Assessing the ecological effects of the Three-North Shelter Forest Program by a novel ecosystem service index

Assessing the ecological effects of the Three-North Shelter Forest Program by a novel ecosystem service index | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 11 November 2025 V1 Latest version Share on Assessing the ecological effects of the Three-North Shelter Forest Program by a novel ecosystem service index Authors : Chao Lin 0009-0007-5244-1405 , Qiang Bie [email protected] , Liang Huajun , Xiangxiang Huang 0009-0009-2484-7701 , and Liu Hang Authors Info & Affiliations https://doi.org/10.22541/au.176284287.78816269/v1 336 views 128 downloads Contents Abstract Introduction Data sources and methods Data source Research Framework Research methodology A novel CES index Results ESs change induced by land use Spatiotemporal variation characteristics of CES TNSFP ecological effects based on spatiotemporal evolution of CES Discussion Limitations and future research Conclusions Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Ecosystem services (ESs) are vital indicators for evaluating ecological restoration outcomes and regional sustainability. The Three-North Shelter Forest Program (TNSFP)—China’s largest and longest-running ecological program—plays a crucial role in improving ecosystem functions and ecological security in arid and semi-arid regions. This study employed the PLUS model coupled with CMIP6 climate scenarios (SSP119, SSP245, SSP585) to simulate future land-use patterns, and applied the InVEST and related models to quantify four key ESs: water yield (WY), habitat quality (HQ), carbon storage (CS), and soil conservation (SC). Furthermore, a comprehensive ecosystem services (CES) index was constructed to evaluate the ecological effects of the TNSFP in arid and semi-arid regions of China from 2000 to 2020. Results indicate: (1) From 2000 to 2020, cropland (2.80×104 km2) and urban (1.19×104 km2) in the study area expanded continuously, while water (1.71×104 km2) declined markedly. During this period, ESs showed distinct trends: WY and HQ remained generally stable, CS increased steadily (+2.32×108 t), and SC improved significantly (+4.29×108 t). By 2040, grassland and unused land exhibit the most pronounced but opposite changes across the three scenarios. SSP119 enhances HQ and CS but weakens WY and SC; SSP245 maintains balance; SSP585 strengthens WY and SC but offers limited HQ and CS gains. (2) The CES showed a pattern of “decline–stabilization–recovery”, decreasing from 0.3786 (2000) to 0.3711 (2020), and then rebounding under all future scenarios (SSP119: 0.3863 > SSP245: 0.3767 > SSP585: 0.3756). High CES values are concentrated in the eastern and southern mountainous regions, forming ecological barrier zones, while low CES areas are distributed in desert basins and arid plains. (3) The TNSFP significantly promoted ecosystem restoration and service enhancement, forming a “point–belt–patch” ecological restoration pattern that strengthened regional stability and service provision. Future low-emission and conservation-oriented development pathways are most beneficial for sustaining and improving overall ecosystem functions. Assessing the ecological effects of the Three-North Shelter Forest Program by a novel ecosystem service index Lin Chao a,b,c , Bie Qiang a,b,c, 11 ∗ Corresponding author. E-mail address: [email protected] (B. Qiang)., Liang Huajun a , Huang Xiangxiang a , Liu Hang a a . Faculty of Geomatics, Lanzhou Jiaotong University, Lanzhou 730070, Gansu, China b . National-Local Joint Engineering Research Center of Technologies and Applications for National Geographic State Monitoring, Lanzhou 730070, Gansu, China c . Key Laboratory of Science and Technology in Surveying & Mapping Gansu Province, Lanzhou, Gansu, China Abstract: Ecosystem services (ESs) are vital indicators for evaluating ecological restoration outcomes and regional sustainability. The Three-North Shelter Forest Program (TNSFP)—China’s largest and longest-running ecological program—plays a crucial role in improving ecosystem functions and ecological security in arid and semi-arid regions. This study employed the PLUS model coupled with CMIP6 climate scenarios (SSP119, SSP245, SSP585) to simulate future land-use patterns, and applied the InVEST and related models to quantify four key ESs: water yield (WY), habitat quality (HQ), carbon storage (CS), and soil conservation (SC). Furthermore, a comprehensive ecosystem services (CES) index was constructed to evaluate the ecological effects of the TNSFP in arid and semi-arid regions of China from 2000 to 2020. Results indicate: (1) From 2000 to 2020, cropland (2.80×104 km2) and urban (1.19×104 km2) in the study area expanded continuously, while water (1.71×104 km2) declined markedly. During this period, ESs showed distinct trends: WY and HQ remained generally stable, CS increased steadily (+2.32×108 t), and SC improved significantly (+4.29×108 t). By 2040, grassland and unused land exhibit the most pronounced but opposite changes across the three scenarios. SSP119 enhances HQ and CS but weakens WY and SC; SSP245 maintains balance; SSP585 strengthens WY and SC but offers limited HQ and CS gains. (2) The CES showed a pattern of “decline–stabilization–recovery”, decreasing from 0.3786 (2000) to 0.3711 (2020), and then rebounding under all future scenarios (SSP119: 0.3863 > SSP245: 0.3767 > SSP585: 0.3756). High CES values are concentrated in the eastern and southern mountainous regions, forming ecological barrier zones, while low CES areas are distributed in desert basins and arid plains. (3) The TNSFP significantly promoted ecosystem restoration and service enhancement, forming a “point–belt–patch” ecological restoration pattern that strengthened regional stability and service provision. Future low-emission and conservation-oriented development pathways are most beneficial for sustaining and improving overall ecosystem functions. Keywords: Land use change; Multi-scenario simulation; Comprehensive ecosystem service; Three-North Shelter Forest Program; Arid and semi-arid region Introduction Ecosystem services (ESs) refer to the benefits provided by ecosystems in forming and maintaining the environmental conditions essential for human survival. These manifest as a series of ecological characteristics, functions, and processes that directly or indirectly influence human sustainable well-being Costanza et al., 2017Raymond et al., 2013(; ). Over the past half-century, approximately 60% of global ESs have experienced varying degrees of degradation Carpenter et al., 2005(). Research indicates that the decline in ESs not only exacerbates climate warming but also triggers a series of ecological problems such as soil erosion and habitat degradation, thereby posing a serious threat to regional ecological security and sustainable development Scholes, 2016Zhang et al., 2024a(; ). In recent years, with increasing attention to this field, quantitative assessment of ESs has become a crucial approach for understanding the interrelationship between human activities and the ecological environment Gaylard et al., 2025Li and Wu, 2025Zhang et al., 2025(; ; ). Comprehensively considering key service functions, such as carbon storage, water conservation, soil retention, and habitat quality, is crucial for revealing the mechanisms of regional ecological function evolution and optimizing ecological spatial patterns Bai et al., 2025Seppelt et al., 2011(; ). Arid and semi-arid regions cover over 40% of the world’s land area, where water scarcity, land degradation, and ecosystem decline pose severe challenges to regional socioeconomic sustainability and ecological security Berg and McColl, 2021Borges et al., 2020(; ). The arid and semi-arid regions of China, situated in the heart of the Eurasian continent, cover a vast expanse. As a typical agro-pastoral transition zone and ecological buffer, this area serves not only as a crucial base for grain and livestock production in China but also as a vital ecological safeguard for the North China Plain and regions like Beijing-Tianjin-Hebei Guo et al., 2025Sun et al., 2022(; ). However, the region exhibits significant interannual variability in extreme precipitation indices. Eastern areas have experienced markedly reduced extreme precipitation, showing pronounced aridification trends, while annual precipitation in western regions has increased at a rate of 5.11 mm/10a Hou et al., 2023(). As climate change intensifies, extreme droughts and precipitation events characterized by long intervals and heavy rainfall will occur more frequently Wu et al., 2025Xu et al., 2021(; ). Under the dual impacts of climate change and human activities, this region faces severe ecological challenges including long-term grassland degradation, desertification expansion, and accelerated soil erosion Feng et al., 2024(). To improve the regional ecological environment, China has implemented a series of major ecological projects since the 1970s, the most representative of which is the Three-North Shelter Program (TNSFP) Li et al., 2012(). This large-scale artificial forestry ecological project involves afforestation across China’s Three-North region (Northwest, North China, and Northeast). To enhance ecological conditions, the Chinese government designated this project as a key national economic initiative in 1978. Planned for a 70-year duration, it is executed in seven phases. The second phase (2000–2020) represents a critical period for deepening implementation, expanding coverage, and sustaining ecological progress Li et al., 2022(). Through measures such as afforestation, grassland restoration, and soil and water conservation, the project aims to enhance regional windbreak and sand fixation capabilities, improve land use patterns, and thereby promote the enhancement of ecosystem service functions and the construction of ecological barriers Qiu et al., 2017(). Existing research indicated that the TNSFP exerted positive impacts on the regional ecological environment to a certain extentJi et al., 2022Zhai et al., 2023(; ). However, against the backdrop of global warming and increasing land resource demands driven by socioeconomic development, the dynamic evolution of ecosystem service functions in this region warrants further investigation. This study focuses on assessing the ecological effect of TNSFP in the arid and semi-arid regions of China. We employed the PLUS model integrated with CMIP6 climate models to simulate future land use changes and assessed key ecosystem services including water yield (WY), habitat quality (HQ), carbon storage (CS), and soil conservation (SC) combined with InVEST model. A comprehensive ecosystem services (CES) index was proposed to explore the spatiotemporal evolution of ecological effect under the implementation of the TNSFP. This research provides scientific support for ecological environment management and ecological security barrier construction in arid and semi-arid regions of China. It holds scientific significance and practical value for evaluating the ecological effects of the TNSFP, optimizing regional ecological construction patterns, and enhancing ecosystem service levels. Data sources and methods Study area Fig. 1. Overview of arid and semi-arid regions of China (a) Geographic location (b) Land use types in 2020 (c) Topography of the study area Based on the Comprehensive Agricultural Zoning of China compiled by the National Agricultural Zoning Committee, this study focuses on China’s arid and semi-arid regions (Fig.1) Chen et al., 2015(). This area encompasses six provinces and autonomous regions: Inner Mongolia, Shaanxi, Gansu, Qinghai, Ningxia, and Xinjiang, covering a total area of approximately 3.24 million km 2 . The terrain generally slopes from west to east, featuring complex topography and landforms that exhibit the typical geographical characteristics of interlaced plateaus and basins. The climate transitions gradually from a temperate monsoon climate in the east to a temperate continental climate in the west. Winters are long and cold, while summers are dry and hot. Overall, the region experiences aridity with low precipitation, high water evaporation, and pronounced extreme weather phenomena Xu et al., 2016(). Furthermore, the arid and semi-arid regions of China form the core implementation area of TNSFP. Within the study region, the project covers 82.61% of the TNSFP’s total area and approximately 85.10% of the study area’s total land, holding a pivotal position in the national ecological security framework. This region not only serves as a vital base for China’s livestock and agricultural production but also harbors abundant mineral resources, holding significant strategic importance for regional economic development. Data source The data used in this study primarily include multi-source spatial data including land use, meteorology, socio-economics, topography, soil, and transportation (Table 1). Land use/land cover (LULC) data were derived from the China Multi-period Land Use Remote Sensing Monitoring Dataset (CNLUCC). This dataset constitutes a national-scale, multi-period thematic database of China’s LULC, constructed through manual visual interpretation using Landsat imagery as the primary information source Xu et al., 2018(). Its classification system employs a two-tier classification framework. This study reclassified the primary categories based on land resource attributes and utilization characteristics as six types: cropland, forest, grassland, water, urban, and unused land. This reclassification effectively captures subtle temporal variations in land use patterns across different periods within the study area (Xu et al., 2018). Climate data encompass both historical and future dimensions, primarily utilizing the China Climate Dataset released by Peng Shouzhang’s team Peng, 2025abcPeng et al., 2019(, ; ). This dataset series builds upon the Climatic Research Unit (CRU) Global 0.5° Climate Dataset and WorldClim global climate data. It employs the Delta spatial downscaling method to generate monthly temperature, precipitation, and potential evapotranspiration data at 1 km resolution. Future-period data are derived from the CMIP6 multi-model ensemble results, utilizing representative scenarios such as SSP119, SSP245, and SSP585. To ensure spatial consistency among multi-source data and comparability in model operations, this study uniformly resampled land cover data with a 30 m resolution and other input data to 300 m resolution. In ecosystem service assessments, moderately reducing resolution filters out fine-scale isolated pixel noise while preserving key land-use spatial patterns. This enhances model computational stability and result robustness, unifies spatial baselines for multi-source data to avoid registration errors, and balances spatial precision with computational efficiency at regional scales (e.g., watersheds, arid zones), thereby optimizing assessment accuracy against computational costsBagstad et al., 2018(). Table 1. Data description and sources. Land Use Data Land use/land cover (LULC) 30 m https://www.resdc.cn/ Natural Environmental Data Digital elevation model (DEM) 250 m Normalized difference vegetation index (NDVI) 250 m Soil 1 km Temperature, Precipitation, Potential evapotranspiration 1 km http://data.tpdc.ac.cn/ Harmonized World Soil Database (HWSD) 1 km https://www.fao.org/home/en/ Depth to bedrock 1 km http://globalchange.bnu.edu.cn/research/cdtb.jsp Socio-economic data China population spatial distribution kilometer grid dataset 1 km https://www.resdc.cn/ Gross Domestic Product (GDP) 1 km China Ecological Functional Reserves Vector Data on Nature Reserve Boundaries in China Vector Nighttime Lighting Dataset 500 m https://www.geodata.cn Distance Accessibility Data Poi, Railway, River, Road, Transport Vector http://www.openstreetmap.cn/ Research Framework This study employed observed land use data to derive land use requirements under the SSP119, SSP245, and SSP585 scenarios. Subsequently, the PLUS model was applied in conjunction with driver data to project land use patterns for 2040 based on SSP-RCP projections. Core ecosystem services including WY, HQ, CS, and SC were then assessed using the InVEST model. Additionally, CES was developed to evaluate the ecological effects of TNSFP in arid and semi-arid regions of China. The complete research framework is illustrated in Fig. 2. Fig. 2. Research framework. Research methodology PLUS multi-scenario simulation forecasting of land use SSP-RCPs coupled scenario setting CMIP6 provides multiple future global climate change scenarios by utilizing shared socioeconomic pathways (SSPs) and representative concentration pathways (RCPs) Zhang et al., 2022a(). This study selected three representative scenarios: SSP119 represents a low-emission and sustainable development pathway emphasizing stringent climate policies and green transition; SSP245 is characterized by no significant deviation from historical trends in the region, with socioeconomic and land-use trajectories remaining unchanged; SSP585 represents a more intense state of human activity, describing a future pathway with high carbon emissions and high economic growth Su et al., 2021(). Land use change simulation This study employs PLUS V1.40 to predict land use patterns in the arid and semi-arid regions of China by 2040 under SSP-RCPs and national ecological red line policy constraints using the Cellular Automaton-Markov (CA-Markov) model Gao et al., 2025(). The model’s advantage lies in its ability to more precisely characterize land use change mechanisms under different policy scenarios. Within the LEAS module, expansion data for six land use types were first extracted. Subsequently, based on the study area’s actual conditions, 16 drivers—including population, GDP, nighttime light, transportation accessibility, soil, NDVI, climate, and topography—were selected. Combined with land use expansion data, these drivers were used to obtain suitability probabilities for each land use type via a random forest algorithm. Finally, within the CARS module, the 2000 land use data served as the baseline. By integrating parameters such as land use development probability, future demand, transition matrices, and neighborhood weights, the 2020 land use pattern was simulated and validated against actual 2020 data Zhu et al., 2025(). Results indicate high simulation accuracy (Kappa coefficient 0.84, overall accuracy 0.89), demonstrating the model’s capability to effectively reflect land use change patterns. Building upon this foundation, the 2020 land use data was used to simulate land use patterns under various scenarios for the year 2040. Ecosystem service functioning assessment In this study, we selected four key ecosystem service indicators (WY, HQ, CS, SC) and quantified and spatially mapped current and future ecosystem services at a 300m×300m spatial resolution to assess the ecosystem service functions within the study area. The detailed calculation method was delineated in Table 2. Table 2. Quantitative methods of ESs. WY Y(x) represents the annual WY (mm) per pixel; AET(x) denotes the annual actual evapotranspiration (mm); P(x) indicates the annual precipitation data (mm); Li et al., 2021() HQ Q xj denotes the HQ index for land use type j in grid cell x ; H j denotes the ecological suitability index for land use type j , ranging from 0 to 1; z is the default constant; k is the semi-saturation coefficient; \(D_{\text{xj}}^{z}\) indicates the degree of habitat degradation within grid cell x for land use type j . Zhang et al., 2024b() CS C tot represents total carbon storage; A i denotes the area of land use type; C above indicates above-ground biomass carbon storage; C below signifies below-ground biomass carbon storage; C soil denotes soil organic carbon storage; C dead represents dead organic matter carbon storage. Gao et al., 2023() SC SC represents soil conservation; RLSE and ULSE denote potential soil erosion and actual soil erosion, respectively; R is the rainfall erosion factor; K is the soil erodibility factor; L is the slope length factor; S is the slope gradient factor; C is the vegetation cover factor; and P is the soil and water conservation measures factor. Wang et al., 2024() A novel CES index To reflect and quantify the overall effects of multiple ecosystem services, this study constructs a CES Index. In existing research, various ESs are often assigned equal weights, yet the importance of different ecosystem service functions varies in reality Watson et al., 2019(). Building upon prior research, this study employs the Analytic Hierarchy Process (AHP) to determine the weighting of each service, thereby constructing the CES to compare the overall level of ESs Zhao and Shao, 2023(). This indicator reveals the comprehensive status of regional ESs at the spatial scale, providing scientific reference for government authorities in formulating territorial spatial planning and ecological conservation policies. The calculation formula is as follows: Among these, CES j denotes the comprehensive ecosystem service index for year j , w i represents the weight of ecosystem service type i , S ij indicates the normalized value of ecosystem service type i in year j , and n signifies the number of ecosystem service types. This study constructs a judgment matrix by pairwise comparison of elements within the influence layer and indicator layer. The weights for each ecosystem service type are presented in the following table (Table 3). Table 3. Weighting of different ecosystem service indicators in the study area. Regulatory Services 0.53 Carbon storage 0.55 0.29 Soil conservation 0.45 0.24 Support Services 0.25 Habitat quality 1 0.25 Supply Services 0.22 Water yield 1 0.22 Results Land use evolution in the past and multi-scenaio future From 2000 to 2020, cropland and urban in the study area expanded continuously, while water declined markedly (Fig. 3). Cropland expanded from 25.70×10⁴ km² to 28.50×10⁴ km², a net increase of 2.80×10⁴ km², indicating sustained expansion. Grassland areas initially decreased before increasing, ultimately recovering to their initial levels. Unused land first expanded and then contracted, with a net decrease of 1.71×10 4 km 2 , reflecting concurrent degradation, development, and restoration. Forest decreased by 0.78×10 4 km 2 , while water decreased by 1.71×10 4 km 2 . Urban continued to expand, growing from 2.02×10 4 km 2 to 3.21×10 4 km 2 —a net increase of 1.19×10 4 km 2 —marking the most significant growth. Spatially, cropland expansion concentrated in the eastern and southern agriculturally suitable zones; grassland changes primarily occurred in the northern and western grassland belts, with high overlap between earlier degraded areas and later restored zones; changes in unused land occurred in the western and northern desertification fringe areas, showing a pattern of initial expansion followed by contraction; forest concentrated in the northeastern and southern forest zones; and urban expansion centered on the central and western regions of the study area. Overall, the expansion of cropland and urban primarily relied on the conversion of unused land, water, and portions of forest. These changes reflect both human activities and natural conditions. Fig. 3. Land-use patterns in the arid and semi-arid zones of China from 2000 to 2020.(a) Land-use cover in 2000; (b) Land-use cover in 2010; (c) Land-use cover in 2020. By 2040, grassland and unused land exhibit the most pronounced but opposite changes across the three scenarios (Fig. 4). Grassland area increase overall, but the growth rate diminishes with increasing development intensity. Under the SSP119, grassland expands by 11.21×10 4 km 2 , while under SSP585, it increases by only 6.94×10 4 km 2 . Unused land decreases under all three scenarios, with the most significant reduction of 11.28×10 4 km 2 occurring under SSP119. Cropland areas continue to expand, reaching 30.11×10 4 km 2 under the SSP585 —an increase of 1.61×10 4 km 2 compared to 2020. Forest remains largely stable under SSP119, while decreasing by 1.85×10 4 km 2 and 2.07×10 4 km 2 under SSP245 and SSP585, respectively. Water exhibits only minor fluctuations. Urban use shows marked divergence: slight contraction under SSP119 versus significant expansion to 4.61×10 4 km 2 under SSP585. Spatially, land use patterns in 2040 will differ markedly from those in 2020. Cropland expansion primarily concentrated in the eastern and southern agriculturally suitable zones, extending into central and western regions under SSP585. Grassland increases mainly occurred in northern and western grassland areas, with SSP119 exhibiting the most extensive spatial coverage and most pronounced expansion. Unused land reduction centered on western and northern marginal areas, showing the most complete conversion and most significant spatial contraction under SSP119. Urban expansion concentrated in densely urbanized central and western zones of the study area, with the most extensive spatial sprawl occurring under the SSP585. Fig. 4. Multi-scenarios land use pattern in arid and semi-arid areas of China in 2040.(a) Land use cover in SSP119; (b) Land use cover in SSP245; (c) Land use cover in SSP585. A comprehensive comparison of historical changes and future scenarios reveals that, whether examining historical trends or simulating outcomes under different future scenarios, the expansion of Cropland and urban primarily occurs through the conversion of unused and portions of ecological land. However, the intensity of expansion and the extent of ecological land affected vary significantly across scenarios, fundamentally reflecting how different socioeconomic development paths drive the evolution of land use patterns. Specifically, under the low-emission scenario (SSP119), ecological land protection is relatively effective, with land conversion moving toward eco-friendly directions. The medium-development pathway (SSP245) exhibits transitional characteristics. In contrast, the high-emission scenario (SSP585) shows the most intense expansion of cropland and urban, with the most significant displacement of grassland and forest, reflecting the strong impact of rapid socioeconomic development and urbanization on ecological spatial patterns. ESs change induced by land use During the period from 2000 to 2020, ESs showed distinct trends: WY and HQ remained generally stable, CS increased steadily (+2.32×10 8 t), and SC improved significantly (+4.29×10 8 t). Changes in each ecosystem service are presented in Table 4 and Fig. 5, where HQ denotes the mean value and the remaining values represent the total quantity. Overall, all ESs demonstrated a spatial distribution characterized by higher values in the east and lower values in the west, as well as higher values in the south and lower values in the north. WY decreased by 10.4×10 8 m 3 . Spatial distribution revealed higher water yields in the northeast, south, and mountainous northern Xinjiang regions, while arid central and western areas exhibited lower yields with minimal variation, indicating overall stability in the spatial pattern of water resources. HQ initially declined before increasing. High-value areas were primarily concentrated in the northeast, southern Gansu and Ningxia, and parts of Xinjiang with higher vegetation coverage, closely related to forest and grassland cover. The overall lower HQ in the drier central and western regions reflects the constraints imposed by natural aridity on ecosystem quality. CS steadily increased from 295.12×10 8 t to 297.44×10 8 t. Its spatial distribution generally aligns with habitat quality, indicating a strong correlation between the two. SC increased by 4.29×10 8 t, with the primary growth occurring between 2000 and 2010, while growth from 2010 to 2020 was relatively limited. The spatial distribution of soil conservation generally aligns with water yield, reflecting the influence of water resource conditions on soil retention capacity. Comprehensively, during this phase, the ESs in the study area exhibited a pattern of generally stable WY, slightly declining HQ, steadily increasing CS, and significantly enhanced SC. This confirms the positive role of vegetation restoration and ecological engineering in strengthening SC and increasing CS, while also providing a reference basis for subsequent ecological conservation and soil and water conservation management. Table 4. ESs from 2000 to 2040. SSP119 SSP245 SSP585 WY×10 8 /m 3 1823.7 1775.27 1813.34 1753.95 1906.21 2289.22 HQ 0.378 0.364 0.373 0.396 0.390 0.380 CS×10 8 /t 295.12 294.36 297.44 308.27 303.37 302.46 SC×10 8 /t 20.67 24.36 24.96 18.24 25.26 39.97 Fig. 5. Spatial distribution of ESs from 2000 to 2020. By 2040, SSP119 enhances HQ and CS but weakens WY and SC; SSP245 maintains balance; SSP585 strengthens WY and SC but offers limited HQ and CS gains (Table 4 and Fig. 6). WY declines markedly under the SSP119 (-59.39×10 8 m 3 ), while it increases to varying degrees under SSP245 (+92.87×10 8 m 3 ) and SSP585 (+475.88×10 8 m 3 ), with the most pronounced enhancement observed under SSP585 with high-value areas primarily concentrated in the northeastern and southern mountainous regions. HQ shows an overall improvement trend, with SSP119 exhibiting the largest increase (+0.023), followed by SSP245 (+0.017), while SSP585 shows the smallest improvement (+0.007). Spatially, improvements remain centered around forested and grassland areas. CS increase across all scenarios, with the most pronounced growth under SSP119 (+10.83×10 8 t), followed by SSP245 (+5.93×10 8 t) and SSP585 (+5.02×10 8 t). High-value areas were predominantly distributed in grassland and forest ecosystems. SC functions diverge significantly across scenarios: SSP119 showed the most severe decline (-6.72×10 8 t), SSP245 remained largely stable (+0.30×10 8 t), while SSP585 demonstrated a substantial increase (+15.01×10 8 t), with enhanced regions concentrated in mountainous and steeply sloped areas. Fig. 6. Spatial distribution of multi-scenario ESs in 2040. A comprehensive comparison of historical changes and future scenarios reveals significant divergence in the impacts of different development pathways on ESs within the study area: The low-emission SSP119 pathway is more conducive to maintaining and enhancing HQ and CS, but WY and SC functions are weakened due to reduced precipitation and increased potential evapotranspiration. Under the medium-emission SSP245 pathway, changes in various ESs are moderate, presenting a relatively balanced development trend overall. The high-emission SSP585 scenario shows marked improvements in WY and SC due to enhanced climatic conditions, yet yields limited gains for HQ and CS. This reflects the trade-offs and synergies among ESs under different development pathways, providing scientific support for regional ecological policy formulation and pathway selection. Spatiotemporal variation characteristics of CES From 2000 to 2040, the overall distribution pattern of the CES in the study area showed higher values in the east and lower values in the west (Fig. 7). Between 2000 and 2020, the overall CES level first declined and then stabilized. It decreased from 0.3786 in 2000 to 0.3707 in 2010, and then remained largely stable at 0.3711 by 2020. The SSP119 mean rose to 0.3863, showing a significant increase compared to 2020. Its CES level exceeded that of other scenarios, indicating that under a pathway of low emissions and enhanced ecological conservation, the region’s comprehensive ecosystem services improved markedly. The SSP245 average reached 0.3767, slightly higher than 2020 levels. This indicates that under a scenario balancing economic development and ecological conservation, ecosystem services show some improvement, though the overall increase remains limited. The SSP585 averaged 0.3756, the lowest CES level among the three scenarios. This reflects constrained improvements in regional ecosystem services under high emissions and intensive development, with some areas potentially facing further degradation risks. Spatially, CES values were generally low in arid and desert regions of Northwest China (e.g., Tarim Basin, Junggar Basin, Hexi Corridor desert areas) and desertified areas of western Inner Mongolia, indicating weak ecosystem service provision. High-value areas are primarily distributed in the northeastern and southern mountainous regions, with elevated CES values also observed in mountainous zones like the Tianshan and Altai Mountains in Xinjiang. These regions feature robust vegetation cover, prominent water conservation and soil retention functions, and form critical ecological barriers. Fig. 7. Spatial distribution of CES from 2000 to 2040. (a) 2000; (b) 2010; (c) 2020; (d) 2040 SSP119; (e) 2040 SSP245; (f) 2040 SSP585. Overall, CES exhibited a declining-to-stable trend from 2000 to 2020, but showed a recovery trend under all emission scenarios by 2040, with SSP119 > SSP245 > SSP585. This indicates that future emission reduction measures and ecological conservation policies are crucial for enhancing comprehensive ecosystem services. Conversely, pursuing high-emission development pathways may weaken the ecological security barrier functions of these regions. TNSFP ecological effects based on spatiotemporal evolution of CES TNSFP is the largest and longest-running ecological protection project in China. This study reveals the ecological effects of the project’s second phase (Fig. 8) through spatio-temporal evolution analysis based on CES, focusing on key construction areas. The fourth phase spanned 2000-2010, while the fifth phase covered 2010-2020. Fig. 8. Spatiotemporal changes in CES. (a) 2000-2010; (b) 2010-2020; (c) 2000-2020. Phase IV Program (2000-2010) During this period, the CES remained largely stable overall, with a marked improvement trend as its core characteristic. The stable zone covered the largest area, reaching 269.37×10 4 km 2 , with notable regional variations: On one hand, focusing on the oasis-desert transition zone, significant CES improvement occurred in the Kunlun Mountains, the periphery of the Tarim Basin, and northern Xinjiang (18.12×10 4 km 2 ). This was achieved through the synergistic effects of natural conditions and governance measures, making it a representative example of ecological improvement in the fourth phase of the project. Conversely, CES declined in western Inner Mongolia, the Tarim Basin, and Junggar Basin regions (16.13×10 4 km 2 ). Overall, the area showing CES improvement slightly exceeded degraded areas during this phase, indicating a positive trend in the ecosystem’s overall condition. Phase V Program (2010-2020) During this period, the study area exhibited characteristics of expanding stable zones within the CES. Degraded areas decreased to 14.00×10 4 km 2 , while improved areas reached 15.04×10 4 km 2 , remaining slightly larger than degraded zones. Overall ecosystem stability further strengthened. By expanding management coverage and deepening existing regional governance, western Inner Mongolia and eastern Tianshan Mountains transitioned into improved zones. Areas resistant to improvement saw enhanced CES through artificial restoration and grazing bans. Stability in the oasis-desert transition zone continued to rise under optimized water resource management and vegetation conservation. The overall second phase of the program (2000-2020) Long term changes from 2000 to 2020 reveal that the CES exhibits an overall trend of continuous improvement and gradually increasing stability. The area of long term stable zones spans 278.34×10 4 km 2 , with the total area of improved zones exceeding that of degraded zones, reflecting the overall positive impact of the project on ecosystems. Spatially, change zones are primarily concentrated in Xinjiang. Stable zones maintain a significant proportion due to synergistic effects of project maintenance and natural ecological conditions. Degraded zones, constrained by climate variations and human activities, show CES declines in some arid areas and desert fringe zones. Improved zones cover key project construction areas and extend to broader regions. Changes in land use and CES collectively demonstrate phased ecological restoration outcomes, gradually forming a “point-band-patch” ecological restoration pattern (Fig. 9). Originally fragmented, ecologically unfavorable land use types shifted toward more optimal directions. In point-like areas, local ecological land use increased, with significant and stable CES improvement. Belt-like areas expanded contiguously along specific directions, forming strip-like distributions of improved zones with coherent ecological restoration. Patch-like areas saw large-scale ecological land use expansion, with widely distributed improved zones demonstrating pronounced scale effects in patch-level ecological restoration. Fig. 9. Typical “point-band-patch” ecological restoration zones in the TNSFP. (a) Land use cover in 2000; (b) Land use cover in 2020; (c) Spatiotemporal changes in CES from 2000 to 2020. It is evident that the TNSFP has played a significant role in promoting CES improvement. Phase IV focused primarily on enhancing key areas, while Phase V further elevated overall stability. The project exhibits a long-term trend of sustained improvement and increasing stability. Under the implementation of the TNSFP and other related policies, a “point-band-patch” ecological restoration pattern has gradually taken shape. However, significant regional disparities exist: while eastern regions have achieved remarkable ecosystem recovery, ecosystems in western desert areas remain relatively fragile. The spatiotemporal evolution of CES reflects the combined effects of the phased implementation of the TNSFP, natural environmental constraints, and bidirectional regulation by human activities. This provides scientific basis for regional ecological construction and ecosystem service management. Discussion Impact of land use types on CES Land use types serve as the critical link connecting human activities and ecosystems, with their variations directly determining the supply capacity and evolutionary characteristics of CES. Based on the average CES values across various land use types in the arid and semi-arid regions of China from 2000 to 2020 and under multiple scenarios for 2040 (Fig. 10), a core pattern emerges: “natural attributes establish the foundation, human activities determine the trend, and policies define the potential” in how land use types influence CES. Fig. 10. Mean values of CES for land use types in arid and semi-arid regions of China. Inherent ecological attributes determine the foundational level of CES. The ecological function differences across land use types directly shape the inherent hierarchy of CES, which remains relatively stable across historical and future scenarios. Forest exhibit the highest CES values, with their high vegetation coverage, rich biodiversity, and robust carbon sequestration capacity establishing them as the core of regional ecological security. Cropland and grassland occupy the upper-middle contribution tier. Cropland, needing to balance food production, is significantly influenced by crop types and cultivation practices, resulting in higher CES variability. Grassland, however, exhibit more stable ecological service capacity due to their relatively consistent herbaceous vegetation cover, serving as transitional links between natural and artificial land types. Although water have lower CES values, they play irreplaceable roles in water conservation and climate regulation. Urban and unused land, constrained by limited ecological functions, consistently occupy the lowest tier of CES contribution. Unused land ecosystems, in particular, exhibit high vulnerability and provide virtually no effective ecological support to the region. Human activities drive CES differentiation through dual “disturbance-conservation” effects, with the TNSFP emerging as a pivotal force in reversing CES decline by strategically regulating land-use structure. From a land use evolution perspective, CES in cropland and water continues to decrease: cropland CES dropped by 4.6%, attributed to urbanization encroaching on high-quality cropland and soil degradation caused by agricultural intensification; whereas water CES decreased by 6.2% due to water scarcity and non-point source pollution. The implementation of TNSFP significantly influenced CES dynamics in forest and grassland areas. For forest, measures like afforestation and closed-pasture management drove CES up from 0.722 to 0.734. For grassland areas, supported by grazing bans and rotational grazing policies, the decline in CES narrowed considerably. Additionally, urban recorded a modest 10.8% increase in CES, challenging the conventional understanding that artificial land use inevitably reduces CES. Scenario simulations reveal the potential of policies to regulate land use and stabilize CES under different development pathways. CES simulation results under different scenarios highlight the regulatory role of policies in shaping the relationship between land use types and CES. As a long-term ecological policy vehicle, TNSFP can enhance the resilience of land categories to scenario responses and mitigate CES fluctuation risks. Under the SSP119, CES for urban increases by 19.4% compared to 2020 levels, but under SSP585 it declines to near 2000 levels, demonstrating its high sensitivity to development patterns. Cropland also recovered under SSP119, reflecting the positive effects of conservation tillage, while it reached its lowest level under SSP585. In contrast, forest and grassland CES exhibit minimal variation, demonstrating strong stability. Therefore, the mechanism by which land use types influence CES can be summarized as: attributes establish the foundation, human activities determine the trend, and policies define the potential. Based on this, enhancing regional ecology requires focusing on three aspects: leveraging the TNSFP as a key initiative to delineate priority implementation zones, optimize afforestation and management, and protect and restore natural land types such as forests and grassland; promoting the ecological transformation of artificial land types like cropland and urban areas; and prioritizing the convergence of land use toward low-carbon sustainable scenarios like SSP119 to achieve synergistic development in land use optimization and ecological service enhancement. Limitations and future research Although this study reveals the spatiotemporal characteristics of ecosystem services in arid and semi-arid regions from 2000 to 2040 and analyzes the ecological effects of the TNSFP in the study area, certain limitations remain. First, the study period primarily focuses on 2000-2020, which reflects the program’s second phase but provides limited insight into longer-term evolutionary trends. Second, given the study area’s vast geographical scope, evaluating only four ESs may not fully represent overall ecological conditions. Additionally, reliance on empirically derived parameter settings for certain variables could introduce biases and uncertainties in the results Zhang et al., 2022b(). Future research could expand in the following directions: incorporating a broader range of ecosystem service types to conduct a more comprehensive assessment of the region’s ecological status; introducing higher-resolution, multi-source data to enhance the accuracy and reliability of model simulations and parameter settings; and extending the temporal scale by integrating multi-temporal remote sensing and long-term ecological monitoring data to reveal the sustained effects of ecological engineering on ecosystem services. This would provide more scientific support for ecological conservation and sustainable development in arid and semi-arid regions. Conclusions We employed the PLUS model coupled with CMIP6 climate scenarios (SSP119, SSP245, SSP585) to simulate future land-use patterns and applied InVEST and related models to quantify four key ESs (WY, HQ, CS, and SC). Based on this, a CES index was constructed to assess the ecological effects of TNSFP in arid and semi-arid regions of China from 2000 to 2020. From 2000 to 2020, land use structure in the study area underwent significant adjustments, with cropland (+2.80×10 4 km 2 ) and urban (+1.19×10 4 km 2 ) continuously expanding, while water (-1.71×10 4 km 2 ) declined markedly. Grassland initially decreased and then recovered to its original extent, and forest areas remained largely unchanged. In contrast, unused land first expanded and later contracted, resulting in a net decrease of 1.71×10 4 km 2 . Concurrently, ESs exhibited differentiated trends: WY and HQ remained generally stable, CS increased steadily (+2.32×10 8 t), and SC improved markedly (+4.29×10 8 t), confirming the positive ecological effects of vegetation restoration and conservation projects. By 2040, notable scenario-based divergences emerged—under SSP119, ecological land protection was most effective with a marked reduction in unused land (−11.28×10 4 km 2 ); under SSP585, cropland and urban expansion intensified, accompanied by forest loss (−2.07×10 4 km 2 ), demonstrating the direct influence of socioeconomic development intensity on ecological spatial patterns. The CES exhibited a temporal pattern of “decline–stabilization–recovery”, decreasing from 0.3786 (2000) to 0.3711 (2020) and rebounding under all future scenarios (SSP119: 0.3863 > SSP245: 0.3767 > SSP585: 0.3756). High CES values were concentrated in the eastern and southern mountainous regions, forming key ecological barrier zones, while low CES areas were mainly located in the western desert basins and arid plains. The TNSFP significantly enhanced ecosystem functions, leading to a “point–belt–patch” restoration pattern across its core zones. Over successive project phases, ecological improvement areas expanded steadily, regional ecological stability increased, and ecosystem functions were continuously reinforced. Although the TNSFP has achieved notable ecological restoration and service enhancement, regional heterogeneity persists. Areas such as the Tianshan and Altai Mountains exhibit high CES and strong ecological resilience, whereas desert basins remain ecologically fragile. Moving forward, policies should prioritize low-emission, conservation-oriented development pathways, strengthen the protection of forests and grassland, promote ecological transitions in cropland and urban areas, and implement differentiated management strategies for arid zones. These measures are essential for consolidating the long-term achievements of the TNSFP and safeguarding the ecological security barrier in arid and semi-arid regions of China. Acknowledgments This study was supported by The Project Supported by the Open Fund of Key Laboratory of Urban Land Resources Monitoring and Simulation，Ministry of Natural Resources (KF-2023-08-01) and West Light Foundation of The Chinese Academy of Sciences (25JR6KA004). References Bagstad, K.J., Cohen, E., Ancona, Z.H., McNulty, S.G., Sun, G., 2018. The sensitivity of ecosystem service models to choices of input data and spatial resolution. Applied Geography 93, 25-36. Bai, J., Wang, X.F., Tu, Y., Zhou, J., Wang, X., Yao, W., Sun, Z., 2025. Integration of ecosystem service composite index and driving thresholds for ecological zoning management: A case study of Qinling-Daba Mountain, China. Journal of Environmental Management 384, 125309. Berg, A.X., McColl, K.A., 2021. No projected global drylands expansion under greenhouse warming. Nature Climate Change 11, 331-337. Borges, C.K., dos Santos, C.A.C., Carneiro, R.G., da Silva, L.L., de Oliveira, G., Mariano, D., Silva, M.T., da Silva, B.B., Bezerra, B.G., Perez-Marin, A.M., de S. Medeiros, S., 2020. Seasonal variation of surface radiation and energy balances over two contrasting areas of the seasonally dry tropical forest (Caatinga) in the Brazilian semi-arid. Environmental Monitoring and Assessment 192, 524. Carpenter, S., Pingali, P., Bennett, E., Zurek, M., 2005. Ecosystems and Human Well-Being: Findings of the Scenarios Working Group v. 2. Washington, DC, Island Press. Also available at Millennium Ecosystem.Chen, X., Hu, R., Jiang, F.Q., Wang, Y., Zhang, J., 2015. Physical Geography of Arid Regions in China. Beijing: Science Press. (In Chinese)Costanza, R., de Groot, R., Braat, L., Kubiszewski, I., Fioramonti, L., Sutton, P., Farber, S., Grasso, M., 2017. Twenty years of ecosystem services: How far have we come and how far do we still need to go? Ecosystem Services 28, 1-16. Feng, S.Y., Zhao, W.W., Yan, J.M., Xia, F., Pereira, P., 2024. Land degradation neutrality assessment and factors influencing it in China’s arid and semiarid regions. Science of The Total Environment 925, 171735. Gao, M.N., Xu, R.H., Huang, J.L., Su, B., Jiang, S., Shi, P., Yang, H., Xing, Y., Wang, D., Jiang, H., Kundzewicz, Z.W., Jiang, T., 2023. Increase of carbon storage in the Qinghai-Tibet Plateau: Perspective from land-use change under global warming. Journal of Cleaner Production 414, 137540. Gao, X., Hu, Y., Zhang, Z.X., Jiao, Y., Ji, W., Qian, Y., He, Y., Pei, L., Yin, Y., Hu, Y., Shi, X., 2025. Analysis of the spatiotemporal evolution of cultivated land in the Chaohu Basin based on the PLUS model and multi-scenario simulation. Sustainable Futures 10, 100945. Gaylard, S., Colella, R., Nelson, M., Lavery, P., Waycott, M., 2025. Incorporating ecosystem service assessments into development planning − impact from a dredging project in South Australia on seagrass. Ecosystem Services 74, 101738. Guo, J., Feng, P.F., Xue, H., Xue, S., Fan, L., 2025. A framework of ecological security patterns in arid and semi-arid regions considering differences socioeconomic scenarios in ecological risk: Case of Loess Plateau, China. Journal of Environmental Management 373, 123923. Hou, C.Z., Huang, D.Q., Gui, D.W., Lei, J., Lu, H., Xu, Z., 2023. Spatiotemporal variations of climate extremes and influential factors in deserts and sandy fields of northern China from 1961 to 2019. Geographical Science 43, 1495-1505. (In Chinese)Ji, P., Shao, Q.Q., Wang, M., Liu, H., Wang, X., Ling, C., Hou, R., 2022. Monitoring and Assessment of Ecological Benefits of the Shelter Forest Program in the Three-North Region during 2001—2020. Scientia Silvae Sinicae 58, 31-48. (In Chinese)Li, C., Zhang, S.Q., Cui, M.Y., Wan, J., Rao, T., Li, W., Wang, X., 2022. Improved Vegetation Ecological Quality of the Three-North Shelterbelt Project Region of China during 2000–2020 as Evidenced from Multiple Remotely Sensed Indicators, Remote Sensing.Li, J.Y., Chen, X., Kurban, A., Van de Voorde, T., De Maeyer, P., Zhang, C., 2021. Coupled SSPs-RCPs scenarios to project the future dynamic variations of water-soil-carbon-biodiversity services in Central Asia. Ecological Indicators 129, 107936. Li, M.M., Liu, A.T., Zou, C.J., Xu, W.-d., Shimizu, H., Wang, K.-y., 2012. An overview of the “Three-North” Shelterbelt project in China. Forestry Studies in China 14, 70-79. Li, X.Y., Wu, C.S., 2025. Sensitivity assessment and simulation of ecosystem services in response to land use change in arid regions: Empirical evidence from Xinjiang, China. Ecological Indicators 171, 113150. Peng, S., 2025a. 1-km monthly mean temperature dataset for china (1901-2024), in: National Tibetan Plateau Data, C. (Ed.). National Tibetan Plateau Data Center.Peng, S., 2025b. 1-km monthly potential evapotranspiration dataset for China (1901-2024), in: National Tibetan Plateau Data, C. (Ed.). National Tibetan Plateau Data Center.Peng, S., 2025c. 1 km multi-scenario monthly potential evapotranspiration dataset for China (2021-2100), in: National Tibetan Plateau Data, C. (Ed.). National Tibetan Plateau Data Center.Peng, S., Ding, Y., Liu, W., Li, Z., 2019. 1 km monthly temperature and precipitation dataset for China from 1901 to 2017. Earth Syst. Sci. Data 11, 1931-1946. Qiu, B.W., Chen, G., Tang, Z.H., Lu, D., Wang, Z., Chen, C., 2017. Assessing the Three-North Shelter Forest Program in China by a novel framework for characterizing vegetation changes. ISPRS Journal of Photogrammetry and Remote Sensing 133, 75-88. Raymond, C.M., Singh, G.G., Benessaiah, K., Bernhardt, J.R., Levine, J., Nelson, H., Turner, N.J., Norton, B., Tam, J., Chan, K.M.A., 2013. Ecosystem Services and Beyond: Using Multiple Metaphors to Understand Human–Environment Relationships. BioScience 63, 536-546. Scholes, R.J., 2016. Climate change and ecosystem services. WIREs Climate Change 7, 537-550. Seppelt, R., Dormann, C.F., Eppink, F.V., Lautenbach, S., Schmidt, S., 2011. A quantitative review of ecosystem service studies: approaches, shortcomings and the road ahead. Journal of applied Ecology 48, 630-636. Su, B.D., Huang, J.L., Mondal, S.K., Zhai, J., Wang, Y., Wen, S., Gao, M., Lv, Y., Jiang, S., Jiang, T., Li, A., 2021. Insight from CMIP6 SSP-RCP scenarios for future drought characteristics in China. Atmospheric Research 250, 105375. Sun, P.L., Qu, L., Liu, Q.G., Xu, Y., Gong, Q., 2022. Spatiotemporal differentiation characteristics and driving factors of ecological restoration in the agro-pastoral ecotone of northern China. Resources Science 44, 943-954. (In Chinese)Wang, X.Z., Peng, S.Z., Wu, J.Z., Zheng, K., Wang, S., Shangguan, Z., Deng, L., 2024. Simulation of the Key Ecosystem Services Changes in China’s Loess Plateau under Various Shared Socioeconomic Pathways Scenarios. Ecosystem Health and Sustainability 10. Watson, K.B., Galford, G.L., Sonter, L.J., Koh, I., Ricketts, T.H., 2019. Effects of human demand on conservation planning for biodiversity and ecosystem services. Conservation Biology 33, 942-952. Wu, H., Li, X.Y., Cai, Z.H., Chen, Y., 2025. Spatially heterogeneous wetting and climatic drivers of precipitation variability in arid and semi-arid Northwest China since 1960. Journal of Atmospheric and Solar-Terrestrial Physics 277, 106632. Xu, L., Zheng, C.L., Ma, Y., 2021. Variations in precipitation extremes in the arid and semi‐arid regions of China. International Journal of Climatology 41, 1542-1554. Xu, X.L., Liu, J.Y., Zhang, S.W., Li, R., Yan, C., Wu, S., 2018. China Multi-period Land Use and Land Cover Remote Sensing Monitoring Dataset (CNLUCC), in: Data Registration and Publication System of Resource and Environment Science Data Center, C.A.o.S. (Ed.).Xu, Y.F., Yang, J., Chen, Y.N., 2016. NDVI-based vegetation responses to climate change in an arid area of China. Theoretical and Applied Climatology 126, 213-222. Zhai, J.J., Wang, L., Liu, Y., Wang, C., Mao, X., 2023. Assessing the effects of China’s Three-North Shelter Forest Program over 40 years. Science of The Total Environment 857, 159354. Zhang, B., Fang, H.Y., Wu, S.F., Li, C., Wang, Y., Siddique, K.H.M., 2024a. Soil erosion prediction and spatiotemporal heterogeneity in driving effects of precipitation and vegetation on the northern slope of Tianshan Mountain. Journal of Cleaner Production 459, 142561. Zhang, K.L., Fang, B., Zhang, Z.C., Liu, T., Liu, K., 2024b. Exploring future ecosystem service changes and key contributing factors from a “past-future-action” perspective: A case study of the Yellow River Basin. Science of The Total Environment 926, 171630. Zhang, S.Q., Yang, P., Xia, J., Wang, W., Cai, W., Chen, N., Hu, S., Luo, X., Li, J., Zhan, C., 2022a. Land use/land cover prediction and analysis of the middle reaches of the Yangtze River under different scenarios. Science of The Total Environment 833, 155238. Zhang, X.H., Zhang, B.P., Yao, Y.H., Wang, J., Yu, F., Liu, J., Li, J., 2022b. Dynamics and climatic drivers of evergreen vegetation in the Qinling-Daba Mountains of China. Ecological Indicators 136, 108625. Zhang, Y., Chen, X.Y., Zhang, Y., Wang, B., 2025. Quantitative contribution of climate change and vegetation restoration to ecosystem services in the Inner Mongolia under ecological restoration projects. Ecological Indicators 171, 113240. Zhao, Q.J., Shao, J.F., 2023. Evaluating the impact of simulated land use changes under multiple scenarios on ecosystem services in Ji’an, China. Ecological Indicators 156, 111040. Zhu, Y., Ma, B., Hu, H.B., Ding, D., Zhou, H., Liu, J., Liu, J., Lin, Z., 2025. Analysis and prediction of spatiotemporal carbon storage changes in the Taihu Lake Basin in Jiangsu Province based on PLUS and InVEST model. Trees, Forests and People 21, 100916. Information & Authors Information Version history V1 Version 1 11 November 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords arid and semi-arid region comprehensive ecosystem service land use change multi-scenario simulation three-north shelter forest program Authors Affiliations Chao Lin 0009-0007-5244-1405 Lanzhou Jiaotong University View all articles by this author Qiang Bie [email protected] Lanzhou Jiaotong University View all articles by this author Liang Huajun Lanzhou Jiaotong University View all articles by this author Xiangxiang Huang 0009-0009-2484-7701 Lanzhou Jiaotong University View all articles by this author Liu Hang Lanzhou Jiaotong University View all articles by this author Metrics & Citations Metrics Article Usage 336 views 128 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Chao Lin, Qiang Bie, Liang Huajun, et al. Assessing the ecological effects of the Three-North Shelter Forest Program by a novel ecosystem service index. Authorea . 11 November 2025. DOI: https://doi.org/10.22541/au.176284287.78816269/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.176284287.78816269/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'a0250910dacc3fe2',t:'MTc3OTg4NTA3NA=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();