Feed aggregator

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

Leaf area index, which measures the ratio of leaf area to ground area, is commonly used by researchers to track foliage coverage and ecosystem health. One consequence of global warming, observed by satellite remote sensing, has been a long-term increase in leaf area index. This Earth “greening” is predicated on increased terrestrial carbon uptake under increasing CO2.

Chang et al. [2026] examine water controls on vegetation greening by isolating structural (e.g. leaf area index, LAI) and physiological (e.g., water use efficiency, WUE) impacts under increasing CO2. Their analysis shows worldwide patterns of increased plant water stress (PWS) and ecological drought, attributed to soil moisture, atmospheric aridity, and regional climate. This study demonstrates water limits on Earth greening emerging from local interdependencies between the carbon and water cycles. 

Attribution of variables, including meteorological factors (air temperature, Ta; and net radiation, Rn), soil and atmospheric aridity (VPD; surface soil moisture, SMsurf; and root-zone soil moisture, SMroot), and atmospheric CO2 to the changes of PWS indicated by ESILAI over the past four decades with a Random Forest model. ESILAI is the evaporative stress index that accounts for dynamical changes in Leaf Arae Index (LAI). a. Spatial distribution of dominant factor; b. Area ratio for the distribution of each dominant factor in a. Credit: Chang et al. [2026], Figure 7

Citation: Chang, Q., Wang, L., Barnes, M. L., Ficklin, D. L., Benson, M. C., & Novick, K. A. (2026). Widespread increase in global plant water stress obscured by greening. AGU Advances, 7, e2025AV002243. https://doi.org/10.1029/2025AV002243

—Ana P. Barros, Editor, 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.

Some 550 million years ago, Earth looked very different than it does today. The Southern Hemisphere was dominated by an enormous single continent known as Gondwana.

A University of the Basque Country-EHU research group has determined the causes of a 255 m (837 feet) crustal uplift of the Atlantic slope and a 167 m (548 feet) uplift of the Mediterranean slope, as well as the development of rivers in these areas.

Peatlands store large amounts of carbon. However, when they are drained, they can become substantial sources of the greenhouse gas carbon dioxide (CO₂).

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

When studying the subsurface of the Earth, scientists commonly use geophysical inverse modeling, which aims to infer key physical properties, like geological structures, from indirect observations. But these modeling techniques are typically high-dimensional and computationally demanding. To overcome these limitations, the scientific community is embracing Bayesian inference methods, which could offer a solution by integrating prior geological knowledge with observed data, enabling systematic uncertainty quantification.

A new article in Reviews of Geophysics explores recent advances in Bayesian methods for modeling the Earth’s subsurface from indirect geophysical data. Here, we asked the authors to give an overview of geophysical inverse modeling, how scientists use Bayesian inference, and what challenges remain.

In simple terms, what is geophysical inverse modeling?

Many decisions depend on what lies beneath our feet: where to drill, whether carbon dioxide can be stored safely, how groundwater may move, or where a hidden fault may create risk. Yet most of the subsurface cannot be observed directly. Geophysical surveys provide clues by measuring seismic waves, electromagnetic fields, gravity, and other responses of the Earth. Inverse modeling works backward from those measurements. Scientists propose an underground model, use physical laws to predict the signals it would produce, compare those predictions with the observations, and revise the model. The result is not a photograph, but a scientifically constrained description of underground structures and properties.

Conceptual overview of the multidisciplinary landscape of geophysical inverse modeling in subsurface applications. Credit: Liu et al. [2026], Figure 1

What are the benefits and limitations of using geophysical inverse modeling?

Geophysical inverse modeling can investigate large and deep regions without extensive drilling or excavation. It helps estimate rock velocity, electrical resistivity, density, porosity, fluid content, and the shape of faults, aquifers, or reservoirs. Its central limitation is that the data are noisy, incomplete, and unevenly informative. Different underground models can therefore produce very similar measurements. A single “best” model may look more certain than the evidence allows, while important alternatives remain hidden. For practical decisions, the key question is not only whether an image looks sharp. It is whether the remaining ambiguity could change a prediction or action, such as a drilling location, a monitoring plan, or a safety assessment.

What is Bayesian inference?

Bayesian inference is a way to update what we believe when new evidence arrives. A “prior” represents geological knowledge and plausible underground scenarios before the latest measurements are used. A “likelihood” describes how well each scenario explains the observed data, while accounting for measurement noise and modeling errors. Combining the two gives a “posterior,” a probability-weighted collection of scenarios that remain plausible after the evidence is considered. Rather than producing only one answer, Bayesian inversion shows which features are strongly supported, which remain uncertain, and how alternative interpretations compare. Those scenarios can then be carried forward into predictions and decisions, connecting data, models, possible outcomes, and action.

How has Bayesian inference improved subsurface modeling?

Bayesian inference has shifted the goal from finding one preferred underground model to evaluating a range of plausible models and predictions. This makes uncertainty an explicit result rather than an afterthought. Scientists can identify poorly constrained regions, reveal trade-offs between properties, combine different data types, and estimate the risk of costly misinterpretation. Bayesian analysis can also help determine which additional measurement would be most valuable. Recent gradient-informed methods use derivatives to navigate large model spaces more efficiently when the simulator is differentiable. Deep learning can represent realistic geological patterns, accelerate expensive physical simulations, and support rapid repeated inference. Together, these advances make uncertainty more useful for monitoring, planning, and decision-making.

Overview of Bayesian inference approaches for geophysical inverse problems. RW-MCMC enables general-purpose uncertainty quantification without requiring gradients. Ensemble-based methods offer efficiency in high-dimensional, weak non-linear, and non-Gaussian settings. Gradient-informed techniques leverage posterior geometry for improved sampling when gradients are available. Differentiable Bayesian frameworks combine physics-based modeling, machine learning, and automatic differentiation. Credit: Liu et al. [2026], Figure 2

What is Differentiable Bayesian Inversion, and what are its benefits?

Differentiable Bayesian Inversion, or DBI, is not a single new sampling algorithm. It is a unifying framework for linking geological knowledge, physical or learned simulators, models of measurement uncertainty, and Bayesian inference in one computational workflow. “Differentiable” means that software can track how a small change in an underground model affects the predicted data and how plausible that model is. Automatic differentiation can calculate these sensitivities through the full workflow. This common computational language allows physics-based models and machine-learning components to work together rather than remain separate boxes. Because the framework is modular, individual components can be improved or replaced. DBI could support faster, geologically realistic, uncertainty-aware inversion for monitoring and operational decisions.

What challenges remain?

The largest challenge is scale. Real three-dimensional problems may contain millions of unknowns, and every candidate model can require an expensive seismic, electromagnetic, or flow simulation. Plausible answers may also form several disconnected geological scenarios that are difficult to explore. AI models used to represent geology and fast surrogate simulators introduce further risks: they may fail when field conditions differ from their training data, and their errors can distort uncertainty estimates. Progress will require efficient differentiable solvers, more robust samplers, better uncertainty calibration, and validation on realistic field data. Gradient-free and hybrid methods will remain essential when models are not differentiable. The aim is not to eliminate uncertainty, but to identify which uncertainties matter and use them wisely.

—Mingliang Liu (mingliangliu@sdu.edu.cn; 0000-0002-5783-4490), Shandong University, China, and Stanford University, United States; Dario Grana (0000-0003-4220-053X), University of Wyoming, United States; Klaus Mosegaard (0000-0001-5292-5249), University of Copenhagen, Denmark; Mrinal K. Sen (0000-0002-5525-0467), The University of Texas at Austin, United States; Minghui Xu (0000-0001-9567-5569), Stanford University, United States; and Tapan Mukerji (0000-0003-1711-1850), Stanford University, United States

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

Citation: Liu, M., D. Grana, K. Mosegaard, M. K. Sen, M. Xu, and T. Mukerji (2026), How can we make better decisions about what lies underground?, Eos, 107, https://doi.org/10.1029/2026EO265036. Published on 10 September 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.

In early 2017 and 2023, California experienced two catastrophic and record-breaking floods that caused billions of dollars in damage, triggered mudslides and forced thousands of people to evacuate. Why this happened left scientists scratching their heads. Both disasters struck during La Niña conditions, which normally bring drier-than-normal weather to California.

Extreme heat has become more common—both research and lived experience document this—but when it arrives is a different question. A new study finds that, rather than occurring only during warm seasons, extreme heat events have become more frequent in the months before and after, too.

This is the highest-resolution image of the Sun ever taken.

It was captured by the Daniel K. Inouye Solar Telescope (DKIST) in Hawaii, the largest solar telescope in the world, on 14 April 2025.

The image zooms in on a small patch of the Sun’s surface in visible light. The whole image spans 5,800 kilometers (3,604 miles) horizontally and 4,350 kilometers (2,703 miles) vertically, covering a mere 0.0004% of the Sun’s surface. Each pixel is 6 kilometers (3.7 miles) per side.

Take a close look at the grid-like pattern and see if you can spot small swirls among the feathered edges.

A high-resolution image of the Sun at 416 nanometers taken by the Inouye Solar Telescope. Credit: NSF/NSO/AURA/MPS, CC BY 4.0

Here are two spots with swirling plasma.

The inset here zooms in on a small patch of the Sun’s surface with the telltale swirl of Kelvin-Helmholtz instabilities (KHI) just tens of kilometers across. Credit: NSF/NSO/AURA/MPS, CC BY 4.0 Another region of the same image shows even more KHI fingerprints, suggesting that these instabilities might be common on the Sun’s surface. Credit: NSF/NSO/AURA/MPS, CC BY 4.0

The tiny swirls and vortices are called Kelvin-Helmholtz instabilities, and this is the first observational confirmation that they exist on the Sun. The swirls are manifestations of a physical process that could play a crucial role in how the Sun’s plasma bubbles and bursts.

“Kelvin-Helmholtz instabilities are one of the fundamental instability processes in magnetofluids, fluids, and gases,” said David Kuridze. Kuridze studies plasma in the solar atmosphere at the National Solar Observatory (NSO) in Boulder, Colo., and is colead researcher on the discovery.

“We did not set out to find Kelvin-Helmholtz instabilities,” added colead researcher Friedrich Wöger, who studies flow dynamics of the solar photosphere at NSO. “The experiment was targeted at finding the most efficient way to reach the diffraction limit of the telescope. And the nice thing is, not only did we succeed with that but we also found Kelvin-Helmholtz instabilities.”

Kelvin-Helmholtz instabilities might be everywhere on the Sun. Credit: NSF/NSO/AURA/MPS, CC BY 4.0 When Plasmas Flow By

Kelvin-Helmholtz instabilities (KHI) are caused by the edges of two streams of fluid flowing past each other at different speeds. And “fluid” is quite a broad term, geophysically speaking. Scientists have spotted these swirls and vortices in estuaries and oceans, as well as in Earth’s clouds; the skies of Mars, Jupiter, and Saturn; and interactions between the solar wind and planetary magnetospheres, Kuridze explained.

Astronomers have long suspected that KHI exist on the Sun’s surface. The solar photosphere is a fluidlike plasma, so it would make sense for it to follow the same rules as any other fluid, albeit with the added complication of the Sun’s magnetic field. What’s more, the existence of KHI could explain some of the Sun’s more mysterious phenomena, like braided magnetic field lines, eruptions, and the ultrahot corona.

“Kelvin-Helmholtz instabilities are a very effective mechanism to twist and bend magnetic structures” and generate magnetic energy, Kuridze said.

“One of the big questions is, How much energy can these vortices create and transport up into the outer atmosphere of the Sun, and is it enough to heat it up to millions of degrees kelvin?” Wöger said.

But until recently, actually spotting these instabilities on the Sun’s surface was impossible. Picking out small vortices within the solar plasma requires specialized instruments installed on solar telescopes, and previous generations of solar telescopes were simply not powerful enough to see them. When DKIST came online in 2019, its 4-meter (13-foot) mirror immediately changed the game by allowing astronomers to view the bubbling, boiling solar surface at several wavelengths and at smaller scales than ever before.

“DKIST’s resolving capacity is equivalent to finding a quarter from a distance of 50 kilometers,” said Michail Mathioudakis, a solar astrophysicist at Queen’s University Belfast in the United Kingdom.

Roil and Toil

Kuridze, Wöger, and their team pushed DKIST’s resolving power to the max to zoom in on a small part of the Sun near a sunspot for a few minutes. This slow roiling is what they saw.

As the Sun’s plasma surface roils, the different flows of plasma shear against each other and create Kelvin-Helmholtz instabilities. Credit: NSF/NSO/AURA/MPS, CC BY 4.0

Here are those same two zoomed-in spots from before.

When scientists zoomed in on the high-resolution video of the Sun, they could watch KHI actively swirling. Credit: NSF/NSO/AURA/MPS, CC BY 4.0

“When we looked at these data and the movie, we immediately recognized the signature of Kelvin-Helmholtz instability,” Wöger said.

After they spotted the telltale signs of KHI, the observing team asked their modeling partners to simulate the motions of the Sun’s magnetically influenced fluidlike plasma, or magnetohydrodynamics, at a similarly high resolution. Creating simulations with the level of detail needed to compare with the DKIST observations requires phenomenal computational power and likely wouldn’t have happened without these observations as motivation, Wöger said.

Those simulations, based on fundamental physics and fluid dynamics principles, confirmed that the vortices captured by the DKIST images were likely created by KHI. The DKIST observations show that KHI might be ubiquitous across the Sun’s surface. This discovery was published in Nature in August.

“Models of the solar atmosphere have shown indications of this instability, but the paper has identified this physical process observationally in some of the smallest astrophysical scales,” said Mathioudakis, who was not involved with the research. “What surprised me the most is that this discovery was made with a relatively simple imaging setup and does not involve complex instrumentation, calibration issues, or data inversions. It will therefore stand the test of time.”

Small Swirls, Big Energy

“This is a very notable observation because Kelvin-Helmholtz instability in photospheric shear flows has been predicted theoretically for decades, but the relevant spatial scales were simply too small to resolve directly,” said Claire Foullon, a solar and space physicist at the University of Exeter in the United Kingdom.

“What is more surprising is how DKIST reveals it to be so widespread,” Foullon added. “Rather than being an occasional phenomenon, the observations suggest that this may be a fundamental part of the small-scale dynamics of the magnetized photosphere.” Foullon was not involved with the new discovery.

“There’s always this nice tension between simulation and observation.”

As KHI swirl around and around in the solar photosphere, they can twist up magnetic fields and store up energy. All that energy has to go somewhere, and many solar scientists think it might contribute to heating up the Sun’s corona.

“We are effectively seeing, for the first time, dynamics on the scales at which the magnetic foot-points of the corona are being continually stirred and restructured,” Foullon said.

The degree to which KHI contribute to coronal heating is still unknown, as is whether KHI exist on even smaller physical scales on the Sun. These might remain open questions for a while, Kuridze said, as observers figure out ways to push DKIST to even smaller physical scales and as simulations stretch farther to match.

“There’s always this nice tension between simulation and observation,” Wöger said. “When we see something [in observations], then we’re pushing the numerical simulations. And sometimes in the numerical simulations you see something, and then we’re trying to see whether it actually exists on the Sun.”

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

Citation: Cartier, K. M. S. (2026), Sharpest-ever image of the Sun shows small but mighty swirls, Eos, 107, https://doi.org/10.1029/2026EO260285. Published on 10 September 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.

Source: AGU Advances

A process known as the biological carbon pump (BCP) moves organic carbon from near the ocean’s surface to the ocean floor. That carbon often stays there for hundreds or thousands of years, making the BCP a primary source of carbon sequestration in the oceans. Scientists have thought that relatively large phytoplankton are more efficient at exporting carbon to the deep ocean than tiny cyanobacteria, but new research from Zhang et al. now calls that belief into question.

Using sediment traps placed at a depth of 1,000 meters in the South China Sea paired with multiple years of observations of surface phytoplankton communities, the authors show that about two thirds of the carbon sequestered in the ocean comes not from eukaryotic microalgae, but from another major group: cyanobacteria. These tiny, often unicellular prokaryotic phytoplankton are so named because of the blue-green color seen when they grow in large colonies. The main reason cyanobacteria play a larger role is that much of the carbon from microalgae is lost on its journey to the deep ocean but carbon from cyanobacteria is not. These findings could change how models of ocean carbon sequestration are constructed and hold implications for our understanding of how the BCP will shift in response to climate change, the authors say.

Evidence of taxon-specific divergence in carbon sequestration comes from the amino acid carbon isotope signatures found 1,000 meters deep, which allowed the researchers to tie the carbon they collected to phytoplankton groups at the surface. The authors found that production of carbon and phytoplankton structure at the surface vary considerably by season and corresponding nutrient availability. But surprisingly, the efficiency and origin of carbon collected deeper down do not change much by season.

The linked processes of aggregation and ballasting, in which carbon from cyanobacteria is incorporated into larger clumps that also contain other minerals, help cyanobacteria carbon sink more efficiently and reduce its loss during sinking by limiting microbial degradation. These findings may explain why high-latitude systems dominated by large phytoplankton do not always exhibit higher organic carbon sequestration efficiency compared to low-latitude oligotrophic gyres rich in cyanobacteria, the authors note. (AGU Advances, https://doi.org/10.1029/2025AV002148, 2026)

—Nathaniel Scharping (@nathanielscharp), Science Writer

Citation: Scharping, N. (2026), Tiny cyanobacteria may be the secret heroes of ocean carbon sequestration, Eos, 107, https://doi.org/10.1029/2026EO260283. Published on 10 September 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.

In November 2025, NASA and its European partners launched the Sentinel-6B satellite to improve hurricane forecasts, help protect infrastructure and benefit commercial industries, including shipping. The satellite is now flying 30 seconds behind its predecessor, Sentinel-6 Michael Freilich. Both satellites are providing precise sea level height measurements during what oceanographers expect to be a historic El Niño, a naturally occurring oceanic phenomenon in which warmer-than-usual Pacific waters shift global weather patterns.

Author(s): K. Zsukovszki, I. Papp, and L. P. Csernai

Using fully kinetic three-dimensional particle-in-cell (EPOCH) simulations, we investigate resonant planar nanoantenna configurations supporting localized surface plasmon resonances in hydrogen-rich media under irradiation by 795 nm femtosecond laser pulses with intensities of 1017–1019 W/cm2. The s…

[Phys. Rev. E 114, 035210] Published Thu Sep 10, 2026

Authors: Lakshman Srikanth and David Petley

This article was first published on the UNDRR PreventionWeb site.

A hazard event occurs. Imagine that the first report says 20 people have died. By the evening, the number is 27. The following day, authorities report 23. A week later, some people initially reported missing are confirmed alive, while others previously listed as injured have died. Which number is correct? 

The obvious answer is the latest one. 

But the problem is more complicated than changing numbers over time.

Before we can count disaster fatalities, we need to answer a set of more fundamental questions: What exactly was the event? Which deaths were caused by it? How directly were they connected? And what evidence do we have?

Disaster fatality data are therefore not simply collected. They are interpreted, classified, attributed, verified, and revised. This is part of a broader challenge in disaster loss and damage databases, where definitions, thresholds, reporting practices and data sources can strongly influence what ultimately gets recorded – a pattern documented in studies of missing data in the Emergency Events Database (EM-DAT) , loss-data fallacies and flood loss databases .

Here we focus specifically on the challenges of accounting for disaster fatalities.

Before counting deaths, we need to define the event and the causal pathway

Consider what happened during Typhoon Maysak in southern China in July 2026 . Heavy rainfall associated with a storm caused flooding in Guangxi, where a reservoir embankment was breached and a snake-breeding farm was flooded. Around 800-900 snakes escaped, including venomous species. Several people were bitten and at least one person died following a snakebite. Reports also indicated delays in accessing medical care contributed to the circumstances surrounding the fatality. An illustrative causal pathway of the reported snakebite fatality is represented in Figure 1:

Figure 1: Illustrative causal pathway for reported snakebite fatality during Typhoon Maysak

So how should this death be recorded? As a snakebite death? A flood-related death? A consequence of the reservoir failure? A typhoon-related death? Or an indirect disaster fatality?

There is no universally obvious answer because each label answers a slightly different question. Mortality registries typically seek the underlying cause of death: the disease, the injury or the circumstances that initiated the chain leading to death. Disaster databases, instead, often link deaths with a particular hazard event, including direct and indirect consequences.

Therefore, the same fatality can occupy several positions in a causal chain. It may be a “snakebite death” in a health or mortality registry, a “reservoir breach consequence” in infrastructure analysis, a “rainfall-related death” or a “flood related indirect death” in a disaster database. None of these labels are necessarily wrong. A disaster fatality can thus pass through multiple interpretive layers before becoming a database entry, and each layer may legitimately produce a different label. 

This distinction matters because disaster databases often use the event as their basic unit of analysis. Yet defining the boundaries of an event can itself be difficult, particularly when hazards cascade or occur simultaneously. A tropical cyclone can generate heavy rainfall, flooding, landslides, infrastructure failures and secondary impacts, while a single fatality may result from several links in that chain. The challenge is therefore not only to count deaths, but to decide which event and causal pathway those deaths belong to.

Typhoon Maysak’s official death toll rose from 39 to159 over the six weeks, according to the South China Morning Post . Several causal chains may be established to better understand the reasons for the deaths due to the event. A hypothetical causal pathway of disaster deaths is presented in Figure 2.

Figure 2: Hypothetical (not event verified) causal pathway associated with a tropical cyclone (darker boxes are hypothetical additions) Defining the main event (start point of the causal chain)

Disaster databases often require an event as the unit of analysis, although defining one event is difficult. A World Meteorological Organization (WMO) / United Nations Development Programme (UNDP)review of 91 post-disaster needs assessments (PDNAs) highlights recent efforts by WMO to develop a globally accepted standard for determining the temporal or spatial extent of a hazard event. 

The same document also highlights that only 20% of the PDNAs involved consultations with the National Hydrological and Meteorological Services. This matters directly for fatality attribution because understanding the initiating hazard and its evolution requires hazard expertise. If a report says “landslide deaths,” hazard experts may help determine whether the event was rainfall-triggered, earthquake-triggered, construction-related, mining-related, or part of a compound flood-landslide event. If a death follows a dam breach, hydrological and infrastructure expertise can help distinguish the initiating rainfall, the failure mechanism, downstream flood wave and later exposure pathway. 

The WMO Cataloguing of Hazardous Events approach offers one potential building block. By providing events with unique identifiers, recording temporal and spatial parameters, it can help link related or cascading hazards and associated impacts. This does not completely remove uncertainty, but it helps prevent complex chains from being flattened into one misleading hazard label. 

Defining the temporal boundary

Not all deaths due to disasters happen immediately. The Hillsborough stadium disaster, for example, caused 94 immediate deaths and 766 injuries, with 300 hospitalizations, in 1989. According to a contemporary timeline , the latest recognized death due to the event – the 97th – was confirmed in 2021, 32 years after the incident. The delayed death was recognized because a documented causal chain remained. Many disasters lack comparable long-term records and institutional follow-up. This creates systematic differences in what can later be recognized.

Physical injury is not the only pathway through which a disaster may contribute to later mortality. Hillsborough survivors also experienced long-term trauma, and some died by suicide . Disasters may contribute to trauma, bereavement, displacement, social isolation or substance-related harm, and these pathways can contribute to later mortality. Establishing causation for an individual case is, however, more difficult than identifying increased mortality at the population level.

While counting deaths, we are actually counting reports about deaths

Much of what we know about disaster fatalities comes through an information system that is itself imperfect. A death that occurs in a remote community with little media coverage may never enter an international database. A death in a major city may generate dozens of independent reports. Official emergency management sources may record events requiring institutional intervention while smaller events remain undocumented. Comprehensive disaster databases therefore often need to combine official records with media reports, local sources and other forms of evidence.

Disaster deaths also get lost in the noise of conflicting reports on the actual death toll. The landslide in Papua New Guinea in May 2024 illustrates this. Early media articles indicated over 2,000 individuals buried while other articles reported 670 deaths and, later, 200. That is a difference of about 1,800 individuals from the earliest estimate. Uncertainty arises not only because sources provide different numbers, but because they may be counting different things — such as estimated people buried, estimated deaths, confirmed or recovered bodies, directly affected people, or retrospective estimates.

Similar irregularities in media reporting that results in high uncertainty are common in outlier events such as earthquakes. The Kashmir earthquake in 2005 with an epicenter in Muzaffarabad affected Azad Jammu and Kashmir region administered by Pakistan, Khyber Pakhtunkhwa, areas in Afghanistan, and some areas of Jammu and Kashmir region, India. The death toll in Pakistan ranges from 87,350 to more than 100,000 – an uncertainty of almost 13,000 people.

Reports may also be in vernacular languages, use local hazard terminology, or translated differently across languages. A term translated as “landslide” may refer to a debris flow, slope collapse, embankment failure, quarry wall failure, construction-site collapse or broader rain-related ground failure. Official district-level statements may give only aggregate numbers such as “five dead, 20 injured”, without explaining whether the deaths occurred by drowning, collapse, electrocution, explosion, disease, evacuation accident or delayed medical access. These examples show that hazard datasets must work with partial visibility, language effects and evolving source quality. 

Actual deaths ≠ reported deaths ≠ verified deaths ≠ database deaths.

The database is therefore not simply a mirror of reality. It is also a record of what became visible, what could be verified and what could be attributed from the available evidence.

This raises a broader question: if disaster fatality data are inherently evolving, uncertain and dependent on how events and causal pathways are defined, what should a good disaster fatality accounting system look like?

What should we do differently? Build on existing data systems

More research is needed on the standardization of disaster fatality accounting and on the requirements for disaster fatality registries at different levels – from local and national systems to global disaster databases. 

There are already useful examples to build on. In landslide death mapping, specific methodologies and databases have been developed, including NASA’s Global Landslide Catalog and the Global Fatal Landslide Database . Recent work has also proposed frameworks for linking landslide inventories with loss and damage reporting . At the global level, EM-DAT provides a major compilation of disaster-related losses and impacts. The emerging WMO approach to cataloguing hazardous events could provide another important building block, by supporting more consistent identification and linking of hazard events.

Connect multiple sources and causal pathways

A key requirement is to move beyond reliance on a single source of information and systematically combine multiple sources, with mechanisms for corroboration, verification and revision.

The opportunity is to move towards interoperable and mineable disaster fatality datasets that recognize the complexity of compiling, interpreting, attributing and maintaining fatality information. 

Such systems should not only record how many people died but also preserve the causal pathway through which deaths occurred. This could allow fatality data to be used not only for counting impacts, but also for understanding where vulnerabilities emerge along the disaster chain and where risk reduction measures might intervene. 

Preserve how the evidence evolves

Finally, the evolution of the evidence itself should be preserved. A fatality database should not simply replace yesterday’s estimate with today’s number. It should retain the trajectory of the information: what was initially reported, what was subsequently verified or revised, what evidence led to the revision, and what remains uncertain.

In other words, a disaster fatality database should preserve not only the number of deaths, but also the event, the causal pathway, the evidence and the evolution of our understanding.

When we say that a disaster killed 20 people, we are not simply reporting a number. We are making a series of decisions about what the event was, how deaths were connected to it, what evidence was available, and how certain we are. Those decisions are often invisible once the number enters a database.

Improving disaster fatality accounting is therefore not simply a data-management exercise. It is part of improving how we understand disaster risk itself.

Why better counting matters

This matters because disaster fatality data are not only used to describe what happened. They are also used to measure whether disaster risk is being reduced. The Sendai Framework includes a global target to substantially reduce disaster mortality by 2030. We therefore use disaster fatality data to assess progress towards reducing disaster risk. But how confidently can we claim progress if the way disaster deaths are identified, attributed, classified and revised remains so difficult to standardize?

Further examples and an evolving compilation of disaster fatality data and sources are available in Deaths, injuries, evacuations due to hazard events as part of the  Disaster Risk Intelligence Hub .

Lakshman Srikanth is a Senior Advisor at Deltares, working on disaster risk management, & resilience. His work focuses on disaster vulnerability and risk assessment, risk management planning, and translating evidence and data into practical approaches for reducing disaster risk.

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.

Fieldwork is a vital part of geoscience research, allowing scientists and students to directly observe, interact with and collect data about Earth processes that cannot be replicated in the lab.

Rising sea levels and storm surges could cost $855 billion in economic losses across Australia by the end of the century, according to new research, the first of its kind in Australia.

From the Vredefort Dome in South Africa (the world's oldest and largest known meteorite impact structure) to the volcanic lakes of the Great Rift Valley in Kenya, Africa's landscapes hold some of Earth's oldest stories. Morocco's M'Goun UNESCO Global Geopark records more than 200 million years of geological history, while Tanzania's Ngorongoro-Lengai landscape brings together a vast caldera, an active volcano and the fossil record of Olduvai Gorge.

The Anthropocene became part of the zeitgeist after Nobel laureate Paul Crutzen launched the idea in 2000. This is the notion that a crescendo of human impacts has recently propelled Earth out of the relatively stable climatic conditions of the Holocene epoch, during which human civilization developed and flourished, to produce a new kind of planet.

Canada's last epishelf lake drained into the Arctic Ocean following the 2020 breakup of the Milne Ice Shelf and will not recover, a new study confirms.

The Arctic is often described as a distant frontier. In reality, it is one of the most consequential regions on the planet.

For 2 decades, the Arctic Report Card (ARC) has provided an annual, peer-reviewed assessment of environmental conditions across the Arctic, helping scientists, decisionmakers, industries, and communities understand how one of Earth’s fastest-changing regions is evolving. Issued annually since 2006, the report has become one of the world’s most trusted syntheses of Arctic observations and research.

The ARC’s importance stems from a simple truth: The Arctic matters far beyond the Arctic.

Changes unfolding across the region influence weather, climate, ecosystems, economies, transportation networks, and security interests around the globe. As Arctic temperatures continue to rise faster than the global average and sea ice continues its long-term decline, what happens in the far north increasingly affects people, organizations, and governments everywhere.

 
Related

•  Arctic Report Card Finds a New Home With AGU
•  Government Pulls Funding for the Arctic Report Card
•  Arctic Report Card Will Move Ahead Without NOAA
 

Despite the ARC’s value, NOAA recently announced that it will no longer facilitate or support the report’s development. Under this decision, NOAA says it will continue to collect and make available Arctic data, but it will no longer provide the important coordination, review management, communications support, and other resources that have historically enabled publication of the report card.

The announcement that a uniquely valuable scientific resource would be disrupted at a time when understanding Arctic change has never been more important raised serious concerns for many in the Arctic research community. Researchers emphasized that the ARC serves not only scientists but also policymakers, educators, industry stakeholders, and communities seeking reliable information about ongoing environmental change.

Fortunately, the ARC will continue.

AGU has announced that it is partnering with the University of Colorado Boulder Cooperative Institute for Research in Environmental Studies (CIRES) and the University of Alaska Fairbanks (UAF) to support the 2026 ARC effort and the report’s editorial team.

This partnership reflects a recognition shared across the scientific community: Continuity matters. Long-term observational records become more valuable with every passing year. Regular assessments such as the ARC provide critical context, helping connect individual observations into a coherent picture of change.

One of the most visible consequences of Arctic change is the opening of new maritime pathways.

As seasonal sea ice retreats, routes that were once inaccessible for much of the year are becoming increasingly navigable. This shift carries not only significant economic opportunities but also environmental challenges. Shipping companies, logistics providers, and governments are all watching closely how changing Arctic conditions could reshape global trade routes, reduce transit times between major markets, and alter patterns of maritime activity.

Richard Spinrad, an oceanographer and former NOAA administrator, presents the 2024 Arctic Report Card at the AGU Annual Meeting in Washington, D.C. Credit: Beth Bagley/AGU

At the same time, operating in Arctic waters remains complex and risky. Weather variability, sea ice dynamics, and rapidly changing environmental conditions require robust observational data and scientific analysis to support navigation, risk management, and long-term planning. The ARC helps provide exactly that kind of information.

The Arctic is also attracting growing attention from a national security perspective.

Several strategically important nations share Arctic coastlines. Increased activity across Arctic waters has heightened interest in maritime awareness, monitoring capabilities, and regional infrastructure. Scientific observations and environmental intelligence help inform understanding of operating conditions across the region, supporting everything from navigational safety to situational awareness.

The value of Arctic information extends below the surface. Interest in seabed resources and terrestrial mineral deposits across and adjacent to the Arctic has grown substantially in recent years. Accessing and operating in these environments requires a deep understanding of changing weather patterns, permafrost conditions, sea ice trends, and environmental risks. Reliable scientific assessments help inform planning decisions, infrastructure investments, and long-term management considerations.

In addition—and critically—the ARC serves far more than just the scientific community.

Commercial fisheries depend on observations and analyses that help them understand changing ocean conditions and ecosystem shifts. Arctic waters support important fisheries in their own right, but the Arctic also serves as a bellwether for broader changes occurring across Northern Hemisphere marine systems. Understanding these trends can help inform fisheries management and operational planning well beyond the Arctic alone.

The Arctic Report Card has never been solely about documenting environmental change. It is about providing society with the information needed to understand and respond to that change.

Similarly, maritime shipping and logistics companies monitor Arctic environmental conditions to evaluate emerging routes, assess operational risks, and support strategic planning. Insurers and reinsurers rely on environmental data and climate information when evaluating risk exposure across a number of sectors affected by changing Arctic conditions, including shipping, fisheries, infrastructure, and more.

In each case, informed decisions are absolutely critical. Those decisions depend on trustworthy observations, expert analysis, and a consistent record of environmental change over time. The ARC has become an essential source of that crucial information.

Maintaining the continuity of this annual assessment ensures that all the various people who need this information, including decision-makers, researchers, communities, and industries, continue to have access to trusted information about one of Earth’s most consequential regions.

The Arctic Report Card has never been solely about documenting environmental change. It is about providing society with the information needed to understand and respond to that change.

As the Arctic continues to transform, the need for rigorous observations, expert assessment, and trusted scientific communication will only grow. Supporting efforts like the ARC is therefore not simply an investment in Arctic science; it is an investment in the information infrastructure that helps communities, economies, and governments navigate an increasingly interconnected world.

And that is why the Arctic Report Card remains so important, not just for the Arctic but for all of us.

—Chris Avery, Director, Science Assessment, AGU; and Kristen Averyt Hinton, Executive Vice President, Science, AGU 

Citation: Avery, C. and Averyt Hinton, K. (2026), The Arctic Report Card matters more than ever; AGU joins partners to publish in 2026, Eos, 107, https://doi.org/10.1029/2026EO260291. Published on 9 September 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.

Indonesia's Mount Anak Krakatau has erupted again this week, sending plumes of volcanic ash high into the sky and disrupting hundreds of flights. More than 300,000 passengers have been affected.

Mexico City has an atmospheric puzzle: Air quality policies have significantly reduced concentrations of many primary pollutants such as carbon monoxide, sulfur dioxide, and nitrogen oxides, but decreases in the amount of ozone, a secondary pollutant, have leveled off in recent years. This apparent contradiction is part of what the scientific literature describes as the “ozone paradox.” Could air temperatures be the key to this conundrum?

The ozone paradox has implications for health and safety. Unlike stratospheric ozone, elevated ozone concentrations near ground level are associated with respiratory tract inflammation, increased susceptibility to lung infections, and aggravation of existing respiratory diseases. Further, ozone can impede the growth of certain plants and can make them more vulnerable to disease and damage from insects and severe weather.

Ozone in the troposphere remains a challenge for Mexico City’s public health and policymakers. Unlike other pollutants emitted directly from car exhaust or industry, ozone forms when precursor pollutants such as nitrogen oxides (NOx) and volatile organic compounds (VOCs) interact with sunlight through photochemical reactions. Under favorable atmospheric conditions, more heat and more sunlight can increase the efficiency of ozone formation. As temperatures rise, so does the efficiency of atmospheric reactions related to ozone production. This dynamic was identified decades ago [Sillman and Samson, 1995; Jacob and Winner, 2009; Nolte et al., 2021] and reaffirmed by our most recent analysis of atmospheric data collected in the Mexico City area between 2000 and 2021 [Castro et al., 2025].

Our data revealed that high-ozone episodes occur most frequently at the end of the dry season, particularly during April and May, and again in winter. The April–May peak coincides with intense solar radiation and rising ambient temperatures, creating ideal conditions for ozone formation. The winter peak, on the other hand, is related to atmospheric stagnation and a compressed boundary layer that traps pollutants near the surface [Castro et al., 2025].

A Complex History of Air Pollution

The ozone paradox is especially relevant in Mexico City because the region has a long and complex history of air pollution management.

The ozone paradox is especially relevant in Mexico City because the region has a long and complex history of air pollution management. The metropolitan area, located at high elevation and surrounded by mountains, has meteorological and topographic conditions that often limit pollutant dispersion.

During the late 20th century, the city became widely known for severe air pollution episodes associated with rapid urban growth, industrial activity, traffic emissions, and frequent atmospheric stagnation. Since then, air quality policies have substantially improved the situation, but ozone remains one of the most persistent challenges.

For more than 20 years, Mexico City has addressed these issues with ambitious environmental policies. The city’s comprehensive “ProAire” programs, for example, included expansion of the Metrobús system, reforestation efforts, and environmental education. Additionally, the city introduced requirements for low-sulfur fuels, vehicle inspection programs, and measures to reduce industrial and transportation emissions.

These efforts should not be underestimated. They have successfully reduced several ozone precursors and primary pollutants and represent one of the most important urban air quality management experiences in Latin America. Mexico City’s long-term air quality monitoring network has also made it possible to evaluate the effectiveness of these policies and to identify new atmospheric challenges as they emerge.

So why do ozone levels remain high, even when levels of its precursor pollutants decrease?

Scientists have concluded that the missing piece of the puzzle may be temperature. The average maximum surface temperature in Mexico City’s metropolitan area increased from 32.5°C to 33.2°C between 2000 and 2021. This increase of almost 1°C, although modest, acts as a catalyst, accelerating the chemical efficiency of ozone formation [Sillman and Samson, 1995; Jacob and Winner, 2009].

Ozone is becoming a defining pollutant of a warming and chemically complex urban atmosphere.

The COVID-19 lockdown offered an unusual opportunity to study this phenomenon. During this period of reduced mobility, emissions of several primary pollutants decreased, but ozone did not decline in the same proportion. These data indicate that reducing NOx alone reduces the chemical removal of ozone by nitric oxide, but VOCs, sunlight, and high temperatures continue to favor ozone production [Peralta et al., 2021].

Mexico City is not an isolated case; other large urban regions around the world have also faced difficulties reducing ozone, even after achieving important reductions in primary emissions. A study of air quality during the COVID-19 lockdown in northern China saw similar results: Surface-level quantities of NOx decreased by about 60%, while surface-level ozone increased by a factor of 1.5 [Shi and Brasseur, 2020]. This broader context suggests that ozone is becoming a defining pollutant of a warming and chemically complex urban atmosphere.

Quantifying the Relationship Between Heat and Ozone

A relatively small temperature increase can have a disproportionately large effect on ozone formation.

Using data from the RAMA (Red Automática de Monitoreo Atmosférico) and REDMET (Red de Meteorología y Radiación Solar) networks, we identified a clear exponential relationship between ozone concentrations and temperatures in Mexico City between 2000 and 2021 (Figure 1), with a model fit to the data explaining more than 70% of the variation in peak ozone levels. Essentially, each additional degree of warming produces a progressively larger peak in ozone concentration [Castro et al., 2025].

Fig. 1. Maximum daily ozone (O3) concentrations show an exponential relationship with maximum daily surface temperature (T) on the basis of measurements collected from 2000 to 2021 by the RAMA (Red Automática de Monitoreo Atmosférico) and REDMET (Red de Meteorología y Radiación Solar) networks. R2 is the coefficient of determination and indicates the proportion of variation in the maximum daily ozone concentration accounted for by the exponential fit. Credit: Castro et al. [2025]

Importantly, our results show that ozone does not respond to temperature in a simple linear way and help to explain why ozone episodes may continue even when long-term emission control policies are in place. A relatively small temperature increase can have a disproportionately large effect on ozone formation when other conditions are favorable. These conditions include intense solar radiation, the presence of stagnant air masses, and sufficient availability of precursor gases.

This dynamic is most visible during the “ozone season” from March to May, when high solar radiation and stagnant air masses prevent pollutants from dispersing in the atmosphere. These conditions are no longer just seasonal anomalies; they are expected to become more frequent or more intense as climate change progresses [Estrada et al., 2023]. In this context, ozone management is not only an air quality issue but also a climate adaptation challenge.

The city is running a race where the finish line, clean air, is receding farther and farther because of global warming.

The relationship between heat and ozone is also connected to urbanization. Large cities modify their local climate through the urban heat island effect, which can increase surface temperatures relative to surrounding rural areas. In Mexico City, this effect interacts with regional climate warming, local emissions, and basin meteorology. Together, these factors can create a more reactive atmosphere than the one that existed 2 decades ago.

The city is running a race where the finish line, clean air, is receding farther and farther because of global warming.

The Way Forward: Integrated Policy

Our findings highlight a critical challenge for urban planning. If temperatures continue to rise, technological advances in vehicles and industry could be partially offset by the increased chemical efficiency of the atmosphere [Nolte et al., 2021]. The next generation of policies must recognize that ozone control requires a more integrated strategy.

First, ozone mitigation should consider the balance between NOx and VOC emissions. Because ozone formation depends on nonlinear chemistry, reducing only one group of precursors may not always produce the expected response. More detailed control of VOC emissions from fuels, solvents, industry, traffic, and consumer products may be necessary, especially during the ozone season.

Second, air quality policy should be linked to meteorological forecasting. High-ozone episodes often occur under predictable combinations of high temperature, intense solar radiation, atmospheric stagnation, and limited vertical mixing. Short-term actions during these periods could help reduce peak exposures, particularly for vulnerable populations such as children, older adults, outdoor workers, and people with respiratory disease.

Air quality strategies cannot exist in isolation. They must be integrated with climate change adaptation and mitigation goals.

Third, reducing the urban heat island effect should be considered part of ozone management. Measures such as increasing urban vegetation, protecting green areas, using reflective materials, improving building efficiency, and reducing heat emissions from transport and infrastructure can contribute to climate adaptation while also helping to reduce conditions favorable for ozone formation.

Fourth, long-term monitoring must continue. Mexico City’s atmospheric monitoring networks are essential because they allow scientists to distinguish between emission-driven trends and climate-driven changes. Without continuous measurements of pollutants and meteorological variables, it would be much more difficult to understand why ozone remains persistent despite reductions in other pollutants.

To protect public health, air quality strategies cannot exist in isolation. They must be integrated with climate change adaptation and mitigation goals. In megacities like Mexico City, reducing emissions is no longer enough; we must also address the urban heat island effect and prepare for a fundamentally more reactive and hotter atmosphere.

Mexico City’s experience with the ozone paradox offers a useful lesson for other megacities. The challenge ahead is not only to emit less but also to understand how a warmer atmosphere changes the effectiveness of traditional pollution control strategies. Ozone is therefore more than an air pollutant; it is a signal of how climate change can reshape urban environmental risks.

References

Castro, T., et al. (2025), Evolution of tropospheric ozone and surface temperature in Mexico City from 2000 to 2021, Atmosphere, 16(12), 1379, https://doi.org/10.3390/atmos16121379.

Estrada, F., et al. (2023), State and perspectives of climate change in Mexico: A starting point, cambioclimatico.unam.mx/.

Jacob, D. J., and D. A. Winner (2009), Effect of climate change on air quality, Atmos. Environ., 43(1), 51–63, https://doi.org/10.1016/j.atmosenv.2008.09.051.

Nolte, C. G., et al. (2021), Regional temperature-ozone relationships across the U.S. under multiple climate and emissions scenarios, J. Air Waste Manage. Assoc., 71(10), 1,251–1,264, https://doi.org/10.1080/10962247.2021.1970048.

Peralta, O., et al. (2021), Ozone over Mexico City during the COVID-19 pandemic, Sci. Total Environ., 761, 143183, https://doi.org/10.1016/j.scitotenv.2020.143183.

Shi, X., and G. P. Brasseur (2020), The response in air quality to the reduction of Chinese economic activities during the COVID-19 outbreak, Geophys. Res. Lett., 47, e2020GL088070, https://doi.org/10.1029/2020GL088070.

Sillman, S., and P. J. Samson (1995), Impact of temperature on oxidant photochemistry, J. Geophys. Res., 100(D6), 11,497–11,508, https://doi.org/10.1029/94JD02146.

Author Information

Telma Castro (telma@atmosfera.unam.mx), Oscar Peralta, and Salvador Reynoso-Cruces, Instituto de Ciencias de la Atmósfera y Cambio Climático, Universidad Nacional Autónoma de México, Mexico City; Harry Alvarez-Ospina, Facultad de Ciencias, Universidad Nacional Autónoma de México, Mexico City; and Alejandro Salcido, Instituto Nacional de Electricidad y Energías Limpias, Cuernavaca, Mexico

Citation: Castro, T., O. Peralta, S. Reynoso-Cruces, H. Alvarez-Ospina, and A. Salcido (2026), The ozone paradox: A growing cost of a warmer atmosphere, Eos, 107, https://doi.org/10.1029/2026EO260284. Published on 9 September 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.