Introduction

Sandstorms predominantly occur in arid and semi-arid regions with severe desertification, representing one of the most hazardous weather phenomena in northern China during spring [1-3]. These events cause atmospheric pollutant concentrations to surge while horizontal visibility plummets within short timeframes, inflicting substantial impacts on ecological environments, public health, transportation, and socio-economic systems, with severe cases resulting in casualties [4-6]. Research utilizing numerical models and remote sensing observations indicates that the primary influencing systems for spring sandstorms are upper-level troughs, Mongolian cold vortices, and cold fronts [7-9]. In the Hexi Corridor region, spring cold-front sandstorms often accompany Mongolian cyclones, where the intensity and position of upper-level troughs, jet streams, and surface cold fronts determine sandstorm severity and extent [10-12].

Drought exacerbates desertification in the Hexi Corridor, thereby increasing sandstorm frequency and intensity [13-14]. Consequently, investigating sandstorm genesis and transport mechanisms against a backdrop of arid climate conditions holds critical significance for safeguarding lives and property, disaster prevention and mitigation, and ecological governance [15-16]. Strong winds, dust sources, and unstable atmospheric stratification constitute the essential conditions for sandstorm formation [17-18]. Recent studies employing numerical value models and remote sensing observations reveal that summer sandstorm influencing systems primarily consist of upper-level short-wave troughs, mesoscale shear lines, and surface thermal lows [19-20]. Sandstorm outbreaks result from coordinated development of upper- and lower-level circulation patterns [21-22], with momentum downward transfer frequently intensifying surface winds and strengthening sandstorms when atmospheric stratification becomes unstable [23-24].

Since 2000, northwest China has exhibited concurrent trends of rising temperatures and increasing precipitation, with sandstorm occurrences showing a decreasing trend [25-26]. However, the summer of 2023 witnessed the most intense regional high-temperature drought event in nearly 60 years across the Hexi Corridor, with wind and dust events increasing markedly and sandstorm counts reaching the highest level since 2000. This extreme climatic background provided favorable conditions for the frequent occurrence of summer and autumn sandstorms in the Hexi Corridor during 2023.

1. Study Area Overview

The Hexi Corridor in Gansu Province is situated in an arid to semi-arid inland region, extending from Wushaoling in the east to Xingxingxia at the Gansu-Xinjiang border in the west [27], with a total length of approximately 1000 km and a width of 40-200 km. Characterized by a temperate continental arid climate, the region's annual precipitation is far less than evaporation, resulting in severe water resource shortages [28]. Desertified areas exceed a certain proportion of the total land area, with Minqin County exhibiting a typical sand-encircled landscape (desert encirclement rate exceeding a specific threshold) and abundant sand source materials, making it a high-vulnerability zone for natural disasters and a region with frequent sandstorm activity [12,27].

2. Data and Methods

This study analyzes two strong sandstorm events that occurred in the Hexi Corridor during summer and autumn 2023 using: (1) surface meteorological observation data from national and regional automatic weather stations (Fig. 1); (2) ERA5 reanalysis data with 0.25° × 0.25° spatial resolution (https://cds.climate.copernicus.eu); (3) hourly PM10 concentration data (https://www.aqistudy.cn); (4) Himawari-8 satellite dust monitoring products; and (5) HYSPLIT model backward trajectory simulation results. Synoptic diagnostic methods are employed to analyze the two strong sandstorm processes. All times mentioned in the text and figures refer to Beijing Time.

3. Results and Analysis

3.1 Weather Conditions

On August 16, 2023, from 06:00 to 08:00, strong sandstorms occurred in Liangzhou District and Minqin County in the eastern Hexi Corridor (Fig. 2), with average maximum wind speeds reaching 9.9 m·s⁻¹ (Beaufort scale 5), instantaneous peak winds of 16.2 m·s⁻¹, and minimum visibility dropping to 300 m. From 15:45 to 22:00 on September 6-7, sandstorms affected Zhangye City in the eastern Hexi Corridor, with Minqin County experiencing a strong sandstorm featuring average maximum winds of 13.2 m·s⁻¹ (Beaufort scale 6), instantaneous peak winds of 22.3 m·s⁻¹, and minimum visibility of just 133 m.

3.2 Pre-event Climate Conditions

Influenced by the aftermath of La Niña and developing El Niño conditions, the Hexi Corridor experienced remarkable climate extremes in 2023 [29], with average temperatures reaching record highs for the period and precipitation at its lowest in six decades. During summer 2023, most areas of the Hexi Corridor experienced the most severe regional high-temperature drought event in nearly 60 years. Precipitation in Minqin County, Jiuquan City, and Jinta County was more than 50-75% below the climatological average. The standardized precipitation index (SPI) for June-August reached severe drought levels. Temperature anomalies at most stations in the central and eastern regions exceeded specific thresholds, with high-temperature days reaching the highest count in nearly 20 years and the latest ending date on record. The high temperatures provided favorable thermal conditions for sandstorm development, while the dry climatic background thickened the surface sand layer, offering abundant material basis for sandstorms [30-31].

3.3 Upper-level Circulation Characteristics

Both sandstorm events exhibited a stepped-trough pattern in upper-level circulation, a typical configuration for summer and autumn severe convective weather in northwest China. Both processes featured strong atmospheric baroclinicity, northwesterly jet streams, and unstable upper-cold/lower-warm stratification. However, the intensities of the western trough cold center, plateau warm center, mid-level jet, low-level jet, and upper-level frontal zone in the September 6-7 event were stronger than those on August 16, resulting in more intense, longer-lasting sandstorms with broader impact (the August 16 event affected central-eastern Hexi Corridor, while September 6-7 impacted only the eastern part).

3.4 Surface Frontogenesis Effects

Surface cold fronts directly triggered both sandstorm events. On August 16, a thermal low over central Inner Mongolia and a cold high over central-western Xinjiang created a stable, slow-moving cold high blocked by the thermal low, allowing cold air to accumulate in the Hexi Corridor and form a strong pressure gradient zone. The cold front passage occurred in the morning on August 16 and in the afternoon on September 6-7. During the latter event, rapid afternoon temperature increases intensified the sandstorm. In both processes, the central-eastern Hexi Corridor was situated within a surface saddle field where intense frontogenesis occurred, significantly strengthening the frontal system as it entered the region.

Both events featured post-cold-front northwesterly winds, though the area experiencing Beaufort scale 6+ winds was smaller on August 16. The September 6-7 event exhibited distinct post-cold-front northwesterlies and cyclonic convergence from low-pressure easterlies, with maintenance of small-scale surface convergence systems that hindered dust dispersion, causing Minqin County's visibility to plummet to 133 m. The saddle field structure and diurnal temperature variation promoted frontogenesis, increased surface wind speeds, and the small-scale wind field convergence was unfavorable for dust diffusion.

3.5.1 Surface Meteorological Elements and PM10 Variation Characteristics

During the August 16 event, pressure transformation from negative to positive and wind direction shift from easterly to northwesterly occurred between 04:00-05:00, with wind speeds rapidly increasing to 9.3 m·s⁻¹ and PM10 concentrations surging to 1520 μg·m⁻³ as visibility dropped to 300 m. For the September 6-7 event, pressure transformation and wind direction shift from northeasterly to northwesterly occurred between 17:00-18:00, with wind speeds increasing to 13.2 m·s⁻¹ and PM10 reaching 1150 μg·m⁻³ as visibility fell to 100 m. In both cases, the sandstorm outbreak synchronized with frontal passage. Minimal visibility before the events (2000-3000 m) rapidly decreased as PM10 concentrations and wind speeds increased simultaneously, indicating that local dust lifting was the primary source, supplemented by external transport.

3.5.2 Near-surface Intensive Meteorological Element Characteristics

Analysis of intensive upper-air data shows the August 16 event occurred in early morning while September 6-7 occurred in the afternoon. Inversion layer analysis reveals that evening inversion layers were thinner and lower than morning ones, creating more unstable stratification. Consequently, frontal passage in the evening produced stronger sandstorms. During the August 16 event (08:00), wind speeds at 1150 m reached 19.2 m·s⁻¹, providing dynamic conditions for sandstorm development. During the September 6-7 event (20:00), strong winds at 630 m (15.9 m·s⁻¹) and 1600 m also supplied sufficient momentum.

3.5.3 Atmospheric Stratification Stability Analysis

The K-index, a parameter for assessing atmospheric stability, is expressed as: [FIGURE:9]. Larger temperature differences between upper and lower layers accumulate more unstable energy, yielding higher K-values and greater instability. K-index charts show that on September 6 at 20:00, the eastern Hexi Corridor exhibited extremely unstable stratification (K-index reaching 37.5°C), more unstable than August 16 at 08:00 (21.3°C), corresponding to the stronger sandstorm intensity.

Near-surface air was warm and dry during both events. Below 1400 m on August 16 and below 1250 m on September 6-7, temperature-dewpoint differences exceeded specific thresholds, indicating very dry conditions. This near-surface warm-dryness enhanced thermal convection development, accelerated soil moisture evaporation, rapidly decreased soil moisture content, and made loose surface dust more easily lifted by strong winds.

3.6 Divergence, Vertical Velocity, and Vorticity

ERA5 0.25°×0.25° reanalysis data were used to diagnose divergence, vertical velocity, and vorticity, with expressions given by formulas (1) and (2). During both events, the divergence field over Minqin County showed low-level convergence and upper-level divergence (Fig. 10a-b). The August 16 event had a convergence center at 850 hPa with intensity -2.4×10⁻⁵ s⁻¹ and a divergence center at 650 hPa with intensity 0.75×10⁻⁵ s⁻¹. The September 6-7 event featured a maximum convergence center at 800 hPa (-0.75×10⁻⁵ s⁻¹) and a divergence center at 600 hPa (0.75×10⁻⁵ s⁻¹). Ascending motion occurred between 800-550 hPa, with maximum upward velocity at 700 hPa reaching -0.6 hPa·s⁻¹ on August 16 and -3.0×10⁻³ hPa·s⁻¹ on September 6-7. Below 700 hPa, consistent positive vorticity existed (Fig. 10e-f), with maximum centers at 800 hPa (3.2×10⁻⁵ s⁻¹) for August 16 and at 700 hPa (0.75×10⁻⁵ s⁻¹) for September 6-7. Strong mid-to-lower tropospheric upward motion provides the primary dynamic mechanism for lifting surface dust.

3.7 Dust Transport Characteristics

Using GDAS 1°×1° data as initial conditions, backward trajectory simulations were conducted from Minqin Station (38.6319°N, 103.0886°E) at 1000 m starting height. Results show that on August 16, dust originated from the Badain Jaran Desert, transported via east-northeasterly paths through central-western Inner Mongolia into the Hexi Corridor. On September 6-7, dust sources were the Gurbantünggüt Desert and Badain Jaran Desert, transported west-east by northwesterly airflow, with additional dust supplementation from the Badain Jaran Desert along the path. Himawari-8 satellite dust products clearly show dust transport paths consistent with HYSPLIT simulations (Fig. 11-12).

3.8 Comparison of Summer/Autumn and Spring Sandstorm Characteristics

Comparing typical cases from summer/autumn 2023 (2 events) and spring (1 event) reveals that spring sandstorms in the Hexi Corridor are primarily influenced by long-wave and transverse troughs at 500 hPa, while summer/autumn events feature short-wave and stepped troughs. Both seasons involve cold fronts at the surface, but spring events develop Mongolian cyclones ahead of the cold front, whereas summer/autumn events exhibit mesoscale shear lines. Analysis of typical cases shows that compared to spring sandstorms, summer/autumn events require stronger mid-to-low level jets, greater vertical velocities, higher near-surface wind speeds, and larger pressure differences across cold fronts (Table 1).

4. Conclusions

Through comparative analysis of two strong summer/autumn 2023 sandstorms and typical cases from both summer/autumn and spring in the Hexi Corridor, the following conclusions are drawn:

1) Both events primarily involved local dust sources supplemented by external transport. The preceding drought conditions provided abundant material basis for sandstorms. When local dust dominates, near-surface air humidity becomes the key factor influencing sandstorm intensity. Sandstorms intensified significantly when surface pressure fields formed saddle patterns, frontal passages occurred in afternoon/evening, and mesoscale convergence existed at the surface.

2) The physical quantity characteristics of the two strong sandstorms featured: convergence below 800 hPa (center at 850 hPa, intensity ≤-0.75×10⁻⁵ s⁻¹); divergence at 650-700 hPa (center at 650 hPa, intensity ≥0.75×10⁻⁵ s⁻¹); ascending motion layer at 800-600 hPa (maximum at 700 hPa, intensity ≤-0.6 hPa·s⁻¹); and positive vorticity below 700 hPa (maximum center at 800 hPa, intensity ≥0.75×10⁻⁵ s⁻¹).

3) External dust transport paths differed: the August 16 event passed through central and western Inner Mongolia via east-northeasterly routes, while September 6-7 followed northwesterly paths through western Mongolia, Xinjiang, and western Inner Mongolia. Continuous dust source supplementation along transport paths resulted in longer duration and greater intensity.

4) Compared to spring sandstorms, summer/autumn events require stronger upward motion, greater near-surface wind speeds, and larger surface pressure differences. When 500 hPa shows stepped or short-wave troughs with surface fronts or mesoscale shear lines, and when mid-level jets ≥20 m·s⁻¹, low-level jets ≥14 m·s⁻¹, maximum upward velocity ≤-0.6×10⁻³ hPa·s⁻¹, and near-surface wind speeds ≥9 m·s⁻¹ occur, sandstorms may develop in the Hexi Corridor during summer and autumn.

References

[1] Qian Li, Li Yanying, Yang Yonglong, et al. Distribution characteristics of strong sandstorm and special case analysis of strong sandstorm initiated by squall line in the eastern of Hexi Corridor[J]. Arid Land Geography, 2010, 33(1): 29-36.

[2] Tu Aiqin, Wang Zhenzhu, Zhu Genghua, et al. Pollution characteristics of two strong dust processes in northern China in March 2021[J]. Journal of Arid Meteorology, 2023, 41(4): 607-619.

[3] Qian Li, Yao Yubi, Yang Xin, et al. An analysis of sandstorm weather process in the Hexi Corridor in summer[J]. Journal of Desert Research, 2016, 36(2): 458-466.

[4] Duan Bolong, Liu Xinwei, Guo Runxia, et al. Cause analysis on severe dust storm in northern China on 15 March 2021[J]. Journal of Arid Meteorology, 2021, 39(4): 541-553.

[5] Zhao Huihua, Luo Xiaoling, Li Yanying, et al. Climate characteristics of extreme weather in 2023 in Wuwei City, eastern Hexi Corridor[J]. South Central Agricultural Science and Technology, 2024, 45(8): 126-128.

[6] Yin Xiaohui. Progress and prospect of research on sand dust weather in China[J]. Journal of Desert Research, 2009, 29(4): 728-733.

[7] Ma Zhuguo, Fu Congbin, Yang Qing, et al. Drying trend in northern China and its shift during 1951-2016[J]. Chinese Journal of Atmospheric Sciences, 2018, 42(4): 951-961.

[8] Li Yaohui, Zhang Shuyu. Review of the research on the relationship between sand dust storm and arid in China[J]. Advance in Earth Science, 2007, 22(11): 1169-1176.

[9] Hua Cong, Liu Chao, Zhang Bihui. Comparative analysis of transport characteristics of two dust events affecting Beijing[J]. Journal of Desert Research, 2019, 39(6): 99-107.

[10] Wu Shuoqiu, Ma Xiaoyan. Analysis of dust vertical and horizontal distribution during dust events in northwest China based on FY-4A, MODIS and CALIPSO satellite data[J]. Acta Scientiae Circumstantiae, 2020, 40(8): 2892-2901.

[11] Ye Qia, Zheng Xiaoshen, Zhao Shangyu. Remote sensing monitoring and transport path analysis of an intense dust weather in northern China in 2021[J]. National Remote Sensing Bulletin, 2023, 27(8): 1821-1833.

[12] Zhang Chunyan, Li Yanying, Ma Xingwei, et al. Analysis of the vertical momentum transmission characteristics of different intensity trough type sandstorm along the Hexi Corridor[J]. Advances in Earth Science, 2022, 37(9): 925-936.

[13] Di Xiaohong, Liu Xinwei, Sha Hong'e, et al. Dynamic mechanism of a severe sandstorm occurred in the Hexi Corridor[J]. Journal of Arid Land Resources and Environment, 2016, 30(8): 145-151.

[14] Wang Xiwen, Liu Zhiguo, Huang Yuxia, et al. Integrated analysis of a severe sand storm in Hexi Corridor in summer[J]. Meteorological Journal, 2006, 32(7): 102-109.

[15] He Yuanping, Zhang Yunwei, Gu Zhaolin. Features, triggering and evolution of very strong sandstorms[J]. China Environmental Science, 2021, 41(8): 3511-3522.

[16] Zhang Chunyan, Li Yanying, Zeng Ting, et al. Case analysis and climatic characteristics of winter sandstorm over eastern Hexi Corridor[J]. Meteorological Monthly, 2019, 45(9): 1227-1237.

[17] Hu Wenjie, Ma Li, Wu Xiuqin, et al. Dust storms forward trajectories and influence range over the Mu Us Desert[J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2021, 57(6): 1161-1171.

[18] Wang Ganquan, Shen Xia. The FY-4 radiometer imager and the application of its data in the satellite meteorology[J]. Chinese Journal of Nature, 2018, 40(1): 1-11.

[19] Li Yun. Operational application of Fengyun satellite in dust weather monitoring[J]. Satellite Applications, 2018(11): 24-28.

[20] Zhang Qiang, Wang Sheng. On physical characteristics of heavy duststorm and its climatic effect[J]. Journal of Desert Research, 2005, 25(5): 675-681.

[21] Wei Qian, Long Xiao, Tian Chang, et al. Multiscale meteorological characteristics during a sandstorm in Minqin[J]. Arid Zone Research, 2018, 35(6): 1352-1362.

[22] Lei Zhengcui, Zheng Yuanyuan, Liu Yinfeng, et al. Causes analyses of a severe continuous fog-haze weather process in Changzhou in 2018[J]. Meteorological Monthly, 2019, 45(8): 1123-1134.

[23] Mao Ye, Zhang Hengde, Zhu Bin. Analysis of the continuous heavy pollution process in the winter of 2016 in Beijing, Tianjin, and Hebei[J]. Environmental Science, 2021, 42(8): 3615-3621.

[24] Ablikmu Alemitijiang, Li Na, Zhao Keming, et al. Characteristics of atmospheric boundary layer during an east irrigated sandstorm in Tarim Basin[J]. Desert and Oasis Meteorology, 2019, 13(5): 55-61.

[25] Zhao Jin, Zhang Chunyan, Chen Jing, et al. Analysis of the causes and sand and dust transport characteristics of a severe sandstorm in the Hexi Corridor in spring 2024[J]. Journal of Agricultural Catastrophology, 2024, 14(10): 201-203.

[26] Fei Bingqiang, Wu Bo, Yin Jie, et al. Changes in the spatial pattern of newly cultivated and abandoned farmland in the Mu Us Sandy Land from 1964 to 2020 and their impact on desertification[J]. Arid Land Geography, 2025, 48(4): 661-672.

[27] Shen Xiaoling, Cen Lulin, Zhang Chaoqin, et al. Analysis of multi-types radar products characteristics of a gust front and the extreme wind after the gust front[J]. Journal of Arid Meteorology, 2025, 43(1): 114-125.

[28] Iraji F, Memarian M H, Joghataei M, et al. Determining the source of dust storms with use of coupling WRF and HYSPLIT models: A case study of Yazd Province in central desert of Iran[J]. Dynamics of Atmospheres and Oceans, 2021, 93: 101197, doi: 10.1016/j.dynatmoce.2020.101197.

[29] Lu X, Mao F Y, Pan Z X, et al. Three dimensional physical and optical characteristics of aerosols over central China from long-term CALIPSO and HYSPLIT data[J]. Remote Sensing, 2018, 10(2): 314, doi: 10.3390/rs10020314.