Based on data from 152 meteorological stations in Northwest China and 16 climate models of CMIP6, the CMIP6 model data were bias-corrected using the RoMBC method. The Standardized Precipitation Evapotranspiration Index (SPEI) was then constructed to analyze the spatial and temporal distribution and variation of drought in Northwest China under the historical and future scenarios (SSP1-2.6, SSP2-4.5, SSP3-7.0, SSP5-8.5). The results are as follows: (1) Under the historical scenario, the northwest area experienced a notable increase in both the temperature and precipitation. The temperature and precipitation have been rising at a rate of 0.15-0.74 ℃ and 2.71-14.83 mm per decade, respectively, and the same is expected for future scenarios. (2) From 1975 to 2014, the annual and seasonal SPEI in Northwest China decreased overall. The maximum decline rate in spring was 0.19 per decade. Droughts in most areas were increasingly intense throughout the year, particularly in spring and winter. In terms of drought frequency in Northwest China, mild and moderate droughts appeared more than severe and extreme droughts, and this type of natural disaster was more frequent in the east of the country than in the west. (3) From 2020 to 2100, Northwest China is likely to suffer from droughts, but there are no distinct drought characteristics identified in the research under the SSP1-2.6 scenario. The northwest region is expected to experience an increase in the number of droughts, trends in drought, and drought frequency under the other three scenarios. The most severe drought conditions were observed under the SSP5-8.5 scenario. This study provides insights into the spatial and temporal development of drought in Northwest China using meteorological and model data. The findings can serve as a basis for drought risk assessment, scientific water resources management, and agricultural production in the region.

From observational data of scattering coefficients, the mass concentrations of aerosols and pollutants, and meteorological elements, collected from July 9, 2020 to July 8, 2023 in Xilinhot, the characteristics of aerosol scattering coefficients—including the variation over time, probability density distribution, and correlation with different types of aerosols and meteorological impact factors—are studied. Consequently, the scattering coefficient levels are classified. The results show that: (1) the overall level of aerosol scattering is relatively low, but the transport of dust aerosol in spring and the high frequency of temperature inversions in winter and at night may increase aerosol scattering. (2) The smaller the aerosol, the higher the correlation between aerosols and scattering coefficients, with the correlation coefficients following the order BC>PM_2.5>PM₁₀, although seasonal differences are observed. In addition, NO₂ is an important factor in increasing scattering in autumn, whereas SO₂ contributes to scattering in summer, autumn, and winter. (3) The increases in correlation coefficients are considered as the contribution rates of current meteorological factors to scattering coefficients, with contribution rates of between 1% and 2%.

Based on the daily minimum temperature measured at 42 meteorological stations in Qinghai Province from 1961 to 2019, the spatial and temporal evolution of different grades of cold days (extreme cold, extremely cold, severe cold, major cold, minor cold, light cold, slightly cold, cool) were analyzed. The results showed that: (1) From 1961 to 2019, the number of cold days in Qinghai Province gradually increased with decreasing levels, mainly dominated by slightly cold days. The total number of cold days showed an overall decreasing trend, with a significant rapid decrease occurring in 1995. The decrease in total cold days was mainly caused by the reduction in extremely cold days. After the climate abrupt change in 1997, the trends of severe cold, major cold, minor cold, light cold, slightly cold, and total cold days decreased, while the trend of extremely cold days increased. The trends of extreme cold and cool days decreased and increased respectively, with a distribution roughly equal. (2) Qinghai Province exhibits significant spatial differences in cold days, with the total number of cold days gradually increasing with altitude, and the trends of higher-level cold days are more pronounced. As the level of coldness decreases, the trends of increase and decrease develop towards lower latitudes and areas with relatively lower altitudes. (3) The numbers of extreme cold, extremely cold, severe cold, light cold, slightly cold, and total cold days gradually decrease with increasing annual mean temperature, while the numbers of major cold, minor cold, and cool days gradually increase with increasing annual mean temperature. (4) Except for severe cold days, the numbers of other levels of cold days in Qinghai Province show persistence, and the downward trend will continue in the future, but the strength of persistence varies.

Reservoirs play a pivotal role in regional economic development and societal well-being. In recent years, Xinjiang has experienced frequent extreme precipitation events, which pose significant challenges to reservoir safety. However, research on precipitation characteristics specific to Xinjiang’s reservoirs remains limited, preventing scientific guidance for water resource utilization and reservoir management. Using daily precipitation data from 1961 to 2022 and hourly precipitation data from 2009 to 2022 in the Xiagou reservoir watershed of Yiwu County, Hami City, this study analyzes long-term variations at different time scales and precipitation levels. The findings indicate following points: (1) During the rainy season from 1961 to 2022, the average precipitation in the Xiagou reservoir catchment area showed a weak increase, and the number of days with discontinuous precipitation enhanced significantly. The total number of precipitation days and the maximum number of continuous precipitation days reduced evidently. Together, these trends may lead to further enhancement of the precipitation intensity in the region. (2) The rainy season is mainly characterized by light rainfall; however, over the past 62 years, the number of light and moderate rain days has declined, unlike heavy and torrential rain days. The highest proportion of rainstorms to total rainfall during the rainy season was approximately 50.0%, which was the main reason for the increase in precipitation during the rainy season in the Xiagou reservoir catchment area. (3) Daily precipitation values displayed a rising pattern as the precipitation intensity strengthened. For instance, high values for light, moderate, heavy, and torrential rain occurred in the afternoon, midday, morning, and early morning, respectively. Except for torrential rain, high values for other precipitation levels were predominantly observed during the daytime. The daily variation curves for the average precipitation intensity exhibit multiple peaks with remarkable variations. The relationship between cumulative precipitation frequency and amount was more closely associated with light and moderate rain than with precipitation intensity. Conversely, the relationship between average precipitation intensity and cumulative precipitation amount was more closely related to heavy and torrential rain than to cumulative precipitation frequency.

This study aimed to spatially and temporally characterize not only land surface temperature (LST) in the complex mountainous terrain of Daqingshan, Inner Mongolia but also the environmental factors affecting it. We used the Weather Research and Forecasting Mode (WRF) used to obtain LST data with high temporal and spatial resolution and analyze the variation of mountain influencing factors. The accuracy of the WRF simulated LST (WRF LST) was verified by the observation values of meteorological stations and MODIS LST values, and the relationship between LST and environmental factors was analyzed by the method of comprehensive impact factor analysis and the method of single impact factor analysis. The comprehensive impact factor analysis is based on regional WRF LST and regional environmental factors. Single impact factor analysis achieves the relationship between WRF LST and single environmental factors by fixing other environmental factors. The results revealed that the correlation coefficients between the simulated and observed values were >0.97 (P<0.001) and the spatial correlation with MODIS LST was 0.73 (P<0.05), indicating that WRF has good practicability in mountainous areas. After comprehensive impact factor analysis, it was found the annual WRF LST had the greatest correlation with elevation (R>0.97), followed by temperature at 2 m and water/air mixing ratio at 2 m (R>0.8), vegetation coverage and slope (R>0.3), and other factors. By single impact factor analysis, LST decrease rate with elevation was 0.83 K·(100m)^-1, 0.79 K·(100m)^-1, 0.80 K·(100m)^-1 and 0.32 K·(100m)^-1 in spring, summer, autumn and winter, and it increased by -0.05 K, 0.17 K, -0.14 K, and 0.02 K for every 10° increase in slope in spring, summer fall winter, respectively. LST also increased for every 10% increase in vegetation cover by 0.31 K, 1.41 K in summer and winter, and was not correlated with fall. The slope direction and average LST for the four seasons were south>southwest>southeast>west>east>northwest>northeast>north. The 2 m water-air mixing ratio increased logarithmically with LST, while the 2 m air temperature increased exponentially with LST. This study demonstrated that the WRF model can be used to simulate the spatial and temporal distribution of LST in mountainous terrain and analyze the LST relationship in complex mountain environments.

Drought prediction has always been a major challenge in the field of drought research. Improving the accuracy of drought prediction is the key to solving the drought problem. The standardized precipitation evapotranspiration index (SPEI) was calculated on the basis of the monthly precipitation and average temperature data from 34 meteorological stations in Xinjiang from 1961 to 2019. Dry and wet changes in the Xinjiang region were analyzed. An empirical mode decomposition (EMD)-Gray Wolf Optimizer (GWO)-long short-term memory network is proposed. A combination prediction model based on the data decomposition of LSTM was used to forecast the drought, and the performance of the model was evaluated. The results were as follows: (1) the drought periodicity was stable and the periodicity was long. (2) EMD can effectively optimize the stationarity of data, GWO can optimize the parameters of the prediction model, and the prediction accuracy of the combination model is significantly higher than that of the single prediction model. (3) The accuracy of the results of the four prediction models in descending order was as follows: EMD-GWO-LSTM, GWO-LSTM, GWO-support vector regression (SVR), and LSTM (goodness of fit: 0.972, 0.939, 0.862, 0.830, respectively). The prediction accuracy of the EMD-GWO-LSTM combination prediction model was higher than that of the other three prediction models. The EMD-GWO-LSTM combination prediction model can effectively improve the accuracy of meteorological drought prediction and provide a novel approach for meteorological drought forecasting and drought mitigation in Xinjiang.

The topographic conditions of the bell in the upper reaches of the Ili River lead to an extremely uneven spatial distribution of precipitation, and it is difficult for limited observation stations to truly determine the spatial and temporal changes in daily precipitation. Therefore, it is necessary to assess the applicability of different precipitation products in the upper reaches of the Ili River. On the basis of seven statistical indicators and the generalized three-cornered hat method, we determined the accuracy and uncertainty of three precipitation products (GPM, ERA5, and CHIRPS) in the upper reaches of the Ili River. The results show that (1) ERA5 showed the highest correlation between POD and FAR, and its moderate and heavy rain precipitation estimates were the most accurate. The root mean square error of GPM was the smallest, and POD and FAR were the lowest. CHIRPS showed the smallest relative bias and mean error, its POD and FAR values were between those of GPM and ERA5, and its light rain precipitation estimates were the most accurate. The accuracy of rainstorm precipitation estimated by the three precipitation products was not high, but ERA5 was better than GPM and CHIRPS. (2) The uncertainty of daily precipitation of ERA5 was between that of GPM and CHIRPS, and the signal-to-noise ratio was the largest. GPM showed the lowest uncertainty of daily precipitation, and the signal-to-noise ratio was between that of ERA5 and CHIRPS. CHIRPS had the largest uncertainty of daily precipitation and the smallest signal-to-noise ratio. (3) The daily precipitation quality of ERA5 was better than that of GPM and CHIRPS, and it can be used to analyze the precipitation characteristics in the upper reaches of the Ili River. GPM had the lowest uncertainty of daily precipitation and is most likely to improve its quality through system calibration. The present findings provide support for hydrological simulation and water resource change analysis in the upper reaches of the Ili River.

Based on data obtained from the Micro Rain Radar (MRR), OTT-PARSIVEL laser raindrop spectrometer, and Rain Gauge (RG) at Zeku Station, the applicability of the MRR in the plateau region was compared and examined for a precipitation weather process on September 17, 2021. The vertical variation characteristics of the MRR observation parameters and raindrop spectrum were investigated at different rain rates. Results show that the observed cumulative rainfall results of the MRR were consistent with those of the raindrop spectrometer and RG, and the MRR 200 m rain rate was highly associated with the raindrop spectrometer inversion value. At various levels of rainfall intensity, differences were found in the vertical distribution of precipitation parameters. Reflectivity, rain rate and liquid water content were affected by evaporation, and they fluctuated from high to low levels in the I stage of rain. The evaporation effect was weakened, and the peak height of each microphysical quantity was lower in the II stage of rain. The increase in particle diameter was due to the intensification of collision and coalescence, and the microphysical quantities increased with the decrease in height in the III stage of rain. Precipitation was dominated by small particles, and the raindrop number concentration contribution of small particles at each height layer was the largest. The contribution rate of 1000-4000 m small particles to the rain rate exceeded 90%, and the contribution rate of medium particles below 1000 m to the rain rate increased with the decrease of height. The contribution rate of large particles to the rain rate in the upper layer was greater than that in the lower layer.

The Qinghai-Tibet Plateau has a unique climate, complex topography, and few meteorological observation stations, which makes it difficult to observe and simulate its regional climate and water cycle processes. Using the regional climate models RegCM and WRF, the spatial and temporal distribution of the climate in this region from 1989 to 2008 was systematically analyzed, and the simulation capability of the RegCM and WRF models was investigated at 10, 25, and 50 km horizontal resolutions in the Qinghai-Tibet Plateau. Results show that the trend of annual average temperature simulated by both models at 10 km horizontal resolution is 1.60-2.12 ℃ lower than the multiyear average temperature simulation at 25 and 50 km horizontal resolution. With increasing horizontal resolution, the simulation biases of annual and seasonal temperatures simulated by the WRF model decrease, and the cold bias of temperature in the central and western parts of the Qinghai-Tibet Plateau improves. The simulated temperature in the RegCM model at a 10 km horizontal resolution has the lowest error, and it is significantly better for simulating the spatial distribution of temperature in the Qinghai-Tibet Plateau. The correlation between the simulated temperature of both models in different seasons and the observation data has been improved. In the precipitation simulation, the WRF model at a horizontal resolution of 25 km has the best correlation with the observed data but has the largest error. With the increase of horizontal resolution, the overestimation of precipitation in the southeastern and southern Qinghai-Tibet Plateau by the WRF model has been significantly improved, and the annual precipitation simulated by the RegCM model gradually approaches the measured values (the overestimation decreases from about 2.73 times to 1.77 times). However, the overall overestimation of precipitation by both models still exists. In the simulation of the five major river sources on the Qinghai-Tibet Plateau, with increasing horizontal spatial resolution, the WRF model reduces the biases of the air temperature in the source region of the Mekong river and Salween River, whereas the RegCM model reduces the biases of the air temperature in the source region of the Brahmaputra River and Mekong river. The largest reduction in precipitation bias was achieved in the Brahmaputra River source region at 10 km horizontal resolution by the WRF and RegCM models. This study can lay the foundation for understanding the impact of climate change on the water cycle process in the Qighai-Tibet Plateau.

Persistent cold air pools (PCAPs) in valley cities lead to the prolonged accumulation of air pollutants, thereby affecting the lives and health of residents. In this study, sounding data and daily air quality data from January 2013 to November 2023 were used to calculate and statistically analyze the characteristics of PCAPs occurrences in the Lanzhou Valley. In addition, the impact of PCAPs intensity on changes in pollutant concentrations was explored, and variations in pollutant concentrations during PCAPs were analyzed and compared with concurrent dust aerosol pollution. Results indicate that from 2013 to 2023, 59 PCAPs occurred, lasting cumulative 197 days. During PCAPs, valley heat deficit and PM_2.5 concentrations were 4.4 J·m^-2 and 52.59 μg·m^-3 higher, respectively, compared with non-PCAPs. The air quality index (AQI), SO₂ concentration, NO₂ concentration, CO concentration, and PM₁₀ concentration increased by 70.37%, 144.3%, 84.3%, 156%, and 73.15%, respectively, whereas O₃ concentration decreased by 60.89% during PCAPs. In PCAPs without dust aerosols, the average PM_2.5:PM₁₀ ratio was 0.58, whereas in PCAPs with dust aerosols, the average ratio was 0.31. During PCAPs with concurrent dust aerosols, PM_2.5 concentration, PM₁₀ concentration, AQI, and O₃ concentration increased by 18.33%, 133.03%, 84.44%, and 8.5%, respectively. However, SO₂ and CO concentrations decreased by 17.54% and 17.88%, respectively. These findings can serve as a reference for atmospheric pollution prevention and management strategies in the Lanzhou region.

Winter drought is a main factor hindering winter livestock production in Inner Mongolia. Thus, quantitative characterization of its spatiotemporal changes and development patterns is of great significance for disaster prevention and reduction and for ensuring the healthy development of agriculture and animal husbandry. Using ERA5-Land reanalysis meteorological data from the winter of 1980 to 2021 (October to March of the following year), the standardized precipitation evapotranspiration index (SPEI) was calculated at monthly and semi-annual scales. Trend analysis, spatiotemporal hotspot analysis, and other methods were used to analyze the winter drought evolution characteristics of the entire Inner Mongolia region and the five main vegetation types. Results show that in the past 40 years, the overall SPEI in Inner Mongolia has shown an increasing trend in winter, and aridification varies among different vegetation and months, with a few vegetation and months tending toward humidification. The change patterns in Inner Mongolia mainly include three types: oscillating hot spots, oscillating cold spots, and undetected patterns. From a seasonal perspective, hotspots are primarily distributed in most areas of western Inner Mongolia, as well as in Xing’an League and Tongliao City in the east. On a monthly scale, hotspots often appear in the central and western regions of Inner Mongolia. With regard to drought frequency and frequency statistics, mild drought events have the highest frequency, whereas winter drought events occur more frequently and seriously in desert grasslands and neighboring desert areas.

By using stable isotope data of atmospheric precipitation from September 2018 to May 2020 in the Taxkorgan River Basin Valley and meteorological data such as temperature, precipitation, and relative humidity from representative weather stations within the valley, this study analyzed the variation in δ¹⁸O, δ²H, and deuterium excess (d-excess) of precipitation. The influencing factors were explored, and the water vapor transport pathways of atmospheric precipitation in the valley were traced and analyzed using the hybrid single- particle Lagrangian integrated trajectory model (HYSPLIT). Results show that the δ²H and δ¹⁸O values of precipitation generally present a seasonal pattern of enrichment in summer and depletion in winter, showing a significant temperature effect (1.33‰·℃^-1), but no significant precipitation effect was observed. The local meteoric water line is δ²H=7.63δ¹⁸O-3.55, which shows distinct arid climate characteristics. The HYSPLIT simulation results indicate that the water vapor of precipitation in the study basin is mainly influenced by the westerly circulation and local water vapor recycling, with local water vapor evaporation accounting for 54.09% in the summer half-year and the long-distance transport of the western route accounting for 45.53% in the winter half-year. Water vapor from the Indian Ocean in August can bypass the Tibetan Plateau and reach the study area. These findings can provide a reference basis for water resource management and climate response in the Taxkorgan River Basin Valley.

The scarcity of water resources is the most critical natural factor impeding high-quality economic and social development and ecological security in Xinjiang. This paper systematically analyzes trends in precipitation, atmospheric water resources, and surface water resources in Xinjiang. It also establishes the conversion relationship between different water resources in Xinjiang. The findings reveal that annual precipitation water resources amount to 2717.12×10⁸ m³, with water vapor input reaching 21115×10⁸ m³, resulting in a net water vapor income of 347.5×10⁸ m³. Between 1961 and 2022, Xinjiang experienced a 12.5% increase in precipitation conversion. The annual total water resources in Xinjiang is 912.3×10⁸ m³, where surface water resources constituted 864.1×10⁸ m³ from 2001 to 2020, resulting in a water yield coefficient of 0.32. The observed trends show a significant increase in annual precipitation in Xinjiang, a slight decrease in total water vapor input, a marginal increase in net water vapor income, and a significant increase in precipitation conversion between 1961 and 2022. Although surface water resources in Xinjiang are abundant, the water yield coefficient exhibited a weak fluctuating decreasing trend from 2001 to 2020. Nevertheless, prominent issues persist in water resources research in Xinjiang, including insufficient studies on precipitation water resource volumes, understanding of cloud water resource characteristics, and continuous monitoring of the physical process of cloud precipitation. To address these challenges, it is imperative to conduct comprehensive scientific field experiments on cloud precipitation physics, including strengthening research on the physical processes of cloud precipitation, refining cloud water resource assessments, and examining precipitation efficiency and water increase effects within typical cloud systems. These studies will aid in developing cloud water resources and air-groundwater resources joint control technology for arid areas.

Based on MOD16 global evapotranspiration data, the Crop Water Stress Index (CWSI) was computed. This was combined with the meteorological station precipitation, temperature, and vegetation index data in the Fenhe River Basin, along with land use data. Employing the difference method, linear trend method, and correlation analysis, the temporal and spatial characteristics of drought in the Fenhe River Basin from 2000 to 2021 were analyzed. The results showed that: (1) CWSI effectively monitored drought in the Fenhe River Basin, displaying a notably negative correlation between CWSI and 10 cm soil relative moisture. (2) The spatial distribution of CWSI in the Fenhe River Basin exhibited significant disparities, illustrating wet conditions in the south and dry conditions in the north. (3) While interannual CWSI variations in the Fenhe River Basin remained relatively stable, monthly fluctuations were substantial, peaking in May annually. (4) Drought conditions varied distinctly during different growing periods in the Fenhe River Basin: significant drought occurred in the early growing season (April to May), encompassing 48.55% of the Fenhe River Basin area. No drought occurred in the mid-growing season (June to August). By the end of the growing season (September to October), only 11.17% of the area experienced drought. (5) Drought occurrences differed among various land use types, ranked by CWSI from smallest to largest: forest land (0.686) < grassland (0.749) < cultivated land (0.751) < unused land (0.758) < urban land (0.765). These study outcomes offer critical scientific data support for drought monitoring and decision-making regarding drought resistance in the Fenhe River Basin.

The Northeast of Qinghai Province is the key area for crop production and but is highly susceptible to hail. Hail forecasting, early warning, and artificial anti-hail operations are crucial strategies for reducing hail disasters. Understanding the features of hail monitoring data is fundamental to improving hail forecasting capabilities and initiating timely performing hail suppression measures. The hailstorm occurrence on June 29, 2021, Northeast of Qinghai, was analyzed using Doppler radar data, raindrop spectrum data, and high and ground data. The results revealed extreme unstable atmosphere stratification due to cold advection transportation at the upper levels, coupled with obvious temperature increases on the ground in this region, which is the weather background of this hailstorm process. During the hailstorm, the average raindrop spectrum and velocity spectrum at Pingan Station exhibited multipeak distribution. Differences between the maximum diameter of the Pingan hailstorm observed artificially and the raindrop spectrometer were insignificant, indicating the raindrop spectrometer’s efficacy in observing hail particles and determining the time of maximum hail occurrence. The developmental stages of the hail cloud were identified, encompassing occurrence, jump, hail formation, and extinction. There was an obvious “V” shaped inflow gap in the low-level radar reflectivity factor before the hail. In the mature stage, the hail cloud displayed a noticeable bounded weak echo area, with obvious southerly inflow in the middle and lower layers. The radial velocity map depicted a discernible “0 line” in the hail cloud, pointing vertically upward through the overhanging echo and bounded weak echo region, indicating the top of the hail cloud. The research results have important guiding significance for hail forecasting Northeast of Qinghai. Moreover, the characteristics of various elements preceding hail occurrence serve as important criteria for scientifically identifying sites and conducting timely and appropriate artificial hail suppression operations.

Precipitation acts as a crucial supply for mountain glaciers, and its water vapor source closely correlates to the amount of precipitation. This study focuses on the modern glacier distribution area of Ulugh Muztagh in the eastern Kunlun Mountains, analyzing water vapor sources in the region from 2005 to 2022 using the Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) model and the Global Data Assimilation System (GDAS). Employing backward trajectory analysis, we reveal the source and regularity of water vapor in the Ulugh Muztagh region and discuss its seasonal changes. The results show that the water vapor source in the Ulugh Muztagh area mainly extends to the Eurasian interior along the midlatitude westerly belt and is divided into three routes entering the Qinghai-Tibet Plateau from the Tianshan Mountains, the Pamir Plateau, and over the high-altitude stratosphere. On the Qinghai-Tibet Plateau, water vapor from the Indian Ocean either moves northward over the Himalayas or turns northwestward to merge with the westerly circulation into the plateau’s hinterland. Land-source water vapor, entering from the Pamir Plateau and Tianshan Mountains, accounts for 62.52% of the total water vapor in the Ulugh Muztagh area. Meanwhile, sea source water vapor, comprising high-altitude water vapor from the westerly belt (Atlantic water vapor) and the Indian Ocean, accounts for 37.48% of the total water vapor. Notably, we find that the proportion of water vapor from the sea source has increased steadily over recent decades. Analyzing multiyear seasonal averages for water vapor, we find a notably high proportion of locally recycled water vapor in the summer, comprising 22.64% of the total. This study’s outcomes offer valuable insights into the water cycle dynamics of the Ulugh Muztagh area in the East Kunlun Mountains.

The proposed Aerxiangou section of the Duku expressway, characterized by high mountainous terrain and canyons, faces frequent avalanches due to heavy snowfall and climate change. In this study, a collaborative investigation using UAV remote sensing interpretation and field research identified 92 avalanche points. In addition, elevation, slope, surface cutting degree, ground roughness, maximum snow depth, maximum wind speed, average temperature, and average snowfall were selected as driving factors. A geographical detector was used to examine the relationship between terrain factors, different resolutions, and avalanche stability. The results of this study revealed strong avalanche activity with generally poor stability in the study area. However, it was reassuring to note that most avalanche release and activity areas are located on mountain slopes. The accumulation area lies on the valley floor, a considerable distance away from the planned road route, thus minimizing its impact. Results from the geographical detector analysis suggest positive correlations between interpretations of slope and ground roughness with snow avalanche stability across varying resolutions. The interactive detection results are both double-factor enhancement and nonlinear enhancement, with the latter being more significant than the former. The combination of slope and other factors is crucial for determining the impact of avalanche stability. This study offers reliable data support for assessing avalanche vulnerability and risks, thereby establishing a solid scientific basis for constructing and operating the Duku expressway.

Based on hourly precipitation observation data from 340 meteorological stations in Gansu Province from April to October 2013 to 2022, the refined evolution characteristics of warm season precipitation in Gansu Province on a diurnal variation scale were revealed. Discussions and analyses were conducted in different regions, providing a scientific reference for the study of extreme precipitation events in Gansu. The results show the following: (1) the daily peak of precipitation and precipitation intensity in the warm season in Gansu mainly occurs between 10:00 and 13:00, the daily peak of precipitation frequency primarily occurs between 22:00 and 01:00 at night, and the daily precipitation variation has obvious seasonal differences. There is a relatively concentrated distribution of autumn rain in the central and southern parts of Gansu. (2) The diurnal variation of precipitation has distinct regional characteristics. Precipitation in the Qilian Mountains and the plateau slopes of central Gansu mainly occurs during the day, with intense precipitation dominating around noon, marking the peak of daytime precipitation. Conversely, in western Hexi, the peak and frequency of daily precipitation generally occur at night, with occasional sudden heavy rainfall between 18:00 and 21:00. In southeast and eastern Gansu, the precipitation is nonuniformly distributed; nighttime rain is common due to the frequency of precipitation peaks during the night, but strong precipitation periods tend to occur in the afternoon and morning, respectively. (3) The precipitation characteristics of different durations are different. For short-term precipitation events with a duration of 6 h and below, the daily variation of precipitation is mostly “bimodal type.” Long-term precipitation events lasting more than 6 h are “unimodal type” and primarily begin in the evening, reach their peak at night, and end at noon.

Based on the hail observation records of 119 national stations in Inner Mongolia and ERA5 reanalysis data from 1959 to 2021, the differences in layer formation, water vapor, typical temperature layer height, vertical wind shear, and cloud microphysical quantities between hail and nonhail were analyzed. A hail prediction model was established to provide an objective and quantitative basis for potential hail forecasting. The results revealed that the K-index, pseudo-equivalent temperature difference, vertical wind shear, specific humidity, rain water mixing ratio, snow water mixing ratio, ice water mixing ratio, and liquid water mixing ratio were not well distinguished between hail and nonhail. The total index was >50 °C, temperature difference between 850 hPa and 500 hPa was ≥28.4 °C, precipitable water vapor was ≤24 mm, height of -20 °C was <7.05 km, and the height of -20 °C to 0 °C was ≤3.15 km. The above environmental parameter thresholds were able to identify >70% of hail samples, and the reverse condition can identify >70% of nonhail samples. Based on the analysis of the environmental parameters of hail and nonhail samples, the Fisher discriminant method was used to establish a hail prediction model using the total index, temperature difference between 850 hPa and 500 hPa, height of -20 °C to 0 °C, and precipitable water vapor. The accuracy of the model discrimination exceeded 80%.

The micrometeorological elements, radiation budget, and surface turbulent data at two sites with land cover types of subalpine shrub and warm steppe in the Qinghai Lake Basin in 2021 were compared to investigate the differences in the land-atmosphere interaction between the various surface types and the biophysical effects of Land Use/Land Cover Changes on surface temperature. June to September was the growing, and January to April was the nongrowing season. There were marked differences in surface, air, and soil temperatures and relative humidity between the two sites. In the growing and nongrowing seasons, the peak temperatures of the topmost 5 cm of the soil in warm steppe were 295.4° K and 277.6° K, while those in subalpine shrub were only 288.6° K and 275.4° K, respectively. During the growing season, the peak surface and air temperatures of the warm steppe were 298.8° K and 288.2° K, and that of the subalpine shrub were 292.5° K and 286.5° K, respectively. In the nongrowing season, there was no significant difference in the daytime surface temperatures between the two stations, and the night surface temperature of the warm steppe was 2.8° K higher than that of the subalpine shrubs. The subalpine shrub had lower surface, air, and soil temperatures than the warm steppe; these differences between the two stations were more evident during the growing season, and the variations in the relative humidity in the nongrowing season were more obvious. Based on the Direct Decomposed Temperature Metric, the influence of radiation budget and surface energy distribution between the two sites regarding the surface temperature differences was analyzed. The subalpine shrub had cooling effects compared with the warm steppe. In the daytime of the growing and nongrowing seasons, the short wave radiation term promoted the cooling effect of the subalpine shrub, and the sensible, latent, and surface-soil heat flux terms inhibited the cooling effect of the subalpine shrub. At night, the radiation and nonradiation terms promoted the cooling effect of subalpine shrubs in the growing season. In contrast, the sensible heat flux terms had a warming effect, and the other terms demonstrated a cooling effect in the nongrowing season. The main contributing factors to subalpine shrub cooling during the daytime were shortwave radiation, surface-soil heat flux, and sensible heat flux terms. The main contributing factor at night was the surface-soil heat flux term.

Based on the Chen-Weng heat exchange parameterized scheme, the average sensible heat flux from 1980 to 2017 in Qinghai Province was calculated using the observation data collected from 35 stations. Temporal and spatial characteristics of the sensible heat fluxes and their impact factors in Qinghai Province were determined using wavelet analysis, Mann-Kendall test, and Empirical Orthogonal Function. The result shows that the seasonal and annual average sensible heat fluxes have risen since 1980. The primary cycle of the annual average sensible heat flux was 28 a, and the secondary cycle was about 18 a. A high correlation between the seasonal and yearly average sensible heat flux with average ground-air temperature difference manifested. The annual average sensible heat flux increased from 2004 to 2017 due to a rise in the average ground-air temperature differences. The correlation of average wind speed with annual, spring, and autumn average sensible heat fluxes was high. The average yearly sensible heat flux decreased from 1980 to 2004 due to a decline in average wind speed. A prominent negative correlation between summer precipitation and sensible heat flux was identified. From the perspective of space, spring and annual average sensible heat fluxes expressed a prominent east-west difference and partly indicated a north-south variation in the autumn and winter.

Based on daily observations from 56 meteorological stations from 1961 to 2019, this study analyzed dry-wet conditions, changes, and their underlying driving factors in Xinjiang. The main findings are as follows: (1) In the past 59 years, the Xinjiang climate has changed significantly from dry to wet, with the aridity index (AI) changing at a rate of 0.01·(10a)^-1 (P < 0.01). The number of stations with a significant upward trend of AI accounted for 57.1%. (2) The annual precipitation in Xinjiang increased significantly at a rate of 8.6 mm·(10a)^-1 from 1961 to 2019, consistent with the change in AI. Conversely, the annual reference evapotranspiration (ET₀) showed a significant decreasing trend at a rate of -15.7 mm·(10a)^-1. However, it is worth noting that ET₀ transitioned around 1990. ET₀ continued to decrease before 1990 and then switched to a fluctuating upward trend. (3) Wind speed and relative humidity primarily influenced ET₀ variation. Wind speed exhibited a consistent decreasing trend across the entire region, while approximately half of the stations observed a decline in relative humidity. The combined effect of these factors significantly decreased ET₀ at nearly 50% of the stations in Xinjiang. Moreover, trends in relative humidity from increasing to decreasing trends and wind speed from decreasing to increasing around 1990 contribute to the differences in ET₀ trends between the two periods. The conclusions achieved here provide valuable insights into understanding the dry-wet changes and their underlying driving factors in Xinjiang and have implications for the rational development and use of water resources in this region.

Using ensemble empirical mode decomposition and the M-K mutation test, the multi-times cale characteristics of the evolution of total solar radiation in Dunhuang city were analyzed based on the meteorological data of total solar radiation, relative humidity, total cloudiness, and dust days in Dunhuang city between 1971 and 2020. The key meteorological factors influencing solar radiation in Dunhuang city were explored. The results show the following: (1) There was a significant upward trend of annual total solar radiation in Dunhuang city between 1971 and 2020, with a linear climate propensity rate of 49.6 MJ·m^-2·(10a)^-1, and the multiyear average annual radiation was 6354.0 MJ·m^-2, belonging to the area with the most abundant solar resources. The annual radiation was lowest in the 1970s and highest in the 2010s. Dunhuang has four distinct seasons of solar radiation, with radiation increasing at rates of 32.5, 13.4, 2.9, and 1.1 MJ·m^-2·(10a)^-1 in summer > spring > fall > winter, respectively. The total solar radiation in Dunhuang city in the last 50 years was dominated by interannual variations of 2.9 and 7.1 years and interdecadal variations of 16.7 years. (2) Monthly solar radiation varied in a “single-peak” pattern, with a sharp increase in March, a peak in May, a gradual decrease in June, and a yearly minimum in December. The hourly distribution of total solar radiation is monomodal, with the maximum occurring between 12:00 and 13:00 a.m. (3) The annual, spring, and summer solar radiation changes were abrupt in 1997, 2000, and 1982. (4) Meteorological factors affecting solar radiation at Dunhuang can be attributed to three factors: atmospheric transparency, illumination, and humidity, and the correlation between each meteorological factor and solar radiation varies according to the seasons.

This study used the Sen+M-K trend analysis, the gravity center movement model, the standardized regression coefficient, and the spatial clustering methods to reveal the spatial characteristics of the variation, trends, and movement of dust days of various types and the contribution rate of their influencing factors in the Tarim Basin based on the annual data of 32 meteorological stations between 1964 and 2022. The results indicated: (1) Floating dust, followed by blowing sand and sandstorms, dominate the Tarim Basin, showing a spatial distribution pattern of more in the south and less in the north, and the dust days of various types show significantly decreasing trends. (2) The gravity centers of dust days of various types in the Tarim Basin tended to move southeast, and the gravity center of sandstorm days moved the most. (3) Warmer temperatures, lower wind speeds, and fewer gale days were the main factors in the decreased dust days in the Tarim Basin, whereas precipitation had the least influence. (4) The high contribution rates of precipitation to dust days are clustered in the western part of the Tarim Basin, mean temperature and mean maximum temperature in the southern and western parts, respectively, and gale days and mean wind speed in the northwestern and southeastern parts, respectively. This study’s results can provide a scientific basis for developing regionally applicable wind-breaking and sand-fixing measures in the Tarim Basin.

Based on daily temperature observation data and reanalysis data of the geopotential height, sea level pressure, and wind field from 1961 to 2020, the characteristics of spatial and temporal change of cold waves, strong cold waves, and exceptionally strong cold waves lasting 24 h, 48 h, and 72 h were studied in Ningxia over the past 60 years. The causes of atmospheric circulation anomalies of cold waves were also revealed. The results show the following: (1) In the past 60 years, cold waves of different intensities and different durations in Ningxia consistently showed the distribution characteristics of “shifting eastwards and northwards.” (2) The cumulative frequencies of cold waves, strong cold waves, and exceptionally strong cold waves in the region accounted for 71.7%, 22.6%, and 5.7% of the total annual cold wave frequencies, respectively, among which cold waves dominated by process lasting 24 h and 48 h. The proportions of various durations for strong cold waves and exceptionally strong cold waves were equivalent. They mainly occured in October to April, during which the accumulated cold waves, strong cold waves, and exceptionally strong cold waves in the region accounted for 99%, 98%, and 95% of those throughout the year, and there was a decreasing trend from January to April and increasing trend from October to December. (3) In the past 60 years, the cold waves, strong cold waves, and exceptionally strong cold waves have been decreasing at a rate of 4.5, 2.8, and 0.18 per station every 10 years, respectively. Among them, the frequencies of waves of 24 h and 48 h as short durations were decreasing, but that of waves of 72 h as a long duration was decreasing. All types of cold waves decreased from the 1960s to the 1990s, reaching a minimum in the 1990s and increasing since the 2000s, with a sudden change around 1990. (4) Under the influence of global warming, the atmospheric circulation showed completely the opposite distribution characteristics between before and after the sudden change of cold waves in Ningxia. The key systems influencing cold waves in Ningxia are consistent. When the blocking high pressure in the Ural Mountains was stronger, the East Asian trough was deeper, the west Pacific subtropical high was weaker, the western side of Lake Baikal was dominated by cyclonic circulation, and cold high pressure at the ground was active, this was conducive to the southward movement of cold air in middle and high latitudes, and more cold waves in Ningxia.

This study was implemented to accurately identify the avalanche flow characteristics and flow information, and comprehensively analyze avalanche motion. This study was based on UAV tilt photography technology to obtain high-resolution aerial photography data, taking the avalanche-prone area of Aerxiangou as an example. Through on-site investigation and UAV remote sensing interpretation to detect avalanche activity in a high-resolution manner, the goals were to determine the input parameters of the RAMMS model, to simulate and reconstruct different types of avalanche events on this basis, and to comparatively analyze the differences among the results of conventional ground-based investigation, UAV remote sensing interpretation, and simulation results to explore avalanche activity in different types and different snow layer release conditions. The results of the study show that (1) the avalanche investigation and analysis system based around tilt photography technology, which combines conventional ground-based investigation methods with UAV remote sensing and numerical simulation to verify each other, improves the accuracy of the assessment of disaster development status. (2) In mid-February, the snow on the slopes of Aerxiangou approaches the critical thickness value, and continuous snowfall destabilizes the snow layer and triggers new avalanches. The investigation is still in the disaster breeding stage, the snow layer cracks intensified deformation, the role of the wind snow eave self-weight gradually increased, there is more than the trend of the breaking strength of the snow, and the overall stability is poor. (3) In slope-type avalanches with a snow platform above the slope surface as the potential release area, the release volume can reach 8.2669×10⁴ m³, the movement duration is about 128 s, and the flow height of the accumulation area peaks in 120 s at about 3.55 m, the flow velocity is about 18.34 m·s^-1, and the impact force is about 32.67 kPa. In addition, the accumulation area is formed into an accumulation with an area of 3369.7 m² and a volume of 1.8525×10⁴ m³ of the pile. Through mutual verification, the slope-type avalanche does not involve release of the snow platform, and there is a discrepancy between the ground-based investigation results and numerical simulation interpretation results. (4) Trench-slope composite avalanches are released by fracture of the snow layer on the trench-slope, where the depth of fracture is only about 60% of the critical thickness value, the avalanche duration is close to 300 s in this case, and the impact range is 1178.5 m² in the accumulation area, with an average accumulation depth of 1.64 m. The flushing-out volume is 3107.76 m³, the maximum flow rate in the accumulation area is 6.58 m·s^-1, and the maximum impact force is 17.97 kPa. The results of the ground-based investigation are roughly the same as those of the numerical simulation based on the 3D model. The results of the study have improved the accuracy with which avalanche event information can be acquired and can provide strong data support and a scientific basis for predicting future avalanche potential hazards, risk avoidance, and disaster emergency response.

Data from 89 meteorological stations in the Yellow River Basin from 1960 to 2020 was used in this investigation. The Mann-Kendall mutation test as well as Morlet wavelet and correlation analyses were conducted to assess the spatial and temporal change characteristics and influencing factors at the beginning of the growing season (GSS), the end of the growing season (GSE), and the length of the growing season (GSL), as well as days with an active accumulated temperature of ≥10 ℃ (AT10) and active accumulated temperature of ≥10 ℃ (DT10) during the growing season. From 1960 to 2020 the GSS significantly advanced at a rate of -2.04 d·(10a)^-1, while the GSE showed a delayed trend with a change rate of 0.85 d·(10a)^-1, and the GSL was significantly prolonged at a rate of 2.88 d·(10a)^-1; there were also significant regional differences. The GSS in the lower reaches of the Yellow River Basin was the earliest (February 23), while that in the upper reaches was the latest (March 30). Furthermore, the GSE in the upper reaches ended early (October 24), while that in the lower reaches was the latest (November 30), and the GSL in lower reaches was the longest (334.03 d), while that in the upper reaches was the shortest (297.33 d). The significant extension of GSL was mainly due to the significant advance of GSS. Over the past 61 years, the growth season indices were found to have a main period of approximately 28 a in the Yellow River Basin. GSS, AT10, and DT10 mutated in 1998, and GSL mutated in 2002. The changing trends for the growth season indices in the upper, middle, and lower reaches of the Yellow River Basin were consistent, with the largest change occurring in the lower reaches, followed by the upper and middle reaches, respectively. Correlation analyses showed that GSS advances in the Yellow River Basin were mainly related to spring warming over the past 61 years, and the delay of GSE was mainly due to autumn warming, the extension of GSL in the upstream and downstream areas was mainly due to spring warming, and the extension of GSL into the middle reaches was mainly related to autumn warming.

Due in large part to global climate change, drought, flood, and high temperature events have increased significantly around the world in recent years. The Shiyang River Basin is in Northwest China and fringes onto a monsoon region, and is consequently, highly sensitive to climate change. The rapid development of oasis agriculture has led to high levels of development and the utilization of water resources in fragile ecological environments. Future climate change will aggravate the uncertainty of water resources in the basin, posing a threat to food security and economic development. Coupled General Circulation Models (GCMs) play an important role in the prediction of future climate change and formulation strategies to help devise adjustments accordingly. Based on the observed data in the historical period (1985-2014), the simulation capabilities of 11 climate models from the 6th international Coupled Model Intercomparison Program (CMIP6) in the Shiyang River Basin were evaluated. The equidistant cumulative distribution function method was applied to downscale climate data to obtain the future climate change trend for the basin as presented in this paper. The results show that the CMIP6 multi-model ensemble has good applicability in the Shiyang River Basin, as it accurately depicts the annual and seasonal distribution characteristics of climate factors, including precipitation, temperature, and potential evapotranspiration. The model performs well when simulating temperatures, in comparison to precipitation. While multimodel ensemble mean data perform better when simulating precipitation and temperature in the Shiyang River Basin, in comparison with other models. Under different future scenarios (2023-2100), precipitation, temperature, and potential evapotranspiration in the basin show a significant upward trend and increase with the radiative forcing increase. The late 21 century shows a greater increase in climate factors than the early and middle periods. Compared to the historical period, precipitation in the future could increase by 45.02% in the winter and 0.38% in the summer, and the greatest temperature increases can occur in spring and autumn. In the future, the aridity index of the Shiyang River Basin will decrease overall. The climate of the basin will tend to warm and humidify, with the summer season becoming drier while the other seasons become wetter than those in the historical period. The Minqin Basin located in the lower reaches of the basin is the area most sensitive to climate change. The research results have important reference value as they will help to address future climate change and ensure sustainable economic and agricultural development in the Shiyang River Basin.

Monthly summer NDVI and meteorological data from February to August, 2001-2019, were used to analyze the changes in NDVI and the time lag effects of NDVI on climate factors in the Qinling-Daba Mountains. The results showed that NDVI presented an overall increasing trend in the Qinling-Daba Mountains, and the area with a highly significant increase in NDVI accounted for the largest proportion, with a value of 77.1%. There were positive correlations between NDVI and the temperature and precipitation, in which the correlations between NDVI and temperature were higher than those with precipitation. The response of the NDVI to climate factor changes showed significant time lag effects, and the spatial distributions of the lag time had regional differences. In the western regions of the Qinling-Daba Mountains, NDVI had a timely response to temperature changes and a 2-month lag response to precipitation changes. In the central regions, the lag time of the NDVI responses to temperature and precipitation changes varied with latitude. There were generally 1-and 0-month lag responses for temperature changes and 0-and 3-month lag responses for precipitation changes from north to south, respectively. In the northeastern regions there was a 3-month temporal lag in NDVI to the changes in temperature and precipitation. In the southeastern regions, NDVI had a timely response to temperature change and a 3-month lag response to precipitation change. This study provides scientific basis for summer storm and geological disaster warnings in the Qinling-Daba Mountains.

Meteorological elements are the key factors used to assess the earth’s hydrothermal processes. Accurate acquisition of meteorological data is thus of great significance to the development of ecological protection systems and agricultural research. Loess hilly areas typically have hills and gullies, and consequently, the terrain seriously affects the interpolation of meteorological data and its accuracy. Based on the professional meteorological interpolation software ANUSPLIN, this study has used the daily temperature and rainfall data from 105 meteorological stations in and around the Yanhe River Basin from 2010 to 2021 as the basis, and three digital elevation models (DEM) with resolutions of 25 m, 90 m, and 1 km as covariables to interpolate the grid of temperature and precipitation in the loess hilly region. The spatiotemporal variation of precipitation was analyzed, and the applicability of the ANUSPLIN interpolation method in the loess hilly region evaluated. The results show that temperatures in the eastern extension area of the Yanhe River Basin were higher, when compared with those in the western area. The rainfall was lower in the central and northwestern regions when compared with the eastern regions. The distributions of temperature and rainfall were consistent with the laws of the previous meteorological station data. The ANUSPLIN model had a good adaptability to the spatial interpolation of temperature and rainfall in the loess hilly-gully region. In three different DEM resolution simulation scenarios, the accuracy of temperature interpolation was ranked as 25 m > 90 m > 1 km, and the rainfall interpolation accuracy was ranked as 90 m > 25 m > 1 km.