Atmospheric and Oceanic Sciences

Abstract In snow‐fed, mountainous karst regions, aquifers sustain and are connected to streamflow via springs and distributed groundwater exchanges (DGE) including inflows (DGI) and losses (DGL). Despite their significance, there is limited understanding of the complex flow paths linking aquifers and streams, particularly regarding flow path source and residence time, and thus system resilience to future shifts in recharge. This study advances a combined approach using discharge, solute, and solute isotope ( 87 Sr/ 86 Sr, 234 U/ 238 U) mass balances alongside spring environmental tracer data to quantify DGE and characterize DGI in a snow‐fed karst watershed. The integration of DGI solute isotope composition and mineral saturation indices, along with a comparison to spring 3 H‐informed apparent ages, allowed for a novel assessment of flow path dynamics. Results reveal the dominance of karst‐conduit‐influenced contributions to DGI during high and low flow. Karst‐conduit‐influenced springs and DGI showed large increases in residence times from high to low flow periods. This is explained by a shift in the presence from flow in quick response conduit and shallow soil flow paths during spring runoff to carbonate matrix flow paths during baseflow. Because of the significant spring and DGI contribution to river baseflow, insight into these groundwater sources has important implications for system resilience and the potential storage volume available that can buffer impacts of climate variability.

Abstract. Deep Learning (DL) has recently emerged as a promising approach for statistical climate downscaling. In this study, we investigate the use of pre-training in this context, building on the DeepESD model developed for the Spanish National Adaptation Plan (PNACC), which uses ERA5 predictors and the 5 km ROCIO-IBEB national gridded predictand dataset. We evaluate the effectiveness of different fine-tuning strategies to adapt this pre-trained model to alternative national and regional station (point-based) datasets. The objective is to develop downstream downscaling methods that maintain consistency with the original national-scale model while capturing the specific characteristics of regional and local datasets. We analyze the benefits of fine-tuning, focusing on the improved consistency and robustness of the resulting models. Using eXplainable Artificial Intelligence (XAI) techniques, we examine the relationships learned by the models and compare the resulting climate change signals. Our results demonstrate that pre-training provides a robust foundation for statistical downscaling, particularly in cases with limited spatial and/or temporal data availability (e.g. local high-resolution datasets available only for short periods), thereby reducing epistemic uncertainty and improving the reliability of future climate projections. Overall, this approach represents a step toward standardizing DL-based downscaling models to ensure more coherent and consistent climate projections across national and regional scales.

Study region Upper Great Ruaha River Basin, Tanzania. Study focus Climate change impacts on water availability were assessed using the Soil and Water Assessment Tool Plus (SWAT+) model forced by climate projections from ten bias-adjusted and downscaled global climate models under SSP1–2.6, SSP3–7.0, and SSP5–8.5 scenarios. Simulations were conducted for a historical period (1961–1990) and two future periods (2036–2065 and 2070–2099). Climate impacts were quantified by comparing future and historical simulations, while hydrological extremes were evaluated using flow duration curves. New hydrological insights (1) Annual temperature was projected to increase under all scenarios by 1.8–1.9°C (SSP1–2.6), 2.3–4.0°C (SSP3–7.0), and 2.6–4.9°C (SSP5–8.5). (2) Wet-season rainfall was projected to increase by up to 24% under SSP5–8.5 in the far-future, while reductions of up to 12.5% are projected under SSP1–2.6 in the mid-future during the onset of the rainy season. (3) These changes increased evapotranspiration, surface runoff, and water-yield, while reducing percolation, baseflow, and river discharge by up to 7.8% under SSP1–2.6 in the mid-future. (4) High-flows were projected to increase by more than 75% in the mid-future across all scenarios, while low-flows declined by more than 10% both future periods. The changes pose substantial risks to agriculture, water supply, energy generation, and local livelihoods. The findings provide local evidence to support climate-resilient water resources management and adaptation planning in the basin.

Abstract The Orbiting Carbon Observatory‐3 (OCO‐3) provides precise observations of the column‐averaged dry air mole fraction of carbon dioxide () with the ability to target specific areas using the Snapshot Area Maps (SAMs). This study utilizes OCO‐3 observations in a Bayesian inversion framework, together with the column version of the Stochastic Time‐Inverted Lagrangian Transport (X‐STILT) model driven by the Weather Research and Forecasting (WRF) model to estimate emissions from the Bełchatów power plant in Poland. Many previous studies quantifying power plant emissions have relied on Gaussian plume modeling (GPM), cross‐sectional flux (CSF), and Integrated Mass Enhancement (IME) approaches. These methods have been effective for estimating point source emissions, but they do not explicitly simulate the emission plume. We used the WRF model at a 1 km resolution to simulate the emission plume from the Bełchatów power plant. We used nine SAMs over Bełchatów between April 2020 and October 2022, together with nitrogen dioxide () observations from the TROPOspheric Monitoring Instrument (TROPOMI) to isolate the plume in the OCO‐3 data. We found that the satellite‐derived emissions are in good agreement with independent emission estimates derived from hourly reported power generation data as well as estimates from previous studies. Our results demonstrate the potential of Bayesian inverse modeling with WRF‐X‐STILT for quantifying emissions from power plants.

Geospatial monitoring is crucial for landslide research and hazard mitigation. This paper provides a comprehensive overview of contemporary landslide monitoring methods and lays the groundwork for a unified monitoring framework. An in-depth bibliometric analysis and critical review of state-of-the-art approaches developed over the past decade are presented. The study proposes a new classification and systematization of geospatial monitoring methods based on dimensionality (1D, 2D, and 3D) and referencing approach (absolute or relative). The reviewed methods include geodetic techniques, photogrammetry, laser scanning, global satellite navigation systems, UAVs, radar interferometry, and various sensors. The operational characteristics, advantages, and limitations of the existing methods are analyzed with respect to monitoring accuracy, spatial coverage, temporal resolution, and applicability to different deformation conditions. A comparative analysis and systematization of monitoring methods according to landslide velocity classes are presented. This framework links achievable observation accuracy and monitoring frequency to landslide dynamics. Based on the analysis, a refined workflow for geospatial landslide monitoring is proposed. The workflow integrates monitoring design, observation network configuration, data integration, statistical analysis, and forecasting stages. The analysis indicates that effective landslide monitoring requires integrated multi-sensor systems. Future developments are expected to focus on geospatial and non-geospatial data integration, monitoring automation, and next-generation monitoring system design.

Vegetation Optical Depth (VOD) from L-band microwave radiometry is widely used to monitor vegetation water status and ecosystem dynamics, but existing SMAP and SMOS products at 25–40 km resolution cannot resolve fine-scale canopy heterogeneity. We develop and evaluate a framework for estimating forest Global Navigation Satellite System Transmissometry (GNSS-T)-calibrated VOD at 30 m resolution by fusing Sentinel-1A Synthetic Aperture Radar (SAR), Harmonized Landsat–Sentinel-2 optical reflectance, GEDI lidar canopy structure, and climate variables (precipitation and air temperature aggregated over same-day, 7-, 30-, and 60-day windows), all calibrated using ground-based (GNSS-T). A self-supervised learning (SSL) framework based on Vision Transformers, pretrained on unlabeled satellite imagery and fine-tuned with GNSS-derived VOD from five forest sites in the United States and Switzerland, generally outperforms Random Forest and XGBoost across SAR-only, optical-only, and fusion configurations. Feature importance analysis reveals that medium-term climate drivers, particularly 60-day precipitation and 7-day temperature, consistently outrank SAR backscatter, while SAR captures rain-driven VOD peaks that optical sensors miss due to cloud screening. The framework demonstrates robust temporal generalization across deciduous and evergreen forest types, though spatial generalization remains the critical next step as the GNSS-T network expands. These results establish GNSS-T-calibrated multi-sensor fusion with domain-specific SSL pretraining as a viable pathway toward operational 30 m L-band VOD estimation.

Abstract Mineral dust drives Arctic climate variability, but accurate source attribution remains difficult due to plume mixing from the different regions. Using observations and HYSPLIT modeling, we characterize a record‐breaking trans‐continental dust intrusion (March 13–19, 2013) traversing from North Africa to the Arctic. Distinct from single‐source transport, a novel “cumulative‐snowballing” phenomenon‐Saharan dust uplifted into the free troposphere mixed with Central and East Asian emissions‐was identified. This blended plume was dominated by East (45.22%) and Central (23.62%) Asian sources, while retaining a non‐negligible Saharan contributions (17.34%). Driven by extreme multi‐regional emissions, the mid‐latitude westerly jet exported this dust across the North Pacific in the middle to lower troposphere, further a blocking‐like anticyclone opened a meridional pathway for continuous poleward intrusion. These findings demonstrate that Arctic aerosol loading can result from consecutive multi‐continental injections, highlighting the importance of considering complex mixing states in climate assessments.

Accurate forecasting of irregular ocean waves is important for offshore engineering safety and wave energy applications. Irregular wave sequences exhibit strong non-stationarity, multiscale dynamics, and sparse extreme events. To address this problem, this study proposes MM-DAF, a multi-model dynamic adaptive fusion forecasting model. It integrates two structurally complementary submodels in parallel: PatchTST (Patch Time Series Transformer) for long-sequence dependency modeling and N-HITS (Neural Hierarchical Interpolation for Time Series) for hierarchical interpolation-based multiscale decomposition and reconstruction. In addition, an LSTM-based gating network is constructed to encode the historical wave context, enabling adaptive dynamic weight fusion and bias compensation. Visualization of the learned coefficients further indicates that the gating network adapts across non-stationary windows. Bayesian optimization is employed to tune the hyperparameters of the submodels. Experiments are conducted on AQWA-generated numerical datasets and wave-tank experimental data over multiple forecasting horizons. The results show that MM-DAF outperforms representative baselines and static-fusion baselines in both conventional accuracy metrics and extreme-wave error metrics. On Dataset A, MM-DAF reduced RMSE by 44.8% and 28.3% relative to the best single model at the 32-step and 96-step forecasting horizons, respectively; on Dataset B, the corresponding RMSE reductions were 27.2% and 38.0%. Overall, MM-DAF provides an interpretable and practical framework for irregular wave forecasting, and offers a basis for future extensions to multivariable inputs and improved reliability under extreme events.

Abstract There have been significant wet biases in the simulation of precipitation over the Tibetan Plateau (TP) for a long time. One of the important reasons is that current numerical models cannot accurately describe the influence of small‐scale orography. In the orographic gravity wave drag (OGWD) parameterization used to describe small‐scale orographic effects, this study develops a high‐precision calculation method of sub‐grid mountain sharpness that varies with geographical areas to revise and optimize the current constant mountain sharpness parameter (results in a constant drag coefficient at the surface). Batch numerical experiments on precipitation over the TP are carried out using the Yin‐He Global Spectral Model. The results show that the newly calculated drag coefficient of revised OGWD scheme improves the simulation of precipitation and atmospheric circulation over the TP. The new mountain sharpness overall increases the OGWD over the TP on the westerly circulation. According to the vertical vorticity equation, changes in OGWD enhance the positive vorticity over the southern TP. Compared to the original scheme, the revised OGWD experiment exhibits a cyclonic circulation difference over the southern TP, which weakens the flow of water vapor into the TP. The water vapor budget of the main precipitation area over the western TP decreases by about 12.66%, thus reducing precipitation and wet biases over the western TP.

What controls the occurrence of giant earthquakes remains a central question in seismology. We reveal a notably simple-but previously overlooked-relationship between earthquake size and fault dip in subduction zones. Global earthquake statistics show a robust dependence of earthquake size on fault dip, with the probability of extreme events peaking on ultralow-angle faults. Through three case studies, we demonstrate that earthquakes tend to grow larger when the fault geometry is aligned with the regional stress field. We further discuss how such favorable stress configurations are generated along subduction interfaces and emphasize the critical importance of monitoring the temporal evolution of stress in subduction zones to better forecast future megaearthquakes.

Abstract Changes in cloud fraction associated with Central Pacific and Eastern Pacific El Niños are analyzed using monthly cloud data from the International Satellite Cloud Climatology Project, meteorological variables from the European Center for Medium Range Weather Forecasting reanalysis version 5, and radiation data from the Clouds and the Earth's Radiant Energy System. We observe robust increases in low cloud fraction in the tropical Southeast Pacific (SEP, 120°–80°W 0°–20°S) during El Niños. The increases in SEP low‐cloudiness are associated with CP‐El Niño variability. Increases in SEP Sea Surface Temperature during CP‐El Niños are accompanied by a warmer tropical free troposphere, increased cold advection, and reduced overlying cirrus, facilitating the growth of low clouds. Northeast Pacific cloud decks decrease during El Niños, consistent with weaker free‐tropospheric heating outside the tropics. Coupled CMIP6 model experiments are unable to replicate increased SEP low cloud due to a misrepresentation of El Niño Southern Oscillation diversity.

Hydroclimatic whiplash, characterized by rapid transitions between wet and dry conditions, poses a growing challenge for water resource management and ecosystem stability. However, the spatial and statistical characteristics of wet–dry transitions across the contiguous United States are not fully understood. This study analyzed Palmer Drought Severity Index (PDSI) and Palmer Hydrological Drought Index (PHDI) records from 344 NOAA climate divisions spanning 1895–2024. Whiplash was defined as a transition between severe drought and severe wetness within 12-month rolling windows. Statistical descriptors including lag-1 autocorrelation, variance, and skewness were evaluated for whiplash and non-whiplash windows to assess differences between them. Whiplash hotspots were identified in Texas, Southern California, northern Colorado, western New Mexico, and parts of the Midwest, with PDSI detecting higher whiplash frequency than PHDI. This difference reflects PDSI’s greater short-term sensitivity to near-surface soil moisture compared to PHDI’s longer-term hydrological persistence. Whiplash windows showed elevated variance and autocorrelation relative to baseline conditions across both indices. These findings demonstrate that hydroclimatic whiplash is strongly dependent on index selection and threshold definition, and that elevated variance and autocorrelation may serve as statistical indicators of transition periods with important implications for drought monitoring and climate adaptation planning.

Groundwater levels are usually mapped as water table contours produced from point data, that is, hydraulic heads measured in modest numbers of observation wells distributed across a given region. The typically sparse distributions of wells, especially in remote areas, severely limit the number of observations that can be made and may lead to ambiguous groundwater-level estimates at unsampled locations. Satellite altimetry provides reliable estimates of hydraulic heads wherever surface water and groundwater intersect, regardless of how remote the location is. Therefore, we tested the use of Ice, Cloud, and Land Elevation Satellite 2 (ICESat-2) and Surface Water and Ocean Topography (SWOT) measurements of water levels in interdune water table lakes to characterize groundwater levels in the Nebraska Sandhills (central USA), the largest dune field in the Western Hemisphere. Our satellite altimetry estimates of groundwater levels in the Nebraska Sandhills closely approximate the measurements made in nearby observation wells. ICESat-2 showed a root-mean squared error (RMSE) of 0.68 m with ± 0.45 m standard deviation (SD). SWOT estimated an RMSE of 0.75 m with ± 0.76 m SD. Monthly groundwater levels were estimated using kriging with an external drift and generalized additive models, with RMSEs ranging from 1.9 m to 3.3 m and with unbiased errors (mean error of −0.003 m to 0.153 m). We conclude that satellite altimetry has potential for the remote measurements of groundwater levels under certain geographic conditions, especially where groundwater-dominated lakes are prevalent.

Abstract The southeastern Vietnamese coast is a typical upwelling region, where understanding long‐term spatiotemporal variations of the underwater light field—commonly quantified by the diffuse attenuation coefficient () and Secchi disk depth ()—and its regulation by physical processes (such as the upwelling, the northeastward jet, and the Mekong River plume) is essential for evaluating ecosystem health and comprehending marine carbon cycling. Based on in situ measurements, along the southeastern Vietnamese coast was simulated using Hydrolight 6.0 (HL60) and analyzed in combination with chlorophyll a (Chla) and hydrophysical parameters. The consistent or opposite distribution patterns among , Chla, salinity, and temperature were found during sampling. On this basis, the long‐term spatiotemporal variability of was studied using an empirical model applied to monthly remote sensing products. Through a comprehensive analysis, the combined effects of the upwelling, the northeastward jet, and the Mekong River plume on were clarified, revealing (a) significant correlations between and sea surface temperature (SST) ( R = 0.500), Chla ( R = −0.894), and wind speed ( R = −0.662), (b) stronger negative correlations between 3‐month‐lagged and both surface runoff ( R = −0.712) and precipitation ( R = −0.709), (c) distinct seasonal patterns of that vary spatially, and (d) an increase in during the weakened upwelling and Mekong River plume in 2010 and 2020. This study provides support for the regional ecological protection and sustainable development of fishery resources and further establishes a crucial theoretical foundation for understanding the coupling of ocean dynamics with biogeochemical processes, as well as regional responses to climate change in coastal upwelling regions worldwide.

Abstract To understand the physical processes leading to extreme and sustained near‐surface temperatures, three classes of summertime days in Victoria, Australia are compared: they are upper‐quartile warm days (above 75th percentile in daily maximum temperature), single hot days (those above the 90th percentile) and three‐day heatwaves (those above the 90th percentile). In each class, the physical processes are similar: in the composite mean there is cold advection in the boundary layer and strong diabatic heating during the day. However, short‐wave radiation, and hence the surface sensible heating, decreases as hot days or heatwaves progress. At night on hot days and heatwaves, a nocturnal jet advects warm air into the region above the boundary layer, and this warmer air is mixed to the surface after sunrise. During the day, the temperature advection anomaly is warm (relative to warm days), with stronger anomalous warm advection on heatwaves than hot days. The stronger and slower‐moving anticyclones during heatwaves account for their longevity. The arrival of a coastal front terminates the sequence of high surface temperatures. Ten‐day backward trajectories computed from ERA5 show that the air mostly originates from the South Indian Ocean 10 days before warm, hot and heatwave days. During heatwaves, the air descends from greater heights, resulting in stronger adiabatic compression and hence greater warming.

Abstract Accurate prediction of nocturnal convection remains a forecasting challenge, particularly in stable nighttime environments where atmospheric bores can influence convective development. Previous studies suggest that low-level jets (LLJs) and nocturnal stable boundary layers provide favorable environmental conditions for atmospheric bore generation, yet their relative roles remain unclear. This study performs a series of idealized numerical simulations to isolate and quantify the impacts of LLJ strength and thickness, as well as the strength and depth of the low-level stable layer, on the formation and maintenance of cool-pool-driven bores. The results show that bore intensity increases with LLJ strength, whereas a thicker LLJ, associated with weaker vertical wind shear, weakens wave trapping and shortens bore longevity. In the absence of an LLJ, a cold pool can still produce a bore in environments characterized by a moderately deep stable layer, corresponding to a partially blocked regime, with stronger low-level stability supporting longer-lived bore propagation. In contrast, extremely shallow or excessively deep isolated stable layer cannot support bore propagation without LLJ support. Overall, both the LLJ and low-level stable layer play essential and complementary roles in promoting bore generation and sustaining bore evolution under various environmental conditions.

Study region Jiziwan region of the Yellow River basin, China. Study focus In the Yellow River basin, ecological protection and the development of high-quality infrastructure are of great strategic significance for China. To date, the ecological protection of the Yellow River basin has faced severe challenges, among which frequent droughts have become key limiting factors. However, the spatiotemporal structure and occurrence pattern of drought in the Jiziwan region of the Yellow River basin are difficult to accurately represent, and the mechanisms by which the driving factors affect drought are unknown. Therefore, in this study, the temperature–vegetation drought index (TVDI), 3D drought identification algorithm, structural equation modelling (SEM) system and multivariate cross-wavelet transform were used to reveal the spatiotemporal characteristics of drought events and their response mechanisms to driving factors New hydrological insights for the region From 2001–2020, the annual TVDI increased significantly (0.011/year), with a spatial distribution showing a pattern of higher values in the west and lower values in the east. Seasonal differences were evident, with spring being the most severe drought season and winter showing the fastest increase in TVDI (0.012/year). A total of 15 drought events were identified, with the 2015–2016 event being the most severe (24-month duration, 52.58% of the area affected, and an intensity of 4.635 ×10 5 km²·month). Soil moisture mitigates drought through direct and indirect pathways, while surface temperature and solar radiation exacerbate drought via energy balance processes. Precipitation is the key negative moderator of drought, with the largest total impact (-0.69), highlighting its critical role in alleviating drought. Atmospheric circulation factors (AO, DMI, PNA) interact with drought (TVDI) through multiple timescale resonance cycles (4–56 months), collectively influencing drought development by altering regional water-heat balance.

Obstructive sleep apnea (OSA) is characterized by recurrent respiratory events that trigger autonomic arousals and blood pressure (BP) surges, contributing to elevated cardiovascular risk. Photoplethysmography (PPG)-derived timing markers such as pulse arrival time (PAT) are frequently used as noninvasive surrogates of BP dynamics, yet their interpretation is confounded by the pre-ejection period and peripheral vascular effects. Here, we used detrended fluctuation analysis (DFA) to quantify short- and long-term scaling exponents of continuous blood pressure (Portapres), PPG-, and PAT-derived signals across sleep stages in healthy individuals and patients with OSA. Directly measured systolic and diastolic BP exhibited a robust short- to long-term crossover across all sleep stages, with elevated short-range exponents (α1>1) and lower long-range exponents (α2<1), reflecting well-organized autonomic and vascular control. In OSA, this crossover persisted but was visibly attenuated, consistent with reduced short-term adaptability of cardiovascular regulation. In contrast, PAT-based indices showed substantially weaker short-range correlations and minimal crossover structure. Systolic PAT displayed almost no separation between α1 and α2, and PPG-derived measures exhibited scaling patterns that differed fundamentally from BP. Across modalities, PAT (whether derived from BP or PPG) failed to reproduce the multiscale organization characteristic of beat-to-beat BP dynamics. Group comparisons further identified systolic BP scaling, particularly the short-range exponent α1, as the most sensitive marker of cardiovascular dysregulation in OSA, whereas PAT and PPG provided complementary but physiologically distinct information related to peripheral vascular and autonomic modulation. These findings demonstrate that PAT and PPG timing measures should not be used as surrogates for BP in fractal or scaling analyses and underscore the unique diagnostic value of BP-derived scaling behavior for assessing cardiovascular regulation during sleep.

Study Region: The Nile River Basin (NRB), spanning 11 countries in East Africa and supporting over 300 million people. Study Focus: This study synthesizes three decades of research using an AI-enhanced PRISMA review to assess impacts of climate change, land-use change, and infrastructure on water availability and risks across the basin. New Hydrological Insights for the Region: The review reveals pronounced spatial heterogeneity in hydrological responses. In the Blue Nile, discharge trends are mixed and often statistically insignificant, with recent changes increasingly linked to land-use dynamics rather than climate alone. In the White Nile, hydrology shows multi-decadal variability rather than consistent long-term trends. A major regime shift in 1961, associated with a positive Indian Ocean Dipole (IOD), raised Lake Victoria by about 2 m and increased baseline outflows. Flows declined through the 1990s, dropped sharply in the early 2000s due to dam operations, and reached record highs in 2020 following IOD-driven extreme rainfall. Attribution studies indicate climate change has likely contributed to increased White Nile flows in recent decades, although uncertainties remain regarding the roles of climate, land use, and human interventions. Future rainfall and river flow projections show wide divergence, ranging from − 10 % to + 25 % for rainfall and − 20 % to + 40 % for discharge by late century, highlighting uncertainty in future water availability and increasing pressure on transboundary water management.

Abstract The rough seabed morphology is widely distributed in the marine environment and significantly impacts the dynamic processes of submarine landslides. However, the underlying mechanisms remain challenging and unresolved. In this study, we simplified the submarine landslide using the well‐known immersed granular collapse model and conducted 18 simulation groups, considering both basal roughness and initial packing volume fraction. Our results reveal that the mobility of the particles initially decreases and then increases with increasing roughness, demonstrating a nonlinear and non‐monotonic variation. When the diameter ratio of basal particles to flowing particles reaches a critical value (between 1.25 and 1.5), particle mobility is at its lowest. Analysis of the runout distance and particle kinetic energy shows that basal roughness influences the dynamic processes by modulating the particles' x ‐direction kinetic energy during the horizontal spread stage. The mechanism behind the increase in mobility as roughness surpasses the critical value can be summarized as follows: the filling of flowing small particles between the basal particles reduces the effective roughness, which leads to an increase in particle‐base collisions. These intense collisions generate excess pore water pressure near the bed, strengthening the hydroplaning effect of the system, which, in turn, reduces frictional resistance. As a result, mobility increases once the roughness exceeds the critical value. In summary, our results enhance the understanding of how rough seabed morphology affects the dynamic processes of submarine landslides at the particle scale. This study also provides theoretical support for the disaster prevention and mitigation of large‐scale submarine landslides.

Abstract This study investigates the role of moisture in double rainbands over South China through combining ensemble sensitivity analysis and Lagrangian backward tracing. During 10–11 May 2022, a high‐impact double‐rainband event hit South China with coexisting inland and coastal heavy rainfall. Ensemble sensitivity analysis reveals that the coastal rainfall was most sensitive to upstream moisture conditions over the South China Sea (SCS). By comparison, the inland rainfall was more related to nearby moisture. Using the Hybrid Single‐Particle Lagrangian Integrated Trajectory model, the respective moisture pathways and sources for the two rainfall areas were further quantified and linked to the sensitivity results. Dominant moisture pathways (∼80%) for coastal rainfall originated from the western Pacific and the SCS. Most moisture was absorbed from the SCS along these pathways, and was finally transported by low‐level southerly/southwesterly jets toward the target area, making the SCS the primary moisture source (&gt;30%) and sensitive region for coastal rainfall. For inland rainfall, southerly and northerly trajectories related to the frontal system accounted for a comparable proportion. Northern sources, including local moisture recycling, contributed more for inland rainfall than for coastal rainfall. Analysis of two additional double‐rainband events confirmed moisture paths were more diverse for inland rainband with increased moisture contribution from northern sources. The frontal system and monsoon activity significantly affected the relative importance of northern sources, the Bay of Bengal and the Arabian Sea, respectively. The findings underscore the necessity of considering area‐specific moisture features to improve forecasts of double rainbands.

Abstract This study investigates the co‐evolution of geomagnetic field disturbances with ionospheric parameters and their characteristic latitudinal signatures by conducting a location and time‐dependent analysis of the 07–12 October 2024 events, to fill crucial gaps in understanding the coupled response of the magnetosphere‐ionosphere system. This has been carried out utilizing integrated multi‐instrument approach combining ground‐based magnetometer‐derived geomagnetic fields deviations from the mean (ΔBx, ΔBy, ΔBz), Digisonde‐derived critical frequency and maximum electron density at the F2‐ionospheric layer (foF2, hmF2), and Global Navigation Satellite System Total Electron Content (GNSS TEC) measurements. Significant disturbances in the ΔBx component are observed across all sectors in the low‐latitude region, particularly when the storm intensity increases; while in the mid‐latitude region, all magnetic field components exhibit significant disturbances across American stations. Mid‐ and low‐latitude stations displayed more stable but still disturbed patterns in both storm cases, with phase dependent and sector‐specific ionospheric responses. During the initial phase, prompt penetration electric fields increased TEC in low‐latitude regions, particularly during evening hours (19:00–23:00 UT) when the equatorial ionization anomaly was strongest. The American sector experienced the most pronounced and coincide geomagnetic field and ionospheric perturbations during both storms, showing the largest ΔBx and TEC enhancements relative to quiet‐time maxima. Notably, the main phase of the 10–12 October 2024 storm resulted in the largest depletion of TEC and foF2, measured against quiet‐time minima. Larger TEC enhancements also coincides with substantial hmF2 depression and foF2 reduction than higher hmF2 and foF2 when the intensity of the storm enhances.

ABSTRACT This study provides a comprehensive evaluation for the prediction of wind power ramping events in the Belgian Offshore Zone. These rapid, large‐scale power fluctuations pose significant challenges to grid reliability. The research uses operational Numerical Weather Prediction (NWP) models from the Royal Meteorological Institute of Belgium, as well as its version enhanced with Wind Farm Parameterization (WFP). Power predictions are generated with both typical power curves and machine learning approaches. Standard verification metrics, such as Mean Absolute Error (MAE), often fail to capture the operational significance of ramp events. To address this, we develop a flexible verification framework designed to assess ramp forecast performance. This framework incorporates adjustable time and power buffers, which tolerate minor, operationally acceptable discrepancies in the timing and magnitude of predicted events. Application of this framework to both intraday and day‐ahead forecasts reveals that WFP‐enhanced models consistently improve ramp predictions over the operational baseline. Further analysis reveals that while the WFP model with power curves effectively reduced false alarms, it comes at the cost of more misses. In contrast, ML‐based approaches achieve slightly higher overall skill scores by striking a better balance between reducing these error types. Moreover, we introduce the Ramp Alignment Score (RAS), an event‐based metric that quantifies the temporal alignment between predicted and observed ramps, to supplement the model evaluation by lead time. RAS analysis demonstrates that WFP models achieve better temporal alignment and reveals a distinct diurnal cycle in ramping prediction errors. Finally, we investigate the impact of a specific meteorological driver, finding an association between severe precipitation and large, highly predictable ramp events. Conversely, moderate and light precipitation are linked to a higher incidence of missed events and false alarms. This work provides both an operationally relevant evaluation methodology and insights into ramp predictions under specific meteorological conditions.

Gross primary productivity (GPP) is central to terrestrial carbon cycling and forest carbon sink assessment. Using Google Earth Engine, MODIS GPP, ERA5-Land meteorological data, and forest extent masks, this study examined GPP dynamics and climatic controls in China’s northeast, southern, and southwest forest regions from 2005 to 2025. GPP increased overall in all three regions, with higher values in the south and lower values in the north. Climatic drivers differed regionally: in the northeast, GPP responded positively to temperature, while VPD slightly exceeded temperature in the dominant-control area; in the southern region, temperature was the main driver but VPD remained important; in the southwest, temperature dominated larger areas, whereas moisture-related controls showed stronger spatial heterogeneity. Piecewise analysis identified temperature–VPD turning points of 11.74 °C, 10.43 °C, and 25.64 °C for the northeast, southwest, and southern regions, respectively. Two-dimensional temperature–VPD binning further revealed nonlinear GPP distributions and distinct optimal hydrothermal combinations across regions. These results show that warming effects on forest productivity are region-specific and constrained by atmospheric dryness, providing evidence for assessing China’s forest carbon sink responses to climate change.