Preamble

ARID LAND GEOGRAPHY
Vol. 48 No. 8 Aug. 2025

Spatiotemporal Variation Characteristics and Main Driving Factors of Soil Moisture in the Qilian Mountains

ZHAO Jianwen¹,², LI Jinlin¹,³, WANG Shengjie¹,²

¹College of Geography and Environmental Science, Northwest Normal University, Lanzhou 730070, Gansu, China
²Key Laboratory of Resource Environment and Sustainable Development of Oasis, Gansu Province, Lanzhou 730070, Gansu, China
³Northwest Institute of Eco-Environment and Resources, Chinese Academy of Sciences, Lanzhou 730000, Gansu, China

Abstract: Soil moisture is crucial for vegetation growth and terrestrial ecosystems, particularly for ecosystem services provided by mountain forests in arid and semiarid regions. This study investigates the temporal dynamics and spatial differentiation of soil moisture in the Qilian Mountains from 2017 to 2021 using remote sensing soil moisture data, trend analysis, correlation analysis, and the geographical detector method. The analysis examines the effects of mean annual temperature, mean annual precipitation, normalized difference vegetation index (NDVI), slope, aspect, and elevation on the spatiotemporal variation of soil moisture. The results reveal that: (1) Soil moisture in the Qilian Mountains remained relatively stable during 2017–2021, with a trend slope of 0.000018, although interannual variability was significant, as indicated by a coefficient of variation of 0.183. (2) During the growing season (May–October), soil moisture exhibited significant spatial heterogeneity (0.068–0.214 cm³/cm³), with an overall pattern of higher moisture in the east and lower moisture in the west, attributed to the influence of the East Asian monsoon. (3) Both correlation analysis and geographical detector results demonstrate that precipitation and NDVI were the dominant factors driving spatiotemporal soil moisture variation, with explanatory power (q-values) of 0.761 and 0.722, respectively, both exceeding 70%. In contrast, topographic factors had minimal effects, with q-values below 0.1. Environmental factors significantly influenced the spatial distribution of soil moisture in the Qilian Mountains, exhibiting interaction effects characterized by two-factor enhancement and nonlinear relationships. These findings provide a scientific basis for developing ecological protection policies and measures for the Qilian Mountains, thereby promoting environmental conservation in the region.

Keywords: spatiotemporal variation; soil moisture; geographical detector; Qilian Mountains

1. Introduction

Soil moisture constitutes a vital component of soil and a key element in the global energy cycle, directly impacting food security, human health, and ecosystem function [1,2]. As a critical link and driving force for material and energy transfer among different spheres of the Earth's surface, soil moisture regulates the spatial patterns and processes of land surface systems [3]. It controls surface water and heat fluxes and influences the partitioning of available land energy into sensible and latent heat fluxes [4], making it a crucial variable in hydrology, meteorology, ecology, agriculture, and climate change research. Soil moisture is also a key factor affecting water regulation services in mountain ecosystems, reflecting plant water utilization status and land-atmosphere interaction patterns through ecological processes such as infiltration, runoff, and evapotranspiration [5]. Consequently, soil moisture exerts important influences on the water cycle while exhibiting high variability and nonlinearity in both temporal and spatial dimensions.

Soil moisture detection methods can be categorized into traditional approaches and remote sensing techniques. Traditional methods such as oven drying, neutron probes, and frequency domain reflectometry offer high measurement accuracy but suffer from limitations including low spatiotemporal resolution, high costs, and operational complexity, making them suitable primarily for small-scale applications [6]. In contrast, microwave remote sensing technology has become an effective means for large-scale dynamic environmental monitoring due to its all-weather detection capability, penetration advantages, and continuous observation characteristics [7]. Microwave remote sensing exploits the significant dielectric constant differences between dry soil and liquid water to directly retrieve soil moisture [8]. Based on sensor operation modes, it can be divided into passive and active types: passive microwave remote sensing retrieves water content by receiving microwave radiation signals emitted by the soil itself [9], while active microwave remote sensing analyzes radar backscatter coefficients [10]. Optical remote sensing technology primarily relies on multispectral data from Landsat/MODIS to indirectly reflect soil moisture conditions by constructing drought indices or vegetation indices [11], but its effectiveness is often constrained by cloud cover and lighting conditions.

The spatiotemporal differentiation of soil moisture is driven by multiple coupled factors, with mechanisms that can be classified into climate-driven and underlying surface response categories. The climate system dominates the dynamic balance of soil moisture through hydrological processes including precipitation input, temperature regulation, and evapotranspiration output [10,17-18]. Underlying surface attributes establish multi-scale water redistribution mechanisms through vegetation [19], slope [20], soil properties, topography [21], and land use type [22]. Revealing these interaction effects on soil moisture transport pathways can improve spatiotemporal distribution modeling of soil moisture.

The Qilian Mountains serve as a critical ecological barrier in the arid regions of northwest China, and the stability and health of their ecological environment are vital for regional ecosystem security and sustainable socioeconomic development [23]. Previous studies have extensively investigated soil moisture spatiotemporal variability and influencing factors in the Qilian Mountains. For instance, Che et al. [24] found that soil temperature variation was small while water content fluctuated significantly in the western Qilian Mountains grasslands, with a quadratic spatial distribution and linear temporal evolution. Hu et al. [25] confirmed in the Pailugou watershed that soil temperature and moisture along vertical profiles and during the growing season exhibited nonlinear patterns of initial increase followed by decrease. However, due to limitations imposed by high-altitude complex terrain and sampling conditions, existing research has primarily focused on local point observations, with systematic analysis of regional-scale soil moisture heterogeneity and its controlling factors remaining limited.

The objectives of this study are to: (1) analyze the spatiotemporal differentiation characteristics of surface soil moisture at the regional scale; (2) explore the coupling mechanisms of multiple environmental factors on soil moisture; and (3) identify the dominant factors controlling the spatial distribution of surface soil moisture in the mountain system. The findings will enhance understanding of alpine hydrological cycles and provide theoretical support for ecological environmental protection in the Qilian Mountains.

2. Data and Methods

2.1 Study Area

The Qilian Mountains are located in the northeastern Tibetan Plateau, spanning northeastern Qinghai Province and western Gansu Province between 94°52′–103°09′E and 36°26′–40°01′N [26]. Comprising a series of mountains and valleys in China's northwestern arid and semiarid regions, the Qilian Mountains serve as the source area for inland rivers including the Heihe, Shiyang, and Shule Rivers, fulfilling the ecological function of a "mountain water tower" [27]. Situated at the intersection of the Tibetan Plateau, northwestern arid region, and eastern monsoon region, the area is jointly influenced by monsoon and Tibetan Plateau circulation systems [28]. The central and eastern regions exhibit a continental semiarid alpine steppe climate, while the western region has a continental semiarid desert climate [29]. During winter, the Mongolian high dominates with dry and cold air masses, while in summer, the Tibetan thermal low pressure and East Asian monsoon jointly drive moist air currents [30], with mean annual precipitation of 250–500 mm decreasing from southeast to northwest [31]. Vegetation shows distinct vertical distribution, transitioning from desert steppe, mountain steppe, mountain forest steppe, alpine shrub meadow, alpine meadow to alpine sparse meadow with increasing elevation [32]. According to the second glacier inventory, the Qilian Mountains contain 2,684 glaciers covering a total area of 1,597.81 km², which are facing accelerated retreat under global warming [33].

2.2 Data Sources

Soil moisture data were obtained from the 0.05° soil moisture dataset for the Qilian Mountains provided by the National Tibetan Plateau Data Center (https://data.tpdc.ac.cn). This dataset was downscaled from the SMAP L3 passive microwave 36 km soil moisture product (SMAP L3 Radiometer Global Daily 36 km Grid Soil Moisture, 0–5 cm) using a random forest optimization downscaling model coupled with wavelet analysis [34]. Precipitation, temperature, and normalized difference vegetation index (NDVI) data were obtained from the National Earth System Science Data Center of the National Science and Technology Infrastructure Platform (www.geodata.cn), including 1 km resolution monthly precipitation and mean temperature datasets (2017–2021) and monthly MODIS NDVI data. A 30 m ASTER DEM was used to extract slope and aspect data for the study area. All data were clipped using ArcGIS 10.8 software, with soil moisture data resampled to 1 km resolution using Kriging interpolation and nearest-neighbor methods.

2.3 Methods

2.3.1 Geographical Detector

The geographical detector is a statistical method for exploring spatial differentiation characteristics and driving forces of geographic elements, proposed by Wang et al. [38]. It comprises four components: factor detection, interaction detection, risk detection, and ecological detection [39].

Factor Detection. The factor detector calculates the explanatory power (q-value) of each influencing factor to quantitatively analyze its effect on the spatial differentiation of soil moisture in the Qilian Mountains. The formula is:

$$
q = 1 - \frac{1}{N\sigma^2}\sum_{i=1}^{L} N_i\sigma_i^2
$$

where q ranges from [0, 1]; N_i and N represent the number of units in layer i and the entire region, respectively; and σ_i² and σ² represent the variance of Y values in layer i and the entire region, respectively.

Interaction Detection. Interaction detection identifies whether the combined effect of two factors on soil moisture spatial differentiation is enhanced or weakened. The criterion is: if the q-value of the interaction exceeds that of either single factor alone, it indicates two-factor enhancement; if the interaction q-value exceeds the sum of the two individual q-values, it indicates nonlinear enhancement.

Ecological Detection. Ecological detection compares whether the effects of different factors on soil moisture spatial distribution are significantly different, using an F-statistic for testing:

$$
F = \frac{N_x(N_x-1)}{N_y(N_y-1)} \times \frac{SSW_x}{SSW_y}
$$

where N_x and N_y are the sample sizes of two factors, and SSW_x and SSW_y are the within-layer variance sums formed by the two factors.

Indicator Selection. Based on previous research experience, this study selected six natural factors affecting soil moisture variation. Using the optimal parameters geographical detector model in R to calculate optimal classification methods and numbers avoids subjective bias from traditional manual classification [40]. Data were processed as follows: mean annual temperature and slope were classified using the quantile method into 6 and 8 classes, respectively; mean annual precipitation and elevation were classified using the natural breaks method into 9 and 10 classes, respectively; NDVI and aspect were classified using the standard deviation method into 6 and 8 classes, respectively.

3. Results and Analysis

3.1 Temporal Variation Characteristics

3.1.1 Intra-Annual Variation

Analysis of intra-annual soil moisture variation from 2017 to 2021 revealed that monthly-scale soil moisture fluctuated between 0.092 and 0.138 cm³/cm³ [FIGURE:2]. As spring temperatures gradually increased, snowmelt contributed significantly to soil moisture replenishment, with phase transitions during freeze-thaw processes also contributing to seasonal variations. From May to October, the synergistic effect of East Asian monsoon precipitation and alpine meltwater resulted in peak soil moisture values. In autumn, decaying water vapor transport from the westerlies combined with soil freezing processes caused sharp declines in soil moisture content, which remained low and stable during winter.

3.1.2 Inter-Annual Variation

The average soil moisture value in the Qilian Mountains decreased to 0.114 cm³/cm³ in 2017, reaching the minimum value of 0.108 cm³/cm³ in 2018. In 2019, the average value increased rapidly to 0.138 cm³/cm³, the maximum during the study period, likely due to abundant precipitation that year. Subsequently, soil moisture content showed a continuous declining trend, reaching 0.112 cm³/cm³ by 2021. The interannual variation of soil moisture was relatively stable during 2017–2021, which has important reference value for maintaining ecological balance and water resource management in the region.

Soil moisture at different elevations showed consistent trends from 2017 to 2021, with relatively stable moisture content in the mid-elevation zone (2987–3888 m), while high- and low-elevation areas exhibited greater fluctuations [FIGURE:2], indicating that these areas are more sensitive to climate change. Notably, soil moisture did not decrease monotonically with elevation; moisture content in the 3452–3888 m range was higher than in the 2987–3458 m range, suggesting complex elevation effects.

3.2 Spatial Variation Characteristics

During the growing season (May–October), soil moisture in the Qilian Mountains maintained a stable pattern of higher moisture in the east and lower moisture in the west, with annual mean values decreasing slightly from 0.129 cm³/cm³ in 2017 to 0.128 cm³/cm³ in 2018, then to 0.120 cm³/cm³ in 2019, and stabilizing at 0.120–0.121 cm³/cm³ in 2020–2021 [FIGURE:3]. The spatial heterogeneity of soil moisture was most pronounced in 2019, with a maximum coefficient of variation of 0.174 cm³/cm³. Trend analysis revealed that significant wetting areas were concentrated in the easternmost regions and some western local areas [FIGURE:3], which is highly coupled with monsoon water vapor transport pathways and topographic uplift effects.

3.3 Correlation Analysis Results

3.3.1 Pearson Correlation Analysis

Pearson correlation analysis was first employed to examine relationships among environmental factors and soil moisture. Soil moisture showed significant positive correlations with mean annual temperature, mean annual precipitation, and NDVI (P < 0.01), with correlation coefficients of 0.312 and 0.761 for precipitation and NDVI, respectively [TABLE:1]. However, except for the correlation between soil moisture and precipitation, other correlation coefficients did not exceed 0.5.

Spatial correlation analysis between soil moisture and precipitation, NDVI, and temperature revealed very strong positive correlations between soil moisture and precipitation (correlation coefficient up to 0.761) and between soil moisture and NDVI (correlation coefficient 0.722), confirming a positive vegetation-water feedback mechanism under hydrothermal coupling [FIGURE:4]. Soil moisture and temperature also showed positive correlation (coefficient 0.312), indicating that temperature increases may help enhance soil moisture, though this effect is less direct than precipitation and may be modulated by evaporation and other factors.

Significance test results showed that correlations between soil moisture and precipitation, NDVI, and temperature passed the 0.01 significance test only in small eastern areas, with large regions showing non-significant correlations [FIGURE:4].

3.4 Geographical Detector Analysis Results

To further quantify the effects of environmental factors and analyze their combined impacts, this study employed geographical detector analysis. The q-values for each factor were calculated [FIGURE:5], revealing that among the six factors studied, precipitation and NDVI had the greatest influence on temporal soil moisture variation in the Qilian Mountains, with q-values significantly higher than other factors. Except for 2017 when NDVI was the primary factor, precipitation was the dominant factor in all other years.

All selected factors influenced soil moisture variation to varying degrees, with the ranking of influence being: mean annual precipitation > NDVI > elevation > mean annual temperature > slope > aspect. Precipitation and NDVI had the highest q-values at 0.761 and 0.722, respectively, while topographic factors had minimal effects with q-values below 0.1. This confirms that precipitation is the primary factor determining soil moisture spatial distribution.

Interaction detection revealed that all factors exhibited interactive effects on soil moisture, showing two-factor enhancement and nonlinear enhancement relationships, indicating that soil moisture spatial differentiation in the Qilian Mountains results from multiple factors acting together. Compared with single-factor effects, q-values increased significantly under two-factor interactions, with the strongest interactions being: mean annual precipitation ∩ NDVI, mean annual precipitation ∩ temperature, and mean annual precipitation ∩ elevation, with q-values of 0.883, 0.851, and 0.847, respectively, all exceeding 80% explanatory power.

Ecological detection demonstrated significant differences among factors [FIGURE:5], indicating that each factor influences soil moisture spatial patterns in unique ways, with these differences being statistically meaningful.

3.5 Environmental Factor Variation Analysis

3.5.1 Precipitation Variation

Intra-annual precipitation variation in the Qilian Mountains aligned closely with monthly soil moisture patterns, increasing continuously from January to July when it reached the maximum value of 358.21 mm, then gradually decreasing to the minimum of 80.12 mm in December [FIGURE:6]. Notable sharp increases and decreases occurred in June and October. Interannual variation showed a trend of initial increase, then decrease, then increase again, peaking at 366.72 mm in 2019 and dropping to the minimum of 332.58 mm in 2021 [FIGURE:6]. Precipitation distribution showed clear spatial differentiation, gradually decreasing from east to west [FIGURE:7].

3.5.2 Temperature Variation

Mean monthly temperature remained below zero for five months, gradually increasing from January, reaching the maximum of 9.02°C in July, then decreasing to the minimum of -16.76°C in December [FIGURE:6]. Interannual variation was relatively small, fluctuating between -3.87°C and -3.24°C [FIGURE:6]. The temperature distribution pattern showed lower temperatures in the southwest and higher temperatures in the northeast with increasing elevation [FIGURE:7].

3.5.3 NDVI Variation

Intra-annual vegetation coverage in the Qilian Mountains was poor from January to April, improved significantly from May, reached the maximum of 0.206 in August, then continuously decreased [FIGURE:6]. Interannual variation was small, ranging from 0 to 0.206 [FIGURE:6]. Spatially, NDVI decreased from east to west, with vegetation mainly distributed in low-elevation areas of the central and eastern regions with favorable hydrothermal conditions [FIGURE:7].

4. Discussion

The Qilian Mountains form the natural boundary among the Tibetan Plateau, Mongolian-Xinjiang Plateau, and Loess Plateau. The complex landforms and hydrothermal conditions result in significant regional differences in soil moisture spatiotemporal variation [41]. This study systematically analyzed soil moisture spatiotemporal variation characteristics and influencing factors from 2017 to 2021, revealing significant temporal dynamics and spatial heterogeneity.

Precipitation has a decisive influence on soil water content and distribution [42]. It not only directly affects soil moisture but also creates spatiotemporal heterogeneity due to its uneven temporal and spatial distribution [43]. Precipitation amount directly relates to the dynamic balance of soil moisture, influencing surface water and groundwater recharge as well as evaporation and plant transpiration [44]. In the Qilian Mountains, precipitation shows distinct wet and dry seasons and clear spatial distribution, gradually decreasing from east to west—a trend that matches soil moisture spatial patterns. The eastern Qilian Mountains, significantly influenced by the Asian monsoon, receive abundant precipitation as warm, moist monsoon air rises over the terrain and condenses. With increasing elevation, decreasing temperatures, and steeper terrain, the orographic lifting effect becomes more pronounced, increasing rainfall and providing superior soil moisture conditions [27,45]. The correlation coefficient of 0.761 between precipitation and soil moisture, with a factor detection q-value of 0.761, confirms precipitation as the primary factor affecting soil moisture variation, consistent with findings by Che et al. [24] and Hu et al. [25].

The Qilian Mountains constitute an important water source area in China, where vegetation plays a key role in maintaining soil moisture. Vegetation effectively captures and retains water through its root systems, reducing evaporative losses [46], while canopy interception further reduces surface runoff, promotes infiltration, and enhances water conservation functions [47]. Vegetation growth and root activity improve soil structure and increase porosity, facilitating water infiltration and storage [48], thereby enhancing soil water retention. Vegetation cover also reduces surface temperature and solar radiation, effectively decreasing soil evaporation rates [49] and helping maintain soil moisture. Recent ecological restoration measures such as returning farmland to forest and grassland and closing mountains for afforestation have significantly increased vegetation coverage [50]. The vegetation distribution pattern of higher coverage in the east and lower coverage in the west [FIGURE:7] aligns closely with precipitation and soil moisture conditions, as lower elevations and abundant precipitation in the eastern region provide favorable conditions for vegetation growth [51].

Elevation indirectly affects soil moisture dynamics by influencing meteorological factors such as precipitation, temperature, and solar radiation [52]. By altering hydrothermal conditions, elevation affects plant species and distribution, thereby influencing soil moisture distribution and availability [53]. During the ablation period, high-elevation areas receive more snow and ice meltwater, while lower temperatures, weaker soil evaporation, and less plant biomass result in lower water consumption, helping maintain soil moisture. The Qilian Mountains host extensive glaciers that are accelerating their retreat under global warming, particularly in the central and eastern basins [35,49], releasing substantial water that significantly supplements soil moisture. Additionally, phase transitions during freeze-thaw processes increase soil moisture. The temperature distribution pattern of lower temperatures in the southwest and higher temperatures in the northeast with elevation [FIGURE:7] creates relatively stable temperature conditions in areas with lush vegetation, benefiting plant growth and ecosystem balance. However, lower temperatures at high elevations may reduce photosynthetic rates and inhibit vegetation growth [54], affecting soil moisture retention and recharge. In arid regions with relatively low soil water content, temperature's effect on soil moisture through evapotranspiration is relatively limited. Therefore, the interaction between precipitation and vegetation remains the key factor maintaining soil moisture in arid and semiarid regions.

Since the 21st century, northwest China's arid region has shown a significant warming and wetting trend [55], particularly after 2008, with this wetting trend not only intensifying but also expanding into the monsoon region, demonstrating an accelerated "warm-wetting" phenomenon [56]. Increased precipitation is considered the main cause of wetting in northwest China's arid region [57]. This long-term climate change has profoundly affected regional precipitation patterns, providing an important background for understanding soil moisture variation in the Qilian Mountains from 2017 to 2021.

5. Conclusions

This study investigated the spatiotemporal variation characteristics and driving factors of soil moisture in the Qilian Mountains from 2017 to 2021. The main conclusions are:

1. Soil moisture in the Qilian Mountains remained relatively stable during 2017–2021, with a trend slope of 0.000018, although interannual variability was significant, reflected by a coefficient of variation of 0.183.

2. During the growing season (May–October), soil moisture exhibited significant spatial heterogeneity (0.068–0.214 cm³/cm³), with an overall pattern of higher moisture in the east and lower moisture in the west, attributed to the influence of the East Asian monsoon.

3. Both correlation analysis and geographical detector results demonstrate that precipitation and NDVI were the dominant factors driving spatiotemporal soil moisture variation, with explanatory power (q-values) of 0.761 and 0.722, respectively, both exceeding 70%. In contrast, topographic factors had minimal effects, with q-values below 0.1. Environmental factors significantly influenced the spatial distribution of soil moisture in the Qilian Mountains, exhibiting interaction effects characterized by two-factor enhancement and nonlinear relationships.

These findings provide a scientific basis for ecological protection and water resource management in the Qilian Mountains, offering valuable insights for sustainable development in the region.

References

[1] McColl K A, Alemohammad S H, Akbar R, et al. The global distribution and dynamics of surface soil moisture[J]. Nature Geoscience, 2017, 10(2): 100-104.

[2] Humphrey V, Berg A, Ciais P, et al. Soil moisture atmosphere feedback dominates land carbon uptake variability[J]. Nature, 2021, 592: 65-69.

[3] Gou Q, Zhu Q. Response of deep soil moisture to different vegetation types in the Loess Plateau of northern Shannxi, China[J]. Scientific Reports, 2021, 11: 15098, doi: 10.1038/s41598-021-94584-2.

[4] Hsu H, Dirmeyer P A. Soil moisture evaporation coupling shifts into new gears under increasing CO2[J]. Nature Communications, 2023, 14: 1162, doi: 10.1038/s41467-023-36794-5.

[5] Li W, Migliavacca M, Forkel M, et al. Widespread increasing vegetation sensitivity to soil moisture[J]. Nature Communications, 2022, 13: 3959, doi: 10.1038/s41467-022-31667-9.

[6] Qiao L, Zuo Z, Zhang R, et al. Soil moisture atmosphere coupling accelerates global warming[J]. Nature Communications, 2023, 14: 4908, doi: 10.1038/s41467-023-40641-y.

[7] Zhao Tianjie. Recent advances of L band application in the passive microwave remote sensing of soil moisture and its prospects[J]. Progress in Geography, 2018, 37(2): 198-213.

[8] Zheng Mandi, Liu Zhong, Xu Zhaohui, et al. Research progress of soil moisture estimation based on microwave remote sensing[J]. Acta Pedologica Sinica, 2024, 61(1): 16-28.

[9] Pan Ning, Wang Shuai, Liu Yanxu, et al. Advances in soil moisture retrieval from remote sensing[J]. Acta Ecologica Sinica, 2019, 39(13): 4615-4626.

[10] Wang Junxia, Pan Yaozhong, Zhu Xiufang, et al. A review of researches on inversion of eigenvariance of soil water[J]. Acta Pedologica Sinica, 2019, 56(1): 23-35.

[11] Zhou Lei, Wu Jianjun, Zhang Jie. Remote sensing based drought monitoring approach and research progress[J]. Scientia Geographica Sinica, 2015, 35(5): 630-636.

[12] Zhang K, Wang Q, Chao L, et al. Ground observation based analysis of soil moisture spatiotemporal variability across a humid to humid transitional zone in China[J]. Journal of Hydrology, 2019, 574: 903-914.

[13] Chen Mingyu, Shao Ming'an, Li Tongchuan, et al. Characteristics and influencing factors of spatiotemporal distribution of soil moisture in typical gully of the Loess Plateau[J]. Acta Pedologica Sinica, 2021, 58(2): 381-390.

[14] An Wenming, Han Xiaoyang, Li Zongshan, et al. Effects of different vegetation restoration types on the slope difference of soil water content in the Loess Plateau, China[J]. Acta Ecologica Sinica, 2018, 38(13): 4852-4860.

[15] Han Xinsheng, Liu Guangquan, Xu Hao, et al. Temporal and spatial variation characteristics of soil moisture under three land use types in the semi-arid Loess Region of southern Ningxia[J]. Journal of Soil and Water Conservation, 2022, 36(6): 250-259.

[16] Zhang Zihui, Wu Shixin, Zhao Zifei, et al. Estimation of grassland biomass using machine learning methods: A case study of grassland in Qilian Mountains[J]. Acta Ecologica Sinica, 2022, 42(22): 8953-8963.

[17] Zhou S, Williams A P, Lintner B R, et al. Diminishing seasonality of subtropical water availability in a warmer world dominated by soil moisture atmosphere feedbacks[J]. Nature Communications, 2022, 13: 5756, doi: 10.1038/s41467-022-33473-9.

[18] Dong Jinyi, Luo Min, Meng Fanhao, et al. Spatiotemporal evolution of soil moisture and its driving forces in the Mongolian Plateau[J]. Research of Soil and Water Conservation, 2024, 31(2): 110-121.

[19] Wang Kunyue, Li Xuhong, Yu Xinxiao, et al. Response of soil moisture of typical vegetation to individual rainfall in Beijing mountain area[J]. Acta Ecologica Sinica, 2024, 44(6): 2561-2571.

[20] Li W, Yan D, Weng B, et al. Nonlinear effects of surface soil moisture changes on vegetation greenness over the Tibetan Plateau[J]. Remote Sensing of Environment, 2024, 302: 113971, doi: 10.1016/j.rse.2023.113971.

[21] Thomas A, Yadav B K, Šimůnek J. Water uptake by plants under nonuniform soil moisture conditions: A comprehensive numerical and experimental analysis[J]. Agricultural Water Management, 2024, 292: 108668, doi: 10.1016/j.agwat.2024.108668.

[22] Zhu Qing, Liao Kaihua, Lai Xiaoming, et al. A review of soil water monitoring and modelling across spatial scales in the watershed[J]. Progress in Geography, 2019, 38(8): 1150-1158.

[23] Xue Jian, Li Zongxing, Feng Qi, et al. Spatiotemporal variation characteristics of water conservation amount in the Qilian Mountains from 1980 to 2017[J]. Journal of Glaciology and Geocryology, 2022, 44(1): 1-13.

[24] Che Zongxi, Li Jinjun, Wang Youkui, et al. Characteristics of soil temperature and water content variation in the western Qilian Mountains[J]. Acta Ecologica Sinica, 2018, 38(1): 105-111.

[25] Hu Jian, Lü Yihe, Fu Bojie, et al. Soil hydrothermal variation and rainfall pulses along altitudinal gradient in Pailugou Watershed in the Qilian Mountain[J]. Arid Zone Research, 2017, 34(1): 151-160.

[26] Sun Meiping, Ma Weiqian, Yao Xiaojun, et al. Assessment and spatiotemporal characteristics of service value of glaciers in the Qilian Mountains[J]. Acta Geographica Sinica, 2021, 76(1): 178-190.

[27] Zhang Baiting, Li Zongxing, Feng Qi, et al. Evolution of ecosystem service values in Qilian Mountains based on land use change from 1990 to 2020[J]. Acta Ecologica Sinica, 2024, 44(10): 4187-4202.

[28] Liu Honghong, Liu Shufang. Spatiotemporal evolution and obstacle identification of complex ecosystem resilience in the Qilian Mountain area[J]. Arid Land Geography, 2024, 47(2): 237-247.

[29] Zhou Xueru, Li Yu. Response of dry-wet change to millennial and centennial warm periods in the Qilian Mountains[J]. Acta Geographica Sinica, 2022, 77(5): 1138-1152.

[30] Li Lili, Du Jun, Li Yawei, et al. Changes of vegetation water use efficiency and their responding to scPDSI in frozen ground area of the Qilian Mountains from 2000 to 2020[J]. Journal of Glaciology and Geocryology, 2023, 45(2): 688-698.

[31] Zhang Wenjie, Chen Weiming, Li Baolin, et al. Simulation of the permafrost distribution on Qilian Mountains over past 40 years under the influence of climate change[J]. Geographical Research, 2014, 33(7): 1275-1284.

[32] Wang Lihui, Qin Xiang, Chen Jizu, et al. Reconstruction of the glacier mass balance in the Qilian Mountains from 1961 to 2013[J]. Arid Zone Research, 2021, 38(6): 1524-1533.

[33] Ren Zijian, Wang Jianglin, Xu Henian, et al. Evolution and driving factors of megadrought and pluvial events in the Qilian Mountains during the past 500 years[J]. Arid Land Geography, 2024, 47(2): 214-227.

[34] Zhao Wei, Wen Fengping, Cai Junfei. Methods, progresses, and challenges of passive microwave soil moisture spatial downscaling[J]. National Remote Sensing Bulletin, 2022, 26(9): 1699-1722.

[35] Hu Z, Chai L, Crow W T, et al. Applying a wavelet transform technique to optimize general fitting models for SM analysis: A case study in downscaling over the Qinghai-Tibet Plateau[J]. Remote Sensing, 2022, 14(13): 3063, doi: 10.3390/rs14133063.

[36] Qu Y, Zhu Z, Montzka C, et al. Inter-comparison of several soil moisture downscaling methods over the Qinghai-Tibet Plateau, China[J]. Journal of Hydrology, 2021, 592: 125616, doi: 10.1016/j.jhydrol.2020.125616.

[37] Guillod B P, Orlowsky B, Miralles D G, et al. Reconciling spatial and temporal soil moisture effects on afternoon rainfall[J]. Nature Communications, 2015, 6: 6443, doi: 10.1038/ncomms7443.

[38] Wang Jinfeng, Xu Chengdong. Geodetector: Principle and prospective[J]. Acta Geographica Sinice, 2017, 72(1): 116-134.

[39] Wu Jingjing, Jiao Liang, Zhang Hua, et al. Vegetation coverage variation in the Qilian Mountains before and after ecological restoration[J]. Acta Ecologica Sinica, 2023, 43(1): 408-418.

[40] Wang Zhoufeng, Zheng Bo, He Xiangqi, et al. Spatial-temporal variations and influencing factors of glacial lakes in Tibet based on optimal parameters based geographical detector[J]. Journal of Glaciology and Geocryology, 2023, 45(6): 1950-1960.

[41] Fu Bojie, Yang Zhijian, Wang Yanglin, et al. Mathematical modeling of soil moisture spatial distribution in loess hill slopes[J]. Science in China (Series D: Earth Sciences), 2001, 31(3): 185-191.

[42] Chang Xueshang, Chang Guoqiao. Advances in research and prospect on soil moisture in arid and semi-arid areas[J]. Journal of Desert Research, 2021, 41(1): 156-163.

[43] Xu Bin, He Baojun, Li Zongjie, et al. Characteristics of spatio-temporal variability of soil moisture during the growing season in the mountainous alpine area and its influencing factors[J]. Acta Ecologica Sinica, 2024, 44(18): 8397-8408.

[44] Lin Pengfei, Zhu Xi, He Zhibin, et al. Research progress on soil moisture temporal stability[J]. Acta Ecologica Sinica, 2018, 38(10): 3403-3413.

[45] Yang Fei, Feng Xiang, Zhang Feimin, et al. Characteristics of vegetation change and its relationship with climate in the Qilian Mountains over the past 40 years[J]. Journal of Arid Meteorology, 2024, 42(3): 385-394.

[46] Jin Zizhen, Qin Xiang, Zhao Qiudong, et al. Characteristics of runoff variation during ablation season in Laohugou watershed of western Qilian Mountains[J]. Arid Land Geography, 2023, 46(2): 178-190.

[47] Yang Xin, Xue Huazhu, Dong Guotao, et al. Vegetation changes and driving factors in the Qilian Mountains during 1982–2022[J]. Chinese Journal of Ecology, 2024, 43(6): 1576-1586.

[48] Fu Jianxin, Cao Guangchao, Li Lingqin, et al. Spatiotemporal variation of air temperature in the middle and eastern parts of the Qilian Mountains and the nearby regions during the period of 1960—2014[J]. Arid Zone Research, 2018, 35(3): 549-561.

[49] Chen Fahu, Xie Tingting, Yang Yujie, et al. Discussion of warming and wetting trend and its future variation in the drylands of northwest China under global warming[J]. Scientia Sinica (Terrae), 2023, 53(6): 1246-1262.

[50] Ding Yihui, Liu Yanju, Xu Ying, et al. Regional responses to global climate change: Progress and prospects for trend, causes, and projection of climatic warm-wetting in northwest China[J]. Advances in Earth Science, 2023, 38(6): 551-562.

[51] Zhou S, Williams A P, Lintner B R, et al. Diminishing seasonality of subtropical water availability in a warmer world dominated by soil moisture atmosphere feedbacks[J]. Nature Communications, 2022, 13: 5756, doi: 10.1038/s41467-022-33473-9.

[52] Li W, Yan D, Weng B, et al. Nonlinear effects of surface soil moisture changes on vegetation greenness over the Tibetan Plateau[J]. Remote Sensing of Environment, 2024, 302: 113971, doi: 10.1016/j.rse.2023.113971.

[53] Thomas A, Yadav B K, Šimůnek J. Water uptake by plants under nonuniform soil moisture conditions: A comprehensive numerical and experimental analysis[J]. Agricultural Water Management, 2024, 292: 108668, doi: 10.1016/j.agwat.2024.108668.

[54] Gou Q, Zhu Q. Response of deep soil moisture to different vegetation types in the Loess Plateau of northern Shannxi, China[J]. Scientific Reports, 2021, 11: 15098, doi: 10.1038/s41598-021-94584-2.

[55] Wang Yujie, Qin Dahe. Influence of climate change and human activity on water resources in arid region of northwest China: An overview[J]. Climate Change Research, 2017, 13(5): 483-493.

[56] Zhang K, Wang Q, Chao L, et al. Ground observation based analysis of soil moisture spatiotemporal variability across a humid to humid transitional zone in China[J]. Journal of Hydrology, 2019, 574: 903-914.

[57] Guillod B P, Orlowsky B, Miralles D G, et al. Reconciling spatial and temporal soil moisture effects on afternoon rainfall[J]. Nature Communications, 2015, 6: 6443, doi: 10.1038/ncomms7443.