Atmospheric and Oceanic Sciences

Abstract Extreme wet‐cold events (WCEs), characterized by persistent freezing rain and/or heavy snowfall resulting in damage to power transmission, communication and transportation, are the most severe compound disasters during the cold season in southern China. In this study, we first analyzed the observational trend of WCEs using daily gauged precipitation and 2 m‐temperatures from 753 stations across mainland China from 1960 to 2020. The results show a decreasing trend in the total WCE days. Then the historical runs from the 26 models participating in Coupled Model Intercomparison Project Phase 6 (CMIP6) are analyzed, and the results suggest that the models can simulate this trend generally well but have a greater magnitude of reduction along with evident uncertainty. Thus, we constrain the CMIP6 models by using the lower‐tropospheric temperature inversion as the observable variable based on the emergent constraint. Five models are selected by this constraint, and their simulations significantly reduce the uncertainty in WCE projections, encompassing trend, intensity, occurrence, spatial extent, and duration, relative to the original 26 models. Accordingly, future changes in WCEs under two emission scenarios, SSP245 and SSP585, are projected based on constrained models. The results suggest that WCEs will intensify in the future, while their occurrence, spatial extent and duration will decrease. A further circulation mechanism analysis based on a self‐organizing map suggests that, alongside surface warming, changes in the optimal circulation patterns for WCEs help explain the projections. The finding that the intensity of WCEs will intensify in the future is of special importance for disaster preparedness.

Abstract. Top-down atmospheric CO2 inversions are essential for estimating surface carbon fluxes, yet significant inter-system discrepancies highlight an incomplete understanding of how observational information is transferred to flux estimates. This study introduces a diagnostic strategy to explicitly investigate this information transfer, primarily in an Ensemble Kalman Filter (EnKF) system, with a comparative analysis of 4D-Var. Using Monte Carlo simulations, we analyze the spatial and temporal correlation patterns between CO2 concentrations and fluxes, which play a crucial role in the inversion process by tracing information flow via the influence matrix. Our results reveal that these correlation scales are fundamentally set by the prescribed autocorrelation structure of the prior fluxes, rather than by atmospheric transport processes alone. We identify a resonance-like effect wherein correlated fluxes amplify concentration-flux correlations, while uncorrelated fluxes suppress them. The absence of this suppression for prescribed fluxes (e.g., anthropogenic emissions) can cause systematic signal misattribution. We further demonstrate that 4D-Var relies also heavily on flux autocorrelations due to its cost function's localized gradient. In both methods, the prior's critical role is mediated through the transitivity of strong autocorrelations. Simplified observing system simulation experiments corroborate these diagnostic findings: under the current sparse surface network in East Asia, a relatively longer correlation length (e.g., 600 km) is better than a short length (e.g., 100 km). This process-oriented perspective offers practically useful mechanistic insights for reconciling inversion results, optimizing observing networks, and strengthening carbon budget assessments.

During geomagnetic storms, ionospheric disturbances can undergo substantial spatiotemporal restructuring and affect high-precision GNSS applications. This study investigates the multiscale ionospheric response over Brazil during the intense geomagnetic storm of 12 November 2025 and examines the associated changes in precise point positioning (PPP) convergence. Multi-source observations, including GNSS TEC/dSTEC, ROTI, JPL Global Ionospheric Maps, ionosonde parameters, and three-dimensional ionospheric tomography, were jointly analyzed. The results show that the storm produced pronounced and nonuniform global TEC anomalies, with the Brazilian sector embedded in a disturbed background. Over Brazil, clear traveling ionospheric disturbance (TID) propagation and ROTI enhancement were observed during the main response phase. The TID developed after approximately 02:05 UT and reached its maximum intensity during 02:25–03:00 UT. Ionosonde observations indicated decreased foF2 and increased h′F2, suggesting electron density depletion and an apparent uplift of the F-region reflection height. The GNSS dSTEC-constrained tomographic reconstruction suggested that the relative perturbation structures were more evident at 150–400 km, especially near 250–350 km. PPP analysis further revealed longer convergence times on the storm day, particularly in the vertical component. These results indicate that the Brazilian ionosphere experienced a multiscale response from global anomalies to regional propagation and vertical restructuring, which was associated with delayed PPP convergence performance.

Abstract Reliable large‐sample hydrological predictions require systematic benchmarking and careful model selection. However, this process is challenging due to structural uncertainty among models, high computational demand, and the climatic and physiographic diversity of a region. This study benchmarked 47 conceptual rainfall–runoff models across 159 watersheds in Peninsular India using the Modular Assessment of Rainfall–Runoff Models Toolbox (MARRMoT) and applied a consistent performance threshold to identify plausible simulations. The benchmarking framework enabled systematic evaluation of model adequacy, complexity, and uncertainty, followed by selection of a compact set of models. Results showed that 141 watersheds met the performance threshold, although model performance varied considerably, with semi‐arid regions posing the greatest challenge and tropical regions exhibiting high equifinality. A compact set of five models (HBV, GR4J, MODHYDROLOG, SMA, and Hillslope) was found sufficient to reproduce reliable streamflow in over 85% of watersheds, while an additional five models extended robust flood prediction to the same coverage. Storage analysis revealed that soil moisture was the dominant process representation, followed by routing and interception storages, and the most reliable models required no more than three storages. Parameter complexity also influenced performance: models with intermediate complexity (10–15 parameters) generally performed best, although some lower‐parameter models, such as GR4J, also achieved strong performance. Model skill was further associated with watershed characteristics, including soil porosity, forest cover, and leaf area index. Overall, the results demonstrate that a small, well‐chosen set of models with appropriate storage structures and moderate complexity can reliably represent hydrological processes across Peninsular India.

Abstract Microplastics (MP) are increasingly recognized as important contaminants of agricultural soils. Yet a lack of global estimates of MP inputs to agricultural land hampers systematic quantification of both current contamination levels and accumulation over time at scales. Here, we introduce MPGRIDSv1, the first global geospatial dataset of MP inputs into agricultural soils via the application of MP-contaminated organic matter (OM) used as fertilizer. We collected, through a literature survey, the OM amendment rates, A, and the MP concentration, C, in OM for 17 and 26 countries, respectively. Upon assumptions and simplifications for OM composition, and type and appearance of MPs, we estimated high and low values of A and C for 186 countries using statistical inference methods built on 14 independent national-level covariates including waste management, agricultural practices, and environmental and socio-economic factors. High and low estimates of A and C, and the annual mass of MP inputs estimated for countries worldwide were then spatialized on selected fruit, vegetable and pulse crops in the CROPGRIDSv1 dataset. Globally, 33 to 2,316 (910 median) thousand tonnes of MP were estimated to reach agricultural soils annually, with input rates ranging from less than 0.01 to more than 100 kg-MP ha-1 year-1 worldwide. While these numbers underestimate reality and acknowledging several limitations including paucity of data and uncertainty in existing ones, we expect MPGRIDSv1 to facilitate analyses of agricultural soil degradation globally with geospatial detail.

The ongoing war in Ukraine has triggered large-scale and multi-dimensional environmental degradation affecting soils, freshwater systems, forests, agricultural landscapes, atmospheric quality, and biodiversity. Beyond immediate physical destruction, these impacts compromise ecosystem resilience, human health, food security, and long-term environmental governance. Despite extensive monitoring efforts, there remains a lack of integrated and spatially explicit analytical frameworks that systematically link observable environmental damage to discussions surrounding the concept of ecocide and post-conflict recovery planning. This study provides a comprehensive and evidence-based assessment of war-related environmental degradation across Ukraine using a mixed-methods approach combining multi-temporal geospatial analysis of satellite imagery (including Sentinel-1/2, Copernicus Emergency Management Service, NASA FIRMS, and Global Forest Watch) with a structured qualitative synthesis of governmental, intergovernmental, and civil-society reports. The approach enables the identification, classification, and spatial quantification of observable environmental disturbances across key domains, including industrial infrastructure, hydrological systems, agricultural land, forest ecosystems, protected areas, and urban environments. The results reveal widespread land-cover transformation, soil degradation and potential contamination, disruption of hydrological regimes following critical infrastructure damage, intensified wildfire activity in conflict zones, significant forest loss, and transboundary atmospheric pollution. The study develops a geospatial typology of war-induced environmental impacts and examines how these patterns may be interpreted in relation to the proposed criteria of “severe, widespread, and long-term” environmental harm associated with contemporary ecocide debates. By integrating spatial indicators of conflict-related environmental disturbance with qualitative environmental evidence, this research advances methodological tools for environmental monitoring in conflict settings and supports risk-based zoning, prioritisation of ecological restoration, and the integration of scientific evidence into environmental governance and post-conflict environmental assessment, providing a scalable and replicable framework for analysing ecosystem degradation in war-affected regions.

Abstract Recent advances in generative AI have enabled the integration of scientific knowledge into natural language interfaces. However, existing large language models (LLMs) lack domain‐specific expertise and cannot directly utilize simulation data essential for risk assessment. This study develops a climatology‐specific LLM fine‐tuned on climate science literature and augmented with ensemble projection data set, aiming to support local governments and small enterprises in climate adaptation planning as a preliminary support tool. The framework combines domain‐specific fine‐tuning, retrieval‐augmented generation (RAG) for scientific documents and local guidelines, and future projection data. The developed model demonstrated highly competitive performance compared to general‐purpose LLMs such as Swallow 70B on climate‐specific benchmarks in both English and Japanese, particularly in the categories of “Impacts, adaptation, and vulnerability” and “Mitigation.” A case study in Kumagaya City, Japan, demonstrated that the model can quantitatively update heat‐mitigation measures using probabilistic projections and dynamically retrieved local constraints, generating data‐informed baselines to facilitate human deliberation under multiple warming scenarios. The proposed approach bridges the gap between climate science and the early stages of adaptation planning by enabling AI‐driven access to both textual and numerical climate knowledge. By lowering technical barriers to expert‐level analysis, it facilitates inclusive and data‐driven climate services for diverse stakeholders. This study represents the first systematic demonstration of an LLM that integrates ensemble projection data to support regionally grounded adaptation planning, contributing to the development of next‐generation climate services that enhance local resilience.

Abstract The Last Glacial Maximum (LGM) and Last Interglacial (LIG) provide insights into the response of the North Atlantic–European climate variability to strongly contrasting boundary conditions. Using six CMIP6/PMIP4 models, we compare temperature–precipitation patterns, atmospheric circulation, and variability, with emphasis on the North Atlantic Oscillation (NAO). Across climate states, the leading modes of wintertime sea-level pressure variability remain recognisable, indicating a robust dynamical structure. However, their amplitude, explained variance, and spatial positioning exhibit state dependence. During the LGM, extensive ice sheets were associated with a stronger and more spatially confined NAO dipole, southward-displaced pressure centres, and reorganised circulation patterns, reinforcing colder and drier European conditions with enhanced seasonality. The LIG exhibits a weaker NAO expression with westward-shifted pressure centres, a more zonal circulation, northward-shifted warm sea surface temperatures, and enhanced seasonality. NAO-related interannual atmosphere–ocean coupling persists across climate states, with limitations in capturing low-frequency variability within the available simulation lengths. Summer variability is substantially weaker and more heterogeneous across all periods, with the NAO losing dominance often to the East Atlantic pattern during the LGM. Inter-model differences increase considerably under LGM and LIG conditions compared to the pre-industrial period (PI), underscoring the importance of multi-model approaches and an improved representation of key processes when investigating climate variability in extreme climate states.

Cocoa production in Ghana and Côte d’Ivoire is threatened by climate variability and extremes, particularly droughts and excessive rainfall. However, quantitative evidence on the impacts of climate change on cocoa yields remains limited, constraining the development of effective climate-smart adaptation strategies. This study assessed future climate impacts on cocoa production using regridded (~5 km resolution) ensemble projections from 12 Global Climate Models under a low (SSP1-2.6) and a high (SSP5-8.5) SSP scenario. Precipitation and temperature data were bias-corrected using five approaches: Delta Change, CDFt, SDM, EQM, and LOCI. Among these, the Delta Change method best preserved intra-annual climate variability, while temperature corrections outperformed precipitation corrections. A Random Forest model, trained on bias-corrected climate data, simulated and projected cocoa yields with an accuracy exceeding 85%, although performance varied across regions. Future changes were assessed for the near future (2026–2055) and far future (2056–2085) relative to a historical baseline (1985–2014). Ensemble projections indicate a drying trend across cocoa-growing areas, with precipitation declining by 5–10% under SSP5-8.5 and increasing modestly (around 5%) under SSP1-2.6. At the same time, temperatures are projected to rise across all regions, exceeding 3.5°C under SSP5-8.5 by the late century, particularly in central and northern zones. Projected yield responses vary spatially. Southern and coastal cocoa-growing areas are expected to experience yield declines of about 5%, with losses reaching up to 20% under severe drought conditions in highly vulnerable regions such as Dix-Huit Montagnes in Côte d’Ivoire under SSP5-8.5. In contrast, some northern and central regions may maintain or slightly increase yields under SSP1-2.6. Vulnerability is shaped by climatic, biophysical, and socio-economic factors, with regions such as Sud-Comoé (Côte d’Ivoire) and Brong Ahafo (Ghana) identified as at risk. These findings highlight the need for targeted adaptation strategies to enhance the resilience of West Africa’s cocoa sector.

Abstract A critical factor affecting overshooting tops in deep convection is the entrainment of dry free tropospheric air, which reduces updraft magnitudes and may reduce the overall storm depth. The goal of this work is to improve our understanding of the environmental factors controlling entrainment using a large data set of observed thunderstorms. We accomplish this by investigating the estimated radar‐observed level of neutral buoyancy (LNB), that is, the level of maximum detrainment (LMD), to the LNB predicted for an undiluted parcel, tropopause height, and other environmental parameters from reanalysis data over the central and eastern United States during May‐August of 2021–2022. Specifically, we consider how variability in storm modes, environments, and months might impact the magnitude of convective effective entrainment. Overall, we find that effective entrainment is lowest in environments with stronger lower tropospheric shear, a drier 2–6 km layer, higher lifted condensation level, lower most unstable convective available potential energy (MUCAPE), and higher entraining CAPE (ECAPE)/MUCAPE ratio. In particular, LMD is closer to both the LNB and tropopause within stronger lower tropospheric shear and higher ECAPE/MUCAPE ratio environments, such that these storms have greater tropopause‐overshooting potential. This work presents the first systematic evaluation over the mid‐latitudes of how parcel theory predicts storm depth over a large sample of storms and provides valuable insight into factors that impact effective entrainment. These results can be used to inform our understanding of the relationship between convective environments, the resulting storms, and the potential of those storms to reach the stratosphere.

Abstract This study quantified the modulating effects of ice-phase microphysical processes on surface precipitation intensity in typhoon spiral rainbands using convection-permitting numerical simulations (3-km grid spacing) of Super Typhoon Lekima (2019). The Weather Research and Forecasting model with Morrison double-moment microphysics scheme was employed to diagnose ice process rates and their contributions to surface precipitation. The simulated 48-hour accumulated precipitation was validated on a common 0.1° grid against GPM IMERG (V07B) retrievals, and the simulated peak accumulations along the Zhejiang coast were further compared with regional automatic weather station gauges, supporting the model’s representation of the rainband structure. Results revealed that ice-phase processes dominated precipitation formation, contributing 68–92% of surface rainfall through melting of ice particles below the freezing level. Graupel melting emerged as the primary mechanism, accounting for 52% of surface precipitation in convective cores, followed by snow melting (31%) and cloud ice-origin particles (17%). Systematic differences between inner and outer rainbands were identified, with inner rainbands exhibiting 40% higher ice water content and enhanced graupel production due to stronger updrafts. The storm-centered environmental shear remained within a moderate regime of roughly 11–14 m s −1 throughout the analysis period, and against this large-scale background a grid-point sensitivity analysis showed that vertical wind shear and mid-level humidity strongly modulated ice-phase precipitation efficiency, with moderate shear reducing efficiency from 0.78 to 0.52. Pronounced diurnal variations and marked transformations during landfall highlighted complex interactions between thermodynamic, dynamic, and microphysical processes. These findings provide insights for improving typhoon precipitation forecasts and underscore the importance of accurately representing ice-phase processes in operational weather prediction models for typhoon-affected regions.

Abstract Urban expansion replaces vegetated land with impervious surfaces, significantly reducing urban Gross Primary Productivity (GPP u ). Conversely, rising CO 2 levels and warming associated with climate change can enhance GPP u . The interplay between these opposing effects on GPP u remains poorly understood, resulting in considerable uncertainties for estimating the long‐term dynamics of urban carbon storage. Using an improved Carnegie‐Ames‐Stanford approach (CASA) model that accounts for urban CO 2 , we evaluated the long‐term trends and driving factors of GPP u in 2126 cities worldwide, spanning both historical and future periods. The study revealed that the average urban expansion rate (UER) of global cities from 1982 to 2024 was 21.3 km 2 /year. Rapid urban expansion was predominantly observed in cities in China and North America; whereas, cities in Europe exhibited relatively slower expansion rates. The response of GPP u to urban expansion was found to be dependent on the UER. In cities with faster UERs than 5.83 km 2 /year, the negative impact of urban expansion surpassed the positive effects of climate change on long‐term GPP u , resulting in a downward trend in GPP u (−2.48 g C/m 2 /year). Conversely, in cities with slower UERs (commonly observed in Europe), an opposite pattern was observed, leading to an upward trend in GPP u (1.21 g C/m 2 /year). The projected future response and driving factors of GPP u to urbanization were expected to resemble those observed in the historical period. These findings highlight that the pace of urbanization acts as a fundamental regulator of GPP u , underscoring the need for growth‐rate‐specific strategies to sustain regional carbon sequestration.

The escalating frequency and severity of marine heatwaves are driving factors behind mass coral bleaching, necessitating both rapid emission reductions and the development of scalable intervention tools. Shading coral to reduce irradiance stress is a promising approach, yet most studies focus on static structures, and few include recovery dynamics. Here, we tested the efficacy of a laboratory-scale seawater fogging system as a shading intervention during a simulated thermal/light stress and recovery experiment. An orthogonal design with two temperature levels (control (26.5 °C) vs heat-stressed ((MMM) at the collection site, 29.1 + 3.7 °C) and two light treatments (fogged, 33.3% shading ± 6.46 SD for 6 h daily vs non-fogged) was used to test the response of Acropora hyacinthus and Pocillopora damicornis over 13 days of thermal stress (3 °C-weeks) followed by a 24-day recovery period. Variable shading from seawater fog reduced mortality risk in heat-stressed A. hyacinthus by 55%. Only two mortalities occurred in P. damicornis , both in the heat-stressed treatment without fog. Fog lowered per cent whiteness and improved Fv/Fm in both species, starting at 0.84 °C-weeks, with the effects peaking near 3 °C-weeks. Fogging during the recovery period did not inhibit coral recovery but provided some additional benefits, including increased Fv/Fm and lower catalase activity. Seawater fog can not only be used to delay/reduce bleaching while thermal stress is high but can also enhance recovery and post-stress repair processes of corals. In order to advance seawater fogging as a practical reef intervention tool, engineering advances should be accompanied by in-situ trials that capture complex coral-environment interactions and assess ecosystem-wide responses.

Accelerating climate change and chronic anthropogenic pressures demand connectivity-based design of marine protected area networks (MPANs). Yet integrating dynamic species responses and cumulative human pressures into spatial planning remains a key challenge, particularly in marginal seas at the land–ocean interface. As a biodiversity hotspot under intensive development, the Yellow and Bohai Seas (YBS) exemplify this complexity, yet how global change will reshape connectivity here remains unclear. Using ensemble species distribution models, we projected suitable habitats for 70 representative marine taxa under present-day conditions and two future scenarios (SSP1-2.6 and SSP5-8.5, representing low- and high-emission Shared Socioeconomic Pathways), generating multi-species natural resistance surfaces. These were combined with an entropy-weighted anthropogenic resistance layer—comprising fishing, shipping, mariculture, and pollution—to produce integrated resistance surfaces (IRS) for each scenario. Circuit theory was then applied to quantify ecological corridors, current density, pinch points, and barriers among 61 MPAs. Our results showed that: (1) climate forcing drove pronounced habitat redistribution, with cold-adapted taxa contracting within the Bohai Sea while warm-adapted species shifted poleward, especially under SSP5-8.5; (2) the IRS intensified overall (mean rising from 1.00 to 1.22), especially in the central-northern Bohai, while the southern Yellow Sea—particularly the central trough influenced by the Yellow Sea Cold Water Mass (YSCWM)—retained relatively low resistance, indicating a potential climatological corridor; (3) although total corridor length increased modestly (6,937 to 7,634 km), functional connectivity declined sharply, with Critical Corridors decreasing from 66 to 36 and effective resistance rising by 27%; (4) Grade-1 pinch points expanded by 35% and barrier clusters nearly tripled in area, concentrating where climate-degraded habitats overlap with intense human activities. Together, these drivers create a high-resistance Bohai core and more permeable Yellow Sea margins—a north–south divergence with direct consequences for species persistence and climate-driven range shifts. Embedding global change-adjusted corridor, pinch-point, and barrier metrics into spatial planning is therefore essential for maintaining MPA network resilience in the YBS and other marginal seas.

Extreme storm tide levels, arising from nonlinear cross−scale interactions among surge, astronomical tide, and fluvial flood, threaten estuarine stability and cause major economic losses. The Bay-Inlet-Channel (BIC) system, pivotal to the Greater Bay Area, was severely impacted by Typhoon Hato, which produced record−breaking winds and severe inundation. To quantify the morphological-hydrodynamic resilience of the BIC system against the highest storm tide level (HSTL), the Delft3D model was employed to reproduce characteristics of Hato. Simulation results indicate that HSTL exhibited a sharp gradient along the Bay, with a relative increase of 63.84%, and a more moderate one in the Channel (37.00%), associated with the Channel’s higher resilience ( R G = 0.87). Under a hypothetical “Triple Coincidence” scenario involving a stronger flood discharge, the robustness of the BIC system decreased, with a more pronounced decline for the Channel (Δ R G = −0.23) than for the Bay. Contribution analysis identified surge as the dominant driver of HSTL during Hato (52–75%), followed by astronomical tide (25–51%), nonlinear interactions (−7–6%), and flood (<2%). Surge dominance diminished under “Triple Coincidence” as nonlinear interactions intensified. Momentum and energy analyses showed that lateral HSTL differences were primarily governed by direct wind stress, while longitudinal variations were modulated by morphological heterogeneity. Stronger floods amplified HSTL unevenly, mainly through enhanced nonlinear convection. These discoveries advanced the understanding of the morphological-hydrodynamic resilience and its mechanism regulating HSTL in estuarine systems, providing insights for storm tide risk management.

Abstract. The Characterising CiRrus and icE cloud acrosS the specTrum-Microwave (CCREST-M) aircraft campaign (February–March 2024) was based around the Chilbolton Observatory, UK, using the Facility for Airborne Atmospheric Measurements (FAAM) BAe-146 aircraft together with ground-based multi-frequency radars to provide a testbed for ice-cloud scattering and radiative transfer models across the microwave and sub-millimetre spectrum. Ice particle size distributions (PSDs) are retrieved from the ground-based zenith-pointing radars at the time of the radiometric overpasses, and the aircraft in-situ PSDs are used as an independent validation dataset. We present a novel hybrid retrieval framework for mid-latitude ice PSD parameters (slope λ, intercept No, and shape μ of the gamma size distribution) that combines a machine-learning (ML) ensemble with physics-based multi-frequency radar retrievals using 3, 35, and 94 GHz reflectivities. An ensemble of ML models is trained on observations from the Parameterising Ice Clouds using Airborne ObServationS and triple-frequency dOppler radar (PICASSO) campaign, also centred on Chilbolton Observatory. These models predict PSD moments from temperature, pressure, 3 GHz-retrieved ice water content (IWC), and the mean mass-weighted dimension. The ML predictions are converted into first guess gamma-PSD parameters at each height. A subsequent deterministic optimisation then adjusts No and λ, using a randomly oriented rosette-aggregate scattering model, to enforce simultaneous agreement with the observed 35 and 94 GHz reflectivities. Application of the above method to three CCREST-M cases show that the ML ensemble reproduces PSD moments well for two cases but fails when extrapolating beyond its trained temperature range in the third case. Retrieved IWCs from the 3 GHz radar compare favourably with in-situ measurements of IWC, and exponential (μ=0) and gamma PSD assumptions show comparable performance overall. Retrieved mean PSDs show generally good agreement with in-situ PSDs as a function of temperature for two of the cases, with IWCs within about 50 % of the in-situ measured IWCs over much of the −50 to −10 °C temperature range. The systematic biases seen in one case are attributed to temporal cloud evolution between radar and in-situ sampling. Independent validation using 200 GHz radar reflectivity profiles shows good agreement between the forward-modelled refllectivities and measurements above about 4.5 km. Below 4.5 km the agreement is more sparse owing to the likely presence of dendritic particles, which depart from the rosette-aggregate scattering assumption.

Abstract Hydrological droughts in Alpine regions are increasingly shaped by human regulation, with hydropower operations playing a central role in modifying their statistical characteristics. This study examines how the representation of hydropower systems in hydrological models influences the identification and statistical characterization of drought. Using the HYPERstreamHS hydrological model, we simulate streamflow in the Adige River basin under three configurations of increasing complexity, ranging from natural flow conditions, to simplified rule‐based operations, to detailed representations of reservoir management based on historical operation patterns. Drought events are analyzed from both univariate and bivariate perspectives, with the latter relying on copula‐based frequency analysis to capture the joint statistical behavior of drought duration and severity. Our findings show that conventional performance metrics, such as the Nash–Sutcliffe Efficiency and its logarithmic variant, can mask substantial differences in drought statistics across model configurations. In particular, simplified or naturalized representations systematically underestimate the frequency of hydrological droughts, failing thus to reproduce the observed dependence structure between hydrological drought attributes. Only the configuration that explicitly incorporates reservoir infrastructure and realistic operational rules is able to replicate both the marginal distributions and the joint behavior of observed droughts. These results demonstrate that hydropower regulation not only alters streamflow regimes but also reshapes the statistical properties of droughts. Capturing this influence is essential for drought risk assessment in regulated basins. More broadly, the study highlights the limitations of simplified modeling approaches and underscores the need to integrate realistic storage reservoir operations into hydrological models to support decision‐making under drought conditions.

Abstract This study examines how preconditioning from a Sudden Stratospheric Warming (SSW) alters the ionosphere‐thermosphere (IT) response to a geomagnetic storm using the Specified Dynamics version of the Whole Atmosphere Community Climate Model (SD‐WACCM‐X). Two simulations were performed to study the 2022–2023 Northern Hemisphere winter, when a strong geomagnetic storm (Kp ∼ 6) coincided with the occurrence of an SSW. One simulation included the SSW by nudging the lower atmosphere to MERRA‐2 reanalysis data for January–March 2023, while the other used January–March 2022 MERRA‐2 data to represent no SSW. Both simulations used the same solar and geomagnetic conditions for January–March 2023. Comparison of the two reveals that the storm responses in thermosphere composition and electron density can differ by up to 20%–50% due to the SSW preconditioning. The simulation including the SSW exhibits faster storm‐time recovery, consistent with earlier studies, indicating that lower‐atmosphere forcing influences storm‐time IT dynamics. This study further examines how SSW influences storm‐time processes. Under SSW conditions at mid‐to‐high latitudes, the thermosphere is cooler, and electron density is reduced relative to the no‐SSW case, lowering conductivity, weakening storm‐time Joule heating, and reinforcing thermospheric cooling. Thus, both SSW‐induced background modifications and storm‐background interactions shape the storm‐time temperature and meridional circulation and thus thermosphere composition that is, ΣO/N 2 , which then influences electron density through ion‐neutral chemistry. At lower latitudes, the SSW modifies E × B ion transport and the ion loss through recombination, both of which are key drivers of electron density changes. Consideration of lower‐atmosphere variability is therefore essential for understanding the storm‐time IT response.

Abstract Although planted forests (PFs) contribute greatly to soil and water conservation in the Yangtze River Basin and are central to afforestation efforts in China, their resilience to compound drought and heatwave (CDHW) events remains poorly understood. The unprecedented CDHW event in 2022 provided a unique opportunity to assess the responses of PFs and natural forests (NFs) to such extreme climatic disturbances. Using kernel‐normalized difference vegetation index (kNDVI), gross primary productivity (GPP), and solar‐induced chlorophyll fluorescence (GOSIF), we found that PFs experienced larger declines than NFs during the event year, but recovered more rapidly in the following year under alleviated climatic conditions, revealing a clear resistance–recovery trade‐off. For both kNDVI and GPP, NFs showed weaker declines and PFs showed faster recovery in more than 70% of grids. NFs, with their taller canopies, greater biomass, and higher species diversity, exhibited stronger resistance, whereas PFs, dominated by simpler and younger stands, demonstrated higher recovery potential. Recovery differences between the two forest types became most evident under extreme and exceptional CDHW conditions. XGBoost coupled with SHAP analysis further showed that structural differences between forest types, particularly canopy height differences, were the strongest predictors of recovery divergence, while the effects of edaphic, compositional, and climatic variables varied among indicators. These findings provide new empirical evidence that NFs and PFs play complementary roles in ecosystem stability and recovery, offering critical implications for water resource management under the projected intensification of CDHWs.

Abstract Bayesian inference offers a flexible framework for parameter estimation and uncertainty quantification in eco‐hydrological models. However, simultaneously achieving robust posterior exploration and high computational efficiency for multimodal, high‐dimensional, and computationally intensive targets remains challenging for the widely used Markov chain Monte Carlo (MCMC) and sequential Monte Carlo (SMC) methods. In this study, we developed the Parallel Adaptive Transition Particle Evolution Metropolis Sequential Monte Carlo (PATPEMS) algorithm, which is an adaptive and parallel SMC sampler for posterior distributions of model parameters in offline calibration. PATPEMS employs an adaptive sequence of intermediate distributions to control weight degeneracy and automatically select stages, a flexible scheduling of MCMC proposal kernels used to rejuvenate particles, together with reflection boundary handling to maintain particle diversity, and a particle‐level parallelization scheme to exploit multicore architectures and reduce wall‐clock time for computationally intensive models. Performance is assessed on four case studies: two synthetic targets probing multimodality and high‐dimensional dependence, and a land surface model (LSM) with six parameters constrained by synthetic and real observations. Across all cases, PATPEMS provides close approximations to the target posteriors, judged against the analytic ground truth or reference solutions. For the LSM, parallelization yields substantial wall‐clock speedups over the original non‐parallel implementation. Compared with the original particle evolution Metropolis sequential Monte Carlo (PEM‐SMC) algorithm, these results indicate that PATPEMS provides a more adaptive and parallel framework for robust Bayesian calibration of multimodal, correlated, and computationally demanding land surface and environmental models.

Cryospheric landforms play a critical role in alpine hydrology and ecosystems. Using historical and contemporary data spanning nearly six decades (1967–2024), we assessed elevation change for glaciers, rock glaciers, and perennial snowfields and the thermal response of streams in the Teton Range, Wyoming, United States. Glaciers and snowfields thinned at −0.84 ± 0.07 meters per year (m year −1 ) and −0.59 ± 0.04 m year −1 between 2014 and 2022, a ~7-fold increase relative to 1967–2014, driven by warming summer temperatures. In contrast, rock glaciers are near equilibrium (−0.05 ± 0.05 m year −1 ) and saw no change in rate. Since 2015, snowfield-fed streams have warmed rapidly (+3.4°C), whereas glacier- and rock glacier–fed streams have warmed at lower magnitudes (+0.9° and +0.6°C, respectively). Our results demonstrate the greater resilience of rock glaciers to atmospheric warming, highlighting the critical role that these features will play as glaciers and perennial snowfields are lost.

Municipal wastewater treatment plants (WWTPs) are essential for protecting public health, however, their contribution to greenhouse gas (GHG) emissions has often been overlooked. Achieving carbon-neutral operation requires more than incremental improvements in energy efficiency; it calls for a rethinking of process design, energy flows, and resource recovery strategies. This review examines recent developments across several key pathways, including carbon capture through A-B configurations, energy recovery via anaerobic digestion, and low-carbon nitrogen removal based on autotrophic processes such as partial nitritation–anammox. Emerging technologies, such as microalgal and bioelectrochemical systems, are also reviewed, although their large-scale applicability remains uncertain. Particular attention is given to the trade-offs introduced by advanced treatment for micropollutant removal, which can significantly increase energy demand if not carefully integrated. Beyond individual technologies, the paper highlights the importance of system-level optimization, life-cycle assessment, and data-driven control strategies. A staged roadmap is proposed to distinguish near-term improvements from longer-term transitions. Rather than presenting a single solution, the review emphasizes that feasible pathways depend strongly on local conditions, including influent characteristics, climate, and energy mix.

Abstract Radar variables are commonly used to identify deep convection in thunderstorms. Polarimetric radar provides additional signs of deep convection with columnar regions of increased differential reflectivity ( Z dr ) and specific differential phase ( K dp ). However, the prevalence of polarimetric signals across all trackable storm objects observed by radar is not well understood and is necessary context for understanding the prevalence of polarimetric parameters in deep convection. In this study, we objectively track all thunderstorm objects observed by WSR-88D NEXRAD radar on eight storm days in the southeastern United States. The case days are selected from the Verification of the Origins of Rotation in Tornadoes Experiment-Southeast (VORTEX-SE) and Propagation, Evolution, and Rotation in Linear Storms (PERiLS) field campaigns and encompass a wide range of storm modes, including quasi-linear convective systems, a storm type for which its hazards are notoriously difficult to forecast. Storms are objectively identified and tracked using the open-source analysis tool Tracking and Object-Based Analysis of Clouds (tobac), and over 2800 tracked cells are identified over the eight case days. The tobac provides a feature mask at every time step of a tracked cell. Within each cell shape, we examine the prevalence of Z dr and K dp columns above the melting level and lightning activity. We find that columns of Z dr are widely prevalent, even in thunderstorms without lightning or any other polarimetric column features, but still meeting the tracking threshold of 30 dB Z . However, intense K dp columns are related to lightning production—columns with lightning are over 2 times stronger than those without lightning. The microphysical development in the storm leading to electrification and polarimetric column observations is also discussed. Significance Statement The purpose of this study is to better understand how prevalent microphysical radar signatures are in the presence of lightning. This is important because thunderstorms electrify via collisions between ice and hydrometeors within the cloud. Cloud electrification is initiated within the mixed-phase region of thunderstorms, where populations of these particles are found in large quantities, and can be observed with polarimetric radar. Our results are a step toward understanding relationships between lightning and what we can observe with radar for all storm types.

Abstract Oceanic mesoscale eddies are pivotal in redistributing mass and heat. Despite a warming‐induced increase in global oceanic stratification, previous studies have reported a global increase in mesoscale eddy activity, particularly in eddy‐rich regions, pointing to a complex and regionally contingent response to climate change. Here we show that the eddy kinetic energy (EKE) over the two wings of the tropical Indo‐Pacific Ocean shows an opposite trend over the past three decades, with a pronounced decreasing in the western tropical Pacific Ocean (WTPO) but a significant increasing in the southeastern tropical Indian Ocean (SETIO). We find that the EKE decline in the WTPO is mainly due to weakened intra‐seasonal wind work, while the EKE increase in SETIO is mainly driven by weakened stratification in the north and enhanced shear in the south, both of which are associated with the intensified Indonesian Throughflow.

High Resolution Image Download MS PowerPoint Slide The burden of wildland fires to human health is well established, but less is known about how these impacts are distributed among vulnerable population groups and whether and to what extent wildland fire smoke affects certain parts of the country repeatedly. Moreover, much of the literature has focused on impacts from fine particles, but very few researchers have investigated deaths and illnesses attributable to ozone associated with wildland fires. This paper uses a 12-year time series of CMAQ model-predicted PM 2.5 and ozone concentrations between 2007 and 2018 and the BenMAP-CE tool to quantify the number, distribution, and economic value of these impacts. We calculate measures of inequality, including the Gini coefficient, to determine how these impacts are distributed among populations and locations. We estimate cumulatively tens of thousands of wildland fire PM 2.5 - and ozone-attributable deaths, valued at $130B. Impacts occur disproportionately among those most vulnerable, as characterized by socioeconomic status in the Social Vulnerability Index, who experience a greater air pollution mortality burden as compared to those less vulnerable. A relatively small number of counties in northern California and Idaho experience a disproportionately high number of days of wildland fire PM 2.5 exposure. The Gini coefficient and Atkinson index indicate that counties with vulnerable populations are inequitably impacted, suggesting an opportunity for policies that could promote a more equitable burden of smoke to human health.