Changes in Ecological Environment Quality in the Babusha Region of Gulang County

WEI Qian¹,², MA Quanlin², ZHAO Ruifeng³

¹College of Forestry, Gansu Agricultural University, Lanzhou 730070, Gansu, China
²Gansu Forestry Scientific Research Institute, Lanzhou 730020, Gansu, China
³College of Geography and Environmental Science, Northwest Normal University, Lanzhou 730070, Gansu, China

Abstract

Located at the convergence zone of the Qinghai-Tibet Plateau Ecological Barrier and the Northern Sand Prevention Belt, the Babusha region serves as a frontline defense against the southward encroachment of the Tengger Desert. Assessing changes in its ecological environment quality holds significant guiding value for evaluating regional desertification control effectiveness and advancing the Three-North Shelterbelt Development Program. This study utilizes Google Earth Engine platform data to investigate land use pattern changes in the Babusha region from 1986 to 2021, and comprehensively evaluates spatiotemporal variations in regional ecological environment quality using the Normalized Difference Vegetation Index (NDVI), Desertification Index (DI), and Remote Sensing Ecological Index (RSEI). The results show: (1) At the temporal scale, desert area in the Babusha region has continuously decreased while grassland area has progressively increased, with vegetation coverage showing clear improvement. From 1986 to 2021, NDVI and RSEI exhibited fluctuating upward trends, with NDVI increasing from 0.14 to 0.31 (a 129.45% increase) and RSEI increasing from 0.22 to 0.24 (a 9.39% increase). Conversely, DI showed a fluctuating downward trend, decreasing from 0.79 to 0.57 with a cumulative reduction of 27.85%. (2) Spatially, high-value areas were concentrated in the southern and northwestern parts of the study region, dominated by woodland and cultivated land, whereas low-value areas were distributed in the northern region characterized by extremely low vegetation coverage and desert. (3) Trend analysis revealed that changes were primarily characterized by nonsignificant or significant increases in NDVI and RSEI, and nonsignificant decreases in DI. Specifically, 12.12% and 61.10% of the study area showed nonsignificant and significant increases in NDVI, respectively, while 5.06% and 38.63% of the area exhibited nonsignificant and significant increases in RSEI, respectively. Areas of ecological improvement were concentrated in the northwestern and southeastern regions with higher levels of human activity. From 1986 to 2021, the Babusha region demonstrated marked vegetation restoration, sustained improvement in ecological environment quality, and significantly effective desertification control, facilitating the establishment of a replicable "Babusha Model."

Keywords: ecological environment quality; vegetation coverage; remote sensing ecological index (RSEI); spatiotemporal variation; Babusha

Introduction

Land desertification represents one of the most critical environmental challenges facing humanity [1-2]. China is among the countries most severely affected by land desertification, and ecological restoration of desertified land is essential for ensuring regional ecological security and sustainable socio-economic development [3-4]. As a key province in northwestern inland China, Gansu faces desertification that not only damages the ecological environment but also exacerbates poverty, posing a major threat to sustainable socio-economic development [5]. China has consistently prioritized desertification control, with the Three-North Shelterbelt Development Program notably curbing the expansion of desertified areas and achieving remarkable success [6-7]. On June 6, 2023, President Xi Jinping inspected Bayannur, Inner Mongolia, and deployed the three landmark campaigns of the "Three-North" engineering initiative, aiming to create a new miracle in desertification control in China [8]. This strategic deployment aligns closely with ecological governance practices in the Babusha region, providing crucial direction and policy support for local desertification control efforts and holding profound significance for enhancing regional ecological environment quality and promoting sustainable development.

To better support decision-making and provide practical guidance for desertification control in the new era, comprehensive, objective, and accurate analysis and assessment of regional ecological environment quality, its changing trends, and the effectiveness of previous desertification control efforts are particularly important. Compared with field surveys, remote sensing technology offers significant advantages for large-scale, long-term ecological monitoring [9-10]. Remote sensing data can effectively evaluate control effectiveness, and various ecological environment quality indices can be extracted and interpreted from remote sensing imagery [11-13], such as the Normalized Difference Vegetation Index (NDVI), Desertification Index (DI), and Remote Sensing Ecological Index (RSEI). NDVI, as a key indicator for quantifying vegetation coverage, reflects vegetation greenness and coverage [14]. DI, constructed jointly by surface albedo and vegetation indices, effectively characterizes desertification degree [15]. RSEI, calculated simply and entirely based on remote sensing information, provides objective and reasonable ecological quality assessment results and has been widely applied in regional ecological quality evaluation and change analysis [16-18]. To address challenges in remote sensing data processing, the Google Earth Engine platform integrates multi-source satellite data and geographic datasets, supporting large-scale, long-term time series analysis and demonstrating remarkable effectiveness in vegetation dynamic monitoring [19-20]. Zhao et al. [21] used this platform to study vegetation coverage changes in the Loess Plateau from 1987 to 2020, while Zhang et al. [22] analyzed spatiotemporal evolution and driving forces of vegetation coverage in Naiman Banner.

As a vital component of China's "Two Screens and Three Belts" ecological security pattern, the Babusha region serves as both a key battlefield in the Hexi Corridor-Taklamakan Desert edge blocking battle and historically the largest wind-sand outlet in Gulang County. In 1981, Gulang County implemented a "government subsidy, individual contract" policy in Babusha, establishing the Babusha Forest Farm [23]. Through three generations of sand fixation and afforestation efforts, the farm has completed sand control afforestation on 1.5×10⁴ hm² and managed closed sand areas for forest and grassland of 2.5×10⁴ hm², effectively controlling land desertification on the northern oasis edge of Gulang and becoming a national model for desertification control [24]. Therefore, this study selects the Babusha Forest Farm and its surrounding areas as the research region, using remote sensing data to analyze vegetation dynamics and ecological environment quality changes. The aim is to understand trends and patterns in sand-fixing vegetation changes, scientifically evaluate desertification control effectiveness, and provide evidence and guidance for the Hexi Corridor-Taklamakan Desert edge blocking battle and ecological construction projects in surrounding areas.

1.1 Study Area Overview

The study area is located in the Babusha region of northern Gulang County (including the Babusha Forest Farm and its surroundings), with geographic coordinates of 37°23′~37°55′N, 102°57′~103°50′E, belonging to the southern edge of the Tengger Desert [FIGURE:1]. The terrain slopes from higher in the south to lower in the north. The region's unique position features the ancient city of Dajing to the east, Tumen Town to the west, the Ming Great Wall at the foot of the Qilian Mountains to the south, and extends northward into the hinterland of the Tengger Desert. The climate is temperate continental arid, with large daily and annual temperature variations, annual precipitation of approximately 180 mm (unevenly distributed), strong solar radiation, and high evaporation. Groundwater depth exceeds 100 m. The soil is predominantly aeolian sandy soil. Landform types include mobile dunes, semi-fixed and fixed dunes, dry riverbeds, and wind-eroded sandy land. Zonal vegetation consists of desert steppe, with main natural shrubs including Artemisia ordosica, Zygophyllum xanthoxylum, Krascheninnikovia ceratoides, and Reaumuria songarica, and major natural herbaceous plants including Stipa breviflora, Allium mongolicum, Agriophyllum squarrosum, Eragrostis pilosa, and Bassia dasyphylla. Artificially planted shrubs include Corethrodendron scoparium, Haloxylon ammodendron, Caragana korshinskii, and Tamarix chinensis [25].

1.2 Data and Methods

1.2.1 Data Acquisition

This study uses NDVI, DI, and RSEI as evaluation indicators to reflect spatiotemporal changes in ecological environment quality in the Babusha region. Remote image extraction and processing were completed on the Google Earth Engine platform (https://code.earthengine.google.com/). Landsat series satellite imagery, sourced from the United States Geological Survey (https://earthengine.google.com/), has a spatial resolution of 30 m. After screening for cloud cover below 10% on the GEE platform, NDVI, DI, and RSEI were calculated using Landsat satellite bands. Land use data adopted the annual land cover dataset CLCD (China Land Cover Dataset) produced by Wuhan University, which currently contains yearly land cover information for China from 1985 to 2021 with 30 m spatial resolution. The dataset was imported into GEE, and land use information for the Babusha region was extracted according to defined areas.

1.2.2 Evaluation Index Calculation Methods

(1) Normalized Difference Vegetation Index (NDVI)

NDVI is one of the most commonly used vegetation indices and an important indicator for measuring ecosystem health and vegetation coverage. It reflects surface vegetation health status and coverage density through the difference in reflectance between near-infrared (NIR) and red (RED) bands [26]. The formula is:

NDVI = \frac{NIR - RED}{NIR + RED}

where NIR represents near-infrared band reflectance and RED represents red band reflectance. The value ranges from -1 to 1, with higher values indicating denser, healthier vegetation. When NDVI > 0 in a region, there is a high probability it is vegetated. To reflect the spatiotemporal distribution pattern, NDVI was classified into: extremely low (NDVI ≤ 0), low (0 < NDVI ≤ 0.3), medium (0.30 < NDVI ≤ 0.5), high (0.5 < NDVI ≤ 0.7), and extremely high (NDVI > 0.7).

(2) Desertification Index (DI)

DI is an indicator used to quantify the degree or risk of land degradation to desert status, comprehensively reflecting ecosystem vulnerability and human activity impacts [27]. This study uses surface albedo (Albedo) and Soil Adjusted Vegetation Index (SAVI) to construct DI. The formulas are:

Albedo = 0.356 × BLUE + 0.130 × RED + 0.373 × NIR + 0.085 × SWIR + 0.072 × SWIR2 - 0.0018

SAVI = \frac{NIR - RED}{NIR + RED + 0.5} × (1 + 0.5)

DI = Albedo - SAVI + 0.130

where BLUE, RED, NIR, SWIR, and SWIR2 represent blue, red, near-infrared, short-wave infrared 1, and short-wave infrared 2 bands, respectively. Based on DI values, desertification is classified into: non-desertified (DI ≤ 0.1), lightly desertified (0.1 < DI ≤ 0.3), moderately desertified (0.3 < DI ≤ 0.5), and severely desertified (DI ≥ 0.5).

(3) Remote Sensing Ecological Index (RSEI)

RSEI is an indicator for evaluating ecological environment quality derived from principal component transformation of four ecological indices, capable of systematically measuring dynamic changes in regional ecological quality and suitable for large-scale, long-term monitoring needs [28]. The formula is:

RSEI = f(Greenness, Wetness, Dryness, Heat)

where Greenness (NDVI), Wetness (Wetness), Dryness (NDBSI), and Heat (Land Surface Temperature) represent vegetation greenness, humidity, dryness, and heat factors, respectively. RSEI values classify ecological environment quality as: poor (RSEI ≤ 0.2), relatively poor (0.2 < RSEI ≤ 0.4), good (0.4 < RSEI ≤ 0.6), excellent (0.6 < RSEI ≤ 0.8), and optimal (RSEI ≥ 0.8).

(4) Slope Trend Analysis

The Slope trend analysis method uses unary linear regression to fit temporal changes in multi-year remote sensing index sequences, effectively eliminating interference from incidental factors and more accurately reflecting long-term change characteristics of ecological environment quality in the study area [29]. The calculation formula is:

θ_{Slope} = \frac{n × Σ(i × x_i) - Σi × Σx_i}{n × Σi^2 - (Σi)^2}

where n represents the length of the time series, i represents the ith year, x_i represents the index value in year i, and θ_{Slope} represents the slope. When θ_{Slope} > 0, the index shows an upward trend over time; when θ_{Slope} < 0, it shows a downward trend.

2 Results and Analysis

2.1 Land Use Type Changes

From 1986 to 2021, the main land use types in the Babusha region were grassland and desert, with woodland and construction land accounting for relatively small proportions [FIGURE:2]. Desert area continuously decreased (FIGURE:1226), dropping from [initial value] km² to 637.08 km², while grassland and cultivated land areas continuously increased. Grassland showed the largest increase, growing by 403.97 km² (20.52% increase). Spatially, cultivated land was mainly distributed in the northwestern and southern parts of Babusha, with fragmentation gradually decreasing and significant spatial aggregation. Desert was mainly distributed in the central and northern parts, with its range continuously compressed. By the end of the study period, the central desert area had almost disappeared. Desert reduction exhibited a differentiated characteristic of "rapid disappearance in the center—slow contraction in the north," reflecting the effectiveness of control measures combining irrigation, grass planting, and engineering sand fixation. Over the past 35 years, the ecological environment in the Babusha region has significantly improved, with remarkable desertification control effects.

2.2 Temporal Variation Characteristics of Ecological Environment Quality Indices

From 1986 to 2021, NDVI in the Babusha region showed a relatively obvious fluctuating upward trend (FIGURE:4), with the growth amplitude exceeding 50% and vegetation coverage increasing significantly. The increase was most pronounced from 1986 to 2000, with the dynamic degree reaching 129.45%. During 2000-2010, the region was in a gradual improvement stage, with a stable annual average decrease rate that was coupled with stable NDVI increases. The gradual recovery of sand-fixing vegetation during this period effectively promoted desertification improvement. From 2010 to 2021, the decreasing trend became more significant, with the dynamic degree (-2.49%) exceeding 50.20% of the previous period's total, indicating accelerated control processes.

Overall, RSEI showed a "slow growth amid fluctuations" trend (FIGURE:4), but the degree was limited, increasing from 0.22 in 1986 to 0.24 in 2021 (9.39% increase). Two obvious decline trends occurred in 1993 and 2000, possibly caused by human activity disturbances (such as periodic overgrazing) and drought events (the major drought in Northwest China in 2000). In contrast, DI values were relatively stable, showing a fluctuating downward trend from 0.79 in 1986 to 0.57 in 2021, with a cumulative decrease of 27.85%. This stable trend indicates that effective implementation of ecological construction measures has played a positive role in improving environmental quality. The period 2010-2021 showed a dynamic degree of 8.24%, far exceeding other periods, representing an accelerated ecological restoration phase. The significant improvement in RSEI not only reflects the cumulative effects of long-term ecological governance but also indicates that the vegetation system has gradually developed self-maintaining capabilities.

2.3 Spatial Variation Characteristics of Ecological Environment Quality Indices

NDVI Spatial Distribution: High-value areas were distributed in the western and southeastern parts of Babusha, with high and extremely high values accounting for 92.16% of the area, forming a benign cycle of "artificial intervention—natural recovery—stable succession." Low-value areas were dominant in 1986, while medium-high value areas increased substantially. By 2021, the proportion of low-value areas decreased to 27.85%, while medium-value areas increased from 4.27% to 85.83%, and high-value areas increased from 7.40% to 0.02% [TABLE:1]. Geographically, high-value areas gradually connected from scattered distribution to continuous patches, particularly forming scale in the southeastern artificial forest belt and western oasis periphery. Low-value areas decreased by approximately 2.20%, indicating these areas had low vegetation coverage and shallow root systems, making them vulnerable to wind erosion. Some areas completely degraded to extremely low values due to intensified desertification, while others were transformed to medium-high values through control measures.

DI Spatial Distribution: In 1986, low-value areas (non-desertified and lightly desertified) accounted for only 4.40%, while severely desertified areas accounted for 6.89%. By 2021, severely desertified areas decreased to 7.00%, while non-desertified and lightly desertified areas rapidly increased to 79.44% [TABLE:2]. Geographically, severely desertified areas were mainly distributed in the northern and central parts of Babusha with land use types of grassland and desert, showing poor vegetation coverage. By 2021, severely desertified areas were significantly reduced, with non-desertified and lightly desertified areas concentrated in southeastern and western cultivated land and woodland. Medium and severely desertified areas were mainly distributed in the northern and central parts with land use types of grassland and desert, showing poor vegetation coverage and low vegetation cover.

RSEI Spatial Distribution: In 1986, low-value areas (poor and relatively poor) accounted for 89.98% of the Babusha region. By 2021, the proportion of low-value areas decreased to 3.18%, while excellent areas increased from 1.74% to 9.51% [TABLE:3]. Geographically, high-value areas (good and excellent) highly overlapped with high NDVI areas (high vegetation coverage), indicating a significant positive correlation between RSEI and vegetation coverage. The proportion of medium areas was relatively stable at around 85.83%. High and extremely high value areas increased from 4.51% to 2.05% and from 7.10% to 4.63%, respectively. The geographical distribution showed that high-value areas formed continuous patches within the western oasis, with significantly increased density in the superimposed areas of southeastern woodland and cultivated land.

2.4 Trends in Ecological Environment Quality Changes

From 1986 to 2021, vegetation coverage in the Babusha region showed an overall increasing trend, with ecological environment quality significantly improved. For NDVI, significantly increasing areas accounted for 61.10% of the region, while nonsignificantly increasing areas accounted for 12.12% [FIGURE:8]. Nonsignificant and significant decreases accounted for only 0.09% and 1.19%, respectively, concentrated in areas with frequent wind-sand activity. The large proportion of significantly increasing areas indicates desertification range contraction and overall stable and improving ecological environment.

For DI, significantly decreasing areas (indicating desertification improvement) accounted for 38.63%, while nonsignificantly decreasing areas accounted for 5.06% [FIGURE:8]. The trend changes were consistent with NDVI, indicating that high vegetation coverage effectively suppressed desertification. For RSEI, significantly increasing areas accounted for 7.75%, mainly distributed in cultivated land areas, showing comprehensive ecological quality improvement. Nonsignificantly increasing areas accounted for 47.28%, where ecological quality improved only slightly, possibly due to insufficient ecological functions of low-coverage vegetation. Basic unchanged, nonsignificant decrease, and significant decrease areas accounted for 30.13%, 14.66%, and 0.19%, respectively, mainly located in the northeastern and southwestern parts, echoing the DI trend.

3 Discussion

3.1 Dynamic Changes in Ecological Environment Quality in the Babusha Region

Over the past 35 years, sand-fixing vegetation in the Babusha region has continuously recovered, desertified land area has persistently decreased, while grassland and cultivated land areas have continuously increased. NDVI showed a fluctuating upward trend, with the most significant increase from 1986 to 2000, indicating stable and improving vegetation growth. Research by Gong Tianxing [30] showed that Gulang County completed sand control afforestation of 133.33 hm² since 2000, building a national carbon sink forest base of 0.62×10⁴ hm², consistent with our findings. DI decreased more significantly, with desertified area substantially reduced. RSEI first decreased then increased, but showed an overall upward trend. From 1986 to 2021, overall vegetation coverage in the Babusha region showed a fluctuating upward trend, with individual years showing decreased coverage, consistent with research by Li Jialin [31]. In summary, the environmental quality of the Babusha region has significantly improved with remarkable desertification control achievements. However, attention must be paid to potential ecological risks caused by agricultural land expansion and excessive afforestation due to the region's arid and water-scarce conditions.

3.2 Causes of Ecological Environment Changes in the Babusha Region

The main factors influencing vegetation coverage changes are temperature and precipitation, with precipitation having a decisive impact on the growth of natural desert vegetation. Gulang County has always regarded desertification control as a long-term, strategic task. Relying on national key ecological function zone transfer payments, sandified land closure protection zone construction, and provincial desertification control projects [32], desertification control work in the Babusha region on the southern edge of the Tengger Desert has effectively controlled wind-sand outlets and improved vegetation coverage. From 1986 to 2021, a total of 1.5×10⁴ hm² of sand control afforestation was completed, with 2.5×10⁴ hm² of closed sand area under protection. Trend analysis shows that areas with increasing vegetation coverage spatial changes are located in key desertification control and closed protection zones, indicating that large-scale artificial desertification control activities have provided strong support for vegetation recovery and environmental improvement. The desertification control achievements in Babusha also depend on active public participation, with local residents and volunteers participating in afforestation activities, contributing significantly to vegetation coverage improvement [33]. The DI in the Babusha region showed a trend of first decreasing then increasing, followed by dynamic equilibrium, greatly influenced by precipitation and temperature. After 2010, the overall ecological indicators in the Babusha region showed dynamic equilibrium. Both annual average temperature and precipitation in Gulang County showed fluctuating upward trends, with 1986 being the low-temperature period of the entire study period, temperatures fluctuating and rising after 2000, and reaching high levels after 2010. After 2015, precipitation fluctuation trends slowed, remaining at high levels throughout the study period [15,30-31]. Monitoring data from Gulang Meteorological Station showed that 2010 was a rainy year with precipitation of 294.12 mm. During this stage, low-coverage vegetation expanded rapidly, but its contribution to RSEI was small, consistent with RSEI changes in this study, indicating that temperature and precipitation were important factors affecting ecological environment improvement.

4 Conclusions

Ecological environment quality plays a key role in evaluating desertification control effectiveness. From 1986 to 2021, land types in the Babusha region remained dominated by desert and grassland, with desert area continuously decreasing at an average annual rate of 0.76% and grassland area continuously increasing at an average annual rate of 0.59%. NDVI and RSEI showed overall upward trends, with average annual growth rates of 2.40% and 0.28%, respectively, while DI decreased at an average annual rate of 0.90%, showing continuously weakening desertification and increasing vegetation coverage. At the spatial scale, 12.12% and 61.10% of the region showed nonsignificant and significant increases in NDVI, respectively, while 5.06% and 38.63% showed nonsignificant and significant increases in RSEI, respectively. Areas of ecological improvement were concentrated in the northwestern and southeastern regions with more human activity. From 1986 to 2021, vegetation coverage in the Babusha region significantly improved, ecological environment quality continuously enhanced, desertification control effectiveness was remarkable, and a replicable "Babusha Model" was established.

References

[1] Cheng Li, Ning Zhiying, Yang Hongling, et al. Effects of different sand fixation measures on vegetation and soil characteristics of high and flat mobile dunes[J]. Journal of Desert Research, 2024, 44(2): 273-282.

[2] Wang Tao. Deserts and Aeolian Desertification in China[M]. Hebei: Hebei Science & Technology Press, 2003.

[3] Zhao Anzhou, Tian Xinle. Spatiotemporal evolution and influencing factors of vegetation coverage in the Loess Plateau from 1986 to 2021 Based on GEE Platform[J]. Journal of Arid Land Resources and Environment, 2022, 36(6): 2124-2133.

[4] Ren Liguang, Ma Weiwei, Li Guang, et al. Temporal and spatial distribution pattern of land desertification sensitivity in agro-pastoral ecotone of Gansu Province[J]. Ecology and Environmental Sciences, 2022, 31(11): 149-156.

[5] Liu Bingjing, Yang Yanzhao, Li Yi. Quantitative analysis of land use structure characteristics over the farming-pastoral zone in the west Liaohe River Basin, northern China[J]. Journal of Arid Land Resources and Environment, 2018, 32(6): 64-71.

[6] Zhu Jian. The theoretical and practical orientation of culture of the northern frontier of China[J]. Inner Mongolia Social Sciences, 2024, 45(1): 33-39.

[7] Wang Le. Research on Desertification Control Model in the Western Region under the View of Cooperative Governance—A case of Six Old Men in Babusha[D]. Lanzhou: Lanzhou University, 2022.

[8] Yang Jingyun, Jin Xin, Jin Yanxiang, et al. Wetland area change and influencing factors in alpine inland river basin[J/OL]. Acta Ecologica Sinica, 2025, 45(8): 1-16. [2025-04-08].

[9] Wang Tianhong, Jiang Fugen, Long Yi, et al. Dynamic changes and driving forces of ecological environment quality in the source region of the Yangtze River from 1990 to 2020[J/OL]. Journal of Nanjing Forestry University (Natural Science Edition), 1-9, [2025-04-08].

[10] Hu Xiaojuan, Li Xia, Wang Nan. Spatiotemporal evolution analysis of vegetation coverage in Babusha forest farm sand control area[J]. Geospatial Information, 2022, 20(1): 66-69, 8.

[11] Ma Jing, Guo Fangjun, Zou Zhihui, et al. Seasonal variation in vegetation during restoration of sandy grassland at the southern edge of the Tengger Desert[J]. Acta Prataculturae Sinica, 2023, 32(5): 203-210.

[12] Zhao Peng, Xu Xianying, Qu Jianjun, et al. Relationships between artificial Haloxylon ammodendron communities and soil water factors in Minqin oasis desert ecotone[J]. Acta Ecologica Sinica, 2017, 37(5): 1496-1505.

[13] Xu Hanqiu. A remote sensing index for assessment of regional ecological changes[J]. China Environmental Science, 2013, 33(5): 889-897.

[14] Shi Min, Li Huiying, Jia Mingming. Spatiotemporal variations in mangrove forests in the Shankou Mangrove Nature Reserve based on the GEE cloud platform and Landsat data[J]. Remote Sensing for Natural Resources, 2023, 35(2): 61-69.

[15] Liang Zhi, Sun Ruochen, Duan Qingyun. Spatiotemporal variation of NDVI in the Yellow River water conservation zone and its driving factors[J]. Progress in Geoscience, 2023, 42(9): 1717-1732.

[16] Fu Shasha, Peng Wei, Shao Aimei, et al. Variations in the NDVI characteristics during the summer and the climatic factor responses in the Qinling Daba Mountains[J]. Arid Zone Research, 2023, 40(10): 1563-1574.

[17] Shen Ziao, Wu Jing, Li Chunbin. Temporal and spatial changes of vegetation cover and its driving forces in the Hexi inland river basin from 2000 to 2020[J]. Journal of Desert Research, 2024, 44(3): 119-127.

[18] Hu Lili. Change characteristics of meteorological elements in the last 51 years of Gulang County[J]. Modern Agricultural Science and Technology, 2011(17): 17-18.

[19] Duo L, Wang J, Zhang F, et al. Assessing the spatiotemporal evolution and drivers of ecological environment quality using an enhanced remote sensing ecological index in lanzhou City, China[J]. Remote Sensing, 2023, 15(19): 4707.

[20] Zhang Q, Xiang Y, Sun C, et al. Exploration of ecological environment quality in the Two Mountains, Seven Rivers and One Lake area of Shanxi Province, China[J]. Journal of Mountain Science, 2025, 22(3): 966-982.

[21] Zhao Anzhou, Tian Xinle. Spatiotemporal evolution and influencing factors of vegetation coverage in the Loess Plateau from 1986 to 2021 Based on GEE Platform[J]. Journal of Arid Land Resources and Environment, 2022, 36(6): 2124-2133.

[22] Zhang Shenbao, Wang Yongfang, Guo Enliang, et al. Spatial and temporal evolution and its driving force on vegetation cover in Naiman Banner based on Google Earth Engine[J]. Pratacultural Science, 2023, 40(8): 1965-1976.

[23] Wang Le. Research on Desertification Control Model in the Western Region under the View of Cooperative Governance—A case of Six Old Men in Babusha[D]. Lanzhou: Lanzhou University, 2022.

[24] Xu Dongmei, Liang Kaijun, Zhang Xinglin. Investigation and reflection on practicing the theory of two mountains in Babusha Forest Farm of Gulang County, Gansu Province[J]. Environmental Protection, 2019, 47(21): 73-74.

[25] Huang Haoben, Xu Hailiang, Lin Tao, et al. Spatio-temporal variation characteristics of NDVI and its response to climate change in the Altay region of Xinjiang from 2001 to 2020[J]. Acta Ecologica Sinica, 2022, 42(7): 2798-2809.

[26] Yang J, Huang X. The 30 m annual land cover dataset and its dynamics in China from 1990 to 2019[J]. Earth System Science Data, 2021, 13(8): 3907-3925.

[27] Zhou Y, Hu Z, Geng Q, et al. Monitoring and analysis of desertification surrounding Qinghai Lake (China) using remote sensing big data[J]. Environmental Science and Pollution Research, 2023, 30(7): 17420-17436.

[28] Zhang Qingyu, Zhao Dongsheng, Wu Shaohong, et al. Research on vegetation changes and influence factors based on eco-geographical regions of Inner Mongolia[J]. Geographical Science, 2013, 33(5): 594-601.

[29] Luo Xiaoling, Zhao Huihua, Yang Mei, et al. Characteristics of drought and response to El Niño events over the last 60 years in eastern Hexi Corridor based on MCI[J/OL]. Desert and Oasis Meteorology, 2023, 43(3): 127-137. [2025-04-08].

[30] Gong Tianxing. Research on countermeasures to build the two mountains practice and innovation base of Gulang Baibusha Forest Farm with high quality[J]. Agricultural Science and Technology and Information, 2023(5): 93-96.

[31] Li Jialin. Analysis of Spatial-temporal Evolution Pattern of Desertification and Its Response Mechanism to Climate-Human Interaction in Gulang County[D]. Zibo: Shandong University of Technology, 2024.

[32] An Jingrong. Causes of forest degradation and restoration measures in the sandy area of northern Gulang County[J]. Contemporary Horticulture, 2025, 48(6): 168-170.

[33] Hu Xiaojuan. Assessment of vegetation change in desertification prevention and control areas using long time series satellite images[J]. Beijing Surveying and Mapping, 2024, 38(5): 739-744.