Ningxia Meteorological Bureau

Under the influence of global climate change and human activities, the frequency and intensity of extreme weather events—such as heavy precipitation and severe droughts—have increased markedly. Flood disasters triggered by intense rainfall have severely threatened lives, property, and regional socioeconomic development. To address the challenge of precise prevention and control of short‑duration rainstorm‑induced flash floods in the complex terrain of Northwest China, this study focuses on the Ningxia region, located within China’s arid‑semi‑arid transition zone. By integrating Water Internet technology, big data, and deep learning, we construct an intelligent flash flood disaster prevention and control system.In rainfall forecasting, we have (1) developed a radar‑based precipitation retrieval model through data fusion and calibration, achieving a retrieval accuracy of R² > 0.75 and NMAE < 0.3; (2) proposed an attention‑mechanism‑driven radar echo extrapolation technique that attains over 80% accuracy for a 3‑hour lead time; and (3) built a rapid‑cycle, multi‑source data assimilation rainfall forecast model incorporating GNSS water vapor tomography.For flood forecasting, we (1) introduced a forecasting technique that couples multi‑source rainfall predictions with a distributed hydrological model, yielding accuracy above 80%; and (2) constructed a runoff simulation model for mountainous basins by integrating radar and terrain data with adaptive pooling and attention mechanisms, achieving over 85% forecast accuracy.In the domain of intelligent flood regulation, a real‑time operational model based on rainfall‑runoff forecasting has been developed. By combining flood forecasts with a simplified inundation model, the system enables large‑scale watershed flood analysis.

Field observations provide direct evidence of how does carbon cycling in permafrost ecosystems respond to climate change. This study provides a comprehensive dataset on the impact of warming on carbon cycling and greenhouse gas (GHG) fluxes in permafrost ecosystems. The dataset is extracted and integrated from 132 peer-reviewed studies with 1430 paired observations across eight major permafrost ecosystems, including Arctic and subarctic tundra and wetland, and alpine meadow, steppe, tundra and wetland. This dataset includes 17 variables from experiments conducted during the growing season, covering the plant and soil carbon pools, soil nitrogen pool, and GHG (i.e., CO2, CH4, and N2O) fluxes, among others. Background information on site climate conditions, vegetation and soil characteristics, and details of the warming experiments, including timing, methods, and warming magnitude, are also contained in the dataset. This dataset facilitates a comprehensive understanding of the impact of warming on carbon cycling and GHG fluxes in permafrost ecosystems, and provides supports for meta-analyses and literature reviews, remote sensing data validation, and land model development and parameterization.

In this study, a machine learning-based quality control algorithm for heavy rainfall was developed by integrating automatic weather station observations with remote sensing data, minute-level data, and metadata. Based on heavy rainfall samples from 1 June 2022 to 31 December 2024, the performances of four gradient boosting models—eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), Categorical Boosting (CatBoost), and Gradient Boosted Regression Trees (GBRT)—significantly outperformed precipitation-threshold-based conventional methods, including regional extreme value checks, temporal consistency checks, and others. Specifically, the XGBoost in particular achieves an increase in precision by 0.110 and recall by 0.162. This translates to a substantial reduction in both false alarms (higher precision) and missed detections (higher recall) of anomalous heavy rainfall events, thereby significantly enhancing the reliability of the quality-controlled data. The radar composite reflectivity, satellite cloud-top temperature, and minute-level precipitation were identified as dominant contributors to model predictions. The integration of multi-sensor observations effectively addressed limitations inherent in conventional threshold-based approaches. Through SHapley Additive exPlanations (SHAP)-based interpretability analysis, the model’s decision logic was shown to align with meteorological physical principles. Characteristic patterns such as combinations of low radar reflectivity and elevated cloud-top temperatures were flagged as anomalous rainfall events, typically corresponding to manual operational errors. Moreover, the model identified anomalous minute-level precipitation extremes to be critical signals for detecting instrument malfunctions, data encoding and transmission errors. The physical consistency of the model’s reasoning enhances its trustworthiness and supports its potential for operational implementation in heavy rainfall quality control.

Aiming at the problem of systematic deviation in the regional application of the national meteorological large model “Fenglei-V1”, this paper proposes a lightweight visual correction algorithm based on radar echo images. The algorithm utilizes ConvGRU-based spatiotemporal modeling integrated with dual-attention mechanisms (spatial + temporal) to emulate forecasters‘ decision-making logic of ‘dynamic focusing on critical regions → sequential correction’, achieving intelligent calibration for 0-2h radar echo nowcasting. Validation in Zhejiang region demonstrates a performance breakthrough: the TS score for strong convective cores (≥50dBz) has leaped from 0 to 0.06-0.09, while the Probability of Detection (POD) for ≥35dBz convective echoes has increased by 2 to 6 times; Error control: MAE and RMSE reduced by over 30% compared to the baseline model, with bias approaching the ideal value of 1; Efficiency advantages: Inference speed <0.2 seconds per iteration with seamless integration into business platforms. Case validation demonstrates that the algorithm significantly improves the depiction of typhoon spiral rainband structures and the forecasting of heavy precipitation core intensity, yet limitations persist in predicting newly formed convective cells and forecasts beyond 2-hour lead times. This algorithm has been deployed in grassroots-level nowcasting warning operations, providing forecasters with automated and standardized decision-making support.

ABSTRACT Compound drought and heatwave (CDHW) events are complex climate extremes driven by global climate change, making it challenging to estimate their comprehensive risk using univariate statistics. To address this challenge, we construct a two‐dimensional joint function based on the duration and severity of CDHWs using China as a case study, enabling assessment of joint occurrence probability under diverse scenarios. Whilst previous studies have conducted multi‐dimensional risk assessments, most have focused on individual variables or examined specific aspects of drought‐heatwave relationships. By simultaneously integrating duration and severity through a bivariate joint function, our approach advances this line of research and provides a comprehensive multi‐dimensional framework for assessing the compound characteristics of CDHWs. The results revealed that for China as a whole, changes in the severity threshold had a greater impact on extreme CDHWs, and the joint occurrence probability of severe CDHWs (events with a duration exceeding 7 days and a severity surpassing the 80th, 90th, or 95th percentile) was more sensitive to the severity. The conditional probability rose more rapidly for short‐term events (3 days) than for long‐term events (5–7 days). When the duration exceeded 7 days and the severity exceeded different thresholds, the joint occurrence probability of CDHWs was within the range of 2%–6%. Amongst the different regions, the duration and severity had varying impacts on the joint occurrence probability. The extreme CDHWs in North China, Northeast China, and western Northwest China were more strongly influenced by the duration. Jianghuai, South China, Southwest China, eastern Northwest China, and the Tibetan Plateau were more prone to longer‐lasting extreme CDHWs. In these areas, when CDHWs lasting more than 7 days occurred, changing the severity threshold had a greater impact on their extremity. In North China in 1997, the longest CDHW had joint return periods of over 1000 years when both duration and severity exceeded thresholds, and over 500 years when either exceeded thresholds. The findings demonstrate the variations in the impacts of duration and severity on the joint probability of CDHWs in different regions of China.