Accurate and timely monitoring of spatiotemporal variations in snow is crucial for numerical weather prediction, climate change studies, and disaster weather forecasting. This study leverages multiple datasets, including the atmospheric forcing data for China Meteorological Administration (CMA) Land Surface Data Assimilation System (CLDAS v2.0), Fengyun-4 snow cover data, in-situ snow depth observations, and land surface parameters. Snow simulations of CLM (Community Land Model), Noah (the Community Noah Land Surface Model), and Noah-MP (Noah with Multi-Parameterization Options) land surface models, the "Ensemble Square Root Kalman Filter+Direct Insertion" (EnSRF+DI) snow assimilation technique, the TC (Triple Collocation) covariance-based multi-model integration, and a multi-grid variational analysis method accounting for elevation are employed in this study to develop a multi-source merged snow analysis product over China at the spatial and temporal resolutions of 6.25 km and 1 h respectively. Evaluation based on in-situ observations demonstrates that the quality of this product is generally superior to those of international counterparts, such as GLDAS and ERA5-Land snow depth products. Compared to snow depth simulations solely from land surface models, the assimilation of Fengyun-4 satellite snow cover data reduces the Root Mean Square Error (RMSE) of snow depth estimates by 10%. The product is provided in NetCDF format, with temporal coverage beginning in November 2021 and continuing in near real-time, and an approximate data volume of 45 MB/h. It includes key variables such as snow depth, snow cover fraction, and snow depth change. This product has been recognized as a high-value meteorological dataset by the CMA and has been operationally applied in weather briefings, disaster weather monitoring, and snowmelt flood risk analysis.

Atmospheric reanalysis is a process that utilizes numerical weather prediction (NWP) models and data assimilation systems to integrate historical meteorological observations, thereby reconstructing datasets that represent past atmospheric states. This paper presents the China Meteorological Administration’s (CMA) latest global atmospheric reanalysis product, CRA1.5, which employs the Global Spectral Model (GSM) and the Gridpoint Statistical Interpolation (GSI) Data Assimilation system. CRA1.5 features a model spectral resolution of T_L1534 (equivalent to approximately 13 km in horizontal grid spacing), a product time interval of 1 h, a model top at 0.27 hPa, and a hybrid-4DEnVar scheme along with an Ensemble Kalman Filter for data assimilation. The temporal coverage of CRA1.5 extends from 1979 to near real-time. During the reanalysis production, we assimilated a substantial volume of reprocessed satellite data and extensive conventional observations. These observations include dense networks of conventional observations within China (comprising approximately 120 sounding stations and 2400 meteorological stations) as well as satellite observations from China’s Fengyun, Haiyang, and Yunyao satellite series. Compared to CRA-40, the root-mean-square error (RMSE) of geopotential height at 500 hPa is reduced by 22.46%, the RMSE of temperature at 200 hPa is reduced by 21.7%, and the RMSE of zonal wind at 850 hPa is reduced by 14.41%. CRA1.5 outperforms ERA5 in 100-m wind speed over China (RMSE: 3.21–3.36 m s⁻¹ vs. 3.46–3.59 m s⁻¹) and achieves comparable accuracy to ERA5 in atmospheric precipitable water (RMSE: 3.10 mm vs. 3.13 mm). CRA1.5’s 2-m temperature trend aligns closely with HadCRUT5, capturing global warming accurately. With higher spatiotemporal resolution (0.1° × 0.1°, hourly) and advanced assimilation, CRA1.5 provides a critical dataset for climate monitoring, NWP, and AI-based meteorological research. CRA1.5 has been implemented for real-time operation at the CMA. Updates are available up to 6 h after the analysis time, with a final update provided at a 3-day lag.

In order to study the simulation effects of different land surface models on soil temperature in China, the Noah and Noah-MP land surface models are driven by the atmospheric driving data of the China Meteorological Administration (CMA) land surface data assimilation system (CLDAS) to simulate soil temperature in China. Soil temperature simulations of GLDAS_Noah, CLDAS_Noah and CLDAS_Noah-MP are evaluated from the perspectives of spatial distribution, different seasons, time series in different regions, etc. based on soil temperature observations collected at 2380 sites during 2010—2018 in China and the Noah soil temperature from the United States Global Land Data Assimilation System (GLDAS_Noah). This study realizes the comparative analysis of soil temperature simulated with different driving data, with same land surface models and same driving data, and with different land surface models. The results show that the GLDAS_Noah, CLDAS_Noah and CLDAS_Noah-MP can reasonably simulate spatial distributions of soil temperature at 10 and 40 cm-depth in China from a qualitative point of view, but there are certain differences in magnitude, which mainly occur in snow-covered areas of Northeast China, Xinjiang and Qinghai-Tibet Plateau. From a quantitative perspective, with the same land surface model and different driving data, CLDAS_Noah is better than GLDAS_Noah in different seasonal assessments based on bias spatial distributions and RMSE (Root Mean Square Error) time series in different regions. This result can indirectly show that CLDAS atmospheric driving data is better than GLDAS Atmospheric driving data and the atmospheric driving data is one of the important factors to improve the accuracy of soil temperature simulation. With the same driving data and different land surface models, the overall effect of CLDAS_Noah-MP is better than that of CLDAS_Noah. Among them, the errors of CLDAS_Noah winter soil temperature at 10 and 40 cm depths in snow-covered areas are significantly greater than that of CLDAS_Noah-MP, which may be related to the improvement of Noah-MP parameterization scheme in snow-covered areas. However, the spring soil temperature simulation errors of CLDAS_Noah-MP at 10 and 40 cm depths in Northeast, North China, and Qinghai-Tibet Plateau are significantly larger than that of CLDAS_Noah, which may be related to the snow melting parameterization scheme in the model. In short, this research provides certain references for subsequent development of soil temperature multi-model integration research and in-situ soil temperature data assimilation research and the final development of high-quality soil temperature dataset in China.

Due to lack of a dense network of ground observations in China before 2008, the China Meteorological Administration’s Land Data Assimilation System (CLDAS) faces challenges in directly generating high-resolution and high-quality land assimilation products prior to 2008. To address this issue, this paper proposes a deep learning model based on the Hybrid Attention Transformer (HAT), aiming to improve the downscaling accuracy of high speed winds in the CLDAS2.0 10-m wind field from 6.25 to 1 km by (1) incorporating digital elevation information (DEM), (2) enhancing the loss function, and (3) employing a prediction error method. We utilized data in 2020–2021 for training and validation, and data in 2019 for testing, conducted ablation experiments to verify the effectiveness of each module, while comparing the results with those of the traditional bilinear interpolation method and the UNET model coupled with a dual cross-attention mechanism. The ablation experiment results indicate that in terms of wind speed categories, HAT with DEM performs the best for wind speeds below level 3 on the Beaufort scale, while HAT with DEM, loss function, and prediction error improvements excels for wind speeds above level 4. Specifically, for wind speeds above level 6, the HAT with all the three improvement measures achieves decent results, with mean absolute error (MAE), probability of detection (POD), and threat score (TS) of 0.825 m s⁻¹, 0.813, and 0.607, respectively, when evaluated against CLDAS3.0 as the ground truth. The model performs better in March–May and November, while its performance is the weakest in June–August; it also performs better during the day than at night and shows suboptimal performance over the plains. The model is closer to the ground truth in reconstructing the structural details of wind fields and outperforms the annual average during most high wind weather events, indicating better predictive capability and adaptability for such events. Overall, the HAT with all the three proposed improvements demonstrates significant progress in downscaling predictions of high winds and provides insights into generation of high-resolution historical meteorological gridded data.