Southwest China is characterized by complex topography, diverse climatic conditions, and frequent drought events. To improve the representation of spatiotemporal characteristics in drought prediction, this paper developed a coupled Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) model using precipitation, air temperature, total solar radiation, wind speed, and other factors as inputs. The model was employed to analyze the spatiotemporal evolution characteristics and driving mechanisms of meteorological drought in Southwest China and project drought conditions in 2030. The results indicate that the CNN-LSTM model effectively captures the spatiotemporal variability of drought and exhibits high predictive accuracy, with a mean absolute error of approximately 0.17. Meteorological drought in Southwest China is jointly influenced by precipitation, air temperature, wind speed, and total solar radiation. The projected results for 2030 suggest that drought conditions are likely to expand toward low-hill and plain regions, accompanied by an overall increase in drought severity.
Studying the variation characteristics of regional drought in Shandong Province is of great significance for drought disaster prevention and mitigation. Using daily temperature and precipitation data from 123 national meteorological stations in Shandong Province, this study identified regional meteorological drought processes from 1961 to 2024 based on the MCI (Meteorological Drought Composite Index) and the monitoring and assessment method of regional drought process, and analyzed the spatio-temporal evolution features of these drought processes and their responses to climate warming. The results show that Shandong experienced 1-2 regional meteorological drought events annually with distinct seasonal patterns, predominantly concentrated in spring and summer, peaking from April to July. The frequency of consecutive spring-summer droughts reached 42%. Regional droughts exhibited clear geographical distribution: except for the eastern Jiaodong Peninsula and southeastern Linyi, the average drought duration exceeded 50 days in most remaining regions, with high-value zones concentrated in central Shandong and the southwestern side of Mount Tai. The northwestern Shandong regions showed a decreasing trend, with most areas experiencing a reduction of over 2 days per decade, while the southeastern regions demonstrated an increasing trend, particularly in the southern Jiaodong Peninsula and southeastern coastal areas, where the increase exceeded 2 days per decade. The frequency and affected extent of regional drought processes showed a slight upward trend, with the highest incidence observed from the 1980s to the early 21st century, followed by a marked decline and thereafter a renewed intensifying trend in recent years. Against the background of climate warming, precipitation across Shandong showed a decreasing-then-increasing trend, and the climate shifted from warm-dry to warm-wet conditions. Spatially, western Shandong tended toward warming-wetting, whereas eastern Shandong remained characterized by warming-drying. As warming accelerates, greater attention should be paid to guarding against more severe regional meteorological drought risks.
The flower bud stage is an important phenological stage marking the onset of fruit development of grapevines, and its interannual variation directly affects fruit quality and yield. This paper adopted phenological data of the Thompson Seedless grape (the dominant cultivar in Turpan City, Xinjiang) during its flower bud stage (from budburst to the end of flowering) from 1991 to 2024 and matching meteorological observation data. Climate trend rate analysis, correlation test and stepwise regression were applied to investigate the effects of climatic factors and last spring frost on the flower bud stage of the Thompson Seedless grape in extreme arid regions. The results show that under the background of warming and drying climate, the onset dates of all grape phenological stages exhibited advancing trends to varying degrees. The inflorescence stage advanced the most significantly at a rate of 5.02 d·(10 a)-1 (p<0.01), and the duration of grape flower bud stage exhibited a notable lengthening trend. The date of last spring frost advanced significantly at a rate of 6.43 d·(10 a)-1 (p<0.01). Climate change characteristics during the first decade of the 21st century were most prominent, with the flower bud duration extending by 21.94 d·(10 a)-1 (p<0.01), and frost damage frequency reaching once every 2 years. Climate warming and drying in extreme arid regions significantly increased the risk of frost damage during the grape flower bud stage. The day-of-year of each grape phenological stage was significantly correlated with variations in meteorological factors (p<0.05), and relative humidity and soil temperature were identified as the dominant ten-day climatic factors influencing grape flower bud phenology. From budburst to inflorescence stage, lower relative humidity and higher soil temperature resulted in earlier phenological development, whereas for the onset of flowering, higher relative humidity was associated with earlier development while lower relative humidity was associated with later development. Clarifying the variation characteristics of grape flower bud phenology and last spring frost in Turpan City can provide scientific guidance for grape cultivation in extreme arid regions.
Solar energy development is the key way for China to achieve dual carbon goals. China’s arid regions cover a vast territory and abound in solar energy resources. However, limited by scarce ground observation sites and insufficient in-situ data, the spatiotemporal variations and influencing factors of solar radiation in this region under climate warming since 1980 have not been fully clarified, which restricts the site selection of photovoltaic power stations and the development and utilization of regional solar energy resources to a certain extent. Therefore, based on solar radiation observations from 25 national ground radiation stations, cloud cover data from 218 national benchmark climate stations, and high spatiotemporal resolution ERA5 surface downward shortwave solar radiation reanalysis data from the European Centre for Medium-Range Weather Forecasts, this study investigates the variation characteristics and influencing factors of solar radiation in arid regions of China during 1980-2020. The results show that the daily mean solar radiation in the study area presents an increasing trend with obvious zonal characteristics. High-value radiation areas are mainly distributed in southern Xinjiang, western Hexi Corridor of Gansu, and parts of Qinghai, with radiation values exceeding 50 MJ·m-2, with obvious decadal variation characteristics. The annual daily mean solar radiation increased slowly during 1981-1990, rose rapidly from 1991 to 2010, and fluctuated mildly and tended to be stable overall during 2011-2020. The daily mean solar radiation for annual, cold-season, and warm-season periods in the arid regions of China exhibited significant abrupt changes around 1995, which mainly occurred in northern Xinjiang and eastern Inner Mongolia. Low cloud cover variation is the possible cause of solar radiation variation in the regions, but its influence mechanism is complex and does not follow a simple linear relationship, and may be related to the strong land-atmosphere interaction in the area, so more in-depth research is needed.
The Gobi desert region of China is a core area for current and future large-scale wind and photovoltaic power base construction. Conducting climate risk research on extreme events in this region is therefore crucial for the development of energy bases. Based on bias-corrected datasets from the Coupled Model Intercomparison Project Phase 6 (CMIP6), this study adopted existing quantitative relationships among dust disasters, extreme low temperatures, extreme high temperatures, and power generation efficiency losses and incorporated them into a climate risk assessment framework. The WRF-Chem model was used to project the climate risks of wind and solar power development in the Gobi desert region of China and its sub-regions under different emission scenarios from 2030 to 2060. The results indicate that the impact of a single dust event on wind and solar power generation efficiency in the Gobi desert region of China shows an increasing trend in the future. The hazard of extreme high temperatures generally increases, while that of extreme low temperatures generally decreases. Under the influence of extreme events, the overall climate risk for wind and solar resource development in the Gobi desert region of China tends to intensify. Specifically, dust-induced climate risks for wind power development increase significantly in Xinjiang and Qinghai, whereas those for solar power development increase significantly in Qinghai. Extreme high temperatures lead to significant increases in wind power climate risks in Xinjiang, Gansu-western Inner Mongolia, and eastern Inner Mongolia, and cause significant increases in solar power climate risks across all sub-regions. Under different emission scenarios, the wind power development risks caused by extreme low temperatures exhibit notable regional differences.
Under the background of climate warming, changes in extreme climate events have received increasing attention. In this study, three machine learning models, namely Random Forest, Extremely Randomized Trees, and Ridge Regression, were used to evaluate their capability in simulating extreme climate indices over China, including the warm day index (TX90p), cold night index (TN10p), maximum 1-day precipitation (RX1day), and maximum consecutive 5-day precipitation (RX5day). Their performances were compared with that of the traditional globally bias-corrected multi-model ensemble method to determine the optimal model schemes. Furthermore, the spatial distributions and temporal trends of the selected extreme climate indices during 2024-2100 under different emission scenarios, including SSP1-2.6, SSP2-4.5, and SSP5-8.5, were analyzed relative to the baseline period of 1961-1990. The results show that the machine-learning-based schemes can improve the simulation capability for extreme climate events to varying degrees and effectively reduce simulation biases. During 2024-2100, extremely warm events over China increase significantly under all emission scenarios, with the largest increase under the high-emission scenario SSP5-8.5, reaching approximately 52% relative to the baseline period. In contrast, extremely cold events decrease significantly, and the reduction becomes more pronounced with increasing emissions. Spatially, TX90p shows relatively large increases in North China, the middle and lower reaches of the Yangtze River, the Sichuan Basin, and parts of South China, while the increases are relatively smaller over the Qinghai-Xizang Plateau and some high-elevation areas in Northwest China. Extreme precipitation indices, including RX1day and RX5day, show increasing trends under all emission scenarios, with the most pronounced increase under the high-emission scenario. The enhancement of RX5day is generally stronger than that of RX1day. Northeast China and North China show the strongest responses in extreme precipitation, especially for RX5day, indicating higher sensitivity to climate warming.
Under the background of global warming, extreme precipitation events occur frequently in the Tuhai-Majia River Basin, which can easily trigger basin-wide floods and urban waterlogging. Focusing on the extreme precipitation event that occurred in this basin on 26 August 2024, this study comprehensively applies multi-source new observational data, including dual-polarization radar, wind profiler radar, millimeter-wave cloud radar, and raindrop disdrometer, to analyze its formation mechanism. The results show that the Yellow River cyclone followed an uncommon track and moved slowly, resulting in a prolonged duration of heavy rainfall. The water vapor originated from the East China Sea and the Bohai Bay, converging strongly over the basin and providing exceptionally high-humidity environmental conditions for the extreme precipitation. The establishment of low-level jets and the intrusion of mid-level cold air further enhanced the dynamic conditions.Dual-polarization radar and raindrop size distribution data clearly reveal the evolution characteristics of the microphysical mechanisms of the extreme precipitation. During the warm-front precipitation stage, the number concentration of raindrops of all sizes was generally higher, the precipitation was the strongest, with hourly rainfall exceeding 100.0 mm and the maximum minute-scale rainfall reaching 3.1 mm. During the cold-front precipitation stage, the raindrop size distribution width increased, but the number concentration of medium-to-large raindrops decreased. During the precipitation stage influenced by the cyclone center, the raindrop size distribution width was the largest, while the number concentration of raindrops was the lowest, resulting in the weakest rainfall intensity. Cloud radar observations show that strong updrafts promoted cloud development, while downdraft enhancement and speed spectral width variation reflected phase transitions and collision-coalescence processes, corresponding well with precipitation intensity.
In order to explore the triggering mechanism of convective rainstorms in the northwest plateau of Yunnan, this paper used multi-source detection data to diagnose a typical convective rainstorm process in Lijiang on June 24, 2024. The results show that: (1) This process was primarily influenced by the convergence zone between the subtropical high pressure and the Yunnan-Burma high pressure at 500 hPa, in conjunction with a shear line at 700 hPa, and combined with cold air moving southward, it formed an unstable atmospheric structure, providing sufficient dynamic condition and a favorable circulation pattern for heavy precipitation. (2) This heavy rainfall was caused by the β- and γ-mesoscale convective cells generated locally, corresponding to a maximum radar reflectivity of 59.5 dBZ, and there was velocity blurring on the velocity map, along with features such as vortices and weak mesocyclones. (3) The moisture of the heavy rainfall originated from the Bay of Bengal and the South China Sea. Under the influence of topographic convergence and lifting, a zone of strong moisture convergence existed from the central to the southeastern part of Lijiang, which corresponded well with the position of the shear line. The southwesterly jet stream transported warm and moist air and intensified low-level convergence, and the upper-level northwesterly jet stream strengthened upward motion through a suction effect. The intrusion of cold air further promoted the maintenance of convection. The strong water vapor accumulation combined with the dynamic uplift conditions of lower-level convergence and upper-level divergence was the main cause of the local heavy rainstorm. (4) The combined effect of the north-south trending mountain in Lijiang and the surface convergence line triggered and intensified convection through topographic lifting and airflow convergence, becoming an important factor in continuous triggering and maintenance of convection during this heavy rainfall event. An easterly airflow served as a reliable indicator for the occurrence of severe convection. (5) In the precipitation echoes of cumulus clouds, the negative flash peak occurred approximately 40 minutes ahead of the heavy precipitation, while in the precipitation echoes of stratiform mixed clouds, there was a phenomenon of a precipitation peak occurring earlier.
To further understand the circulation characteristics and formation mechanisms of summer extreme precipitation over the Qinling Mountains and surrounding areas, this study used precipitation data from automatic weather stations and ERA5 reanalysis data from the European Centre for Medium-Range Weather Forecasts from 2008 to 2025. Using spectral clustering method, the regional hourly extreme precipitation (RHEP) in the north and south of the Qinling Mountains (divided along 34°N) was classified into three and four typical circulation patterns, respectively. The dominant circulation features, water vapor, thermal, and dynamic conditions of summer RHEP were further investigated. The results are as follows: (1) At 200 hPa, all circulation types are influenced by divergence near the periphery of the South Asian High; at 500 hPa and below, these patterns are predominantly characterized by the combined effects of warm, moist air transport associated with the Western Pacific Subtropical High and short-wave troughs, shear lines, and similar features. Abundant moisture is transported by the peripheral airflow of the Subtropical High toward the Qinling Mountains and surrounding regions, producing pronounced convergence over regions of extreme precipitation. The combined effects of different circulation configurations and their interactions with complex terrain through orographic lifting and convergence intensify vertical motion, which is a key factor contributing to the spatial variability of RHEP occurrence. (2) In the lower troposphere, the periphery of the Subtropical High in each circulation pattern is characterized by anomalous southwesterly, southerly, or southeasterly airflow, with some types accompanied by anomalous northerly airflow on the northeastern side of the Qinghai-Xizang Plateau, favoring the confluence of cold and warm air and thereby promoting the occurrence of RHEP. (3) Water vapor diagnostic analysis reveals that multi-source water vapor from the Arabian Sea, Bay of Bengal, South China Sea, Western Pacific Ocean, and East China Sea converges over South China and East China, and is subsequently transported to the Qinling Mountains and its adjacent areas. In the areas north and south of the Qinling Mountains, most circulation patterns are predominantly influenced by water vapor input from the southern boundary of the lower-to-middle troposphere, while certain circulation types are predominantly influenced by water vapor input from the eastern boundary. (4) The peak frequency of RHEP for each circulation type occurs near specific topographic features, generally characterized by low convective inhibition (CIN) and high K-index values, indicating pronounced atmospheric instability, where even weak lifting mechanisms are sufficient to trigger the development of convection.
The Altay region of Xinjiang is a high-frequency area for cold waves in China, and cold wave weather poses a serious threat to its ice and snow economy. Based on the daily minimum temperature data from seven national meteorological stations in the region from 1960 to 2023, the frequency, intensity, and variation characteristics of regional cold waves were analyzed, and a comprehensive cold wave intensity index was constructed. The results show that: 1) A total of 833 regional cold waves occurred in Altay from 1960 to 2023, with an annual average of 13.0 times. The regional cold waves in winter (November to March of the following year) accounted for 72.5%, and regional cold waves were concentrated from late September to mid-March of the following year, accounting for 84.3%. The regional extremely strong cold waves in mid-February accounted for 52.9% of the total number of regional cold waves during this period, indicating that the activity of strong cold waves was highly concentrated in the middle and late winter. 2) The frequency of regional cold waves has shown a significant decreasing trend (-0.5 times·(10 a)-1, P<0.05). However, from 1996 to 2023, the number of regional extremely strong cold waves increased to an average of 4.0 times per year, approaching the peak level during the period from 1961 to 1990, presenting an evolution characteristic of “total reduction and enhanced extremality”. 3) The intensity of cold waves showed significant seasonal variations: in winter, extremely strong cold waves were predominant (accounting for 40.4%), while in spring and autumn, general cold waves were more common. The maximum 24-hour temperature drop of regional extremely strong cold waves averaged 10.9 ℃ (with a maximum of 18.5 ℃), and the cumulative temperature drop averaged 17.6 ℃ (with a maximum of 30.1 ℃). The temperature drop in winter was significantly higher than that in spring and autumn. 4) The comprehensive intensity index of cold waves in winter was 0.38, significantly higher than that in spring (-0.97) and autumn (-1.06). Over the past 30 years, the intensity of regional extremely strong cold waves has shown an upward trend, indicating that the intensity of extremely cold waves has further increased recently. 5) The synchronicity between the anomaly of winter precipitation and the anomaly of cold wave frequency reached 84.1% (P<0.01), and the synchronicity between the anomaly of cold wave frequency and the anomaly of the comprehensive intensity of cold waves reached 77.7% (P<0.01). This indicates that years with more precipitation in winter tend to have more frequent and stronger cold waves and a higher risk of disasters.
Stratiform-convective mixed cloud precipitation occurs frequently in the Liupanshan region. Understanding the vertical structure characteristics of clouds and the influence mechanism of mountainous topography is of great importance for improving local weather modification operations. Based on multi-source observations from Ka-band cloud radar, X-band weather radar, ERA5 reanalysis, and microwave radiometers, this study selected 20 light precipitation processes associated with stratiform-convective mixed clouds and systematically analyzed the spatial differences in cloud macroscopic structures, radar reflectivity characteristics, and dynamic and water vapor conditions. The results show that the cumulative precipitation amount and precipitation duration at Liupanshan Station are significantly higher than those at Dawan Station on the eastern slope and Longde Station on the western slope. The vertical structure of clouds shows a good spatiotemporal correspondence with precipitation and exhibits evident topographic differentiation. Cloud-top height, cloud thickness, and radar reflectivity at Liupanshan Station are significantly greater than those at the slope stations. During light precipitation processes of stratiform-convective mixed clouds in the Liupanshan Mountain area, the cloud vertical structure exhibits characteristics of both stratiform and convective clouds. During relatively stable precipitation periods, a distinct 0 °C bright band appears at 4-5 km altitude. The middle and lower layers below the bright band serve as the main precipitation contribution region, with a mean radar reflectivity of 4.1-9.2 dBZ. During periods of frequent precipitation, cloud thickness increases significantly, and the intensity of strong radar echoes in the lower layers is markedly enhanced. Typical precipitation case analysis indicates that the coupling of upper- and lower-level circulation and water vapor transport associated with the periphery of the subtropical high provides necessary dynamic and moisture conditions for precipitation. Topography modulates the cloud vertical structure through dynamic lifting and windward slope effects. The water vapor supply and cloud liquid water accumulation capacity on the eastern slope are significantly stronger than those on the western slope, which is an important cause of spatial differences in precipitation.
Analyzing the mechanism of the impact of extreme climatic conditions on vegetation productivity is of great significance for regional adaptation and the mitigation of the effects of climate change. Based on the Gross Primary Productivity (GPP) data of vegetation and the extreme climate indices of the same period, two-dimensional Copula joint analysis, the univariate linear regression method and Pearson correlation coefficient method, etc. were used to analyze the differences in the influence of four extreme climate indices on the GPP of different regions and different vegetation types in Shaanxi Province. The results show that: 1) From 1981 to 2018, 97.9% of the province’s vegetation GPP showed an increasing trend, indicating a steady improvement in carbon sequestration capacity, with the highest GPP growth rate observed in northern Shaanxi. 2) The spatial response of vegetation GPP to extreme climates showed remarkable spatial heterogeneity. Extremely low temperature and extreme drought strongly suppressed GPP in northern Shaanxi and Guanzhong regions, whereas extremely high temperature and extreme precipitation significantly had a prominent promoting effect on vegetation GPP in southern Shaanxi. 3) The monthly scale was the critical time window for the impact of extreme climate on vegetation productivity, and its influence intensity was much greater than that at the annual and growing season scales. 4) The effects of different extreme climate events on GPP of forest, grassland, and cropland exhibited clear vegetation specificity and seasonal dynamics.
To investigate the mechanisms underlying dust events in southern China, this study examined a representative dust intrusion that affected Hunan Province from 12 to 14 April 2025. ERA5 reanalysis data from the European Centre for Medium-Range Weather Forecasts (ECMWF), wind-profiler radar observations, air-quality monitoring data, and the Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) backward-trajectory model combined with topographic data were used to examine the role of Hunan’s terrain in pollutant transport and retention during the dust event. The results are as follows: (1) Although Hunan lies in the humid region of southern China, it can be significantly affected by long-range dust transport. The dust originating from Mongolia and southern Xinjiang moved southward along the northwest path, forming the first province-wide dust event since 2000, with pollutants dominated by exogenous coarse dust particles. (2) The dual “channel-retention” effect of Hunan’s horseshoe-shaped terrain was the major cause of this multi-day dust event. The northern plain accelerated the southward spread of dust, while the surrounding mountains blocked airflows and prolonged pollution retention in southern Hunan, and this effect may be common in complex-terrain regions of southern China. (3) A conceptual model for dust intrusion into Hunan was established, and the key forecasting and early-warning indicators were identified as follows: strong dust storms lasting for more than 2 d over northwestern China and adjacent regions, the coupling of a Mongolian cyclone with a surface cold front, and widespread northerly winds exceeding Beaufort Force 5. Under these conditions, dust aerosols were transported over long distances along a westerly pathway, reached Hunan within 48-72 h, and were rapidly removed after the prevailing wind shifted to a southerly direction.
Atmospheric aerosols have significant impacts on climate change and regional air quality. Based on 1 km resolution aerosol optical depth (AOD) data from MODIS MAIAC Collection 6.1 product during 2003-2023, this study systematically explored the spatiotemporal variation patterns of AOD in Ningxia and its influencing factors by comprehensively applying methods such as Theil-Sen trend analysis, Mann-Kendall significance test, and Spearman correlation analysis. The results are as follows: (1) The AOD showed a distribution pattern of “high in the north and west, low in the south and east”. The AOD high-value areas are mainly distributed in the urban belt along the Yellow River and the Qingshui River Basin where human activities are intensive. (2) The annual average AOD presented a significant decreasing trend at a rate of -0.003 a-1, with a particularly marked decline after reaching the peak in 2011. Over the past 21 years, the AOD has decreased significantly in 75.12% of the entire region, and the area with the most significant decrease (-0.006 a-1) is highly consistent with the densely populated urban belt along the Yellow River, reflecting the effect of anthropogenic emissions reduction. (3) The AOD followed the pattern of “high in spring (0.33), falling in summer (0.27), low in autumn (0.20), and rising in winter (0.25)”, and the spring peak is mainly dominated by dust activities. (4) The interannual variation of AOD is synchronized with the reduction of anthropogenic particulate matter emissions, while on the seasonal scale, there is a dynamic shift of dominant factors, namely a significant positive correlation with wind speed in spring, a highly significant negative correlation with vegetation index in summer, and a negative correlation with wind speed in autumn.
Based on APS-3321 (Aerodynamic Particle Sizer), real-time and continuous measurements of particle size distributions in the 0.5-20 μm range of dust aerosols were conducted in a desert source area of Jiayuguan during August 2024. The number and mass concentrations and their spectral distribution characteristics of dust aerosol were comprehensively analyzed under different weather patterns by using threshold values of mass concentration method. The results show that the number concentrations of dust particle under clear-sky conditions presented a unimodal distribution, with the peak particle size of 0.626 μm. The number concentrations of dust particles under floating dust and blowing dust conditions exhibited a bimodal pattern, and the main peak particle sizes were both located at a submicron of 0.626 μm, while the secondary peak diameters were 1.114 μm and 1.286 μm respectively. By contrast, the number concentrations of dust aerosols under heavy dust events also displayed a bimodal distribution, and the main peak particle size was located at a coarse-mode of 1.486 μm, while the secondary peak was 0.583 μm, which differed from those under floating and blowing dust scenes. This indicated that heavy dust storm events contribute significantly to the number concentration of large coarse-sized particles in the desert source area of Jiayuguan. However, the mass concentrations of dust particles under clear-sky, floating and blowing dust conditions all showed a bimodal structure, and the main peak sizes were 4.371, 3.523 and 3.278 μm respectively, and the secondary peaks occurred at 19.810, 15.960 and 15.960 μm. By contrast, the mass concentrations of dust particles under heavy dust events exhibited a trimodal distribution, with the peak particle sizes of 4.371 μm, 6.264 μm and 13.820 μm. During the whole period, the average mass concentration under floating dust condition varied from 80 to 200 μg·m-3, and the corresponding maximal concentrations under blowing and heavy dust events were 600 μg·m-3 and 2 400 μg·m-3, respectively, which occurred at 06:00-09:00 and 09:00-18:00. The occurrences of heavy dust events led to a significant decrease in the percentage of dust particles at 0.5-1.0 μm from 80.9% to 39.9%, but they didn’t remarkably alter the percentage proportion of the mass concentration of coarse-mode dust particles in Jiayuguan.
The variation characteristics of gust coefficient and basic wind pressure are of great significance for gust forecasting, wind energy resource development, and the design of wind parameters for large-scale engineering. Based on hourly wind observations from 24 national meteorological stations over Shandong Peninsula during 2015-2023, together with annual maximum wind speed data from station establishment to 2023, the spatiotemporal variations of gust coefficient， 50-year return level wind speed and basic wind pressure were analyzed. The results show that the gust coefficient exhibits significant spatial variability across Shandong Peninsula, ranging from 1.380 to 1.998. Seasonally, it is relatively higher from August to December and lower from January to March. The gust coefficient shows a pronounced diurnal variation, with maximum values around noon and minimum values before dawn. In addition, the gust coefficient generally decreases with increasing wind speed. The 50-year return level maximum wind speed ranges from 18.6 to 33.1 m·s-1. Relatively lower values are observed in parts of the western and eastern regions of the peninsula, while higher values occur in the northeastern coastal areas, including Weihai, Chengshantou, Haiyang, and Laoshan. The basic wind pressure ranges from 0.21 to 0.68 kN·m-2, with lower values (<0.30 kN·m-2) in parts of the western and eastern regions, and higher values (>0.41 kN·m-2) along the northeastern and southern coasts. The maximum value is observed at Chengshantou station, reaching 0.68 kN·m?2.
To deepen the understanding of the topographic influence mechanism of lightning-ignited fires in the Greater Khingan Range forest area of Inner Mongolia, identify the key driving factors of fire suppression demand, and provide a scientific basis for the monitoring and early warning of lightning-ignited fires as well as the optimal allocation of fire suppression resources in the forest area, based on historical lightning-ignited fire data and high-precision topographic data from 2015 to 2024 in the key state-owned forest area of the Greater Khingan Range of Inner Mongolia, this study systematically analyzed the spatial distribution characteristics of lightning-ignited fire points across multiple topographic factors including elevation, slope gradient, slope aspect and topographic relief. The XGBoost (eXtreme Gradient Boosting) algorithm optimized by the Newton-Raphson-Based Optimizer (NRBO) combined with the SHAP (SHapley Additive exPlanations) method was adopted to model and analyze fire suppression demand, and the contribution degrees of burned area and various topographic factors to the number of fire suppression personnel were revealed. The results show that lightning-ignited fire points in the study area present a gradient characteristic of “concentrated at medium-low elevations and sparse at high/low elevations” in terms of elevation; 83.06% of lightning-ignited fires occur in the medium-low elevation zone of 500-1 000 m. The regions with the highest frequency of lightning-ignited fire points are concentrated between 52.5°N-53.0°N and 120.5°E-122.5°E, while the absolute high-value areas of burned area appear at more southerly latitudes. Lightning-ignited fire points are mainly distributed in medium-slope areas of 5°-35°,and the slope range of 2° to 5° accounts for 37.83% of the total burned area. The medium-relief terrain with a relief value of 75 to 200 m serves as the “core interval” and “optimal” environment for lightning-ignited fires, contributing to 70.97% of such fires. Sunny slopes such as the south, southwest, and southeast slopes are high-risk areas for lightning-ignited fires, accounting for a relatively high proportion of the total fire occurrences. Burned area is the dominant factor in predicting the number of fire suppression personnel, and its contribution is significantly higher than that of topographic factors such as elevation, relief, slope gradient, and slope aspect. The distribution of the model’s prediction errors showed a significant leptokurtic pattern, with the peak closely around the zero error line. The R2 value of the model’s prediction on the test set reached 0.723 9, and the prediction interval coverage probability reached 84.7%.
To meet the demand for high-precision monitoring of cloud physical characteristics in artificial weather modification operations in Inner Mongolia region, this study utilizes multi-source data from Fengyun-4A/B satellites (FY-4A/B), CloudSat/CALIPSO, Himawari-8, Moderate Resolution Imaging Spectroradiometer (MODIS), and the fifth-generation reanalysis data of the European Centre for Medium-Range Weather Forecasts (ERA5), and through preprocessing steps of radiometric calibration and geometric correction for FY-4 Satellite data, then applies artificial intelligence algorithms such as random forest to construct a cloud physical characteristic parameter inversion algorithm based on FY-4 Satellite. This algorithm achieves cloud detection and the retrieval of cloud top height, cloud top temperature, supercooled layer thickness, cloud optical thickness as well as cloud effective particle radius. Furthermore, the algorithm accuracy verification and adaptability analysis are conducted, and an operational platform for cloud parameter retrieval and a data release website are developed, forming a complete technical chain of “data-algorithm-platform-application”. The results show that the overall accuracy of the self-developed cloud detection algorithm is 90.07%, which is 1.11% higher than that of the official algorithm of the FY-4 Satellite; the determination coefficients (R2) of the inversion model for cloud top height and cloud top temperature are 0.928 and 0.922 respectively, and the root mean square errors are 0.901 km and 5.963 K respectively; the R2 of the inversion model for ice clouds and water clouds optical thickness are 0.693 and 0.582 respectively, and the R2 of the inversion of effective particle radius is 0.562 and 0.809, respectively.
To further improve the scientific basis and effectiveness of aircraft artificial precipitation enhancement operations in Ningxia and enhance regional water resource utilization efficiency, this study conducts a systematic analysis of an aircraft precipitation enhancement event that occurred in Ningxia during the spring of 2025. The analysis is based on products from the Cloud Precipitation Explicit Forecast System (CPEFS) issued by the Weather Modification Center of the China Meteorological Administration, combined with multi-source observationals including radar, satellite, automatic weather station precipitation, raindrop spectrometer and aerosol particle size spectrometer, and reanalysis datasets. The study focuses on operational conditions, operational rationality, and the physical response of the cloud-precipitation system to seeding. The results indicate that this event was a typical systematic precipitation process, with high supercooled liquid water content primarily located in southern Ningxia. During the aircraft seeding operations, the temperature of the seeding layer remained generally below -7 °C, while the relative humidity was consistently maintained above 80%, indicating favorable conditions for cloud seeding. After the implementation of seeding, precipitation in the operation area and its downwind region increased significantly. Precipitation intensity, maximum raindrop diameter, and radar reflectivity factor all exhibited increasing trends, accompanied by an overall enhancement in radar echo intensity. These changes reflect increases in both the size and concentration of hydrometeor particles, as well as enhanced development of the cloud-precipitation system. Meanwhile, surface aerosol number concentration decreased during the precipitation development stage, indicating a pronounced wet scavenging effect of precipitation on aerosol particles.
An extreme gale event occurred in Urumqi from April 29 to May 3, 2022. The duration of gale-force winds in the southern suburbs lasted for 65 h, representing the longest duration in nearly a decade. Based on meteorological observation data, reanalysis data, and high-resolution numerical simulations, the three-dimensional structure and triggering mechanism of this gale event were systematically analyzed. The results show that the eastward movement of the Mongolian High, together with the pressure decrease ahead of the frontal system in northern Xinjiang, established and maintained a pressure pattern characterized by higher pressure to the south and lower pressure to the north across the Tianshan Mountains. The pressure difference extended from the surface to the upper levels, with a top height reaching about 4 000 m, providing a stable dynamical condition for the continuous acceleration of airflow within the canyon. Driven by a strong pressure gradient force, the airflow entered the Tianshan canyon from the southern end and was continuously accelerated under the effects of topographic forcing and funneling, eventually developing into a typical topographic low-level jet. As the airflow entered the canyon, gravity waves were triggered by topographic lifting, and the low-level jet subsided along the leeward slope with the propagation of gravity waves. The strong wind zone gradually expanded toward the near-surface layer, thereby triggering a downslope windstorm. In addition, during the gale period, persistent stable stratification existed above the leeward slope and was accompanied by evident subsidence, which effectively suppressed vertical mixing and enhanced downward momentum transport, playing an important role in the development and maintenance of the downslope windstorm.
To enhance the forecasting and early warning capability for short-duration heavy rainfall (HHR) and optimize disaster prevention and mitigation planning in central Hunan, based on terrain partitioning (Regions I-IV) and hourly precipitation observations, the multi-scale spatiotemporal distribution characteristics and evolution of HHR during the flood seasons (April-September) in central Hunan from 2016 to 2024 were systematically analyzed. The results show that the spatial distribution of short-duration heavy rainfall during the flood season in central Hunan exhibits significant differences. High-value areas of mean precipitation amount and frequency are mainly located in the northeastern part of Region I and the northwestern and southwestern parts of Region IV, among which the northwestern part of Region IV is the center of extreme precipitation intensity, while the southern parts of Regions II and III are characterized by relatively low values. On the interannual scale, precipitation amount and frequency of short-duration heavy rainfall vary consistently and show significant fluctuations, with the largest values occurring in 2024 and the smallest in 2022, whereas the interannual variation of precipitation intensity is relatively small. On the ten-day scale, precipitation amount and frequency both exhibit a unimodal distribution, with the peak appearing in late June, while precipitation intensity shows a multimodal variation pattern. The diurnal variations of precipitation amount and frequency both exhibit a bimodal structure. The primary peak occurs in the early morning over Region IV (mountainous area), whereas the primary peak appears in the afternoon over Regions I, II, and III (plain, basin, and hilly areas). The peak of precipitation intensity generally lags behind those of precipitation amount and frequency. The spatial distribution of diurnal peaks in precipitation amount and frequency is highly consistent, exhibiting a west-to-east sequential propagation pattern (Region IV→Region III→Region II→Region I). A clear northeast-southwest-oriented transition zone is formed between Region IV and Region III along the Xuefeng Mountains, indicating that the diurnal variation of short-duration heavy rainfall in the region is dominated by different precipitation mechanisms. The diurnal peak of precipitation intensity mainly occurs in the afternoon, and the regional consistency of its spatial distribution is relatively weak.
The study of hail days over the Qinghai-Xizang Plateau plays a crucial role in regional hail disaster mitigation and response strategy formulation. Based on observational data from 89 meteorological stations over the Qinghai-Xizang Plateau and ERA5 reanalysis datasets from the European Centre for Medium-Range Weather Forecasts (ECMWF) in the warm season (May-September) from 1969 to 2024, this study examined the spatiotemporal variation of hail days and the causes of the reduction of hail days using methods such as linear trend estimation, concentration degree and concentration period analysis, and correlation analysis. The results show that the number of hail days has decreased significantly at a rate of 1.7 d·(10 a)-1 in the warm season over the Qinghai-Xizang Plateau from 1969 to 2024, and has remained consistently below average, especially since 2008. Hail events are frequent in the high-altitude regions of the central Qinghai-Xizang Plateau, while low-frequency areas are scattered across the Qaidam Basin, the Hehuang Valley, and the plateau margins. Moreover, the rate of decrease of hail days accelerates with increasing altitude. Over the past 56 years, hail concentration degree has increased, and the concentration period has been delayed over the Qinghai-Xizang Plateau. Especially since the beginning of the 21st century, the concentration period has been significantly delayed further, and interannual variability has increased. Hail events have become more concentrated in low-altitude areas, whereas the concentration period occurs later in high-altitude regions. Under climate warming, enhanced mid-to-upper tropospheric warming has significantly elevated the 0 °C and -20 °C isotherms. With the -20 °C isotherm rising more rapidly, the upward shift of these critical levels has compressed the vertical growth space of hail embryos and has intensified hail melting during their descent. Concurrently, the marked rise in near-surface minimum temperature and the reduction in diurnal temperature ranges have impaired the atmospheric energy accumulation process. The number of days with moderate dew point depression (5-15 °C) has decreased, thereby further suppressing the favorable surface conditions required for hail occurrence. Since the 21st century, the atmospheric profile over the Qinghai-Xizang Plateau has shown strong low-level moistening but weak upper-level moistening. In the mid-lower troposphere, the pseudo-equivalent potential temperature has increased significantly, with the greatest amplification in the lower layers near the hail-freezing layer (600-500 hPa). In contrast, the atmospheric stability in the hail growth layer (400-300 hPa) has tended to increase. These combined factors have suppressed hailstorm development. Overall, the reduction in warm-season hail over the Qinghai-Xizang Plateau results from the coordinated changes in atmospheric thermal structure, stability, and moisture conditions under climate warming.
To reveal the spatiotemporal characteristics of cloud water resources in Xinjiang, this study systematically analyzes the variations in cloud amount and cloud water parameters (cloud water content, liquid water content, and ice water content) using satellite data from CloudSat and MODIS for 2007-2019 and reanalysis data from ERA5 for 1979-2024. The results show that all three datasets consistently capture the primary spatial pattern of cloud water parameters, characterized by higher values in mountainous areas and lower values in basins, with ice water content generally exceeding liquid water content. ERA5, with its higher spatial resolution, can more clearly represent terrain-related local features. MODIS shows good agreement with ERA5 in spatial distribution, although its cloud water content is systematically overestimated. CloudSat exhibits differences in spatial distribution in some local regions, while its overall magnitude is comparable to that of ERA5. Cloud water content is higher in summer and lower in winter, with spring and autumn serving as transitional seasons. From 1979 to 2024, cloud amount and cloud water content show significant decreasing trends in the northern Xinjiang Basin and the northern part of the southern Xinjiang Basin, while a significant increasing trend is observed in the central and eastern Kunlun Mountains.
Gansu Province is one of China’s three national-level seed maize breeding bases, with the Hexi Corridor being a major production area for seed maize. Under climate warming, frequent regional high temperature events seriously threaten the high and stable yield of seed maize. Therefore, heat disaster prevention for seed maize is crucial to national seed industry development and food security. Using daily and hourly temperature data from 14 meteorological stations in the main seed maize producing areas of the Hexi Corridor, Gansu Province during 2004-2025, combined with seed maize yield data, this study determined high temperature damage indicators for seed maize, analyzed the spatio-temporal distribution characteristics of heat damage, constructed a high temperature damage risk assessment index, and carried out risk zoning. The results show that the heat damage indicators considering daily high temperature duration have a stronger correlation with meteorological yield. The high temperature days, defined as days with a daily cumulative high temperature （≥35 ℃） duration greater than or equal to 4 hours, and the corresponding high temperature accumulated heat were selected as the hazard factors of high temperature heat damage. From 2004 to 2025, both high temperature days and high temperature accumulated heat generally showed an increasing trend with obvious interannual variations in the main seed maize production areas, and their spatial distribution decreased from the northwest to the southeast. Based on the high temperature heat damage risk assessment index, risk zoning was carried out. The extremely high-risk and high-risk areas of seed maize heat damage in the Hexi Corridor are mainly concentrated in Guazhou County and Jinta County of Jiuquan City, Ganzhou District, Gaotai County and Linze County of Zhangye City, Jinchuan District of Jinchang City, and Minqin County of Wuwei City. Counties and districts with large planting areas should strengthen targeted prevention measures against high temperature heat damage.
Against the backdrop of global warming and intensifying human activities, extreme drought and flood events have shown a trend of increasing frequency and intensity. However, research on sub-seasonal drought-flood abrupt alternation events remains relatively limited. In light of this, this study utilizes daily precipitation data from 20 national meteorological observation stations in Ningxia from 1961 to 2023 and employs the multi-threshold run theory to propose a method for sub-seasonal drought-flood abrupt alternation events, which is subsequently validated. Based on this method, drought-flood abrupt alternation events in Ningxia from April to October during 1961-2023 are statistically analyzed, with a focus on examining the spatiotemporal evolution characteristics in terms of event frequency, transition timing, and intensity. The results indicate that the proposed identification method performs well in detecting both drought and flood events and effectively captures the transition process from drought to flood. Since the 1990s, the frequency of drought-flood abrupt alternation events in Ningxia has increased significantly, reaching its peak in the 2010s. The occurrence time of turning points, ranked from most to least frequent, are the sixth pentad of July, the third pentad of July, the fourth pentad of October, the first pentad of July, and the fifth pentad of September. The spatial distribution of event frequency is uneven, showing a clear regional pattern of increase from south to north. Severe drought-flood abrupt alternation events mainly occurred in the 1970s and the 2000s and were largely concentrated in the northern Yellow River irrigation area. Although this region has the lowest annual precipitation, both light and moderate drought-flood abrupt alternation events show an increasing trend.
Under global warming, compound events of heat and drought pose a far greater threat to agricultural production than individual extremes. Based on daily precipitation and maximum temperature data from 92 meteorological stations in Sichuan Province druing 1981-2022, this study employed the Standardized Precipitation Evapotranspiration Index (SPEI) and the Standardized Temperature Index (STI) to characterize drought intensity and high temperature intensity， respectively. A joint probability distribution model of heat and drought intensities was constructed using the Copula function, and the spatiotemporal characteristics of compound heat-drought events during the critical rice growth periods (booting-heading and heading-maturity) were analyzed systematically. The results indicated that: 1) The Generalized Extreme Value (GEV) distribution was identified as the optimal marginal distribution for both drought intensity and heat intensity at most stations across the two growth stages, with better goodness-of-fit observed at the booting-heading stage than the heading-maturity stage. 2) The Gumbel Copula was selected as the optimal joint distribution model for the majority of stations, revealing a tendency for heat and drought events to occur in tandem. 3) In the seven rice-planting subregions, compound events characterized by mild heat and mild drought exhibit the highest occurrence probability, with joint return periods ranging from 9 to 20 years during the booting-heading stage. Spatially, the central Sichuan Basin, southern Sichuan Basin, and basin peripheral areas are high-frequency regions. The booting-heading stage was the critical growth period with a higher occurrence probability of compound stress. 4) Compared with the period 1981-2000, both the frequency and intensity of compound heat-drought events increased across all seven rice-planting subregions during 2001-2022, and the increases in heat frequency and intensity were more pronounced than drought. In summary, over the past two decades, the risk of compound heat-drought events during the rice growth period in Sichuan Province has intensified, with heat-related risk rising at a particularly rapid rate.
Beijing is the political, economic, cultural, and international exchange center of China, where major state events are fre-quently held in October. Cold waves and their associated weather conditions can significantly affect the meteorological support for these events, making it important to investigate their formation mechanisms. Based on ERA5 reanalysis data from the European Centre for Medium-Range Weather Forecasts (ECMWF) and conventional meteorological observations, this study comparatively analyzes two re-gional cold wave events that occurred in Beijing during 16-18 October 2022 (cold wave I) and 18-20 October 2024 (cold wave II), with a focus on their circulation characteristics and formation mechanisms. The results show that both cold waves developed under the back-ground of relatively high temperatures in the preceding period. The cold wave I was mainly characterized by strong winds, whereas the cold wave II featured more intense cooling accompanied by strong winds and precipitation. The intensity of cold advection was not syn-chronized with the magnitude of temperature decrease, and under clear-sky and weak-wind conditions, diabatic processes contributed more significantly to the temperature drop. A strong surface cold high and the 3-h pressure tendency gradient were the key dynamic fac-tors responsible for the strong winds, while persistent subsidence in the boundary layer showed a positive correlation with both the in-tensity and duration of strong winds. These results provide scientific support for the forecasting and early warning of cold waves in Bei-jing during October and for meteorological services for major events.
From July 27 to August 2, 2023, an extreme heavy rainfall event occurred in North China (referred to as the “23·7” event), and significant discrepancies are found among different numerical weather prediction models. In this study, three global models and seven regional models are selected to systematically evaluate the forecast performance for this heavy rainfall event under complex terrain conditions, based on a categorical analysis of 3-hourly accumulated precipitation, combined with precipitation verification and surface wind field analysis. The results show that the global models perform relatively well in predicting precipitation of ≥0.1 mm and ≥1 mm, but with significant overforecasting, and exhibit obvious underprediction for precipitation of ≥10 mm. The regional models exhibit superior overall forecasting performance to global models, especially for precipitation of ≥10 mm, for which their forecasting skill is significantly higher. Model resolution is found to exert a considerable influence on the prediction of rainfall at different intensities. Better performance is achieved by high-resolution models compared with low-resolution models, particularly for precipitation of ≥5 mm, though exceptions do exist. With regard to the fitting of precipitation frequency distribution against precipitation intensity, regional models are demonstrated to be overall distinctly superior to global models. Analysis based on the Froude number (Fr) reveals the synergistic mechanism between terrain and wind fields. Overestimation of wind speed or underestimation of terrain height in models leads to the misrepresentation of flow-around as orographic updraft, inducing biases in the predicted precipitation location. In contrast, the characteristics of orographic lifting or flow around can be better preserved in regions with reasonable local Fr values.
The changes in snow cover on the Qinghai-Xizang Plateau have significant impacts on weather, climate, and hydrological processes. In the context of global warming, the climate change in the complex terrain area in the eastern part of the plateau shows altitude dependence, but the characteristics of snow cover changes with altitude on the plateau are still unclear. The spatio-temporal variation characteristics of snow cover frequency in the eastern Qinghai-Xizang Plateau from 2003 to 2021 and its main influencing factors were analyzed by using the daily cloud-free satellite remote sensing snow cover dataset and gridded meteorological data. The results indicate that: 1) High values of snow cover frequency are mainly located in the high-altitude mountainous areas in the southern of the study area. In the southern part of the eastern plateau in spring, the frequency of snow cover is higher than in winter, while in the inland, the frequency of snow cover is higher in winter than in spring. The frequency of snow cover generally increases first and then stabilizes with respect to altitude, reaching its peak at about 6 000 meters. Above 4 000 meters, it shows a bimodal pattern, with peaks occurring in November and from March to April; below 4 000 meters, it follows a unimodal pattern, with the peak in January. 2) Except for the significant decrease trend observed in autumn snow cover, the overall change trends of snow cover in spring, winter and the annual average are not significant. However, the snow cover in all periods in areas above 6 000 meters in altitude decreases significantly. 3) Snow cover is generally negatively correlated with temperature, being significantly in winter and spring. It is positively correlated with precipitation, with the strongest and most extensive correlations in winter. Significant positive correlations are also observed in autumn in the southern, inland, and the Qilian Mountain regions, and in spring in the southeastern and northeastern mid-to-high altitude areas. 4) Compared with past studies based on shorter time series of original MODIS snow cover data, the snow cover variation characteristics reflected by the long-term cloud-free dataset show distinct differences and greater reliability.
Evapotranspiration acts as an intermediate link in the terrestrial water and energy cycles while also serving as a crucial nexus connecting soil, vegetation, and atmospheric processes. Investigating changes in evapotranspiration holds significant scientific importance for the scientific management of water resources, addressing challenges posed by climate change, and ensuring regional eco-hydrological security. This study utilizes observational data from representative sites in the Yellow River Basin, namely the source region, the Hetao region, and the downstream region, which correspond to Haibei Station, the Semi-Arid Climate and Environment Observatory of Lanzhou University, and Yucheng Station. The purpose is to evaluate the performance of the 6th Phase of the Coupled Model Intercomparison Project (CMIP6) in simulating evapotranspiration across different regions of the Yellow River Basin. Based on this, the spatiotemporal variations of evapotranspiration across different regions of the Yellow River Basin under historical (1980-2014) and future (2026-2100) scenarios are analyzed using the multi-model ensemble mean results. The results show that the evapotranspiration derived from the CMIP6 multi-model ensemble mean exhibits good correlation and high Taylor skill scores in the source region, the Hetao region, and the downstream region of the Yellow River. Therefore, it is considered a suitable tool for investigating the spatiotemporal distribution of evapotranspiration. Furthermore, the annual evapotranspiration derived from the CMIP6 multi-model ensemble mean shows an increasing trend, with the highest change rate of 3.45 mm·(10 a)-1 identified in the source region of the Yellow River, while the increasing rates in the Hetao and downstream regions are relatively slower. Evapotranspiration shows increasing trends in spring and winter across the Yellow River Basin. However, the trends in summer and autumn exhibit spatial heterogeneity, with evapotranspiration rising in the source region but decreasing in the Hetao and downstream regions. Notably, the downstream region shows pronounced decreasing trends, with rates of 1.13 mm·(10 a)?1 and 0.73 mm·(10 a)?1 in summer and autumn, respectively. Under all future scenarios, evapotranspiration is projected to continue increasing throughout the 21st century in the source region, Hetao, and the downstream regions of the Yellow River Basin, peaking around the year 2100. As anthropogenic emissions increase, the rate of evapotranspiration increase is expected to accelerate further, with the most significant acceleration projected for the downstream region.
The source region of the Yellow River, located in the northeastern part of the Tibetan Plateau, is the largest runoff-producing area in the Yellow River Basin. Studying the future runoff variation characteristics in this region is of great significance for the rational allocation and efficient utilization of water resources in the Yellow River Basin. This study utilized observed monthly runoff data from the Tangnaihai Station during 1976-2018, gridded meteorological observation datasets, the Soil and Water Assessment Tool (SWAT) model, and four machine learning algorithm models to simulate and analyze historical runoff at Tangnaihai Station in the source region of the Yellow River. Through the evaluation of simulation results and comparison of the performance of different models, the Random Forest (RF) model was identified as the most suitable for runoff prediction in this region. Based on the RF model and meteorological data from six models of the Coupled Model Intercomparison Project Phase 6 (CMIP6) under different emission scenarios (SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5), future runoff at Tangnaihai Station was projected and analyzed. The results show that the runoff at Tangnaihai Station in the source region of the Yellow River simulated by the SWAT model and RF model was in good agreement with the observations. The RF model achieved a coefficient of determination (R2) and Nash-Sutcliffe Efficiency (NSE) both above 0.83 during the training period, while the SWAT model achieved R2 and NSE values above 0.70 during both the calibration and validation periods. Moreover, the bias of these two models is relatively small compared with other models. Under future climate scenarios, annual precipitation in the source region of the Yellow River shows a gently fluctuating upward trend. The precipitation trend under the SSP1-2.6 scenario is relatively small, with an increase rate of 2.00 mm per decade, while under the SSP5-8.5 scenario, precipitation increases at a rate of 19.52 mm per decade, the fastest among the four emission scenarios. Under different emission scenarios, future runoff displays significant fluctuating variations. The multi-year average runoff under low emission scenarios (SSP1-2.6 and SSP2-4.5) is 673.49 m3·s-1 and 670.37 m3·s-1, representing increases of 3.37% and 2.90%, respectively, relative to the historical period. In contrast, under high emission scenarios (SSP3-7.0 and SSP5-8.5), the multi-year average runoff is 646.68 m3·s-1 and 623.08 m3·s-1, representing decreases of 0.74% and 4.36%, respectively, compared with the historical period.
To assess the vegetation ecological quality in the ecologically fragile region of the upper Yellow River and evaluate the effectiveness of national key ecological restoration projects, this study selected the Baiyin Section of the Yellow River Basin as the research area and delineated typical ecological functional areas, including the sandstorm control area, the irrigation area along the Yellow River, and cropland-to-forest area. Based on multi-source remote sensing data from 2000 to 2020, the vegetation ecological quality index (EQI) was constructed, and trend analysis and spatial statistical methods were employed to systematically reveal the spatiotemporal evolution characteristics and future variation trends of EQI. The results show that, during the past 21 years, the EQI of the study area exhibited a significant upward trend; the irrigation area along the Yellow River had the highest mean EQI (49.8), while the grain for green area showed the fastest growth rate (0.63 a?1), indicating that ecological projects played a key driving role in improving regional ecological quality. The EQI presented a gradient pattern of “high in the south and low in the north” accompanied by an “ecological island” phenomenon spatially; high-value areas were mainly distributed in the Yellow River irrigation area and mountainous forest regions such as Hasi Mountain and Tiaoshan Farm, whereas low-value areas were concentrated in the Jingtai-Jingyuan arid belt and the loess hilly region in the north, reflecting the combined influence of water conditions, topography, and human activities. Historical analysis indicates that 80.6% of the region experienced ecological improvement, and future projections suggest that 21.4% of the area may continue to recover, while ecological reversal risks remain in northern Pingchuan, northern Jingyuan, and southeastern Jingtai. The study demonstrates that national key ecological governance projects have achieved remarkable effectiveness in ecological protection and restoration in the Baiyin Section of the Yellow River Basin.
Accurately characterizing the spatiotemporal distribution of surface solar radiation is crucial for solar energy resource assessment and regional renewable energy planning. In this study, ground-based radiation observations in Gansu Province were used as the reference to bias-correct the hourly surface downward solar radiation from the fifth-generation ECMWF (European Centre for Medium-Range Weather Forecasts) reanalysis (ERA5) using a machine learning approach. Based on the corrected data, the spatiotemporal variability of surface downward solar radiation in Gansu Province during 2000-2024 was systematically analyzed, and annual cumulative radiation totals are quantified for each prefecture-level administrative region. The results demonstrate that the machine learning-based method significantly improves the accuracy of the ERA5. The correlation coefficient between the corrected data and ground observations increases by 12.04%, while the root mean square error decreases by 36.45%. Compared with the CARE (Cloud Remote Sensing, Atmospheric Radiation and Renewal Energy Application) satellite remote sensing product released by the Aerospace Information Research Institute, Chinese Academy of Sciences, the correlation coefficient between them reaches 0.87, and the remaining biases are mainly concentrated along the northeastern margin of the Tibetan Plateau. Over the study period, the provincial mean surface downward solar radiation is 206.73 W·m-2, corresponding to an annual cumulative total of 1 659.60 kWh·m-2, which is higher than the national average. Spatially, the radiation exhibits a distinct pattern of being higher in the northwest and lower in the southeast. The radiation in Jiuquan area reached 1 828.44 kWh·m-2, indicating excellent solar energy development potential. Moreover, no significant interannual fluctuation trend was observed across the province.
With the increasing frequency of warm-sector heavy rain events in North China, it is of great significance to study the occurrence and evolution mechanism of mesoscale convective systems during warm-sector heavy rain processes to improve the forecasting ability of warm-sector heavy rain. This research conducts a numerical simulation on the circulation background, thermodynamic structure, and moisture transportation characteristic of an extreme warm-sector rainstorm event in the Beijing-Tianjin-Hebei region using the high-resolution (3 km) WRF (Weather Research & Forecasting Model) mesoscale model, combined with the 0.25°×0.25° reanalysis data from the fifth generation global climate reanalysis dataset (ERA5) of the European Centre for Medium-Range Weather Forecasts, along with conventional and radar observation data for rapid assimilation updates. The results demonstrate that: (1) The high-resolution WRF model, which has been rapidly updated with assimilated observational data, can effectively simulate this warm-sector rainstorm process, accurately representing the radar echo characteristics and propagation mechanisms of meso-small scale systems, verifying the model’s capability to characterize key processes of warm-sector rainstorms. (2) The dynamic characteristics of this process are characterized by synergy of “three jet streams”: the 950 hPa ultra-low-level jet, the 850 hPa low-level jet, coupled with the strong divergence in the exit region on the right side of the 200 hPa upper-level jet, forming a vertical suction structure. The phased evolution characteristics of the low level (the establishment of the ultra-low-level jet, the fluctuation of the low-level jet intensity, the enhancement and maintenance of the low-level jet) are the key factors for the occurrence and maintenance of the heavy precipitation process. (3) Enhanced upward motion induced by low-level jet intensity fluctuation and convergence continuously lifts warm-moist airflow, promoting water vapor condensation and precipitation. Meanwhile, downward intrusion of mid-level weak dry air into high warm-moisture areas triggers the release of unstable energy, further intensifying the process. (4) The low-level high-humidity environment provides abundant water vapor conditions for the heavy rain. With the strengthening of the low-level southeast jet stream, water vapor from the Bohai Bay continuously flow into the Beijing-Tianjin-Hebei region. The strong accumulation of water vapor, combined with powerful dynamic conditions, is the main cause of local short-term heavy precipitation.
The formation and evolution of haze involve multi-scale atmospheric physical and chemical processes. “High humidity” is a typical pollution-related meteorological characteristic of the Sichuan Basin and an important influencing factor for haze development. Based on ERA5 reanalysis data from 2015 to 2018 and ground-based conventional environmental meteorological observations, this study systematically analyzed the evolution characteristics of water vapor and its relationship with atmospheric visibility during winter haze processes in the Sichuan Basin. The results show that: 1) The mean regional net water vapor budget during winter haze processes in the Sichuan Basin is (3.40±2.92)×106 kg·s-1, indicating an overall water vapor surplus; the western and southern boundaries are the main water vapor input pathways, the eastern boundary shows net output, and water vapor transport across the northern boundary exhibits uncertainty. 2) As the haze processes evolve from the formation to the development and persistence stages, the lower-tropospheric (below 700 hPa) water vapor content increases continuously, and the water-vapor high-value tongue extends northward with an expanding coverage. 3) The increase in lower-tropospheric water vapor facilitates the hygroscopic growth of near-surface aerosols, thereby increasing the mass extinction coefficient and consequently reducing atmospheric visibility.
The complex influence of snow cover on surface energy processes constitutes a critical source of uncertainty in wintertime numerical simulations over complex terrain and therefore warrants further investigation. Comparative simulation experiments were conducted for a snow-covered period (18-26 February) and a snow-free period (11-19 January) in 2014 over the Lanzhou New Area using the Weather Research and Forecasting (WRF) model version 4.3. Four land surface models (LSMs), SLAB, Pleim-Xiu, RUC, and NoahMP were systematically evaluated against observations from four meteorological towers to reveal the impact of snow cover on simulation accuracy and scheme sensitivity. Satisfactory performance was achieved during the snow-free period: correlation coefficients (R) of air temperature ranged from 0.80 to 0.97, with normalized centered root mean square errors (NCRMSE) of 0.27-0.60. The R of wind speed ranged from 0.46 to 0.82, and the absolute bias was generally below 0.5 m·s-1, successfully reproducing slope wind circulation. Conversely, simulation accuracy declined significantly during the snow-covered period. R of air temperature for half of the LSMs decreased below 0.80, cold biases exceeded 5.00 ℃, and NCRMSE increased to 0.38-0.79. Wind speed NCRMSE increased to 0.77-2.52, while wind direction frequency errors doubled. Taylor diagram analysis demonstrated that snow cover enhanced the sensitivity to LSMs, indicated by increased dispersion in normalized standard deviation among the schemes. NoahMP exhibited the superior performance with the lowest cold bias under snow-covered conditions (R≈0.9; NCRMSE<0.5), emphasizing the significance of accurate snow process representation for improving winter meteorological simulation in complex terrain.
From 25 to 29 November 2024, Heilongjiang Province experienced an extreme precipitation event associated with a northeast cold vortex (NECV), during which precipitation at multiple observation stations exceeded historical records. Using hourly observations from surface meteorological stations in Heilongjiang Province and ERA5 reanalysis data from the European Centre for Medium-Range Weather Forecasts (ECMWF), the evolution characteristics of the NECV and the formation mechanisms of sustained heavy precipitation were investigated. The results indicate that the cold-core structure of the NECV initially appeared in the mid-troposphere, extended downward during its development, and retreated to the mid-levels during the weakening stage. During the development and mature stages, subsidence dominated on the southern side of the vortex, while pronounced upward motion and deep moist layers were present on the northern and eastern sides. Throughout the heavy precipitation period, the precipitation center remained on the eastern side of the NECV. The southeasterly low-level jet and super low level jet acted as warm conveyor belts, continuously transporting moisture and heat to the precipitation area, and exhibited a pronounced diurnal variation, with jet intensification and downward extension of strong winds from early morning to afternoon, accompanied by significant vertical wind shear. Heavy precipitation showed a strong correspondence with the 925 hPa moisture convergence zone. The long-term maintenance of sustained moisture transport and convergence near Hegang was a necessary condition for the occurrence of extreme precipitation. In addition, terrain-induced convergence and uplift, together with the coupling of upper- and lower level jets, significantly enhanced low level ascent, leading to prolonged and extreme precipitation. Extreme precipitation mainly occurred on the windward slopes of the eastern foothills of the Xiaoxing’an Mountains.
To deepen the understanding of precipitation patterns in complex mountainous terrain, hourly precipitation data from three meteorological observation stations at different altitudes on the eastern side of the Fanjing Mountain in Guizhou Province in the flood season from May to October during 2022-2023 were used to analyze the diurnal variation characteristics of precipitation at the foot, mid-slope and summit stations. The results show that the amount and intensity of precipitation from night to morning increase with increasing altitude, while both decrease with increasing altitude from afternoon to evening. The periods with the large precipitation amount at stations on the foot and middle of the mountain occur from the late afternoon to evening, while at the summit station it concentrates from the early morning to morning. The precipitation at three stations mainly originates from rainfall events lasting from 2 to 18 hours. At the summit station, the precipitation amount during rainfall events lasting less than 8 hours is greater at night than during the day, while at the foot and mid-mountain stations, this characteristic is only observed in rainfall events lasting less than 3 hours.The peak period of precipitation shows a systematic delay with altitude increasing. From the night to the morning at the station at the foot of the mountain, then from the night to the noon at the station on the middle of the mountain, and finally transitioning from the noon to the early morning at the top station of the mountain, it exhibits a trend of “spreading from the morning to the noon and then to the early morning”. Short duration heavy precipitation (with precipitation amount greater than or equal to 25 mm) mostly occurs in the morning and has the highest frequency, while long duration precipitation has the highest proportion of precipitation amount.
To gain an in-depth understanding of the fine-scale characteristics of short-term heavy precipitation under Zhengzhou’s complex terrain, based on hourly precipitation data from national and regional stations from 2013 to 2022, conventional observation data, and high-precision geographic information data, this study systematically analyzes the multi-temporal scale variations and spatial distribution patterns of short-term heavy precipitation in Zhengzhou and quantitatively explores the relationships between precipitation intensity, frequency and topographic factors. Combining the case study of the extreme torrential rain event occurring in July 2021 (“21·7”) in Zhengzhou, the study reveals the thermodynamic mechanisms through which terrain triggers and enhances short-term heavy precipitation. The results indicate that the station-based frequency of short-term heavy precipitation in Zhengzhou shows a fluctuating increasing trend, July and August are the peak occurrence periods. The active period is between 14:00 and 20:00 (Beijing Time, the same as below), peaking from 18:00 to 20:00. The probability of daytime occurrence in mountainous areas is significantly higher than in plains. The short-term heavy precipitation events with rainfall intensity greater than or equal to 20 mm·h?¹ occur mostly in mountainous areas, whereas extreme events with rainfall intensity greater than or equal to 50 mm·h?¹ are more likely in the Zhengzhou main urban area and Xinmi City, reflecting a spatial distribution pattern where mountainous areas experience higher frequency but relatively lower intensity, while urban areas exhibit stronger extremity. Circulation classification shows that under weak synoptic-scale forcing backgrounds, the number of station occurring short-term heavy precipitation in mountainous areas is significantly greater than that in plain areas. Terrain’s influence on rainfall intensity distribution of short-term heavy precipitation is not significant, but it has a clear impact on its frequency. During the “21·7” torrential rain process, the triggering effect of the terrain convergence line and the mechanism of convective enhancement caused by the uplift on the windward slope and the thermal difference of the underlying surface are particularly prominent.
Based on hourly precipitation data from high-density regional automatic stations and national stations in Shaanxi Province during 2009-2023, the spatiotemporal characteristics of short-term heavy rainfall (hourly precipitation greater than or equal to 20.0 mm) in different regions of Shaanxi were comparatively analyzed to provide a scientific basis for refined forecasting and early warning of short-term heavy rainfall. The results show that: (1) The frequency and precipitation extremes of short-term heavy rainfall in Shaanxi generally increase from north to south, with the highest values occurring in southern Shaanxi, where the maximum hourly precipitation reaches 108.7 mm. (2) The normalized frequency of short-term heavy rainfall exhibits a significant increasing trend in the Guanzhong region; short-term heavy rainfall in all regions is mainly concentrated from June to August, with a peak in late July. From April to June and in September, short-term heavy rainfall in southern Shaanxi is significantly more frequent than that in Guanzhong and northern Shaanxi. Precipitation extremes in all three regions show increasing trends, and the occurrence time of peak extremes is progressively delayed from south to north. Precipitation intensity increases in Guanzhong and southern Shaanxi, with the maximum intensity in all regions occurring in early August. The variation characteristics of the normalized frequency of extreme short-term heavy rainfall are generally consistent with those of short-term heavy rainfall. (3) The diurnal variation of the normalized frequency of short-term heavy rainfall in all regions reaches its maximum at 19:00. Northern Shaanxi exhibits a bimodal pattern, with a primary peak during 14:00-23:00 and a secondary peak during 03:00-05:00. Guanzhong shows a unimodal pattern, with a high-frequency period from 16:00 to 01:00 of the following day. Southern Shaanxi displays pronounced nocturnal rainfall characteristics, with a high-frequency period from 16:00 to 04:00 of the following day, and short-term heavy rainfall during the late night mainly occurring in the central and western parts of the region. Compared with short-term heavy rainfall, the peak period of extreme short-term heavy rainfall is delayed by approximately 1 hour in Guanzhong and advanced by approximately 1 hour in southern Shaanxi.
