Atmospheric and Oceanic Sciences

Accurate three-dimensional representations of tree structure are essential for fire modeling, radiative transfer simulation, synthetic data generation, and digital twins of forests, yet detailed 3D structure is rarely available at required scales. Current approaches approximate crowns with smooth geometric primitives, discarding the clumping, gaps, and irregular branching present in real trees. We present TreeFlow, a conditional flow matching model that generates realistic 3D tree point clouds from species, acquisition platform, and height. The model uses a transformer trained on real laser scanning data from the FOR-species20K benchmark to learn a velocity field transporting samples from a Gaussian distribution to the source data distribution. We evaluate generation quality by comparing conditioning and distributional fidelity metrics to scans of real trees. Generated trees match or approach the intra-class baseline on five of six metrics, with a Chamfer distance of 0.581 m versus 0.559 m for real trees of the same genus and height class. Performance is strongest below 25 m and degrades with increasing height. TreeFlow generates individual-tree point clouds conditioned on scalar inventory attributes using a model trained entirely on real laser scanning data.

Detecting weak radar targets in complex sea conditions is inherently challenging due to non-stationary sea clutter and sea spikes. Furthermore, traditional dictionary learning algorithms for clutter suppression suffer from high computational complexity. To address these issues, this paper proposes an efficient sea clutter suppression method cascading Block Coordinate Descent (BCD)-accelerated dictionary learning with Tunable Q-factor Wavelet Transform (TQWT) denoising. During dictionary learning, a BCD strategy replaces global Singular Value Decomposition (SVD) with analytical optimization. Combined with an adaptive soft-thresholding operator, this enables low-complexity joint optimization of dictionary atoms and sparse coefficients, drastically reducing training time. Subsequently, a batch-adaptive Orthogonal Matching Pursuit (OMP) algorithm featuring Gram matrix precomputation and a dual-stop mechanism achieves efficient reconstruction and preliminary cancellation of clutter components. Finally, TQWT is applied to filter out residual non-stationary clutter and noise by leveraging its narrowband feature representation and shift invariance. Experiments on measured radar data from the IPIX database and datasets published by the Journal of Radars demonstrate that the proposed method significantly outperforms traditional K-SVD-based algorithms. Specifically, it improves the average signal-to-clutter-plus-noise ratio (SCNR) by 17.48 dB and requires a total execution time of only 7.99 s, achieving a highly favorable trade-off between suppression performance and computational efficiency.

Spaceborne imaging spectroscopy has created new opportunities for monitoring soil properties at regional scales. Its use for predicting soil heavy metal concentrations in mountainous environments, however, remains insufficiently tested, especially when EMIT hyperspectral data are used. In this study, EMIT Level-2A surface reflectance data were integrated with DEM-derived terrain variables to estimate soil arsenic (As), copper (Cu), and zinc (Zn) concentrations in Renhuai, Guizhou Province, Southwest China. Only soil samples falling within valid EMIT coverage were used for element-specific modeling, resulting in 139 samples for As, 136 for Cu, and 130 for Zn. To reduce redundancy among predictors, EMIT spectral variables and terrain factors were screened before model construction. Random forest and XGBoost models were then tested using repeated spatial cross-validation. The best-performing model for As combined EMIT predictors with elevation and achieved a validation R2 of 0.460. Model performance was considerably weaker for Cu, with a validation R2 of 0.188. For Zn, the model failed to outperform the mean-based benchmark, producing a negative validation R2 of −0.028. The spatial prediction maps and residual patterns suggested that the EMIT-based prediction showed moderate potential for As, limited predictive value for Cu, and exploratory rather than reliable mapping capability for Zn under the current sample and predictor conditions.

Small object detection is fundamentally constrained by the lack of discriminative fine-grained features. Although introducing higher resolution detection scales can improve performance, it also amplifies background noise. In addition, the independently decoupled design of conventional detection heads is insufficient to address the persistent challenges of missed detections and false positives for small objects. To this end, we propose MSIA-YOLO, a YOLOv11-based detector with multi-scale semantic interaction and alignment, optimized from three complementary perspectives: feature modeling, high resolution semantic compensation, and task coordinated alignment. First, Receptive Field Attention Convolution (RFAConv) is integrated into the backbone to enhance critical local details, such as edge and texture cues, via receptive field aware attention. Second, to alleviate fine detail attenuation caused by repeated downsampling, we construct a CHSP-P2 small object detection framework with an additional P2 branch. A scale sequence fusion mechanism is further introduced to perform high resolution semantic compensation through cross scale hybrid inputs. Finally, we design a DTIA-Head (Dynamic Task Interaction and Alignment Head), which promotes joint optimization of classification and localization through dynamic task interaction and spatial alignment. Extensive experiments on the public datasets VisDrone, TinyPerson, and RSOD show that, compared with the YOLOv11n baseline, MSIA-YOLO improves mAP50 by 7.7%, 10.3%, and 1.0%, respectively, while also outperforming several advanced detectors. These results demonstrate the effectiveness and generalization capability of the proposed method in small object, dense object, and complex scene object detection scenarios.

In aerial image small-object detection, complex imaging perspectives, arbitrary object orientations, and long-tailed category distributions jointly exacerbate sample imbalance, which significantly degrades detection stability and leads to frequent misclassification of minority categories. To address these challenges, this paper proposes a novel training framework termed SCUD. Specifically, in the label noise suppression strategy (LNSS), a contrastive learning mechanism based on semantic consistency is introduced to constrain the aggregation of similar samples in the feature space, thereby reducing the adverse impact of noisy samples on model optimization. In addition, a scale-aware resampling strategy (SARS) is designed to alleviate noise amplification and overfitting caused by excessive repetition of small objects during training. Furthermore, an adaptive instance selection mechanism (AISM) is developed by jointly modeling prediction uncertainty and global statistical priors, enabling the model to dynamically emphasize learning from informative samples. Extensive experiments are conducted on two publicly available unmanned aerial vehicle (UAV) aerial image datasets to validate the effectiveness of the proposed approach. The proposed method achieves an mAP50 of 70.7% on the DOTA-v1.0 dataset and 88.1% on the DIOR dataset. Notably, the detection accuracy of several rare categories is significantly improved, further demonstrating the effectiveness of the proposed method in addressing sample imbalance in aerial image small-object detection.

Transmission towers are fundamental components of electric power networks. However, their structure, scale and background textures vary substantially across remote sensing images acquired from different geographic regions. These discrepancies often reduce the detection accuracy of a model trained in one region when it is applied to another region. This paper proposes an enhanced DINO-based framework for cross-domain transmission tower detection that incorporates three lightweight optimisation modules. First, a Query-level Objectness Gating (QOG) module adaptively reweights decoder queries by estimating per-query objectness scores, thereby suppressing background-dominated queries. Second, MPDIoU regression is used to improve the localisation accuracy of elongated transmission tower targets. Third, a Quality-aware Scoring Module (QSM) calibrates classification confidence using predicted localisation-quality logits, thereby reducing high-confidence false detections caused by poor box alignment. Experiments are conducted on two remote sensing image datasets from different geographic regions. Under the 10% target-domain annotation setting, the proposed method achieves a precision of 0.8947, a recall of 0.8199, an F1-score of 0.8556 and an mAP@0.5 of 0.8684, outperforming the original DINO baseline and mainstream detectors including YOLOv8, YOLOv9 and YOLOv11. The results demonstrate that the proposed framework provides an effective solution for robust cross-domain detection of slender transmission tower targets in remote sensing images.

Abstract The rugged terrain, complex near-surface geology and harsh deep subsurface conditions in Nandan, Guangxi, restrict shale gas reservoir research. This paper adopts magnetotelluric (MT) surveys down to 4 km to analyse shale gas enrichment rules, controlling factors and reservoir properties. The results show that regional strata are stable with dip angles below 10° and weak tectonic deformation. Shale gas occurs in Lower Carboniferous Luzhai Formation (C1lz) mud shale, with thickness between 200–500 m and burial depth of 300–3000 m. Its resistivity varies from 1 to 7420 Ω m, averaging 164 Ω m. This shale contains abundant organic matter: 85.96% of samples have TOC above 3%, with a maximum value near 12%. The organic matter has reached the over-mature thermal evolution stage, showing strong hydrocarbon generation capacity. Sufficient shale thickness and moderate burial depth create favourable accumulation conditions. The upper mudstone of the Luzhai third member and tight limestone of the Baping Formation form intact cap rocks with little denudation, offering good sealing for gas preservation. The study area underwent slow tectonic uplift (15.77 m/Ma), avoiding intense fracturing and helping retain shale gas. Without man-made seismic sources, MT surveys fit the mountainous topography well and accurately map deep structures and shale reservoirs. This study verifies the reliability of MT prospecting in Nandan and supplies key geophysical support for shale gas evaluation across South China.

The accurate detection of unmanned aerial vehicles (UAVs) in various sizes played an important role in the practical applications. Yet the preceding works suffered from the missing inference, the false alarms, and the poor accuracy due to the the adverse scene conditions, as well as the mutable scales. To solve the problems, a hierarchical attention promoted cross-scale learning framework was proposed in this paper. First, the hierarchical attention mechanism was introduced in the backbone to generate the multi-scale features of targets, so they can be discerned and located at different scales. The resulting features were further delivered to the neck, in which two branches of features were built, respectively. The former was obtained by the target-specific feature operator, while the latter was generated by the upsampling operation. The dual branches were further connected in the quasi-residual structure. So the content of targets can be protected well, and the detail information can be reconstructed. Finally, the dynamic focusing loss measurement was presented to regress the bounding box of the target, so the learning effectiveness of presented the architecture can be promoted. To verify the proposed method, multiple rounds of experiments were performed. The results demonstrated that small and weak drones can be detected accurately, especially in adverse lighting and weather conditions. The evaluation metric of mean average precision rate (mAP) can be improved by 18.5% (YOLO6) on the collected dataset.

Measured radar point-target detection under ground clutter is difficult because weak target echoes often compete with clutter peaks, and image-space detections do not directly represent physical range-Doppler coordinates. This paper proposes a waveform- and physics-guided center detector (WPG-Center) for measured radar maps. The method reformulates detection as continuous physical-coordinate center localization, constructs a four-channel clutter-aware range-Doppler (RD) representation, uses an anisotropic backbone with range-clutter response modulation, and supervises the center heatmap with waveform-resolution-aware Gaussian targets. Conditional physical-prior decoding is used as a targeted candidate reranking step for low-bandwidth, near-zero-Doppler cases. Experiments are conducted on the measured LSS-Ku-1.0 dataset using a strict blocked 5-fold linear frequency-modulated (LFM) protocol with physical localization tolerances and decoded-center detection metrics. WPG-Center achieves a probability of detection (Pd) of 0.850±0.101 and a range root-mean-square error (RMSE) of 18.16±12.19 m, giving the best average decoded-center detection probability and range accuracy among the compared learned and constant false-alarm rate (CFAR)-style baselines, including an independent radar feature pyramid network (Radar-FPN) heatmap detector. The fixed-sigma ablation is substantially weaker, supporting the need for resolution-aware supervision. Fold-wise, qualitative, and stepped-frequency waveform (SFW) analyses are reported as auxiliary evidence to define the scope of the measured LFM conclusion.

Non-uniform sound velocity profiles (SVPs) cause sound-ray refraction and propagation-path bending. The straight-line mapping among propagation time, propagation distance, and target position is, therefore, disrupted, leading to systematic errors in constant-sound-speed localization. To improve the consistency between propagation correction and geometric localization, an iterative ray-path correction method based on propagation-time consistency is proposed. The method contains three coupled steps. First, a path-dependent local layered SVP model is constructed for each target-to-base-station path, rather than using a global or fixed sound-speed model. Second, the ray parameter is inverted under the constraint of measured time-of-arrival (TOA), so that the corrected ray path remains consistent with the observed propagation time. Third, the corrected slant range obtained by layered ray tracing is fed back into a known-depth weighted least squares (WLS) localization model, forming a closed-loop position update. The method is evaluated through long-baseline (LBL) simulations with multiple SVPs and propagation geometries and is validated using measured TOA data and an observation-derived SVP. The simulation results show that sub-meter accuracy can be achieved under the tested TOA-noise conditions. In measured-data validation, the planar localization error is reduced from 4.6866 m to 0.1923 m. No divergence is observed in the tested small SVP-perturbation cases.

LiDAR point clouds are widely used in remote sensing perception scenarios, such as autonomous driving. However, LiDAR-based perception models remain vulnerable to adversarial perturbations, which may compromise the reliability of safety-critical 3D perception systems. Among different attack paradigms, transfer-based attacks are particularly practical because they generate adversarial examples on accessible surrogate models and apply the generated examples directly to unknown target models. Nevertheless, existing transferable attacks on point clouds often perturb regions that are discriminative for the surrogate model but insufficiently stable across different architectures, leading to limited transferability and noticeable geometric distortion. To address this problem, we propose SAGE, a Saliency And Geometry Enhanced transferable attack framework for LiDAR point cloud perception in remote sensing. Specifically, SAGE unifies point-coordinate priors with source-model gradient signals to generate a saliency map, which serves as a transferable indicator of vulnerable local structures. SAGE further leverages this map through saliency-guided perturbation allocation and explicit geometric constraints to enhance transferability while preserving point-cloud geometry. To demonstrate the effectiveness of SAGE, we evaluate SAGE on point-cloud classification benchmarks and further validate it on LiDAR-based 3D object detection using KITTI and nuScenes. Experimental results show that SAGE consistently outperforms existing transferable attack methods in attack success rate while preserving favorable geometric quality of adversarial point clouds. These findings demonstrate that SAGE offers an effective and practical framework for assessing the transfer robustness of LiDAR-based remote sensing perception systems.

Abstract Lakes and reservoirs are estimated to be globally important sources of nitrous oxide (N 2 O) to the atmosphere but recent evidence of N 2 O uptake across a broad range of lakes have called the accuracy of emission estimates into question. Here, we use a new national-scale dataset of dissolved N 2 O concentration and a Bayesian hierarchical model to predict summertime N 2 O concentration and emission rates in 465,896 waterbodies in the conterminous U.S. (CONUS). We found that N 2 O undersaturation was pervasive throughout the CONUS during the summer of 2017, with an estimated 72.9% (95% credible interval: 68.9–76.6%) of lakes functioning as N 2 O sinks. The model predicts dissolved N 2 O concentrations reasonably well based partly on interactions between nitrate concentration, waterbody surface area, and water temperature. Despite working with the largest aquatic N 2 O dataset to date, our national-scale estimate of summertime N 2 O emissions from CONUS lakes is poorly constrained, with a 95% credible interval ranging from net uptake to net emission (−282 − 482 metric tons N 2 O summer −1 ). Pervasive N 2 O undersaturation in CONUS waterbodies during the summer highlights the need to revisit N 2 O models which presume surface waters are a N 2 O source.

Abstract Observational data sparsity in Antarctica and the Southern Ocean results in lower forecast skill for atmospheric circulation over the Southern Hemisphere (SH) than the Northern Hemisphere. Additional Antarctic radiosonde observations at existing observation stations can enhance accurate forecasting of severe events over the SH; however, the number of daily radiosonde observations is limited, owing to financial and operational limitations. Using a data assimilation system, this study investigates the impact of hourly wind profiles from the PANSY radar at the Japanese Antarctic Syowa Station. Assimilation of PANSY data leads to differences in the initial atmospheric conditions for weather forecasts, even including extra radiosonde observations from the Year of Polar Prediction in the Southern Hemisphere (YOPP-SH) campaign in 2022. These differences in initial atmospheric condition persist in Antarctica and the Southern Ocean. We show that Antarctic radar data with relatively high temporal resolution has an impact on the reproducibility of atmospheric basic state (wind speed, temperature and geopotential height) over the high latitudes and the prediction of atmospheric circulation and integrated water vapor associated with atmospheric rivers over the mid latitude in SH.

Pre-monsoon (April-June) tropical cyclones (TC) over the North Indian Ocean (NIO) pose severe coastal hazards. Surprisingly, in recent years like in 2025 and 2026 pre monsoon season there was not a single named TC reported over NIO. In this study, we explored the drivers for such a suppressed activity. We considered 46 years (1980-2025) of India Meteorological Department (IMD) best-track data and found that such situation occurred in 12 specific years during the analysis period. We considered those 12 years as non-TC years and to examine the contrast, we identified 10 TC-active years, where at least one TC occurred in both Bay of Bengal (BoB) and Arabian sea (AS) basins. Genesis Potential Parameter (GPP) composites reveal a basin-wide suppression during non-TC years, with a statistically significant difference of 1.0 × 10⁻⁵ concentrated over the BoB and AS. Logarithmic decomposition of the GPP shows that low-level vorticity is the primary driver of enhanced GPP over both basins during TC years, while reduced vertical wind shear provides additional positive support over the BoB. In general, the conditions are less favourable for the TC formation over the AS and are favourable over the BoB. But, during TC years, warmer SST, higher mid-level moisture and strong low-level positive vorticity supported the TC activity over the AS. Enhanced atmospheric moisture transport (IVT) and stronger large-scale upper-tropospheric diabatic heating during TC years reflect a pre-existing convective environment driven by organized large-scale forcing rather than TC-induced latent heat release. We further identify the MJO as the dominant large-scale modulator of pre-monsoon cyclogenesis, with TC activity systematically phase-locked to long duration and high-amplitude MJO phases 2-4 during active years, while suppressed years are characterized by weak or inactive MJO convective envelopes (RMM amplitude < 1.0). Wavenumber-frequency spectral analysis exhibits enhanced power in equatorial Rossby, Kelvin, and MRG waves alongside active MJO during TC years. Analysis of ENSO and IOD states reveals no systematic difference between TC and non-TC years. These findings suggest role of intraseasonal variability in modulating the large-scale kinematic forcing, alongside local thermodynamic conditions behind the cyclogenesis activity over the NIO.

Abstract Oxygen and other light elements comprise up to 5 wt% of the Earth’s outer-core, and may significantly influence its physical properties and the operation of the geodynamo. Here we report in situ X-ray diffraction measurements of Fe, Fe + 4.5 FeO (atomic proportion), and Fe 2 O 3 melts at 177-440 GPa, achieved using laser-driven shock compression at an x-ray free-electron laser. The melts exhibit Fe-O coordination numbers between 4.0(0.4) and 4.5(0.4), indicating predominantly four-fold coordination environments. These coordination states are significantly smaller than those of Fe-bearing lower-mantle phases such as bridgmanite and ferropericlase. Shorter Fe-Fe interatomic distances in compressed iron oxide melts drive the denser packing relative to ambient melts, while the structural differences between Fe + 4.5 FeO and Fe 2 O 3 melts under shock indicate that the oxidation state modulates oxygen solubility in liquid Fe. At 177 GPa ( ~ 380 km below the core-mantle boundary) and 3800 K, Fe 2 O 3 melts exhibit higher Fe-O coordination, suggesting that local variations in oxygen content could contribute to the stratification in the uppermost outer-core inferred from seismological and geomagnetic observations.

Innate immunity is traditionally viewed as a broad defense system with limited specificity. However, increasing evidence suggests that innate immune cells can discriminate between distinct microbial partners. How such specificity arises in early-diverging animals remains unclear. Here, we identify in the sea anemone Nematostella vectensis a selective host innate immune mechanism mediated by nematosomes, motile multicellular bodies that differentially process bacterial cells. Nematosomes preferentially engulf non-native Vibrio isolates while showing reduced uptake of native host-associated strains. We identify the transcription factor cJUN as a key regulator of this process. CRISPR/Cas9-mediated knockout of cJUN reduces nematosome abundance, impairs lysosomal response, alters microbiome assembly, and increases susceptibility to bacterial infection. These results link immune gene function to microbial selectivity and demonstrate that even early-diverging animals exhibit sophisticated innate immunity mechanisms for microbiome regulation. Our findings support the idea that immune specificity can arise through repurposing deeply conserved pathways and may have deep evolutionary origin.

A critical factor in the transition towards renewable energies is offshore wind, and therefore efficient site investigation to assure the safety and stability of the wind turbines' foundations is needed. Conventionally, borehole drilling and in-situ testing are used. Although these techniques can provide accurate geological data, their costs are prohibitive and their scope of survey is restricted. This drives the need for complementary geophysical survey methods. The Single-Channel Seismic (SCS) method is a cost-effective and rapid technique that can be employed for regional seabed survey. However, low signal-to-noise ratio (SNR) and poor stratigraphic continuity result when applying conventional processing workflow on seismic data obtained from complex settings, such as thick sand layers, shallow water environment with the presence of strong multiples, and sea surface swell conditions. To overcome these problems, an optimized SCS processing workflow which introduces three new techniques is proposed: (i) eigenvalue based swell correction with adaptive sliding window smoothing, (ii) shearlet transform based sparse representation of seismic data to remove random noise, and (iii) combined predictive deconvolution (for short period multiples) and SRME with Bayesian separation (for long period multiples). This algorithm has been applied to a field data survey at Fangchenggang, Guangxi (water depths: 0-25 m, total area is 99 km 2 ), and a substantial increase is observed: SNR enhancement from about 8-12 dB (from comparison with F-K spectrum analysis), the stratigraphic continuity improvement by factor of about 2.5 (estimated by reflection event tracing), and elimination of over 70% multiple energy within targeted range. The final result successfully delineates bedrock topography and structural settings required for wind farm installation site investigation. This new approach makes it possible to enhance shallow-marine seismic data quality even with difficult geological conditions and extract information indispensable for offshore wind farm development.

Magnetic anomaly generated by magnetic target is widely used in many areas. In this paper, a real-time magnetic target localization method based on second-order scalar magnetic gradients at the closest point of approach (CPA) is proposed. By exploiting the geometric symmetry of the magnetic anomaly field at the CPA point, closed-form expressions of the target position and magnetic moment are derived directly from the second-order spatial derivatives of scalar magnetic anomaly under the induced-magnetization assumption, thereby avoiding iterative global optimization. Furthermore, a residual-based error index is constructed to evaluate the consistency between measured and reconstructed second-order scalar magnetic gradients, enabling automatic determination of the CPA point during platform motion. The proposed method is validated by the experiment. The results show that the CPA point on the trajectory can be accurately identified using the proposed error index, and the localization accuracy is significantly improved near the CPA point. At the CPA point, the relative errors of the estimated distances and angles between the target and the two sensors are 0.89% and 0.38%, and 1.0% and 0.52%, respectively, while the relative error of the estimated magnetic moment magnitude is 4.85%. Therefore, the proposed method has great value in target localization based on a mobile magnetic anomaly detection system.

Abstract Global surface ozone (O 3 ), intensified by climate change, poses increasing health and ecosystem threats. Despite stringent air policies, China’s persistent O 3 pollution exemplifies a global challenge intertwined with climate actions reshaping emission pathways. Optimal mitigation remains contentious, primarily due to inconsistent conclusions regarding the sensitivity of summer regional O 3 formation. Here we show the path dependency in O 3 mitigation strategies for synergistic clean air and climate action goal achievement over multidecade scales. This path dependence is validated by observed concurrent plateaus (2020-2023) in deweathered O 3 concentrations and sensitivity trends across Chinese megacity clusters. Leveraging this understanding of path dependency, we quantitatively reveal that the optimal future strategy involves prioritizing early volatile organic compound reductions, as their high effectiveness for regional O 3 mitigation gradually diminishes towards 2050. This challenges prevailing nitrogen oxides priority paradigms. Our work reframes O 3 control, providing a paradigm for resilient air quality-climate governance.

Small-scale turbulent mixing in the ocean interior is vital in governing ocean circulation and tracer distributions, and hence global climate. However, the planetary extent of this role and its dependence on the microphysics of mixing remain inadequately understood. Here, we emphasize the variety of spatio-temporal scales on which such interior turbulent mixing can shape the climate system. In addition to its well-established role in facilitating the equilibration of deep branches of ocean circulation on centennial-to-millennial timescales, interior turbulent mixing is a leading determinant of oceanic tracer budgets on timescales as short as sub-annual. We highlight the importance of the co-dependence of vertical (diapycnal) mixing and lateral (isopycnal) stirring in establishing the large-scale impacts of oceanic turbulence. We conclude with a summary of theoretical, observational and computational bottlenecks in the way of a sufficiently accurate representation of mixing in Earth System Models, and discuss emerging opportunities for making progress in these areas.

Sand dunes develop when there is a source of sediment and adequate wind. Dune morphology and occurrence can then be used to infer sediment source distribution and formative climate conditions. This is useful where direct climate observation is challenging on Earth, other planets, and the past. However, there has been no complete and accurate digital map of the occurrence of Earth’s sand dunes with distinguishable morphologies. Here we present that map and demonstrate that in arid environments dune presence is mostly explained by convergent transport and source proximity, whereas in wetter climates wind strength is an additional constraint. By limiting analysis to dunes identified from globally available imagery and topographic data, we produce a consistent dataset useful for inferring myriad aspects of geology and climate, and improving understanding of aeolian landscapes. We provide an example, using barchan dune orientations, to demonstrate a trade-off between inference of sediment and wind characteristics. A global dune map, combined with climate and geologic data, reveals how sediment availability, wind-driven transport, and flux convergence control dune formation, enabling improved prediction of dunes across past and future climates.

This study examines the long-term surface temperature variability across Jammu and Kashmir using ground-based observations and reanalysis data during 1980–2024. The region shows a clear but spatially heterogeneous warming, with the strongest annual mean temperature (T mean ) rise at mid-elevation stations such as Bhaderwah (+ 0.3 °C/dec) and weak or insignificant trends at lower elevations like Jammu (about − 0.1 °C/dec). Minimum temperature (T min ) shows the most rapid acceleration, by + 0.1 to 0.5 °C/dec at several mid-to-high elevation regions, whereas daytime maximum temperature (T max ) trends remain modest (about 0–0.2 °C/dec). These spatial and seasonal contrasts, along with enhanced warming at higher altitudes in specific seasons (e.g., pre-monsoon T min up to + 0.6 °C/dec), indicate the presence of elevation-dependent warming (EDW) in these mountainous regions. The annual T mean increases by 0.18 °C/km/dec and winter T max shows the strongest altitude dependence of 0.43 °C/km/dec, whereas T min exhibits no significant EDW across seasons. Multiple linear regression analysis suggests that wintertime altitude-dependent temperature trends are statistically associated with albedo-related surface processes, whereas the annual and seasonal increases in Tmin are more closely associated with atmospheric moisture and longwave radiative conditions. Overall, the region has warmed by up to nearly 1 °C in last two decades at several high-altitude locations (e.g., Pahalgam and Gulmarg), highlighting the sensitivity of the Himalayan environment to ongoing climate change with severe implications for high-altitude hydrology, cryosphere stability, and regional climate resilience.