EOS

Want updates from a trusted source about how scientists monitor groundwater and drought? Preference Eos in your searches! Go to Google

From 2020 to 2022, California faced one of its driest periods on record, in part because of human-caused climate change. Though the drought has eased, scientists have noticed lasting effects on the state’s aquifers that may affect their ability to store water in the future.

A new study published in the Proceedings of the National Academy of Sciences of the United States of America found that the drought and resulting overpumping of groundwater caused some aquifers in the Sacramento Valley to collapse and become irreversibly damaged. The study is the first to capture this kind of aquifer collapse in high-resolution satellite data and provides evidence in support of giving the Sacramento Valley’s aquifers additional protections, the authors write.

“How fast part of the Sacramento Valley is subsiding was quite surprising to us,” said Stacy Larochelle, a geophysicist at the University of California, Los Angeles, and lead author of the new study. “This is really one of the first times we could capture this transition in so much detail.”

“It’s sad that the Sacramento Valley is seeing more subsidence, but that it’s being recognized, documented, and looked at in more detail is a good thing,” said Claudia Faunt, a retired hydrologist at the U.S. Geological Survey California Water Science Center who was not involved in the new study.

Sinking Sacramento Valley

Larochelle did not set out to measure the drought’s impact on Sacramento Valley aquifers. Rather, she noticed a periodic trend in regional Global Navigation Satellite System (GNSS) data, which track ground deformation using fixed, ground-based instruments. The data showed that aquifers were being depleted and refilled at a regular, seasonal pace from 2016 to 2020. She noticed a striking change around 2021, when parts of some aquifers suddenly sank.

Wanting more information, Larochelle and the research team looked to interferometric synthetic aperture radar (InSAR) data, which track land surface deformation over time using radio waves sent from satellites. InSAR data, GNSS data, and groundwater monitoring wells all showed the same pattern.

When humans pump groundwater out of an aquifer, they remove water between the aquifer’s rocks and sediment. This removal creates empty space, and the aquifer shrinks. Usually, this process is elastic: Once the aquifer fills with water again, the spaces between the rocks and sediment swell, and the aquifer rebounds.

“You have a permanent collapse of the pore space.”

But if too much water is removed too rapidly, the structure of the aquifer’s sediment deforms irreversibly, and it loses its ability to recharge and store water in the future. “You have a permanent collapse of the pore space,” Larochelle said.

When water was removed during the 2020–2022 drought, for example, some areas of the Sacramento Valley subsided at rates up to 30 centimeters per year, enough to damage the aquifer’s infrastructure.

The end of the drought and influx of strong precipitation events in 2023–2025 have not returned these damaged aquifers to their former state, though researchers say more study is needed. “From preliminary observations, we don’t see a huge rebound,” Larochelle said. “We’re not seeing a full recovery; we’re not seeing the ground go back up.” She said she does not expect some parts of the aquifers to ever recover.

In the long term, that means the aquifer will not be able to store as much water for future generations, which could also increase flood risk as the ground is unable to absorb as much precipitation.

Protecting Aquifers

In 2014, California’s Sustainable Groundwater Management Act (SGMA) created statewide regulations guiding groundwater pumping. Water managers designated many basins in the San Joaquin Valley, adjacent to the Sacramento, as “critically overdrafted,” changing the basins’ monitoring requirements. (The Sacramento and San Joaquin Rivers interact with their valleys’ aquifers in complex ways. Though river flow is one indicator of the health of an aquifer, the California Department of Water Resources mainly monitors groundwater flow from wells.)

The Sacramento Valley, drained by the Sacramento River, is the northernmost part of California’s large Central Valley aquifer. It borders the San Joaquin River Basin, which has many “critically overdrafted” subbasins (colored in red). Click image for larger version. Credit: California Department of Water Resources

“Maybe we should change the designation of the Sacramento Valley,” Larochelle said, “so that we’re as careful in the way we extract groundwater from it as we are with the southern part of the valley.”

Faunt agreed, saying that the California Department of Water Resources should determine if changes to SGMA classification are warranted in areas where scientists are beginning to see a permanent loss of storage. “The authors have a point,” she said.

The team’s methods also provide a way to track the impacts of groundwater pumping on aquifers in real time using satellite data. Such tracking could give communities an early warning that they’re pumping too much groundwater, before an aquifer is irreversibly damaged, Larochelle said. Though that’s not a completely new idea, Faunt explained that using InSAR data, GNSS data, and ground-based data all together in one method is a newer approach that helps scientists see how the different types of data align or diverge.

Another of the study’s findings was that most of the ground-based groundwater monitoring data in the Sacramento Valley are for shallow parts of the region’s aquifers. That disparity indicates a need to monitor all levels of an aquifer to be able to gather a three-dimensional picture of which layers are being affected by pumping, Faunt said.

The study could also help scientists better measure aquifers without ground monitoring data, Larochelle said. California has one of the best groundwater monitoring systems in the world, so using a combination of satellite data and ground-based measurements there can help calibrate satellite measurements in other parts of the world that may not have a dense network of ground-based instrumentation.

“That’s the dream, long term,” Larochelle said. “To be able to just use satellites to see what’s going on in the groundwater system.”

—Grace van Deelen (@gvd.bsky.social), Staff Writer

Citation: van Deelen, G. (2026), California’s drought irreversibly damaged Sacramento Valley aquifers, Eos, 107, https://doi.org/10.1029/2026EO260267. Published on 21 August 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.

body {background-color: #D2D1D5;} Research & Developments is a blog for brief updates that provide context for the flurry of news that impacts science and scientists today.

Earlier this year, NASA announced that one of its powerhouse telescopes, the Neil Gehrels Swift Observatory, was falling from orbit much faster than anticipated. Although Swift’s low-Earth orbit had remained relatively steady for more than 2 decades, unusually strong solar activity in 2024 destabilized it. The agency predicted that the telescope would burn up in Earth’s atmosphere by the end of 2026.

NASA soon after announced a commercial effort by U.S. company Katalyst Space Technologies to rescue the telescope by using another spacecraft to boost Swift’s orbit. The mission launched successfully on 3 July. However, after a month of technological issues with the rescuing craft’s maneuvering thrusters, NASA and Katalyst have pulled the plug on the mission, leaving Swift to its fiery fate.

 
Related

•  FAQ About Swift
•  About the rescue mission attempt
•  NASA Updates Next Steps for Commercial Swift Boost Mission
 

“This is not the outcome we were working toward, but it does not change why this mission was worth attempting,” NASA administrator Jared Isaacman said in a 19 August statement. “The team moved with extraordinary speed to give Swift a chance to carry out more science while advancing capabilities America will need for satellite servicing in the future.”

Swift launched in 2004 with the goal of monitoring unpredictable high-energy astrophysical phenomena called gamma-ray bursts. The mid-sized mission operates out of a small building near Pennsylvania State University in State College with the picturesque Allegheny Mountains as a backdrop.

The telescope was designed with the ability of spotting these short-lasting explosions and quickly pivoting to monitor the events and their aftermaths in multiple wavelengths. But during its 21 years of science operations, Swift became a workhorse of multiwavelength astronomy observations, studying everything from the active centers of distant galaxies to supernovae near and far, as well as ravenous black holes, brown dwarfs, and interstellar objects. Take a look back at some of its most notable discoveries.

The BOAT: Over and over since its science operations began, Swift announced the detection of powerful gamma-ray bursts that shattered previous records, including the brightest of all time (BOAT). Since launching in 2004, it has made 829,336 observations and spotted more than 1,800 gamma-ray bursts.

Left: Swift has detected nearly 2,000 gamma-ray bursts since its launch in 2004. Credit: NOIRLab/NSF/AURA/M. Garlick, CC BY 4.0

Interstellar Object Chemistry: Swift also observed objects much closer to home, though these objects had distant and exotic origins, too. In 2019, Swift’s ultraviolet instrument detected water coming off of the interstellar object 2I/Borisov. In 2025, it performed similar observations of the interstellar comet 3I/ATLAS.

Right: Swift detected the presence of outgassed water coming from interstellar comet 2I/Borisov. Credit: NASA, ESA, and D. Jewitt (UCLA)

“Rosetta Stone” Supernova: In 2008, Swift spotted an X-ray burst in galaxy NGC 2770. It turned out to be the precursor to a supernova, spotted in an earlier stage than ever before. Because of Swift’s rapid response, astronomers around the world could observe the supernova throughout its evolution in multiple wavelengths. With so many types of observations of a single event, SN 2008D has been called the “Rosetta stone of supernova studies.”

Left: Some supernovae detected by Swift resulted in magnetars. Credit: NASA E/PO, Sonoma State University, Aurore Simonnet

Black Holes Snacking on Stars: When a star gets too close to a black hole, the black hole’s gravity can break the star apart into a stream of gas. The black hole can gobble up the gas and burp out some intense radiation. In 2023, Swift observed one black hole repeatedly taking bites out of an unlucky star that ventured too close. Every few weeks the black hole—with the mass of a whopping 200,000 Suns—swallows three Earth-masses of material from the star and belched out x-rays.

Right: Swift observed bursts of energy as a black hole snacked on a star. Credit: NRAO/AUI/NSF/NASA

Brown Dwarf Wandering By: Swift joined forces with NASA’s Spitzer Space Telescope in 2016 to observe a microlensing event, which occurs when a close-by passing object distorts the light coming from a more distant one. With this technique, Swift helped discover a brown dwarf 80 times the mass of Jupiter that orbits close to a Sun-like star. Swift and Spitzer’s discovery, OGLE-2015-BLG-1319, is one of the few brown dwarfs found to orbit their host stars within a few Earth-Sun distances.

Left: Swift aided in the discovery of a rare brown dwarf using gravitational microlensing. Credit: NASA/JPL-Caltech

Milky Way Magnetars: When some stars die they become small, dense balls of neutrons with extremely intense magnetic fields. These magnetars can release short, strong bursts of energy as they interact with surrounding material. Swift has studied many of these magnetars, including one that hid in the center of the Milky Way and masqueraded as our galaxy’s supermassive black hole, and another one surrounded by a “wind nebula.”

Right: Magnetars can release intense bursts of radiation, which have been picked up by Swift. Credit: ESO/L. Calçada, CC BY 4.0 Science After Swift

Swift launched with a nominal mission lifetime of 2 years. It has lasted nearly 22 years, and yet its demise will leave a gaping hole in NASA’s ability to study the high-energy universe. Although NASA said that it would “continue to prioritize finding new options to react readily to cosmic events,” no telescope in NASA’s current or upcoming portfolio will look at the universe at the same wavelengths as Swift or with the same ability to monitor unexpected transient events.

“We knew this was a high-risk, high-reward [rescue] mission—a first-of-its kind attempt, developed on an unprecedented timeline driven by the Sun’s activity,” Shawn Domagal-Goldman, director of NASA’s Astrophysics Division, said in a statement. “We were all hoping for more science from Swift. But we knew the takeaways from this mission would be worthwhile either way.”

—Kimberly M. S. Cartier (@astrokimcartier.bsky.social), Staff Writer

These updates are made possible through information from the scientific community. Do you have a story about science or scientists? Send us a tip at eos@agu.org. 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.

Source: Journal of Geophysical Research: Biogeosciences

Methane-munching microbes in soil might be more important than previously thought, a new study finds. Soil is an important carbon sink, and scientists are still learning much about the diversity of its microbial communities. They’re being uncovered from Arctic soils to desert sands, and some of them are sucking down methane—a greenhouse gas about 27–30 times more potent than carbon dioxide over 100 years.

Soil methanotrophs are organisms capable of biologically removing methane from the atmosphere. Current estimates vary widely, but soil methanotrophs may store an average of 28–35 gigatons of methane per year globally. And even that could be an underestimate, scientists suspect.

Previous efforts to estimate the global biological methane source from wetlands and inland fresh waters primarily used process-based modeling, which focuses on biogeochemical processes, and atmospheric inversion modeling, which starts with methane concentrations in the atmosphere and works backward to determine emission sources.

But estimates from these two approaches tend to differ. The bottom-up, process-based estimates of methane emissions from wetlands and inland fresh waters were higher than the top-down, atmosphere-based estimates. A larger soil sink could help offset some of these discrepancies, bringing net bottom-up estimates closer to those inferred from the atmosphere.

Oh et al. dig in to reconcile that discrepancy and refine the estimate of how much methane-munching soil microbes contribute to the global methane sink.

The authors added a third kind of modeling: data-driven machine learning. By running the three kinds of models in parallel and comparing their results, the researchers hoped to home in on a more reliable estimate with smaller uncertainties. They also tweaked the microbial dynamics in the process-based model and included previously overlooked places and microbes.

The three-pronged approach worked. Both process-based and machine learning models yielded similarly sized sinks. According to their estimates, microbes in soils take up 40–45 gigatons of global methane per year, significantly higher than estimates from older approaches. That value is also larger than estimates in global climate assessments, such as that of the Intergovernmental Panel on Climate Change. When incorporated in top-down atmospheric inversions, this larger soil methane sink also improved the models’ ability to reproduce observed atmospheric methane and its stable carbon isotope composition.

The findings suggest that the microbial soil methane sink has been underestimated and that the revised three-model approach may improve global carbon cycle modeling. (Journal of Geophysical Research: Biogeosciences, https://doi.org/10.1029/2025JG009668, 2026)

—Rebecca Dzombak, Science Writer

Citation: Dzombak, R. (2026), It takes three to model methane right, Eos, 107, https://doi.org/10.1029/2026EO260268. Published on 20 August 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: Tectonics

The Eastern and Southern Alps deform slowly compared with many tectonically active regions worldwide, making their ongoing movements especially difficult to detect. Grützner et al. [2026] provide an unusually detailed synthesis of decades of research across nine countries, combining evidence from different datasets, methods, catalogues, and research traditions. By comparing and reconciling these sources, the study creates a comprehensive framework for understanding how active faults and earthquakes relate to climate, the structure of Earth’s outer rocky layer, and deeper processes within the planet.

The results show that present-day deformation is controlled mainly by a strong crustal block beneath the Dolomites pushing into the Alps, movement between relatively rigid blocks of rock, and uplift as the land rebounds following the melting of ice-age glaciers. The slow movement of hot mantle rock deep within Earth appears to play a smaller role than previously proposed. The areas of greatest earthquake hazard are concentrated along the southern edge of the Alps and where the Alps meet the Dinarides. This synthesis will provide an important reference for researchers studying the Alps and other slowly deforming regions, where weak tectonic signals and varied datasets make earthquake-hazard assessment particularly challenging.

Citation: Grützner, C., Petersen, G., Serpelloni, E., Metzger, S., Moernaut, J., Ustaszewski, K., et al. (2026). Active tectonics of the eastern and southern Alps – Crustal response to deep processes? A review. Tectonics, 45, e2025TC009267. https://doi.org/10.1029/2025TC009267

—Lothar Ratschbacher, Associate Editor; and Djordje Grujic, Editor, Tectonics

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.

At least 107 people were killed in major failure at an artisinal mine on the border with Cameroon.

In the afternoon of 18 August 2026, a large landsldie occurred at an artisinal gold mine at Zamboye (also spelt Zamboï) in the Central African Republic. The landslide was caught on videos that have been widely shared (there are two different angles). Please be aware that it makes uncomfortable viewing:-

I am unsure of the precise location of this accident. It is described as being close to Garoua-Boulaï, which is located at [5.888, 14.550].

News reports indicate that 107 bodies have been recovered to date, but that the toll might be higher.

At first sight, the behaviour of the miners looks odd – there is a huge crowd directly in the path of what is clearly an unstable slope. I suspect though that the videos capture just one of a series of failures. This screenshot is from early in the recorded failure sequence:

Screenshot of the video of the 18 August 2026 gold mine landslide at the Zamboye (Zamboï) mining site in the Central African Republic.

It appears to me that this landslide moves onto, and mobilises, debris from an earlier failure, which can be seen in the bottom right hand corner of the image. I would speculate that the failure event(s) captured in the videos might be just one of a series of failures.

Were the crowds trying to rescue people already buried when this recorded landslide occurred?

Of course, this serves to remind us of the horrific dangers faced in artisinal mines, such as the one at Zamboye.

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.

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

Every lightning flash has one defining moment: the return stroke, an intense surge of electric current that rockets upward along the ionized channel at a sizable fraction of the speed of light, unleashing the blinding flash, the crack of thunder, and the burst of radio energy that detection networks use to pinpoint strikes worldwide. Despite decades of study, a full, self-consistent explanation for why this current takes the shape it does — its rapid rise, its slower decay, its weakening and spreading as it climbs — has remained elusive.

In a new article published in Reviews of Geophysics, Caitano da Silva and colleagues at New Mexico Tech show that the Telegrapher’s Equations, a compact framework describing how electrical signals propagate along any conductor, can be adapted into a physically transparent model that derives all of these features from first principles — and reconciles them with decades of field and laboratory measurements. Here, the authors answer a few questions about their work.

In simple terms, what is the “lightning return stroke current”?

When a downward-moving leader from a thundercloud gets close enough to the ground, a channel of ionized air connects cloud to ground. At that instant, a powerful surge of electric current — the return stroke — rushes upward along this newly formed conducting path at a sizable fraction of the speed of light, carrying tens of thousands of amperes. This surge is what produces the visible flash, heats the air explosively to create thunder, and radiates the burst of radio waves that lightning detection networks use to pinpoint strikes. It’s the most energetic and consequential part of a lightning flash, even though it typically lasts only tens of microseconds.

Why are return strokes important to study?

The return stroke current is responsible for most of lightning’s real-world impacts. It causes billions of dollars in damage annually to power transmission lines and communication infrastructure, and it’s a leading ignition source for wildfires. It is also the atmosphere’s main natural source of nitrogen oxides, which influence atmospheric chemistry on regional and global scales. On top of that, the radio pulse the return stroke emits is exactly what national and global lightning-detection networks measure to locate strikes, supporting both hazard mitigation and weather forecasting. Understanding the physics that shapes this current — its peak strength, its speed, and how quickly it weakens — is therefore essential for protecting infrastructure and for interpreting the remote-sensing data scientists rely on.

What are the main types of models scientists use to simulate the return stroke?

Researchers have taken a few different approaches. “Gas-dynamic” models solve the detailed physics of how the current heats and expands the channel of air, which is useful for calculating channel temperature and chemical byproducts, but they need the current as an input rather than predicting it. To calculate the current and resulting electromagnetic fields directly, three families of models exist: “engineering models,” which simply assume a plausible mathematical shape for the current and how it weakens with height; “antenna-theory models,” which apply full numerical electromagnetics; and “distributed-circuit models,” which treat the lightning channel as an electrical transmission line. This last approach, governed by the Telegrapher’s Equations, is the focus of this review.

What are the Telegrapher’s Equations, and what can they tell us about return strokes?

The Telegrapher’s Equations describe how current and voltage evolve along any conductor with distributed resistance, inductance, and capacitance — they are widely used to model how signals travel in power transmission cables. We model the lightning channel as two concentric cylinders: a thin core that carries the current and a wider sheath that stores the associated charge. Solved this way, the equations self-consistently explain the current’s signature shape at ground level — a fast rise, set by how quickly the leader tips connect and thermalize, followed by a slower decay, governed by the channel’s electrical resistance. We also explain why the current wave travels at a fraction of light speed, why it weakens as it climbs, and why the current pulse disperses over distance — all derived from a handful of physical parameters rather than assumed curve shapes.

(a) Photograph of a lightning return stroke. (b-c) Schematic representation of the return stroke as a charge-neutralization wave. (d) Numerical discretization of the problem. Credit: da Silva et al. [2026], Figure 1

What are the benefits and limitations of this approach compared to other techniques?

Its biggest strength is speed paired with insight: the model runs orders of magnitude faster than full electromagnetic or gas-dynamic simulations, yet still yields exact analytical solutions in several limiting cases and explains why the empirical “engineering models” long used in industry take the mathematical forms they do. That transparency has made it a teaching tool in its own right — the model anchors how the return stroke is taught in the “Physics of Lightning” graduate course at New Mexico Tech, letting students derive lightning’s key features from first principles rather than take them on faith. Its main limitation is a simplifying assumption baked into the mathematics: it treats the electromagnetic fields as purely transverse to the channel, which breaks down for real, tortuous, branching channels. The model also simplifies the charge-storing corona sheath and requires care at the channel’s upper boundary to avoid artificial wave reflections.

What remaining questions or knowledge gaps need more research?

Two open questions stand out. The first is how to extend this framework — built around a single, straight channel — to capture the full complexity of a real lightning flash: its three-dimensional, branching geometry, and the sequence of multiple return strokes and other subprocesses, such as M-components and continuing currents, that typically follow the first stroke down the same channel.

The second is how to properly incorporate corona sheath dynamics into the Telegrapher’s Equations themselves. The present model treats the sheath’s charge as spreading out instantaneously over a fixed radius, but in reality, the sheath expands and charges on a finite timescale, and self-consistently coupling that behavior to the equations remains unresolved. Solving both problems — geometric realism and sheath physics — would sharpen predictions of lightning’s electromagnetic fields and its hazards to infrastructure, wildfires, and atmospheric chemistry.

—Caitano da Silva (caitano.dasilva@nmt.edu, 0000-0003-3728-3035), New Mexico Institute of Mining and Technology, United States; Logan Baeza, New Mexico Institute of Mining and Technology, United States; Jacob Wemhoner (0000-0002-8917-3009), New Mexico Institute of Mining and Technology, United States; and Saulo Orizaga (0009-0000-9635-8100), New Mexico Institute of Mining and Technology, 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: da Silva, C., L. Baeza, J. Wemhoner, and S. Orizaga (2026), Decoding the origins of lightning’s violent currents, Eos, 107, https://doi.org/10.1029/2026EO265031. Published on 19 August 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.

Source: AGU Advances

This is an authorized translation of an Eos article. 本文是Eos文章的授权翻译。

北极地区冻土覆盖的山坡上会呈现出一系列几何纹理，从圆形、条纹到多边形图案，不一而足。此外，还存在冻融泥流（solifluction）形态，即部分融化的永久冻土沿坡下滑时留下的痕迹。这些冻融泥流形态看起来像是平坦的梯田状土壤，如同一个巨大的阶梯，与梯田底部圆形的土壤块状物组合在一起。

随着气候变化加剧冻土融化的速度，了解这些图案的形成方式对于预测和修复北极地区不稳定的坡面至关重要。同时，这也有助于研究火星过去的气候状况，因为科学家已经在火星表面发现了类似的地形。然而，冻融泥流形态的形成机制一直难以解释。在一项新的研究中，Glade等人利用数学和物理模型以及遥感技术，解释了冻融泥流地形的形成机制。

冰冻土壤的移动速度非常缓慢，每年仅移动几毫米到几厘米，其行为也十分复杂，有时像流体，有时又像固体。这种复杂性源于水分和温度的季节性变化，以及土壤本身的固有物理规律。

研究人员排除了其他常见的流体类比物，例如墙上的油漆滴落、熔岩褶皱中的屈曲不稳定性以及滚动波。研究人员回顾了土壤学文献，运行了基于物理过程的计算机模型，对梯田与叶状结构的形成进行了模拟，并对不同流体行为进行了数学建模。

最终，他们找到了一个合适的类比物：在一种名为“Oobleck”的非牛顿流体中形成的波浪。Oobleck是一种由玉米淀粉和水混合而成的非牛顿流体，在不同应力下，其速度会发生变化，而且与直觉相反，施加的压力越大，它就越难被推动。

由于Oobleck具有独特的物理性质，它常被用于教学实验，且与研究人员在自然界观测到的冻土地貌特征高度吻合。土壤湿度的差异可能导致土壤流速不同，从而形成空间上不均匀的土壤堆积，最终在泥流过程再次开始之前坍塌。

研究人员指出，目前这一模型仍有其局限性。Oobleck反映的仅仅是流变特性，也就是所研究土壤（或流体）的物质构成。现实世界中的冻土，远比玉米淀粉和水的简单混合物要复杂得多。

除此之外，地形和植被等因素也会影响其形态，而不仅仅是物质组成。土壤堆积的前提是必须存在一个隆起结构，同时还需要足够的土壤湿度来积累冰层。

研究人员希望在实地验证他们的模型，但由于这些地貌特征的形成需要数百年甚至更长时间，直接观察其演变过程极为困难，但也并非不可能，他们决心尝试。(AGU Advances, https://doi.org/10.1029/2026AV002392, 2026)

—科学撰稿人Rebecca Dzombak

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: Earth Surface

Many exciting topics in geomorphology revolve around boundaries. The boundaries between hillslopes and river channels may seem obvious in an active landscape, but are far from trivial to detect objectively from digital elevation data. A pragmatic and routine solution is to use a fixed contributing catchment area to separate hillslopes from channels. The resulting geometry of the channel network then allows estimates of local steepness as a key metric of how rivers incise in response to rock uplift, and more generally, how landscapes respond to tectonic drivers. Yet, these estimates ultimately hinge on the choice of where channels begin.

Fox et al. [2026] explore how this choice matters: they showcase a numerical model of hillslope and channel evolution that predicts that the boundary between the two domains systematically shifts with varying rates of rock uplift. According to the model, more rapid uplift tends to lengthen hillslopes such that they can extend well beyond the arbitrary minimum catchment area used to characterize channels exclusively. The effect is that hillslope geometry contaminates estimates of channel steepness, and thus any inference about how river incision responds to changes in rock uplift. What is commonly reported as “channel steepness” as a metric of river form and adjustment might indeed carry an undesired contribution of hillslopes and their processes such as soil creep or debris flow. Clearly it is time to acknowledge a more flexible perspective of where channels begin, especially if using their geometry in models of landscape evolution.

Citation: Fox, M., Goren, L., & Adams, B. A. (2026). Non-linear hillslopes produce apparent non-linear river erosion models. Journal of Geophysical Research: Earth Surface, 131, e2025JF008753. https://doi.org/10.1029/2025JF008753   

—Oliver Korup, Associate Editor, JGR: Earth Surface

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: Journal of Advances in Modeling Earth Systems (JAMES)

Photovoltaic technology, most commonly seen as the bulky solar panels used on solar farms, is expected to become a dominant energy source by 2050. But these panels are often installed on land that might otherwise be used to grow crops for feeding a burgeoning population.

Agrivoltaics aims to solve this problem by planting crops around or underneath rows of solar panels, allowing for more efficient land use. In previous studies, solar panels were shown to help shade and protect certain crops as well as increase soil moisture, suggesting that carefully designed systems could support both agriculture and clean energy production.

Existing agrivoltaic research, however, tends to focus on one aspect of this process at a time—for example, light availability or crop growth—rather than addressing the nuanced interactions between microclimates, crop type, light, and panel type. Hosseini et al. share a new model that can simulate the microclimates beneath solar panels and even addresses the heat stress that workers might face in actual conditions.

The new model simulates the interactions between solar panels, crops, soil, air and water movement, and carbon dioxide uptake by tracking how energy, momentum, and mass move through the agrivoltaic system. The researchers used agrivoltaic site data, including leaf temperature measurements taken in Davis, Calif., and soil temperatures taken in Chicago City, Minn., to assess how the model’s efforts matched real-world conditions.

They then applied the model to a hypothetical agrivoltaic tomato farm using weather data from a hot, humid day in Princeton, N.J., a representative location for the densely populated mid-Atlantic region, where food and energy are both in high demand.

Compared to tomatoes grown in an open field, tomatoes grown under solar panels experienced leaf temperatures that were 1.84°C cooler during the day overall and up to 7.56°C cooler during peak afternoon heat, reducing water loss through evapotranspiration by 22.4%. Even though the simulated crops received 47% less sunlight, their carbon uptake declined by only 31%, suggesting that more temperate conditions lowered heat stress and partially offset the effects of increased shade.

The solar panels themselves were also 5.6°C cooler during the daytime than panels in bare soil, allowing them to recover about 15% of the efficiency that is lost during hotter temperatures. The average perceived temperatures for humans decreased by 4.46°C during working hours, implying important occupational health and safety benefits for farmworkers. The researchers suggest this model can be used to examine the benefits of agrivoltaic farms as well as other climate and crop combinations. (Journal of Advances in Modeling Earth Systems (JAMES), https://doi.org/10.1029/2025MS005588, 2026)

—Rebecca Owen (@beccapox.bsky.social), Science Writer

Citation: Owen, R. (2026), Solar panels can cool crops—and workers, Eos, 107, https://doi.org/10.1029/2026EO260241. Published on 19 August 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.

The external review of the Mount Maunganui Beachside Holiday Park landslide is extremely critical of the local authority in their approach to the management of the well-documented landslide hazard and risk at the site.

The external review commissioned by Tauranga City Council (TCC) into the 22 January 2026 Mount Maunganui Beachside Holiday Park landslide released its report today. It has been extensively covered in the New Zealand media. The conclusions are quite devastating for TCC, but are also likely to have wider implications for the management of risk from natural hazards across New Zealand.

The 22 January 2026 landslide at Mount Maunganui in New Zealand. Image from the TCC external review.

It is important to remember the terms of reference of this external review, which was undertaken by Hon. Paul Davison KSO KC:

“the external review examines the facts, timeline and decision-making processes leading up to the landslide, the adequacy of risk assessments and monitoring systems, and any lessons or improvements needed to strengthen future safety. The external review is separate from the Government Inquiry and the investigations currently being undertaken by the Coroner, Police and WorkSafe.”

So this review does not look at the initiation of the landslide – that will come in due course – it is really about the management of the risk. The conclusions are summarised in this section of the report (p. 204):-

What this Review does find is that this tragedy was, in the end, preventable. I do not mean by that that anyone could have known that this slope would fail at this moment, on this particular morning — no one could. I mean something more deeply troubling: that the hazard was known, that an effective and inexpensive means of managing the risk it posed had twice been recommended, and that the Council had every opportunity to put it in place. Had it done so, the most consequential decision of 22 January — whether to move people away from the foot of the slope — would not have been left to the unaided judgement of whoever happened to be on site that morning. It would have been made in advance, against defined criteria that the readily observable conditions of that day would plainly have met. What was required was not extraordinary foresight. It was the robust operational processes and discipline required of a well-run organisation: that a known risk to life be owned by someone, recorded, escalated, and followed through until an effective means of mitigating it was in place, and that those left in charge of the campground were equipped to recognise the danger, and ready to act quickly and effectively if ever required.

The report highlights that multiple studies showed that the risk to life at the campsite was unacceptably high – indeed, by some calculations the impact of this event was lower than had been feared. This is primarily because individuals at the site recognised that the risk was high and started to raise the alarm. If the Mount Maunganui Beachside Holiday Park landslide had occurred in the early hours of the morning the toll would probably have been higher.

There is little in this report with which I disagree, and the consequences for TCC are likely to be serious. I’m unsure as to the detail of the New Zealand judicial system, but in the UK this report would open the path to both civil and criminal court action, with the latter potentially occurring at both the institutional and individual level.

If there is one thing that worries me, it is that the the report does not fully recognise that early warning systems and evacuation plans are not a magic bullet. There is strong evidence from multiple settings that they can be dogged by uncertainty, equipment failures, false alarms and a lack of willingness from individuals to respond in the way that is planned.

But that is not an excuse for failing to have such a system in place.

Apart from the direct impact on TCC and its members, this report is likely to have profound implications for landslide risk management in New Zealand. Take this recommendation for example:

“For every populated site controlled by TCC, exposed to a natural-hazard risk to life assessed as Medium or higher, TCC should require — not merely consider — the development and maintenance of a Trigger Action Response Plan. The TARP should: be capable of immediate implementation in a simple initial form (for example, staged evacuation of defined runout zones when rainfall exceeds a defined return interval threshold over a defined duration), and be refined over time as a living document; link specific, observable or measurable triggers — defined rainfall thresholds, antecedent soil-moisture, and visible signs such as turbid water from the toe of a slope, tension cracking, bulging or changes in seepage — to specific, staged actions up to and including full evacuation; empower designated site personnel to act on a trigger immediately, without needing approval from someone higher in the TCC organisation; identify who holds each responsibility and the communication and escalation protocol; and recognise that evacuation/avoidance is a higher-order control than engineered consequence-reduction.”

Whilst this recommendation is TCC specific, it is inevitably going to have implications across all local authorities in New Zealand. The country inevitably has many, many sites in which “natural-hazard risk to life” is medium or above. This recommendation is going to require a very extensive revisit of risk management at those locations. In a country with a small population and a very large, hazard-prone landmass, that is going impose a major burden.

It is worth noting that TCC has “today accepted the findings and recommendations from the external review into the landslide at Mount Maunganui Beachside Holiday Park on 22 January 2026”.

The reverberations of the tragic Mount Maunganui Beachside Holiday Park landslideare going to continue for a long time.

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.

body {background-color: #D2D1D5;} Research & Developments is a blog for brief updates that provide context for the flurry of news regarding law and policy changes that impact science and scientists today.

The Arctic Report Card, an annual, peer-reviewed evaluation of the state of our warming Arctic, will still be published this year despite a lack of financial or logistical support from NOAA. 

On 10 August, editors of the Arctic Report Card were informed that NOAA will no longer support its publication, meaning the agency will not provide the report with coordinating editors, coordinate its external peer review, or fund the report’s web presence, publicity, graphics, or summary video. 

After a meeting of non-NOAA editors and chapter authors, however, the Arctic Report Card team has decided to move forward with publishing the 2026 report. Many of the report’s authors had been well into their work on their chapters when they learned that NOAA would no longer support that work.

“While many of the details are not final, we will be producing an ARC sans NOAA this year,” Rick Thoman, an Arctic scientist at the University of Alaska Fairbanks and an editor of the 2026 report, wrote in a message to Eos. Thoman said the team is not yet sharing which organization will host the 2026 Arctic Report Card, but that there has been a “tremendous outpouring of support for the Arctic Report Card, from individuals to large organizations.”

 
Related

•  Government Pulls Funding for the Arctic Report Card
•  Amid the Arctic’s Hottest Year, Arctic Science Faces a Data Deficiency
•  Get Involved: AGU Science Policy Action Center
 

He said the team plans to move forward with the original plan to present the report’s findings in a press conference at the 2026 AGU Annual Meeting in San Francisco, Calif. this December. Thoman also said the group of editors and authors is “pursuing multiple avenues” to produce the report’s graphics and summary video, and that they are planning to establish a non-NOAA website to serve as a long-term home for the Arctic Report Card in future years. 

“A lot of what we’re doing right now is, ‘How do we get this year’s report card out?’” he said. “But we are looking at the longer term, [also].”

The Arctic Monitoring and Assessment Programme (AMAP), part of the Arctic Council (an international Arctic policy and research collaboration), previously found reviewers and carried out the review process for the report. According to Thoman, AMAP has committed to providing that service for the 2026 report card. 

Science for Stakeholders

Having support from NOAA gave the report a heightened credibility for stakeholders, said Walter Meier, a sea ice scientist at the National Snow and Ice Data Center and a longtime author of the report’s sea ice chapter. “NOAA has had a strong reputation for providing high quality data, high quality observations, analysis, and research,” he said. “When you have that NOAA stamp of approval, people trust in that.” 

“I’m confident that the science will be of the same high quality.”

Still, he said not much would change about the scientific content of the report once it is hosted by a new organization. “I’m confident that the science will be of the same high quality,” he said. “I don’t feel there will be any loss [of quality], which is really the most important part.”

The report being outside of NOAA’s purview also “allows much more flexibility in language,” for the authors, Thoman said. He added that creating an Arctic Report Card without NOAA may allow the report to be more internationally focused than it has been previously.

—Grace van Deelen (@gvd.bsky.social), Staff Writer

These updates are made possible through information from the scientific community. Do you have a story about how changes in law or policy are affecting scientists or research? Send us a tip at eos@agu.org. 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.

body {background-color: #D2D1D5;} Research & Developments is a blog for brief updates that provide context for the flurry of news regarding law and policy changes that impact science and scientists today.

The U.S. Forest Service has moved forward with a rule to withdraw protections from 45 million acres of pristine national forests by rescinding the 2001 Roadless Area Conservation Rule. The U.S. Department of Agriculture (USDA) first announced its intent to rescind the Roadless Rule in June 2025 and it has now taken the next step to do so.

In today’s announcement, Secretary of Agriculture Brooke Rollins claimed that the Roadless Rule increases wildfire risk by preventing land managers from maintaining national forests.

“For too long, outdated restrictions have kept tens of millions of forested acres off-limits to the very treatments that improve forest health and reduce wildfire risk to our communities,” Rollins said in a statement. “Today, we filed a proposal to restore authority to local forest managers who know the land best, removing the barriers that have kept them from doing the work the land demands. It’s time to turn the page on the failed roadless rule and return our forests to health and productivity.”

However, this justification is not backed up by data. Research has shown that nearly 85% of wildland fires begin with human activity, either accidentally or deliberately. What’s more, human-caused fires are much more destructive than naturally-caused fires from, for example, lighting strikes. Building roads would bring more people into former wildlands, thereby increasing wildfire risk despite the increased access for maintenance, several forest experts told the New York Times.

The Trump Administration “is trying to walk away from the most successful land conservation policy in modern U.S. history.”

“The truth is that roadless forests are our greatest defense against wildfire,” David Jenkins, president of the nonprofit Conservatives for Responsible Stewardship, said in a media statement. “Not only are old growth forests naturally resistant to fire, but 90% of all wildfires in the U.S. occur within a half mile of a road. Any claims to the contrary are demonstrably false.”

Most of the inventoried roadless areas are in western states, with more than 95% located in Alaska, Arizona, California, Montana, Nevada, New Mexico, Oregon, Utah, Washington, and Wyoming. However, the proposed rule does not apply to Colorado and Idaho, which have state-specific roadless regulations.

Some suspect that the administration’s true intent is not to reduce wildfire risk but to drastically increase logging production.

“The Trump administration is moving to gut our most iconic national forests by giving loggers and drillers free rein to turn them into industrial wastelands. It is trying to walk away from the most successful land conservation policy in modern U.S. history,” Andrew Wetzler, senior vice president for nature at the nonprofit Natural Resources Defense Council, said in a statement.

Jenkins added, “This is one of the most boneheaded and fiscally irresponsible actions this administration could take. The agency doesn’t have the budget to take care of the 368,000 miles of roads already riddling our national forests. In fact, there is currently a $10.8 billion maintenance backlog.”

 
Related

•  Trump Administration Moves to Allow Logging in Pristine National Forests
•  Read the Forest Service’s Proposal
•  Context from Eos: USDA Moves to Rescind Roadless Rule Protecting 45 Million Acres of Wild Area
•  Get Involved: AGU Science Policy Action Center
 

The Roadless Rule was enacted by the Clinton administration in 2001 to prevent logging and development on millions of acres of pristine forest across the country. For 25 years, this rule has protected crucial habitats for migratory species, headwaters for municipal water supplies, and old-growth forests which act as significant carbon sinks. If the rule was rescinded, those forests and the ecosystem services they provide would be at risk.

“Once we cut a road through these forests, we don’t get their values back,” Wetzler said. “The forest fragments. The streams and rivers are degraded by runoff. Iconic wildlife struggles to find suitable habitat. And we get stuck with even more roads in a vast network the agency already can’t maintain.”

The Roadless Rule itself is hugely popular, garnering around 75% public support. Nearly 626,000 public comments were submitted on the USDA’s 2025 draft proposal to eliminate the rule, with more than 99% of comments calling for the rule to stand. Former U.S. Forest Service chiefs, lawmakers, Tribal leaders, scientists, and wildland firefighters have also defended the rule.

The Forest Service’s proposal will be published in the Federal Register on 20 August, after which it will accept public comments before finalizing the rule change.

—Kimberly M. S. Cartier (@astrokimcartier.bsky.social), Staff Writer

These updates are made possible through information from the scientific community. Do you have a story about how changes in law or policy are affecting scientists or research? Send us a tip at eos@agu.org. 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: AGU Advances

Droughts are becoming an increasing global concern as their impacts intensify. Understanding how climate change affects drought frequency and severity is essential for projecting future drought risk. In this context, the question raised by Helpap et al. [2026]—”are modern droughts unprecedented?”—has a high scientific relevance. Here, 20 ensemble members from a 600-year climate simulation and reanalysis data from the period 1950-2024 are analyzed to place recent droughts in a multi-century context. It is concluded that drought extremes are intensifying more rapidly beyond the range experienced in the past than moderate droughts. Therefore, an increasing likelihood of droughts reaching unprecedented magnitudes is expected in the near future, thus highlighting the urgency for adaptation measures in water resources and environmental management.

Citation: Helpap, P., Brönnimann, S., Hand, R., Franke, J., Raible, C. C., & Stocker, B. D. (2026). Are modern droughts unprecedented? AGU Advances, 7, e2026AV002536. https://doi.org/10.1029/2026AV002536

—Alberto Montanari, Editor-in-Chief, AGU Advances

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 Landslide Blog is written by Dave Petley, who is widely recognized as a world leader in the study and management of landslides.

On social media, the Department of Mineral and Petroleum Resources, Republic of South Africa has released two aerial images of the aftermath of the 13 August 2026 tailings storage facility failure at the Semancor Dikwena Chrome Mine in South Africa.

The first shows the tailings storage facility (TSF) itself, including the loss of slimes:-

The aftermath of the TSF failure at the Samancor Dikwena Chrome Mine tailings storage facility near Brits. Image released by the Department of Mineral and Petroleum Resources on social media.

It is interesting to note the small-scale slope failures on the tailings wall in the foreground (I’d be interested to know if this was expected as part of the design), but also the fact that this TSF has been raised above those on either side.

The second shows the breach itself:-

The aftermath of the TSF failure at the Samancor Dikwena Chrome Mine tailings storage facility near Brits. Image released by the Department of Mineral and Petroleum Resources on social media.

I’m no expert on TSF construction, but this one has a complex history. There is a very detailed archive of images on Google Earth. So for example, this was the facility in August 2015:-

Google Earth image of the TSF that failed at the Samancor Dikwena tailings storage facility near Brits. Image from 2015.

Note that this was a single facility at that stage.

This is the same site in June 2022:-

Google Earth image of the TSF that failed at the Samancor Dikwena tailings storage facility near Brits. Image from 2022.

The eastern end of the facility (which failed) had become vegetated with little sign of human activity. There was some work going on at the western end.

If we jump to July 2025, we see the facility had been raised and subdivided, and that active deposition was occurring in both sections. The one that failed (known as 1b) was notably higher than the one to the west.

Google Earth image of the TSF that failed at the Samancor Dikwena tailings storage facility near Brits. Image from 2025.

It is good to see that Semancor has taken full responsibility for “the repair of any infrastructure and the rehabilitation of the affected area“, noting of course that the global mining industry does not have a good track record of actually delivering against such commitments.

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.

This story was originally published by Grist. Sign up for Grist’s weekly newsletter here.

Matthew LaPlante doesn’t have to go to a lab to see his research in action. On some mornings, he can just look out his own window in the mountains high above Salt Lake City. After a warm night when the temperature stays above freezing, he’ll wake up and see that the snow level has dropped by inches, revealing more of the aspens he taps for syrup.

“It feels like a monster came and just in the middle of the night, took a bite out of a snowpack.”

“It feels like a monster came and just in the middle of the night, took a bite out of a snowpack,” said LaPlante, a journalist and climate scientist at Utah State University.

So it’s fitting that scientists have started calling these kinds of heat waves, marked by unusually high temperatures in the spring and early summer, “snow eaters.”

LaPlante was part of a recent study, published in the journal Science Advances, that attempted, for the first time, to identify what conditions exactly make for a “snow eater.” Compared to normal warm spells or heat waves, the researchers determined that these events happen when temperatures stay above freezing through both day and night for multiple days, typically three to five. These events can roughly double the rate at which snow melts, causing flooding and making it challenging to manage water resources.

“Snow eaters” appear to be occurring earlier in the year and becoming more widespread in the Western United States as the climate warms. Since the 1850s, the study found, the area affected by snow eaters has increased by an average of about 40,000 square miles per century, and the first snow eater of the season has been arriving about one month earlier per century.

The term “snow eater” has a murky history. By at least the 1880s, people in the West were talking about “snow-eating” chinooks, warm mountain winds that make snow disappear quickly. More recently, the phrase “snow-eater heat wave” first made headlines in March, when an early heat wave enveloped much of the West, quickly wiping away snowpack in the Colorado Rockies and California’s Sierra Nevada. The scientists hope the catchy, evocative term can help draw more attention to this type of heat wave, since there’s still a lot to learn. The study only looked at the Western U.S., but snow eater heat waves almost assuredly occur elsewhere, LaPlante said.

Solar radiation is a major, but sometimes overlooked, driver of snowmelt, said Noah Molotch, a professor of geography at the University of Colorado Boulder who was not involved in the new study. When it interacts with heat waves, those impacts are amplified. As snow crystals warm up, they lose some of their structure and light-reflecting abilities, causing snowpacks to absorb more sunlight and melt faster. “It’s a little bit of a—no pun intended—a snowball effect,” he said.

Much of the Western U.S. saw record-low snowpack this spring. What was really unusual about it, Molotch said, was how widespread it was. Colorado received less precipitation than normal this winter, while California got lots of precipitation, but in the form of rain instead of snow. But across the region, “the one thing in common was above-average air temperatures,” Molotch said.

These conditions have likely helped fuel exceptionally severe wildfires in the West, from Utah to Spokane, Washington, where hundreds of homes burned earlier this month. “Drought stress for mountain forests around the Western U.S. is heavily dictated by the snow that accumulates each winter and then melts through the spring and summer,” Molotch said. “There is a direct connection there in terms of the water availability and drought stress that can provide one of the important ingredients for increases in wildfire intensity and frequency.”

“At night, it used to be quiet. And now it’s not quiet, because everything’s melting all the time.”

If scientists are able to better predict what will happen to snowpack, it could help water managers plan for what’s coming. Early or rapid snowmelt poses problems for managing water resources in the West, where snowpack serves as a key source of fresh water in the drier summer months. “Water that would otherwise be stored as snow comes out early, and then we have to deal with it as a hazard instead of a resource at our reservoirs and along the rivers and streams,” said Ben Hatchett, a co-author on the study and a scientist at Colorado State University’s Cooperative Institute for Research in the Atmosphere.

Snow-eater heat waves also may pose risks to skiers, hikers, and anyone else on or near mountains. They could be linked to hazards such as avalanches, glacial collapses, and permafrost melt, Hatchett said, though scientists are still investigating those connections. He lives in the Sierra Nevada, and he’s noticed one tangible change over his lifetime: The soundscape has changed.

“At night, it used to be quiet,” he said. “And now it’s not quiet, because everything’s melting all the time.”

—Katie Yoder (@katemyoder), Grist

This article originally appeared in Grist at https://grist.org/science/new-heat-wave-category-snow-eaters-study/.

Grist is a nonprofit, independent media organization dedicated to reporting on climate change.

A major tailings accident occurred last week in a Chrome mine at Dikwena near Brits.

On 13 August 20926, a major failure occurred in a large tailings storage facility at the Dikwena facility, owned by Samancor Chrome in the Brits area of South Africa. There are news reports from IOL and the Daily Maverick about this major accident.

The Daily Maverick article includes a video of the aftermath of the TSF failure, collected by Iaan Myburgh, showing the aftermath of the failure. This is a still from that video:-

The aftermath of the TSF failures at the Samancor Dikwena tailings storage facility near Brits. Still from a video collected by Iaan Myburgh.

The failure appears to have originated in a failure of the dam wall of one of the TSF’s at the site:-

The dam wall collapse responsible for the TSF failure at the Samancor Dikwena tailings storage facility near Brits. Still from a video collected by Iaan Myburgh.

The location of the failure appears to have been 25.65236° S, 27.85177° E [-25.65236, 27.85177]. This is a Google Earth image of the site, with the location of the breach marked, collected on 15 July 2026:-

Google Earth image highlighting the site of the dam wall collapse responsible for the TSF failure at the Samancor Dikwena tailings storage facility near Brits

It is interesting to note that this TSF does not appear to have been active in July.

On LinkedIn, Adam Thomas has posted a Sentinel-2 image of the aftermath of the failure. Peter McGough has also posted a range of images and videos.

Finally, the Daily Maverick article has an interesting interview with a tailings expert, Alastair Bovim, about this site:-

“What we believe occurred is that the original TSF was re-mined between 2019 and 2022 and then in 2022 a new facility was designed, which then started operation in 2024,” said Bovim.

He said that from 2022 to 2024, historical satellite imagery shows new dams and walls being constructed.

This event asks many questions. I wonder how many answers we’ll get.

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.

Want updates from a trusted source about how climate change is altering pollution pathways? Preference Eos in your searches! Go to Google

Antarctica is 600 miles (about 1,000 kilometers) from any major city, almost entirely uninhabited, and surrounded by a cold and inhospitable ocean. It might seem like the last place to expect anthropogenic pollution. But the continent’s freezing temperatures trap airborne and marine pollutants, a situation that has historically made it a sink for contaminants such as mercury, a heavy metal that can harm human and wildlife health.

“Global mercury pollution and climate change are not independent environmental problems.”

According to a new study published in the Proceedings of the National Academy of Sciences of the United States of America, mercury is accumulating in Antarctica much faster thanks to increasing anthropogenic mercury emissions.

Additionally, the study found the continent’s ice cap is releasing mercury faster than ever before as Antarctic ice melts in a warming climate and frees up long-stored mercury that then flows into the surrounding ocean. Together, these processes are turning Antarctica into an active source of mercury pollution rather than a long-term storage area, the authors write.

“Climate change acts as an amplifier of mercury cycling,” Maodian Liu, an environmental scientist at Peking University in China and coauthor of the new study, wrote in an email. “Global mercury pollution and climate change are not independent environmental problems.”

Dual Pathways

Previous research has shown that mercury reaches Antarctica through the atmosphere, traveling from sources such as coal plants and mining operations to settle on Antarctic ice and the Southern Ocean. And studies have suggested that mercury accumulation has increased in some regions of Antarctica over the past 2 centuries.

But there has been little evidence to tease apart two mechanisms of mercury accumulation: atmospheric deposition from human sources and the remobilization of trapped, frozen mercury as Antarctic glaciers melt.

To further investigate, Liu and the research team analyzed 16 sediment cores collected from the continental shelf off the Antarctic Peninsula. They used a variety of chemical analyses to determine the amount and sources of mercury that was deposited in the sediment over the past 200 years.

Nearly all the sediment cores showed a significant increase in mercury accumulation since industrialization (the mid-18th century). On average, cores in the region showed a 160% increase in accumulation since industrialization, with rates accelerating particularly quickly beginning in the 1950s.

“Climate change is not only increasing the amount of glacial meltwater but is also releasing legacy mercury that was previously stored on the Antarctic Peninsula.”

Further simulations of Earth’s mercury cycle and budget showed that glacier melt has greatly increased the amount of mercury that Antarctic glaciers are releasing into the ocean: The analysis showed a 550% increase in such releases of mercury since industrialization. The research team estimated that about 56% of the mercury detected in their Antarctic Peninsula seafloor cores originated from the atmosphere; the rest likely originated from glacier melt, weathering, and erosion of Antarctic ice and sediment.

“Climate change is not only increasing the amount of glacial meltwater but is also releasing legacy mercury that was previously stored on the Antarctic Peninsula,” Liu wrote.

The fact that the study indicates changes in the transport and processing of mercury in such a remote area makes the research noteworthy, said Charles Driscoll, an environmental engineer at Syracuse University who was not involved in the new research. Similar processes are happening elsewhere in the world: In the Arctic, permafrost thaw has begun to release trapped mercury into ecosystems, and in the northeastern United States, increasingly intense storms are accelerating erosion and mobilizing nutrients and metals, including mercury, Driscoll said.

The new study adds context to other indications of a changing global mercury cycle and shows a consistent global pattern, Driscoll said.

Methylmercury Matters

Once mercury accumulates in sediment, it can transform into methylmercury, a more toxic form that more readily accumulates in living organisms. Methylmercury enters and travels through the Antarctic food web, taken up by plankton, then krill, fish, seabirds, and marine mammals.

At sufficiently high doses, methylmercury can harm animals’ neurology, behavior, growth, and reproduction, Liu wrote. And because some Antarctic krill and fish make their way into commercial fisheries, mercury pollution has the potential to affect human health, too.

Driscoll said he’d like to see more research to answer the question of how much mercury in the cold waters of the Antarctic shelf could be transformed into methylmercury; the process typically happens faster in warmer temperatures.

Liu hopes future research will determine whether the trends he and the research team noticed around the Antarctic Peninsula are representative of Antarctica as a whole.

Even if new mercury emissions from coal combustion, mining, and other human activities decrease, mercury stored in Antarctic ice will continue to be released as ice melts, meaning “environmental mercury burdens are unlikely to decline immediately in step with emission reductions,” Liu wrote.

—Grace van Deelen (@gvd.bsky.social), Staff Writer

Citation: van Deelen, G. (2026), Antarctica is releasing mercury faster than ever because of climate change, Eos, 107, https://doi.org/10.1029/2026EO260262. Published on 14 August 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.

Editors’ Highlights are summaries of recent papers by AGU’s journal editors. Source: Community Science

As flooding becomes more frequent, nature-based solutions (NBS), such as restoring wetlands or improving soil health, can help reduce flood impacts while supporting healthy landscapes. Financial incentive programs are available to help landowners pay for these practices, yet participation remains lower than expected.

Moore et al. [2026] explore landowner perspectives of NBS to better understand potential challenges to adoption (such as policy, financial, and administrative factors), building on previous work around a FloodWise program developed by North Carolina State University and partners. Focus groups with landowners in North Carolina’s inner plain provided insights into their experiences with conservation financial assistance programs. Participants described confusing application processes, unclear program requirements, long wait times, and inconsistent technical support as major barriers.

At the same time, landowners expressed strong conservation values and offered practical ideas for improving participation, including simpler program guidance, better technical assistance, and more opportunities to learn from other landowners. By reducing these administrative hurdles, conservation programs can become more accessible and help more communities adopt nature-based approaches that strengthen resilience to flooding.

Citation: Moore, S. J., Hovis, M., Bardon, R., Cubbage, F., & Lupek, M. (2026). “Welcome to the club”: Unpacking administrative burdens in conservation incentive programs supporting natural flood resilience. Community Science, 5, e2026CSJ000191. https://doi.org/10.1029/2026CSJ000191

—Kathryn Semmens, Deputy Editor, Community Science Exchange

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: Journal of Geophysical Research: Biogeosciences

Salt marshes are coastal wetlands that are alternately inundated and exposed by tides. These marshes, like other coastal ecosystems, are hot spots of carbon cycling.

But salt marshes sit at the complex intersection of land and marine ecosystems, and there is still uncertainty over how they cycle, store, and release carbon. One major uncertainty is how much carbon is transported by water rather than stored in marsh soils. Once transported, this carbon may be buried, used by organisms and released into the atmosphere, or exported to the ocean. But this lateral transport between salt marshes and coastal waters has largely been overlooked in research on salt marshes’ carbon storage potential.

He et al. dove into this understudied process by focusing on tidal creeks, which run high or dry out based on the state of the tides. These creeks can move water—and its dissolved components—around the landscape or out to sea, so quantifying their carbon movement is an important part of quantifying salt marshes’ carbon budgets.

The researchers installed sensors to collect data on water flow, water chemistry, and environmental parameters in a salt marsh creek in coastal Louisiana. They also collected water samples to track dissolved inorganic and organic carbon and total alkalinity.

These observations continued for 3 years, building a rare and continuous water chemistry dataset that captures daily, seasonal, and annual variability. In contrast, most studies of salt marsh carbon transport use short-term monitoring, an approach that can miss hydrological and biogeochemical changes that occur with tides, seasons, and storms.

After taking water flow into account, the team calculated how much carbon was stored or exported. They found that more dissolved organic and inorganic carbon was exported from the marsh than was stored. That net flux was driven primarily by different concentrations of carbon in falling versus rising tides. This pattern held at daily, monthly, and seasonal timescales. Lateral transport, then, can act as a dynamic control on a salt marsh’s carbon sequestration, boosting or damping it over time.

Total alkalinity also varied with the tides, suggesting that salt marshes can affect the acidity of downstream waters.

The findings suggest that the capacity of tidal streams to transport dissolved carbon out of salt marshes could be larger than that of sediment burial, making lateral carbon transport in tidal creeks a major overlooked component of coastal carbon models and blue carbon assessments.

Quantifying lateral carbon transport in more salt marshes will improve those assessments’ accuracy and deepen our understanding of the role salt marshes play in the global carbon budget, the authors say. (Journal of Geophysical Research: Biogeosciences, https://doi.org/10.1029/2026JG009760, 2026)

—Rebecca Dzombak, Science Writer

Citation: Dzombak, R. (2026), The under-the-radar creeks that are sweeping carbon out of salt marshes, Eos, 107, https://doi.org/10.1029/2026EO260261. Published on 11 August 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.

On 4 July 2026, 19 people were killed in a massive landslide in Zamora Chinchipe province, Ecuador. Planet Labs imagery shows that this was caused by a c.1 km long landslide that occurred 13 km upstream of Cantzama. This transitioned into a channelised debris flow that destroyed the community.

On 4 July 2026, a massive landslide swept through Cantzama in Zamora Chinchipe province, Ecuador. News reports indicate that 19 people were killed and around 50 houses were destroyed or damaged. As is often the case with disasters of this type in places like Ecuador, this event garnered little press coverage in the west, but it is quite significant event.

There is some Youtube drone footage of the aftermath of the disaster:-

Fundacion Pachamama has quite a detailed analysis of the event, compile using radar imagery, which describes a large, long run out debris flow / mudflow event. They were unable to pinpoint a cause of the flow, but I think Planet Labs optical imagery allows us to diagnose what happened.

The location of the most seriously damaged parts of Cantzama is 3.85480° S, 78.89506° W [-3.85480, -78.89506]. This is a Google Earth view of that site, with the village in the foreground and the tributary valley down which the landslide flowed in the background:-

Google Earth image of Cantzama in Ecuador.

This is a Planet Labs image, dated 5 July 2026, showing the aftermath of the landslide. The path of the flow is very clear, as is the damage to a range of buildings around the channel:-

Planet Labs image of Cantzama in Ecuador, overlain into Google Earth, showing the aftermath of the landslide. Image copyright Planet Labs, used with permission, collected on 5 July 2026.

If we go upstream from Cantzama we reach this point:-

Planet Labs image of the area upstream of Cantzama in Ecuador, overlain into Google Earth, showing the landslide that triggered the catastrophic debris flow. Image copyright Planet Labs, used with permission, collected on 5 July 2026.

The location of the crown of this major landslide is 3.88474° S, 79.00312° W [-3.88474, -79.00312]. The initial failure of the landslide is about 1 km long and 300 m wide. This major failure has transitioned into a channelised debris flow – the linear distance to the Cantzama is about 13.5 km – this is an extremely long runout event. It appears that the landslide has entrained a significant amount of sediment from the channel, leading to an extremely damaging flow.

As such this is an unusual landslide. There is another interesting and intriguing aspect. The image below, collected on 04 December 2025, shows the area around the main failure of the landslide. The marker shows the location of the crown of the 4 July 2026 landslide:-

Planet Labs image of the site of the Cantzama landslide in Ecuador, overlain into Google Earth, showing the multiple shallow failurses at the site of the landslide that triggered the catastrophic debris flow. Image copyright Planet Labs, used with permission, collected on 5 July 2026.

There is a host of shallow failures across the hillslopes, which appear to have been triggered between 22 June 2025 and 08 August 2025. Presumably this was a response to a heavy rainfall event. Note that the site of the 4 July 2026 landslide had suffered a substantial failure in this earlier rainfall event.

Is it possible that this 2025 rainfall event destabilised the slope, creating a situation in which it then destabilised in the next major rainstorm? It would be interesting to look at the InSAR data for this slope.

As for Cantzama, there is a proposal to relocate the population located close to the channels, redesignating the site for tourism. That is probably wise.

Acknowledgement

Many thanks to Planet Labs for making the imagery available to me.

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.