ISSN 1000-0526
CN 11-2282/P
CN 11-2282/P
Volume 52,2026 Issue 3
indexed by
- Recommended Topics
-
- The Future Projections of Extreme Weather, Climate and Water Events and Strategic Responses
- A Review of Downburst Genesis Mechanism and Warning
- The Pattern Structure and Thermodynamic and Dynamic Processes of Severe Storms Associated with Linear Convective Gales
- Analysis on Extremity and Characteristics of the “21·7” Severe Torrential Rain in Henan Province
- Diagnostic Analysis on Water Vapor and Jet Characteristics of the July 2021 Severe Torrential Rain in Henan Province
- Current Issue
- Online First
- Archive
- Most Downloaded
Select AllDeselectExport
Display Method:
2026,52(3):257-272, DOI: 10.7519/j.issn.1000-0526.2026.030902
Abstract:
The Meiyu frontal heavy rainfall is a heavy precipitation phenomenon of China’s Yangtze River Basin and the East Asian Region. Conducting integrative field experiments on Meiyu frontal heavy rainfall to thoroughly study the thermodynamic and dynamic processes, moisture transport, microphysical structure, and their evolution mechanisms has important scientific value. In summer, over 40% of the moisture for precipitation in the Yangtze River Basin originate from the Indian Ocean and the Bay of Bengal in South Asia. Moreover, there is also a moisture pathway from the Qinghai-Xizang Plateau, and more than 60% of the precipitation are closely related to the plateau and weather systems on its eastern side. Therefore, the observational study of the South Asian moisture transport pathway to the Yangtze River Basin and of the vertical structural evolution of eastward-moving plateau cloud clusters is very important. To this end, the Wuhan Institute of Heavy Rain of China Meteorological Administration conducted in-depth studies on the mechanism for moisture to be transported from South Asia and the southern slope of Qinghai-Xizang Plateau into the Yangtze River Basin, on the mechanism how eastward-moving plateau cloud clusters affect Meiyu frontal heavy rainfall, and on the frontal structure of the Meiyu frontal surface, cloud microphysical processes, and the upstream to downstream effects of the Meiyu frontal system. They have established a field experiment system for heavy rainfall spanning from moisture source regions and plateau-system source areas to the middle and lower reaches of Yangtze River, and carried out a series of field experiments, which include observation of Meiyu frontal surface and mesoscale systems of heavy rainfall, observation of typical orographic heavy rainfall, observation of the vertical structure of eastward-moving plateau cloud clusters, observation of the characteristics and evolution of the moisture channel in the Yarlung Zangbo Grand Canyon, and tracking observation of extreme heavy rainfall systems. The research results can provide useful references for organizing similar experiments in the future.
The Meiyu frontal heavy rainfall is a heavy precipitation phenomenon of China’s Yangtze River Basin and the East Asian Region. Conducting integrative field experiments on Meiyu frontal heavy rainfall to thoroughly study the thermodynamic and dynamic processes, moisture transport, microphysical structure, and their evolution mechanisms has important scientific value. In summer, over 40% of the moisture for precipitation in the Yangtze River Basin originate from the Indian Ocean and the Bay of Bengal in South Asia. Moreover, there is also a moisture pathway from the Qinghai-Xizang Plateau, and more than 60% of the precipitation are closely related to the plateau and weather systems on its eastern side. Therefore, the observational study of the South Asian moisture transport pathway to the Yangtze River Basin and of the vertical structural evolution of eastward-moving plateau cloud clusters is very important. To this end, the Wuhan Institute of Heavy Rain of China Meteorological Administration conducted in-depth studies on the mechanism for moisture to be transported from South Asia and the southern slope of Qinghai-Xizang Plateau into the Yangtze River Basin, on the mechanism how eastward-moving plateau cloud clusters affect Meiyu frontal heavy rainfall, and on the frontal structure of the Meiyu frontal surface, cloud microphysical processes, and the upstream to downstream effects of the Meiyu frontal system. They have established a field experiment system for heavy rainfall spanning from moisture source regions and plateau-system source areas to the middle and lower reaches of Yangtze River, and carried out a series of field experiments, which include observation of Meiyu frontal surface and mesoscale systems of heavy rainfall, observation of typical orographic heavy rainfall, observation of the vertical structure of eastward-moving plateau cloud clusters, observation of the characteristics and evolution of the moisture channel in the Yarlung Zangbo Grand Canyon, and tracking observation of extreme heavy rainfall systems. The research results can provide useful references for organizing similar experiments in the future.
2026,52(3):273-286, DOI: 10.7519/j.issn.1000-0526.2025.091301
Abstract:
In order to study the characteristics of convective cloud precipitation in Mount Oomolangma Region (MOR), the comprehensive observation with multi-source remote sensing detection equipment for convective cloud precipitation on the northern slope of Mount Oomolangma about 6 years was conducted based on the second comprehensive scientific investigation and research plan of the Qinghai-Xizang Plateau. In this paper, the methods such as data quality control, data retrieval and data statistics are adopted to carry out comparative analysis, and the results show that precipitation in the MOR has unique structural characteristics of convective cloud precipitation compared to the precipitation in other regions. In terms of macroscopic characteristics, the MOR precipitation occurs frequently but has a short duration, most of which are isolated convective cell. The horizontal scale and vertical extension thickness of the precipitation are much smaller than in other areas, and rainfall rate is weak in the boundary layer. As for microscopic characteristics, there are two phases of precipitation (graupel and liquid raindrops) in the boundary layer under different vertical profile distributions of atmospheric temperature, with distinctive characteristics of a narrow raindrop size distribution width, higher raindrop number concentration and smaller raindrop equivalent diameter. This study may bridge the gap in detailed observations of the synoptic structure of the convective cloud precipitation on the northern slope of the MOR in summer and provide a significant reference for further researches on the changes in the weather conditions in Qinghai-Xizang Plateau and their potential impacts on the East Asian climate.
In order to study the characteristics of convective cloud precipitation in Mount Oomolangma Region (MOR), the comprehensive observation with multi-source remote sensing detection equipment for convective cloud precipitation on the northern slope of Mount Oomolangma about 6 years was conducted based on the second comprehensive scientific investigation and research plan of the Qinghai-Xizang Plateau. In this paper, the methods such as data quality control, data retrieval and data statistics are adopted to carry out comparative analysis, and the results show that precipitation in the MOR has unique structural characteristics of convective cloud precipitation compared to the precipitation in other regions. In terms of macroscopic characteristics, the MOR precipitation occurs frequently but has a short duration, most of which are isolated convective cell. The horizontal scale and vertical extension thickness of the precipitation are much smaller than in other areas, and rainfall rate is weak in the boundary layer. As for microscopic characteristics, there are two phases of precipitation (graupel and liquid raindrops) in the boundary layer under different vertical profile distributions of atmospheric temperature, with distinctive characteristics of a narrow raindrop size distribution width, higher raindrop number concentration and smaller raindrop equivalent diameter. This study may bridge the gap in detailed observations of the synoptic structure of the convective cloud precipitation on the northern slope of the MOR in summer and provide a significant reference for further researches on the changes in the weather conditions in Qinghai-Xizang Plateau and their potential impacts on the East Asian climate.
2026,52(3):287-300, DOI: 10.7519/j.issn.1000-0526.2025.101501
Abstract:
On the night of 12 July 2023, a rarely-seen extreme torrential rain event occurred near Feixian County in Shandong Province, under the influence of a severe precipitation supercell. This study comprehensively uses multi-source observational data and ERA5 reanalysis data in investigating the circulation background, coupling mechanism of upper- and lower-level jets, characteristics of meso- and small-scale weather systems, and microphysical features of precipitation associated with this extreme event. The results show that this event was driven by the synergistic effect of multi-scale systems. Favorable circulation patterns, ambient wind fields, and water vapor transport provided large-scale background conditions for the short-time heavy rainfall. The rapidly developing southwest boundary layer jet (SW-BLJ) coupled with the synoptic-scale southwest low-level jet (SW-LLJ) over southern Shandong enhanced upward motion in the middle and lower troposphere. Meanwhile, convergence in the middle and lower troposphere, combined with upper-level divergence in the northeastern divergent zone of the South Asian high, further strengthened deep-layer upward motion. Under the combined influence of the double low-level jets, mesoscale convective systems developed intensely in southern Shandong, promoting the occurrence of heavy rainfall. Furthermore, the event exhibited mixed characteristics of both tropical maritime and continental convective types. At the initial stage of precipitation, there was a sharp increase in raindrop number concentration and a significant enlargement of raindrop diameters. During the peak rainfall period, small particles with a mass-weighted mean diameter (Dm)<2 mm accounted for nearly 85.00%, and the number of large raindrops increased with the intensification of rainfall intensity. In terms of contribution to precipitation, although medium-sized particles (2-3 mm) only accounted for 11.77%, they contributed most to precipitation, reaching 32.74%, followed by small particles (1-2 mm). Although particles with Dm<1 mm were the most numerous, accounting for 43.35%, their contribution to precipitation was less than 4.00%. The severe precipitation supercell stage showed an “upscaling” feature. Particles in the 3-4 mm range, although accounting for less than 6.00% in terms of frequency, contributed 30.04% of the total precipitation, while particles with Dm>6 mm contributed 8.81% to the precipitation.
On the night of 12 July 2023, a rarely-seen extreme torrential rain event occurred near Feixian County in Shandong Province, under the influence of a severe precipitation supercell. This study comprehensively uses multi-source observational data and ERA5 reanalysis data in investigating the circulation background, coupling mechanism of upper- and lower-level jets, characteristics of meso- and small-scale weather systems, and microphysical features of precipitation associated with this extreme event. The results show that this event was driven by the synergistic effect of multi-scale systems. Favorable circulation patterns, ambient wind fields, and water vapor transport provided large-scale background conditions for the short-time heavy rainfall. The rapidly developing southwest boundary layer jet (SW-BLJ) coupled with the synoptic-scale southwest low-level jet (SW-LLJ) over southern Shandong enhanced upward motion in the middle and lower troposphere. Meanwhile, convergence in the middle and lower troposphere, combined with upper-level divergence in the northeastern divergent zone of the South Asian high, further strengthened deep-layer upward motion. Under the combined influence of the double low-level jets, mesoscale convective systems developed intensely in southern Shandong, promoting the occurrence of heavy rainfall. Furthermore, the event exhibited mixed characteristics of both tropical maritime and continental convective types. At the initial stage of precipitation, there was a sharp increase in raindrop number concentration and a significant enlargement of raindrop diameters. During the peak rainfall period, small particles with a mass-weighted mean diameter (Dm)<2 mm accounted for nearly 85.00%, and the number of large raindrops increased with the intensification of rainfall intensity. In terms of contribution to precipitation, although medium-sized particles (2-3 mm) only accounted for 11.77%, they contributed most to precipitation, reaching 32.74%, followed by small particles (1-2 mm). Although particles with Dm<1 mm were the most numerous, accounting for 43.35%, their contribution to precipitation was less than 4.00%. The severe precipitation supercell stage showed an “upscaling” feature. Particles in the 3-4 mm range, although accounting for less than 6.00% in terms of frequency, contributed 30.04% of the total precipitation, while particles with Dm>6 mm contributed 8.81% to the precipitation.
2026,52(3):301-311, DOI: 10.7519/j.issn.1000-0526.2025.092801
Abstract:
This article utilizes North China regional radar composite reflectivity factor mosaic products, single-site radar data from northern Shanxi, ERA5 reanalysis data, and surface observation data collected in the warm season (May-September) of 2021-2023, and conducts statistical analyses on the changing characteristics of squall lines descending mountains in this region. The results show that a total of 29 squall lines are identified, and following the moving directions of the squall lines, they are classified into four types: west-moving, northwest-moving, north-moving, and basin-originating. Then, based on their intensity changes when descending mountains, they are classified into three types: intensifying, weak-ening, and maintaining types, of which the weakening type is the most common, accounting for 67% of the total. All west-moving squall lines belong to the weakening type upon descent, while all north-moving types of squall lines are of the intensifying type. The northwest-moving type includes squall lines that intensify, weaken or maintain the intensity upon descent. Analysis of the environmental background ahead of the descending path for intensifying and weakening types within the northwest-moving squall lines reveals that, compared to weakening squall lines, the intensifying ones exhibit slightly stronger dynamic conditions, while conditions related to moisture, convective available potential energy and vertical wind shear are comparable or slightly poorer. Therefore, it’s difficult to accurately forecast whether a squall line will intensify or weaken upon descent based solely on the environmental conditions ahead of its path. For short-time nowcasting, radar data can be used to forecast whether a squall line will intensify or weaken upon descent. The intensifying squall lines upon descent typically exhibit stronger echo intensity, with moving speed around 17 m·s-1. Large gradient zone of reflectivity factor is concentrated in the front section of a squall line, with an overall bow-shaped squall line accompanied by a gust front. Radial velocity cross-section shows a distinct organized structure with forward inflow ascending slantwise along the rear inflow. In contrast, the weakening squall lines upon descent typically have weak or moderate echo intensity, and their moving speeds are generally below 10 m·s-1. Large gradient zone of reflectivity factor appears at the rear part of a squall line, which has overall a straighter shape, not accompanied by gust front. On the radial velocity cross-section, there is no distinct organized structure with forward inflow ascending slantwise along the rear inflow.
This article utilizes North China regional radar composite reflectivity factor mosaic products, single-site radar data from northern Shanxi, ERA5 reanalysis data, and surface observation data collected in the warm season (May-September) of 2021-2023, and conducts statistical analyses on the changing characteristics of squall lines descending mountains in this region. The results show that a total of 29 squall lines are identified, and following the moving directions of the squall lines, they are classified into four types: west-moving, northwest-moving, north-moving, and basin-originating. Then, based on their intensity changes when descending mountains, they are classified into three types: intensifying, weak-ening, and maintaining types, of which the weakening type is the most common, accounting for 67% of the total. All west-moving squall lines belong to the weakening type upon descent, while all north-moving types of squall lines are of the intensifying type. The northwest-moving type includes squall lines that intensify, weaken or maintain the intensity upon descent. Analysis of the environmental background ahead of the descending path for intensifying and weakening types within the northwest-moving squall lines reveals that, compared to weakening squall lines, the intensifying ones exhibit slightly stronger dynamic conditions, while conditions related to moisture, convective available potential energy and vertical wind shear are comparable or slightly poorer. Therefore, it’s difficult to accurately forecast whether a squall line will intensify or weaken upon descent based solely on the environmental conditions ahead of its path. For short-time nowcasting, radar data can be used to forecast whether a squall line will intensify or weaken upon descent. The intensifying squall lines upon descent typically exhibit stronger echo intensity, with moving speed around 17 m·s-1. Large gradient zone of reflectivity factor is concentrated in the front section of a squall line, with an overall bow-shaped squall line accompanied by a gust front. Radial velocity cross-section shows a distinct organized structure with forward inflow ascending slantwise along the rear inflow. In contrast, the weakening squall lines upon descent typically have weak or moderate echo intensity, and their moving speeds are generally below 10 m·s-1. Large gradient zone of reflectivity factor appears at the rear part of a squall line, which has overall a straighter shape, not accompanied by gust front. On the radial velocity cross-section, there is no distinct organized structure with forward inflow ascending slantwise along the rear inflow.
2026,52(3):312-324, DOI: 10.7519/j.issn.1000-0526.2025.122501
Abstract:
Analog forecasting is a widely adopted statistical method in operational meteorological services. Traditional single-layer similarity approaches have such limitations as the lack of three-dimensional spatial information, the unstable performance of single similarity criteria, and the frequent interference from synoptic system pattern and intensity (magnitude). To address these challenges and explore the feasibility of deep learning models in synoptic situation recognition and forecasting, in this study we develop a novel approach using the ECMWF fifth-generation reanalysis (ERA5) dataset. We construct a deep learning architecture that integrates convolutional neural networks (CNN) with Transformer modules, incorporating self-attention mechanisms. Verification shows that this model can effectively capture three-dimensional spatial features of synoptic situation. Then, utilizing the extracted feature vectors, we design a comprehensive similarity framework that combines three complementary metrics: Pearson correlation (empha-sizing pattern shape), Euclidean distance (emphasizing magnitude), and Chebyshev distance (considering both shape and magnitude). This integration forms our proposed method: Synoptic Similarity Net. Finally, the operational application effect of this method is tested and evaluated in detail. The results indicate that this method can achieve the highest average structural similarity index (SSIM) and lowest mean squared error (MSE) relative to the traditional methods, demonstrating significant improvements in both metrics. Case studies across seasons confirm that the historical analogs identified by Synoptic Similarity Net exhibit both greater numerical accuracy and superior spatial pattern consistency compared to the original synoptic fields. These results demonstrate the promising potential of this method for meteorological operational applications.
Analog forecasting is a widely adopted statistical method in operational meteorological services. Traditional single-layer similarity approaches have such limitations as the lack of three-dimensional spatial information, the unstable performance of single similarity criteria, and the frequent interference from synoptic system pattern and intensity (magnitude). To address these challenges and explore the feasibility of deep learning models in synoptic situation recognition and forecasting, in this study we develop a novel approach using the ECMWF fifth-generation reanalysis (ERA5) dataset. We construct a deep learning architecture that integrates convolutional neural networks (CNN) with Transformer modules, incorporating self-attention mechanisms. Verification shows that this model can effectively capture three-dimensional spatial features of synoptic situation. Then, utilizing the extracted feature vectors, we design a comprehensive similarity framework that combines three complementary metrics: Pearson correlation (empha-sizing pattern shape), Euclidean distance (emphasizing magnitude), and Chebyshev distance (considering both shape and magnitude). This integration forms our proposed method: Synoptic Similarity Net. Finally, the operational application effect of this method is tested and evaluated in detail. The results indicate that this method can achieve the highest average structural similarity index (SSIM) and lowest mean squared error (MSE) relative to the traditional methods, demonstrating significant improvements in both metrics. Case studies across seasons confirm that the historical analogs identified by Synoptic Similarity Net exhibit both greater numerical accuracy and superior spatial pattern consistency compared to the original synoptic fields. These results demonstrate the promising potential of this method for meteorological operational applications.
2026,52(3):325-336, DOI: 10.7519/j.issn.1000-0526.2025.091601
Abstract:
Based on thunderstorm gale cases in Sichuan Basin from March 1 to September 30 in 2018-2022, combined with three-dimensional radar mosaic data and surface maximum wind observations, this paper constructs a thunderstorm gale sample dataset and develops a grid-point thunderstorm gale warning model. Independent validation is performed on thunderstorm gale events in 2023 and the warning performance of four models is evaluated. The results show that the LightGBM model achieves the highest probability of detection (POD), reaching 0.536 at a 15 min lead time and a 10 km evaluation radius, but it also exhibits the highest false alarm rate (FAR). The random forest model demonstrates the optimal comprehensive performance, with the highest critical success index (CSI) being 0.306 at a 30 min lead time and a 10 km evaluation radius. Both CSI and POD decrease significantly with prolonging warning lead time or decreasing evaluation radius, with a particularly notable decline in CSI when the lead time extends from 30 to 45 min. Synoptic conditions significantly influence the warning performance. Under pronounced cold air influence, factors such as echo intensity, echo top height, and 45 dBz echo top height are more likely to have high values, favoring the development of severe convection. However, newly initiated storms at convective fronts often lead to the increase in missed detections. In the absence of strong cold air, thunderstorm gales mainly occur at the leading edge of convective systems, resulting in higher POD. The temporal variation of vertically integrated liquid water content contributes most to the decision-making of models, followed by vertically integrated liquid water content density, echo top height, and maximum reflectivity factor. This highlights the central role of deep convection in the generation of thunderstorm gales. In the scenarios without cold air intrusion, downdrafts play a dominant role in thunderstorm gale warnings. Analysis of key feature values and high SHAP values reveals that temporal variations in convective echoes are critical for effective warnings. Samples with high echo-tracking wind speeds often correspond to positive SHAP values, indicating an increasing probability of convective wind events when echo motion accelerates.
Based on thunderstorm gale cases in Sichuan Basin from March 1 to September 30 in 2018-2022, combined with three-dimensional radar mosaic data and surface maximum wind observations, this paper constructs a thunderstorm gale sample dataset and develops a grid-point thunderstorm gale warning model. Independent validation is performed on thunderstorm gale events in 2023 and the warning performance of four models is evaluated. The results show that the LightGBM model achieves the highest probability of detection (POD), reaching 0.536 at a 15 min lead time and a 10 km evaluation radius, but it also exhibits the highest false alarm rate (FAR). The random forest model demonstrates the optimal comprehensive performance, with the highest critical success index (CSI) being 0.306 at a 30 min lead time and a 10 km evaluation radius. Both CSI and POD decrease significantly with prolonging warning lead time or decreasing evaluation radius, with a particularly notable decline in CSI when the lead time extends from 30 to 45 min. Synoptic conditions significantly influence the warning performance. Under pronounced cold air influence, factors such as echo intensity, echo top height, and 45 dBz echo top height are more likely to have high values, favoring the development of severe convection. However, newly initiated storms at convective fronts often lead to the increase in missed detections. In the absence of strong cold air, thunderstorm gales mainly occur at the leading edge of convective systems, resulting in higher POD. The temporal variation of vertically integrated liquid water content contributes most to the decision-making of models, followed by vertically integrated liquid water content density, echo top height, and maximum reflectivity factor. This highlights the central role of deep convection in the generation of thunderstorm gales. In the scenarios without cold air intrusion, downdrafts play a dominant role in thunderstorm gale warnings. Analysis of key feature values and high SHAP values reveals that temporal variations in convective echoes are critical for effective warnings. Samples with high echo-tracking wind speeds often correspond to positive SHAP values, indicating an increasing probability of convective wind events when echo motion accelerates.
ZHANG Yefang, WU Qishu, LI Tingting, FENG Zhenzhen, LIU Bing, HUANG Lingguang, HE Qingfang, LIAO Kuo
2026,52(3):337-347, DOI: 10.7519/j.issn.1000-0526.2026.010501
Abstract:
Based on the dual-polarization radar particle classification products and cloud-to-ground lightning location data from Fuzhou, Xiamen and Longyan from April 2022 to June 2024, in this study we select 54 convective storm samples with or without cloud-to-ground lightning and investigate the temporal variation characteristics of graupel particle layer thickness during the developing processes of these samples. The results show that the graupel particle layer thickness of convective storms with cloud-to-ground lightning reaches at least 2.26 km at the initiation moment of the first cloud-to-ground lightning, while 95% of the samples without cloud-to-ground lightning, have graupel particle layer thickness less than 2.2 km throughout their lifecycles. Considering both the lead time and accuracy of cloud-to-ground lightning initiation forecasts, we propose to use a graupel particle layer thickness greater than 2 km as a forecast indicator for cloud-to-ground lightning initiation in Fujian Province. The forecast potential is evaluated, and the calculated TS score of the sample is 0.864 and the average forecast lead time is 28.13 min. Taking four convective storms on 14 June 2022 in southern Fujian Province as examples, the above-mentioned forecast indicators are applied and analyzed in these cases. The results show that the proposed forecast indicator can accurately forecast whether cloud-to-ground lightning will occur in the four convective storms, and the forecast lead time for cloud-to-ground lightning initiation is 6 min.
Based on the dual-polarization radar particle classification products and cloud-to-ground lightning location data from Fuzhou, Xiamen and Longyan from April 2022 to June 2024, in this study we select 54 convective storm samples with or without cloud-to-ground lightning and investigate the temporal variation characteristics of graupel particle layer thickness during the developing processes of these samples. The results show that the graupel particle layer thickness of convective storms with cloud-to-ground lightning reaches at least 2.26 km at the initiation moment of the first cloud-to-ground lightning, while 95% of the samples without cloud-to-ground lightning, have graupel particle layer thickness less than 2.2 km throughout their lifecycles. Considering both the lead time and accuracy of cloud-to-ground lightning initiation forecasts, we propose to use a graupel particle layer thickness greater than 2 km as a forecast indicator for cloud-to-ground lightning initiation in Fujian Province. The forecast potential is evaluated, and the calculated TS score of the sample is 0.864 and the average forecast lead time is 28.13 min. Taking four convective storms on 14 June 2022 in southern Fujian Province as examples, the above-mentioned forecast indicators are applied and analyzed in these cases. The results show that the proposed forecast indicator can accurately forecast whether cloud-to-ground lightning will occur in the four convective storms, and the forecast lead time for cloud-to-ground lightning initiation is 6 min.
2026,52(3):348-357, DOI: 10.7519/j.issn.1000-0526.2025.071701
Abstract:
Low-altitude flight activities are highly susceptible to complex wind fields, so civil aviation me-teorological departments place significant emphasis on refined monitoring and early warning technologies of low-altitude winds. This study analyzes a cold front strong wind case at Guanghan Airport on 26 December 2021, by using the data from Doppler wind lidar and airport automatic observation equipment, as well as reanalysis data. The results indicate that the Doppler wind lidar can clearly capture that during the influence of cold air, surface strong winds occur approximately 4-6 h after the low-level jet weakens, with significant in homogeneity in downward momentum transfer and vertical motion changes. Sinking motion enhances the downward momentum transfer of the low-level jet, which result in increased surface wind speeds, while upward motion hinders downward momentum transfer, leading to intermittent surface strong winds. The 12° PPI scanning of the Doppler wind lidar can provide early warnings of potential strong winds along the glide path and at the surface 1.0-1.5 h in advance. The intrusion of residual cold air behind the front in the study case caused a secondary wind event, which should be taken into account when forecasting strong winds. This study demonstrates that Doppler wind lidar can significantly improve the monitoring precision of low-altitude wind field evolution, and could provide valuable insights for ensur-ing the safety and efficiency of low-altitude flight operations.
Low-altitude flight activities are highly susceptible to complex wind fields, so civil aviation me-teorological departments place significant emphasis on refined monitoring and early warning technologies of low-altitude winds. This study analyzes a cold front strong wind case at Guanghan Airport on 26 December 2021, by using the data from Doppler wind lidar and airport automatic observation equipment, as well as reanalysis data. The results indicate that the Doppler wind lidar can clearly capture that during the influence of cold air, surface strong winds occur approximately 4-6 h after the low-level jet weakens, with significant in homogeneity in downward momentum transfer and vertical motion changes. Sinking motion enhances the downward momentum transfer of the low-level jet, which result in increased surface wind speeds, while upward motion hinders downward momentum transfer, leading to intermittent surface strong winds. The 12° PPI scanning of the Doppler wind lidar can provide early warnings of potential strong winds along the glide path and at the surface 1.0-1.5 h in advance. The intrusion of residual cold air behind the front in the study case caused a secondary wind event, which should be taken into account when forecasting strong winds. This study demonstrates that Doppler wind lidar can significantly improve the monitoring precision of low-altitude wind field evolution, and could provide valuable insights for ensur-ing the safety and efficiency of low-altitude flight operations.
ZHOU Xingyan, CHEN Yixiao, YANG Guowei, WANG Yaqi, QIAO Qi, HONG Haixu, YIN Manman, YIN Yizhou, ZHONG Hailing
2026,52(3):358-365, DOI: 10.7519/j.issn.1000-0526.2026.021401
Abstract:
In 2025, the global average surface temperature was 1.40℃ above pre-industrial levels and 0.52℃ above the 1991-2020 average, ranking among the top three warmest years since meteorological records began. The global ocean heat content reached a new record high, while the annual maximum Arctic sea ice extent in March fell to the lowest level in the satellite era. Against such a backdrop, extreme weather and climate events occurred frequently in many parts of the world, leading to serious disaster consequences. Rainstorms and floods severely struck Sudan, Nigeria, Indonesia, Sri Lanka and the Philippines, triggering catastrophic floods and landslides. Extreme heatwaves affected vast regions of North America, East Asia and southern Europe, fueling severe forest fires. In addition, tropical cyclones impacted the Philippines, Vietnam and the Caribbean, severe winter storms gripped North America and East Asia, and moreover, destructive tornados occurred several times in the United States.
In 2025, the global average surface temperature was 1.40℃ above pre-industrial levels and 0.52℃ above the 1991-2020 average, ranking among the top three warmest years since meteorological records began. The global ocean heat content reached a new record high, while the annual maximum Arctic sea ice extent in March fell to the lowest level in the satellite era. Against such a backdrop, extreme weather and climate events occurred frequently in many parts of the world, leading to serious disaster consequences. Rainstorms and floods severely struck Sudan, Nigeria, Indonesia, Sri Lanka and the Philippines, triggering catastrophic floods and landslides. Extreme heatwaves affected vast regions of North America, East Asia and southern Europe, fueling severe forest fires. In addition, tropical cyclones impacted the Philippines, Vietnam and the Caribbean, severe winter storms gripped North America and East Asia, and moreover, destructive tornados occurred several times in the United States.
ZENG Hongling, ZHAO Lin, ZHONG Hailing, GAO Hui, ZOU Xukai, ZHENG Zhihai, WANG Ling, SUN Mingyang, SUN Linhai, ZHAI Jianqing, YIN Yizhou, WANG Youmin, ZHOU Xingyan, ZHU Xiaojin, DAI Tanlong, YIN Manman, ZHANG Yingxian, FU Shuo, ZHAO Yuheng, ZHAO Junhu, LYU Zhuozhuo
2026,52(3):366-373, DOI: 10.7519/j.issn.1000-0526.2026.030301
Abstract:
Using station observation data and reanalysis data, a summary was conducted on the climatic characteristics of China in 2025. The results are as follows. China exhibited distinctly warm and wet climatic characteristics in 2025. The annual average temperature in China was 10.9℃, which is 1.0℃ above normal and ties with that in 2024, marking the highest level since 1951. The annual precipitation was 668.0 mm on average, which is more than normal by 4.5%. The major weather and climate events in 2025 are as follows. In summer, extremely severe rainstorms occurred in North China, Northeast China and Inner Mongolia, with the central and eastern parts of China experiencing the fourth-severest heat wave events since 1961. Autumn typhoons influenced South China more than normal. In addition, severe convective weather events led to significant localized disasters, while the regional meteorological droughts had periodic characteristics clearly. Moreover, there were more cold wave processes and more high wind days recorded. In spring, sand-dust events became a frequent phenomenon.
Using station observation data and reanalysis data, a summary was conducted on the climatic characteristics of China in 2025. The results are as follows. China exhibited distinctly warm and wet climatic characteristics in 2025. The annual average temperature in China was 10.9℃, which is 1.0℃ above normal and ties with that in 2024, marking the highest level since 1951. The annual precipitation was 668.0 mm on average, which is more than normal by 4.5%. The major weather and climate events in 2025 are as follows. In summer, extremely severe rainstorms occurred in North China, Northeast China and Inner Mongolia, with the central and eastern parts of China experiencing the fourth-severest heat wave events since 1961. Autumn typhoons influenced South China more than normal. In addition, severe convective weather events led to significant localized disasters, while the regional meteorological droughts had periodic characteristics clearly. Moreover, there were more cold wave processes and more high wind days recorded. In spring, sand-dust events became a frequent phenomenon.
2026,52(3):374-384, DOI: 10.7519/j.issn.1000-0526.2026.020101
Abstract:
The main characteristics of atmospheric circulation in December 2025 are as follows. The polar vortex in the Northern Hemisphere exhibited a dipole pattern with a negative geopotential height anomaly. The mid-to-high latitude circulation presented a four-wave pattern, while the Eurasian Continent was governed by a meridional pattern with high pressure in the west and low pressure in the east. The national average precipitation was 8.6 mm, decreasing by 28% compared to the long-term average for the same period. Furthermore, the spatial distribution of the precipitation anomaly percentage was characterized by higher values in northern China and lower values in southern China. The monthly average temperature was -1.2℃, which is 1.8℃ above that in the same period of normal years, ranking as the second highest since 1961. The cold airs were frequent, manifesting as multiple waves of intrusion occurring in stages with relatively weak intensity. In December, China experienced four cold air processes. During mid-December, the widespread cold wave occurred with the first wide range of rain and snow in winter, accompanied by a sand-dust event. Additionally, two fog-haze episodes appeared during the intermittent periods of cold air processes in mid-to-late December.
The main characteristics of atmospheric circulation in December 2025 are as follows. The polar vortex in the Northern Hemisphere exhibited a dipole pattern with a negative geopotential height anomaly. The mid-to-high latitude circulation presented a four-wave pattern, while the Eurasian Continent was governed by a meridional pattern with high pressure in the west and low pressure in the east. The national average precipitation was 8.6 mm, decreasing by 28% compared to the long-term average for the same period. Furthermore, the spatial distribution of the precipitation anomaly percentage was characterized by higher values in northern China and lower values in southern China. The monthly average temperature was -1.2℃, which is 1.8℃ above that in the same period of normal years, ranking as the second highest since 1961. The cold airs were frequent, manifesting as multiple waves of intrusion occurring in stages with relatively weak intensity. In December, China experienced four cold air processes. During mid-December, the widespread cold wave occurred with the first wide range of rain and snow in winter, accompanied by a sand-dust event. Additionally, two fog-haze episodes appeared during the intermittent periods of cold air processes in mid-to-late December.
Select AllDeselectExport
Display Method:
Available online: April 18, 2026 , DOI: 10.7519/j.issn.1000-0526.2026.041601
Abstract:
Based on visibility observation, millimeter wave radar, EC reanalysis data, and wrfv4.2 version, IAP-LACS (3M), Morrison05 (2M) and Thompson 08 (2M) tests were designed for a high concentration, wide range and high impact sea fog process in April 2021. SW99 algorithm and FSL algorithm were used to diagnose and analyze visibility, and the following main conclusions were obtained: from April 1 to 2, 2021, 850hPa Zhejiang was between southwest wind and easterly shear. The southwest wind in the South was conducive to water vapor transport and the development of temperature ridge. There was an inversion layer near the surface along the coast. The junction is stable and the water vapor condition is good. In this process, the FSL algorithm diagnosis visibility results are obviously strong, and there is no gradient. SW99 algorithm is close to the actual situation in terms of intensity and range. The IAP-LACS (3M) of SW99 algorithm performs best in strength and range compared with the other two schemes. IAP-LACS (3M) better simulated the high value area of low-level cloud water content in Zhoushan and Taizhou Wenzhou coast, and better simulated the sea fog enhancement process along the coast of Sanshan gate from 22:00 on the 1st to 00:00 on the 2nd. The improvement of this scheme on the simulation of low-level cloud water content in the sea fog process is an important reason to improve the prediction effect of the sea fog. The comparative analysis of the three groups of schemes on the element simulation of Yuhuan station shows that the temperature simulation of 2m in the three groups of schemes from 12:00 to 16:00 on the 1st day is high, but the dew point temperature is close to the actual situation, so the relative humidity in the corresponding time is relatively small, which may be one of the reasons for the missing report of sea fog in Taizhou Wenzhou coastal area at the corresponding time. The IAP-LACS (3M) has the best temperature and dew point temperature simulation effect after 20:00 on the 1st day. In this process, IAP-LACS (3M) has a good matching relationship between the distribution of the high value area of numerical concentration output and the distribution of low visibility. At present, most cloud micro schemes do not output numerical concentration, and the droplet number concentration is an important parameter affecting visibility. IAP-LACS (3M) provides conditions for the introduction of numerical concentration into visibility diagnosis scheme.
Based on visibility observation, millimeter wave radar, EC reanalysis data, and wrfv4.2 version, IAP-LACS (3M), Morrison05 (2M) and Thompson 08 (2M) tests were designed for a high concentration, wide range and high impact sea fog process in April 2021. SW99 algorithm and FSL algorithm were used to diagnose and analyze visibility, and the following main conclusions were obtained: from April 1 to 2, 2021, 850hPa Zhejiang was between southwest wind and easterly shear. The southwest wind in the South was conducive to water vapor transport and the development of temperature ridge. There was an inversion layer near the surface along the coast. The junction is stable and the water vapor condition is good. In this process, the FSL algorithm diagnosis visibility results are obviously strong, and there is no gradient. SW99 algorithm is close to the actual situation in terms of intensity and range. The IAP-LACS (3M) of SW99 algorithm performs best in strength and range compared with the other two schemes. IAP-LACS (3M) better simulated the high value area of low-level cloud water content in Zhoushan and Taizhou Wenzhou coast, and better simulated the sea fog enhancement process along the coast of Sanshan gate from 22:00 on the 1st to 00:00 on the 2nd. The improvement of this scheme on the simulation of low-level cloud water content in the sea fog process is an important reason to improve the prediction effect of the sea fog. The comparative analysis of the three groups of schemes on the element simulation of Yuhuan station shows that the temperature simulation of 2m in the three groups of schemes from 12:00 to 16:00 on the 1st day is high, but the dew point temperature is close to the actual situation, so the relative humidity in the corresponding time is relatively small, which may be one of the reasons for the missing report of sea fog in Taizhou Wenzhou coastal area at the corresponding time. The IAP-LACS (3M) has the best temperature and dew point temperature simulation effect after 20:00 on the 1st day. In this process, IAP-LACS (3M) has a good matching relationship between the distribution of the high value area of numerical concentration output and the distribution of low visibility. At present, most cloud micro schemes do not output numerical concentration, and the droplet number concentration is an important parameter affecting visibility. IAP-LACS (3M) provides conditions for the introduction of numerical concentration into visibility diagnosis scheme.
Available online: April 17, 2026 , DOI: 10.7519/j.issn.1000-0526.2025.100701
Abstract:
Using conventional observation data and Suizhou S-band dual polarization weather radar observation data, short-time heavy rain is classified into strong weather scale system forcing type (referred to as "strong forcing type") and weak weather scale system forcing type (referred to as "weak forcing type") according to the weather circulation pattern. The distribution characteristics and dual polarization radar echo characteristics of the two types are compared and analyzed.The results show that:(1)Strong forcing type have more occurrences, wider ranges, more stations, and longer durations.(2)In the strong forcing type, July has the highest frequency, followed by June, and the lowest in August. Dahong Mountain and Tongbai Mountain being the most frequent areas.The daily variation characteristics show a three peak distribution, with the highest occurrence occurring from 05:00 to 08:00, followed by 23:00 to 00:00, and the third occurring around 15:00. In the weak forcing type, the frequency of occurrence is basically the same in July and August, with the least in June. The distribution of high-frequency areas is scattered, but mountainous areas are still high-frequency areas. The daily variation follows a unimodal distribution, with the peak occurring at 15:00.(3)Both two types have the characteristics of higher rainfall intensity, stronger echo intensity, higher concentration of liquid particles, higher echo top, larger diameter of liquid particles. (4)In the same rainfall intensity level, there is no significant difference in the combined combined reflectivity(CR), 1 km horizontal reflectivity(Zh), and echo top height (ET) of two types of short-term heavy precipitation, and they all have low centroid characteristics, with the strong forcing type having more obvious low centroid characteristics. However, the interval and average values of the 1 km differential reflectivity(Zdr) and specific differential phase (Kdp) for the strong forcing type are significantly smaller than those for the weak forcing type, while the interval and average values of the 1 km correlation coefficient(CC) are slightly larger than those for the weak forcing type. This indicates that the weak forcing type has larger particle diameters in the lower layers and is more prone to incomplete melting of ice phase particles in the lower layers.
Using conventional observation data and Suizhou S-band dual polarization weather radar observation data, short-time heavy rain is classified into strong weather scale system forcing type (referred to as "strong forcing type") and weak weather scale system forcing type (referred to as "weak forcing type") according to the weather circulation pattern. The distribution characteristics and dual polarization radar echo characteristics of the two types are compared and analyzed.The results show that:(1)Strong forcing type have more occurrences, wider ranges, more stations, and longer durations.(2)In the strong forcing type, July has the highest frequency, followed by June, and the lowest in August. Dahong Mountain and Tongbai Mountain being the most frequent areas.The daily variation characteristics show a three peak distribution, with the highest occurrence occurring from 05:00 to 08:00, followed by 23:00 to 00:00, and the third occurring around 15:00. In the weak forcing type, the frequency of occurrence is basically the same in July and August, with the least in June. The distribution of high-frequency areas is scattered, but mountainous areas are still high-frequency areas. The daily variation follows a unimodal distribution, with the peak occurring at 15:00.(3)Both two types have the characteristics of higher rainfall intensity, stronger echo intensity, higher concentration of liquid particles, higher echo top, larger diameter of liquid particles. (4)In the same rainfall intensity level, there is no significant difference in the combined combined reflectivity(CR), 1 km horizontal reflectivity(Zh), and echo top height (ET) of two types of short-term heavy precipitation, and they all have low centroid characteristics, with the strong forcing type having more obvious low centroid characteristics. However, the interval and average values of the 1 km differential reflectivity(Zdr) and specific differential phase (Kdp) for the strong forcing type are significantly smaller than those for the weak forcing type, while the interval and average values of the 1 km correlation coefficient(CC) are slightly larger than those for the weak forcing type. This indicates that the weak forcing type has larger particle diameters in the lower layers and is more prone to incomplete melting of ice phase particles in the lower layers.
Available online: April 15, 2026 , DOI: 10.7519/j.issn.1000-0526.2025.082001
Abstract:
Based on rice phenological stages and meteorological observations from nationwide single-season and double-season rice regions, combined with major agro-meteorological disaster indicators, disaster intensity and hazard indices were constructed to analyze the changing characteristics of heat stress, cold stress, drought, and compound heat and drought stresses from 1991 to 2024. The results indicate that in the period of 1991-2024, heat stress in single-season rice has significantly intensified, with the peak observed in 2022 while the area percentage exposed to heat stress. Meanwhile, early rice heat stress has shown fluctuating changes and has been gradually increasing since 2017. Obstruction-type chilling injury in rice in Northeast China has been weakened, while it weakens after 2017 and merely occurs in recent two years. The intensity of cold dew wind in late rice has been stronger after 2017 and peaked in 2020 after that it decresed. Drought has exhibited large fluctuations, with the most severe conditions occurring in single-season and late rice in 2022, and the area percentage exposed to drought in single-season ric and late rice is higher than that in early rice. The compound heat and drought stresses in single-season rice has been more severe than in early rice. At the spatial scale, areas for severer heat stress in single-season rice have been concentrated in northeastern Sichuan, central Chongqing, and eastern Hubei, while for early rice, they have been located in southeastern Hunan central Zhejiang, central-southern Jiangxi, and central Fujian where the heat stress intend to intensify. The obstruction-type chilling injury in Northeast China have been identified from southwest to northeast, and cold dew wind for late rice tends to increase from south to north. The drought in single-season rice is severe in Northeast China in comparison with that in southern areas, and that in early rice decreases from south to north which is opposite in late rice. High intensity for compound heat and drought stresses in single-season rice have been concentrated in northeastern Sichuan, central Chongqing, eastern Hubei and western Hunan, while for early rice, they have been located in northern Zhejiang and southeastern Jiangxi. In these areas exposed to high intensity of disasters, the hazard is generally heavy, which can help guide for taking prevention of specific disasters. Keywords: rice, heat stress, cold stress, drought, combined disaster, intensity index
Based on rice phenological stages and meteorological observations from nationwide single-season and double-season rice regions, combined with major agro-meteorological disaster indicators, disaster intensity and hazard indices were constructed to analyze the changing characteristics of heat stress, cold stress, drought, and compound heat and drought stresses from 1991 to 2024. The results indicate that in the period of 1991-2024, heat stress in single-season rice has significantly intensified, with the peak observed in 2022 while the area percentage exposed to heat stress. Meanwhile, early rice heat stress has shown fluctuating changes and has been gradually increasing since 2017. Obstruction-type chilling injury in rice in Northeast China has been weakened, while it weakens after 2017 and merely occurs in recent two years. The intensity of cold dew wind in late rice has been stronger after 2017 and peaked in 2020 after that it decresed. Drought has exhibited large fluctuations, with the most severe conditions occurring in single-season and late rice in 2022, and the area percentage exposed to drought in single-season ric and late rice is higher than that in early rice. The compound heat and drought stresses in single-season rice has been more severe than in early rice. At the spatial scale, areas for severer heat stress in single-season rice have been concentrated in northeastern Sichuan, central Chongqing, and eastern Hubei, while for early rice, they have been located in southeastern Hunan central Zhejiang, central-southern Jiangxi, and central Fujian where the heat stress intend to intensify. The obstruction-type chilling injury in Northeast China have been identified from southwest to northeast, and cold dew wind for late rice tends to increase from south to north. The drought in single-season rice is severe in Northeast China in comparison with that in southern areas, and that in early rice decreases from south to north which is opposite in late rice. High intensity for compound heat and drought stresses in single-season rice have been concentrated in northeastern Sichuan, central Chongqing, eastern Hubei and western Hunan, while for early rice, they have been located in northern Zhejiang and southeastern Jiangxi. In these areas exposed to high intensity of disasters, the hazard is generally heavy, which can help guide for taking prevention of specific disasters. Keywords: rice, heat stress, cold stress, drought, combined disaster, intensity index
Available online: April 13, 2026 , DOI: 10.7519/j.issn.1000-0526.2025.122001
Abstract:
In mid-December 2023, Shanxi Province experienced a weather event of low temperatures, rain, snow, and icing. On 13 December, wire icing occurred in Yuanqu County, located in southern Shanxi, and significantly disrupted people"s lives and production activities. Conventional meteorological observations, Doppler weather radar data and the ERA5 reanalysis datasets are used to investigate the formation mechanism of this severe disaster-causing wire icing event on 13 December in 2023. The results show that: this wire icing event in Yuanqu belonging to the form of mixed rime and glaze is caused by the combined effects of short-duration freezing rain in the early stage and prolonged freezing fog in the subsequent period. Freezing fog plays a more significant role than freezing rain. In the north frontal zone at 500 hPa, the horizontal trough in front of the high pressure ridge near the Ural Mountains and a deep cold vortex near the Sea of Okhotsk both remain stable and move slowly. The short-wave trough ahead of the bottom of the horizontal trough moves rapidly eastward and affects Shanxi. In the south frontal zone, the southwest jet stream at 700 hPa is strong. The cold temperature trough in the northeast-southwest direction at 850 hPa is located over North China. The surface inverted trough in Hetao area develops strongly and interacts with the return-flow weather pattern. The southwest warm and humid airflow ascends along the low-level cold air cushion, providing a favorable large-scale circulation background for the formation of freezing rain and freezing fog in Yuanqu. In the vertical direction, warm temperature advection above and cold temperature advection at bottom lead to a cold-warm-cold temperature structure in Yuanqu from low to high altitudes, with the middle-layer temperature above 0℃ and the lower-layer temperature below 0℃. The short-term freezing rain in Yuanqu, triggered by the short-wave trough, is attributed to the melting mechanism. The vertical humidity advection configuration of "wet above and dry at bottom", coupled with sinking motion together make the lower atmosphere highly saturated. Additionally, temperature inversion continues in the lower atmosphere. These conditions collectively facilitate the long-term persistence of freezing fog in Yuanqu. Abundant supercooled water droplets, undergoing two phases of rapid growth, continuously impinge on the wire surfaces, causing ice accretion to thicken progressively. Southeasterly airflow ascends along the windward slopes of the horn-shaped topography, which plays a certain role in promoting the thickening of accretion on wires.
In mid-December 2023, Shanxi Province experienced a weather event of low temperatures, rain, snow, and icing. On 13 December, wire icing occurred in Yuanqu County, located in southern Shanxi, and significantly disrupted people"s lives and production activities. Conventional meteorological observations, Doppler weather radar data and the ERA5 reanalysis datasets are used to investigate the formation mechanism of this severe disaster-causing wire icing event on 13 December in 2023. The results show that: this wire icing event in Yuanqu belonging to the form of mixed rime and glaze is caused by the combined effects of short-duration freezing rain in the early stage and prolonged freezing fog in the subsequent period. Freezing fog plays a more significant role than freezing rain. In the north frontal zone at 500 hPa, the horizontal trough in front of the high pressure ridge near the Ural Mountains and a deep cold vortex near the Sea of Okhotsk both remain stable and move slowly. The short-wave trough ahead of the bottom of the horizontal trough moves rapidly eastward and affects Shanxi. In the south frontal zone, the southwest jet stream at 700 hPa is strong. The cold temperature trough in the northeast-southwest direction at 850 hPa is located over North China. The surface inverted trough in Hetao area develops strongly and interacts with the return-flow weather pattern. The southwest warm and humid airflow ascends along the low-level cold air cushion, providing a favorable large-scale circulation background for the formation of freezing rain and freezing fog in Yuanqu. In the vertical direction, warm temperature advection above and cold temperature advection at bottom lead to a cold-warm-cold temperature structure in Yuanqu from low to high altitudes, with the middle-layer temperature above 0℃ and the lower-layer temperature below 0℃. The short-term freezing rain in Yuanqu, triggered by the short-wave trough, is attributed to the melting mechanism. The vertical humidity advection configuration of "wet above and dry at bottom", coupled with sinking motion together make the lower atmosphere highly saturated. Additionally, temperature inversion continues in the lower atmosphere. These conditions collectively facilitate the long-term persistence of freezing fog in Yuanqu. Abundant supercooled water droplets, undergoing two phases of rapid growth, continuously impinge on the wire surfaces, causing ice accretion to thicken progressively. Southeasterly airflow ascends along the windward slopes of the horn-shaped topography, which plays a certain role in promoting the thickening of accretion on wires.
Available online: April 11, 2026 , DOI: 10.7519/j.issn.1000-0526.2026.030903
Abstract:
Aiming at the challenge of fine-scale forecasting of 2-meter temperature (T2m) in complex terrain areas, this study selects the Guangxi region, a typical area with complex terrain, as the research object and proposes a physics-constrained deep learning forecasting model named PSD-Net, which integrates the forecast products of the Pangu-Weather model. The forecast products of Pangu-Weather (PANGU) are used as feature variables input.The generator based on the Super-Resolution Generative Adversarial Network (SRGAN) is employed to extract multi-scale features.Power-spectral-density (PSD) and Kullback–Leibler (KL) divergence terms are explicitly injected into the loss function as regularizers, enforcing strict spectral fidelity and distributional congruence between the forecast and the observed state.Comparing the T2m forecasting performance of the ECMWF, SCMOC and PANGU products in the Guangxi region in 2024, the results show that PSD-Net outperforms the comparative products at both grid and station forecasting. The grid absolute mean error is reduced by 37.6 % compared with PANGU, and the accuracy is improved by 17 percentage points. The error growth is gentle from 25 to 72 hours. This study verifies the effectiveness of the physics-constrained deep learning framework in fine-scale T2m forecasting and provides a new approach for the combination of meteorological and AI models.
Aiming at the challenge of fine-scale forecasting of 2-meter temperature (T2m) in complex terrain areas, this study selects the Guangxi region, a typical area with complex terrain, as the research object and proposes a physics-constrained deep learning forecasting model named PSD-Net, which integrates the forecast products of the Pangu-Weather model. The forecast products of Pangu-Weather (PANGU) are used as feature variables input.The generator based on the Super-Resolution Generative Adversarial Network (SRGAN) is employed to extract multi-scale features.Power-spectral-density (PSD) and Kullback–Leibler (KL) divergence terms are explicitly injected into the loss function as regularizers, enforcing strict spectral fidelity and distributional congruence between the forecast and the observed state.Comparing the T2m forecasting performance of the ECMWF, SCMOC and PANGU products in the Guangxi region in 2024, the results show that PSD-Net outperforms the comparative products at both grid and station forecasting. The grid absolute mean error is reduced by 37.6 % compared with PANGU, and the accuracy is improved by 17 percentage points. The error growth is gentle from 25 to 72 hours. This study verifies the effectiveness of the physics-constrained deep learning framework in fine-scale T2m forecasting and provides a new approach for the combination of meteorological and AI models.
Available online: April 07, 2026 , DOI: 10.7519/j.issn.1000-0526.2026.040202
Abstract:
This study classified the weather situations of heavy precipitation events in the Yangtze River Delta region from September 2022 to September 2024 and evaluated the forecast performance of the numerical models CMA-MESO, CMA-GFS, CMA-TYM, CMA-SH9, and ECMWF under four main weather patterns. The research results show that: 1) In the 24-hour precipitation forecasts, light rain tends to have a high false alarm rate, while heavy rain and above are difficult to accurately predict. In the 3-hour forecasts, the CMA-MESO model performs best for light precipitation. However, under cold shear and low vortex shear patterns, heavy precipitation becomes harder to capture, and the forecast performance weakens.2) In terms of the evaluation of temporal characteristics, under each weather pattern, the models have the highest accuracy in forecasting the start time of precipitation, followed by the end time, and the accuracy in forecasting the peak time is relatively low. 3) In terms of spatial feature evaluation, in the north-south direction, most models exhibit a northern systematic error in the low vortex shear and subtropical high combined with low trough types, while the errors are generally southern in the typhoon body and peripheral weather types. In the east-west direction, models generally exhibit an eastern bias in the typhoon body and peripheral weather types, while in other weather types, most models show a western systematic error.
This study classified the weather situations of heavy precipitation events in the Yangtze River Delta region from September 2022 to September 2024 and evaluated the forecast performance of the numerical models CMA-MESO, CMA-GFS, CMA-TYM, CMA-SH9, and ECMWF under four main weather patterns. The research results show that: 1) In the 24-hour precipitation forecasts, light rain tends to have a high false alarm rate, while heavy rain and above are difficult to accurately predict. In the 3-hour forecasts, the CMA-MESO model performs best for light precipitation. However, under cold shear and low vortex shear patterns, heavy precipitation becomes harder to capture, and the forecast performance weakens.2) In terms of the evaluation of temporal characteristics, under each weather pattern, the models have the highest accuracy in forecasting the start time of precipitation, followed by the end time, and the accuracy in forecasting the peak time is relatively low. 3) In terms of spatial feature evaluation, in the north-south direction, most models exhibit a northern systematic error in the low vortex shear and subtropical high combined with low trough types, while the errors are generally southern in the typhoon body and peripheral weather types. In the east-west direction, models generally exhibit an eastern bias in the typhoon body and peripheral weather types, while in other weather types, most models show a western systematic error.
Available online: April 03, 2026 , DOI: 10.7519/j.issn.1000-0526.2026.040201
Abstract:
The persistent drought event cross the winter and spring in Yunnan (YN-WSPDE) have a significant impact on the planting of major autumn crops and the safety of people"s life and property, and it is of great scientific significance for meteorological disaster prevention and reduction to investigate the change and cause of this events. As a fundamental part of the research on YN-WSPDE, how to define the YN-WSPDE standard is studied based on daily precipitation and temperature observation data of 125 stations in Yunnan from 1961 to 2024, as well as the NCEP/NCAR daily reanalysis atmospheric circulation data during the same period. By analyzing the spatiotemporal characteristics of drought in winter and spring in Yunnan, it was found that the drought distribution in Yunnan is characterized by less in the east and more in the west, and more in the north and less in the south. The regions with the highest drought occurrence are mainly concentrated in the eastern part of northwest Yunnan, including Lijiang, Dali, and northern Chuxiong, while the southern parts of Honghe and Xishuangbanna have less drought occurrence. The WSPDEs standard for single station and provincial region in Yunnan are defined as: the station standard refers to the occurrence of light or above level drought in both winter and spring in a year, with moderate or above level drought occurring in one of the seasons. When 25 stations (1/5 of the total number of stations in the Yunnan) experience the WSPDE in a year, and the average precipitation anomaly percentage in both winter and spring is ≤-10%, it is defined as a WSPDE in provincial area. Preliminary analysis results show that the continuously anomalous geopotential height difference at 500-hPa between the north and south in the Eastern of East Asia, as well as anomalous anticyclone at 700-hPa over the Bay of Bengal basin, have a significant impact on the YN-WSPDE.
The persistent drought event cross the winter and spring in Yunnan (YN-WSPDE) have a significant impact on the planting of major autumn crops and the safety of people"s life and property, and it is of great scientific significance for meteorological disaster prevention and reduction to investigate the change and cause of this events. As a fundamental part of the research on YN-WSPDE, how to define the YN-WSPDE standard is studied based on daily precipitation and temperature observation data of 125 stations in Yunnan from 1961 to 2024, as well as the NCEP/NCAR daily reanalysis atmospheric circulation data during the same period. By analyzing the spatiotemporal characteristics of drought in winter and spring in Yunnan, it was found that the drought distribution in Yunnan is characterized by less in the east and more in the west, and more in the north and less in the south. The regions with the highest drought occurrence are mainly concentrated in the eastern part of northwest Yunnan, including Lijiang, Dali, and northern Chuxiong, while the southern parts of Honghe and Xishuangbanna have less drought occurrence. The WSPDEs standard for single station and provincial region in Yunnan are defined as: the station standard refers to the occurrence of light or above level drought in both winter and spring in a year, with moderate or above level drought occurring in one of the seasons. When 25 stations (1/5 of the total number of stations in the Yunnan) experience the WSPDE in a year, and the average precipitation anomaly percentage in both winter and spring is ≤-10%, it is defined as a WSPDE in provincial area. Preliminary analysis results show that the continuously anomalous geopotential height difference at 500-hPa between the north and south in the Eastern of East Asia, as well as anomalous anticyclone at 700-hPa over the Bay of Bengal basin, have a significant impact on the YN-WSPDE.
Available online: April 02, 2026 , DOI: 10.7519/j.issn.1000-0526.2026.031101
Abstract:
Based on multi-source observations and ERA5 reanalysis data, this study systematically analyzed an extreme rainfall event in North China in late July 2025. It revealed the diversity in the initiation, organization, and propagation mechanisms of Mesoscale Convective Systems (MCSs) during three extreme rainfall days (July 24, 25, and 27) under a similar large-scale warm sector background, and investigated their synoptic causes.The results indicate that: (1) The event was characterized by significant extremity with high rainfall intensity and large accumulation. The multiple extreme precipitation centers within the southern warm-sector rainband posed the main forecast challenge. (2) The MCS evolution varied significantly by day. On July 24, convection initiated at the foothills, and the extreme rainfall was dominated by a "training effect" from backward-propagating and band-shaped convection, resulting in the most widespread rainfall area spanning mountains, foothills, and plains. On July 25, convection originated in the mountains, and the extreme rainfall was associated with a quasi-stationary system, leading to the most localized precipitation. On July 27, convection mostly initiated in the plains and intensified after moving to the foothills, with extreme rainfall linked to a "training effect" from line-shaped convection.(3) The primary cause for this diversity lies in the varied interactions between the environmental field and complex terrain. On July 24, the strongest synoptic-scale forcing, characterized by a strong and deep Low-Level Jet (LLJ) coupled with orographic lift and sustained moisture transport, played a key role in convective initiation and back-propagation. On July 25, under the control of the subtropical high, a weaker and shallower LLJ limited downstream propagation, making orographic lift the dominant mechanism for triggering and enhancement, which was accompanied by a mesocyclone-like structure that produced the strongest local hourly rainfall. On July 27, under the weakest synoptic-scale forcing, convection was triggered by boundary layer easterlies and a weak low-level convergence line, while a mesoscale front and orographic lifting of southwesterly flow facilitated its intensification and maintenance.
Based on multi-source observations and ERA5 reanalysis data, this study systematically analyzed an extreme rainfall event in North China in late July 2025. It revealed the diversity in the initiation, organization, and propagation mechanisms of Mesoscale Convective Systems (MCSs) during three extreme rainfall days (July 24, 25, and 27) under a similar large-scale warm sector background, and investigated their synoptic causes.The results indicate that: (1) The event was characterized by significant extremity with high rainfall intensity and large accumulation. The multiple extreme precipitation centers within the southern warm-sector rainband posed the main forecast challenge. (2) The MCS evolution varied significantly by day. On July 24, convection initiated at the foothills, and the extreme rainfall was dominated by a "training effect" from backward-propagating and band-shaped convection, resulting in the most widespread rainfall area spanning mountains, foothills, and plains. On July 25, convection originated in the mountains, and the extreme rainfall was associated with a quasi-stationary system, leading to the most localized precipitation. On July 27, convection mostly initiated in the plains and intensified after moving to the foothills, with extreme rainfall linked to a "training effect" from line-shaped convection.(3) The primary cause for this diversity lies in the varied interactions between the environmental field and complex terrain. On July 24, the strongest synoptic-scale forcing, characterized by a strong and deep Low-Level Jet (LLJ) coupled with orographic lift and sustained moisture transport, played a key role in convective initiation and back-propagation. On July 25, under the control of the subtropical high, a weaker and shallower LLJ limited downstream propagation, making orographic lift the dominant mechanism for triggering and enhancement, which was accompanied by a mesocyclone-like structure that produced the strongest local hourly rainfall. On July 27, under the weakest synoptic-scale forcing, convection was triggered by boundary layer easterlies and a weak low-level convergence line, while a mesoscale front and orographic lifting of southwesterly flow facilitated its intensification and maintenance.
LIU Junjie, Lu Bo, Li Hao, Chen Lei, Zhong Xiao Hui, Zhou Chen Guang, Hu jia hui, wu jie, zhao chun yan, xin yu hang, zhao yang, Qian QiFeng
Available online: April 01, 2026 , DOI: 10.7519/j.issn.1000-0526.2026.032901
Abstract:
Using hindcast datasets from the Fengshun model and the S2S-ECMWF model during 2017–2022, together with CMA-RA and NCEP reanalyses and station temperature observations, the pentad-scale prediction skill for midsummer (July–August) temperature over China is evaluated. Three metrics are adopted: temporal anomaly correlation coefficient (TCC), spatial anomaly correlation coefficient (ACC), and integrated prediction score (IPS). The verification results based on different reference datasets (station observations, CMA-RA1.0, and NCEP) are generally consistent. Overall, the Fengshun model shows higher prediction skill than the S2S-ECMWF model, with TCC, ACC, and IPS values improved by 7.9%, 18.4%, and 1.5%, respectively. Higher TCC skill is found in the Huang-Huai, Jiang-Huai, Central China, South China, East China, and Xinjiang regions, while lower skill appears in Northeast China, Inner Mongolia, the Tibetan Plateau, and Southwest China. The Fengshun model performs better at 1-pentad and 4–8-pentad leads, with the highest skill at a 6-pentad lead within the subseasonal range, mainly due to its improved prediction of 500 hPa geopotential height anomalies over key circulation regions..
Using hindcast datasets from the Fengshun model and the S2S-ECMWF model during 2017–2022, together with CMA-RA and NCEP reanalyses and station temperature observations, the pentad-scale prediction skill for midsummer (July–August) temperature over China is evaluated. Three metrics are adopted: temporal anomaly correlation coefficient (TCC), spatial anomaly correlation coefficient (ACC), and integrated prediction score (IPS). The verification results based on different reference datasets (station observations, CMA-RA1.0, and NCEP) are generally consistent. Overall, the Fengshun model shows higher prediction skill than the S2S-ECMWF model, with TCC, ACC, and IPS values improved by 7.9%, 18.4%, and 1.5%, respectively. Higher TCC skill is found in the Huang-Huai, Jiang-Huai, Central China, South China, East China, and Xinjiang regions, while lower skill appears in Northeast China, Inner Mongolia, the Tibetan Plateau, and Southwest China. The Fengshun model performs better at 1-pentad and 4–8-pentad leads, with the highest skill at a 6-pentad lead within the subseasonal range, mainly due to its improved prediction of 500 hPa geopotential height anomalies over key circulation regions..
Available online: March 27, 2026 , DOI: 10.7519/j.issn.1000-0526.2025.080101
Abstract:
In this paper, the Spectrum GAN (SGAN) and the Multiscale GAN (MSG) are proposed by optimizing the Generative Adversarial Networks (GAN). Both of them are then combined with PhyDNet to form two deep learning based nowcasting methods for heavy precipitation in the Jianghuai region, namely PhySGAN and PhyMSG, which can provide precipitation forecasts for the next 3 hours. Through the forecast skill score, the performance evaluation in "complex scenarios" and the analysis of typical application examples, the performance of the two methods in the short-term heavy precipitation (precipitation ≥ 20 mm/h) forecast in Jiangsu during the flood season in 2024 is analyzed, and the following conclusions are obtained: (1) The TS of short-term heavy precipitation within 3 hours in different verification periods of PhySGAN and PhyMSG are significantly improved compared with those of the basic experiment PhyDNet and the general GAN experiment PhyGAN. The two new methods correct the low frequency problem of short-term heavy precipitation forecasts of PhyDNet and PhyGAN, so that the TS increases with the increase of the forecast lead time, thereby effectively extending the nowcasting lead time of short-term heavy precipitation.(2) Judging from the forecast performance shown by each method in "complex scenarios", deep learning can reflect the evolution of the generation and dissipation of heavy precipitation compared with traditional extrapolation methods. PhySGAN and PhyMSG show better forecast performance than PhyDNet and PhyGAN. The former has a better ability to depict local details such as the shape and intensity of heavy precipitation, while the latter has a better representation of the overall contour and position of the heavy precipitation rain band. (3) Combined with the application of typical heavy precipitation cases during the flood season, it is found that both PhySGAN and PhyMSG can forecast the precipitation enhancement process in advance in both systematic heavy precipitation and local heavy precipitation cases, which effectively guides the early warning before disasters. In addition, PhyMSG has a certain indicative effect on extreme rainfall intensities above 50 mm/h, while PhySGAN can better reflect the changes in the shape and position of the rain band.
In this paper, the Spectrum GAN (SGAN) and the Multiscale GAN (MSG) are proposed by optimizing the Generative Adversarial Networks (GAN). Both of them are then combined with PhyDNet to form two deep learning based nowcasting methods for heavy precipitation in the Jianghuai region, namely PhySGAN and PhyMSG, which can provide precipitation forecasts for the next 3 hours. Through the forecast skill score, the performance evaluation in "complex scenarios" and the analysis of typical application examples, the performance of the two methods in the short-term heavy precipitation (precipitation ≥ 20 mm/h) forecast in Jiangsu during the flood season in 2024 is analyzed, and the following conclusions are obtained: (1) The TS of short-term heavy precipitation within 3 hours in different verification periods of PhySGAN and PhyMSG are significantly improved compared with those of the basic experiment PhyDNet and the general GAN experiment PhyGAN. The two new methods correct the low frequency problem of short-term heavy precipitation forecasts of PhyDNet and PhyGAN, so that the TS increases with the increase of the forecast lead time, thereby effectively extending the nowcasting lead time of short-term heavy precipitation.(2) Judging from the forecast performance shown by each method in "complex scenarios", deep learning can reflect the evolution of the generation and dissipation of heavy precipitation compared with traditional extrapolation methods. PhySGAN and PhyMSG show better forecast performance than PhyDNet and PhyGAN. The former has a better ability to depict local details such as the shape and intensity of heavy precipitation, while the latter has a better representation of the overall contour and position of the heavy precipitation rain band. (3) Combined with the application of typical heavy precipitation cases during the flood season, it is found that both PhySGAN and PhyMSG can forecast the precipitation enhancement process in advance in both systematic heavy precipitation and local heavy precipitation cases, which effectively guides the early warning before disasters. In addition, PhyMSG has a certain indicative effect on extreme rainfall intensities above 50 mm/h, while PhySGAN can better reflect the changes in the shape and position of the rain band.
Available online: March 10, 2026 , DOI: 10.7519/j.issn.1000-0526.2026.011201
Abstract:
Abstract:On 17th (hereinafter referred to as “1217”)and from 24th to 26th December 2021 (hereinafter referred to as “1225”), Shandong Peninsula experienced two Bohai sea-effect snowfall events.Compared with “1217”,the cold air intensity was stronger and snowfall duration was longer for “1225”, but the total snowfall amount was less and intensity of snowfall was weaker than “1217”. In order to clarify the reasons, thermodynamic and dynamical characters were analyzed and compared by using data of moored buoy station,S-band Doppler weather radar,conventional sounding, hourly snowfall observation data from ground automatic meteorological stations and ERA5( ECMWF Reanalysis v5). The main results are as follows:During the first snowfall phase of "1225", sea-air temperature difference(temperature difference between sea surface and 850hPa air) was 28°C, which exceeded "1217" (24°C). During the second snowfall phase, both events exhibited comparable sea-air temperature between 24°C and 25°C. However, throughout the snowfall period,"1225" displayed an?intermittent weakening of low-level cold advection, with?advection intensities consistently weaker than "1217". This weaker thermal advection?impaired the development of?lower-level thermal and moisture conditions?for the "1225". During "1217", the intense core of cold advection extended vertically to 700～500 hPa, whereas during "1225" it was confined to 875～750 hPa and was weaker than"1217". Consequently, the inversion layer developed above the cold advection center in "1225" displayed both lower height and reduced thickness relative to "1217". This structural difference further suppressed depth of the conditionally unstable (or neutral) layer beneath the inversion,thereby limiting the vertical extent of shallow convective clouds and ultimately diminishing snowfall intensity. During the second snowfall phase of "1225", steering wind direction (mean wind direction from 1000 hPa to 700 hPa) over the Bohai Sea and its upstream regions was 327°, deviating by 11° from "1217" with a larger westerly component. Consequently, the low-level tongue, mesoscale shear line and heavy snow echo band all displayed larger zonal component and positioned predominantly offshore resulting in inconspicuous snowfall over the peninsula ultimately. Cold advection constitutes a primary physical driver modulating the thermal structure of sea-effect snowfall. Simultaneously, it indirectly governs the initiation and evolution of mesoscale shear lines by inducing a low-level warm tongue, thereby providing dynamical lifting and organizational mechanism over snowfall echo structures,ultimately shaping the dynamical structure of snowfall. For forecasting applications, it is essential to make an integrated dynamic analysis of cold advection developmental stages, three-dimensional spatial structure and intensity evolution on low-level. Focus on sea-air temperature difference,absolute moisture content within the boundary layer, inversion layer height, thickness of the conditionally(or neutral) unstablelayer and steering wind direction. Diagnostics of mesoscale characteristics such as the position, orientation and intensity of low-level warm tongue and shear line are imperative.
Abstract:On 17th (hereinafter referred to as “1217”)and from 24th to 26th December 2021 (hereinafter referred to as “1225”), Shandong Peninsula experienced two Bohai sea-effect snowfall events.Compared with “1217”,the cold air intensity was stronger and snowfall duration was longer for “1225”, but the total snowfall amount was less and intensity of snowfall was weaker than “1217”. In order to clarify the reasons, thermodynamic and dynamical characters were analyzed and compared by using data of moored buoy station,S-band Doppler weather radar,conventional sounding, hourly snowfall observation data from ground automatic meteorological stations and ERA5( ECMWF Reanalysis v5). The main results are as follows:During the first snowfall phase of "1225", sea-air temperature difference(temperature difference between sea surface and 850hPa air) was 28°C, which exceeded "1217" (24°C). During the second snowfall phase, both events exhibited comparable sea-air temperature between 24°C and 25°C. However, throughout the snowfall period,"1225" displayed an?intermittent weakening of low-level cold advection, with?advection intensities consistently weaker than "1217". This weaker thermal advection?impaired the development of?lower-level thermal and moisture conditions?for the "1225". During "1217", the intense core of cold advection extended vertically to 700～500 hPa, whereas during "1225" it was confined to 875～750 hPa and was weaker than"1217". Consequently, the inversion layer developed above the cold advection center in "1225" displayed both lower height and reduced thickness relative to "1217". This structural difference further suppressed depth of the conditionally unstable (or neutral) layer beneath the inversion,thereby limiting the vertical extent of shallow convective clouds and ultimately diminishing snowfall intensity. During the second snowfall phase of "1225", steering wind direction (mean wind direction from 1000 hPa to 700 hPa) over the Bohai Sea and its upstream regions was 327°, deviating by 11° from "1217" with a larger westerly component. Consequently, the low-level tongue, mesoscale shear line and heavy snow echo band all displayed larger zonal component and positioned predominantly offshore resulting in inconspicuous snowfall over the peninsula ultimately. Cold advection constitutes a primary physical driver modulating the thermal structure of sea-effect snowfall. Simultaneously, it indirectly governs the initiation and evolution of mesoscale shear lines by inducing a low-level warm tongue, thereby providing dynamical lifting and organizational mechanism over snowfall echo structures,ultimately shaping the dynamical structure of snowfall. For forecasting applications, it is essential to make an integrated dynamic analysis of cold advection developmental stages, three-dimensional spatial structure and intensity evolution on low-level. Focus on sea-air temperature difference,absolute moisture content within the boundary layer, inversion layer height, thickness of the conditionally(or neutral) unstablelayer and steering wind direction. Diagnostics of mesoscale characteristics such as the position, orientation and intensity of low-level warm tongue and shear line are imperative.
Available online: March 09, 2026 , DOI: 10.7519/j.issn.1000-0526.2026.030901
Abstract:
Based on observations from 2374 meteorological stations in China and the NCEP/NCAR reanalysis dataset, the spatiotemporal characteristics and causes of climate anomalies over China during autumn 2025 have been analyzed. The autumn of 2025 was generally warmer than normal across China, with the northern region exhibiting an intraseasonal “warm-cold-warm” temperature variation and the southern region showing a “warm in the early stage and cold in the late stage” pattern. The national average precipitation was the highest since 1961 for the same period, with an uneven spatiotemporal distribution. In the early autumn, large-scale excessive precipitation with prominent extremeness occurred in southern North China, northern East China, northern Central China, and eastern Northwest China, leading to severe autumn flood. The autumn rain in West China started early, ended late, had a long duration, and recorded the highest rainfall in history. The formation of autumn flood in northern China was closely related to East Asian atmospheric circulation anomalies. The abnormally strong Western North Pacific subtropical high (WPSH) with an extremely northward ridge line, coupled with a robust low-level anticyclone over the Yellow Sea, provided favorable water vapor transport conditions, serving as the basic circulation background for the autumn flood in northern China. Additionally, the extreme meridional stability of the WPSH (i.e., little north-south movement) was another key circulation feature contributing to the flood. Furthermore, during the autumn, the equatorial central-eastern Pacific was in a developmental stage from cold water conditions to La Ni?a, accompanied by an extreme negative phase of the tropical Indian Ocean dipole (TIOD). Both factors, through pathways such as atmospheric teleconnections and local meridional-zonal circulation coupling, worked synergistically to influence the anomalous configuration of the East Asian circulation, serving as an important oceanic external forcing background for the autumn flood in northern China.
Based on observations from 2374 meteorological stations in China and the NCEP/NCAR reanalysis dataset, the spatiotemporal characteristics and causes of climate anomalies over China during autumn 2025 have been analyzed. The autumn of 2025 was generally warmer than normal across China, with the northern region exhibiting an intraseasonal “warm-cold-warm” temperature variation and the southern region showing a “warm in the early stage and cold in the late stage” pattern. The national average precipitation was the highest since 1961 for the same period, with an uneven spatiotemporal distribution. In the early autumn, large-scale excessive precipitation with prominent extremeness occurred in southern North China, northern East China, northern Central China, and eastern Northwest China, leading to severe autumn flood. The autumn rain in West China started early, ended late, had a long duration, and recorded the highest rainfall in history. The formation of autumn flood in northern China was closely related to East Asian atmospheric circulation anomalies. The abnormally strong Western North Pacific subtropical high (WPSH) with an extremely northward ridge line, coupled with a robust low-level anticyclone over the Yellow Sea, provided favorable water vapor transport conditions, serving as the basic circulation background for the autumn flood in northern China. Additionally, the extreme meridional stability of the WPSH (i.e., little north-south movement) was another key circulation feature contributing to the flood. Furthermore, during the autumn, the equatorial central-eastern Pacific was in a developmental stage from cold water conditions to La Ni?a, accompanied by an extreme negative phase of the tropical Indian Ocean dipole (TIOD). Both factors, through pathways such as atmospheric teleconnections and local meridional-zonal circulation coupling, worked synergistically to influence the anomalous configuration of the East Asian circulation, serving as an important oceanic external forcing background for the autumn flood in northern China.
Available online: March 06, 2026 , DOI: 10.7519/j.issn.1000-0526.2025.082002
Abstract:
The surface wind speed prediction is the diagnostic value at the height of 10 meters above the terrain of the numerical model, so the accuracy of the terrain has a great influence on the wind speed prediction performance. Therefore, SRTM data with high resolution is used to construct high-precision terrain, based on the high-precision terrain data, the wind speed prediction scheme is optimized. The numerical simulation test verify the 24-hour wind speed forecast results using more than 10,000 hourly-resolution automatic weather station observations on November 20 and July 20, 2023, the test results shows the mean deviation of wind speed forecast is reduced by 10.0% and 10.9% after using the optimized scheme based on high-precision terrain data, and the root mean square error of forecast is reduced too. The results of two months of continuous experiments show that the average deviation of the 24-hour wind speed forecast can be reduced by 50.0% and 50.6% with the optimized scheme for July and November 2023. It is shown that the optimized wind speed prediction and diagnosis scheme can produce obvious positive effect.
The surface wind speed prediction is the diagnostic value at the height of 10 meters above the terrain of the numerical model, so the accuracy of the terrain has a great influence on the wind speed prediction performance. Therefore, SRTM data with high resolution is used to construct high-precision terrain, based on the high-precision terrain data, the wind speed prediction scheme is optimized. The numerical simulation test verify the 24-hour wind speed forecast results using more than 10,000 hourly-resolution automatic weather station observations on November 20 and July 20, 2023, the test results shows the mean deviation of wind speed forecast is reduced by 10.0% and 10.9% after using the optimized scheme based on high-precision terrain data, and the root mean square error of forecast is reduced too. The results of two months of continuous experiments show that the average deviation of the 24-hour wind speed forecast can be reduced by 50.0% and 50.6% with the optimized scheme for July and November 2023. It is shown that the optimized wind speed prediction and diagnosis scheme can produce obvious positive effect.
Available online: March 05, 2026 , DOI: 10.7519/j.issn.1000-0526.2026.022701
Abstract:
To enhance the forecast accuracy of average wind speed and direction, this study utilized the 2-minute average wind speed and direction data from ground observations provided by the China Meteorological Administration (CMA) from January 2021 to December 2022, in conjunction with the 24-hour forecasted 10-meter wind field data from the deterministic model of the European Centre for Medium-Range Weather Forecasts (ECMWF). A two-step bias correction method tailored for multi-site wind speed and direction forecasts was developed for 662 national observation stations located within the range of 110°E-120°E and 30°N-40°N. Given that wind direction is a circular variable, traditional statistical post-processing methods are challenging to directly apply for wind direction correction. Therefore, this study adopted the u and v wind components, which simultaneously contain wind speed and direction information, as key variables. Considering the significant differences in u and v wind forecast errors among stations, a site-specific modeling strategy was employed to enhance the relevance and precision of the model. Although the joint distribution of u and v winds approximates a two-dimensional Gaussian distribution, and joint modeling could potentially improve wind speed and direction forecasts, it may also intertwine wind speed and direction forecast errors, increasing model instability. Consequently, this study constructed separate univariate regression correction models for the u and v winds at each station. After correcting the u and v winds, the wind direction forecast errors decreased at most stations. However, when the numerical model forecasted excessively small u and v winds, a sign reversal occurred after correction, primarily in cases of low forecasted wind speeds. Regarding wind speed forecasts, despite the u and v wind corrections, deficiencies still existed in forecasting stronger winds. To address this, this study utilized the quantile matching method to further correct the wind speeds synthesized from the corrected u and v winds, referred to as the two-step bias correction method. To further validate the effectiveness of the two-step bias correction method in practical operational applications, an independent sample test was conducted using data from January to December 2023. In the comparison of wind direction forecasts, the two-step bias correction method demonstrated good generalization ability and forecast performance. Among various wind speed forecast comparisons, this method exhibited the smallest mean error and root mean square error (RMSE) for wind speeds of 10 m/s and above. Additionally, in cases of strong winds such as those associated with cold fronts and typhoons, the method also showed significant forecast advantages. For forecasts of all wind speeds and those specifically at 10 m/s and above, the RMSE after two-step bias correction was reduced by 18.6% and 29.6%, respectively, compared to the uncorrected numerical model.
To enhance the forecast accuracy of average wind speed and direction, this study utilized the 2-minute average wind speed and direction data from ground observations provided by the China Meteorological Administration (CMA) from January 2021 to December 2022, in conjunction with the 24-hour forecasted 10-meter wind field data from the deterministic model of the European Centre for Medium-Range Weather Forecasts (ECMWF). A two-step bias correction method tailored for multi-site wind speed and direction forecasts was developed for 662 national observation stations located within the range of 110°E-120°E and 30°N-40°N. Given that wind direction is a circular variable, traditional statistical post-processing methods are challenging to directly apply for wind direction correction. Therefore, this study adopted the u and v wind components, which simultaneously contain wind speed and direction information, as key variables. Considering the significant differences in u and v wind forecast errors among stations, a site-specific modeling strategy was employed to enhance the relevance and precision of the model. Although the joint distribution of u and v winds approximates a two-dimensional Gaussian distribution, and joint modeling could potentially improve wind speed and direction forecasts, it may also intertwine wind speed and direction forecast errors, increasing model instability. Consequently, this study constructed separate univariate regression correction models for the u and v winds at each station. After correcting the u and v winds, the wind direction forecast errors decreased at most stations. However, when the numerical model forecasted excessively small u and v winds, a sign reversal occurred after correction, primarily in cases of low forecasted wind speeds. Regarding wind speed forecasts, despite the u and v wind corrections, deficiencies still existed in forecasting stronger winds. To address this, this study utilized the quantile matching method to further correct the wind speeds synthesized from the corrected u and v winds, referred to as the two-step bias correction method. To further validate the effectiveness of the two-step bias correction method in practical operational applications, an independent sample test was conducted using data from January to December 2023. In the comparison of wind direction forecasts, the two-step bias correction method demonstrated good generalization ability and forecast performance. Among various wind speed forecast comparisons, this method exhibited the smallest mean error and root mean square error (RMSE) for wind speeds of 10 m/s and above. Additionally, in cases of strong winds such as those associated with cold fronts and typhoons, the method also showed significant forecast advantages. For forecasts of all wind speeds and those specifically at 10 m/s and above, the RMSE after two-step bias correction was reduced by 18.6% and 29.6%, respectively, compared to the uncorrected numerical model.
Available online: March 03, 2026 , DOI: 10.7519/j.issn.1000-0526.2026.022801
Abstract:
The low-frequency oscillation characteristics of continuous rainstorm processes in Guangdong during the first rainy season in 2023 were analyzed by the methods of wavelet analysis and Lanczos temporal filter, focusing on revealing the mean atmospheric circulation field and it’s evolution characteristics in different phases of quasi-biweekly oscillations and the source of the low-frequency signals of the continuous warm-sector rainstorms with strong southwesterlies in the northwest of Guangdong. The results show that the three continuous rainstorm exhibited a quasi-7-18-day periodic oscillation, and they are in the positive phase of two quasi-30-60-day intraseasonal oscillations. The continuous warm-sector rainstorms with strong southwesterlies that occurred in the northwest of Guangdong on 22-26 June, there were stable“west blocking”and“east blocking” at the middle and high latitudes, the northwest of Guangdong was located at the bottom of the plateau trough and the edge of the subtropical high at 500 hPa, and at the left side of southwest wind axis and the right side of the cyclonic circulations at 850 hPa. During the intermittent- start-peak period of the continuous warm-sector rainstorms process with strong southwesterlies , the low-frequency signals in the mid-to-upper levels originate from the southern branch wave train with the eastward propagation and southward extension of baroclinic low-frequency anticyclones and cyclones within or south of the Iranian Plateau-Tibetan Plateau，and the low-frequency signals in the lower levels arise from the eastward movement and southward extension of the Mongolian Plateau low-frequency anticyclone and the progressively intensifying low-frequency cyclone east of the Sichuan Basin and Yunnan-Guizhou Plateau. They inducing the center of South Asian High to gradually move eastward, the western Pacific subtropical high to weaken and retreat eastward, the Qinghai-Tibet Plateau is controlled by low-frequency cyclonic circulation in intermittent periods gradually turning into the low-frequency anticyclone circulation in the peak period，and the North China - South China region is controlled by low-frequency anticyclone circulation in intermittent periods gradually turning into the low-frequency cyclonic circulation in the peak period. They can provide a reference for medium-to-extended range forecasting of continuous warm-sector rainstorm.
The low-frequency oscillation characteristics of continuous rainstorm processes in Guangdong during the first rainy season in 2023 were analyzed by the methods of wavelet analysis and Lanczos temporal filter, focusing on revealing the mean atmospheric circulation field and it’s evolution characteristics in different phases of quasi-biweekly oscillations and the source of the low-frequency signals of the continuous warm-sector rainstorms with strong southwesterlies in the northwest of Guangdong. The results show that the three continuous rainstorm exhibited a quasi-7-18-day periodic oscillation, and they are in the positive phase of two quasi-30-60-day intraseasonal oscillations. The continuous warm-sector rainstorms with strong southwesterlies that occurred in the northwest of Guangdong on 22-26 June, there were stable“west blocking”and“east blocking” at the middle and high latitudes, the northwest of Guangdong was located at the bottom of the plateau trough and the edge of the subtropical high at 500 hPa, and at the left side of southwest wind axis and the right side of the cyclonic circulations at 850 hPa. During the intermittent- start-peak period of the continuous warm-sector rainstorms process with strong southwesterlies , the low-frequency signals in the mid-to-upper levels originate from the southern branch wave train with the eastward propagation and southward extension of baroclinic low-frequency anticyclones and cyclones within or south of the Iranian Plateau-Tibetan Plateau，and the low-frequency signals in the lower levels arise from the eastward movement and southward extension of the Mongolian Plateau low-frequency anticyclone and the progressively intensifying low-frequency cyclone east of the Sichuan Basin and Yunnan-Guizhou Plateau. They inducing the center of South Asian High to gradually move eastward, the western Pacific subtropical high to weaken and retreat eastward, the Qinghai-Tibet Plateau is controlled by low-frequency cyclonic circulation in intermittent periods gradually turning into the low-frequency anticyclone circulation in the peak period，and the North China - South China region is controlled by low-frequency anticyclone circulation in intermittent periods gradually turning into the low-frequency cyclonic circulation in the peak period. They can provide a reference for medium-to-extended range forecasting of continuous warm-sector rainstorm.
Available online: February 25, 2026 , DOI: 10.7519/j.issn.1000-0526.2025.081201
Abstract:
To investigate the impact of meteorological conditions on the key quality components (protein, fat, starch, and amino acids) of maize grains, stepwise regression-based quality prediction models were constructed based on interval sowing tests and tested. The values calculated by the models were converted into quality grades and compared with the actual grades. The results showed that all stepwise regression-based prediction models passed the significance test, establishing a relationship between grain quality and meteorological factors from tasseling to milk stage and from milk stage to maturity. The linear relationship between grain quality and meteorological factors was quantified. The results of the model test and forecasting test indicated that the mean absolute percentage errors for all four quality components were less than 15 % and the predictions for starch and protein were closer to the actual values compared to those for fat and amino acids. The actual and predicted contents of grain quality components (regardless of cropping system) were converted into grades for validation. For protein, fat, and starch, the combined proportion of samples with predicted grades matching or within one grade of actual grades exceeded 90 % (reaching 100 % for starch), and was 76.67 % for amino acids. The predicted grades aligned well with the actual grades, indicating that the prediction models exhibited high accuracy and can be used for forecasting and evaluating grain quality. The findings can offer an objective and quantitative basis for optimizing environmental resource utilization to improve maize quality and for informing maize ecological zoning.
To investigate the impact of meteorological conditions on the key quality components (protein, fat, starch, and amino acids) of maize grains, stepwise regression-based quality prediction models were constructed based on interval sowing tests and tested. The values calculated by the models were converted into quality grades and compared with the actual grades. The results showed that all stepwise regression-based prediction models passed the significance test, establishing a relationship between grain quality and meteorological factors from tasseling to milk stage and from milk stage to maturity. The linear relationship between grain quality and meteorological factors was quantified. The results of the model test and forecasting test indicated that the mean absolute percentage errors for all four quality components were less than 15 % and the predictions for starch and protein were closer to the actual values compared to those for fat and amino acids. The actual and predicted contents of grain quality components (regardless of cropping system) were converted into grades for validation. For protein, fat, and starch, the combined proportion of samples with predicted grades matching or within one grade of actual grades exceeded 90 % (reaching 100 % for starch), and was 76.67 % for amino acids. The predicted grades aligned well with the actual grades, indicating that the prediction models exhibited high accuracy and can be used for forecasting and evaluating grain quality. The findings can offer an objective and quantitative basis for optimizing environmental resource utilization to improve maize quality and for informing maize ecological zoning.
Available online: February 12, 2026 , DOI: 10.7519/j.issn.1000-0526.2025.083001
Abstract:
Using dynamic synthesis analysis methods, a comparative analysis was conducted on the reasons for heavy rainfall induced by tropical cyclones entering Jiangsu, both with and without the influence of cold air. A meteorological model of heavy rainfall associated with tropical cyclones in Jiangsu was established, and the following conclusions were obtained: (1) Regardless of the presence of cold air influence, tropical cyclones entering Jiangsu can induce heavy rainfall. The key reasons include the evolution of tropical cyclone"s asymmetric structure, the continuous water vapor transport by low-level jet streams, and the sustained strong upward motion near the tropical cyclone. The differences primarily involve the maintenance mechanisms of dynamic lifting and the growth way of atmospheric instability. (2) When cold air is present, baroclinicity and conditional instability within the outer circulation of tropical cyclones intensify rapidly. Slantwise ascent in the baroclinic zone promotes the sustained lifting of warm-moist air and enhances convective development, with heavy rainfall primarily located on the northern to northeastern sides of tropical cyclones. When cold air is absent, continuous transport of warm-moist air by low-level and ultra-low-level jets significantly enhances the conditional instability near and around the center of tropical cyclones. The deep vertical updraft, sustained by convergence within the warm sector and further enhanced by latent heat release, favors the development of convective precipitation, with heavy rainfall primarily located on the eastern to northeastern sides of tropical cyclones.
Using dynamic synthesis analysis methods, a comparative analysis was conducted on the reasons for heavy rainfall induced by tropical cyclones entering Jiangsu, both with and without the influence of cold air. A meteorological model of heavy rainfall associated with tropical cyclones in Jiangsu was established, and the following conclusions were obtained: (1) Regardless of the presence of cold air influence, tropical cyclones entering Jiangsu can induce heavy rainfall. The key reasons include the evolution of tropical cyclone"s asymmetric structure, the continuous water vapor transport by low-level jet streams, and the sustained strong upward motion near the tropical cyclone. The differences primarily involve the maintenance mechanisms of dynamic lifting and the growth way of atmospheric instability. (2) When cold air is present, baroclinicity and conditional instability within the outer circulation of tropical cyclones intensify rapidly. Slantwise ascent in the baroclinic zone promotes the sustained lifting of warm-moist air and enhances convective development, with heavy rainfall primarily located on the northern to northeastern sides of tropical cyclones. When cold air is absent, continuous transport of warm-moist air by low-level and ultra-low-level jets significantly enhances the conditional instability near and around the center of tropical cyclones. The deep vertical updraft, sustained by convergence within the warm sector and further enhanced by latent heat release, favors the development of convective precipitation, with heavy rainfall primarily located on the eastern to northeastern sides of tropical cyclones.
Available online: February 10, 2026 , DOI: 10.7519/j.issn.1000-0526.2025.080501
Abstract:
A rare weather event featuring a long-lived period of freezing rain and ice pellets transitioning to snow occurred in the southern Shandong region from February 20-21, 2024. Using conventional observational data, wind profiler radar data, raindrop spectrum data, and ERA5 reanalysis data to analyze this event, the results indicate:?(1)? The event was characterized by return flow precipitation. A southwest warm and humid airflow ahead of the upper-level trough slid upwards along a cold pad caused by a significant temperature drop due to the low-level strong cold air. The interaction between the mid-level warm and humid airflow and the low-level northeastern airflow formed a strong inversion, and the "cold-warm-cold" vertical temperature structure was maintained for a long time, leading to a 12-hour period of freezing rain and ice pellets??(2)? Both freezing rain and ice pellets exhibit typical characteristics of ice-phase melting. The ground temperature during freezing rain is higher, the starting height of the warm layer is lower, and both the thickness and intensity of the warm layer are greater, while the thickness and intensity of the cold layer are smaller. ??(3)? Raindrop spectrum data showed no significant differences in the particle size distribution characteristics between freezing rain and ice pellets. Both had features of small particle size, low number concentration, low precipitation intensity, and high falling speed. The velocity curves of both were similar to those of rainfall but with slight differences. Freezing rain had slightly larger particles and fewer particle numbers than ice pellets. During the snowfall phase, the spectrum widened, characterized by large particle size, high number concentration, high precipitation intensity, and low falling speed.
A rare weather event featuring a long-lived period of freezing rain and ice pellets transitioning to snow occurred in the southern Shandong region from February 20-21, 2024. Using conventional observational data, wind profiler radar data, raindrop spectrum data, and ERA5 reanalysis data to analyze this event, the results indicate:?(1)? The event was characterized by return flow precipitation. A southwest warm and humid airflow ahead of the upper-level trough slid upwards along a cold pad caused by a significant temperature drop due to the low-level strong cold air. The interaction between the mid-level warm and humid airflow and the low-level northeastern airflow formed a strong inversion, and the "cold-warm-cold" vertical temperature structure was maintained for a long time, leading to a 12-hour period of freezing rain and ice pellets??(2)? Both freezing rain and ice pellets exhibit typical characteristics of ice-phase melting. The ground temperature during freezing rain is higher, the starting height of the warm layer is lower, and both the thickness and intensity of the warm layer are greater, while the thickness and intensity of the cold layer are smaller. ??(3)? Raindrop spectrum data showed no significant differences in the particle size distribution characteristics between freezing rain and ice pellets. Both had features of small particle size, low number concentration, low precipitation intensity, and high falling speed. The velocity curves of both were similar to those of rainfall but with slight differences. Freezing rain had slightly larger particles and fewer particle numbers than ice pellets. During the snowfall phase, the spectrum widened, characterized by large particle size, high number concentration, high precipitation intensity, and low falling speed.
Available online: February 09, 2026 , DOI: 10.7519/j.issn.1000-0526.2026.020601
Abstract:
Turbulence is recognized as one of the most prevalent phenomena affecting aircraft flight safety, characterized by complex causes and significant research challenges. Domestic and international research progress is systematically reviewed, with a comprehensive assessment conducted on four key aspects: the definition and causes of turbulence, turbulence indices, impacts of turbulence on flight operations, and turbulence detection and forecasting technologies. Although significant achievements have been made in understanding turbulence and enhancing forecasting capabilities, several scientific challenges persist.These include the in-depth exploration of turbulence generation mechanisms, enhanced forecasting of turbulence across spatiotemporal scales, improved detection and perception through multi-source data fusion, and the advancement and application of new technologies such as Artificial Intelligence (AI). This review is intended to provide valuable insights and essential references for scholars and technical professionals in aviation operations, aircraft design, and aviation meteorology.
Turbulence is recognized as one of the most prevalent phenomena affecting aircraft flight safety, characterized by complex causes and significant research challenges. Domestic and international research progress is systematically reviewed, with a comprehensive assessment conducted on four key aspects: the definition and causes of turbulence, turbulence indices, impacts of turbulence on flight operations, and turbulence detection and forecasting technologies. Although significant achievements have been made in understanding turbulence and enhancing forecasting capabilities, several scientific challenges persist.These include the in-depth exploration of turbulence generation mechanisms, enhanced forecasting of turbulence across spatiotemporal scales, improved detection and perception through multi-source data fusion, and the advancement and application of new technologies such as Artificial Intelligence (AI). This review is intended to provide valuable insights and essential references for scholars and technical professionals in aviation operations, aircraft design, and aviation meteorology.
Available online: February 03, 2026 , DOI: 10.7519/j.issn.1000-0526.2025.072402
Abstract:
The complex topography of the hilly and mountainous regions in Fujian Province gives rise to diverse hail cloud structures, notably isolated convective cells and squall lines. This study addresses the lack of comparative research on the microphysical characteristics of these two types of hail-bearing convection. Based on a persistent severe convective event that occurred in March 2023 over the hilly terrain of southeastern coastal China, S-band dual-polarization radar data and surface precipitation particle spectra observations are employed to comparatively analyze the ground-based spectral characteristics of precipitation particles associated with isolated cells and squall-line systems.The results show that isolated convection during the concentrated hailfall stage exhibits larger particle sizes, more complex spectral structures, more pronounced fluctuations in total particle number concentration, and greater dispersion in the velocity–size spectra, indicating more vigorous ice-phase microphysical processes. In contrast, squall-line convection is characterized by high-density hail, a relatively high concentration of small raindrops, and simpler spectral structures. For both convective types, the peak in precipitation particle concentration appears near the front side of the cloud system in the heavy rainfall region following the main hailfall stage, accompanied by a narrowing of the spectral width and a reduction in large particles. The precipitation process is further divided into four distinct stages, each exhibiting markedly different spectral characteristics.These findings enhance the understanding of the microphysical processes of isolated and squall-line hail clouds and provide scientific support for short-term nowcasting and hail suppression operations.
The complex topography of the hilly and mountainous regions in Fujian Province gives rise to diverse hail cloud structures, notably isolated convective cells and squall lines. This study addresses the lack of comparative research on the microphysical characteristics of these two types of hail-bearing convection. Based on a persistent severe convective event that occurred in March 2023 over the hilly terrain of southeastern coastal China, S-band dual-polarization radar data and surface precipitation particle spectra observations are employed to comparatively analyze the ground-based spectral characteristics of precipitation particles associated with isolated cells and squall-line systems.The results show that isolated convection during the concentrated hailfall stage exhibits larger particle sizes, more complex spectral structures, more pronounced fluctuations in total particle number concentration, and greater dispersion in the velocity–size spectra, indicating more vigorous ice-phase microphysical processes. In contrast, squall-line convection is characterized by high-density hail, a relatively high concentration of small raindrops, and simpler spectral structures. For both convective types, the peak in precipitation particle concentration appears near the front side of the cloud system in the heavy rainfall region following the main hailfall stage, accompanied by a narrowing of the spectral width and a reduction in large particles. The precipitation process is further divided into four distinct stages, each exhibiting markedly different spectral characteristics.These findings enhance the understanding of the microphysical processes of isolated and squall-line hail clouds and provide scientific support for short-term nowcasting and hail suppression operations.
包红军, zeng sihai, wangjianwen, yunxiaobo, lin jian, zhangbo, lizhijia, luanchengmei, wangmeng, xufengwen
Available online: January 30, 2026 , DOI: 10.7519/j.issn.1000-0526.2026.011301
Abstract:
In late July 2025, Chaobai River Basin experienced the largest flood since 1959. A flood simulation and forecasting model is developed for Chaobai River Basin based on a distributed hydrological model, aiming to retrospectively analyze the characteristics of the basin’s "25·7" regional flood in this paper. The Zhangjiafen cross section of the Bai River, the Xiahui cross section of the Chao River, and the Putaoyuan cross section of the Qingshui River were taken for test hydrological sections. Flood simulation and forecasting model based on the GMKHM distributed hydrological model was developed with taking hourly precipitation observation data from CMA regional meteorological stations as model forcing input, incorporating terrain curve numbers and terrain indices to develop a DEM-based runoff and production model, adding a deep groundwater module in the water source separation module. Results show that the peak discharge simulation errors is -1.8% for the Zhangjiafen hydrological section, 3.6% for the Xiahui hydrological section. The model determination coefficient is 0.87 for the Zhangjiafen hydrological section, and 0.89 for the Xiahui hydrological section. for the Putaoyuan section of the Qingshui River, the peak discharge error was 0.9%, and the determination coefficient reached 0.92. GMKHM distributed hydrological model perform well for simulation of "25·7" Flood events in the Chaobai Basin.
In late July 2025, Chaobai River Basin experienced the largest flood since 1959. A flood simulation and forecasting model is developed for Chaobai River Basin based on a distributed hydrological model, aiming to retrospectively analyze the characteristics of the basin’s "25·7" regional flood in this paper. The Zhangjiafen cross section of the Bai River, the Xiahui cross section of the Chao River, and the Putaoyuan cross section of the Qingshui River were taken for test hydrological sections. Flood simulation and forecasting model based on the GMKHM distributed hydrological model was developed with taking hourly precipitation observation data from CMA regional meteorological stations as model forcing input, incorporating terrain curve numbers and terrain indices to develop a DEM-based runoff and production model, adding a deep groundwater module in the water source separation module. Results show that the peak discharge simulation errors is -1.8% for the Zhangjiafen hydrological section, 3.6% for the Xiahui hydrological section. The model determination coefficient is 0.87 for the Zhangjiafen hydrological section, and 0.89 for the Xiahui hydrological section. for the Putaoyuan section of the Qingshui River, the peak discharge error was 0.9%, and the determination coefficient reached 0.92. GMKHM distributed hydrological model perform well for simulation of "25·7" Flood events in the Chaobai Basin.
Available online: January 29, 2026 , DOI: 10.7519/j.issn.1000-0526.2025.072902
Abstract:
To improve the refined forecasting capability for large urban temperatures, this study employed 2m temperature forecast products from ECMWF and observational 2m temperature data from critical locations in Xi"an during the period 2023-2024, to assess the performance of ECMWF model temperature forecasts. Optimal feature factors were identified using subjective prior knowledge and lag correlation analysis. Subsequently, three machine learning models(XGBoost, LightGBM, CatBoost) were applied to separately correct biases in the 2m temperature forecasts at critical locations in Xi"an. To further improve the accuracy of temperature forecasts, these individual models were then integrated using a machine learning stacking approach. The results show that:(1) Overall, the ECMWF 2m temperature forecasts at critical points in Xi"an align well with the actual trends, but the temperatures forecast levels tended to be lower than the observed temperatures. The 2m temperature forecast error exhibits a diurnal variation characteristic, being smaller during the day and larger at night. During significant cooling and precipitation weather events, the 2m temperature forecasts at key points in the urban area of Xi"an exhibit a notable cold bias.(2)The three machine learning temperature forecasting models effectively reduced the forecast bias of 2m temperatures at key points in the urban area of Xi"an as predicted by ECMWF. During nighttime, the root mean square error (RMSE) reduction rate reached 30% to 46%. Bayesian optimization was employed to fine-tune the hyperparameters of the three individual temperature forecasting models. After parameter tuning, the RMSE of the three models on the test set across all time periods decreased by 0.037°C, 0.021°C, and 0.024°C, respectively, compared to before the tuning.(3)By integrating prior knowledge, seven physical quantities closely related to 2m temperature were introduced. Then, using lag correlation analysis, different upper-air region characteristic factors closely associated with each forecast lead time within 24h period were constructed. After optimizing the feature factors for the three individual models, the RMSE on the test set across all time periods decreased by 0.259°C, 0.243°C, and 0.272°C, respectively, compared to before the optimization.(4)The comparison between stacking ensemble and weighted ensemble methods indicated that the former performs better for 2m temperature forecasting. Specifically, the RMSE of the stacking ensemble is reduced by 0.045°C compared to the weighted ensemble, while the accuracy of forecasts within 2°C was improved by 0.021.At each forecast lead time, the stacking ensemble approach demonstrates improved forecast quality compared to individual model forecasts. During significant cooling and precipitation weather events, the stacking ensemble approach reduced temperature forecasts RMSE by 7% to 20% compared to individual model forecasts. Overall, the temperature forecasting model for key points in Xi"an, based on ensemble correction using multiple machine learning methods, effectively reduces the ECMWF model temperature forecast error and further enhances the quality of large urban temperature forecast.
To improve the refined forecasting capability for large urban temperatures, this study employed 2m temperature forecast products from ECMWF and observational 2m temperature data from critical locations in Xi"an during the period 2023-2024, to assess the performance of ECMWF model temperature forecasts. Optimal feature factors were identified using subjective prior knowledge and lag correlation analysis. Subsequently, three machine learning models(XGBoost, LightGBM, CatBoost) were applied to separately correct biases in the 2m temperature forecasts at critical locations in Xi"an. To further improve the accuracy of temperature forecasts, these individual models were then integrated using a machine learning stacking approach. The results show that:(1) Overall, the ECMWF 2m temperature forecasts at critical points in Xi"an align well with the actual trends, but the temperatures forecast levels tended to be lower than the observed temperatures. The 2m temperature forecast error exhibits a diurnal variation characteristic, being smaller during the day and larger at night. During significant cooling and precipitation weather events, the 2m temperature forecasts at key points in the urban area of Xi"an exhibit a notable cold bias.(2)The three machine learning temperature forecasting models effectively reduced the forecast bias of 2m temperatures at key points in the urban area of Xi"an as predicted by ECMWF. During nighttime, the root mean square error (RMSE) reduction rate reached 30% to 46%. Bayesian optimization was employed to fine-tune the hyperparameters of the three individual temperature forecasting models. After parameter tuning, the RMSE of the three models on the test set across all time periods decreased by 0.037°C, 0.021°C, and 0.024°C, respectively, compared to before the tuning.(3)By integrating prior knowledge, seven physical quantities closely related to 2m temperature were introduced. Then, using lag correlation analysis, different upper-air region characteristic factors closely associated with each forecast lead time within 24h period were constructed. After optimizing the feature factors for the three individual models, the RMSE on the test set across all time periods decreased by 0.259°C, 0.243°C, and 0.272°C, respectively, compared to before the optimization.(4)The comparison between stacking ensemble and weighted ensemble methods indicated that the former performs better for 2m temperature forecasting. Specifically, the RMSE of the stacking ensemble is reduced by 0.045°C compared to the weighted ensemble, while the accuracy of forecasts within 2°C was improved by 0.021.At each forecast lead time, the stacking ensemble approach demonstrates improved forecast quality compared to individual model forecasts. During significant cooling and precipitation weather events, the stacking ensemble approach reduced temperature forecasts RMSE by 7% to 20% compared to individual model forecasts. Overall, the temperature forecasting model for key points in Xi"an, based on ensemble correction using multiple machine learning methods, effectively reduces the ECMWF model temperature forecast error and further enhances the quality of large urban temperature forecast.
LUO Ling, ZHANG Zhicha, ZhaoJunPing, CHEN Lie, HUANG Xuanxuan, WANG Liying, LI Wenjuan, LUO Ran, PENG Xiayun, HUANG Juan, YU Pei
Available online: January 26, 2026 , DOI: 10.7519/j.issn.1000-0526.2025.061101
Abstract:
The severe imbalance in sample distribution,characterized by a sharp decrease in the occurrence frequency of severe convective wind with increasing wind speed，is identified as the predominant factor hindering the accurate intensity-based classification of severe convective wind by various existing algorithms.To address this issue，this paper proposes using a non-differentiable Probability of Detection (POD) as the loss function for a Convolutional Neural Network (CNN)，with Bias as its constraint condition.Subsequently，the Multi-objective Optimization Immune Evolution Algorithm (MOIEA) was employed to optimize all the model parameters of the CNN.This led to the development of an identification algorithm for identifying severe convective wind of 17.2、20.8、24.5 m?s-1 and above，named Severe Convective Wind Identification Network (SCWINet). SCWINet leverages data from 2022 to 2024，including radar vertical liquid water content，three-dimensional radar reflectivity，lightning location data，and minute-level ground-based automatic observation station data from Zhejiang Province.It achieves different levels of severe convective wind identification with a temporal resolution of 6 minutes and a spatial resolution of 0.01°.The performance of SCWINet was compared to approaches using the same CNN structure but with the differentiable loss functions of Weighted Mean Squared Error (WMSE)and Balanced Mean Squared Error (BMSE).The applicability of SCWINet was then evaluated based on Threat Score (TS)，Bias，POD，False Alarm Ratio (FAR) using the neighborhood method (with a scanning radius of 5 km)，and the planar distribution characteristics of severe convective wind.The main results are as follows: SCWINet effectively identifies severe convective wind of 17.2、20.8、24.5 m?s-1 and above, associated with both systematic and scattered severe convective systems，with better performance observed in identifying severe convective wind triggered by systematic convection compared to scattered convection.Furthermore，the identification effectiveness generally decreases as wind speed increases，with increased false alarms and missed detections being the primary causes of this pattern.The commonly used WMSE and BMSE approaches，however, showed no ability to identify severe convective wind，with all identified severe convective wind being below 17.2 m?s-1.Nonetheless，the data used in this study are still somewhat limited in terms of feature completeness and volume.Future enhancements in identification accuracy could be achieved by incorporating additional features and data，such as radar radial velocity，Specific Differential Phase (KDP), Differential Reflectivity (ZDR)，and satellite data，which could extend the applicability to even higher wind speeds.
The severe imbalance in sample distribution,characterized by a sharp decrease in the occurrence frequency of severe convective wind with increasing wind speed，is identified as the predominant factor hindering the accurate intensity-based classification of severe convective wind by various existing algorithms.To address this issue，this paper proposes using a non-differentiable Probability of Detection (POD) as the loss function for a Convolutional Neural Network (CNN)，with Bias as its constraint condition.Subsequently，the Multi-objective Optimization Immune Evolution Algorithm (MOIEA) was employed to optimize all the model parameters of the CNN.This led to the development of an identification algorithm for identifying severe convective wind of 17.2、20.8、24.5 m?s-1 and above，named Severe Convective Wind Identification Network (SCWINet). SCWINet leverages data from 2022 to 2024，including radar vertical liquid water content，three-dimensional radar reflectivity，lightning location data，and minute-level ground-based automatic observation station data from Zhejiang Province.It achieves different levels of severe convective wind identification with a temporal resolution of 6 minutes and a spatial resolution of 0.01°.The performance of SCWINet was compared to approaches using the same CNN structure but with the differentiable loss functions of Weighted Mean Squared Error (WMSE)and Balanced Mean Squared Error (BMSE).The applicability of SCWINet was then evaluated based on Threat Score (TS)，Bias，POD，False Alarm Ratio (FAR) using the neighborhood method (with a scanning radius of 5 km)，and the planar distribution characteristics of severe convective wind.The main results are as follows: SCWINet effectively identifies severe convective wind of 17.2、20.8、24.5 m?s-1 and above, associated with both systematic and scattered severe convective systems，with better performance observed in identifying severe convective wind triggered by systematic convection compared to scattered convection.Furthermore，the identification effectiveness generally decreases as wind speed increases，with increased false alarms and missed detections being the primary causes of this pattern.The commonly used WMSE and BMSE approaches，however, showed no ability to identify severe convective wind，with all identified severe convective wind being below 17.2 m?s-1.Nonetheless，the data used in this study are still somewhat limited in terms of feature completeness and volume.Future enhancements in identification accuracy could be achieved by incorporating additional features and data，such as radar radial velocity，Specific Differential Phase (KDP), Differential Reflectivity (ZDR)，and satellite data，which could extend the applicability to even higher wind speeds.
Available online: January 23, 2026 , DOI: 10.7519/j.issn.1000-0526.2026.010602
Abstract:
This study is based on LiDAR wind profiler data and uses the Wind Profile Exponent (WPE) method to calculate the wind profile exponent within the 50–500 meter height range under different surface wind conditions. The results were validated using data from 50 meteorological stations, each with observation heights exceeding 40 meters. The findings indicate that the LiDAR wind profile exponent suffers from a systematic overestimation, with errors closely related to the building density and height characteristics of the observation area. To address this issue, the study innovatively introduces urban building parameters and proposes a multi-factor-driven dynamic correction algorithm for the wind profile exponent, overcoming the limitations of the traditional single roughness parameter approach. After correction, 91.7% of the wind speed calculation errors were controlled within a ±1 Beaufort scale range, with 50.1% showing zero error. Building upon this, a three-dimensional high-altitude wind field reconstruction model tailored for urban complex building environments was further developed. This model has a horizontal resolution of 100 meters and vertical coverage from 50 to 500 meters, enabling real-time monitoring every 5 minutes and forecast outputs up to 168 hours. The model provides a low-cost, high-precision wind disaster prevention and control technology pathway for urban high-altitude operations, offering significant business application potential and engineering value.
This study is based on LiDAR wind profiler data and uses the Wind Profile Exponent (WPE) method to calculate the wind profile exponent within the 50–500 meter height range under different surface wind conditions. The results were validated using data from 50 meteorological stations, each with observation heights exceeding 40 meters. The findings indicate that the LiDAR wind profile exponent suffers from a systematic overestimation, with errors closely related to the building density and height characteristics of the observation area. To address this issue, the study innovatively introduces urban building parameters and proposes a multi-factor-driven dynamic correction algorithm for the wind profile exponent, overcoming the limitations of the traditional single roughness parameter approach. After correction, 91.7% of the wind speed calculation errors were controlled within a ±1 Beaufort scale range, with 50.1% showing zero error. Building upon this, a three-dimensional high-altitude wind field reconstruction model tailored for urban complex building environments was further developed. This model has a horizontal resolution of 100 meters and vertical coverage from 50 to 500 meters, enabling real-time monitoring every 5 minutes and forecast outputs up to 168 hours. The model provides a low-cost, high-precision wind disaster prevention and control technology pathway for urban high-altitude operations, offering significant business application potential and engineering value.
Available online: January 21, 2026 , DOI: 10.7519/j.issn.1000-0526.2026.011602
Abstract:
Fixed-hourly precipitation data may underestimate the characteristics of heavy rainfall at the hourly scale. To gain a deeper understanding of the variation patterns of hourly heavy rainfall, this study utilizes minute-level precipitation observations from 123 national meteorological stations in Shandong Province from June to September during the flood seasons of 2004–2024. It compares the differences between the 60-minute sliding window and fixed-hourly statistical methods in identifying heavy rainfall events, aiming to quantitatively assess the impact of different methods on the frequency, intensity, and spatio-temporal distribution of heavy rainfall. The results indicate that the 60-minute sliding window method captures heavy rainfall events more comprehensively than the fixed-hourly method. The average frequency of heavy rainfall across the province was 124.3 times using the sliding method, which is 1.48 times that of the fixed-hourly method (84.3 times), with regional ratios ranging between 1.31 and 1.81. The maximum precipitation captured by the sliding method was on average 10.0 mm higher, exceeding 30.0 mm at some stations. While the spatial distributions of the two methods are generally consistent, the sliding method exhibits greater spatial variability and extremity. It also more clearly reveals increasing trends in both the frequency and intensity of heavy rainfall, with climatic tendency rates of 1.31 times·(10a)?1 and 4.514 mm·(10a)?1, respectively. Moreover, monthly and diurnal variation characteristics are more distinct with the sliding method, particularly during the high-frequency periods from July to August and from afternoon to early morning. The traditional fixed-hourly statistical system underestimates the frequency and extreme intensity of short-duration heavy rainfall. The 60-minute sliding window method provides a more accurate representation of heavy rainfall characteristics, offering important applications for improving the accuracy of storm monitoring and enhancing early warning capabilities for disaster risk.
Fixed-hourly precipitation data may underestimate the characteristics of heavy rainfall at the hourly scale. To gain a deeper understanding of the variation patterns of hourly heavy rainfall, this study utilizes minute-level precipitation observations from 123 national meteorological stations in Shandong Province from June to September during the flood seasons of 2004–2024. It compares the differences between the 60-minute sliding window and fixed-hourly statistical methods in identifying heavy rainfall events, aiming to quantitatively assess the impact of different methods on the frequency, intensity, and spatio-temporal distribution of heavy rainfall. The results indicate that the 60-minute sliding window method captures heavy rainfall events more comprehensively than the fixed-hourly method. The average frequency of heavy rainfall across the province was 124.3 times using the sliding method, which is 1.48 times that of the fixed-hourly method (84.3 times), with regional ratios ranging between 1.31 and 1.81. The maximum precipitation captured by the sliding method was on average 10.0 mm higher, exceeding 30.0 mm at some stations. While the spatial distributions of the two methods are generally consistent, the sliding method exhibits greater spatial variability and extremity. It also more clearly reveals increasing trends in both the frequency and intensity of heavy rainfall, with climatic tendency rates of 1.31 times·(10a)?1 and 4.514 mm·(10a)?1, respectively. Moreover, monthly and diurnal variation characteristics are more distinct with the sliding method, particularly during the high-frequency periods from July to August and from afternoon to early morning. The traditional fixed-hourly statistical system underestimates the frequency and extreme intensity of short-duration heavy rainfall. The 60-minute sliding window method provides a more accurate representation of heavy rainfall characteristics, offering important applications for improving the accuracy of storm monitoring and enhancing early warning capabilities for disaster risk.
Available online: January 20, 2026 , DOI: 10.7519/j.issn.1000-0526.2026.011701
Abstract:
This study employs a modified Jenkinson-Collison (J-C) circulation type classification method to classifyand diagnose 56 snowfall events that occurred during 2013—2020. The results are as followed. The improved method achieves a classification success rate of 92.9%, identifying cyclonic type (12.5%), low-pressure trough type (42.9%), and westerly advection type (37.5%), with only 7.1% of events remaining unclassified—significantly lower than that from the traditional method. The three circulation types exhibit distinct characteristics. The cyclonic type, corresponding to plateau vortices, features the strongest dynamic and moisture conditions, with the largest mean snowfall (3.4 mm) and longest duration (358 min). The low-pressure trough type, influenced by the southern branch trough, shows the most unstable atmospheric conditions but relatively weak dynamics and moisture, resulting in the lowest mean snowfall (1.2 mm) and shortest duration (170 min). The westerly advection type, dominated by warm ridges and upper-level jets, has the most stable atmospheric conditions, with a “low-level convergence–upper-level divergence” moisture structure, and intermediate snowfall characteristics (mean 1.8 mm; duration 280 min).
This study employs a modified Jenkinson-Collison (J-C) circulation type classification method to classifyand diagnose 56 snowfall events that occurred during 2013—2020. The results are as followed. The improved method achieves a classification success rate of 92.9%, identifying cyclonic type (12.5%), low-pressure trough type (42.9%), and westerly advection type (37.5%), with only 7.1% of events remaining unclassified—significantly lower than that from the traditional method. The three circulation types exhibit distinct characteristics. The cyclonic type, corresponding to plateau vortices, features the strongest dynamic and moisture conditions, with the largest mean snowfall (3.4 mm) and longest duration (358 min). The low-pressure trough type, influenced by the southern branch trough, shows the most unstable atmospheric conditions but relatively weak dynamics and moisture, resulting in the lowest mean snowfall (1.2 mm) and shortest duration (170 min). The westerly advection type, dominated by warm ridges and upper-level jets, has the most stable atmospheric conditions, with a “low-level convergence–upper-level divergence” moisture structure, and intermediate snowfall characteristics (mean 1.8 mm; duration 280 min).
Available online: January 07, 2026 , DOI: 10.7519/j.issn.1000-0526.2026.010601
Abstract:
For basins characterized by long-sequence precipitation data but complex runoff generation mechanisms and a lack of flow process data, this study employed the rational formula method to estimate the early warning time window for flood risk in small and medium-sized rivers. The Pearson Type III frequency curve, commonly used in hydrological frequency analysis, was applied to construct flood-triggering critical areal rainfall thresholds. These thresholds were calibrated using nearly 10 years of flood peak discharge data from small and medium-sized rivers in the middle reaches of the Yellow River. The methodology was then applied and tested for six flood events occurring in 2021 within the basin above the Dongwan hydrological station of Yi River, which lacks measured hydrological characteristics. The results demonstrate that using the rational formula to determine the warning time window is straightforward and practical. Calibrating the critical thresholds based on the same frequency method using historical flood data significantly improved the hit rate of risk warnings to 71.8%, while reducing the miss rate and false alarm rate to 20% and 29.4%, respectively. This forecast accuracy is comparable to the current flood forecasting standards in northern China. The method also performed well when applied in basins lacking hydrological characteristic values. Overall, the flood risk early warning method for small and medium-sized rivers, calibrated using the same frequency approach, effectively addresses the challenge of obtaining long-sequence hydrological data across different regions. It also fully leverages the advantage of meteorological departments possessing long-term precipitation records. This method can be further extended to small and medium-sized watersheds without hydrological stations, providing valuable technical reference for meteorological flood disaster warning efforts in similar basins. Future work could involve classifying small and medium-sized basins based on underlying surface conditions or establishing distinct calibration models for basins dominated by saturation-excess runoff by categorizing soil moisture levels, thereby further enhancing risk warning precision.
For basins characterized by long-sequence precipitation data but complex runoff generation mechanisms and a lack of flow process data, this study employed the rational formula method to estimate the early warning time window for flood risk in small and medium-sized rivers. The Pearson Type III frequency curve, commonly used in hydrological frequency analysis, was applied to construct flood-triggering critical areal rainfall thresholds. These thresholds were calibrated using nearly 10 years of flood peak discharge data from small and medium-sized rivers in the middle reaches of the Yellow River. The methodology was then applied and tested for six flood events occurring in 2021 within the basin above the Dongwan hydrological station of Yi River, which lacks measured hydrological characteristics. The results demonstrate that using the rational formula to determine the warning time window is straightforward and practical. Calibrating the critical thresholds based on the same frequency method using historical flood data significantly improved the hit rate of risk warnings to 71.8%, while reducing the miss rate and false alarm rate to 20% and 29.4%, respectively. This forecast accuracy is comparable to the current flood forecasting standards in northern China. The method also performed well when applied in basins lacking hydrological characteristic values. Overall, the flood risk early warning method for small and medium-sized rivers, calibrated using the same frequency approach, effectively addresses the challenge of obtaining long-sequence hydrological data across different regions. It also fully leverages the advantage of meteorological departments possessing long-term precipitation records. This method can be further extended to small and medium-sized watersheds without hydrological stations, providing valuable technical reference for meteorological flood disaster warning efforts in similar basins. Future work could involve classifying small and medium-sized basins based on underlying surface conditions or establishing distinct calibration models for basins dominated by saturation-excess runoff by categorizing soil moisture levels, thereby further enhancing risk warning precision.
Available online: January 06, 2026 , DOI: 10.7519/j.issn.1000-0526.2025.072803
Abstract:
Based on the operational real-time typhoon data of National Meteorological Centre and other sources of observations including the ERA5 reanalysis data from the European Centre for Medium-Range Weather Forecasting, we analyzed the characteristics of typhoon activities in western North Pacific in 2024. We revealed the different features of typhoons between summer and autumn in 2024 from three aspects: typhoon genesis, typhoon activity, and typhoon landfall. In the summer of 2024, the typhoons genesis number was lower than normal, with the genesis region shifted northward and eastward, and weaker maximum intensities. No typhoons were generated from June 1 to July 18, while the peak summer season saw frequent typhoon occurrences, mostly following northwestward and northward tracks. The number of typhoons landing in China was less than usual, with weaker landfall intensities. However, typhoons were more active in autumn. The typhoon genesis number ranked the third highest in history, with their genesis location shifting northward and westward, and their intensities being unusually strong. Typhoon activities were characterized by distinct periodicity and clustering, with most following westward or northwestward tracks. The number of typhoon landfalls was relatively higher, and the landfall intensities were stronger than the average. Further research showed that the significant differences between summer and autumn typhoons in 2024 were closely related to multiple factors such as the lag effect of sea surface temperature in the year following El Ni?o, the abnormal Western Pacific subtropical high, the phase transition of the tropical intraseasonal oscillation, the onset and withdrawal of the South China Sea summer monsoon, etc.
Based on the operational real-time typhoon data of National Meteorological Centre and other sources of observations including the ERA5 reanalysis data from the European Centre for Medium-Range Weather Forecasting, we analyzed the characteristics of typhoon activities in western North Pacific in 2024. We revealed the different features of typhoons between summer and autumn in 2024 from three aspects: typhoon genesis, typhoon activity, and typhoon landfall. In the summer of 2024, the typhoons genesis number was lower than normal, with the genesis region shifted northward and eastward, and weaker maximum intensities. No typhoons were generated from June 1 to July 18, while the peak summer season saw frequent typhoon occurrences, mostly following northwestward and northward tracks. The number of typhoons landing in China was less than usual, with weaker landfall intensities. However, typhoons were more active in autumn. The typhoon genesis number ranked the third highest in history, with their genesis location shifting northward and westward, and their intensities being unusually strong. Typhoon activities were characterized by distinct periodicity and clustering, with most following westward or northwestward tracks. The number of typhoon landfalls was relatively higher, and the landfall intensities were stronger than the average. Further research showed that the significant differences between summer and autumn typhoons in 2024 were closely related to multiple factors such as the lag effect of sea surface temperature in the year following El Ni?o, the abnormal Western Pacific subtropical high, the phase transition of the tropical intraseasonal oscillation, the onset and withdrawal of the South China Sea summer monsoon, etc.
Available online: January 05, 2026 , DOI: 10.7519/j.issn.1000-0526.2025.071001
Abstract:
Based on 7 S-band dual polarized weather radars in Ningbo and its surrounding areas, the paper develops a monitoring and early warning method of regional radar network for severe convective weather. The method is quantitatively evaluated through 43 cases of severe convective weather., and the result shows that: 1. The amount of data provided per minute by the regional radars network is about 2.66 times that of Ningbo Radar. And it can provide more information for the observation of boundary layers below 1km; 2. The method can respond promptly to severe convective weather in the observation area. Hail warning and downburst warning can be advanced about 79 minutes and 42.6 minutes respectively ahead of live weather;3. Compared to Ningbo radar station, the monitoring and early warning method of regional radars network identifies the convective initiation about 4 minutes in advance, predicts hail process about 12.2 minutes in advance, and predicts downburst process about 13.3 minutes in advance. At the same time, the early warning accuracy rate of the regional radar network for downburst triggered by Mγ CS reaches 100%; 4. The method can be applied to all the central eastern part of China"s land area as well as a small portion of the western and northeastern regions, covering approximately 44000km^2 of land. And the method could have a good application potential in the Yangtze River Delta region, Hubei to Guangzhou area, and Anhui to Shandong area.
Based on 7 S-band dual polarized weather radars in Ningbo and its surrounding areas, the paper develops a monitoring and early warning method of regional radar network for severe convective weather. The method is quantitatively evaluated through 43 cases of severe convective weather., and the result shows that: 1. The amount of data provided per minute by the regional radars network is about 2.66 times that of Ningbo Radar. And it can provide more information for the observation of boundary layers below 1km; 2. The method can respond promptly to severe convective weather in the observation area. Hail warning and downburst warning can be advanced about 79 minutes and 42.6 minutes respectively ahead of live weather;3. Compared to Ningbo radar station, the monitoring and early warning method of regional radars network identifies the convective initiation about 4 minutes in advance, predicts hail process about 12.2 minutes in advance, and predicts downburst process about 13.3 minutes in advance. At the same time, the early warning accuracy rate of the regional radar network for downburst triggered by Mγ CS reaches 100%; 4. The method can be applied to all the central eastern part of China"s land area as well as a small portion of the western and northeastern regions, covering approximately 44000km^2 of land. And the method could have a good application potential in the Yangtze River Delta region, Hubei to Guangzhou area, and Anhui to Shandong area.
Available online: September 02, 2025 , DOI: 10.7519/j.issn.1000-0526.2025.080601
Abstract:
Based on the S-band dual-polarization radar data from Qingpu (Shanghai), Nantong (Jiangsu), Hangzhou, Jiaxing and Huzhou (Zhejiang), the effectiveness evaluation of three volume scan patterns (VCP21D, VCP11D, and VCP216D) are assessed under general precipitation, severe convective and typhoon conditions. The evaluation is based on three methods: subjective identification of characteristic tracer factors, interpolation of reflectivity factor isosurfaces, and wind field retrieve using dual-radar data. The results indicate that both VCP21D and VCP11D can accurately identify the zero-degree bright band characteristic. VCP21D, compared to VCP11D, shows better stability in recognizing the melting layer height, with a smaller standard deviation and a better match to actual conditions. The convective scan patterns (VCP11D and VCP216D) significantly enhance vertical observational resolution compared to the precipitation mode (VCP21D). This improvement is crucial for detecting key severe weather phenomena, such as columns and mesoscale cyclones. Meanwhile, the additional 1.0° elevation angle in VCP216D volume scan pattern is particularly effective in capturing mesoscale features such as low-level gust fronts and sea breeze fronts. Also, this additional scan cut eliminates the effect from ground clutter echoes for data quality improvement. The data from VCP11D and VCP216D volume scan pattern shows great advantages in generating constant altitude plan position indicator (CAPPI), characterize the details special above 5 km altitude, compared to VCP21D volume scan pattern. in both isosurface interpolation and wind field inversion. In the comparison of wind field retrieve, the availability and accuracy of the retrieved data significantly improved, from double VCP11D volume scan pattern radar compared to VCP21D mode.
Based on the S-band dual-polarization radar data from Qingpu (Shanghai), Nantong (Jiangsu), Hangzhou, Jiaxing and Huzhou (Zhejiang), the effectiveness evaluation of three volume scan patterns (VCP21D, VCP11D, and VCP216D) are assessed under general precipitation, severe convective and typhoon conditions. The evaluation is based on three methods: subjective identification of characteristic tracer factors, interpolation of reflectivity factor isosurfaces, and wind field retrieve using dual-radar data. The results indicate that both VCP21D and VCP11D can accurately identify the zero-degree bright band characteristic. VCP21D, compared to VCP11D, shows better stability in recognizing the melting layer height, with a smaller standard deviation and a better match to actual conditions. The convective scan patterns (VCP11D and VCP216D) significantly enhance vertical observational resolution compared to the precipitation mode (VCP21D). This improvement is crucial for detecting key severe weather phenomena, such as columns and mesoscale cyclones. Meanwhile, the additional 1.0° elevation angle in VCP216D volume scan pattern is particularly effective in capturing mesoscale features such as low-level gust fronts and sea breeze fronts. Also, this additional scan cut eliminates the effect from ground clutter echoes for data quality improvement. The data from VCP11D and VCP216D volume scan pattern shows great advantages in generating constant altitude plan position indicator (CAPPI), characterize the details special above 5 km altitude, compared to VCP21D volume scan pattern. in both isosurface interpolation and wind field inversion. In the comparison of wind field retrieve, the availability and accuracy of the retrieved data significantly improved, from double VCP11D volume scan pattern radar compared to VCP21D mode.
Select AllDeselectExport
Display Method:
2012,38(12):1482-1491, DOI: 10.7519/j.issn.1000-0526.2012.12.005
Abstract:
By using the conventional meteorological data, Doppler radar data and NCEP/NCAR reanalysis data, the characteristics of Doppler radar’s reflectivity, environmental condition and trigger mechanism of the heavy rain are analyzed and compared between two abrupt heavy rain processes occurring in Sichuan Basin on 3 July (7.3) and 23 July (7.23) 2011. The results show that: the “7.3” heavy rain happened under a typical circulation background, and moisture transporting to the heavy rain area from the South China Sea was smoothly, thus the heavy rainfall maintained so long, but the “7.23” heavy rain occurred behind the upper cold vortex, and convective unstable energy was abundant and vertical wind shear was strong, thus this heavy rain process happened with hail and thunderstorm weather accompanied, its radar reflectivity was 5 dBz stronger than “7.3” case and had the characteristics of severe storms such as the low level weak reflectivity and the upper echo overhang. As a whole, the non equilibrium force is contributed to the occurrence of heavy rain and it is the excited mechanism of the two heavy rainfalls, and the change of the divergence evolvement is consistent with the strength and the position of the heavy rain which would happen 6 hours later．
By using the conventional meteorological data, Doppler radar data and NCEP/NCAR reanalysis data, the characteristics of Doppler radar’s reflectivity, environmental condition and trigger mechanism of the heavy rain are analyzed and compared between two abrupt heavy rain processes occurring in Sichuan Basin on 3 July (7.3) and 23 July (7.23) 2011. The results show that: the “7.3” heavy rain happened under a typical circulation background, and moisture transporting to the heavy rain area from the South China Sea was smoothly, thus the heavy rainfall maintained so long, but the “7.23” heavy rain occurred behind the upper cold vortex, and convective unstable energy was abundant and vertical wind shear was strong, thus this heavy rain process happened with hail and thunderstorm weather accompanied, its radar reflectivity was 5 dBz stronger than “7.3” case and had the characteristics of severe storms such as the low level weak reflectivity and the upper echo overhang. As a whole, the non equilibrium force is contributed to the occurrence of heavy rain and it is the excited mechanism of the two heavy rainfalls, and the change of the divergence evolvement is consistent with the strength and the position of the heavy rain which would happen 6 hours later．
2017,43(7):769-780, DOI: 10.7519/j.issn.1000-0526.2017.07.001
Abstract:
The spatial distributions of severe convective wind (SCW) and nonsevere thunderstorms (NT) over South China, occurring between 08:00 BT and 20:00 BT during spring and summer in 2010-2014, were analyzed by using the observational data from China Meteorological Administration. And then, their environmental characteristics were compared between SCW and NT in spring and summer. It was found that SCW in summer is more frequently than that in spring and that NT in summer is about 3.6 times the counts of NT in spring. SCW events mainly concentrate in the western Guangdong to the Pearl River Delta Region. Compared to NT, SCW is generally associated with stronger baroclinity, instability and stronger dynamic forcing. The precipitable water and averaged relative humidity between 700-500 hPa of SCW tend to be higher than those of NT in spring, while the opposite is the case for the pattern in summer. In conclusion, it is obvious that the dynamic forcing for SCW in spring is much better than these in summer, while the thermal condition is more significant in summer.
The spatial distributions of severe convective wind (SCW) and nonsevere thunderstorms (NT) over South China, occurring between 08:00 BT and 20:00 BT during spring and summer in 2010-2014, were analyzed by using the observational data from China Meteorological Administration. And then, their environmental characteristics were compared between SCW and NT in spring and summer. It was found that SCW in summer is more frequently than that in spring and that NT in summer is about 3.6 times the counts of NT in spring. SCW events mainly concentrate in the western Guangdong to the Pearl River Delta Region. Compared to NT, SCW is generally associated with stronger baroclinity, instability and stronger dynamic forcing. The precipitable water and averaged relative humidity between 700-500 hPa of SCW tend to be higher than those of NT in spring, while the opposite is the case for the pattern in summer. In conclusion, it is obvious that the dynamic forcing for SCW in spring is much better than these in summer, while the thermal condition is more significant in summer.
2010,36(3):9-18, DOI: 10.7519/j.issn.1000-0526.2010.3.002
Abstract:
Potential vorticity (PV) is one of the important concepts in advanced synoptic and dynamic meteorology. This paper is a brief introduction to the theory of potential vorticity, including the concept of PV, the conservation and invertibility of PV, PV thinking, moist PV (MPV), and the application of PV theory.
Potential vorticity (PV) is one of the important concepts in advanced synoptic and dynamic meteorology. This paper is a brief introduction to the theory of potential vorticity, including the concept of PV, the conservation and invertibility of PV, PV thinking, moist PV (MPV), and the application of PV theory.
FU Jiaolan, MA Xuekuan, CHEN Tao, ZHANG Fang, ZHANG Xidi, SUN Jun, QUAN Wanqing, YANG Shunan, SHEN Xiaolin
2017,43(5):528-539, DOI: 10.7519/j.issn.1000-0526.2017.05.002
Abstract:
An extremely severe precipitation event took place in North China in 19-20 July 2016. It was characterized by large rainfall, persistent rainfall, warm cloud rainfall, strong local rainfall intensity and orographic precipitation. Its rainfall was larger than that of the extreme rainfall in 3-5 August 1996, and only next to the amount of the 2-7 August 1963 extreme rainfall event. It occurred under the circulation background of the South Asia high moving eastward, the West Pacific subtropical high moving northwestward and the low vortex in the westerlies developing in mid high latitude. The abnormal development of Huanghuai cyclone, southwest and southeast low level jets, and the abnormally abundant moisture indicates that the dynamic lifting and moisture conditions favored this severe rainfall process significantly. The whole rainfall event presented clearly the phase characteristics, and could be divided into two stages. The first stage was the orographic rainfall caused by the easterly winds ahead of the trough from the early morning to the daytime of 19 July, while the second part was produced by spiral rain bands in the north side of Huanghuai cyclone from the night of 19 to the daytime of 20 July. In the first stage, the easterly low level jet was lifted by the Taihang Mountains, which continuously triggered the convective cells along the east edge of the mountains. The weak dry and cold advection at mid level and the strong warm and wet advection at low level jointly maintained the convective instability. The cold pool generated by heavy rainfall and the mesoscale frontogenesis process created by local orographic effect provided favorable conditions for severe convections to occur continuously. The second stage rainfall was mainly related to the development of cut off vortex and Huanghuai cyclone. The blocking of the high pressure system slowed the steps of Huanghuai cyclone in North China, thus leading to the long lasting rainfall process.
An extremely severe precipitation event took place in North China in 19-20 July 2016. It was characterized by large rainfall, persistent rainfall, warm cloud rainfall, strong local rainfall intensity and orographic precipitation. Its rainfall was larger than that of the extreme rainfall in 3-5 August 1996, and only next to the amount of the 2-7 August 1963 extreme rainfall event. It occurred under the circulation background of the South Asia high moving eastward, the West Pacific subtropical high moving northwestward and the low vortex in the westerlies developing in mid high latitude. The abnormal development of Huanghuai cyclone, southwest and southeast low level jets, and the abnormally abundant moisture indicates that the dynamic lifting and moisture conditions favored this severe rainfall process significantly. The whole rainfall event presented clearly the phase characteristics, and could be divided into two stages. The first stage was the orographic rainfall caused by the easterly winds ahead of the trough from the early morning to the daytime of 19 July, while the second part was produced by spiral rain bands in the north side of Huanghuai cyclone from the night of 19 to the daytime of 20 July. In the first stage, the easterly low level jet was lifted by the Taihang Mountains, which continuously triggered the convective cells along the east edge of the mountains. The weak dry and cold advection at mid level and the strong warm and wet advection at low level jointly maintained the convective instability. The cold pool generated by heavy rainfall and the mesoscale frontogenesis process created by local orographic effect provided favorable conditions for severe convections to occur continuously. The second stage rainfall was mainly related to the development of cut off vortex and Huanghuai cyclone. The blocking of the high pressure system slowed the steps of Huanghuai cyclone in North China, thus leading to the long lasting rainfall process.
2012,38(10):1255-1266, DOI: 10.7519/j.issn.1000-0526.2012.10.012
Abstract:
Precipitation characteristics, environment conditions, generation and development of the mesoscale convective system that brought about the extreme torrential rain in Beijing on 21 July 2012 were analyzed comprehensively in this paper by using various conventional and unconventional data. The results showed that the extreme torrential rain had the characteristics of long duration, great rainfall and wide coverage area and its process consisted of warm area precipitation and frontal precipitation. The warm area rainfall started earlier, the severe precipitation center was scattered and lasted long while the frontal rainfallprocess contained several severe rainfall centers with high precipitation efficiency, lasting a short time.Environment conditions of the mesoscale convective system that triggered this extreme severe rainfall were analyzed. The results showed that interactions of high level divergence, the wind shear and convergence with the vortex in the lower troposphere and the surface wind convergence line provided favorable environment to the severe extreme rain. The warm humid airs from the tropical and sub tropical zones converged over the torrential rain region, continuous and sufficient water vapor manifested as high atmospheric column of precipitable water and strong low level water vapor convergence and other extreme vapor conditions for the torrential rain. In addition, the intense precipitation was triggered by the vortex wind shear, wind disturbance on low level jet, surface wind convergence line and the effect of terrain under the condition of the plentiful water vapour and maintained. With the cold front moved eastward, heavy frontal rainfall was brought by the development and evolution of convective system made by the cold air and the suitable vertical wind shear.Generation and development processes of the mesoscale convective system were also studied. The findings suggested that stratiform cloud precipitation and dispersed convective precipitation occurred firstly in the precipitation process. The warm and steady stratiform cloud precipitation changed to be highly organized convectional precipitation as the cold dry air invaded. Many small scale and mesoscale convective clusters developed into mesoscale convective complex (MCC), leading to the extreme severe precipitation. Since all the directions of the echo long axis, terrain and echo movement were parallel, train effect was obviously seen in the radar echo imegery during this precipitation process. Meanwhile, the radar echo had the characteristics of backward propagation and low centroid which was similar to tropical heavy rainfalls. Finally, a series of scientific problems were proposed according to the integrated analysis on the observation data of this rare torrential rain event, such as the causes for the extreme torrential rain and the extreme rich water vapor, mechanisms for the warm area torrential rain in the north of China, the mechanism for the train effect and backward propagation, mechanisms for the organization and maintenance of the convective cells, the simulation and analysis ability of the numerical models to extreme torrential rains and so on.
Precipitation characteristics, environment conditions, generation and development of the mesoscale convective system that brought about the extreme torrential rain in Beijing on 21 July 2012 were analyzed comprehensively in this paper by using various conventional and unconventional data. The results showed that the extreme torrential rain had the characteristics of long duration, great rainfall and wide coverage area and its process consisted of warm area precipitation and frontal precipitation. The warm area rainfall started earlier, the severe precipitation center was scattered and lasted long while the frontal rainfallprocess contained several severe rainfall centers with high precipitation efficiency, lasting a short time.Environment conditions of the mesoscale convective system that triggered this extreme severe rainfall were analyzed. The results showed that interactions of high level divergence, the wind shear and convergence with the vortex in the lower troposphere and the surface wind convergence line provided favorable environment to the severe extreme rain. The warm humid airs from the tropical and sub tropical zones converged over the torrential rain region, continuous and sufficient water vapor manifested as high atmospheric column of precipitable water and strong low level water vapor convergence and other extreme vapor conditions for the torrential rain. In addition, the intense precipitation was triggered by the vortex wind shear, wind disturbance on low level jet, surface wind convergence line and the effect of terrain under the condition of the plentiful water vapour and maintained. With the cold front moved eastward, heavy frontal rainfall was brought by the development and evolution of convective system made by the cold air and the suitable vertical wind shear.Generation and development processes of the mesoscale convective system were also studied. The findings suggested that stratiform cloud precipitation and dispersed convective precipitation occurred firstly in the precipitation process. The warm and steady stratiform cloud precipitation changed to be highly organized convectional precipitation as the cold dry air invaded. Many small scale and mesoscale convective clusters developed into mesoscale convective complex (MCC), leading to the extreme severe precipitation. Since all the directions of the echo long axis, terrain and echo movement were parallel, train effect was obviously seen in the radar echo imegery during this precipitation process. Meanwhile, the radar echo had the characteristics of backward propagation and low centroid which was similar to tropical heavy rainfalls. Finally, a series of scientific problems were proposed according to the integrated analysis on the observation data of this rare torrential rain event, such as the causes for the extreme torrential rain and the extreme rich water vapor, mechanisms for the warm area torrential rain in the north of China, the mechanism for the train effect and backward propagation, mechanisms for the organization and maintenance of the convective cells, the simulation and analysis ability of the numerical models to extreme torrential rains and so on.
2006,32(10):64-69, DOI: 10.7519/j.issn.1000-0526.2006.10.010
Abstract:
Based on the data of CINRAD Doppler Radar which located at Xinle of Hebei Province,the hail,strong wind and heavy rainfall weather events in mid-south Hebei in 2004 are statistically analyzed.The routine radar products,such as echo reflectivity,radial velocity,Vertically Integrated Liquid(VIL)Water,hail index,mesocyclone,velocity azimuth display wind profile,etc.are used in this statistics.The results show that hail's VIL value is larger than generic thunder storm's.At the same time,greater VIL value and longer sustaining will bring about greater diameter hail and larger effect area.It is the very useful index to indicate strong wind in mesocyclone products and the wind direction sudden change in radial velocity products.A reference based on analyzing this type synoptic forecast with radar system in future is proposed.
Based on the data of CINRAD Doppler Radar which located at Xinle of Hebei Province,the hail,strong wind and heavy rainfall weather events in mid-south Hebei in 2004 are statistically analyzed.The routine radar products,such as echo reflectivity,radial velocity,Vertically Integrated Liquid(VIL)Water,hail index,mesocyclone,velocity azimuth display wind profile,etc.are used in this statistics.The results show that hail's VIL value is larger than generic thunder storm's.At the same time,greater VIL value and longer sustaining will bring about greater diameter hail and larger effect area.It is the very useful index to indicate strong wind in mesocyclone products and the wind direction sudden change in radial velocity products.A reference based on analyzing this type synoptic forecast with radar system in future is proposed.
Zhou Yuquan, Chen Yingying, Li Juan, Huang Yimei, He Xiaodong, Zhou Feifei, Wu Menxin, Hu Bo, Mao Jietai
2008,34(12):27-35, DOI: 10.7519/j.issn.1000-0526.2008.12.004
Abstract:
Cloud macro and micro physical characteristic parameters play an important role not only in the field of the analysis and forecast of the weather and climate, but also in the field of weather modification to identify the seeding c ondition. Based on the data from FY-2C/D stationary satellite and SBDART radiati on transfer model, associated with the sounding data and surface information, a method retrieving cloud macro and micro physical parameters is established in th is research. These parameters include cloud top height, cloud top temperature, d epth of super-cooled layer, depth of warm layer, cloud bottom height, depth of c loud, cloud optical thickness, cloud effective particle radius and cloud liquid water content. It has been run operationally. In this paper, the correlated info rmation such as physical meaning, retrieving method and technology, retrieving p rocess and data format are simply introduced. Furthermore, comparing with the ob servation of Cloudsat up to the minute, the retrieving results of main cloud par ameters are proved to be reasonable and usable. By contrast with same kind produ cts of MODIS, it also shows good corresponding relationship.
Cloud macro and micro physical characteristic parameters play an important role not only in the field of the analysis and forecast of the weather and climate, but also in the field of weather modification to identify the seeding c ondition. Based on the data from FY-2C/D stationary satellite and SBDART radiati on transfer model, associated with the sounding data and surface information, a method retrieving cloud macro and micro physical parameters is established in th is research. These parameters include cloud top height, cloud top temperature, d epth of super-cooled layer, depth of warm layer, cloud bottom height, depth of c loud, cloud optical thickness, cloud effective particle radius and cloud liquid water content. It has been run operationally. In this paper, the correlated info rmation such as physical meaning, retrieving method and technology, retrieving p rocess and data format are simply introduced. Furthermore, comparing with the ob servation of Cloudsat up to the minute, the retrieving results of main cloud par ameters are proved to be reasonable and usable. By contrast with same kind produ cts of MODIS, it also shows good corresponding relationship.
ZHANG Xiaoling, ZHANG Tao, LIU Xinhua, ZHOU Qingliang, CHEN Yun, ZHOU Xiaoxia, ZHENG Yongguang, ZHAO Surong
2010,36(7):143-150, DOI: 10.7519/j.issn.1000-0526.2010.7.021
Abstract:
Mesoscale severe weather forecasting ability is limited, in some sense for a lack of valid analysis on mesoscale convective systems and its favorable environments. This paper introduces the mesoscale weather chart analysis techniq ue which was tested in the National Meteorological Center (NMC). Mesoscale weath er chart analyzes the favorable environmental conditions of mesoscale convective systems based on observational data and numerical weather forecast outputs. It includes upper air composite chart and surface chart. In the upper air composite ch art, by analyzing wind, temperature, moisture, temperature change and height change, the diagnostic systems and features in all the lower, middle and upper t roposphere isobaric layers are combined into one plot, which can clearly displa y the available environments and synoptic pattern of severe convective weather. In the surface chart, the analysis contents are pressure, wind, temperature, moi sture, convective weather phenomena and all kinds of boundaries (fronts). The te st in NMC shows that mesoscale weather chart analysis is a dependable means for severe convective weather outlook forecasting.
Mesoscale severe weather forecasting ability is limited, in some sense for a lack of valid analysis on mesoscale convective systems and its favorable environments. This paper introduces the mesoscale weather chart analysis techniq ue which was tested in the National Meteorological Center (NMC). Mesoscale weath er chart analyzes the favorable environmental conditions of mesoscale convective systems based on observational data and numerical weather forecast outputs. It includes upper air composite chart and surface chart. In the upper air composite ch art, by analyzing wind, temperature, moisture, temperature change and height change, the diagnostic systems and features in all the lower, middle and upper t roposphere isobaric layers are combined into one plot, which can clearly displa y the available environments and synoptic pattern of severe convective weather. In the surface chart, the analysis contents are pressure, wind, temperature, moi sture, convective weather phenomena and all kinds of boundaries (fronts). The te st in NMC shows that mesoscale weather chart analysis is a dependable means for severe convective weather outlook forecasting.
2013,39(10):1284-1292, DOI: 10.7519/j.issn.1000-0526.2013.10.006
Abstract:
Based on the fog observation data during 24-27 December 2006 (advection radiation fog), NCEP NC reanalysis data (2.5°×2.5°) and GDAS global meteorological data (1°×1°), detailed trajectory analysis of the boundary layer characteristics and water vapor transport of the fog is investigated, combined with the weather condition, meteorological elements and physical quantity field. The results show that: (1) there is thick inversion layer, even multi layer inversion throughout the dense fog event. Temperatures of different inversion tops in the middle and high levels are 2-5℃ higher than the surface temperature. The thickness of inversion layer is more than 200 m, and it gets to 500 m at 08:00 BT 26 December, indicating the atmosphere is very stable and conducive to the convergence of water vapor before the fog forms. However, it is not favorable for the divergence of water vapor after the formation of fog, which helps the development and maintenance of the fog, causing the fog to last about 64 hours with dense fog (visibility <50 m) about 37 hours; (2) The divergence of water vapor flux in low level is negative in the advection fog event. The upper air has persistent moisture convergence and the strongest moisture convergence appears at 02:00 BT 25 December, being -30×10-7 g·s-1·cm-2·hPa-1. The accumulation of low level water vapor makes fog form and develop while the divergence of water vapor flux speeds up its dissipation. 〖JP2〗The long lasting advection radiation fog is mainly caused by the continuous water vapor convergence; (3) The water vapor path is from the coastal area in easten China to Nanjing. The water vapor is continuously supplied from sea during the fog event, with the water vapor flux maximum getting to 2 g·s-1·hPa-1·cm-1. The sufficient supply and supplementary of water vapor determines the duration of the fog.
Based on the fog observation data during 24-27 December 2006 (advection radiation fog), NCEP NC reanalysis data (2.5°×2.5°) and GDAS global meteorological data (1°×1°), detailed trajectory analysis of the boundary layer characteristics and water vapor transport of the fog is investigated, combined with the weather condition, meteorological elements and physical quantity field. The results show that: (1) there is thick inversion layer, even multi layer inversion throughout the dense fog event. Temperatures of different inversion tops in the middle and high levels are 2-5℃ higher than the surface temperature. The thickness of inversion layer is more than 200 m, and it gets to 500 m at 08:00 BT 26 December, indicating the atmosphere is very stable and conducive to the convergence of water vapor before the fog forms. However, it is not favorable for the divergence of water vapor after the formation of fog, which helps the development and maintenance of the fog, causing the fog to last about 64 hours with dense fog (visibility <50 m) about 37 hours; (2) The divergence of water vapor flux in low level is negative in the advection fog event. The upper air has persistent moisture convergence and the strongest moisture convergence appears at 02:00 BT 25 December, being -30×10-7 g·s-1·cm-2·hPa-1. The accumulation of low level water vapor makes fog form and develop while the divergence of water vapor flux speeds up its dissipation. 〖JP2〗The long lasting advection radiation fog is mainly caused by the continuous water vapor convergence; (3) The water vapor path is from the coastal area in easten China to Nanjing. The water vapor is continuously supplied from sea during the fog event, with the water vapor flux maximum getting to 2 g·s-1·hPa-1·cm-1. The sufficient supply and supplementary of water vapor determines the duration of the fog.
2014,40(4):400-411, DOI: 10.7519/j.issn.1000-0526.2014.04.002
Abstract:
Based on the synoptic environment analysis of about 100 severe convection cases in China since 2000 and the reference of related literatures, from the perspectives of the three essential conditions for the development of severe convection, namely the thermal instability, lift and moisture, five basic synoptic situation configurations of severe convection in China are proposed and expounded. They are cold advection forcing category, warm advection forcing category, baroclinic frontogenesis category, quasi barotropic category and elevated thunderstorm category. The typical characteristics of the upper cold advection forcing category is that the mid upper strong cold advection above 500 hPa strengthens and reaches the boundary warm convergence zone. The warm advection forcing category is characterized by trough with special structure moving over low level strong warm and moist advection. The deep convection produced by the mid lower layer convergence of cold and warm air features the baroclinic frontogenesis category. The quasi barotropic category mostly occurs at the northern and the southern edges or the interior of summer subtropical high and the area with weak baroclinicity, where the dynamic forcing and the surface inhomogeneous local heating play major roles. The features of elevated thunderstorms are the southwest jet in 700-500 hPa lifted by boundary cold wedge and the instable energy is from above 700 hPa. The classification based on the difference of the formation mechanisms can grasp accurately the synoptic characteristics, the situation configurations, the dynamic and thermal properties and the key points in analyzing short term potential forecast, providing more technical support to further enhance the level of weather prediction.
Based on the synoptic environment analysis of about 100 severe convection cases in China since 2000 and the reference of related literatures, from the perspectives of the three essential conditions for the development of severe convection, namely the thermal instability, lift and moisture, five basic synoptic situation configurations of severe convection in China are proposed and expounded. They are cold advection forcing category, warm advection forcing category, baroclinic frontogenesis category, quasi barotropic category and elevated thunderstorm category. The typical characteristics of the upper cold advection forcing category is that the mid upper strong cold advection above 500 hPa strengthens and reaches the boundary warm convergence zone. The warm advection forcing category is characterized by trough with special structure moving over low level strong warm and moist advection. The deep convection produced by the mid lower layer convergence of cold and warm air features the baroclinic frontogenesis category. The quasi barotropic category mostly occurs at the northern and the southern edges or the interior of summer subtropical high and the area with weak baroclinicity, where the dynamic forcing and the surface inhomogeneous local heating play major roles. The features of elevated thunderstorms are the southwest jet in 700-500 hPa lifted by boundary cold wedge and the instable energy is from above 700 hPa. The classification based on the difference of the formation mechanisms can grasp accurately the synoptic characteristics, the situation configurations, the dynamic and thermal properties and the key points in analyzing short term potential forecast, providing more technical support to further enhance the level of weather prediction.
2012,38(2):164-173, DOI: 10.7519/j.issn.1000-0526.2012.02.004
Abstract:
Many weather forecasters seem to have acquaintance with most of basic concepts or fundamental theories which are connected with severe convection, but some of them are misapplied frequently by some forecasters when they are engaged in severe convective weather analysis or forecasting argumentation. Due to the above problem, some basic concepts and fundamental theories should be explained from the view of forecasting application. The following issues are discussed in this paper. They are the relationship between humidity and water vapor content, the role of clod air during the precipitation process, the fundamental theories connected with thermal and dynamic instability, the sounding analysis related to instability parameters, the relationship between helicity or moist potential vorticity and instability, the relationship among the convergence line, lifting velocity and convective vertical movement, and the essential connection between the synoptic patterns and severe convective phenomena.
Many weather forecasters seem to have acquaintance with most of basic concepts or fundamental theories which are connected with severe convection, but some of them are misapplied frequently by some forecasters when they are engaged in severe convective weather analysis or forecasting argumentation. Due to the above problem, some basic concepts and fundamental theories should be explained from the view of forecasting application. The following issues are discussed in this paper. They are the relationship between humidity and water vapor content, the role of clod air during the precipitation process, the fundamental theories connected with thermal and dynamic instability, the sounding analysis related to instability parameters, the relationship between helicity or moist potential vorticity and instability, the relationship among the convergence line, lifting velocity and convective vertical movement, and the essential connection between the synoptic patterns and severe convective phenomena.
2009,35(1):55-64, DOI: 10.7519/j.issn.1000-0526.2009.1.007
Abstract:
A strong rainstorm is analysis which occurred in Xinghua located the north of Ji angsu province on 25 July 2007. Results show that wind disaster originated from two kinds of rainstorm. One kind was the gust front which occurred at the front of the storm. Strong wind of grade 7-9 was attained when it happened. Another ki nd was the downburst arose in the multi cell storm. The original height of refl ectivity core was higher than -20℃ isotherm. It had the characteristics of conv ergence on the mid level and descending of reflectivity core. The strong wind ab ove grade 10 was attained, when the descending airflow diverged strongly on the ground. A new cell was combined with the former storm above the gust front, thus the storm enhanced. When the downburst happened, the storm weakened, and another new cell was combin ed with the former storm. The downburst happened continuously, and the impact of gust front persisted.
A strong rainstorm is analysis which occurred in Xinghua located the north of Ji angsu province on 25 July 2007. Results show that wind disaster originated from two kinds of rainstorm. One kind was the gust front which occurred at the front of the storm. Strong wind of grade 7-9 was attained when it happened. Another ki nd was the downburst arose in the multi cell storm. The original height of refl ectivity core was higher than -20℃ isotherm. It had the characteristics of conv ergence on the mid level and descending of reflectivity core. The strong wind ab ove grade 10 was attained, when the descending airflow diverged strongly on the ground. A new cell was combined with the former storm above the gust front, thus the storm enhanced. When the downburst happened, the storm weakened, and another new cell was combin ed with the former storm. The downburst happened continuously, and the impact of gust front persisted.
2015,41(2):212-218, DOI: 10.7519/j.issn.1000-0526.2015.02.009
Abstract:
From 1 May to 8 June 2013 CMA Meteorological Observation Centre conducted an experiment of cloud height observations by using cloud radar (35 GHz), whose observation data are the echo power value and temporal resolution is 1 min and a ceilometer whose observation data are the back scattering intens data with 1 min temporal resolution. The result of analyzing the data observed from the 39 d experiment indicates that: (1) the data acquisition ratio of cloud radar is 26% larger than that of ceilometer; (2) the ratio is 51% in fog haze weather; (3) relatively, precipitation has more significant effect on cloud base height measured by laser ceilometer than that by cloud radar; (4) height of cloud base measured by cloud radar is almost consistent with the height by ceilometer because their average deviation is less than 300 m.
From 1 May to 8 June 2013 CMA Meteorological Observation Centre conducted an experiment of cloud height observations by using cloud radar (35 GHz), whose observation data are the echo power value and temporal resolution is 1 min and a ceilometer whose observation data are the back scattering intens data with 1 min temporal resolution. The result of analyzing the data observed from the 39 d experiment indicates that: (1) the data acquisition ratio of cloud radar is 26% larger than that of ceilometer; (2) the ratio is 51% in fog haze weather; (3) relatively, precipitation has more significant effect on cloud base height measured by laser ceilometer than that by cloud radar; (4) height of cloud base measured by cloud radar is almost consistent with the height by ceilometer because their average deviation is less than 300 m.
2014,40(2):133-145, DOI: 10.7519/j.issn.1000-0526.2014.02.001
Abstract:
By using the NCEP reanalysis data, the vapor budget of the area covered by the severe torrential rain over the northeast of North China on 21 July, 2012 is calculated according to the vapor budget equation. The results show that meridional water vapor transportation is dominant while the extremely heavy rain hits Beijing Region, where most moist vapor comes from the southern boundary below 500 hPa. The low level regional moisture convergence is consistent with the time and space when the torrential rain breaks out and develops. Above the middle level the vertical vapor transport is more prominent. Then the variation features of the vapor transport corridors and their moisture contributions are got through the HYSPLIT mode. The backward trajectory analyses illustrate two major vapor transport corridors. The moistest vapor derived from Yellow Sea and East China Sea along the low level make the main moisture contribution during the heavy precipitation. Moisture from the South China Sea and the Bay of Bengal strengthens the water vapor in the region when the heavy rain starts and develops. Also the drier vapor corridor along the high level from the northwest of China plays an important role in this case.
By using the NCEP reanalysis data, the vapor budget of the area covered by the severe torrential rain over the northeast of North China on 21 July, 2012 is calculated according to the vapor budget equation. The results show that meridional water vapor transportation is dominant while the extremely heavy rain hits Beijing Region, where most moist vapor comes from the southern boundary below 500 hPa. The low level regional moisture convergence is consistent with the time and space when the torrential rain breaks out and develops. Above the middle level the vertical vapor transport is more prominent. Then the variation features of the vapor transport corridors and their moisture contributions are got through the HYSPLIT mode. The backward trajectory analyses illustrate two major vapor transport corridors. The moistest vapor derived from Yellow Sea and East China Sea along the low level make the main moisture contribution during the heavy precipitation. Moisture from the South China Sea and the Bay of Bengal strengthens the water vapor in the region when the heavy rain starts and develops. Also the drier vapor corridor along the high level from the northwest of China plays an important role in this case.
2012,38(1):1-16, DOI: 10.7519/j.issn.1000-0526.2012.01.001
Abstract:
In this paper, the modulation of atmospheric MJO on typhoon generation over the northwestern Pacific and its mechanism are first studied by using the MJO index. The results show that the MJO plays an important modulation role in typhoon generation over the northwestern Pacific: The proportion of typhoon number is 21 between active period and inactive period; During the MJO active period, the proportion of typhoon number is also 2:1 between phases 5-6 and phases 2-3 of MJO. The composite analyses of atmospheric circulation show that there are different circulation patterns over the northwestern Pacific in different phases of the MJO, which will affect the typhoon generation. In phases 5-6 (2-3), the dynamic factor and convective heating patterns over western Pacific are favorable (unfavorable) for typhoon generation. Then， the comparing analyses of the 30-60 day low frequency kinetic energy in lower and higher levels of the troposphere show that the atmospheric intraseasonal oscillation over the northwestern Pacific has a clear impact on the typhoon generation. There is an evident positive (negative) anomaly area of 30-60 day low frequency kinetic energy in the more (less) typhoon years over the northwestern Pacific east of the Philippines, which means that strong (weak) atmospheric intraseasonal oscillation (ISO) over the northwestern Pacific is favorable (unfavorable) for typhoon generation. The analyses of 200 hPa velocity potential show that there is a clear divergence (convergence) pattern over the northwestern Pacific in the more (less) typhoon years, which is favorable (unfavorable) for typhoon generation. The modulation of the intraseasonal oscillation on the typhoon tracks over the northwestern Pacific is studied by observational data analyses. We classified the main classes of typhoon tracks into 5 types as straight west moving typhoons (I), northwest moving typhoons (II), recurving to Korea/west of Japan typhoons (III), landing on Japan typhoons (IV) and recurving to the east of Japan typhoons (V). Then the composite analyses of atmospheric low-frequency wind fields at 850, 500 and 200 hPa, corresponding to the typhoon forming date, for every typhoon track are completed. The analysis results of relationships between the low-frequency (ISO) wind fields and typhoon tracks have indicated that the typhoon tracks will be affected by wind pattern of the ISO. The low frequency positive vorticity belt (the maximum value line of cyclonic vorticity) associated with low-frequency cyclone (LFC) at 850 hPa is so closely related to the typhoon track, that the maximum value line (belt) of low frequency cyclonic vorticity can be an important factor to predicate the typhoon tracks over the northwestern Pacific. And the typhoon tracks will be also affected by the ISO circulation pattern at 200 hPa, particularly the strong low frequency wind associated with low frequency anticyclone (LFAC).
In this paper, the modulation of atmospheric MJO on typhoon generation over the northwestern Pacific and its mechanism are first studied by using the MJO index. The results show that the MJO plays an important modulation role in typhoon generation over the northwestern Pacific: The proportion of typhoon number is 21 between active period and inactive period; During the MJO active period, the proportion of typhoon number is also 2:1 between phases 5-6 and phases 2-3 of MJO. The composite analyses of atmospheric circulation show that there are different circulation patterns over the northwestern Pacific in different phases of the MJO, which will affect the typhoon generation. In phases 5-6 (2-3), the dynamic factor and convective heating patterns over western Pacific are favorable (unfavorable) for typhoon generation. Then， the comparing analyses of the 30-60 day low frequency kinetic energy in lower and higher levels of the troposphere show that the atmospheric intraseasonal oscillation over the northwestern Pacific has a clear impact on the typhoon generation. There is an evident positive (negative) anomaly area of 30-60 day low frequency kinetic energy in the more (less) typhoon years over the northwestern Pacific east of the Philippines, which means that strong (weak) atmospheric intraseasonal oscillation (ISO) over the northwestern Pacific is favorable (unfavorable) for typhoon generation. The analyses of 200 hPa velocity potential show that there is a clear divergence (convergence) pattern over the northwestern Pacific in the more (less) typhoon years, which is favorable (unfavorable) for typhoon generation. The modulation of the intraseasonal oscillation on the typhoon tracks over the northwestern Pacific is studied by observational data analyses. We classified the main classes of typhoon tracks into 5 types as straight west moving typhoons (I), northwest moving typhoons (II), recurving to Korea/west of Japan typhoons (III), landing on Japan typhoons (IV) and recurving to the east of Japan typhoons (V). Then the composite analyses of atmospheric low-frequency wind fields at 850, 500 and 200 hPa, corresponding to the typhoon forming date, for every typhoon track are completed. The analysis results of relationships between the low-frequency (ISO) wind fields and typhoon tracks have indicated that the typhoon tracks will be affected by wind pattern of the ISO. The low frequency positive vorticity belt (the maximum value line of cyclonic vorticity) associated with low-frequency cyclone (LFC) at 850 hPa is so closely related to the typhoon track, that the maximum value line (belt) of low frequency cyclonic vorticity can be an important factor to predicate the typhoon tracks over the northwestern Pacific. And the typhoon tracks will be also affected by the ISO circulation pattern at 200 hPa, particularly the strong low frequency wind associated with low frequency anticyclone (LFAC).
2014,40(4):389-399, DOI: 10.7519/j.issn.1000-0526.2014.04.001
Abstract:
Thunderstorm potential forecasting based on three ingredients has been widely accepted. This article aims to discuss some basical questions in operational forecast applications, and clarify some easily confused concepts. The content includes atmospheric instablility and convection, thunderstorms trigger mechanism and lifting and its relationship with snoptical weather system, how to deal with the three elements of the thunderstorm “enough”, the combination of pattern recognition and ingredients based forecasting methodology. Atmospheric instablility is one of the three ingredients of convection initiation, and it is also very important to thunderstorm short time forecasting and analysis. This paper discusses various mesoscale instability related to the thunderstorm, and inicates how to estimate the spatial and temporal evolution of CAPE. In addition, the definition and criterion for potential instability and symmetric instability are discussed profoundly.
Thunderstorm potential forecasting based on three ingredients has been widely accepted. This article aims to discuss some basical questions in operational forecast applications, and clarify some easily confused concepts. The content includes atmospheric instablility and convection, thunderstorms trigger mechanism and lifting and its relationship with snoptical weather system, how to deal with the three elements of the thunderstorm “enough”, the combination of pattern recognition and ingredients based forecasting methodology. Atmospheric instablility is one of the three ingredients of convection initiation, and it is also very important to thunderstorm short time forecasting and analysis. This paper discusses various mesoscale instability related to the thunderstorm, and inicates how to estimate the spatial and temporal evolution of CAPE. In addition, the definition and criterion for potential instability and symmetric instability are discussed profoundly.
2014,40(7):816-826, DOI: 10.7519/j.issn.1000-0526.2014.07.005
Abstract:
In term of precipitation data of 2400 stations from 1981 to 2010, annual, seasonal and monthly distribution and evolution characteristics of rainstorm were analyzed. The results show that the processes of rainstorm have been increased evidently since 21 century especially in the south of China, but the duration is relatively short. Rainstorm days have been increased, but the amount of precipitation is not as much as in 1990s. Variation trend of the annual (monthly) precipitation amount is in accordance with that of rainstorm days, but rainfall is averagely more while the rainstorm days are less during spring rainfall phase over the south of Yangtze River. Distribution of the maximum annual rainstorm days is very similar with that of the annual mean rainstorm days, revealing the feature of more in south and east but less in north and west. Maximum annual rainstorm days are more than double of annual average rainstorm days with multi centers due to the effect of topography. The months of maximum monthly rainstorm days over different regions of the same province are incompletely same as the result of the impact of different weather systems. Generally, rainstorm days have been increased since 2000, rainstorm begins earlier, ends latter and lasts longer than before. Nowadays, as the extreme rainfall events and secondary disasters happen frequently, it is conducive for the forecast of quantitative precipitation forecast (QPF) to learn the spatio temporal distribution and evolution features of rainstorm.
In term of precipitation data of 2400 stations from 1981 to 2010, annual, seasonal and monthly distribution and evolution characteristics of rainstorm were analyzed. The results show that the processes of rainstorm have been increased evidently since 21 century especially in the south of China, but the duration is relatively short. Rainstorm days have been increased, but the amount of precipitation is not as much as in 1990s. Variation trend of the annual (monthly) precipitation amount is in accordance with that of rainstorm days, but rainfall is averagely more while the rainstorm days are less during spring rainfall phase over the south of Yangtze River. Distribution of the maximum annual rainstorm days is very similar with that of the annual mean rainstorm days, revealing the feature of more in south and east but less in north and west. Maximum annual rainstorm days are more than double of annual average rainstorm days with multi centers due to the effect of topography. The months of maximum monthly rainstorm days over different regions of the same province are incompletely same as the result of the impact of different weather systems. Generally, rainstorm days have been increased since 2000, rainstorm begins earlier, ends latter and lasts longer than before. Nowadays, as the extreme rainfall events and secondary disasters happen frequently, it is conducive for the forecast of quantitative precipitation forecast (QPF) to learn the spatio temporal distribution and evolution features of rainstorm.
2011,37(10):1262-1269, DOI: 10.7519/j.issn.1000-0526.2011.10.009
Abstract:
Based on the daily precipitation data at 110 observational stations during 1961-2008 in South China， the climatic characteristics and variation of torrential rain days， rainstorm intensity and contribution which is in annual， the first and second flood seasons in South China were studied by using statistical and diagnostic methods， such as linear regression analysis， Mann Kendall test， wavelet analysis and the computation of trend coefficients. The results have shown that the annual mean torrential rain days have a decreasing trend from coastal regions to inland in South China in recent 48 years， the highest center is in Dongxing of Guangxi （14.9 d）， and the lowest center is in Longlin of Guangxi （3.2 d）. About 72% of the total torrential rain days occurred in the flood seasons with about 45% in the first season and 27% in the second season. The mean torrential rain days have increased faintly in annual， the first and second flood seasons in South China， but it is not obvious. There are the characteristics of interannual and interdecadal changes. The mean rainstorm intensity has increased faintly in annual and in the first flood season in South China. However, since 2005 it has become obviously. The mean rainstorm intensity has declined in the second flood season， but it is not obvious. The annual mean rainstorm contribution to the total rainfall has increased obviously， but the mean contribution is not obvious in the first and second flood seasons. The wavelet analysis has shown that the changes of torrential rain days， intensity and contribution which is in annual， the first and second flood seasons in South China have two significant periods of 2-3 a and 3-4 a.
Based on the daily precipitation data at 110 observational stations during 1961-2008 in South China， the climatic characteristics and variation of torrential rain days， rainstorm intensity and contribution which is in annual， the first and second flood seasons in South China were studied by using statistical and diagnostic methods， such as linear regression analysis， Mann Kendall test， wavelet analysis and the computation of trend coefficients. The results have shown that the annual mean torrential rain days have a decreasing trend from coastal regions to inland in South China in recent 48 years， the highest center is in Dongxing of Guangxi （14.9 d）， and the lowest center is in Longlin of Guangxi （3.2 d）. About 72% of the total torrential rain days occurred in the flood seasons with about 45% in the first season and 27% in the second season. The mean torrential rain days have increased faintly in annual， the first and second flood seasons in South China， but it is not obvious. There are the characteristics of interannual and interdecadal changes. The mean rainstorm intensity has increased faintly in annual and in the first flood season in South China. However, since 2005 it has become obviously. The mean rainstorm intensity has declined in the second flood season， but it is not obvious. The annual mean rainstorm contribution to the total rainfall has increased obviously， but the mean contribution is not obvious in the first and second flood seasons. The wavelet analysis has shown that the changes of torrential rain days， intensity and contribution which is in annual， the first and second flood seasons in South China have two significant periods of 2-3 a and 3-4 a.
2011,37(5):599-606, DOI: 10.7519/j.issn.1000-0526.2011.5.012
Abstract:
Using the diurnal snow data of 120 meteorological stations in Yunnan Province during 1961-2008, the temporal and spatial distribution characteristics and the trend of climatic change of the annual and monthly snow fall are analyzed. It is pointed out that the total trend of snow frequency and covering stations has been decreasing in Yunnan in the recent 50 years. And the annual snow frequency has declined at a mean rate of 4.5 times per year. The temporal trends of monthly snow frequency and covering stations are all negative. Moreover the reduction of snow frequency in December is the largest in magnitude, therefore, it is the most remarkable. And the reduction of snow stations in April is the largest. As far as the spatial change of the secular trend variation of annual snow frequency is concerned, the reduction of annual snow frequency is larger in Northwest Yunnan than in its northeast and east, where the reduction rate is 0.44 times per year. And the temporal changes of annual snowfall and depth of snow cover are studied, the results show that the secular trends of annual snowfall and the maximum depth of snow cover are all positive. This means that in the nearly 50 years the heavy snow frequency has increased over Yunnan Province.
Using the diurnal snow data of 120 meteorological stations in Yunnan Province during 1961-2008, the temporal and spatial distribution characteristics and the trend of climatic change of the annual and monthly snow fall are analyzed. It is pointed out that the total trend of snow frequency and covering stations has been decreasing in Yunnan in the recent 50 years. And the annual snow frequency has declined at a mean rate of 4.5 times per year. The temporal trends of monthly snow frequency and covering stations are all negative. Moreover the reduction of snow frequency in December is the largest in magnitude, therefore, it is the most remarkable. And the reduction of snow stations in April is the largest. As far as the spatial change of the secular trend variation of annual snow frequency is concerned, the reduction of annual snow frequency is larger in Northwest Yunnan than in its northeast and east, where the reduction rate is 0.44 times per year. And the temporal changes of annual snowfall and depth of snow cover are studied, the results show that the secular trends of annual snowfall and the maximum depth of snow cover are all positive. This means that in the nearly 50 years the heavy snow frequency has increased over Yunnan Province.
Mobile website
® 2026 All Rights Reserved
Supported by:Beijing E-Tiller Technology Development Co., Ltd.
ISSN:1000-0526 CN:11-2282/P