This study evaluates the 1995-2020 global ocean-sea ice simulation using the unstructured-mesh model for prediction across scales (MPAS)-ocean/sea ice model within energy exascale earth system model (E3SM) version 2.1 (E3SMv2-MPAS) at 60 km to 10 km resolution. Multi-source observational data are utilized to validate sea surface temperature/salinity, sea ice, three-dimensional thermal-saline structures, mixed layer depth, ocean heat content, and sea surface height. Key results show the following: (1) E3SMv2-MPAS captures seasonal-to-decadal variability in surface fields and sea ice, but shows systematic biases in sea surface temperature of western boundary currents (inadequate eddy parameterization) and Arctic sea surface salinity (misrepresented freshwater fluxes and mixing processes). (2) The model robustly represents three-dimensional climate variability, yet underestimates mixed layer depth in key regions (Antarctic Circumpolar Current and North Atlantic), revealing deficiencies in extreme mixing. (3) Ocean heat content distributions are well-simulated. (4) Sea surface height spatial patterns and interannual variability are accurately reproduced. This work identifies critical refinements for unstructured-mesh models: mesoscale eddy parameterization, polar ocean-sea ice coupling, and multi-scale energy processes, advancing high-resolution climate model development and laying the groundwork for improved ocean forecasting systems.

The Yin–He global spectral model (YHGSM) is a dry-mass-conserving hydrostatic global spectral model that employs spectral transforms to compute horizontal derivatives. This study investigated the performance of the mass conservation ocean model (MaCOM) coupled with the YHGSM (YHGSM_MaCOM), focusing on its application in forecasting Super Typhoon Hinnamnor (2022) and its interaction with the upper ocean over the northwest Pacific. Hinnamnor presented a marked forecasting challenge due to its abnormal track and extreme intensity changes, including rapid intensification along a westward track followed by rapid weakening along a northward sudden-turning track. The results demonstrate that YHGSM_MaCOM significantly improved Typhoon Hinnamnor intensity forecasts, particularly during its sharp-turning period. This improvement is attributed to the coupled model’s ability to capture sea surface cooling during the typhoon event (5.60 °C, whereas satellite observations recorded 5.57 °C), thereby more accurately assessing oceanic feedback on the typhoon, which corresponded to a more accurate forecast of typhoon intensity. Compared with the pure YHGSM, the mean absolute error of the 0- to 5-d intensity forecasts using YHGSM_MaCOM was reduced by 14.9% to 18.3% across different coupling schemes. When the coupling frequency was increased from 3 to 1 h, the typhoon intensity forecast was further improved; however, the typhoon track errors increased. Overall, YHGSM_MaCOM improved Typhoon Hinnamnor track and intensity forecasting, regardless of the coupling scheme, compared to the pure atmospheric model, demonstrating superiority of the ocean model.

Cloud-native climate services increasingly expose analysis-ready Zarr archives through web APIs and Python/Xarray workflows. These layouts accelerate coordinateto-value subsetting, yet they are less efficient for value-tocoordinate workloads, including threshold-based extreme-event discovery and regional top-$k$ retrieval. In such queries, geometryoverlapping chunks are often read and decompressed before value predicates can be evaluated, causing unnecessary data access. This paper presents an external value-aware metadata layer that preserves the original archive while enabling pruning before data access. The proposed sidecar maintains lightweight chunk synopses, coarse upper bounds, and hot metadata caches to transform geometry-only candidates into value-aware candidates. ERA5 experiments show up to $5 \times$ fewer chunk reads and up to $6.3 \times$ lower median latency than a geometry-only Zarr+Xarray pipeline, demonstrating a practical direction for low-latency extreme-event services without migrating data into a dedicated array database.

Despite significant progress, the high computational cost makes existing precipitation downscaling methods difficult to deploy on resource-constrained devices. To address this challenge, we propose the LU-SSD, which jointly enhances the physical fidelity and spatial resolution of total precipitation from ERA5 through a synergistic strategy of "constructing foundational skeletons" and " decorating with detailed structures" in both spatial and frequency domains. Specifically, we integrate 20 key meteorological variables, utilizing the output of WRFDA as ground-truth to perform 8× precipitation downscaling. First, we analyze the key factors contributing to the computational complexity of State Space Duality Networks (SSDs) and accordingly propose a Separable Channel State Space Duality Module (SC-SSD). A lightweight U-shaped backbone network stacked with SC-SSDs effectively models long-range spatial dependencies among meteorological variables at a cheap computational cost, accurately establishing the foundational skeletons of precipitation areas and rain bands. Second, multi-scale convolutional modules are incorporated to capture local spatial details, thereby reconstructing the fine-scale structures. Furthermore, we analyze systematic biases between input and ground truth in the frequency domain, and design a Precipitation Inverse-frequency Attention Module (PIAM), which corrects low-frequency component biases and reconstructs extreme precipitation events. Finally, a Grid-aware Intensity Consistency Module (GICM) is proposed to serve as a regularization step that anchors the downscaling process to large-scale statistical features before the final spatial refinement. Extensive experiments demonstrate that the LU-SSD achieves state-of-the-art performance. Notably, it requires only 678 MB of GPU memory during inference, enabling efficient deployment on edge devices and offering a viable pathway toward high-resolution meteorological services.

Climate data services support many scientific workflows, but climate datasets are large and costly to store and deliver. Lossy compression can reduce data volume, but climate applications also require controlled error. Motivated by this requirement, we propose ChunkZ, a fast chunked climate data compression method using spatio-temporal predictors. First, we propose a chunk-based parallel pipeline to overlap input and output with compression, thereby improving end-to-end efficiency. Second, we introduce spatio-temporal predictors to improve compression ratios on climate variables that exhibit strong spatio-temporal variation patterns, such as wind, temperature, and humidity fields. Third, we propose a chunkwise predictor selection strategy to improve overall compression ratio with low tuning overhead. Last, we conduct comprehensive experiments on widely used reanalysis datasets at two spatial resolutions, and experimental results show that ChunkZ achieves the best compression throughput on all four physical variables with speedups of up to 7.3 times over runner-up baselines and the best compression ratio on all four variables with up to 65.3 percent improvement under the same error bound.

Research on tropical cyclones (TCs) requires the joint analysis of geostationary satellite imagery and meteorological reanalyses. However, these datasets are typically stored in heterogeneous formats across separate archives, making integrated retrieval and computation inefficient. As a result, researchers must manually download, parse, and coregister multiple sources, leading to tedious and error-prone workflows. To address this challenge, we propose TC-Zarr, an analysis-ready storage framework that fuses geostationary imagery and reanalysis fields into a unified multidimensional Zarr data cube indexed by time, latitude, longitude and variable. TC-Zarr employs a Guided, Mass-conserving Resolution Fusion (GMRF) algorithm to align high-resolution satellite imagery with lower-resolution reanalysis data on a common target grid. The aligned datasets are stored in the Zarr cube, enabling efficient retrieval and seamless joint analysis. In a case study of Super Typhoon Doksuri (2023), TC-Zarr integrates Himawari-8 imagery with ERA5 reanalyses and achieves significant improvements in data access efficiency compared with conventional file-based workflows.

Long-horizon robotic manipulation remains a critical challenge in robotics. Hierarchical reinforcement learning offers a promising solution, but often suffers from an imbalance dilemma: simplifying skill learning increases the complexity of planning, thereby expanding the solution space and computational burden of planning. To tackle this challenge, we propose a Hierarchical Reinforcement Learning framework with Dynamic Kolmogorov-Arnold Network (DyKAN) based Actor Critic, named HIKER. Firstly, HIKER innovates with a dual-chain design that decomposes the complex task into two intersecting sub-chains, reducing the optimization conflict across subtasks and alleviating the burden on the planning model. Secondly, we develop DyKAN, a scalable neural network for both actor and critic in the skill model of HIKER. DyKAN adaptively adjusts grids and basis functions while preserving learned knowledge, enabling efficient learning of complex manipulation skills. Furthermore, to optimize DyKAN's performance, we design a per-layer update module that uses Dynamic Tanh (DyT) and low-rank decomposition to ensure stable, low-cost updates during training. Finally, experiments on long-horizon robotic manipulation tasks demonstrate that HIKER significantly improves efficiency and robustness, yielding higher-quality skill models and achieving a 10.9% increase in task success rate under the high noise condition. Further insights are available on the website: https://sites.google.com/view/hikerdykan.

Climate archives are expanding at the petabyte scale, while existing climate data formats and floating-point compressors deliver only limited compression efficiency. Recent neural image compression approaches achieve higher ratios but lack strict error guarantees, limiting their applicability in scientific domains. To address this gap, we propose EBCF (Error-Bounded Climate Flow), a discrete normalizing-flow–based compressor that enforces absolute reconstruction error bounds for numerical climate data. EBCF introduces three key innovations: (i) an error-bounded normalization algorithm that ensures quantization errors remain within user-specified tolerances; (ii) a precision-preserving discrete flow model that learns probability distributions from climate latents without multiplication/division error; and (iii) a cache-based entropy coding strategy for efficient end-to-end compression. Experiments on ERA5 reanalysis data demonstrate that EBCF reaches highest average compression ratio (4.27×) compared to both format-level compressors (1.38×) and state-of-the-art error-bounded methods (3.48×), while maintaining fidelity required for downstream climate applications.

Similar weather retrieval asks a meteorological archive to return historical multi-day weather processes that are physically analogous to a query process. Here, similar weather is defined as the comprehensive similarity of time-elastically aligned multi-day processes in event-level evolution, multivariate physical-field trends, and numerical intensity; it is not singlevariable, instantaneous, or purely pixel-level statistical resemblance. We study a generative retrieval formulation in which a query process is mapped directly to discrete identifiers of relevant historical processes. The formulation reuses published physicsaligned weather fingerprints as compact daily tokens and organizes archive segments with hierarchical identifiers whose prefixes encode coarse-to-fine physical similarity. Identifier generation enables direct table lookup, while physics-aligned prefix learning and time-elastic supervision preserve meteorological meaning. Experiments on ERA5 weather-process segments demonstrate that, under both strict and time-elastic protocols, the proposed method achieves low latency while maintaining high similarity.

From 19 to 21 July 2021, extreme rainfall occurred in Henan, China, with a single hour of 201.9 mm breaking all records in China. The orographic drag of the mountains around Henan is an important influencing factor. In addition, during this extreme rainfall event with abundant moisture, moisture significantly affects vertical air movement and atmospheric stability. As a result, the orographic gravity wave drag (OGWD) parameterization based on the “dry air” assumption becomes less effective. It is necessary to analyze the impact of moisture effects on the vertical propagation of gravity waves and how these affect this extreme rainfall event. This study simulates the “21.7” extreme rainfall event in Henan, respectively, using the dry and moist OGWD parameterization. The results show that the moisture effect reduces the buoyancy frequency, which will lead to an increase in the amplitude of the gravity wave and a decrease in the Richardson number. These are conducive to the breaking of gravity waves, which increases orographic drag in the troposphere in the OGWD parameterization. The change in OGWD results in a cyclonic circulation difference in northern Henan, which weakens the airflow moving toward the northwest mountainous areas and enhances the airflow moving northward. These changes in circulation significantly alter the water vapor budget in the heavy rainfall region, increasing the net water vapor budget by approximately 33%, thereby reducing the negative biases of average rainfall by about 14.59%. These indicate the moisture effect can significantly alter the vertical propagation of gravity waves and improve the intensity of heavy rainfall.

The planning and operation of renewable energy, especially wind power, depend crucially on accurate, timely, and high-resolution weather information. Coarse-grid global numerical weather forecasts are typically downscaled to meet these requirements, introducing challenges of scale inconsistency, process representation error, computation cost, and entanglement of distinct uncertainty sources from chaoticity, model bias, and large-scale forcing. We address these challenges by learning the climatological prior distribution of a target region with a generative model, using its high-resolution numerical weather simulations. An optimal combination of this learned high-resolution climatological prior with coarse-grid large scale forecasts yields highly accurate, fine-grained, full-variable, large ensemble of weather pattern forecasts. Using observed meteorological records and wind turbine power outputs as references, the proposed methodology verifies advantageously compared to existing numerical/statistical forecasting-downscaling pipelines, regarding either deterministic/probabilistic skills or economic gains. Moreover, a 100-member, 10-day forecast with spatial resolution of 1 km and output frequency of 15 min takes 1 hr on a moderate-end GPU, as contrast to CPU hours for conventional numerical simulation. By drastically reducing computational costs while maintaining accuracy, this paradigm paves the way for more efficient and reliable renewable energy planning and operation.

Ku-band scatterometers lose extensive Sea Surface Vector Wind (SSVW) observations under extreme winds, heavy precipitation, or instrument anomalies, degrading forecast and assimilation skill. Traditional interpolation fails to reconstruct non-linear wind structures, whereas existing deep learning inpainting is hampered by scarce public datasets, high computational cost and insufficient continuity modeling. We propose WMamba, an Attention-Structured State Space Duality (ASSD)-based framework that exploits wind continuity to encode global dependencies with O(N) complexity for accurate SSVW inpainting. A Grouped Multiscale Attention Block (GMAB) ensures accurate fine-scale wind detail reconstruction by mitigating local pixel degradation. We also introduce L-WMamba, a lightweight 0.36 M-parameter variant suitable for resource-limited devices. Moreover, we release the SSVW Inpainting Dataset (WID), comprising 123,841 high-wind HY-2B HSCAT samples (2018–2022), as an open benchmark. Experiments demonstrate that WMamba outperforms GRL (state-of-the-art) decreasing the RMSE for wind speed and direction by 11.4% and 6.3%, respectively, while achieving a 94.7% reduction in parameters. In particular, WMamba effectively inpaints wind details, as evidenced by the highest MS-SSIM and RAPSD scores. This framework and dataset establish a robust baseline for extreme-weather SSVW recovery.

Cloud-native metocean data cubes are increasingly served from object storage, but heterogeneous OGC API-EDR and OGC API-Tiles requests can trigger many small remote reads and unstable tail latency. We present a Zarr-native metocean service that implements the five core OGC API-EDR query endpoints together with OGC API-Tiles over two fully materialized sharded layouts stored as independent Zarr stores in MinIO. A QoS-aware semantic router maps each request to a storage-aware plan using the request template, estimated chunk/shard footprint, and runtime statistics. We further add a lightweight learning-augmented planner, implemented as an MLP-based AI-assisted plan selector, that chooses among six bounded layout/parallelism actions and is trained on 8,000 recorded prototype requests collected under controlled workload generation on the deployed service. We optimize a QoS-oriented objective centered on tail latency, but report mean, P50, and P95 latency for completeness. On a held-out replay set derived from these recorded prototype requests, the learned planner reduces mean latency from 774.6 ms to $722.7 \text{ms}(6.7 \%)$ relative to the rule-based heuristic, closes 86.8% of the gap to a bounded oracle, and adds negligible model-selection cost relative to endto-end request latency. These results show that standards-aligned semantic routing plus learning-augmented plan selection can improve QoS for metocean data access on sharded object storage.

Similar weather retrieval is a critical task in geoscientific information processing. Existing approaches often fail to meet the physical interpretability, spatiotemporal coupling, and scalability required in practice. We present Sim-Weather, a label-free retrieval framework for meteorological fields. The method learns physically aligned fingerprints whose Euclidean distances are statistically matched to field-space RMSE via a distance-preserving objective. We design physics-consistent dual-view augmentations separately for scalar and vector variables, employ Swin Transformer 3D with Hierarchical Temporal Attention (HTA) to model spatiotemporal dependencies, and stabilize representations through keyframe reconstruction. The resulting 256-d per-variable fingerprints can be concatenated and searched using scale-adaptive ANN indexing (HNSW or IVFPQ), enabling fast, training-free queries. Experiments demonstrate that Sim-Weather consistently outperforms ClimaX, VideoMAE, and Video Swin Transformer in numerical error, correlation, and perceptual quality. Ablation studies further confirm the complementary roles of each component.

In recent years, Artificial Intelligence (AI)-based weather prediction models have emerged as powerful tools in meteorology, capable of learning complex dependencies from extensive weather datasets and generating rapid forecasts after training. These models achieve prediction accuracies comparable to state-of-the-art Numerical Weather Prediction (NWP) systems. However, these models remain not fully operational due to their dependence on computationally intensive Data Assimilation (DA) systems for generating accurate initial fields. Recent advances in AI techniques offer a potential pathway to develop more efficient and accurate DA systems, advancing the operational feasibility of end-to-end AI-based weather forecasting. Despite growing interest, research in AI-based DA remains fragmented. Therefore, a comprehensive review is necessary to clarify the current progress, identify challenges, and guide the future development of next-generation AI-based DA systems. This review categorizes AI-based DA research into two primary domains. The first domain is AI-empowered DA, where AI enhances individual components such as observation operators, tangent linear and adjoint models, and uncertainty quantification. It also includes latent DA, which helps reduce computational costs. The second domain is AI-based end-to-end DA models, which integrate observations and short-range weather predictions within unified AI frameworks to generate accurate initial fields. We further discuss key challenges and opportunities, including dataset standardization, model evaluation protocols, assimilation of extended observation types, enforcement of physical constraints, and addressing operational scalability. Finally, we emphasize the importance of interdisciplinary collaboration across AI and meteorology in developing practical and reliable AI solutions to enhance DA processes and support more accurate weather forecasting. This review offers practical insights to the research community to expedite the development and operationalization of AI-based DA and end-to-end weather forecasting systems.

Numerical weather prediction is the cornerstone of modern weather forecasting, yet its operational implementation demands vast computational resources. While artificial intelligence (AI)-based forecasting models offer a computationally efficient alternative, these purely data-driven approaches often sacrifice physical consistency. Here, we bridge physics-based and AI-based models through a novel, efficient hybrid framework that integrates a low-resolution atmospheric dynamical core with a neural operator in the multigrid architecture. This framework achieves performance comparable to that of state-of-the-art medium-range global weather forecasting models, while incurring much lower training costs, and simultaneously enhances the physical consistency that black-box models often lack. Furthermore, our framework provides substantial flexibility in the choice of dynamical cores, since the training process of the neural network does not require gradient propagation through the dynamical core, which ensures scalability to a wide range of operational forecasting systems.

Accurate reconstruction of tidal harmonic constants in complex marginal seas like the South China Sea (SCS) is critical for understanding regional tidal dynamics and improving coastal ocean forecasting. This study integrates the two-dimensional (2D) adjoint tidal model with the Equidistant Nodes Orthogonal Polynomial Fitting (ENOPF) method to reconstruct high-resolution gridded tidal harmonic constants for four tidal constituents—N2, K2, P1, and Q1—across the SCS. The study region is divided into five sub-areas, with ENOPF applied in the central basin and a combination of model assimilation and polynomial fitting used in coastal regions. To address discontinuities at subregional boundaries, four smoothing techniques—Savitzky-Golay filter, Gaussian filter, Bilateral filter, and linear distance weighting—are tested. Results show that the linear distance weighting provides the reasonable cotidal charts while maintaining data fidelity, which is selected for final implementation. The reconstructed results are validated against satellite altimetry (X-TRACK) and tide gauge observations, and benchmarked against six global tidal models, including TPXO10, FES2022, and EOT20, etc. Across all constituents, the reconstruction demonstrates superior or comparable accuracy in both amplitude and phase lag, particularly in coastal areas. This study provides an effective and scalable approach for high-precision tidal reconstruction in marginal seas, with implications for operational oceanography, coastal management, and regional climate studies.

In recent decades, extreme precipitation events have increased dramatically due to global warming, resulting in significant casualties and economic losses. Quantitative Precipitation Nowcasting (QPN), which predicts precipitation intensity within a six-hour timeframe, plays a critical role in public safety, infrastructure protection, transportation management, outdoor event planning, and flood prevention systems. Building on successful innovations in computer vision, deep learning advancements have substantially improved prediction accuracy and transformed QPN methodologies. However, despite the proliferation of research in this rapidly advancing field, comprehensive surveys that systematically examine mainstream techniques and identify key challenges remain limited. This paper provides a thorough review of current deep learning approaches in precipitation nowcasting, examining important challenges, analyzing methodologies across the QPN development lifecycle, and exploring promising research directions. Through systematic synthesis of emerging developments, we aim to foster interdisciplinary collaboration and stimulate continued innovation in this essential field.