While there is widespread evidence for the cyclic advance and retreat of Northern Hemisphere ice sheets throughout the late Pleistocene, the scarcity of absolute-dated records has made it challenging to determine the precise relationship between high-latitude insolation cycles and cryosphere response. Because the δ18O records of stalagmites from northwest Iberian coastal caves track the δ18O value of the North Atlantic Ocean, they can provide new constraints on periods of surface-ocean freshening due to meltwater from retreating ice sheets. Our new, absolute-dated stalagmite δ18O record from 216 to 111 ka reveals two precessionally paced phases of Northern Hemisphere ice sheet retreat and regrowth between the start of marine isotope stage (MIS) 6 and Termination II. This variability in Northern Hemisphere ice sheets is simulated in some intermediate-complexity coupled climate−ice sheet models but has not been evident in benthic foraminifera δ18O stacks or sea-level curves derived from them. One explanation is that antiphased Antarctic ice sheet expansion compensated for the Northern Hemisphere ice volume in the global average ocean δ18O, implying a more dynamic Northern and Southern Hemisphere cryosphere in response to high-amplitude precessional cycles in MIS 6.

This research investigates the interactions between the Atlantic Meridional Overturning Circulation (AMOC) and the Antarctic Ice Sheet, which are both recognized as critical climate tipping elements in the Earth system. We conduct a computer simulation with artificial freshwater input in the North Atlantic to collapse the AMOC for 1500 years. Our findings show no destabilization of the Antarctic Ice Sheet induced by changes in the Southern Ocean during this period.

San Salvador Island, along the eastern margin of the Bahamas−Caicos Archipelago, is a key locality for reconstructing Pleistocene carbonate stratigraphy and Last Interglacial (marine isotope stage [MIS] 5e, 129−116 ka) sea-level history. A high-resolution dataset, including airborne light detection and ranging (LiDAR) data, drone photogrammetry, differential global positioning system (GPS), petrography, field mapping, and a 44 core database, establishes a refined stratigraphic framework and geologic history. This framework identifies interglacial deposits from MIS 11, MIS 9, MIS 5e (subdivided into MIS 5e_1 and MIS 5e_2a−e), MIS 5a, and MIS 1. MIS 11 and MIS 9 units are constrained by superposition and paleoelevation relative to established sea-level stands. MIS 5e is the most extensively exposed and stratigraphically significant interval for interpreting former sea level across the Bahamas−Caicos Archipelago. Early MIS 5e deposition is represented by the lower Cockburn Town reef (“reef I”; 129−125 ka) at <1 m relative sea level (RSL) prior to intra−MIS 5e erosion. Upper MIS 5e_2 subunits record transgression and highstand variability. Subunits MIS 5e_2a−c consist of composite parabolic dune ridges and associated lagoonal, tidal-flat, tidal-channel, and strand-plain facies formed at +3 m RSL. MIS 5e_2d consists of shingled dune sheets indicating increasing aridity during RSL rise to +7 m. MIS 5e_2e deposition forms down-stepping strand plains initiating at +8 m and descending below present sea level, recording late-stage sea-level fall. Final MIS 5e reef development (“reef II”; 122−119 ka) includes patch reefs localized along headlands at +1 m to +3 m RSL. This revised stratigraphy refines Last Interglacial sea-level dynamics and models carbonate-platform response to rapid climate change.

Nature Geoscience, Published online: 03 August 2026; doi:10.1038/s41561-026-02055-5 Northward movement of the Southern Ocean subtropical front diminished salt transport by Agulhas Leakage in the late Pliocene, although this had no apparent influence on the strength of the Atlantic Meridional Overturning Circulation, according to proxy records from marine sediment cores.

Abstract Marine Isotope Stage (MIS) 11 (424–374 ka) has long been an analog for Holocene climate because it is the most recent interglacial period with similar orbital conditions. However, there is global and regional variability in the climate response to MIS 11 warmth. Here, we review sediment core records from across the Bering Sea to investigate changes in paleoceanographic conditions from late MIS 12 to early MIS 10. Sea ice was present over much of the Bering Sea during MIS 11, but today, none of the sites investigated are ice‐covered. This suggests that sea ice regimes in the Bering Sea during MIS 11 were different from those of the Holocene. There are also regional differences in the response of sea ice to warming. At the Umnak Plateau (southeastern Bering Sea), sea ice concentrations decline during deglaciation, but they remain high at the slope sites (western Bering Sea) until peak MIS 11. During peak interglacial warmth, sea ice re‐advances over the Umnak Plateau but declines over the slope sites. Late MIS 11 is characterized by high concentrations of seasonal sea ice at the Umnak Plateau, whereas sea ice at the slope sites fluctuates between consolidated and unconsolidated ice cover. This east‐west dichotomy may be explained by changes in the behavior of the Aleutian Low pressure system. Productivity increases dramatically during deglaciation, due to increased upwelling and sea level rise bringing fresh nutrients into the oceans. This is characterized by increased diatom productivity, increased terrestrial carbon deposition, and laminations at all sites.

Abrupt transitions are a central concern in climate and ecological research and may arise when critical thresholds known as tipping points are crossed. However, previous work has shown that finite-time overshoots of tipping points can be safe, and that such behavior is captured by an inverse-square-law criterion when overshoots are sufficiently small and slow. So far studied in isolated systems with external drivers, (un)safe overshoots may also emerge from interactions between subsystems. Here, we investigate safe-overshoot phenomena in unidirectionally coupled slow-fast systems featuring both nonlinear interactions and coupling through time derivatives. Specifically, we derive a criterion for the occurrence of safe overshoots analogous to the inverse-square law for isolated systems, but adapted to interactive settings, and expressed explicitly in terms of the timescale separation and coupling strength between subsystems. We illustrate these results using two conceptual models in which the Atlantic Meridional Overturning Circulation interacts with either the Amazon rainforest or the Greenland Ice Sheet.

Abstract. The stability of Antarctic ice shelves is closely linked to the properties of the firn layer, which regulates meltwater retention and influences ice shelf vulnerability to hydrofracturing. Firn Densification Models (FDMs) provide valuable insights into the firn structure, but key properties such as grain size are often parameterized using simple approximations, leading to significant uncertainties, especially in regions lacking in-situ validation. Here, we infer an effective firn grain-size parameter across Antarctic ice shelves, constrained by 15 years (2007–2021) of active microwave observations from the C-band Advanced Scatterometer (ASCAT). Within this framework, we use the Institute for Marine and Atmospheric Research Utrecht Firn Densification Model (IMAU-FDM) to prescribe the state of the firn layer (layer thickness, density, temperature, and liquid water content) and couple it with the Snow Microwave Radiative Transfer (SMRT) model to simulate radar backscatter. Grain size is treated as an unknown microstructural parameter and is optimized by minimizing the misfit between FDM–SMRT simulations and ASCAT observations. The retrieved parameter provides an observationally constrained measure of effective firn microstructure within the adopted FDM–SMRT framework, although independent validation is required before interpreting it as an absolute physical grain-size measurement. The framework is further used to examine how the sensitivity of ASCAT backscatter varies across firn regimes, and how this influences the interpretation of backscatter in terms of firn air content (FAC; the vertically integrated pore-air content of the firn column). Our results show broad consistency between the optimized effective grain size and IMAU-FDM estimates in high-FAC regions, where ASCAT backscatter is most sensitive to interannual variability in grain size. In contrast, larger discrepancies emerge in intermediate- to low-FAC regions, particularly on the Amery Ice Shelf, where ice saturation in the firn (pore-space depletion) influences grain growth. The sensitivity experiments indicate that inversion constraints are regime-dependent, with the strongest sensitivity to grain size in intermediate firn regimes and weaker constraints in strongly depleted firn. Furthermore, by statistically reducing grain-size-driven scatter after inversion, we present a proof of concept for a more interpretable backscatter–FAC relationship, which may support future development and evaluation of FAC-oriented diagnostic retrieval frameworks. These findings provide a basis for improving firn model parameterization and supporting large-scale diagnostic monitoring of firn-state variability across Antarctic ice shelves in a warming climate.

Moisture and energy transport across mid-latitude Asia, controlled jointly by the Westerlies and the East Asian Monsoon, constitutes a critical element of the global climate system. However, a significant climatic-environmental divergence exists between Central and East Asia, which has become increasingly prominent in recent years, highlighting the need to investigate its past evolution. Here, we trace the orbital-scale spatiotemporal evolution of the climate-environment in Central and East Asia since the Late Pliocene, based on high-resolution spectral parameters and geochemical data from a deep core in the Tibetan Plateau, and a compilation of multiple paleoclimate records from both regions. The results indicate a marked climatic divergence between the two regions since ca. 2 Ma, characterized by synchronously increased precipitation in East Asia and enhanced aridification in Central Asia. We attribute this divergence to concurrent expansion of the Antarctic Ice Sheet. The resulting Southern Hemisphere cooling created an atmospheric energy deficit, dynamically forcing monsoon precipitation northward into East Asia. Simultaneously, this expansion weakened the Atlantic Meridional Overturning Circulation, steepening the North Atlantic meridional temperature gradient and shifting the Northern Hemisphere Westerlies southward to divert moisture away from Central Asia, thereby exacerbating aridity there ca. 1.2 Ma. A consistent shift in dominant periodicity occurred in records of both regions, transitioning from the previous ∼400 k.y. forcing to a co-dominance of ∼400 k.y., ∼100 k.y., ∼41 k.y., and ∼21 k.y. cycles. This shift suggests enhanced sensitivity of Asian climate to orbital forcing since 1.2 Ma, potentially due to reduced climate stability caused by high-latitude ice expansion.

arXiv:2607.28468v1 Announce Type: new Abstract: The Atlantic Meridional Overturning Circulation is monitored and emulated through reduced indices, but such projections discard thermohaline structure and may require either explicit physical state or memory of the observed index. We compare these strategies in 30 branch-consistent CMIP6 trajectories from eight model families using leave-one-family-out validation. Salinity, temperature and density information improves direct 20-year forecasts, whereas compact scalar memory is top-ranked at every recursive horizon and yields the lowest case-averaged Brier score. A matched ablation confirms that feedback from memory improves long-horizon prediction. Physical state and recent trends also predict future ocean-state changes beyond the emissions pathway, most robustly at five years. NorESM under SSP5--8.5 identifies a forcing-dependent limit of scalar compression, while MIROC shows negative long-horizon transfer. A resolvent analysis explains why stable memory components do not guarantee stability of the complete learned model. Physical augmentation and memory compression therefore serve different AMOC prediction tasks.

Abstract. In the context of global warming, studies have shown that the Antarctic Ice Sheet (AIS) has been experiencing continuous mass loss, constituting a significant contribution to global sea level rise. Utilizing an optimized input-output method (IOM), this study evaluates the 2013–2022 mass balance of eight widely studied Antarctic drainage basins. By integrating RACMO2.4 surface mass balance (SMB), BedMachine ice thickness, and ITS_LIVE annual velocity, we quantify the sensitivity of mass balance estimates to flux gate (FG) location. Results demonstrate that while flux estimates along disparate radar tracks vary by <4%, discrepancies between grounding line and flight-line-based gates can reach 20%, highlighting the importance of flux gate selection. Regional assessments reveal a pronounced regime shift in East Antarctica: basins previously in near-balance or slight loss—including the Amery region and Basins 12–13—transitioned to significant mass accumulation post-2020, primarily driven by a sharp increase in SMB. In contrast, the Amundsen Sea sector in West Antarctica exhibited sustained and accelerating mass loss, with net deficits increasing from −163.9±15.3 Gt a−1 (2013–2019) to −179.2±15.5 Gt a−1 (2020–2022). This divergence stems from the dominance of anomalous precipitation in the East versus persistent ice flux acceleration in the West.

We generated five new and compiled previously published Holocene temperature timeseries from southwestern Greenland. We found temperature maxima between 7500 and 5000 years ago, with temperatures approx. 2–5 °C higher than 1994–2024. Afterward, temperatures gradually cooled towards the present, following maximum summer insolation. Local variations in the magnitude and timing of the thermal maximum are related to topography and proximity to the ice sheet and the ocean.

arXiv:2607.28220v1 Announce Type: new Abstract: Atmospheric predictability declines rapidly beyond the next ten days, such that forecasts at longer lead times primarily convey large-scale trends rather than specific states. Yet in a warming world, improving early warnings of extreme heat is an increasingly critical challenge. Here we evaluate six state-of-the-art deep learning weather emulators - Pangu-Weather, FuXi, ArchesWeather, AIFS, GraphCast and Aurora - alongside leading dynamical systems and statistical baselines in forecasting global near-surface temperature and extreme heat at lead times of 10-15 days. We find that several emulators rival or even surpass physics-based forecasts in deterministic temperature skill, but do so at the cost of reduced spectral fidelity, in a process widely known as blurring. While all models show some degree of predictive skill for extreme heat, most emulators under-represent peak intensities, and IFS recall is greater than that of any of the emulators. These results highlight both the emerging potential of AI to enhance extended range temperature prediction, and the remaining challenges in delivering reliable, actionable early warnings in a changing climate.

Reconstructing past climate variability requires computationally efficient models able to capture the behaviour of complex climate systems. We develop a suite of climate model emulators that improve the representation, reconstruction, and prediction of spatial climate variability compared to traditional approaches. Results highlight the importance of predictability-oriented representations and nonlinear dynamical memory for scalable emulators suited to paleoclimate data assimilation frameworks.

The accelerating effects of climate change, particularly rising sea levels resulting from melting land ice, require accurate predictions of ice sheet dynamics. Traditional numerical models, such as the Parallel Ice Sheet Model (PISM), are widely used; however, they rely on simplifying assumptions and are computationally intensive, limiting their ability to make long‐term predictions. This study addresses these limitations by employing machine learning techniques to emulate ice sheet dynamics, offering a faster, more efficient alternative to traditional numerical models. Using data generated by PISM for the Antarctic ice sheet spanning 2015 to 2100, we develop proof‐of‐concept machine learning models that predict key ice properties—ice mask, ice thickness, and ice velocity—based on climate inputs. Our models tackle both spatial and temporal challenges by incorporating the ice properties of neighbouring cells and past temporal values, thereby improving predictive accuracy. Additionally, explainable AI (XAI) techniques are applied to enhance trust and explainability. Our best‐performing model, a Random Forest with an expanding window, achieves an average RMSE of 3.36 for thickness and 65.6 for velocity and misclassifies on average only 3 grid points per year for the ice mask out of 2257, demonstrating good accuracy comparable to that of the original simulator at a fraction of the computational cost. This approach not only improves the computational efficiency of scenario modelling but also contributes to the broader field of climate science by enabling more frequent and extensive analyses of climate change impacts on Antarctic ice sheets.

We developed a machine learning emulator of a climate model simulating melting over Greenland that performs as well as the original model but it is much faster. We show that this emulator can be used as powerful tools to complement regional climate models by enabling computationally efficient ensemble simulations and physically interpretable attribution of past and future Greenland surface melt. Development of regional climate models should go hand in hand with ML-based tools.

Abstract. Fluvial valley networks in mountain ranges record the interactions between climate, tectonics, and lithology. While drainage network analysis has transformed our understanding of these interactions in subaerial settings, the landscape evolution of ice-covered orogens is poorly known. The Gamburtsev Subglacial Mountains are a ∼ 600 km-long mountain range situated beneath the East Antarctic Ice Sheet. These mountains were an important nucleation site for the ice sheet approximately 34 million years ago and are now buried beneath ∼ 2 km of ice. Airborne radar surveying has revealed that the Gamburtsevs are characterised by a rugged, incised landscape, but their geological structure and uplift history remain enigmatic. Here we use a compilation of radar survey data to extract and quantify valley longitudinal profiles from the Gamburtsevs and in turn infer details of their tectonic and geomorphic development. We use χ-mapping and stream power incision modelling to show that valley network morphology is consistent with a fluvial landscape that was locally overprinted by early mountain glaciation. The spatial distribution of channel steepness indices allows us to determine the position of major geological boundaries at the edges of the mountains. We also use independent estimates of denudation rates to evaluate competing scenarios for the timing of mountain uplift and valley incision, finding that uplift of the modern Gamburtsevs most likely commenced in the Mesozoic or early Cenozoic. Regional geomorphic analysis suggests that base level for some Gamburtsev fluvial catchments was set by enclosed interior basins associated with extensional faulting. These depocentres may retain detrital sedimentary material eroded from the Gamburtsevs prior to Antarctic glaciation and are potential targets for future sub-ice drilling campaigns.

Lake Balaton, the largest shallow lake in Central Europe, is highly sensitive to evaporation–inflow balance and provides an important archive of Holocene hydroclimate change. We reconstructed Holocene lake-level and moisture-availability changes using a high-resolution multi-proxy sediment record integrating age-modelled sediment stratigraphy, micro-XRF geochemistry, stable isotopes (δ 13 C, δ 18 O), grain size, XRD mineralogy and nanoscale TEM characterization of carbonate phases to identify the climatic drivers of hydrological change. Two major evaporative phases at ∼8050–7350 and ∼5450–4500 cal yr BP are marked by elevated Mg/Ca ratios and synchronous δ 13 C and δ 18 O maxima, indicating strongly reduced inflow, enhanced evaporation and substantially lower lake levels. The earlier interval represents the strongest evaporative phase in the record and followed the 8.2 ka event. The later interval reflects fluctuating hydrological conditions and alternating low- and high-stand phases during the onset of Neoglacial cooling. The first evaporative event coincided with post-8.2 ka summer warming linked to reinvigoration of the Atlantic Meridional Overturning Circulation (AMOC), whereas the second corresponded to widespread mid-Holocene hydroclimatic instability across Europe and the Mediterranean. This event was followed by the deepest lake phase between 4500 and 1900 cal yr BP. Our results demonstrate that Lake Balaton sensitively recorded both North Atlantic and Mediterranean hydroclimate forcings and provide new evidence for prolonged post-8.2 ka summer drought and mid-Holocene hydrological instability in Central Europe.

We introduce a new Earth system model experiment protocol to help researchers understand how Earth might respond to positive, zero, and negative carbon emissions. This protocol enables different models to be compared following similar warming and cooling rates. Researchers use the models to explore how the Earth reacts to different climate futures, including the risk of tipping points being exceeded and whether changes can be reversed. The results will support improved long-term climate policy.

Journal of Geophysical Research: Oceans, Volume 131, Issue 8, August 2026.

Geological archives provide the primary evidence for reconstructing Quaternary climate variability, yet differences among proxy records are commonly interpreted using broad archive categories. Such classifications can obscure the measurable controls governing the recording and preservation of climate signals. Here, a generalized quantitative framework is developed in which temporal resolution and signal fidelity depend primarily on accumulation rate, sampling integration scale, post-depositional mixing, diffusive smoothing, and ecological response time. Analytical formulations and numerical forward experiments show that increasing effective temporal integration attenuates abrupt events, broadens transition durations, suppresses high-frequency variability, redistributes variance toward longer timescales, and can generate apparent timing offsets between records responding to the same forcing. The framework further shows that uncertainty reduction in multi-record composites is limited when errors are correlated. Representative applications to Greenland ice cores, Chinese speleothems, and North Atlantic marine sediments show how readily available archive metadata can be translated into first-order estimates of effective temporal resolution, climate-signal attenuation, duration inflation, and preserved climatic bandwidth. These results indicate that Quaternary proxy records should be interpreted as filtered representations of climate variability whose characteristics depend on archive-response processes. Explicit integration of effective temporal resolution is essential for cross-archive comparison, proxy synthesis, and model-data evaluation.

Nature Geoscience, Published online: 03 August 2026; doi:10.1038/s41561-026-02054-6 Eukaryotic microorganisms within a subglacial brine-fed system in the McMurdo Dry Valleys of Antarctica are largely of marine origin, a legacy of past incursions of seawater into the area, according to RNA sequencing and metatranscriptomic analyses.

Abstract This study examines loess–palaeosol sequences (LPS) from three closely situated sites along a slope transect in Langenlois (lower Austria), focusing on palaeoenvironmental variability during Marine Isotope Stage 2 (MIS 2). One objective is to disentangle the pedosedimentary impact of local slope processes from that of regional or global climate changes to evaluate the potential of the profiles for regional environmental reconstructions. By integrating high-resolution colour, grain-size, and geochemical analyses, we assess local depositional dynamics and identify weak pedogenic signals that are macroscopically hard to distinguish in the field. Luminescence dating and relative stratigraphy place these findings into a chronological framework. The results indicate a pronounced landscape transition from tundra to polar desert around 24 ka, a tundra gley corresponding to GI-2, and a variety of smaller fluctuations during MIS 2. Drill core LAN 4 provides an LPS mostly unaffected by slope processes, retaining loess from the end of the last glacial dating to 16.2 ± 0.7 ka. The findings highlight the sensitivity of LPS in a transitional location, as well as the trade-off between signal intensity and increased sensitivity to local environmental and topographic influences. These weak pedogenic signals can be deciphered by applying a multiproxy, multi-profile approach.

Nature Geoscience, Published online: 31 July 2026; doi:10.1038/s41561-026-02060-8 Only 10% of organic carbon from thawing Pleistocene permafrost in the Arctic that is buried in marine sediments ends up re-entering the active carbon cycle, according to a carbon isotope analysis of sediments from off Qikiqtaruk (Herschel Island) in Canada.

Glaciers play an important role in shaping landscapes because of efficient bedrock erosion at their base. Understanding the contribution of glacial erosion to landscape evolution and the associated feedbacks between climate, erosion, and tectonics requires identifying the factors controlling glacial erosion. However, measuring glacial erosion rates is difficult due to the multifaceted combination of erosional processes occurring in glacial environments. Empirical validation of proposed controls on glacial erosion rates usually rely on indirect measurements of basin-wide sediment yields, which may include remobilized sediment and nonglacial sources of erosion that can obscure the contribution of glacial erosion. Here, we investigate glacial erosion using an alternative proxy: measurements of cosmogenic nuclides in bedrock samples ( n = 28) at a glacier in southeastern Alaska. We leverage the spatial arrangement of paired in situ 14 C- 10 Be measurements in bedrock to simultaneously determine the Holocene history of glacial cover and bedrock erosion rates. Measurements of erosion from a glacial landform suggest that quarrying rates are locally five times greater than abrasion rates. The observed spatial variability in abrasion rates (0.01 to 0.44 mm/y) supported by numerical ice-flow model results suggests that glacial erosion depends on a combination of sliding velocity and basal shear stress (“basal power”). Millennial-scale cosmogenic nuclide-derived erosion rates are an order of magnitude lower than sediment-derived erosion rates, which may suggest that sediment-derived values of bedrock erosion are overestimated due to timescale biases and/or contributions from nonglacial sources of sediment.

Preprint under review for ESSD (discussion: open, 0 comments) CMIP6-MedPlus is an open-access climate dataset for the extended Mediterranean region, developed from CMIP6 simulations to support climate impact studies. It provides daily precipitation and near-surface air temperature at 0.25° spatial resolution from 1985 to 2100. The dataset is generated through a two-step statistical downscaling framework, in which a convolutional neural network first refines global climate model output and quantile delta mapping is then applied to correct biases.

Geophysical Research Letters, Volume 53, Issue 15, 16 August 2026.

Nature Communications, Published online: 04 August 2026; doi:10.1038/s41467-026-76310-z Antarctic Peninsula surface melt reversed its decline, significantly increasing 3% year-1 from 2009–2022. This intensified melt is dictated primarily (80%) by surface energy budget thermodynamics, rather than changes in atmospheric circulation.

arXiv:2601.05841v3 Announce Type: replace Abstract: The increasing occurrence of extreme weather events since the beginning of the 21st century has led to the development of new methods to attribute extreme events to anthropogenic climate change. The way in which the extreme event is defined has a major influence on the attribution result. A frequently overlooked aspect concerns the temporal dependence of extremes. This study presents an approach for attributing complete time series during extreme events to anthropogenic forcing. The approach is based on a non-stationary Markov process using bivariate extreme value theory to model the temporal dependence of the time series. We calculate the likelihood ratio of an observational time series from ERA5 given the distributions as estimated from CMIP6 simulations with historical natural-only and natural and anthropogenic forcing scenarios. The spatial fields are condensed by the extremal pattern index (EPI) as a compact description of spatial extremes. In addition, the study examines the extent to which attribution statements about the occurrence of extreme heat events change when the effect of the mean warming is eliminated. The resulting attribution statement provides very strong evidence for the scenario with anthropogenic drivers over Europe, especially since the beginning of the 21st century. For central and southern Europe, the influence of anthropogenic greenhouse gas emissions on heatwaves could already have been proven in the 1960s using today's knowledge. There is no reliable signal apart from a general shift in the temperature distribution, neither in terms of the temporal dependence of extreme heat days nor in terms of the shape of the extreme value distribution.

arXiv:2602.13847v5 Announce Type: replace-cross Abstract: A central challenge across science and engineering is to build data-driven reduced-order models of turbulent dynamical systems that reproduce stationary statistics, predict responses to external perturbations, and remain practical for real-world applications. To this end, we introduce an abstract discrete-time formulation of turbulent dynamical systems with exact energy-conserving nonlinearities. Parameterizing this structure with neural networks yields stable, physics-constrained reduced-order models. We then use the fluctuation-dissipation theorem (FDT) to validate the emulators' forced responses from unperturbed data alone, testing models beyond stationary statistics. The FDT also identifies candidate direct causal links, which are used as regularization terms only when they pass validation. We first test the framework on two idealized models of geophysical turbulence: the proposed emulators reproduce stationary statistics and accurately predict responses to weak and strong forcings, despite being trained solely on unperturbed data. We then move beyond idealized systems to model tropical climate dynamics from reanalysis data, where data scarcity, partial observability, and sensitivity to choices of stochastic parameterizations become central challenges. The resulting physics-constrained model reproduces key statistics of the El Ni\~no-Southern Oscillation (ENSO) and qualitatively captures its cumulative responses to perturbations; augmenting it with a non-Markovian stochastic closure substantially improves quantitative agreement with the FDT benchmark. This response-validated model is then used to characterize long-term causal drivers of ENSO variability. The proposed methodology establishes a modular framework for stable reduced-order models capable of probing causal mechanisms in realistic, partially observed turbulent systems.

Abstract Surface air temperature (SAT) is a key determinant of the surface mass balance of the Greenland Ice Sheet, but its regional disparities, decadal drivers, and seasonally modulated responses remain elusive. Leveraging a novel high-resolution (0.1°×0.1°) station-based reconstruction of SAT over Greenland, this study investigates the SAT variability from 1950 to 2024 and its relationships with the Atlantic Multidecadal Oscillation (AMO) and the Pacific Decadal Oscillation (PDO). We find that annual mean surface temperature decreased significantly from 1950 to the mid-1990s, after which it shifted to a rapid warming, with strong interdecadal variability in all seasons. Spatially, warming has been stronger in northern Greenland than in southern Greenland, and the ablation zone exhibits significant year-round warming. The AMO is the primary driver of decadal‑scale temperature variability, especially in summer, while the PDO modulates this influence most effectively in autumn. Since the late 1990s, the concurrence of a positive AMO and a negative PDO has jointly enhanced anticyclonic conditions over Greenland, driving a sustained warming. These findings significantly improve the predictability of Greenland SAT, thereby supporting more accurate projections of future surface mass balance of the ice sheet.

Nature Communications, Published online: 29 July 2026; doi:10.1038/s41467-026-75648-8 Through satellite observations and numerical modelling, this study reveals intermittent regrounding beneath Pine Island Ice Shelf and shows it added little resistance to flow, but may have helped trigger rifting responsible for causing acceleration.

The response of ice sheets to warming is a critical unknown in projections of ongoing climate change. The Miocene Climate Optimum is one of the best analogs for modern warming. The response of the Antarctic ice sheet to Miocene warming, and subsequent cooling in the Mid-Miocene Climatic Transition, is recorded in the ANDRILL AND-2A sediment core. To accurately interpret these records of ice growth and loss, we must understand the distribution of time within the core. We develop a theory for statistical quantities that dictate time’s distribution in stratigraphy— completeness and the compensation time scale and length scales—for glaciomarine sedimentation. We hypothesize that the compensation length scale, which quantifies the relief on the sedimentary surface, is set by the vertical amplitude of grounding line motion. This value should be tens of meters for astronomically forced Miocene ice sheets. We develop a novel Bayesian age modeling methodology that probabilistically constrains hiatus duration throughout the stratigraphy. This method yields a new age model for AND-2A and constrains the compensation time scale—the longest hiatus resulting from sedimentary dynamics—to 197−478 k.y. This time scale exceeds the time scales of most astronomical forcings and most reported values for fluvio-deltaic systems. The corresponding compensation length scale is 29−71 m, which aligns with our hypothesis that astronomically forced grounding line motion sets the compensation length scale for glaciomarine sedimentation. Our new age model suggests complete Miocene sedimentation from ca. 16 Ma to ca. 14.5 Ma, followed by a decline in accumulation rates during the expansion of the Antarctic ice sheet during the Mid-Miocene Climatic Transition.

Study region The Vilcanota Urubamba Basin in southern Peru includes fragile wetland ecosystems that play a key role in mountain hydrology and support grazing for Andean communities. Study focus Mapping of wetlands variability is missing, limiting our understanding of their characteristics, seasonality and link with the cryosphere. We characterise wetland distribution, seasonality and persistence and evaluate their spatial association with glaciers and seasonal snow. Using Landsat 7 and 8 imagery, we build three-month seasonal land cover maps from 2013 to 2022 using a Random Forest classification and an Albedo Retrieval approach. New hydrological insights for the region Wetland area decreases by 38% from the end of the wet season (October to December) to the end of the dry season (July to September). Pixel transitions indicate that wetlands primarily transform to and from agricultural and pasture lands. Highly persistent wetlands are located above 4600 m a.s.l. and closer to glaciers than less persistent wetlands. We identified three wetland seasonal drying patterns. Basins with delayed and slow dry-out wetlands were more common at higher elevations but were not always in glacierised catchments, suggesting meltwater may maintain wetlands in the early dry season. We provide the first large-scale picture of wetland seasonality, and the basis for modelling the processes that sustain wetlands in tropical high mountains.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 8, August 2026.

Past oceanic characteristics can be reconstructed via δ 18 O in bivalve shells, enabling the study of climatic patterns that affected marine and human communities. Through the novel analysis of δ 18 O in Pinctada mazatlanica shells from a 14 C dated Holocene coastal camp we calculated Holocene summer SSTs of La Paz Bay, Baja California Sur, México. Presently summer δ 18 O of P. mazatlanica is −1.8 ± 0.2‰, varying between −1.3‰ and −1.9‰ during the last 8.4 ka. For the years 8413 BP, 7726 BP and 7046 BP, δ 18 O was 0.2–0.1‰ higher than current values, suggesting slightly colder sea surface temperature (SST), while in 7795 BP, values were 0.1‰ lower, suggesting slightly warmer SST. In 6936 BP and 2093 BP, values were 0.4–0.5‰ higher, suggesting significantly colder SSTs. Periods of low (high) SST were characterized by arid (humid) conditions and high (low) marine primary productivity (MPP), favouring human presence (absence) throughout the Holocene. Our results, like those of other paleoceanographic studies, align with orbital forcing and its effects on SST and MPP, which affected human settlement patterns at regional and global scales.

This article examines the structure, composition, and textural features of the Griva Sequence in the Volchya Griva section and the paragenetically associated relief. Additional data supporting the aeolian origin of the Griva Sequence and the Griva relief are presented. Optically stimulated luminescence and infrared stimulated luminescence dating of the Volchya Griva section indicate that the Griva Sequence formed during the last glaciation and fully corresponds to Marine Isotope Stage (MIS) 2. The findings confirm that aeolian relief formation and the intensification of aeolian processes occurred during cold glacial periods of the Late Pleistocene, when southwesterly winds prevailed. The textural and structural characteristics of the Griva Sequence are similar to those of Upper Pleistocene niveo-aeolian sands that formed in cold, periglacial, arid regions of the world and are distributed along the southern margin of the last glacial ice sheets from Northwestern Europe to Western Siberia and Alaska. The formation of a deflationary truncation and a system of mega-yardangs in the northern part of the Lake Chany basin indicates a significant intensification of aeolian processes at the transition between the deposition of the lower and upper members of the Griva Sequence. The orientation of the Griva relief, which coincides with the prevailing direction of modern winds, suggests that wind directions during the cold and warm epochs of the Quaternary in southern Western Siberia were similar. The complete stratigraphy of the Volchya Griva section, its correspondence to the last glaciation and MIS 2, the presence of the largest Eurasian mammoth faunal locality, its distinct lithological characteristics, and the relatively high degree of its study allow this section to be considered one of the reference sections for the Upper Pleistocene of Siberia. It should be regarded as the stratotype of the widely distributed Griva Sequence in this area, which is proposed as a separate Baraba horizon (formation) for the south of Western Siberia, comprising the Eltsovka and Bagan loesses and the Suma soil that separates them.

Journal of Geophysical Research: Earth Surface, Volume 131, Issue 7, July 2026.

We study how regional climate and climate impact indicators may respond to different emissions scenarios. Their possible outcomes are shaped by uncertainties in future emissions, global warming, regional effects of global warming, and the chaotic climate system. We introduce RIME-X (Rapid Impact Model Emulator Extended), an emulator that combines multiple tools and datasets to estimate probabilistically how any emissions path may influence regional outcomes.

South American Andean glaciers are the least well known in their futures. We can use numerical models to estimate the change of these glaciers to future climate change. However, options within the numerical model which effect the results are unknown. We vary these selected important options to understand their effect on the numerical model output across the South American Andes. We find that how the model approximates sediment friction, and sliding, to be important on model output.

Nature Geoscience, Published online: 30 July 2026; doi:10.1038/s41561-026-02040-y Co-occurring atmospheric and river heatwaves have tripled since the 1980s across the USA and Europe, and are projected to become the norm by 2100, according to model simulations.

Conventional geology interprets Pleistocene deposits, erosional landforms, transported erratics, buried tunnel valleys, and megafaunal remains within a framework of repeated glacial–interglacial cycles. Most modern young-earth creationist models instead place a single Ice Age after the Genesis Flood, commonly proposing approximately 500 years of ice-sheet growth followed by 200 years of melting. However, these post-Flood models face difficulties explaining how sufficient snowfall could accumulate over continental interiors while the continents remained warm and precipitation from heated oceans tended to fall as rain. This paper proposes an alternative model in which major ice sheets formed during the Flood itself and subsequently underwent a comparatively rapid post-Flood “Melt Age.” In this model, explosive release from the fountains of the great deep, volcanic activity, high atmospheric moisture, reduced solar heating beneath extensive cloud cover, and hail formation within towering storm systems produced rapid atmospheric cooling and floating polar ice caps during the inundatory stage of the Flood. Meanwhile, thick deposits of warm sediment and widespread igneous activity maintained elevated temperatures within the continental crust. During the recessive stage, differential vertical tectonics raised the continents toward the floating ice masses, grounding them and forcing pressurized Floodwater and sediment-laden subglacial flows across the emerging land surfaces. These catastrophic flows are proposed to account for deeply incised tunnel valleys, channelized erosion, transported megalithic erratics, and the removal of large volumes of Pleistocene sediment into offshore basins such as the Gulf of Mexico. Long-distance rock transport traditionally attributed to slow glacial creep is instead interpreted as the result of rapid subglacial debris flows. Pleistocene megafauna and Paleolithic human remains are interpreted as Flood-buried pre-Flood organisms rather than inhabitants of a prolonged post-Flood Ice Age. Following the grounding of the ice caps, residual continental heat drove rapid melting, filling subglacial channels and completing a short-lived Melt Age. This model places Pleistocene glaciation, erosion, sedimentation, and mass extinction within the final stages of a single catastrophic Flood year.

Reduced-complexity models are an important tool in climate science, helping us understand and estimate future climate change. We present the experimental protocol for the next phase of the reduced-complexity model intercomparison project, which aims to compare results from many such models to better understand their behaviour. This knowledge will guide how these models are developed and used in the future, including in the upcoming Intergovernmental Panel on Climate Change (IPCC) assessment report (AR7).

Abstract Accurate estimates of snow water equivalent (SWE) are critical for hydrologic forecasting and water resource management in snow-dominated regions. While in situ observations, such as those from the United States Department of Agriculture snow telemetry (SNOTEL) network, provide direct measurements, their sparse spatial coverage limits their broader applicability. Numerical models such as Noah Multi-Parameterization (Noah-MP) provide spatially continuous SWE estimates but are computationally demanding. This study develops a deep learning–based emulator to reproduce SWE simulated by Noah-MP using a Convolutional Long Short-Term Memory (ConvLSTM) network trained on meteorological forcings from the Weather Research and Forecasting model and augmented with static physiographic variables and remotely sensed snow cover and leaf area index data. Incorporating spatial coupling substantially improves performance relative to a pixel-wise Long Short-Term Memory baseline, increasing the fraction of grid cells with non-negative Nash–Sutcliffe Efficiency (NSE) by 15–27% during the test period. The emulator reproduces large-scale SWE patterns consistent with both Noah-MP and Snow Data Assimilation System (SNODAS) and maintains competitive NSE and root mean square error performance relative to SNOTEL observations. Seasonal sensitivity experiments yield high R 2 values up to 0.96 during accumulation and 0.97 during melt. Once trained, the emulator generates SWE predictions faster than the parent physics-based simulation, enabling rapid, high-resolution estimates that support real-time monitoring and ensemble forecasting.

Abstract Mid-late Holocene hydroclimate changes in the Thar Desert in northwest India have been linked with the Indus cultural transformation. Regional hydroclimate reconstructions present a spatially heterogeneous picture; in particular, the role of seasonality remains unclear. Here, we use multi-model simulations from the Paleoclimate Modelling Intercomparison Project Phase-4 (PMIP4) to evaluate the seasonality of mid-Holocene precipitation and mechanisms governing spatial moisture variability. Results show strong agreement with proxy data on strengthened summer monsoon across the Thar Desert during the mid-Holocene, with north receiving double the effective precipitation than south, which is synchronous with development of agriculture alongside continued foraging. High winter rainfall changes in the south as compared to north explains the spatial heterogeneity in paleoclimate records and possibly the distinct cropping strategies that emerged across the Indus occupied regions. These findings highlight the critical role of seasonal precipitation contrasts in shaping both environmental conditions and cultural adaptations in climatically sensitive regions.

Geophysical Research Letters, Volume 53, Issue 15, 16 August 2026.

Proceedings of the National Academy of Sciences, Volume 123, Issue 31, August 2026. SignificanceAntarctica is often viewed as remote from industrial pollution, yet we show that the Antarctic Peninsula has become a hotspot of mercury accumulation. By combining geochemical and isotope records from 16 sediment cores with an observation-...

arXiv:2509.04627v2 Announce Type: replace Abstract: Numerical weather prediction requires initial estimates of the atmospheric state. Since the atmospheric density field is intricately woven into the atmosphere's governing equations, advancing atmospheric density estimation will improve numerical weather prediction. However, current meteorological instrumentation cannot directly measure the atmospheric density field over large volumes. Existing techniques rely on sparse point measurements, limiting our ability to accurately estimate the three-dimensional atmospheric density field. One potential solution is to employ measurements of the atmospheric muon flux. Atmospheric muons are particles produced when energetic atomic nuclei (cosmic rays) collide with nuclei in the upper atmosphere, producing a shower of secondary particles (muons) that propagates to the Earth's surface. The surface atmospheric muon flux is known to be proportional to the local atmospheric density field, implying that this technique can be used as a measurement of atmospheric density. This study examines the potential for using atmospheric muon flux measurements to improve atmospheric state estimation via a case study of simulated atmospheric muon observations in the path of tropical cyclone Freddy. We show that improvement in data assimilation performance can be achieved using data from a relatively small astroparticle detector, well within the capabilities of existing astroparticle technology. We additionally show that the improvements to atmospheric state estimates associated with muon flux assimilation are at least partially unique to muon flux measurements, as comparable surface pressure point measurements do not reproduce a similar effect.

Understanding future rainfall is essential for managing floods and water resources in Australia. We tested different ways of helping a regional climate model better represent large-scale weather patterns during the extreme 2010–2011 La Niña event and compared the results with observations. The best approach produced more realistic atmospheric conditions and rainfall, increasing confidence in future climate projections and helping improve information for planning and climate adaptation

Nature, Published online: 04 August 2026; doi:10.1038/d41586-026-02426-3 Put health systems at the centre of European heatwave plans

Bubastis is one of the most important ancient cities of the southeastern Nile Delta, continuously inhabited from around 3250 B.C.E. until its decline in the Roman Period. Although recent investigations have significantly advanced our understanding of its geomorphological development and settlement history, the region's vegetation and agricultural history has remained largely unexplored. This study provides the first Holocene paleovegetation reconstruction for Bubastis by integrating phytolith analyses from 17 sediment cores with established lithostratigraphic and paleotopographic frameworks. The results show that during the transition from the Late Pleistocene to the Early Holocene, the landscape around Bubastis was dominated by grasses, shrubs and trees, with palms occurring only sporadically. With the establishment and expansion of the settlement from the Old Kingdom onward, vegetation patterns changed markedly. The appearance of cereal phytoliths in fluvial deposits ( unit II ) near the western Kom provides the earliest indication of localized cereal cultivation in the Early Old Kingdom, while higher cereal proportions in anthropogenic deposits ( unit I ) across the settlement—especially in area S1—reflect intensive cereal processing and storage within residential zones. Throughout the Late Period to Roman times, agricultural activities shifted increasingly to the settlement periphery as the urban area expanded and natural vegetation declined.

Earth's Future, Volume 14, Issue 8, August 2026.

arXiv:2608.01546v1 Announce Type: new Abstract: We present FESOM2-JAX, a Python re-implementation of the Finite-volumE Sea ice-Ocean Model (FESOM2) in JAX. The model retains the unstructured-mesh, cell-vertex finite-volume formulation of the original, runs unchanged from a laptop CPU to 256 GPUs, and is end-to-end differentiable. FESOM2-JAX is a code shadow of the Fortran model: a projection onto the Python ecosystem, translated with large language models and verified kernel by kernel against the original. It is built to lower the barrier to experimentation, from new numerics and parameterizations to gradient-based calibration and hybrid physics-machine-learning components, while remaining close enough to the original so that what is developed in the shadow can be transferred back. In a 1958-2019 hindcast at 1$^{\circ}$ equivalent resolution with identical physics and forcing, the mean states of the JAX and Fortran versions differ from each other by two orders of magnitude less than either differs from observations, and the two runs agree for six decades in global temperature, salinity, heat content, and sea ice. The complete 1$^{\circ}$ configuration fits on a single GPU, a node of four GH200 superchips integrates $\sim$113 simulated years per wall-clock day, and meshes of up to 7.4 million surface vertices ($\sim$5 km) scale to 128 GPUs. What limits the model is communication rather than arithmetic. What the shadow adds to the original is the gradient: a single reverse-mode pass through the full time loop returns the sensitivity of a model diagnostic to a parameter at every mesh vertex, verified against finite differences. To our knowledge, FESOM2-JAX is the first global ocean-sea-ice model of CMIP-class complexity written natively in a differentiable framework, and the first on an unstructured mesh.

arXiv:2512.04452v3 Announce Type: replace Abstract: NORi is a machine learning (ML) parameterization of ocean boundary layer turbulence that is physics-based and augmented with neural networks. NORi stands for neural ordinary differential equations (NODEs) Richardson number (Ri) closure. The physical parameterization is controlled by a Richardson number-dependent diffusivity and viscosity. The neural ODEs are trained to capture the entrainment through the base of the boundary layer, which cannot be represented with a local diffusive closure. The parameterization is trained using large-eddy simulations in an a posteriori fashion, where parameters are calibrated with a loss function that explicitly depends on the actual time-integrated variables of interest rather than the instantaneous subgrid fluxes, which are inherently noisy. NORi conserves tracers by design, uses realistic nonlinear thermodynamics, and demonstrates excellent prediction and generalization capabilities in capturing entrainment dynamics under different convective strengths, background stratifications, rotation, and wind forcings. NORi is shown to simulate the seasonal evolution of the boundary layer at Ocean Weather Station Papa with similar performance to the state-of-the-art two-equation k-epsilon closure. When implemented in a double-gyre simulation, it is numerically stable for at least 100 years, despite only being trained on two-day horizons, and can be run with time steps as long as one hour. Combining highly expressive neural networks with a physically grounded base closure proves to be a robust paradigm for designing parameterizations for climate models: data required and training cost are drastically reduced, inference performance can be directly optimized as a primary objective, and numerical stability is implicitly promoted through training.

Introduction Glaciers in the Garhwal Himalaya are highly sensitive to climate change and play a critical role in sustaining downstream water resources, ecosystems, and local livelihoods. This study evaluates glacier dynamics and thermal variability in the upper Bhilangana watershed, with a particular focus on the Khatling Glacier and its tributaries, while incorporating local perceptions of environmental change. Methods Multi-temporal Landsat imagery from 2010, 2013, 2016, 2019, 2022, and 2025 was analyzed to derive the Normalized Difference Snow Index (NDSI), Land Surface Temperature (LST), and debris-cover extent. Glacier and debris-cover boundaries were delineated through image interpretation and digitization. LST was retrieved using emissivity-corrected thermal infrared data and validated with observations from the Indian Meteorological Department (IMD). To complement the geospatial analysis, household surveys, transect walks, and focus group discussions were conducted across seven villages within the watershed. Results The analysis revealed a significant decline in snow cover, indicated by decreasing NDSI values, alongside increasing LST and the expansion of debris-covered glacier surfaces between 2010 and 2025. A moderate inverse relationship between NDSI and LST suggests that reductions in snow cover are associated with higher surface temperatures. Community-based observations were consistent with remote sensing results, with respondents reporting warmer summers, declining snowfall, reduced spring discharge, altered rainfall patterns, and increasing agricultural challenges. Discussion The strong agreement between remotely sensed observations and local perceptions highlights the importance of integrating scientific assessments with community knowledge to understand climate-induced environmental change. The findings demonstrate that ongoing glacier retreat and thermal variability are affecting both ecological conditions and livelihood security in the upper Bhilangana watershed. These results underscore the need for climate-resilient resource management, adaptive agricultural practices, and community-based monitoring systems to strengthen long-term adaptation strategies in Himalayan mountain regions.

Abstract. Predicting the trajectory of aquatic deoxygenation under global warming requires a mechanistic understanding of lacustrine responses to rapid climate shifts. We investigated how climate-driven changes in catchment vegetation and local iron-rich lithology regulated lake stratification and ecosystem resilience in the maar lake Holzmaar (Central Europe). We focused on the Late Glacial, specifically on transitions during Dansgaard-Oeschger Event 1 (DOE-1; ca. 14 690–11 700 cal yr BP), a period of rapid natural warming and cooling that serves as an analogue for future high amplitude climate variation and for modern Arctic lakes undergoing rapid climate-driven transitions. Combining non-destructive hyperspectral imaging (HSI) of sedimentary pigments with high-resolution XRF geochemistry, we resolved parts of the ecosystem trajectory during DOE-1. The primary producer community shifted from an oligotrophic cyanobacterial and low-light Pleniglacial assemblage to a stable, stratified Allerød assemblage, characterized by the planktonic diatom Stephanodiscus minutulus and anoxygenic purple sulphur bacteria (PSB) in the photic zone. While regional warming (mean summer temperature increased ∼ 2.8 °C) provided the physical potential for lake stratification, our data suggest that intense anoxia was primarily triggered by the expansion of Betula in the watershed. This afforestation stabilized the water column through wind shielding. The termination of the anoxic phase coincided with the onset of the Younger Dryas cooling and increased aridity, which effectively destabilized the existing stratification. While the shift from Betula to Pinus forest may have caused a change in the terrestrial-aquatic linkage, the primary driver of the transition was the physical forcing (lake mixing) of the climatic shift (cooling). Geochemically, the lake exhibited remarkable resilience. Unlike carbonate-dominated systems prone to internal phosphorus loading, Holzmaar efficiently sequesters nutrients via a dual mechanism of reactive iron binding (authigenic vivianite) and stable mineral burial. The phosphorus trap prevents nutrient release by permanently sequestering P in the sediment, allowing rapid ecosystem recovery without delay once the specific climate and vegetation drivers shift. Our findings demonstrate that in volcanic maar lakes, catchment vegetation characteristics and local lithology can modulate, and even override, the direct effects of climate warming on aquatic anoxia.

The environmental evolution of volcanic barrier-lake systems is shaped by complex interactions between climate forcing and local geomorphological processes. This study presents a high-resolution multi-proxy record from the sedimentary sequence of Pla de les Preses volcanic barrier-palaeolake (Vall d'en Bas, NE Iberia), spanning the Late Glacial–Early Holocene transition. Particular emphasis is placed on the contribution of non-pollen palynomorphs (NPPs) to palaeoenvironmental reconstruction. Pollen data indicate that regional vegetation dynamics broadly follow the expected transition from open Late Glacial landscapes to temperate deciduous forests during the Early Holocene. However, notable deviations occur, including a marked decline in broadleaf forests, suggesting a strong influence of local disturbance processes. Sedimentological and palaeobiological proxies document the formation of a volcanic barrier lake following the damming of the Fluvià River, and its subsequent evolution through three main stages: (i) shallow, unstable conditions with pioneer aquatic communities, (ii) progressive deepening and establishment of a persistent lacustrine system, and (iii) lake shallowing, wetland expansion and final terrestrialisation linked to barrier erosion and fluvial reactivation. Hydrological changes do not always align with regional climatic trends, indicating a decoupling between climate and local environmental processes. NPP assemblages provide key insights into these dynamics, recording variations in water depth, aquatic productivity, trophic conditions and lake–catchment interactions. Algae and cyanobacteria reflect limnological variability, while pteridophyte spores and fungal remains document wetland development, soil erosion processes and terrestrial inputs. In addition, NPPs capture short-term ecological responses to volcanic activity and fire. This study demonstrates the value of NPPs for reconstructing highly dynamic lacustrine systems, complementing pollen-based approaches and enabling a more detailed understanding of palaeoenvironmental change.

arXiv:2511.05392v2 Announce Type: replace Abstract: Traditional methods for improving tropical cyclone (TC) intensity from climate model outputs or projections have primarily relied on either dynamical or statistical downscaling. With recent advances in deep learning (DL) techniques, an important question is how DL can provide an alternative approach for enhancing TC intensity and structure retrieval from climate data. Using a common DL architecture based on convolutional neural networks (CNN) and a set of key environmental features relevant to TCs, we show that TC intensity and size can be effectively retrieved from climate reanalysis data without requiring super-resolution enhancement as in previous studies, even when applied to coarse-resolution climate data. This approach allows for retrieving TC intensity metrics and size that are dynamically constrained by the data, rather than estimating these quantities independently. Our results highlight that TC intensity and size are governed not only by TC internal processes but also by local environments during TC development for which DL models can learn and capture. The performance of our DL model depends on several factors such as season, the stage of TC development, or ocean basins, with root-mean-square errors ranging from $\approx$8-10 m s$^{-1}$ for the maximum 10-m wind, $\approx$10-13 hPa for minimum central pressure, and $\approx$13-21 km for the radius of the maximum wind. Although these errors are better than any direct vortex detection or statistical downscaling methods applied to the same data, their wide ranges also suggest that a 0.5$^\circ$-resolution climate data may contain limited TC information for DL models to learn from, regardless of model optimizations or architectures. Possible improvements and challenges in addressing the lack of fine-scale TC information in coarse-resolution climate reanalysis datasets are discussed.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 8, August 2026.

Abstract Mountain snowpack in the western US can be difficult to predict multiple months in advance, especially within the Colorado Rocky Mountains, where snowpack prediction varies widely by location and derives little skill from El Niño-Southern Oscillation (ENSO). Using a high-resolution snow-water equivalent depth (SWE) reanalysis dataset, we investigate how seasonal snowpack prediction using sea surface temperature (SST) in the North Pacific and North Atlantic Oceans varies across prominent crests and valleys in the Colorado Rocky Mountains. Possible atmospheric mechanisms are proposed to explain SST patterns associated with enhanced SWE using self-organizing maps. We find that southern areas of the domain receive significant prediction skill from SSTs within a portion of the tropical North Atlantic, and that this SST pattern is associated with the frequency of surface cyclone passages southwest of the domain. Northern areas, by contrast, receive significant prediction skill from SSTs further north in the North Atlantic—an SST pattern related to the strength of the North Pacific High. Prediction skill increases west to east, where a larger fraction of snow accumulation events coincides with the passage of surface cyclones. A forecast experiment is constructed using December SSTs to predict April 1 SWE, which produces an average correlation of 0.51 between predicted and observed SWE across the domain, and a maximum correlation of 0.75 on the eastern slope of the Colorado Front Range for the years analyzed. This promising result highlights the potential to predict snowpack in small mountain watersheds more accurately, which may benefit water management in these locations.

Abstract Robust hazard assessment at dormant stratovolcanoes requires high-resolution eruption chronologies, yet these are often limited by incomplete proximal deposits and poorly constrained ages. Speleothems can preserve tephra particles and volcanogenic trace element signals datable using high-precision 230 Th/U chronologies. However, accurate eruption timings are affected by transfer delays of volcanogenic signals in distal cave systems beneath thick epikarst. Here, we investigate Late Holocene (1094 AD –recent) stalagmites from a shallow cave proximal to Taranaki volcano (New Zealand), where cave geometry and location along the primary tephra dispersal axis provide an opportunity to test whether primary tephra preservation improves reconstruction of eruption timing and duration. Laser-induced breakdown spectroscopy mapping of Si, K, and Ti identified several particle-rich layers, of which two comprise pristine tephra based on clustered glass and titanomagnetite compositions as well as grain morphologies indicating narrow grain-size distributions and increased particle angularity. These incorporated tephra particles correspond to the Te Popo and Burrell episodes and were dated using 230 Th/U at 1196 AD ± 31 and 1568 AD ± 41, respectively. Each tephra layer comprises three distinct tephra-rich laminae, indicating multiple eruption pulses. Growth rates across tephra-bearing laminae indicate eruptive pulses lasted several months and were separated by short repose periods, suggesting that each episode lasted around 2.5 years. Particle-rich layers deposited in the late nineteenth–twentieth centuries contain non-volcanic detritus and heterogeneous grains, consistent with dust and soil-derived material mobilised during landscape modification. These results demonstrate that proximal speleothems can be used to improve volcanic hazard assessments by providing precise information on the timing and duration of volcanic eruptions.

Journal of Advances in Modeling Earth Systems, Volume 18, Issue 7, July 2026.

The cold, dry, high-altitude region of Ladakh in the Indian Himalaya relies nearly entirely on water from glaciers and snowmelt - hence its inclusion among the more climate-sensitive parts of the region. The combination of these effects—warming, changing precipitation, melting glaciers, shrinking snowpack and the increased frequency of extreme weather events—is altering the water balance in the area and threatening water security for the future

Journal of Advances in Modeling Earth Systems, Volume 18, Issue 8, August 2026.

Proceedings of the National Academy of Sciences, Volume 123, Issue 31, August 2026. SignificanceGroundwater extraction can cause both reversible and irreversible ground deformation, with the latter leading to permanent land subsidence and loss of groundwater storage capacity. Satellite observations of land surface deformation reveal that ...

Preprint under review for ESSD (discussion: open, 0 comments) This article combines measurements from an uncrewed aerial system, a tethered balloon system and a surface flux gradient system to study aerosol particles and meteorological parameters at the Arctic site Ny-Ålesund, Svalbard, in spring 2024. The vertical and horizontal observations from the ground to 1.3 kilometres above sea level show strong aerosol variability in correlation with specific meteorological conditions, providing a valuable basis for understanding Arctic aerosol processes.

arXiv:2511.19938v2 Announce Type: replace Abstract: We propose a novel framework based on Topological Data Analysis (TDA) to identify and track cyclonic and anticyclonic structures in the Northern Hemisphere. Using persistent homology applied to seven decades of daily sea-level pressure anomalies (1948--2023), we represent the atmospheric field as a cubical complex and compute sublevel- and superlevel-set filtrations. This approach allows us to identify 1-dimensional topological features (1-holes) that correspond to coherent pressure systems, which we term 1-cyclones and 1-anticyclones. The structural intensity of these features is quantified through their topological depth, while their dynamical evolution is followed using an optimal matching procedure based on the Wasserstein distance between consecutive persistence diagrams. Our results reveal robust seasonal patterns characterized by winter maxima and summer minima in total persistence, frequency, and spatial extent. We show that cyclonic activity is topologically more fragmented and intense, consistent with the seasonal deepening of the Icelandic Low, whereas anticyclones exhibit a heavier long-duration tail associated with persistent blocking episodes. Crucially, we demonstrate that TDA metrics can differentiate between distinct dynamical regimes of atmospheric blocking, distinguishing the quasi-stationary, ``frozen'' topology of the 2003 European heatwave from the more volatile and unstable configuration of the 2012 cold spell. Compared to classical geometric tracking algorithms, this framework provides an objective, multiscale, and noise-robust characterization of the atmospheric skeleton, offering a unified mathematical description of the organization and stability of mid-latitude circulation.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 8, August 2026.

The currently observed climate changes are most intense in the Arctic region. This study utilized some of the oldest existing meteorological measurement series in this region. Atmospheric pressure and anemological conditions were analysed and then compared with contemporary measurements. This showed that the weakest low and strongest high systems occur in the same months. Prevailing wind directions have also remained unchanged, while the northerly wind carries less frigid air than in the past.

Abstract To better understand the effects of anthropogenic forcings on tropical cyclone (TC) activity, we investigate the individual impacts of anthropogenic aerosols (AAER), greenhouse gases (GHG), and biomass burning aerosols (BMB) on TC frequency using CESM2 large ensemble simulations. Changes in TC genesis are estimated using a dynamic genesis potential index (DGPI), which reveals distinct impacts from AAER and GHG. Increasing GHG emissions result in a decrease and southward shift in DGPI over the North Atlantic and eastern North Pacific since 1940, whereas the time-dependent AAER emissions lead to a decrease and southward shift in DGPI before 1980s and an increase and northward shift thereafter. In addition, BMB emission increases lead to a slight increase in TC frequency in the North Atlantic. Various dynamic and thermodynamic parameters act cooperatively to change TC frequency associated with GHG and AAER over the North Atlantic, while the increase in DGPI associated with BMB is mainly due to the changes in dynamic variables. AAER effects largely dominate DGPI changes in earlier decades, but GHG effects become increasingly dominant in recent decades. Near-future projections, characterized by continued increases in GHG emissions and steady or declined AAER emissions, suggest decreased DGPI over the North Atlantic and eastern North Pacific driven primarily by GHG forcing. A comparison between DGPI and an alterative, thermodynamic GPI shows similar increases or decreases in TC genesis potential with quantitative differences. However, comparisons between CESM2, CESM1, and six CMIP6 models show considerable inter-model spread in DGPI response, highlighting model uncertainties and the need for caution when applying and interpreting multi-model ensemble means.

Journal of Geophysical Research: Atmospheres, Volume 131, Issue 14, 28 July 2026.

Abstract Climate change is affecting smallholders’ livelihoods and cultural landscapes in Konso, Ethiopia. The objective of this paper is therefore to identify the changes that have occurred in the livelihoods of smallholder farmers as a result of climate change and to assess the implications for the cultural landscape, recognizing that livelihood vulnerability and heritage sustainability are analytically distinguishable but empirically interlinked. We acknowledge that the cross-sectional design limits causal attribution of "changes over time," and findings primarily reflect associations and perceived trends rather than definitive causal pathways. We gathered data using a cross-sectional survey design together with interviews and focus group discussions. Quantitative data were empirically collected from 371 smallholders in three districts delineated as a UNESCO World Heritage site. Vulnerability indices were generated using the Livelihood Vulnerability Index approach of the Inter-governmental Panel on Climate Change. Principal component analysis was used to assign weights and create index for vulnerability-contributing indicators. The study revealed that high exposure indices were observed across the study districts which indicates that rainfall and temperature are extremely changing. Prolonged drought and erratic rainfall massively affect smallholders’ livelihoods as well as cultural heritages. The ecosystems across the study districts exhibited high sensitivity to climate change. The study also indicates that smallholder’s adaptive capacities were weakening due to the increasing effects of climate change and they are forced to focus on individual and short-term survival needs instead of collective and long-term heritage management. The high scores of the three major vulnerability contributing factors indicate that smallholders’ livelihood is vulnerability to climate change. This has serious implications for sustainable management of the UNESCO World Heritage site, especially in culturally rich but economically fragile areas such as Konso. This demands to take into account the changing needs and interests of smallholders living in the cultural landscapes and their need for better livelihoods. The findings underscore the need for tailored adaptation strategies to improve the livelihoods of smallholders in the Konso and the perpetuation of the unique cultural landscape registered as UNESCO World Heritage site. We recommend continuous environmental and cultural capacity-building programs particularly targeting the youth to encourage their active participation in the conservation of the Konso cultural landscape.

Geophysical Research Letters, Volume 53, Issue 15, 16 August 2026.

Journal of Geophysical Research: Atmospheres, Volume 131, Issue 15, 16 August 2026.

We present an implementation of the Discontinuous Galerkin approach, a numerical solver of the Euler equations (called BRIDGE (Basic Research for ICON with DG Extension)), which is in particular well suited for numerical models for weather and climate prediction and atmospheric research. Two widespread formulations of the Euler equations with different thermodynamic variables are compared by the inspection of idealised benchmark test cases to assess the properties of the Discontinuous Galerkin method.

Abstract. Over the past decades, snow cover trends in North America have been analyzed, providing vital information to the Global Climate Observatory System and other stakeholders about the looming signals of climate-driven snow declines. Detecting daily changes in snow parameters (e.g., depth of snow, extent of snow cover, and snow water equivalent) is, however, fraught with challenges, including internal variability unrelated to climate signals. We used GlobSnow’s passive microwave remote sensing data and a convolutional Siamese U-Net to track how snow water equivalent (SWE) changes daily over North America’s mid- and high-latitude regions. Daily changes in SWE were detected with F1-scores of 94.8% and 100.0% in locations where it was not trained, and 99.3% at the location where it was primarily trained; this suggests the model’s generalization potential to different climatologies and geographic locations. Using the model, we computed a similarity vector to compare SWE trends. We found that although lake-effect snowfall may be prevalent in the Great Lakes Basin during the winter months, the region consistently records the highest frequency of daily changes in SWE. Alaska, Yukon, and the Northwest Territories tended to have minimal daily changes in SWE, suggesting that latitudinal gradients may dominate changes in the snow regime and cryosphere’s processes in the warming climate scenarios.

Geophysical Research Letters, Volume 53, Issue 15, 16 August 2026.

Journal of Geophysical Research: Earth Surface, Volume 131, Issue 8, August 2026.

Journal of Geophysical Research: Oceans, Volume 131, Issue 8, August 2026.

Journal of Geophysical Research: Oceans, Volume 131, Issue 8, August 2026.

The North Atlantic Oscillation (NAO) is an important driver of year-to-year weather variability in Europe and North America, but its behavior on longer timescales is disputed. We present a new reconstruction of the NAO over the last millennium using water isotope proxy data from ice cores, speleothems, trees, and other archives. We find pronounced but irregular multidecadal to centennial variability. Better accounting for such variability in climate models is crucial to better project future climate.

Geophysical Research Letters, Volume 53, Issue 15, 16 August 2026.

Preprint under review for ESSD (discussion: open, 0 comments) This study presents a new method for estimating the three-dimensional geostrophic currents in the Mediterranean Sea from 10 to 2000 m depth using Argo float observations and satellite sea-level data. The method provides long-term, three-dimensional information on ocean circulation, enabling researchers to better understand climate variability, heat and salt transport, and changes in Mediterranean water masses.

Proceedings of the National Academy of Sciences, Volume 123, Issue 30, July 2026. SignificanceMolecular inverse design is a classically exponential problem that could greatly benefit from quantum computing. Here, we introduce the quantum ensemble variational optimization (QEVO), a quantum algorithm that encodes molecules into a ...

Proceedings of the National Academy of Sciences, Volume 123, Issue 30, July 2026. Recent analyses have suggested that the decadal rate of Arctic winter sea-ice extent decline weakened in the early 2020s, with 20-y trends becoming statistically insignificant. Here we show from more up-to-date observations that the exceptionally low 2025 ...

Proceedings of the National Academy of Sciences, Volume 123, Issue 30, July 2026. In the spring of 2026, anomalously low snow conditions in the western United States threatened winter recreation and water supplies. Here, we investigate: to what extent was this snow drought attributable to the climate change that has occurred since the ...

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 7, July 2026.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 7, July 2026.

Seasonally frozen ground strongly influences land–atmosphere interactions, yet its simulation remains uncertain in land surface models. This study evaluates 2 standalone LSM simulations forced by reanalysis data (1986-2022) and validates results against reference observations at 26 Russian sites. Results show contrasting biases linked to snow insulation and freeze–thaw processes, highlighting the need for improved snow and soil parameterizations.

Nature, Published online: 29 July 2026; doi:10.1038/s41586-026-10840-w Climate and econometric modelling reveal that human-induced climate change is shifting the burden of childhood malaria in Africa.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 8, August 2026.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 8, August 2026.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 8, August 2026.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 8, August 2026.

Nature Climate Change, Published online: 30 July 2026; doi:10.1038/s41558-026-02694-x Climate change communication differs across regions, shaped by unique political, economic, social and environmental conditions. These influence how climate messages are framed, received and acted upon by the local public. To explore this diversity, Nature Climate Change spoke to globally distributed scholars for their perspectives on the communication patterns, challenges and priorities in their regions.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 8, August 2026.

Coccolithophores are vital to marine ecosystems and play a central role in the global carbon cycle. Although abundant in the temperate North Atlantic and Barents Sea, their long-term bloom dynamics remain poorly understood. Using two decades of satellite observations, we reveal a substantial decline in coccolithophore blooms between 2003 and 2022, with a 36.1% reduction in bloom area and a 51.2% decrease in bloom frequency. The decline is most pronounced in the temperate North Atlantic and western Barents Sea, while the eastern Barents Sea shows notable increases. Decreasing sea surface temperatures (SSTs) in the North Atlantic and warming in the eastern Barents Sea are identified as key drivers of these trends. However, SST changes alone do not explain the asymmetric patterns across the Barents Sea, suggesting additional environmental drivers and phytoplankton competition. Our study provides the most spatially and temporally resolved assessment of coccolithophore bloom trends in this region to date, with broader implications for marine ecosystems and carbon cycling.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 8, August 2026.

This paper presents data collected at the Zarafshon River Basin and the Hydrographic Party Glacier (GGP), Tajikistan. In annual field expeditions from 2018–2025, data has been collected on the glacier, as well as including meteorological measurements and aerosols. Presented results highlight the environmental conditions that are driving environmental change in Central Asia.

Abstract. Ground based time lapse cameras provide continuous, high frequency observations of glacier dynamics; however, automated analysis of these image streams remains challenging due to fog, snowfall, lens contamination, and variable illumination. This study investigates the potential of adapting the foundation segmentation model Segment Anything Model 2 (SAM2) for glacier segmentation from ground-based monitoring. To enable integration into automated pipelines, SAM2 is configured in image mode with a learned prompt generation strategy, while fine-tuning is restricted to the prompt encoder and mask decoder. In addition, the internal Intersection over Union (IoU) prediction head is utilized as a confidence estimator to assess segmentation reliability. Experimental results demonstrate that the adapted model achieves stable segmentation under moderate environmental variability, while degrading under severe visibility loss. This stability is consistent across model scales and input resolutions. The confidence estimation further provides a meaningful signal for identifying uncertain predictions, supporting reliability-aware processing in downstream workflows.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 8, August 2026.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 8, August 2026.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 8, August 2026.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 8, August 2026.

Arctic clouds containing both liquid droplets and ice crystals are difficult to simulate. We tested how ice-related changes in a weather model, designed for temperate regions, perform in the Arctic. Ice crystal shape is the dominant factor: making crystals spherical substantially reduces cloud ice and shifts it to snow. These modifications are moderate in temperate regions but stronger in the Arctic. As this rests on a single case, they should be validated across diverse cases and climates.

arXiv:2608.02205v1 Announce Type: new Abstract: The wintertime Arctic boundary layer is characterized by two distinct states, radiatively clear (cloud-free) and radiatively opaque (cloudy), which is reflected in observations as a bimodal distribution. Previous studies hypothesized that these regimes are related to an underlying bistability of the system's dynamics. We apply a drift-diffusion framework based on the two leading terms of the Kramers-Moyal expansion to observations from 15 winter periods in Ny-Alesund, Norway. Our model captures the bimodality; we find a single stable point for the deterministic drift, however, which questions the existence of bistable deterministic dynamics. Instead, the observed regimes can be attributed to state-dependent stochastic fluctuations with maximum amplitude close to the clear-cloudy transition. The position of the stable point depends on the meteorological conditions. Controlling for synoptic variability does not remove the state-dependence of diffusion, however. The cloudy regime is associated with southwesterly winds, high specific humidities, and uplifting motion, the latter two being conducive for cloud formation, while the cloud-free regime occurs under drier conditions and northeasterly winds. Through an illustrative example, we suggest a possible process-based interpretation of our statistical analysis. The nonlinear onset of cloud formation, once atmospheric humidity reaches saturation, could map state-independent fluctuations in humidity to state-dependent fluctuations in cloudiness.

arXiv:2608.01864v1 Announce Type: cross Abstract: Predicting drought risk is essential for anticipating impacts on water resources, agriculture, ecosystems, and climate adaptation planning. Yet drought forecasts remain uncertain because variability can substantially alter regional precipitation and evaporative demand. Treating this variability as unstructured noise ignores the fact that internal variability has spatial, seasonal, and temporal structure and thus contains information that can be used to improve drought forecasting. We propose a deep-learning-based forecasting framework for European drought prediction and extend it with an uncertainty-aware drought bound that explicitly incorporates internal forecast variability from a large climate model ensemble. This bound represents a physically plausible lower-tail trajectory of future drought conditions and marks how severe drought could plausibly become under an unfavourable realisation of internal variability, giving adaptation planning a conservative, risk-averse reference. We compare the proposed bound with a lower bound derived from reanalysis data only and show that our proposed ensemble-informed bound is better calibrated across most regions and seasons. This is specifically true during anomalously dry conditions, when historical reanalysis alone underestimates lower-tail drought risk. Our results show that internal variability should be treated as a forecast quantity in its own right. More broadly, large ensembles provide a practical way to transfer physically plausible climate variability into machine-learning drought forecasts, yielding risk-aware bounds that are more informative for drought assessment under shifting climate conditions.

Warm spells are becoming an increasingly important type of extreme event under climate change, particularly in mountain regions where they strongly influence snowmelt processes. This study analyses winter warm spells (WWS) in the Pyrenees over 1960–2024, focusing on their temporal evolution, associated impacts, and large-scale atmospheric drivers. Results show that WWS have become both more frequent and more intense, based on modelling S2M (1960–2024) and cross-validation with homogenized and reconstructed spatial interpolations of maximum air temperature in the Pyrenees CLIMPY (1981–2020), showing consistent positive WWS trends. Their occurrence has increased by approximately 0.4 days per decade since 1990, while maximum air temperature during events has risen significantly (0.24 °C per decade), with stronger trends in the southern Pyrenees. These events substantially enhance snowmelt, increasing mean daily melt from ~1.3 mm day −1 under average winter conditions to ~2.2 mm day −1 during WWS (~65% increase). More than 50% of WWS are associated with circulation weather types characterized by anticyclonic southwesterly flow, which promotes subsidence and the advection of warm air masses. The increasing frequency of these circulation weather types during 1960–2024, together with thermodynamic warming under the same synoptic conditions, demonstrates that the observed intensification of WWS results from the combined influence of dynamic and thermodynamic processes. These results have important implications for winter snowmelt and for the environmental and socioeconomic sectors that depend on seasonal snow cover.

Nature Climate Change, Published online: 04 August 2026; doi:10.1038/s41558-026-02680-3 Social media have become an important arena for climate change communication. This Review synthesizes climate-related content on social media, its effects on climate attitudes, knowledge and behaviour, and its role in reshaping the broader communication landscape.

Nature Communications, Published online: 05 August 2026; doi:10.1038/s41467-026-76445-z Higher-order topology has been well established in insulators and semimetals. Here, authors evidence higher-order topology in metallic systems, and unveil a type of topological corner states that are naturally embedded in the bulk continuum with no need for intentional engineering.

Extreme sea levels during storms can lead to coastal flooding and sediment displacement. In the paper, we present an expanded global dataset of sea levels, produced using a global ocean model driven by ERA5 reanalysis – a state-of-the-art dataset of historical weather and climate. The strengths and limitations of the dataset are discussed by comparing it to observed sea levels. The dataset covers a period of 1950–2024, improving our understanding of the global frequency of extreme water levels.

This study addresses a gap in understanding how turbulent mixing closure schemes and convective adjustments interplay in the Mediterranean Sea. Coupled ocean-wave simulations were performed with different mixing parameterizations and model results were compared against Argo float observations across different space and time scales. Results show that the Generalised Length Scale closure scheme best reproduces observed mixed layer properties and variability, without needing convective adjustment.

Global water models help us understand how human activities and climate change affect water resources. One of them is the CoSWAT Global Model. In this study we enhanced this model by adding a better representations of lakes, reservoirs, and irrigation demand. Evaluation shows these changes improve river flow simulation and enable explicit assessment of lake and reservoir water balances, producing a more robust tool for global freshwater studies.

Nature, Published online: 28 July 2026; doi:10.1038/s41586-026-10923-8 Editorial Expression of Concern: Carcinoma-produced factors activate myeloid cells through TLR2 to stimulate metastasis

The spatial variability of snow depth distribution in avalanche release areas is key for avalanche forecasting but is strongly influenced by the interaction of wind with the terrain. We generate maps of high resolution snow depth changes during three snowfall events by using low-cost terrestrial laser scanner measurements and compare them to the results of selected snow depth distribution models. We show that basic terrain derivations have the highest correlations to our measurement results.

Sea-dumped munitions are a threat to economic activities at sea. This study presents the first comprehensive dataset of munition piles in the German Baltic Sea. It was created by mapping, annotating, and analysing three major dump sites. The results show that the piles differ greatly in their size and composition. Many piles lie close to protected areas and a sediment dump site. The dataset can aid discussions on monitoring, risk assessment and prioritisation of piles for clearance.

Evapotranspiration (ET) describes the transfer of water from land to the atmosphere and is fundamental to understanding agriculture, ecosystems, and drought. We implemented the established DisALEXI model on Google Earth Engine (GEE-DisALEXI), enabling scalable, high-resolution ET mapping over large regions. This paper evaluates this version's accuracy, highlights example applications, discusses limitations, and outlines opportunities for future improvements.

In this paper, we present a global high-resolution hydrological model to investigate how water may have once flowed and accumulated on Mars. Using detailed topography, the model tracks how lakes and seas form, grow, merge, overflow, and dry out over time. It reveals how a vast northern ocean could emerge from smaller bodies of water. This approach links surface landforms to past climates, offering new perspectives on Mars' watery history and its potential habitability.

We present a continuous dataset of carbon and energy fluxes measured over an alpine steppe in the Tibetan Plateau using the Eddy Covariance technique at 19 m. Covering a larger footprint (~ 30 ha) than the existing long-term 3 m measurements, it enables landscape-scale flux estimation, improved alignment with satellite observations, and supports studies of ecosystem modelling and satellite product validation.

Due to the limitations of past data-based models and the high cost of numerical weather prediction computing, accurately forecasting precipitation proximity remains challenging. A dual encoder-decoder framework is proposed to enhance short-term forecasting and reduce underestimation in extreme precipitation by using spatio-temporal information conversion equations and adaptive weighted gradient loss. Demonstrates better accuracy than existing deep learning methods in precipitation datasets.

Proceedings of the National Academy of Sciences, Volume 123, Issue 31, August 2026. We analyze and decompose administrative data on all 1.6 million US Immigration and Customs Enforcement (ICE) arrests from October 2015 to March 2026. Our results reveal that the reality of immigration enforcement diverges sharply from the public narrative ...

Proceedings of the National Academy of Sciences, Volume 123, Issue 31, August 2026. SignificanceFlowers use scent to attract pollinators, yet animals are not known to produce floral chemical signals to manipulate these interactions. Here we show that larvae of the parasitic blister beetleMeloe proscarabaeuschemically mimic flowers by ...

Nature, Published online: 04 August 2026; doi:10.1038/d41586-026-02358-y Diseases don’t stop at borders. Pandemics, climate change and conflicts demand a collective approach.