of sediments and rocks. We apply here this technique to image the Harmalière landslide in Isère (East of France). This landslide includes a mudflow of glacio-lacustrine silty clays overlying the Toarcian clayrock formation with local occurrences of compact pebbles from a paleochannel of the Drac river. A petrophysical study is used to characterize the properties of the four main lithofacies occurring in the area with a total of 22 samples. We performed complex conductivity measurements in the frequency range 10 mHz-45 kHz at different salinities (NaCl brines and in situ pore waters). We also measured the porosity and CEC of the samples. We calibrate the relationships between surface conductivity, quadrature conductivity, and normalized chargeability with both the porosity and CEC. The relationship between the formation factor and the porosity conforms to Archie's law with a cementation exponent close to 2.0 ± 0.2. In the field, we performed a time-domain induced polarization survey using a 1.26 km-long cable (including roll along of the electrodes) with an electrode spacing of 20 m. The landslide is imaged down to a depth of 220 m. The inversion of the data (788 electrodes, 15.7 km of profiling, 13 012 apparent resistivity and 5539 apparent chargeability data) is done with the least-square technique penalizing the roughness of the tomograms using Occam inversion. The resulting 3D electrical conductivity and normalized chargeability tomograms are analyzed in conjunction with the petrophysical data to image the extension of the lithofacies at the field scale. Furthermore, the water content and CEC of the formations are imaged. We demonstrate that the compacted pebbles of the Drac paleochannel form both a mechanical and hydraulic barrier that is locally breached by the mudflow before entering the Monteynard-Avignonet Lake. This study demonstrates the ability of induced polarization to finely characterize the anatomy of such landslide and image its water content.

AbstractThis study investigates the following three issues in numerical models of the thermal structure of subduction zones, using the Tohoku region in Northeast Japan as an example: (1) a steady state is often assumed in models, (2) quantitative assessment of the uncertainty in the predicted temperatures is lacking, and (3) surface heat flow has been used to constrain many of the models. I found that, at least under the model setting of this study, a steady state may be safely assumed as long as only surface heat flow within 150 km of the trench is used to constrain the model. I used Bayesian inference to predict the thermal structure, with surface heat flow near the trench and the location of the blueschist-out boundary in the oceanic crust as observational constraints. The depth of slab–mantle kinematic decoupling, effective friction coefficient, and rate of radiogenic heat production in the upper island arc crust were constrained simultaneously to be ∼80–100 km, 0.03–0.08, and 1.5–2.16 μW m−3, respectively, although the decoupling depth is sensitive to the assumed location and temperature of the blueschist-out boundary. The uncertainties in slab temperature reach ∼450 K at depths of <100 km and 100 K for greater depths, which are substantial. To reduce these uncertainties, it is necessary to reduce the uncertainty in the input parameters and obtain additional observational constraints.

SummaryWater content and pore fluid pressure increases have been recognized as important drivers of shallow landslides, especially through the role of strong rainfalls promoting gravitational instabilities. Less recognized is the role of vertical hydraulic barriers impeding the flow of ground water at the feet of areas prone to landslides. Induced polarization is a non-intrusive geophysical technique able to image hydraulic properties of the shallow subsurface. Recently developed petrophysical models bridging the gap between hydraulic and electrical properties of soft sediments, soils, and rocks have been developed. Thanks to these relationships, this geophysical method can be used to image the water and clay contents of the formations and their permeability. Therefore, induced polarization can be used to image the occurrence of vertical permeability barriers. We focus our approach on a large landslide that occurred in March 1931 (reactivated in 1971–1972) above Le Châtelard village (Bauges, France). This landslide started inside a kilometer-scale syncline hosting clayey formations and moraines. We performed a 2.2 km profile crossing the syncline and the sliding area including resistivity, induced polarization, and self-potential measurements. In addition, 22 samples were taken from the different formations outcropping at the field site including limestones, sandstones, and clayey formations. The petrophysical investigations are combined with the field data to image the water content and cation exchange capacity as well as their permeability. The dataset shows the existence of a vertical permeability barrier at the bottom of the landslide corresponding to the tight Urgonian limestone formation. We combine the permeability distribution, the resistivity, and self-potential data by forward modeling the groundwater flow and electrokinetic response. We then invert the self-potential measurements to refine the image of the Darcy velocity distribution. The results show a strong upflow of the ground water just above the Manauds canyon where several gravitational instabilities occured in the past.

Methods to enhance the ocean's uptake of carbon dioxide (CO₂) are being explored to help tackle the climate crisis. However, some of these approaches could significantly exacerbate ocean deoxygenation. Their potential impact on marine oxygen must therefore be systematically considered when assessing their suitability.

SummaryHere, we investigate how continental rifts initiate and propagate across cratons by exploring the crustal structure of northwestern tip of the East African Rift System (EARS), hosting the volcanic-rich Edward-George and non-volcanic Albertine-Rhino rifts, and their termination at the Precambrian Aswa Shear Zone. We conducted a derivative analysis of magnetic data, utilized power spectral analyses, and implemented a two-dimensional (2D) forward modeling of gravity data constrained by the seismic results obtained from the region. A magnetic derivative map indicates that the border faults of the Albertine Rift, at regional-scale, trend parallel to the Mesoproterozoic Madi-Igisi fold belt (MIFB) structures, representing the suture zone between two Archean microcratons. Our results show a pronounced thinned crust (∼24–30 km) beneath the southern segments of the rift zone, particularly the Edward-George rift, the Rwenzori Mountains, and the southern Albertine graben, consistent with previous seismic studies. In general, we observe that: 1) the rift system follows the boundary between a broadly thinner crust (21–41 km) to the southeast in Uganda, and thick crust (34–41 km) to the northwest in Congo, and 2) within the rifts, the crustal thickness along the axes exhibits a strong gradient that attenuates northwards beneath the Albertine-Rhino graben. We supplement the geophysical results with field observations of an exhumed Permian ‘Karoo’ rift (Entebbe Graben) in central Uganda, indicating the possible source of inherited thinner crust to the southeast of the Albertine-Rhino Rift. We propose that the northwestern tip of the EARS exploited a cratonic crustal thickness-gradient, assisted by structural inheritance from crustal metamorphic fabrics, and potentially, thermo-mechanical weakening of the deeper crust by partial melts beneath some of the rift segments.

Coastal planners take heed: Newly uncovered evidence from fossil corals found on an island chain in the Indian Ocean suggests that sea levels could rise even more steeply in our warming world than previously thought.

A team of planetary scientists, ecologists, and marine biologists affiliated with several institutions in the U.S. and one in the U.K., has found evidence suggesting that parts of the world's oceans have already passed what has come to be known as a planetary boundary.

Forest fires are a fundamental force in Earth's dynamics with a direct impact on human health, food security, and biodiversity. From air quality to landscape configuration and resource availability, the consequences of fire have influenced the development of society throughout history. Their effects on the oceans, though less known, are equally significant.

California produces more than a third of the vegetables and three quarters of the fruits and nuts in the United States. But water constraints are leaving more and more fields unplanted, or “fallowed,” particularly in the state’s famed farming hub, the Central Valley.

In a study published in Communications Earth and Environment, researchers showed that these fallowed agricultural lands are producing a different problem: dust storms, which can cause road accidents and health problems and can have far-reaching environmental impacts. Using remote sensing methods, the team found that 88% of anthropogenic dust events in the state, such as dust storms, come from fallowed farmland.

California’s frequent droughts could mean a rise in fallowed farmland. In 2014, the state passed the Sustainable Groundwater Management Act (SGMA), a policy aimed at ensuring the sustainability of groundwater resources. A report by the Public Policy Institute of California suggested that to meet the SGMA’s demands, farmers may need to fallow hundreds of thousands of additional acres, potentially worsening dust events.

Tracking Down Agricultural Dust

Dust can come from both natural sources, such as wind blowing across a desert, and anthropogenic sources, such as when transportation, construction, or agricultural activities kick up particles. Previous studies identified agriculture as a significant source of human-generated dust, but study author Adeyemi Adebiyi and his colleagues wanted to narrow down which agricultural practices produced the most.

“If you stop irrigating the land, it becomes dry, and we’re already in a dry climate. It’s easy for it to become a new dust source.”

Fallowed land was a logical culprit. “If you stop irrigating the land, it becomes dry, and we’re already in a dry climate,” said Adebiyi, an atmospheric scientist at the University of California, Merced. “It’s easy for it to become a new dust source.”

The researchers started by pinpointing fallowed land across California between 2008 and 2022 using U.S. Department of Agriculture datasets. The data showed that 77% of the state’s fallowed land was in the Central Valley. 

The team then examined NASA satellite images of atmospheric aerosols, identifying which aerosols were dust particles on the basis of the way they scatter light. When they overlaid the regions that regularly experienced dust events with the agricultural data, they saw that dust events were tightly associated with fallowed fields.

The problem appears to be getting worse. Between 2008 and 2022, both the area of fallowed land and corresponding dust levels have increased: In this period, the amount of dust in the atmosphere over the Central Valley grew by about 36% per decade.

Having grown up in California and spent the first decade of his career studying dust in the Central Valley, Thomas Gill, an Earth scientist at the University of Texas at El Paso who wasn’t involved in the study, has long worried that land use changes could lead to dust issues. “This study by Adebiyi et al., unfortunately, shows that my worries have been coming true,” he said.

“These fallowed land locations are emblematic of the properties you would normally see in a typical desert-type location.”

Daniel Tong, an atmospheric scientist at George Mason University who also wasn’t involved in the study, agreed that the work provides some much-needed conclusive data on the connection between land use and dust levels. “This is a very useful study,” he said.

Adebiyi’s team used additional remote sensing data to determine that compared with nearby nonfallowed land, fallowed fields have lower soil moisture and are about 4.2°C hotter. Combined with a lack of vegetation, these factors work together to make such areas more prone to wind erosion. “These fallowed land locations are emblematic of the properties you would normally see in a typical desert-type location,” Adebiyi said.

Far-Reaching Effects

The dust from fallowed fields has wide-reaching consequences. “California is already the state with the largest number of fatalities caused by dust storms,” said Tong, who authored a 2023 study about windblown dust fatalities in the United States. One concern, he said, is that more dust storms could increase road accidents. Dust also contributes to respiratory problems and cardiovascular disease and carries the Coccidioides fungus, which causes the dangerous infection valley fever. Cases of valley fever increased by 800% in California between 2000 and 2018.

“There’s also been a great population increase in the Central Valley,” Gill said. “So not only do you have more particulate matter, but you have more people living there who are vulnerable to its effects.”

Fallowed fields and the dust they produce may also work counter to the groundwater management goals of the SGMA. The Central Valley dust blows east into the Sierra Nevada Mountains, where it speeds snowmelt, a significant reservoir of water for the state. The researchers also found that the heat concentrated in fallowed fields can spread out to nearby fields, causing surrounding crops to need more water. “It’s a double whammy,” Adebiyi said.

He noted the importance of preventing fields from becoming completely bare while still conserving water. One strategy is to plant native, drought-resistant plants that protect the soil from wind erosion without needing much irrigation.

The researchers are now conducting similar studies on the connection between fallowed lands and dust in other agricultural states, such as Kansas, Montana, and Washington. Their findings suggest that addressing dust problems will become increasingly important nationwide.

“The implications are beyond California,” Adebiyi said. “They’re across the United States.”

—Andrew Chapman (@andrewchapman.bsky.social), Science Writer

Citation: Chapman, A. (2025), Fallowed fields are fueling California’s dust problem, Eos, 106, https://doi.org/10.1029/2025EO250223. Published on 13 June 2025. Text © 2025. 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.

Aerosols could hold the key to stopping potentially destructive cyclones in their tracks, according to a first-of-its-kind study from The Australian National University (ANU).

Author(s): Kathryn M. Sturge, Noah Hoppis, Brian L. Beaudoin, Ariana M. Shearin, Ethan T. Basinger, Bryson C. Clifford, Jack R. FitzGibbon, Emily H. Frashure, James E. Krutzler, Abraham A. Levitan, Patrick O'Shea, Holly J. Wilson, and Timothy W. Koeth

Electrostatic discharges occur in numerous media on a range of length scales, from microscopic discharges inside electronic materials to kilometer-long channels in air during natural lightning events. To study the mechanisms and behavior of electrostatic discharges inside materials, we measured the …

[Phys. Rev. E 111, 065207] Published Fri Jun 13, 2025

Author(s): M. W. C. Dharma-wardana and Dennis D. Klug

Experiments at the U.S. National Ignition Facility (NIF) [Döppner et al., Nature (London) 618, 270 (2023)] have created highly compressed hot hydrogenlike Be plasmas. Published analyses of the the NIF experiment have used finite-T multiatom density-functional theory with Molecular dynamics, and Pat…

[Phys. Rev. E 111, 065208] Published Fri Jun 13, 2025