AGU Advances
A cross-disciplinary, gold open-access journal publishing full length, high-impact research articles across all of the Earth and space sciences. Submit your research!
Volume 7 Issue 2 | May 2026
- EDITORIALS
- COMMENTARY
- VIEWPOINTS
- RESEARCH ARTICLES
- EDITORS' PICKS FROM OTHER JOURNALS
- FEATURED AGU SPECIAL COLLECTION
FROM THE EDITOR IN CHIEF
This issue presents 17 articles addressing exciting research challenges in planetary sciences, space physics, and space weather (3 contributions); cryospheric and atmospheric sciences (4); biogeosciences (2); ocean sciences (2); solid Earth, Earth surface, and Earth system modeling (3); and wildfires (2).
We are also very pleased to present an editorial acknowledging the essential and outstanding support provided by 389 peer reviewers, who voluntarily completed 507 reviews for AGU Advances in 2025. Serving as an editor offers the privilege of learning a great deal from reviewers and witnessing firsthand the dedication of community members whose work supports the progress of science. I would like to reiterate here our sincere gratitude and heartfelt “thank you” to our reviewers.
Among the excellent papers featured in this issue, I would like to first highlight two studies focused on predicting the future of glaciers and the Greenland Ice Sheet. Ryan et al. propose an innovative method for exploring glacier melt dynamics using deep learning techniques applied to SkySat imagery. They find that the filling and draining of large, interconnected supraglacial lakes may contribute in previously undocumented ways to faster-than-expected sea level rise. These findings resonate with the conclusions of Martin et al., who developed a model representation of the rapidly retreating Pine Island Glacier in West Antarctica. Their results could have important implications for predictions of future glacial melt, as they suggest that ice viscosity values used in state-of-the-art ice flow prediction models may lead to an underestimation of sea level contributions by 21%. Together, these innovative studies shed new light on our understanding of glacier melt rates and sea level rise under climate change.
Increasingly, important research questions are focused on improving prediction models for the Earth system. In this context, machine learning and artificial intelligence are gaining growing traction. However, their performance relative to physically based approaches remains a subject of debate. Chen et al. present results from a set of idealized experiments demonstrating that physics-informed neural networks may indeed perform comparably to conventional physics-based Earth system models when simulating Earth’s weather and climate across timescales ranging from seasonal to annual, and multi-decadal, while using computational resources more efficiently. Their conclusions open the door to exciting developments in scenario analysis.
Overall, the contributions presented here provide an inspiring overview of current and compelling research challenges in geosciences, resulting from a steadily increasing number of contributions to AGU Advances. The vitality, enthusiasm, and diversity of our community are stimulating optimism, particularly for early career researchers.
EDITORIALS
Thank You to Our 2025 Peer Reviewers
COMMENTARIES
Modeled and Observed Stratospheric Temperature Changes: Implications for Fingerprint Studies
Santer et al. warn that a recent U.S. Department of Energy report misrepresents their work and downplays the role of human activity in global warming. [Press Release]. Santer et al.
The Multi-Faceted Water Footprint of Data Centers
Data centers powering artificial intelligence consume significant amounts of water, highlighting the need for greater transparency regarding water use in both existing and planned facilities. Privette et al.
VIEWPOINTS
Explaining Glacial-Interglacial CO2 changes requires multiple ocean processes
Muglia highlights the work of Gray et al. (2026) from our first issue of the year to our understanding of glacial-interglacial changes in the carbon cycle. Muglia
RESEARCH ARTICLES
Global Observations Reveal Rapid Reorganization of Ocean Nutrients
Data reveal that changes in nutrient levels vary depending on depth and distance from shore—and that these changes are happening more quickly than scientists realized. Martiny et al.
The Fate of the Greenland Ice Sheet: Deep Learning from SkySat Images
Surface meltwater ponding and drainage in the Greenland Ice Sheet are analyzed at high spatial and temporal resolution through SkySat imagery and deep learning. Ryan et al.
Machine Learning Could Enhance Earth System Modeling
Based on tests of a machine learning-based (ML) hybrid model, combining ML with established physics-based frameworks represents a promising path toward developing ML-based Earth system models. Chen et al.
Glaciers May Flow into the Ocean More Quickly Than We Think
New research found that adjusting a key model variable may give more accurate predictions of glacial retreat. Martin et al.
Timing of Geomagnetic Storms Shapes Their Impact
The impact of geomagnetic storms, which can disrupt satellites, GPS, and power grids, is shown to depend on their onset timing. Ghag et al.
Stealth Superstorms Reveal Lightning on Jupiter: Beyond the Superbolt
On the gas giant, the strength and frequency of lightning appear to be more diverse than previously thought. Wong et al.
Toward Marine Cloud Brightening at Scale – A Science Agenda
Marine cloud brightening is a solar radiation management solution that could cool the planet by changing the albedo of low-altitude marine clouds to increase reflected shortwave radiation. Doherty et al.
The Effects of Carbonate Precipitation on the Physical Properties of Basalt
Lab-scale experiments monitoring the internal structure of basalt samples while injecting reactive fluids reveal how seismic waves could be used to monitor carbon mineralization in the field. Simpson et al.
More Braided Rivers from Increasing Flow Variability
Global analysis of satellite data and river flow records show that higher flow intermittency after climate change may lead to an increasing number of threads in braided rivers, thus impacting ecosystems. Zhao et al.
Trees Shed Their Leaves to Adapt to Droughts
The browning or loss of tree leaves that can be observed during droughts may be a coping mechanism to deal with dry circumstances by avoiding additional water stress. Quetin et al.
Managed Agriculture Hinders Predictability of Critical Zone Features
Shifts in the critical zone over time are attributed to environmental drivers using a data-driven approach and traced to changes in intensive management. Goodwell et al.
Amazon River Breezes Mimic Pollution in Clouds
Natural river breezes create clouds over the Amazon that mimic the signs of pollution, complicating climate impact assessments. Christensen et al.
What Makes Mars’s Magnetotail Flap?
Spacecraft reveal a key driver of up-and-down motions of thin, current-carrying plasma sheets on the nightside of Mars. Wen et al.
Future Wildfire Burned Areas in U.S. West May Be Overestimated
Using soil moisture as an indirect measure of fuel dryness leads to lower predictions for the total area of U.S. western forests burned by wildfires as global temperatures rise. [Also mentioned in This Week From AGU] Cheng et al.
Pre-Fire Fuel Conditions Are Dominant Drivers of Burn Severity in the 2025 Los Angeles County Fires
Estimations of fuel conditions from remote sensing could help guide pre-fire severity hazard assessments for more proactive land management. Ward-Baranyay et al.
EDITORS' PICKS FROM OTHER JOURNALS
Cosmic Silence
Bansal et al. [2026] introduce TranQuiL, a groundbreaking system that revolutionizes Radio Quiet Zone enforcement by enabling long-range detection and precise localization of Wi-Fi and Bluetooth interference with unparalleled accuracy.–Jothiram Vivekanandan, Radio Science
Wildfire Radar
Laureau [2026] demonstrates the use of operational weather radar measurements to track long-range ember fallout and rapid spread of intense wildfires in real time.–William J. Randel, JGR: Atmospheres
Ice Sheet Mass Loss
Glen’s Law describes the simple physics of ice flow that underpins ice sheet models, but Lilien et al. [2026] show that parameter choices substantially influence the outcome of model projections.–Ann Rowan, JGR: Earth Surface
FEATURED AGU SPECIAL COLLECTION
This collection will further understanding of recent and ongoing progress related to projecting sea-level change and its drivers, as well as user perspectives on what has and has not worked in the use and communication of prior sea-level projections. We welcome original research both on process understanding and on deriving statistical and/or physical relationships from process understanding that can be represented in the emulators underlying probabilistic projection frameworks. The collection will accompany the Chapman Conference, "Updating Usable Projections of Future Sea Level", which will synthesize the current state of the science to build an actionable roadmap to the next generation of integrated sea-level projections.


