Centennial-scale changes in Atlantic Meridional Overturning Circulation (AMOC) strength might disturb global climate by altering interhemispheric heat transport and CO2 partitioning between the ocean and the atmosphere. Due to the short instrumental record and lack of high-resolution paleo records that so far only resolve millennial-scale changes, centennial-scale changes remain elusive. Here we use radiocarbon ventilation ages from a western equatorial Atlantic sediment core with a remarkably high sedimentation rate to reconstruct AMOC variability during the last deglaciation, with a focus on Heinrich Stadial 1 (HS1; 17.8-14.8 ka). Results from model simulations indicate that ventilation ages serve as a sensitive proxy for AMOC variability under weak overturning conditions (i.e., during HS1) at our study location. Notably, within an overall weakened AMOC condition during HS1, our record shows two centennial-scale AMOC intensifications: one from 16.5 to 15.8 ka and another at ~15.4 ka. These centennial-scale episodes of intensified AMOC briefly revitalized Atlantic interhemispheric heat transport during HS1, resulting in decreased precipitation over northeastern Brazil and short-lived but intense changes in climate elsewhere. These episodes of AMOC intensification likely transported substantial volumes of CO2-rich water from the mid-depth Atlantic to the Southern Ocean, where the CO2 was rapidly outgassed to the atmosphere. Using radiocarbon ages from Atlantic sediments, AMOC variability during Heinrich Stadial 1 is reconstructed, revealing brief intensifications that altered heat transport, reduced rainfall over northeastern Brazil, and likely enhanced atmospheric CO2.

Abstract Sea-level reconstructions and numerical models suggest that the vulnerable West Antarctic Ice Sheet underwent major mass losses during the late Pleistocene, raising global sea level by several metres. However, direct evidence from ice-sheet proximal records for such retreat has remained inconclusive. Here we investigate variations in clay mineral assemblages and past ocean temperatures archived in West Antarctic continental margin sediments during the last 800,000 years. Our results demonstrate that the supply of glacigenic detritus sourced from the main outlet glaciers in the Amundsen Sea sector of the West Antarctic Ice Sheet ceased almost entirely during all previous warm periods, documenting the loss of their drainage basins. Concomitant seawater warming, revealed by a microfossil proxy, is attributed to intensified deep-water advection towards Antarctica’s Pacific margin, which is corroborated by the clay mineralogical evidence. Our findings show that ocean forcing can drive major deglaciation of West Antarctica and related global sea-level rise.

Abstract As anthropogenic emissions continue to increase Earth’s surface temperature, the Atlantic Meridional Overturning Circulation (AMOC) is at risk of weakening substantially and crossing a tipping point. Then, multiple Earth system components may be significantly and irreversibly altered. Here we investigate the responses of ocean carbon uptake, net primary productivity (NPP), oxygen concentration, and carbon storage to an AMOC collapse induced by extremes in the internal variability in the subpolar gyre under a moderate forcing scenario. While similar ensemble splitting has been found in other models, the NASA Goddard Institute for Space Studies (GISS) Earth system model (ESM) was the first to report an AMOC bifurcation due to internal variability, without the need to prescribe freshwater hosing. We find severe regional impacts, including collapsed NPP and deoxygenation in the subpolar North Atlantic, and reduced carbon uptake in the Arctic and subtropical North Atlantic. We identify the physical drivers of these ocean biogeochemical impacts and, for the first time, provide estimates of their lag and latency, i.e. the timescales over which these impacts begin to occur and are fully realized, respectively. Timescales are generally found to increase southward from the deep convection sites, but are remarkably similar between basins at similar latitudes.

Greenland ice-core records show an approximately 15-fold increase in dust concentration during cold stadial periods compared to warmer interstadial periods in the last glacial period (LGP). While multiple regions including East Asian deserts and the Sahara have been suggested as potential source areas (PSAs) for Greenland dust, millennial-scale variability of their relative contributions associated with Dansgaard–Oeschger climatic fluctuations during the LGP remains poorly constrained. To address this, we investigate high-resolution (~800 y) Pb and Sr isotopic, mineralogical, and geochemical records from the North Greenland Eemian Ice Drilling ice-core samples covering the entire LGP. We found that the Pb and Sr isotopic compositions, kaolinite-to-chlorite ratios, and chemical index of alteration values were higher during stadials than during interstadials. Comparison with PSAs suggests that East Asian deserts remained the dominant contributors to Greenland dust throughout the LGP, but contributions from the Sahara increased during stadial intervals. This shift likely reflects climate-driven changes in atmospheric circulation patterns responsible for transporting dust to Greenland. Our findings provide insights into the relationship between dust cycle dynamics and abrupt climate changes on millennial timescales during the LGP.

The polar regions play a central role in the Earth system through their influence on ocean circulation, ice sheets, sea level, carbon cycling, and climate variability. Rapid anthropogenic warming is transforming these regions, highlighting the need to understand the processes, feedbacks, and thresholds that govern ocean–cryosphere interactions across a wide range of timescales. Here we synthesize major scientific advances achieved at the Alfred Wegener Institute during 2021–2025 within the Helmholtz research program Changing Earth – Sustaining Our Future. Combining paleoclimate archives, sustained observations, process studies, and Earth-system modeling, these studies provide new insights into the sensitivity of the climate system under past, present, and future climate conditions. Key findings include new constraints on Antarctic Ice Sheet evolution during past greenhouse climates, improved understanding of links between Southern Ocean circulation and climate–carbon feedbacks, advances in reconstructing Northern Hemisphere ice-sheet dynamics and deglaciation processes, and evidence that regional climate variability is stronger and more persistent than represented in many current climate models. Observations and simulations further reveal accelerating changes in Greenland outlet glaciers, important couplings between cryospheric change and the carbon cycle, and a growing role of mesoscale ocean dynamics in shaping future Arctic and global climate change. Together, these results demonstrate the value of integrating observations, paleoclimate evidence, and high-resolution modeling to constrain Earth-system sensitivity, identify critical feedbacks, and improve projections of future climate change in a warming world.

Ice shelf melt rates by a warm ocean have a critical impact on ice sheet mass balance in Antarctica and in turn on the rates of sea level rise. Validating and tuning modeled ice shelf melt rates is a challenge. Here we discuss the issues by presenting validation of melt rates calculated from the Southern Ocean high-resolution model (SOhi) over the year 2007. SOhi includes full physics, tides, and an updated bathymetric compilation. We compare the model results to two satellite-derived estimates. Overall, we find that the model overestimates ice shelf melt, which we attribute to the ocean model initial (warm) conditions in some areas, bathymetric uncertainties in others, simplified physics on some ice shelves, and also to the inherent limitations of the satellite-derived estimates. The SOhi model nevertheless provides a comprehensive, detailed mapping of ice shelf melt rates around Antarctica; it captures higher melt near grounding lines and ice fronts, areas of freezing versus areas of high melt water production. The identified discrepancies between model and satellite-estimates provide guidelines for model improvements, requirements for additional observations of ice shelf cavities, and assembling of calibration/validation datasets to constrain the models.

Earth’s climate oscillations between interglacial and glacial periods have driven significant sea level changes. Sri Lanka, located in the Indian Ocean, provides an ideal natural setting for studying paleo sea level changes. This study compiles 92 Holocene sea level records and identifies last interglacial stratigraphic evidence from Point Kudrimalai to reconstruct paleo sea level fluctuations in Sri Lanka. In addition, this study evaluates underutilized sea level indicators such as beachrock, coastal barrier sands, and mangrove swamps to address methodological challenges in reconstructing past sea levels. Results show that diverse sea level indicators, including beachrock, coastal barrier sands, raised marine deposits, and marine shell beds, primarily reflect Holocene sea level changes and their geological impacts. Using standardized sea level indicators such as coral and marine shell beds, this study reveals Holocene sea levels approximately 1.0–2.0 m above present between 6000 and 2000 cal BP. The 6–9 m elevation associated with Point Kudrimalai can possibly be correlated with regional Pleistocene sea level highstands. In contrast, the Palk Strait land bridge can be identified as a potential paleogeographic connection during periods of substantially lower sea level. Accordingly, these findings elucidate that the impact of sea level changes in Sri Lanka has likely influenced coastal geomorphology. Therefore, this reconstruction enhances understanding of regional sea level variability and its role in shaping coastal landscapes in the Indian Ocean.

Abstract Growing observational and modeling evidence suggests that the Atlantic Meridional Overturning Circulation (AMOC) has weakened since the 1980s, with reductions in North American and European (NAEU) aerosols identified as a potential contributing factor. Here we use fully coupled climate model experiments to demonstrate that the radiative forcing associated with decreased NAEU aerosols weakens the AMOC by enhancing freshwater transport from the Arctic and subtropical regions into the deep convection region. Reduced aerosol loading induces large-scale atmospheric circulation adjustments, including cyclonic wind stress anomalies over the Arctic Ocean and the subtropical North Atlantic. Over the Arctic Ocean, the cyclonic wind stress anomaly weakens the Beaufort Gyre, releasing freshwater that accumulates north of the Canadian Arctic Archipelago and Greenland. This freshwater is then exported through Baffin Bay into the Labrador Sea, suppressing deep convection there. Simultaneously, enhanced cyclonic wind stress over the subtropical North Atlantic weakens the North Atlantic Current (NAC), reducing the northward transport of saline water and promoting the freshwater accumulation within the subpolar gyre. This freshwater anomaly is then advected toward the Labrador Sea, further suppressing deep convection and weakening the AMOC. Together, our results identify a remotely driven pathway through which reductions in NAEU aerosols weaken the AMOC via wind-driven freshwater advection. The robustness of this mechanism is further supported by the multi-model ensemble mean from the Coupled Model Intercomparison Project Phase 6 (CMIP6) simulations, which exhibits consistent oceanic and atmospheric circulation responses.

The buttressing effect of landfast sea ice is crucial for the stability of Antarctic outlet glaciers. This study investigated spatiotemporal variations in flow velocity and ice thickness at the Shirase Glacier, East Antarctica, from 2016 to 2025 using ALOS-2/PALSAR-2 and CryoSat-2/SIRAL data. Flow velocity increased, and ice thickness decreased downstream, exhibiting a significant negative correlation (r = −0.66, p < 0.05). Notably, during periods of unstable landfast ice (2017–2018 and 2023–2025), substantial ice flow acceleration occurred in the downstream ice tongue. Furthermore, basal melt rates calculated using the Eulerian flux divergence method revealed high melting (16–20 m a−1) near the grounding line, decreasing to 2–4 m a−1 mid-stream, and increasing to 4–8 m a−1 near the ice front. This distribution is consistent with the subglacial oceanic vertical thermal structure. These results quantitatively demonstrate the significant control of both landfast ice stability and subglacial ocean circulation on the spatiotemporal dynamics of the Shirase Glacier.

This study provides a statistical characterization of surface oceanography in the Indonesian Maritime Continent (IMC) during the Last Glacial Maximum (LGM; 23–19 ka) and Late Holocene (LH; 4–0 ka). Using the lgmDA v2.1 paleoclimate data assimilation product, we analyze sea surface temperature (SST), sea surface salinity (SSS), and seawater $\delta^{18}$O. Results indicate the LGM was significantly cooler, saltier, and isotopically enriched, with reduced spatial heterogeneity compared to the LH. Comprehensive nonparametric testing confirms these differences reflect fundamental changes in distributional properties, not merely mean-state shifts. The findings suggest a major oceanographic regime shift, likely driven by lower sea levels, a weakened ITF, enhanced stratification, and a drier regional climate during the LGM. This work offers a statistical framework for understanding IMC sensitivity to glacial-interglacial forcing.

Abstract. The vast majority of studies examining the impact of freshwater from ice sheet melting on the Atlantic Meridional Overturning Circulation (AMOC) use climate models that cannot resolve mesoscale ocean processes and do not include an accurate spatio-temporal distribution of the freshwater forcing. These two factors critically affect the nature of the AMOC response. Our study fills that gap with a set of three hosing experiments performed with the global configurations of the eddy-rich climate model EC-Earth3P-VHR. The model is forced for 21 years with a spatial and monthly distribution of Greenland meltwater fluxes derived from observations, equal to 0.04 Sv on an annual average. Within the first year, we observe a response of reduced salinity in the Greenland and Labrador currents. This is accompanied by an acceleration and a cooling along the currents that lead to a rapid weakening of the AMOC at subpolar latitudes. Around year 7, deep mixing in the Labrador Sea begins to weaken due to as freshwater anomalies accumulate through lateral exchanges with the boundary currents. This shallowing of the mixed layer further weakens the AMOC, resulting in a stronger reduction that reaches also the subtropical latitudes. By the end of the simulation, the AMOC has weakened by almost 3 Sv at subppolar latitudes (i.e. a decrease of around 20 %), with an average relative decrease of 10 % for the whole Northern Hemisphere. The reduction in the AMOC is strong enough for some global climate impacts to emerge, such as the “bipolar seesaw” temperature response.

We studied how the Antarctic ice sheet surface mass balance may change by 2100 using three high-resolution climate models forced by the same future climate scenario. While the models agree for present-day conditions, they project very different futures, especially over floating ice shelves. These differences mainly come from how melting, refreezing, and temperature are represented. Our results show that future sea level projections strongly depend on how well models simulate today’s climate.

The stability of the Atlantic Meridional Overturning Circulation (AMOC) is a major uncertainty in projections of future climate. However, direct evidence for AMOC tipping behavior on orbital timescales remains scarce. Here, we reconstruct the strength of Iceland-Scotland Overflow Water, a key lower branch of the deep Atlantic circulation, over the past 800,000 years by combining 12 current-speed records from North Atlantic sediment cores with a data-driven hybrid model. The resulting reconstruction reveals persistent alternation between weak (glacial) and strong (interglacial) circulation modes, pervasive millennial-scale variability within both regimes, and systematic asymmetry between weak-to-strong recoveries and strong-to-weak collapses. Threshold and phase-space analyses show that regime shifts occur within a narrow window of atmospheric carbon dioxide and sea level and often follow hysteresis-like loops. Although bistability persists throughout the last 800,000 years, its sensitivity and timing evolve through time. These findings provide long-term, high-resolution, data-driven constraints on the bistability of Nordic Seas overflow waters and its implications for AMOC resilience in a warming world.

Dansgaard-Oeschger (DO) events during the last ice age are thought to reflect rapid shifts in the Atlantic Meridional Overturning Circulation (AMOC). Greenland ice cores suggest warming transitions within only a few decades—or even just over 10 years. Using precisely dated speleothem oxygen-isotope records from the Asian monsoon region, we have shown that the same DO warmings unfolded over more than a century. Isotope modeling revealed that Greenland records primarily capture a rapid atmospheric reorganization triggered by the initial sea-ice retreat in the northeastern North Atlantic, whereas the parallel AMOC recovery and hemispheric thermal adjustment proceeded much more slowly. These findings reveal that natural AMOC transitions evolved more slowly than the rapid North Greenland oxygen-isotope shifts would imply, thereby refining our understanding of “abruptness” during DO events.

ABSTRACT Understanding Earth‐surface processes along the fringes of large ice sheets is integral to deciphering cold‐region environmental variability and anticipating slope instability under climate change. We investigate the Meseta Chile Chico, a volcanic plateau in the central Patagonian Andes, located at the eastern margin of the former Patagonian Ice Sheet. Here, glacier retreat patterns are well‐constrained, while the influence of periglacial, paraglacial, and postglacial processes is not. We employ geomorphological mapping integrated with geological and palaeoclimate reconstructions to provide a spatio‐temporal model of landscape evolution. Enhanced erosion since the late Pliocene, caused by differential tectonic uplift and glacial isostatic rebound, produced the rugged morphology that characterizes the Meseta Chile Chico. During the Last Glacial Maximum, glaciers occupied surrounding valleys, while the surface of the plateau was reshaped by long‐lasting periglacial freeze‐thaw cycles and nivation. Postglacial landscape adjustment included intensified landslide activity along the edges of the plateau (linked to paraglacial de‐buttressing at its bordering valleys) and regional base‐level changes. During the Holocene, flow‐type landslides became increasingly controlled by climatic variability. The geomorphology of the Meseta Chile Chico reflects an interplay between inherited geology, (de)glaciation, and climate, highlighting the importance of an interdisciplinary approach for understanding recent landscape dynamics in deglaciating mountain regions.

ABSTRACT The hypothesis that temperate tree taxa survived the Last Glacial Maximum (LGM; 26.5–19 ka cal BP) in cryptic northern refugia has significant implications for European biogeography and postglacial vegetation history. Recently, sedimentary ancient DNA recovered from offshore sediment cores in the Southern River System of the North Sea Basin has been interpreted as evidence for the local persistence of temperate woodland taxa during the LGM. This review critically evaluates that hypothesis by synthesising the currently available palaeobotanical and sedimentary ancient DNA evidence from the North Sea Basin and adjacent terrestrial regions. Independent palaeoclimate reconstructions consistently indicate cold, dry conditions, continuous permafrost and temperatures well below the ecological requirements of temperate woodland trees. These reconstructions are broadly congruent with pollen, plant macrofossil, vertebrate and coleopteran records, which collectively indicate the persistence of low‐productivity arctic to subarctic steppe–tundra vegetation throughout the region during the LGM. This strong convergence among independent palaeoenvironmental datasets contrasts with limited support for the local persistence of temperate woodland communities throughout the LGM in the North Sea Basin.

Abstract Future atmospheric warming could cause an abrupt increase in ocean temperature beneath the Filchner–Ronne Ice Shelf, Antarctica, from −2.2°C to more than 0°C in a few decades. In simulations, such a transition leads to a twenty‐fold increase in sub‐shelf melt rates, driving a retreat of the ice sheet. Here we investigate the evolution and reversibility of the ocean and ice sheet states using coupled ice‐sheet–ocean model simulations. We show that sub‐shelf melt rates increase only half as much as in the uncoupled system, and we find that, after a period of climate cooling, the cavity is reversible from warm to cold conditions, independently of ice‐ocean interactions related to regional changes in sub‐shelf melting, cavity geometry, and ocean dynamics. In contrast to the abrupt warming transition, the recovery of a cold ocean state takes more than a century. Meanwhile the ice sheet continues to lose mass and contribute to sea level rise with a maximum rate of up to 0.9 mm/a.

Abstract Future changes in South Asian climate are expected to amplify flooding and drought, yet the spatial and temporal variability of the South Asian monsoon (SAM) under different climate regimes remains poorly constrained. Here, we present a new marine pollen record from the Bay of Bengal (BoB), informing on the vegetation of the Ganges‐Brahmaputra (G‐B) River basin whose outlet via the lower Meghna lies near 21.3°–21.8°N, and compare it with bulk and leaf‐wax δ 13 C and δD, and planktonic foraminiferal δ 18 O records, all from the same site. The G‐B basin was characterized by an open landscape with sparse subtropical trees during Marine Isotope Stage (MIS) 2 indicating drier and colder conditions. This is in sharp contrast with the development of tropical forests during MIS 1. During the Holocene, changes in the forest composition, characterized by a progressive increase in dry tropical trees relative to moist taxa, showed large‐scale drying, with more pronounced winter dryness after 8 ka. The comparison with the pollen records located off Mahanadi (20°N) and Godavari (18°N) catchments, both within the Indian core monsoon zone reveals that the composition of the forests differs strongly, showing a persistent north–south hydroclimate gradient. At orbital time scales, vegetation records from the western BoB support that the SAM variability is intrinsically linked to the ITCZ migration and largely affected by the Northern hemisphere ice‐sheets and insolation forcing; this is particularly significant for the Holocene. However, the spatial heterogeneity of hydroclimate conditions suggests additional modulation by Himalayan orography, and vegetation through moisture recycling.

Abstract The timing and duration of monsoon rainfall shape regional hydroclimate, yet their response to abrupt climate change remains poorly constrained. Here, we use 25-km-resolution climate simulations to investigate the seasonal migration of the East Asian summer monsoon rain belt during Heinrich Stadials, periods associated with collapsed Atlantic meridional overturning circulation (AMOC). We find that the rain belt migrated northward 18 days later than during interstadials. This delay kept the monsoon front longer over central China, prolonging early-summer rainfall despite an overall weaker monsoon. The delayed migration was linked to later northward advances of the western North Pacific subtropical high and upper-level westerly jet. North Atlantic cooling contributed to these circulation changes through Rossby wave propagation and a weakened land–sea thermal contrast. Our results suggest that abrupt Atlantic changes can reshape both the amount and seasonal timing of East Asian summer rainfall. These findings offer a palaeoclimate perspective on potential changes in monsoon seasonality under future AMOC weakening.

Abstract. An external perturbation to the climate system from anthropogenic or natural activity first impacts the climate by inducing a perturbation to Earth’s energy budget, known as a radiative forcing. The characteristics of the radiative forcing, such as its global-mean magnitude and spatial pattern, determine the subsequent climate response. Therefore, forming accurate projections of climate change first requires diagnosing radiative forcing and evaluating its persistent uncertainty in Global Climate Models. As part of the Coupled Model Intercomparison Project phase 7 (CMIP7), the second iteration of the Radiative Forcing Model Intercomparison Project (RFMIP2.0) will enable the systematic characterization of effective radiative forcing and its components across state-of-the-art climate models, through a set of fixed-Sea Surface Temperature timeslice and transient experiments. The protocol for RFMIP2.0, introduced here, will in part serve as a continuity and an expansion of core RFMIP experiments first introduced in CMIP6, some of which have now been incorporated into the overarching CMIP7 DECK and FastTrack protocols given their broad utility. This will allow for a consistent estimate for radiative forcing across multiple model generations, which is valuable for model evaluation and future development. RFMIP2.0 also includes new experiments that will address open questions about the definition of radiative forcing, such as its sensitivity to evolving surface conditions, and will further enhance an ever-growing swath of science applications that rely on an understanding of Earth’s energy budget.

Long-term satellite monitoring of rift propagation is essential for assessing Antarctic ice shelf stability and projecting sea-level rise. Currently, long-term tracking of rifts still relies on time-consuming manual interpretation, significantly affecting its feasibility for large-scale rift monitoring. Here, we present a novel framework for automatically tracking rift propagation, which fuses the fine-tuned visual foundation model (Segment Anything Model, SAM) and spatiotemporal correction. The SAM enables automated rift extraction using only a few samples without any manual prompts, while the spatiotemporal correction improves the precision of rift mapping for geometric measurements. We applied this framework to automatically monitor key rifts on the Amery Ice Shelf using 768 Landsat 8/9 and Sentinel-2 images from 2015 to 2025, achieving an F1 score of 98.6% across 11 independent scenes. Rift length and width measurements were in agreement with manual interpretation, with R2 values of 0.979 and 0.999, respectively. Furthermore, our extended monitoring revealed recent structural changes within the rift system on the Amery Ice Shelf. This study offers a new way for automated rift tracking with the foundation model for the first time, having significant potential for long-term monitoring of pan-Antarctic rifts.

Geophysical Research Letters, Volume 53, Issue 21, 16 November 2026.

Understanding energetic processes at the surface is important for understanding how the Greenland ice sheet melts. However, most temperature observations of firn are insufficient for studying energetic processes near the surface where melt happens. In this work we present new ground observations of high-resolution temperature profiles in the top meter of firn near the southwest coast of Greenland. Using these observations we characterize how temperature and energy vary during the melt season.

Abstract Exceeding global warming of 1.5 °C in the near term is now unavoidable. An overshoot pathway, in which exceedance is followed by decline to or below the threshold through net-negative emissions, is now the only remaining route back to 1.5 °C. Permanent exceedance forecloses the recovery from climate impacts that an overshoot pathway may permit, yet even an overshoot pathway leaves lasting legacies. To characterise the reversibility of temporary overshoot impacts, we propose a three-layer analytical framework distinguishing hazards (with four dimensions: magnitude, duration, rate of exceedance, and rate of decline), biophysical responses (reversible vs persistent change), and socioeconomic outcomes (reversible vs irreversible impact). We introduce the socioeconomic commitment threshold, endogenous to the overshoot trajectory: how much overshoot intensity a system can experience before the losses it accumulates persist after temperatures decline, explained by three mechanisms: non-substitutability, threshold-crossing, and lock-in. A typology linking biophysical persistence with socioeconomic irreversibility illustrates how reversible biophysical change can produce irreversible socioeconomic loss, while persistent biophysical change need not produce irreversible outcomes where adaptive capacity is sufficient. Whether temporary exceedance produces permanent harm is determined primarily by socioeconomic rather than biophysical factors, except where permanent biophysical change exceeds adaptation limits. For systems and communities with low adaptive capacity, overshoot concentrates irreversible legacies even where temperatures subsequently decline, placing equity and justice at the centre of overshoot assessment.

Journal of Geophysical Research: Oceans, Volume 131, Issue 11, November 2026.

Abstract Large‐scale, high‐resolution estimation of snow water equivalent (SWE) in mountainous regions remains challenging. Continental‐scale approaches, including SWE reconstruction and regional climate model (RCM) simulations, are computationally expensive at high resolutions. We present Blender, a variational data assimilation method that improves modeled SWE and snowfall time series by modifying snowpack mass and energy balance while minimizing changes to input precipitation and energy fluxes, subject to remotely sensed snow cover fraction constraints. Blender combines the advantages of RCM simulation and SWE reconstruction and can be applied as a post‐processing step to many gridded snow data sets. We tested Blender at 500 m resolution in the Tuolumne watershed, Sierra Nevada, USA, during water years 2013, 2016, and 2017 (low, average, and high snow accumulation). We tested Blender with prior inputs from an RCM and ERA5‐Land (ERA5L), and evaluated posterior (output) performance using 18 Airborne Snow Observatory (ASO) flights, and compared its performance with the Western United States Snow Reanalysis (WUS‐SR). Blender improved spatial SWE patterns relative to ASO in all years for the RCM prior (RMSEP: 17.2% → 11.5%, R : 0.39 → 0.72) and ERA5L prior (RMSEP: 13.9% → 12.9%, R : 0.003 → 0.59), approaching WUS‐SR performance (RMSEP = 9.5%, R = 0.82). Blender improved snow water storage (SWS) estimates for the RCM prior but not for ERA5L prior, although ERA5L posterior SWS remained reasonably accurate (median absolute error of 27%). Blender provides a flexible and efficient pathway toward higher‐resolution, large‐scale SWE estimation in mountainous regions.

The Indian Ocean regulates global climate through heat and moisture exchanges. To understand past climate variability, we analyzed oxygen isotopes in individual fossil shells of tiny marine organisms. Combining numerical simulations, we show that shell chemistry reflects seasonal temperature changes driven by both the monsoon and the Indian Ocean Dipole. This opens new perspectives for reconstructing past climate extreme events.

The Middle Miocene expansion of the Antarctic Ice Sheet coincides with major sedimentological changes in Malta. This interval also saw reorganisations in ocean circulation linked to closure of the Mesopotamian Seaway (MSR-events) and perturbations in the carbon cycle (CM events). The final CM event (CM-6), slightly lags the main glacial phase and remains difficult to interpret due to competing mechanisms and conflicting p CO 2 reconstructions. To explore links between cryosphere, circulation, hydroclimate, and the carbon cycle, we present new geochemical records from Malta. Our records show that cryospheric expansion was followed by large, precessionally paced variations in sea-surface salinity and a marked regional increase in biological productivity. The results indicate a shift toward stronger monsoonal influence, delivering enhanced freshwater and nutrient fluxes to the region. This enhanced local biological carbon pump may have acted as a positive feedback to a global CO 2 drawdown. However, any contribution to global CO 2 drawdown remains tentative and cannot be confirmed without contemporaneous high-resolution open-ocean records.

Journal of Geophysical Research: Atmospheres, Volume 131, Issue 22, 28 November 2026.

The Hengduan Mountains are a major hotspot for lake-sediment research, but regional vegetation responses to Holocene climate variability remain incompletely understood. Here, we analyzed pollen assemblages and concentrations from Dinggongniang Co (4633 m a.s.l.) to investigate regional vegetation-related changes over the past approximately 4.7 ka. Pollen percentage and concentration analyses were combined with ensemble empirical mode decomposition (EEMD) and Lomb–Scargle spectral analysis. Despite being located above the modern tree line, the lake sediments are dominated by arboreal pollen (89.5%), particularly Pinus, indicating substantial pollen input from lower-elevation forests. Total pollen concentrations were relatively low around 4.2 ka, increased markedly during the Roman Warm Period, and remained relatively elevated during the Medieval Climate Anomaly, suggesting changes in regional arboreal pollen input potentially associated with Asian summer monsoon-related hydroclimatic variability. Furthermore, time-series analyses identified a possible millennial-scale oscillation at approximately 1.3–1.4 ka, broadly consistent with variability documented in North Atlantic and Asian monsoon records. This finding highlights the value of high-altitude lake pollen records for tracing regional arboreal pollen input and suggests possible connections between Asian monsoon variability and broader Northern Hemisphere climate fluctuations. Although chronological uncertainties and multiple factors influencing pollen concentrations introduce some limitations, the observed patterns offer valuable insights into long-term vegetation-related environmental variability in the Hengduan Mountains.

Proceedings of the National Academy of Sciences, Volume 123, Issue 44, November 2026. SignificanceSince Gerard Bond first identified millennial-scale climate events or oscillations in the Holocene from ice-rafted debris records in the North Atlantic, scientists have struggled to explain how these oscillations occurred. Using multiproxy ...

Thawing permafrost accelerates global warming and threatens infrastructure, yet tracking changes over vast areas remains difficult. We combined over two decades of ground observations with artificial intelligence to create a continuous, highly detailed yearly map of permafrost thaw depth across the Northern Hemisphere from 2000 to 2024. Overcoming previous mapping limitations, this reliable dataset provides a vital tool to evaluate environmental hazards and protect vulnerable regions.

This study analysed volcanic impacts on East Asian temperature and precipitation (wetness/dryness) anomalies and investigated the consistency of climate responses between the REACHES reconstruction and paleoclimate reanalysis product (ModE-RA) from the 15th to 20th centuries. Volcanic events were compiled from the eruption chronology based on ice-core sulphate data, and 10 major eruption events were identified. For each event, the eruption year was designated as t0 and used as the reference year for event alignment. Superposed epoch analysis was applied with a ± 8-year window to compare post-eruption years (t = 0–8) against pre-eruption baselines (t = −8–−1) by a significance t-test. For identifying whether extreme years (percentile thresholds from the top 5%–25% in temperature and precipitation fields) are attributed to volcanic forcing or natural variability, we adopted Bayes’ conditional probabilities and applied an exact binomial test to perform reverse inference. Both datasets showed robust post-eruption cooling signals, the strongest being within t0 +3 years. Precipitation responses were controversial: REACHES revealed overall wetting, whereas ModE-RA suggested drying, indicating a consistent cooling effect of major volcanic eruptions and divergent responses in hydroclimates. Future analysis should focus on the hydroclimatic effects of volcanic eruptions using high-resolution reconstruction data and simulations.

Journal of Geophysical Research: Oceans, Volume 131, Issue 11, November 2026.

Journal of Geophysical Research: Oceans, Volume 131, Issue 11, November 2026.

Glacier deposits are used to reconstruct past climate, but the same glacier size can result from different mixes of temperature and snowfall. Using numerical glacier simulations in the Bolivian Andes, we explored how to overcome this ambiguity. We show that combining several neighboring glaciers with independent rainfall records partly resolves it, giving temperatures 0.8°C and 1.3°C lower during the Little Ice Age and Early Holocene (~10,000 years ago), respectively.

arXiv:2611.06539v1 Announce Type: new Abstract: Global climate model resolution is still too coarse to fully resolve mesoscale processes, therefore subgrid closures are required to approximate the feedback of eddies onto the resolved flow. With high-resolution datasets now readily available, there has been a rapid growth in the use of machine learning to develop new eddy closures that could help reduce model biases and uncertainty. In this work we employ an offline learning approach with a convolutional neural network to predict subgrid eddy kinetic energy ($KE_e$), laying the initial groundwork for a machine-learned scale-aware GEOMETRIC eddy closure for the UK's Earth System Model. We address the question of whether global parameterizations are needed or if regional closures are sufficient by evaluating a number of neural network models through region and feature ablation experiments. ML parameterization models generalize to unseen regions when their training data share similar feature and target distributions with those regions. Models that fail to generalize incur distributional shift, or employ different filtering scales for the true eddy kinetic energy field. When holding out features, there is improvement among the generalizable ML models, with some feature spaces leading to improvements in some regions but degrading performance in others. Generally, power spectral density of $KE_e$ tend to diverge at small scales, though some feature spaces are able to minimize this divergence. Overall, the best performing model trained on all regions achieves time-mean and anomaly R2 scores of R2$\approx 0.95$ and R2$\approx 0.8 - 0.9$, respectively.

The Zhejiang region preserves a complete evolutionary trajectory from hunter-gatherer to early state-level societies, which offers critical material for examining social complexity. Nevertheless, the lack of continuous, high-quality Holocene paleoenvironmental archives in this area hampers a thorough assessment of the human–environment interaction over time. This study investigates pollen, ostracod, foraminiferal and particle-size proxies to reconstruct regional environmental dynamics from core TQ1 in Taizhou Bay spanning 14.8–0.4 cal ka BP based on AMS 14C dating. Core TQ1 records a transition from predominantly terrestrial influence during the late deglaciation to increasing marine influence during the early Holocene. Based on paleoenvironmental changes, the Holocene can be further divided into three stages. Stage 1 (11.5–8.5 cal ka BP) was characterized by a coastal environment with stronger hydrodynamics; arboreal pollen declined while herbs, ferns and algae increased. Stage 2 (8.5–4.2 cal ka BP) represented the peak transgression with open marine conditions and maximum microfossil abundance, and forest recovery as terrestrial herbs and algae decreased. Stage 3 (4.2–0.4 cal ka BP) shifted back to nearshore shallow-sea conditions, with persistently high arboreal percentages and sustained pollen input from forested terrestrial source areas. The intense paleoenvironmental changes, coupled with the evolutionary processes of ancient cultures in the adjacent areas, provide new evidence for a deeper understanding of regional cultural evolution.

Geophysical Research Letters, Volume 53, Issue 21, 16 November 2026.

The geometry and ventilation of upper deep-water masses in the Pacific sector of the Southern Ocean are critical to glacial carbon storage, yet their evolution since the Last Glacial Maximum remains poorly constrained. Here, we present the foraminiferal neodymium isotope composition (ε Nd ) from core MD97-2115 (eastern Chatham Rise, 2160 m water depth), which is located at the present-day boundary between Upper Circumpolar Deep Water (UCDW) and modified North Pacific Deep Water (mNPDW) in the Southwest Pacific, to reconstruct the upper deep-water mass changes since 30 ka. Combining quantitative binary mixing calculations with vertical ε Nd gradients relative to Lower Circumpolar Deep Water, we suggest substantial shifts in the geometry and mixing of UCDW and mNPDW since 30 ka. The variation phases include reduced UCDW influence during 30–20 ka, enhanced UCDW contribution between 20 and 14.3 ka, slightly decreased UCDW during 14.3–11.4 ka, and stable Holocene conditions. The glacial dominance of mNPDW is consistent with basin-wide evidence for enhanced storage of respired carbon in Pacific-sourced deep waters, likely contributing to the maintenance of low atmospheric CO 2 during the glacial period. Crucially, the early rapid increase in UCDW at ∼20 ka preceded the penetration of North Atlantic-sourced water into Southern Ocean and the deglacial CO 2 increase by several millennia, implying a critical role for Antarctic and Southern Ocean processes in establishing the oceanic precondition for glacial termination. During the last deglacial period, the decoupling between our reconstruction and atmospheric CO 2 evolution suggests that water-mass geometry alone is insufficient to drive the full deglacial CO 2 increase, pointing to additional roles for processes such as regional outgassing and variability in air–sea gas exchange efficiency.

Journal of Geophysical Research: Atmospheres, Volume 131, Issue 22, 28 November 2026.

Nature Communications, Published online: 05 November 2026; doi:10.1038/s41467-026-76809-5 This paper shows that current temperature and emission-reduction targets provide only limited protection against future heatwaves, which are found to surpass recent exceptional events across multiple regions of the world.

Abstract Snowmelt plays a vital role in sustaining river flows and regulating water resources in mountain ecosystems, particularly under changing climatic conditions. This study aims to quantify the contribution of snowmelt to streamflow by estimating snow cover and snow water equivalent (SWE) in the Greater Caucasus Basin of Azerbaijan. MODIS and FLDAS remote sensing products were analyzed using the Random Forest machine learning algorithm, with data processed on the Google Earth Engine (GEE) platform to improve prediction efficiency and accuracy. Results show that Random Forest significantly enhanced snow cover estimation compared to satellite data alone. MODIS and FLDAS observations revealed peak snow cover in December 2023 (1,310 km²) and maximum SWE in February 2023 (23 kg/m²). Over the longer period 2013–2023, the highest snow cover was recorded in 2016 (5,052 km²), while peak SWE occurred in 2019 (188.7 kg/m²). These findings indicate a weakening role of snow in regional runoff regulation and water supply, driven by declining snowfall and streamflow. The study demonstrates the value of machine learning for snowpack monitoring and highlights the need for long-term simulations and expanded ground-based networks to improve prediction of peak flow events in mountain basins.

Nature Geoscience, Published online: 06 November 2026; doi:10.1038/s41561-026-02142-7 Warming in the early Eocene reduced the efficiency of the Southern Ocean biological carbon pump, limiting carbon sequestration in the ocean interior and enhancing atmospheric CO2 levels, according to foraminifera-bound nitrogen isotope records.

Northeast China lies on a mid-latitude boundary where small warming can shift snowfall to rain and shorten the snow season, but climate models often simulate snow fields with large regional biases. Here we show that satellite-constrained, 1-km machine-learning emulators project snow-cover duration without directly using modeled snow variables. They relate observed daily snow cover during 2001–2020 to temperature, precipitation, snowfall fraction and terrain, and are driven by bias-corrected climate projections. Snow seasons shorten almost everywhere, with more than 98% of land grid cells agreeing on the sign of change. By 2081–2100, domain-mean snow-cover duration decreases by 42, 64, and 85 days under low-, intermediate- and high-emission pathways, respectively, and areas with fewer than 50 snow days expand from 19% to 99% under high emissions. Snow-season shortening is dominated by loss of late-winter and spring snow persistence, especially at low–middle elevations and along the present rain–snow transition zone. The loss of snow persistence during late winter to spring across Northeast China leads to shorter snow season under global warming, as suggested by bias-corrected CMIP6 climate projections based on Satellite-constrained machine learning framework.

arXiv:2611.08700v1 Announce Type: cross Abstract: We developed TitanMPAS, a new Titan global and regional climate model, to simulate atmospheric dynamics and methane cycles. It integrates the non-hydrostatic MPAS dynamical core and TitanWRF physics parameterizations, with corrections to relax the thin-atmosphere assumptions on the gravity and curvature terms. TitanMPAS improves upon TitanWRF by incorporating a better moist convection parameterization and the observed surface properties. Long-term simulations ($>$100 Titan years) with a high resolution of $1^\circ$ and the realistic topography show that TitanMPAS is able to produce the stratospheric superrotation with typical equatorial wind speed over $\rm 120\,ms^{-1}$ and the maximum zonal wind speed exceeding $\rm 170\,ms^{-1}$ . Besides the good match of temperature profiles to the Cassini CIRS observations, the simulated wind, temperature and methane abundance in the troposphere match well with Huygens probe in-situ measurements. TitanMPAS also captures the major features of seasonal methane precipitation distribution patterns inferred from the observed cloud activities, as well as the vertical and latitudinal variations of methane abundance in the reanalysis of Cassini VIMS data. Resolution sensitivity studies show that higher resolution produces faster stratospheric superrotation, as well as stronger near-surface wind due to better resolved terrain variations. We further demonstrate that, despite the weak solar insolation, the near-surface temperature, pressure, wind magnitude and methane abundance all exhibit diurnal variations. Lastly, we show potential impacts of topography on methane transport and eddy momentum flux in both troposphere and stratosphere, which should be investigated in a future study.

Nature, Published online: 09 November 2026; doi:10.1038/d41586-026-03658-z The COP31 climate conference begins, with ‘climate overshoot’ a key talking point. Plus, the surprising benefits of ‘pink noise’ and the role of AI chatbots to support men’s mental health.

Proceedings of the National Academy of Sciences, Volume 123, Issue 44, November 2026. SignificanceNeuromodulators such as neuropeptides and bioamines regulate behavioral states across species, yet their collective computational consequences remain poorly understood theoretically. We show that a single biologically motivated mechanism for ...

Geophysical Research Letters, Volume 53, Issue 21, 16 November 2026.

Earth's Future, Volume 14, Issue 11, November 2026.

Journal of Advances in Modeling Earth Systems, Volume 18, Issue 11, November 2026.

Journal of Advances in Modeling Earth Systems, Volume 18, Issue 11, November 2026.

Abstract We present a unique application of radiocarbon measurements on hemipelagic sediments that uses continuous radiocarbon dating of total organic carbon ( 14 C TOC age) to recognize event deposits. The 14 C TOC age of sediments is not commonly used for age determination because total organic carbon includes various origins, such as marine, terrestrial, and reworked old fragments. In this study, the 14 C TOC ages of some horizons were found to be older than those of the deeper ones through continuous 14 C TOC measurements at 0.5- to 1.0-cm intervals. The reversal 14 C TOC ages resulted not from the settlement of organic particles from the terrestrial environments, such as floodings, but from the submarine slope failures. Therefore, we can regard the reversal 14 C TOC layers as event layers due to geohazards. After reconstruction of the age model using foraminiferal tests obtained from the sediment core used in this study, depositional ages of event layers were identified. Well correlation with historical natural disasters after the 15 th century suggest validity of the method proposed by the current study.

Abstract Under global warming, land surfaces have warmed substantially faster than the oceans, leading to widespread drying across most continental regions. However, across South Asia, an anomalous trend of moistening, indicated by an increase in surface relative humidity (RH), has been observed in recent decades despite continued land-surface warming. We find that this moistening pattern is consistently evident in multiple observational, reanalysis, and station-based datasets and is also present in the CMIP6 multi-model mean, indicating that the observed moistening trend is broadly consistent with the externally forced response simulated by CMIP6. These concurrent warming and moistening trends have important implications for humid-heat conditions relevant to human health, as higher RH under a warming climate can significantly reduce heat dissipation and exacerbate heat stress. We further evaluate these humid-heat implications using wet-bulb temperature and heat index. This study further discusses the possible mechanisms behind the anomalous moistening trend, focusing on the monsoon season, and highlights enhanced moisture convergence as a dynamically consistent and physically plausible contributor to the increase in surface RH under warming. The broad consistency between the observed trends and the CMIP6 multi-model mean suggests an important role for the externally forced climate response and provides a basis for future assessments of whether the South Asian moistening trend may persist or intensify under continued anthropogenic forcing. Such assessments should nevertheless account for the influences of internal variability, regional processes, and remaining model biases.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 11, November 2026.

Geophysical Research Letters, Volume 53, Issue 21, 16 November 2026.

Journal of Geophysical Research: Atmospheres, Volume 131, Issue 21, 16 November 2026.

Nature, Published online: 04 November 2026; doi:10.1038/s41586-026-11368-9 Carbon radical generation by irreversible metal-to-ligand charge transfer

Journal of Advances in Modeling Earth Systems, Volume 18, Issue 11, November 2026.

Abstract Polar amplification is a well-documented phenomenon driven by multiple feedback mechanisms, with latent heat transport (LHT) playing a role that is often underestimated. Although the total atmospheric energy transport into polar regions remains nearly unchanged in warming scenarios, compensating changes in dry static and latent energy transport interact with polar-amplified surface warming: dry static energy transport generally decreases due to a reduction in meridional temperature gradient, while the latent energy transport increases, contributing to polar warming by enhancing moisture. To isolate the impact of LHT changes, we lock the moisture advection in an idealized general circulation model (GCM) and apply it as a forcing. Simulations forced with an increase in LHT, equivalent to that which would be produced by a doubling of CO 2 , consistently demonstrate substantial polar warming, providing a different perspective from the conventional feedback analysis that attributes polar amplification predominantly to the ice-albedo feedback and lapse-rate feedback. The GCM is coupled to a thermodynamic sea ice model, which allows the role of sea ice to be examined by systematically disabling sea ice-LHT interactions. Sea ice loss strongly enhances annual mean polar amplification and wintertime polar warming, and is itself a response to the increase in LHT. We additionally provide a quantitative comparison of different metrics of LHT-induced polar amplification. These findings suggest that diagnostic decompositions, by only quantifying the enhanced LHT into polar regions and neglecting its impact on polar feedbacks, underestimate the role of LHT as a significant driver of polar amplification.

Science, Volume 394, Issue 6824, Page 602-603, November 2026.

arXiv:2611.07094v1 Announce Type: new Abstract: This paper presents an application of machine learning to the problem of the computational expense of atmospheric radiative transfer calculations from physics-based models. While in recent studies individual emulators have been created that emulate certain radiative transfer schemes for different sensors, we show that a general framework can be applied to emulate the radiative transfer method itself. A fully connected deep neural network has been trained to predict top-of-atmosphere microwave radiances in all-sky conditions over various land, ocean and sea ice surfaces at all frequencies from 1 to 300 GHz, and for all incidence angles from 0 to 65 degrees. The reference models against which the emulator has been trained include an Eddington two-stream radiative transfer model for one-dimensional and pseudo-three-dimensional microwave brightness temperature simulations. This neural network architecture has also been separately trained to reproduce the explicit calculations of the Community Radiative Transfer Model (CRTM) for a single microwave imager. In all cases, the model architecture is successfully able predict microwave brightness temperatures by being trained directly on the optical properties of the atmosphere and surface and therefore directly emulating the radiative transfer method. Compared to the explicit schemes used to produce the training data, the GPU-enabled emulator is able to produce brightness temperature simulations more than three orders of magnitude faster and with a root-mean-squared-error less than 0.0348 K, 0.0531 K, and 0.0246 K against the independent test data for the two Eddington models and CRTM, respectively. Additionally, preliminary training using a full 3D Monte Carlo scheme for high-resolution simulations shows promising results for the application of this method to "exact" radiative transfer methods, and these results are discussed in Section IV.

arXiv:2608.07406v3 Announce Type: replace Abstract: As artificial intelligence (AI) moves from research prototypes to operational tools in Earth system science, trust in its predictions becomes increasingly important. Explainable AI (XAI) offers insight into how AI models make predictions. However, many influential XAI methods were developed or popularised for classification, and their interpretation for regression, which dominates geoscience, is less well established. A central obstacle is the lack of ground truth for judging explanations. We show that, for a dynamical system with known governing equations, matched references can be built directly from the physics. We extend established sensitivity methods to provide physical references for both sensitivity and attribution, separately for methods with implicit and explicit baselines, all computed from the governing equations independently of the network. Applied to an AI emulator of the Lorenz-63 system, these references show that method families answer distinct questions rather than approximating one ``true'' explanation. The emulator's sensitivities and attributions can depart most from their physical references where it forecasts best, so predictive skill is not a substitute for the physical realism of what a model has learned. We then review XAI applications to Earth system regression, organised by method family, extending to issues an idealised system cannot address: computational scalability, strongly correlated high-dimensional predictors, and reliance on expert judgement where no physical reference exists. We also discuss how an XAI method can be mathematically valid yet only partially relevant to a given stakeholder group. We close with future directions and practical recommendations for responsible XAI use in geoscientific modelling and forecasting.

Science Advances, Volume 12, Issue 45, November 2026.

Journal of Advances in Modeling Earth Systems, Volume 18, Issue 11, November 2026.

Journal of Geophysical Research: Oceans, Volume 131, Issue 11, November 2026.

Nature, Published online: 09 November 2026; doi:10.1038/d41586-026-03631-w Scientists say governments must commit to carbon dioxide removal if world is to bring temperatures back down to safer limits.

Nature Communications, Published online: 09 November 2026; doi:10.1038/s41467-026-77956-5 Reply to: Comment on “Self-organizing neuromorphic nanowire networks as stochastic dynamical systems”

Nature Communications, Published online: 09 November 2026; doi:10.1038/s41467-026-77954-7 Comment on “Self-organizing neuromorphic nanowire networks as stochastic dynamical systems"

Journal of Geophysical Research: Atmospheres, Volume 131, Issue 22, 28 November 2026.

Proceedings of the National Academy of Sciences, Volume 123, Issue 45, November 2026. SignificanceIdentifying important features with the false discovery rate (FDR) control is critical in many scientific fields. We develop the Bayesian knockoff filter (BKF) by incorporating the knockoff procedure in a fully Bayesian framework. The BKF ...

We estimate the thickness of a glacier in the Himalaya, using recently available thickness measurements and a computational model which uses higher-order physics to match thickness with satellite-observed velocity. Our velocity-constrained estimate achieves similar accuracy to leading estimates, while estimates additionally constrained by thickness show increased accuracy, but limited interpolative power. We make recommendations for future measurement locations and for choosing model parameters.

Glacier Equilibrium Line Altitudes (ELAs) are substantially influenced by topography through processes such as preferential shading, avalanching, or snow redistribution by wind. Here, we present a novel framework with which these topographic controls can be better assessed. The approach involves (1) estimating the ELA from glacier outlines and surface geometries, (2) separating the ELA into a regional climatic average (regELA) and a local topographic residual (ΔELA) via spatial autocorrelation, and (3) correlating ΔELA to terrain parameters characterizing the shape and orientation of glaciers and their catchments. The framework is developed and calibrated using the Swiss Glacier Inventory (SGI) 2016. In the SGI2016, the average absolute ΔELA is around 200 m, and shows significant correlations with potential solar radiation, topographic openness, ice surface curvature, and the relative size of a glacier’s catchment. These terrain parameters are then used to train a random forest-based regression model, which accurately predicts ΔELA for each Swiss glacier (R2 = 0.95) but can also be successfully transferred to earlier inventories within the Swiss Alps (R2 = 0.59 to 0.73). A spatial transfer, on the other hand, works only for regions with similar topographic and glaciological characteristics like New Zealand (R2 = 0.59), while a transfer from Switzerland to Icelandic glaciers proved unsuccessful (R2 = 0.11) and requires local recalibration. The framework developed in this study allows for a better quantification of topographic controls on present-day glaciers but can eventually also be transferred to a palaeoclimatic context, leading to improved temperature and precipitation reconstructions for past glaciations.

Abstract While the term “Anthropocene” is well established across scientific disciplines and social spheres, interpretations are diverse. Taking account of the 2024 rejection by a geological commission to accept the Anthropocene as a geological epoch and the related scientific debate, here we offer a future‐oriented perspective from the viewpoint of Earth system science. We describe different pathways in the Anthropocene up to the year 3,000, systematically characterizing them according to impacts and causes. We discuss the enormous global consequences of anthropogenic pressures on the Earth system and quantify the corresponding long‐term commitment to change. Regarding the causes, we conservatively explore best‐case and middle‐of‐the road emission scenarios, in combination with climate sensitivities drawn from within the IPCC likely range. We also discuss implications for Earth system resilience that could result in what we call worst case scenarios for Anthropocene outcomes. We conclude that, beyond the slow pace of natural climate recovery spanning many millennia, even minimal, unavoidable residual emissions like from the food sector risk perpetuating global warming in the absence of other human forcing. One implication is that if climate or carbon cycle feedbacks shift toward reinforcing warming, they risk not only exacerbating climate impacts but to also surpassing human forcing in relevance. At that point, human influence on the Anthropocene would no longer play the dominant role.

arXiv:2611.05922v1 Announce Type: new Abstract: Evidences show it is questionable to use the trends in Rossby wave phase speed (Cpx) to interpret the relation between the changes of extreme events and westerly jets. We propose, by using 70 years (1954-2023) of ERA5 reanalysis, that the trends in extremes are associated with the shifts of Rossby wave regimes, via the westerly-induced changes in Rossby wave group velocity (Cgx), which are two to three times larger than those in Cpx. We find significant deceleration in the subtropics and polar regions, alongside a salient midlatitude acceleration, of Rossby wave Cgx, especially over the East Asia-Pacific-North America and North Atlantic-Europe sectors. These changes induce robust shifts in spatiotemporal regimes of Rossby waves, from the transient synoptic-scale regime toward the quasi-stationary planetary-scale wave regime where Cgx decelerates, and vice versa. The CESM2 Large Ensemble confirms that the observed pattern over the Eurasia-Pacific-North America sector is a forced response to external forcing, while uncertainties remain over the polar regions and the North Atlantic. We suggest that westerly trends may reshape the regimes of extratropical Rossby wave activity and extreme weather events through the changes in group velocity rather than phase speed of Rossby waves.

Lake-terminating glaciers of the Southern Patagonian Icefield are retreating rapidly, yet dense multi-glacier records of their termini remain scarce. We present a training-free workflow in Google Earth Engine that measures ice-area change within a fixed terminus domain from every usable Sentinel-2 scene (per-scene Otsu threshold on the near-infrared band, with cloud screening restricted to lake pixels) and delineates annual terminus positions automatically. Applied to Perito Moreno, Upsala and Spegazzini glaciers (Lago Argentino, Argentina) for October–March 2016–2026, it yielded 260 scenes. Perito Moreno showed no significant trend until mid-2019 and then lost 0.39 km2 a−1 (p < 0.001); its terminus retreated by about 580 m in nine summers; the two most recent summers showed the largest seasonal losses of the record (≈1.6–1.7 km2); and in March 2026 the front detached from Península de Magallanes. Upsala retreated by about 1010 m (≈130 m a−1), whereas Spegazzini remained within ±100 m of its 2017 position. Warm, dry summers in 2021–2022 affected all three glaciers, so their contrasting responses suggest that basin geometry and lateral support modulate the response to a common climate. Sentinel-1 reproduced the Perito Moreno decline but failed in iceberg-choked Brazo Upsala.

Lake ice phenology is a sensitive indicator of climate change, yet its regional hydroclimatic associations remain poorly understood. Here, we reconstructed daily lake surface water temperature (LSWT) from MODIS for 583 lakes across China and derived freeze-up date (FUD), break-up date (BUD), and ice cover duration (ICD) for 376 seasonally ice-covered lakes. Robustness was evaluated with sensitivity tests, meteorological observations, transition-period comparisons, and passive-microwave ice records; Mann-Kendall tests, Theil-Sen slopes, VIF-constrained XGBoost-SHAP, and structural equation modeling assessed trends and regional associations. FUD was delayed in all three ice-covered regions, whereas BUD was delayed in the Tibetan Plateau Lake Region (TPR) but advanced in the Mongolian-Xinjiang Lake Region (MXR) and Northeast Lake Region (NLR), resulting in shorter ICD in all regions. Regional mean ICD declined by 0.03, 0.35, and 0.47 d yr -1 in the TPR, MXR, and NLR, respectively; significant lake-level changes were concentrated in the TPR, with 72 and 33 lakes showing significant FUD and BUD trends. All nine models retained positive hold-out R 2 (0.535–0.960). SHAP attributions differed by region: radiative variables dominated in the TPR, the MXR showed metric-specific associations with wind, snowfall, geographic, and thermal , and geographic, radiative, and morphometric variables dominated in the NLR. Structural equation models indicated stronger precipitation-related associations in the NLR and snowfall-related associations in the MXR and TPR. These results reveal pronounced regional heterogeneity in lake ice responses to warming and the value of integrating remote sensing, interpretable machine learning, and pathway modeling to resolve these contrasts.

Mountain river basins in Central Asia are highly sensitive to climate-driven changes in snow and glacier melt, yet the magnitude and timing of the resulting hydrological regime shifts remain poorly quantified. To address this gap, we assess present and future hydrological shifts in the transboundary Zarafshon River Basin with the SPHY glacio-hydrological model. We improved the representation of snowmelt and snow infiltration, which redistributes water to delayed baseflow and improves winter and spring discharge. We calibrated the improved model against discharge observations, satellite-derived snow cover and glacier mass balance data through Monte Carlo optimization. Over 1991–2020, snowmelt dominates runoff (52.4%), followed by baseflow (23.3%) and rainfall-runoff (21.1%), whereas glacier melt contributes 3.2% but sustains late-summer low flows. Snow and glacier melt dominate the upstream hydrology, whereas rainfall-runoff and baseflow sustain the flow in the lower basin. Using downscaled ISIMIP3b climate scenarios, we project glacier volume to decline by ∼ 40% under SSP126 and ∼ 80% under SSP585 by 2100. Annual total runoff declines by less than 7% in all scenarios, because gains in rainfall-runoff and baseflow almost offset the melt loss. Peak total runoff declines by 22.6% under SSP585, and the annual peak discharge shifts earlier by 22–38 days. The hydrological regime shift from snowmelt to rainfall-runoff is more pronounced in the Fondarya than in the Matcha River system. These results indicate a shift from a snowmelt to a rainfall-runoff dominated hydrological regime. Such a shift would reduce summer water availability and increase irrigation stress, highlighting the need for adaptive management.

Abstract It was recently reported that a Quaternary basin in southeastern Korea was formed by a meteorite impact, and a 12-cm-thick tephra layer was found in lake sediments in this basin. In this study, we analyzed the geochemical composition and refractive index of glass shards to identify the unknown tephra. Based on both results, this tephra was identified as Kb-Ks tephra that erupted in the Kyushu area, Japan, approximately 520 ka. To our knowledge, this is the first report of Kb-Ks tephra in terrestrial sediments in the Korean Peninsula. These findings indicate that the depositional age of the lake sediments is much older than previously suspected (ca. 63 ka), and that the meteorite impact occurred before 520 ka. Although more detailed studies are needed, the discovery of Kb-Ks tephra provides important information concerning the paleoclimate during the middle Pleistocene, as well as the prediction of volcanic hazards in the Korean Peninsula.

Nature Climate Change, Published online: 09 November 2026; doi:10.1038/s41558-026-02790-y Climate targets are set for global temperatures, but what they mean regionally is less clear. Here the authors quantify the amount of human-induced warming in individual IPCC regions and find that, by 2025, they had reached 90–94% (65–76%) of their expected warming levels at the 1.5 °C (2 °C) global warming thresholds.

Run-of-river small hydropower (SHP) can electrify off-grid mountain communities in Central Asia, but national assessments rely on coarse runoff grids and fixed capacity factors. We assess 949 reaches of the glacierised Sary-Jaz basin (Tien Shan, Kyrgyzstan) by combining twelve winters of Sentinel-1 backscatter to detect river ice, a degree-day discharge model forced by ERA5-Land and CHIRPS, and a levelised cost of electricity (LCOE) framework. Energy is computed from a screening capacity factor or simulated daily flow, and uncertainty is examined with Monte Carlo, multi-SSP and lifetime analyses. Screening LCOE without ice is 72–544 USD/MWh (median 147). Without ice, the 20–50% glacier-fraction class is cheapest (139.5 USD/MWh), but SAR-derived ice derating shifts this to the 5–20% class (146.7 versus 152.3 USD/MWh). The shift persists with flow-based energy (156.8 versus 162.7 USD/MWh) and needs a shutdown fraction of only 0.00–0.14. Because simulated winter flow mostly falls below the minimum turbine flow, this derating is an upper bound; with generation-weighted derating the classes are tied at present, but the 5–20% class remains cheaper in all eight SSP snapshots. With reach-level cost uncertainty, it is cheaper in 63% of 2000 Monte Carlo realisations, a probable tendency rather than a certainty. Under SSP2-4.5, the 50–100% class loses 24% of its design discharge by 2080; its lifetime LCOE rises from 173 to 180 USD/MWh between commissioning years 2030 and 2050, whereas the 5–20% class stays near 152 USD/MWh. No reach approaches the ≈46–52 USD/MWh feed-in tariff; tariff reform or concessional finance is required.

Open Access via the Elsevier agreement

Global warming has become an increasingly urgent issue that must be addressed within a limited timeframe, as the debate over the overshoot pathway intensifies. Existing climate and socio-economic models inform our understanding but are too complex for timely action and non-expert use. To address these challenges, our stochastic framework leverages a one-box Ornstein-Uhlenbeck model with colored-noise forcing to characterize variance scaling in detrended temperatures and links this behavior to an empirical Hurst coefficient. The observed weakening of anti-persistence, as summarized by this coefficient, is then used to quantify changes in temperature variance relative to the pre-industrial era. Our empirical Hurst maps reveal patterns consistent with findings from global climate models, highlighting greater changes in temperature variance in equatorial regions, such as southeastern Amazonia and western Indonesia, than in high-latitude regions. Furthermore, seasonal analyses of African temperature records reveal pronounced heterogeneity in variance changes, identifying vulnerable regions that are masked when variability is assessed at annual timescales. By linking our results to climate solutions that consider fairness in burden-sharing and unequal risk-bearing capacity, we anticipate that the empirical Hurst coefficient will support more equitable and effective climate action.

Nature Climate Change, Published online: 03 November 2026; doi:10.1038/s41558-026-02779-7 By integrating climate risk into health planning, Brazil is redesigning healthcare to withstand floods, heatwaves and other climate shocks.

Proceedings of the National Academy of Sciences, Volume 123, Issue 44, November 2026. SignificanceThe chronology of ancient Mesopotamia provides a critical temporal framework for synchronizing civilizations across the Near East and eastern Mediterranean, yet competing chronological schemes have persisted for decades. We use patterns ...

Proceedings of the National Academy of Sciences, Volume 123, Issue 44, November 2026. SignificanceNeurons extend axons to wire the nervous system, and the same molecular machinery that builds these connections must later stop growth, hold connections stable, and regenerate them after injury. Axons sharing this machinery can elongate ...

We present a low-lag dataset of monthly CO2 emissions from fossil fuel consumption and industrial processes in Japan, derived directly from official statistical data. The monthly dataset can be used to estimate the annual emissions a year ahead of the official estimates submitted to the United Nations Framework Convention on Climate Change (UNFCCC). Such monthly emissions estimates will provide further opportunities for understanding in a timely manner the reasons behind emissions changes, leading to targeted policy changes for climate mitigation.

Science Advances, Volume 12, Issue 45, November 2026.

In this study, we use historical simulations (1902-2014) from the Irrigation Model Intercomparison Project (IRRMIP) to provide the first multi-model quantification of irrigation-induced effective radiative forcing (ERF). We find that the global multi-model mean ERF is positive but small (~0.023 W m⁻²). This small global signal masks regional heterogeneity: South Asia and West Central Asia show higher positive ERF (~0.23 W m⁻²), whereas East Asia exhibits a negative signal (~ −0.03 W m⁻²).

Science, Volume 394, Issue 6824, Page 654-659, November 2026.

Journal of Geophysical Research: Atmospheres, Volume 131, Issue 21, 16 November 2026.

by Yuchen Wei, Fang Guo, Yawen Wang, Zihao Guo, Guozhang Lin, Conglu Li, Shi Zhao, Chao Ren, Kin Fai Ho, Eng Kiong Yeoh, Ka Chun Chong Background Climate change-induced temperature variability poses growing health risks. The effects of this variability on mortality are well established; however, its specific impact during extreme heat and cold conditions remains understudied. This represents a knowledge gap, as the human body’s physiological capacity to adapt to sudden temperature shifts is compromised when ambient temperatures are extreme. To address this, our study examined the relationship between mortality and temperature variability in Hong Kong for a 21-year period. Methods A time-space-stratified case-crossover design was used in this study. Daily weather data from ERA5-Land and PM2.5 levels from ChinaHighPM2.5 were combined with mortality records in Hong Kong from 2000 to 2020. Diurnal temperature range (DTR) and temperature changes between consecutive days (TCN) were used as the temperature variability metrics. Extreme hot and cold conditions were defined by annual temperature percentiles: > 75%, > 90%, > 95%, > 97.5% for hot seasons, and Results DTR and TCN were strongly associated with mortality during heat extremes. For temperatures >97.5th percentile, DTR (odds ratio [OR] = 1.99, 95% confidence interval [CI]: 1.90-2.08) and TCN (OR = 2.24, 95% CI: 2.09-2.40) showed increased risks for all-cause mortality. During cold extremes ( Conclusion Temperature variability was associated with an increased risk of mortality during extreme heat conditions, with mortality risk rising alongside higher extreme temperature levels, while during extreme cold, the associations were generally smaller. At temperatures above the 97.5th percentile, each 1°C increase in DTR and TCN was associated with approximately 99% and 124% higher odds of all-cause mortality, respectively. These findings highlight the need for targeted public health strategies to protect vulnerable populations from climate-induced temperature fluctuations.

arXiv:2611.08148v1 Announce Type: cross Abstract: Weather fronts are transition regions between air masses with contrasting thermodynamic properties and are associated with mid-latitude precipitation and severe weather. We investigate machine learning (ML) methods for detecting weather fronts and quantify the contribution of reanalysis data to performance. We compare two configurations of the same ML model for pixel-wise frontal classification over the North Atlantic and Europe: (1) using only geostationary infrared satellite observations, and (2) additionally using 28 ERA5 reanalysis fields. The satellite-only model achieves a neighbourhood critical success index at a 150 km tolerance (CSI@150) of $0.402$, which lies within the range of agreement between analyses from different meteorological services. The satellite$+$ERA5 model achieves $0.575$, an improvement of $0.173$. Our analysis shows that low-level winds are the least predictable from the satellite representation ($R^2 = 0.23$, compared with $0.88$ for temperature), while removing winds or moisture causes the largest decreases in detection performance.

Climate warming is altering seasonal snow regimes and vegetation phenology in cold regions, but the influence of snowmelt may extend well beyond spring thaw. We examined how snow-related associations with net primary productivity (NPP) changed across the growing season in the Naoli River Basin during 2014–2024. Monthly NPP was reconstructed using the Carnegie–Ames–Stanford Approach and evaluated against MODIS NPP. The growing season was divided into early (EGS), peak (PGS), and late (LGS) stages based on vegetation phenology. Generalized additive models identified contemporaneous, lagged, and cumulative associations of air temperature, precipitation, solar radiation, and snow water equivalent (SWE), while random forest and partial dependence analyses were used to compare driver importance and SWE–NPP responses. SWE showed antecedent associations, with AIC-selected April-referenced windows A0L4 (EGS; R2 = 0.23) and A0L3 (LGS; R2 = 0.24). Solar radiation remained the most important predictor, but its relative importance declined from 53.77% in the EGS to 42.56% in the LGS. SWE was more important than precipitation in the EGS (18.96% vs. 12.38%) and PGS (17.31% vs. 13.86%), whereas precipitation became more important in the LGS (19.75% vs. 15.13%). The SWE–NPP relationship also shifted from an overall declining response in the EGS to a nonlinear response in the PGS and a predominantly positive response in the LGS. These results highlight the value of considering antecedent snowpack conditions alongside growing-season climate in vegetation productivity monitoring, with implications for sustainable watershed management under climate change.

Science Advances, Volume 12, Issue 45, November 2026.

Journal of Advances in Modeling Earth Systems, Volume 18, Issue 11, November 2026.

ABSTRACT Regional variability and long‐term changes of past ecosystems likely had a strong impact on hunter–gatherer population dynamics, including the expansion of anatomically modern humans and the disappearance of Neanderthals. However, our understanding of these ecosystems remains limited, even when looking at large‐scale patterns, such as the extent and distribution of forested areas. Vegetation reconstructions are key to better understanding ecological preferences and their influence on the spatial distribution and dispersals of early humans. Here, we compare four different vegetation reconstruction techniques, two based on pollen data and two vegetation models, between 70 and 20 thousand years before present in Europe. Absolute values in forest cover can differ by up to 40% between reconstruction techniques, highlighting that the choice of method can make the difference between a boreal forest and a tundra. Besides method‐specific and statistical uncertainties, the reason for such a substantial offset lies in the loose definition of forest cover metrics, rendering a direct comparison between methods impossible. At the same time, we show that all reconstruction techniques accurately reconstruct ecosystem dynamics over time on continental to regional scales. We therefore highly recommend that researchers use relative forest cover measures as the more robust method to estimate past landscapes.

Accurately exchanging energy between the atmosphere and the ocean-ice systems through surface heat fluxes is crucial for climate modelling. We present an improved version of Coupled Arctic Prediction System (CAPS) by revising flux coupling for conservation of heat. Our results show that the model with improved flux coupling can better simulate sea ice conditions during the period of Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC).

arXiv:2611.00918v2 Announce Type: replace-cross Abstract: Decadal changes in the Antarctic stratosphere, generally associated with ozone recovery and shifts in polar circulation, are of broad geophysical interest. However, independent long-term measurements of the atmosphere above the Antarctic interior are scarce. We assess whether the atmospheric muon rate recorded by IceCube could constrain decadal changes in the effective atmospheric temperature, $T_\mathrm{eff}$, a weighted average of temperature through the atmosphere that is most sensitive to the stratosphere. Using public atmospheric reanalysis data and records of cosmic-ray, solar, and geomagnetic variability, we reproduce the measured seasonal response of the IceCube muon rate. We then test whether the same response applies to long-term temperature changes with a different vertical temperature profile. The limiting requirement is detector stability rather than counting statistics: an unmodeled muon rate drift of $1\%\,\mathrm{decade^{-1}}$ would mimic a temperature trend of $2.8\,\mathrm{K\,decade^{-1}}$. The long-term rate must therefore be stable to approximately $0.3\%\,\mathrm{decade^{-1}}$. Within this budget, snow accumulation contributions can be constrained. For a 14-year record, atmospheric variability gives a one-sided 95% CL statistical sensitivity of $1.6\,\mathrm{K\,decade^{-1}}$. Given this sensitivity, the $\sim0.9,\mathrm{K,decade^{-1}}$ cooling indicated by the 2011-2025 atmospheric records would be constrained at the 95% confidence level after approximately 22 years of measurement.

Mineral dust is one of the major light-absorbing impurities in snow and ice, and affects snowpack and glacier melting. However, the origin and composition of mineral dust deposited on snow surfaces, and how these vary over time and space, remain poorly understood. This study examined the temporal and spatial variability in the composition of mineral dust in an alpine snowpack in the Tateyama Mountains, Japan and evaluated the relative contributions of long-range-transported and locally derived dust to dust deposition on the snow surface. Surface snow samples collected from April to July were analyzed using microscopy, SEM–EDS, and X-ray diffraction (XRD). The results showed a clear temporal transition in the mineral assemblages. The April samples were dominated by quartz and feldspar, consistent with long-range-transported Asian dust. Conversely, samples from May to July showed increasing abundances of chlorite, biotite, and amphibole, indicating a growing contribution of locally derived dust during the melt season. Coarse, feldspar-rich dust was frequently observed near diorite outcrops. These results indicate that Asian dust dominated in spring, whereas locally derived dust increasingly contributed to the mineral composition of surface snow during the melt season. This suggests that local geology plays an important role in controlling snowpack impurity composition, with potential implications for snow-surface radiative properties.

Geophysical Research Letters, Volume 53, Issue 22, 28 November 2026.

arXiv:2611.13359v1 Announce Type: cross Abstract: Wind profiles in the atmospheric boundary layer (ABL) drive short-range forecasting and wind-energy siting, yet sparse radiosonde coverage forces practitioners to interpolate them from a handful of physical predictors. We benchmark modern tabular learners against analytical parameterizations on 158,097 IGRA soundings. CatBoost, TabM, and FT-Transformer reduce RMSE to 1.3-1.6 m/s, against approximately 3 m/s for power-law and Monin-Obukhov fits; once the predictor set is fixed the three architectures differ by less than 0.05 m/s, so accuracy is driven by predictors rather than architecture. A cohort analysis of within-bin profile dispersion places an empirical floor of approximately 1.77 m/s on any point-wise predictor that uses standard radiosonde features, consistent with prior eddy-viscosity inversions. A one-dimensional Fourier Neural Operator applied as a post-hoc corrector to frozen tabular predictions reduces near-surface RMSE on U.S. stations from 1.5 to 1.2 m/s, while leaving upper-layer error unchanged. Further gains should come from new predictors rather than larger models.

Offshore the Horn of Africa, volcanic ash layers preserved in deep-sea sediments have played a crucial role in establishing and refining chronologies for paleontological sites in the East African Rift System (EARS). However, no such layers younger than 750 ka have been identified, despite evidence for major eruptions during this interval. Here, we investigate the 166-150 ka interval in marine sediment core MD96-2073, retrieved near Socotra Island. We constructed an age-depth model based on δ 18 O tie-points and 14 C dates, using Bayesian statistics (BACON software). This core spans at least 250 kyr, with a mean chronological uncertainty of ± 3.5 ka (95% confidence interval). Although no visible tephra layers were identified, glass-shard counting revealed a distinct cryptotephra horizon dated to 154.6 ka (13 ka as whole 95% confidence interval). Major and trace element analyses of glass shards allow correlation of this cryptotephra with the Konso Silver Tuff, which is considered to derive from the Shala volcano, making it the youngest EARS-derived tephra currently correlated with marine deposits. This cryptotephra represents a widespread tephrostratigraphic marker and provides a new chronological anchor for sediments of comparable age, including lacustrine, paleontological, and archaeological records.

Nature Communications, Published online: 04 November 2026; doi:10.1038/s41467-026-78966-z Author Correction: Asian summer monsoon orbital variability directly paced by CO2 and precession, not eccentricity

Abstract We present a comparison of extraterrestrial 3 He ( 3 He ET ) and excess 230 Th ( 230 Th xs ) as constant flux proxies across a depth transect on the Cocos Ridge, evaluating their behavior as a function of carbonate dissolution and lateral sediment redistribution. Overall, 60% of paired 3 He ET ‐ and 230 Th xs ‐derived mass accumulation rate (MAR) estimates overlap within 1σ uncertainty, indicating broad agreement between the two systems. However, statistically significant discordances between He‐ and Th‐derived MARs occur in three of four cores, with He‐derived MARs exceeding Th‐derived MARs at the ridge crest (positive ΔMAR values) and falling below Th‐derived MARs at flank sites (negative ΔMAR). The discordance at the ridge crest is consistent with sediment winnowing and preferential removal of 3 He‐bearing particles, whereas the smaller discordances in flank cores are best explained by a complementary focusing signal, likely reflecting redistribution of material removed from the ridge top. Evaluation of the negative ΔMAR offsets demonstrates that 230 Th xs is effectively retained during carbonate dissolution, as ΔMAR responses remain small despite strong CaCO 3 preservation gradients, and theoretical constraints indicate negligible thorium mobility. When applied jointly, the two proxies provide complementary constraints on MARs that would remain unresolved using a single proxy. However, their interpretation remains site‐dependent, requiring careful consideration of sediment composition, transport regime, and analytical limitations. Because sedimentary burial fluxes are calculated from component concentrations and MAR, these proxy‐dependent differences directly affect estimates of carbonate and other elemental burial fluxes, with implications for reconstructing both past and modern marine biogeochemical cycles.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 11, November 2026.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 11, November 2026.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 11, November 2026.

Preprint under review for ESSD (discussion: open, 0 comments) Fuel combustion is the largest source of greenhouse gases, but existing datasets either lag by years or cover CO2 only. We present a monthly dataset of CO2, CH4, and N2O emissions from fuel combustion for 153 countries and entities, by sector and fuel, from 1960 to 2025. Emissions for 2024 and 2025 are predicted from monthly activity indicators. The dataset supports carbon budget tracking, Global Stocktake assessment and near-real-time monitoring of sectoral emissions.

Abstract There has been increased attention to the Indian Ocean as a source of predictability for the early winter climate over the North Atlantic European (NAE) region in the form of a North Atlantic Oscillation (NAO)-like pattern. Particularly, the October Indian Ocean Dipole (IOD) triggers a delayed precipitation response over the Indian Ocean in early winter, which mediates the Rossby wave that propagates to the NAE sector. Seasonal forecast models have been found to be severely underestimating this teleconnection, and finding the underlying causes is crucial to the improvement of the models’ performance. We demonstrate that models lack an adequate December Indian Ocean convective response, both in terms of magnitude and spatial pattern. This misrepresentation of the heating influences the generation of Rossby Wave Source (RWS) along the subtropical jet. Employing a dry linear baroclinic model, we also find that the December upper-tropospheric mean state does not support accurate downstream wave propagation in the forecast models. Particularly, a combination of a stronger subtropical jet and erroneous polar waveguide zonally traps the wave train in the models. Furthermore, our linear model simulations also suggest that this inaccurate downstream wave propagation potentially affects the strengthening of the wave over the North Pacific.

Abstract Tree-ring chronologies were used to reconstruct growing-season precipitation for the North American Corn Belt from 1182–2023 CE. The reconstruction explains close to 60% of the variance in instrumental May–July precipitation totals (1901–2014) and reproduces the teleconnections with circulation and surface climate variables detected in the instrumental precipitation data, substantiating the skill of the reconstruction and indicating episodic co-occurrence of drought between major crop production regions of the central United States and Eurasia. The episode with five consecutive years of deficit May through July precipitation totals in the Corn Belt from 1930–1934 during the Dust Bowl drought was exceeded by several uninterrupted droughts reconstructed during the pre-instrumental period. The early 17 th century drought (1627–1639) was the most severe, sustained, and unbroken drought in the reconstruction for at least 600-years and has no analog in the modern instrumental observations. This drought coincided with decadal dryness in China and eastern Europe and provides a paleoclimatic example of simultaneous megadrought in major breadbasket regions of Eurasia and North America. The decade of wetness during the early 21 st century in the instrumental observations appears to have been unmatched for 700-years and may have been enhanced by increased humidity and radiation changes associated with the dramatic intensification of agriculture in the Corn Belt.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 11, November 2026.

We developed a new global ocean model to understand how mercury moves and changes in seawater. The model closely matches observed mercury patterns, providing confidence in its results. It will serve as the ocean part of a fully coupled Earth system mercury model (CESM2-Hg) with air, land, and river processes, improving predictions of future mercury pollution and guiding strategies to protect marine life and human health.

arXiv:2611.07878v1 Announce Type: new Abstract: The correlation ($r$) between ensemble mean forecasts and observations is a frequently-used metric of forecast quality in the weather and climate prediction communities. However, uncorrected estimates of $r$ are biased towards zero at finite ensemble size due to the influence of unpredictable noise that contaminates the sample ensemble mean. This note summarises established ensemble-size adjustment methods and characterises the behaviour of $r_{\infty}$, a correlation estimator that is constructed from an unbiased and ensemble-size-independent estimate of the signal variance. The only requirement is that forecast members are exchangeable with one another such that they can be interpreted as independent draws from a case-specific predictive distribution. The estimator $r_{\infty}$ does not require forecasts and observations to be statistically exchangeable, only that the unpredictable component of each member is uncorrelated with the observed truth, and thus works for both reliable and unreliable forecasts. In some situations, $r_{\infty}$ may be undefined (i.e. the estimated signal variance is negative) or fall outside the admissible range $-1 \leq r_{\infty} \leq 1$ when the predictable signal is small relative to the sampling uncertainty. These inadmissible values can be interpreted as evidence that the true predictable signal cannot be resolved with the available ensemble size and number of cases. The same unbiased signal variance can also be used to estimate the correlation between the predictable signal and individual members to yield an ensemble-size-invariant estimator of the ratio of predictable components (RPC). In contrast, variance-based RPC estimates, including forms that correct only the model predictable component, are biased low by a factor that depends only on the ensemble size and model signal-to-noise ratio.

arXiv:2606.19581v2 Announce Type: replace Abstract: We argue land-sea contrast is an important driver of spatial patterns of observed surface temperature and atmospheric circulation trends, including the negative Inter-Decadal Pacific Oscillation (IPO)-like tendency across the Pacific basin. Observed and modeled trends display an overall decline in sea level pressure over the faster-warming land, with a spatial pattern that resembles the seasonal cycle and modeled response to land heating. Coupled historical forcing climate model simulations underestimate the land-sea warming ratio. It is only in the early response of abrupt CO2 quadrupling climate model simulations that climate models are able to recreate the observed land-sea warming ratio, in which case a negative IPO-like sea level pressure trend and surface warming pattern over the Pacific comparable to observed trends are seen. We propose that some discrepancies between observed trends and those predicted by climate models are explained by the underestimation of the land-sea warming ratio by climate models.

Abstract. The Warwan sub-basin in Jammu and Kashmir is a remote glaciated region containing several glacial lakes and has recently experienced population growth and infrastructure development. Due to inaccessibility and geomorphic masking, multiple mass movement events including avalanches and Glacial Lake Outburst Floods (GLOFs) have remained largely unreported. This study analyzes three major avalanche events from the past two decades. The September 2005 and September 2020 avalanches originated from glaciers GL-B and GL-A within the same glacier complex, while the March 2020 rock–ice avalanche initiated from the headwall of glacier GL-F in an adjacent valley and terminated before reaching its ablation zone. Runout mapping shows that the September 2020 avalanche descended from the headwall of GL-A and impacted its proglacial lake, triggering a GLOF. In contrast, the September 2005 event terminated before reaching the glacial lake that began forming in 1999 at the terminus of GL-B. Geomorphic analysis indicates persistent sediment influx from meltwater streams of GL-D and GL-E into the lake associated with GL-A, progressively infilling the basin. Continued glacier retreat caused meltwater to accumulate behind the sediment infill, dividing the basin into two disconnected lakes. Impact from the September 2020 avalanche led to complete drainage of stored water and sediment, generating a downstream debris flow. Pre- and post-GLOF imagery reveals breaching and widening of the outflow channel and deposition of a debris fan downstream. The repeated history of mass movements and continued growth of glacial lakes raise serious hazard concerns in the Warwan sub-basin. GLOF modeling identifies Lake-B as potentially hazardous lake at the present condition posing significant downstream risk. Settlements such as Youdu and Qaderna, along with bridges, roads, and residential structures, lie within potential GLOF inundation zones. High sediment availability from past mobilization and deposition along the valley increases the potential of debris flow cascades, posing elevated risks to downstream infrastructure and hydropower, underscoring the need for preparedness and mitigation planning. GLOF modeling suggest that early warning system closer to the source will be able to provide a good lead time in case of potential GLOFs in the valley.

The glacier-lake outburst flood (GLOF) or another upstream hydrological eventthat struck Nepal on 26 August 2026 has resulted in catastrophic loss of life and severe disruption of essential infrast...

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 11, November 2026.

Glaciotectonic landforms include a spectrum of hill-hole pairs, composite ridges, push moraines, cupola hills, and drumlins with deformed internal structures. Their regional distribution as well as buried glaciotectonic features are reviewed with an emphasis on Scandinavia, adjacent Russia, and Baltic nations. LiDAR and ground investigations were the keys for recognizing murtoos as a new category of distinct glaciotectonic landforms, which are particularly well documented in Finland and Sweden. Three main zones of glacial landscapes are recognized for ice-sheet glaciation - outer, intermediate, and inner, as well as a transitional boundary between the outer and intermediate zones. Glaciotectonic phenomena are numerous and widespread in all of these zones wherever the substratum consists of sedimentary bedrock or unconsolidated sediments.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 11, November 2026.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 11, November 2026.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 11, November 2026.

Nature Communications, Published online: 09 November 2026; doi:10.1038/s41467-026-78736-x The authors develop a machine learning force field trained only on quantum-chemical data that can reproduce measured key polymer properties such as densities and glass transition temperatures.

Nature Communications, Published online: 09 November 2026; doi:10.1038/s41467-026-78922-x Collective memory shapes how societies respond to climate extremes, but its long-term biases are hard to observe. Here, the authors show that historical severity assessments of droughts and floods in northern China over 430 years are biased toward recent experience, consistent with an agent-based model of generational amnesia.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 11, November 2026.

Nature Climate Change, Published online: 10 November 2026; doi:10.1038/s41558-026-02789-5 Scientific grounds for a net-zero policy assume near-zero committed warming over a few decades after CO2 emissions cease. A modelling study suggests that if prior peak temperatures rise above 2 °C before cessation of CO2 emissions, committed warming increases, making it harder to stabilize global warming.

Nature Climate Change, Published online: 09 November 2026; doi:10.1038/s41558-026-02797-5 It has been surprisingly difficult to calculate how much warming different regions of the Earth have experienced since pre-industrial times. Now, research shows that in most land regions, the temperature thresholds that governments agreed to in Paris to keep below are already dangerously close.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 11, November 2026.

Nature, Published online: 11 November 2026; doi:10.1038/d41586-026-03621-y Agricultural biotechnology once found itself in opposition to environmental science. Now it could reduce the impact of food systems on global warming and biodiversity loss.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 11, November 2026.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 11, November 2026.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 11, November 2026.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 11, November 2026.

Journal of Geophysical Research: Solid Earth, Volume 131, Issue 11, November 2026.

This study investigates a rapidly surging glacier in the Pamir Mountains using daily satellite imagery and unmanned aerial vehicle (UAV) data. It reveals that a surge-detachment sequence can rapidly displace ice downstream and transform valley morphology. The findings show that even small glaciers can produce large, rapid environmental impacts and highlight the need for detailed observations to understand glacier instability and improve hazard assessment in a warming climate.

arXiv:2512.24525v3 Announce Type: replace Abstract: Ensemble forecasting has proven to be a vital tool for predicting extreme, life-threatening or only partially predictable weather events. As well as providing probabilistic products, individual ensemble members provide context and realisations of possible extreme weather. However, many National Meteorological Services in East Africa lack the computing resources required to run their local area models in ensemble mode over the full period of the two-week medium range. In this paper we test the performance of a short and medium-range forecast system recently put into daily use, comprising the global ECMWF ensemble forecast, post-processed using cGAN, a neural network model. We find that cGAN achieves probabilistic forecast skill comparable to that obtained using isotonic distributional regression (IDR), a recent statistical post-processing method, applied to either the ECMWF ensemble or the machine-learning models FuXi and GraphCast. Unlike IDR, cGAN generates individual ensemble-member realisations (potentially many thousands), including exceptional weather events. This ability was deemed important when assessed in an operational context by the National Meteorological and Hydrological Services of Kenya and Ethiopia. cGAN is cheap to train/run and requires no additional post-processing. It is run on laptops making it practical as a high-resolution medium-range ensemble forecast system for Meteorological Services with limited computational resources.

Joint quantification of glacier terminus change and adjacent secondary landslide evolution can be challenging when multi-temporal UAV datasets lack a consistent absolute spatial reference, thereby limiting robust assessment of their stage-dependent covariation. This study investigated the Hailuogou glacier terminus and the Laoguanjingtai (LGJT) landslide using six UAV photogrammetric surveys acquired between 2018 and 2022 and synchronous airborne light detection and ranging (LiDAR) data from 2022. LiDAR reference constraints and elevation-residual assessment over relatively stable terrain supported multi-temporal digital elevation model (DEM) differencing. The glacier terminus area decreased from 2.63×105m2 to 1.43×105m2, representing a reduction of 45.5%. From July 2020 to August 2022, the glacier volume ablation rate reached 1.00×106m3a−1. Meanwhile, the LGJT landslide shifted from deposition-dominated adjustment from 2018 to 2019 to erosion-dominated material export after July 2020, when its erosion rate reached 3.73×105m3a−1 and its net volume change rate decreased to −3.39×105m3a−1. Annualized rates across five unequal intervals showed positive covariation between glacier ablation and landslide erosion and negative covariation with net landslide volume change. These exploratory associations, together with the spatial concentration of elevation change near the glacier–landslide margin, are consistent with a geomorphic linkage. The LiDAR-constrained UAV workflow provides a practical basis for integrated monitoring of glacier retreat and adjacent sediment redistribution.

Snow cover fraction (SCF) reflects the regional snow cover changes in arid mountainous areas. We developed a terrain-aware graph convolutional network–long short-term memory network (GCN-LSTM), where 9 meteorological stations in Xinjiang serve as graph nodes and 15 monthly meteorological variables as dynamic node features, while altitude, slope, and aspect define a fixed terrain information graph. The overall study period spanned March 2000–June 2022; the model uses a 12-month input window and predicts regional snow cover fraction in Xinjiang one month in advance under strict time series partitioning. On the 30-month test set (January 2020–June 2022), the model’s mean seed-level root mean square error (RMSE) was the lowest point estimate (0.0386); paired block-bootstrap intervals for ensemble-prediction differences from monthly climatology and ridge included zero. The model followed the pronounced seasonal cycle, but its five-seed ensemble had a deseasonalized anomaly R2 of −0.009, providing limited evidence of predictive skill beyond seasonality. The terrain-weighted graph had lower test RMSE than the geographic graph but similar error to the terrain-binary graph; intervals for most other graph comparisons crossed zero. Radiation accounted for 31.28% of total grouped SHAP (Shapley Additive Explanations) attribution and Lag-1 meteorological inputs for 36.24% of lag attribution. Removing the Moisture group during retrained ablation increased ensemble test RMSE by 0.0033. Different rankings indicate that attribution analysis and ablation analysis capture different aspects of model dependencies. Across seven retrospective annual outer Test folds (84 months), the ensemble did not show a stable improvement over monthly climatology (pooled RMSE 0.03896 vs.0.03763); terrain information remains a spatial prior whose broader predictive value requires further evaluation.

arXiv:2502.21006v2 Announce Type: replace Abstract: Bottom-intensified turbulent fluxes observed near the seafloor drive abyssal densification and diapycnal downwelling through their vertical divergence. This appears paradoxical under steady stratification because dense bottom waters continuously enter from Antarctic sources. Motivated by evolving Indo-Pacific stratification, we use idealized simulations to examine abyssal overturning under transient and steady-state conditions. Using a kinematic rather than thermodynamic water-mass-transformation approach, we diagnose advective overturning, isopycnal volume tendency, and diapycnal transport in two cooling experiments with different mixing intensities, one warming experiment, and one quasi-equilibrium experiment. Advective upwelling occurs in all four, whereas diagnosed diapycnal transport varies in direction. In Type-I cooling, the isopycnal volume tendency exceeds advective overturning in magnitude, yielding downward diapycnal transport; in Type-II cooling, advective overturning exceeds the isopycnal volume tendency, yielding upward diapycnal transport. The warming and quasi-equilibrium experiments also exhibit upward diapycnal transport. These relationships persist qualitatively in potential-temperature coordinates. Type-I cooling also spontaneously produces an observationally documented bottom-intensified buoyancy-flux profile with an e-folding scale comparable to previous estimates. Thus, cooling alone does not determine diapycnal transport direction; it depends kinematically on the relative strengths of advective overturning and isopycnal volume tendency. Within this idealized framework, the results provide a transient kinematic interpretation of the interior-downwelling conundrum: abyssal waters can rise through the basin interior while undergoing diapycnal downwelling, without requiring an upslope return pathway within the bottom boundary layer.

This study introduces a verification method that accounts for differences in grid resolution when evaluating extreme event forecasts. We apply it to an artificial intelligence-based weather prediction model and a high-resolution numerical weather prediction model. Results show that, when assessed on equivalent neighborhood scales, the high resolution numerical weather prediction model only outperforms the AI system for short lead times in predicting extreme precipitation.

Climate science is increasingly adopting cloud-based infrastructures for data access and analysis. Contrary to the widespread belief that traditional formats such as NetCDF/HDF5 are unsuitable for remote cloud access, this study demonstrates that they can perform efficiently in cloud environments. These findings challenge current assumptions and provide practical insights that could support CMIP7 data workflows while reducing the need for major changes to existing climate data infrastructures.

Polar oceans are covered by many pieces of floating ice, called floes, that vary in size, shape and thickness. This study focuses on how fracturing caused by stresses from wind and ocean currents changes floe sizes. Using satellite images, we track ice motion and floe sizes, and study floe shapes and how deformation alters the distribution of floe sizes. We find that stronger deformation creates more small floes and present a formula to include this effect in sea-ice models.

Proceedings of the National Academy of Sciences, Volume 123, Issue 44, November 2026.

We developed open-source software that simulates how elastic and electromagnetic waves travel through snow and ice with the aim to advance and make accessible ice and snow imaging methods. This helps researchers test survey designs and interpret field measurements without expensive or risky expeditions. Parameters influencing the mechanics of snow and ice are easily input from JSON files, and geometries can be generated using an image editor allowing for time effective production of models.

Injecting fluids underground for energy projects can trigger earthquakes. Predicting these events is difficult because underground pressure changes slowly, while faults can move in a fraction of a second. We developed a computer method that switches between two ways of calculating these processes. In our test, it reduced computing time by 62 % while giving similar estimates of earthquake size and total fault movement. Differences in the fastest movements show that further testing is needed.

High-resolution weather models over mountains need terrain smoothing to avoid crashing, but standard methods smooth everywhere regardless of need. We developed Slope-Constrained Orography, a tool that uses one slope-angle setting to adjust only the points that are too steep. Tested over the southeastern Tibetan Plateau, stable simulations were achieved while modifying far less terrain than standard smoothing, offering a simpler approach to running weather models over steep terrain.

This study presents the first open-source release of a tool that automatically generates chemical mechanisms for the atmospheric oxidation of organic compounds. Simulations for five hydrocarbons, under conditions ranging from urban to remote, are compared with a benchmark mechanism. Results show that describing successive oxidation steps is key to simulating the formation of low-volatility and water-soluble species important for aerosols. A reduced mechanism reproduces most of these results.

Visible images from unoccupied aerial systems (UAS) can be used to map snow, especially where snow is shallow and does not last long. We tested several methods for identifying snow-covered areas using UAS images collected in southern New Hampshire, USA. We found that using the full color information gives the most reliable results. We showed how UAS surveys can be planned to create accurate snow cover maps, helping track snow in changing conditions.

Weather over Southeast Asia is shaped by complex interactions between the air, ocean, and waves. We developed a high-resolution model that links these processes to better understand and predict rainfall patterns in the region. The results show that ocean–atmosphere interactions improve the representation and prediction of squall events that affect daily life in Singapore and surrounding regions. This work supports more accurate weather forecasting in the region.

Many atmospheric models use shallow approximation that work well when the model top is low. As models top now reach higher layers, these approximation need to be relaxed. We develop a deep atmosphere dynamical core(GMCORE), including full deep atmosphere dynamics, and spherical geometry. The tests case with idealized baroclinic waves and tropical cyclones show that the new model reproduces known results and can alter storm trajectory, suggesting value for future high-top and tropical studies.

Hydrological models help scientists understand water movement and predict changes in water resources. However, their increasing complexity makes them difficult to use. This study developed an open-source framework to simplify the setup and application of the Variable Infiltration Capacity model. Tests in multiple basins showed that the framework improves efficiency, provides reliable simulations, and supports wider use of hydrological models.

Particles in the atmosphere can trigger the formation of cloud droplets, affecting cloud properties and climate. This study evaluates a new global dataset of these particles with measurements from 25 sites around the world. The variability in time and space and their conditional formation behaviour is analysed. The authors identify systematic biases and introduce a simple correction based on observations that greatly improves the dataset's accuracy.

We introduce a deep learning approach that extracts shared operating patterns from diverse, data-rich reservoirs and transfers this knowledge to limited-data reservoirs, yielding skillful simulations of storage and release. Using satellite observations, the model also reconstructs historical behavior in no-data reservoirs. Together, this work provide a scalable foundation for addressing pressing data gaps in reservoir operations and better supporting large-scale and long-term water management.

Groundwater strongly influences exchanges of water and energy between land and atmosphere, but detailed simulation is computationally expensive. We developed a faster approach that preserves groundwater–land surface interactions. In a representative watershed within China's Pearl River Basin, it ran 18.24 times faster while closely matching the physics-based model, offering a practical way to represent groundwater in large-scale environmental modeling.

Rapidly evolving clouds over the tropical area, where complex terrain and tropical weather conditions interact, are difficult to predict. This study uses frequent, high-resolution Himawari-9 satellite observations to evaluate how well a weather model predicts cloud amount, cloud type, cloud height, and daily cloud changes over this region. The aim is to support future model development, improve cloud forecasts, and make better use of satellite observations in cloudy areas.

Snow droughts can threaten water supplies, ecosystems, and communities, but studies often use different methods and data, making results hard to compare. We developed an open-source Python tool that combines snow drought detection methods, data download, mapping, and standardized output in one workflow. Tests with global snow data show that the tool identifies annual, monthly, and daily events consistently. This makes snow drought studies easier to repeat, compare, and use in planning.

Proceedings of the National Academy of Sciences, Volume 123, Issue 45, November 2026. SignificanceColonic dysmotility is a problematic and hard-to-manage consequence of opiate use. The effects of opioids are thought to reduce or block propulsive, peristaltic contractions. We found that opioid receptor agonists act primarily on inhibitory ...

Proceedings of the National Academy of Sciences, Volume 123, Issue 45, November 2026. SignificanceQuorum sensing (QS) enables bacteria to coordinate group behaviors via chemical signals. CqsS is a membrane-bound, two-component kinase-phosphatase inVibriothat detects the autoinducer CAI-1 and regulates downstream phosphorylation. Here, we ...

Proceedings of the National Academy of Sciences, Volume 123, Issue 45, November 2026. SignificanceSurname changes have long been considered integral to immigrant assimilation into the American mainstream and an obstacle for longitudinal research on immigrants. Using a newly available source—Social Security applications from 1936 to 1970—we ...