Urbanity in the countryside: interaction of livelihood, lifestyle, connectivity and rural greening

Urbanity in the countryside: interaction of livelihood, lifestyle, connectivity and rural greening | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Urbanity in the countryside: interaction of livelihood, lifestyle, connectivity and rural greening Zhaxi Dawa, Weiqi Zhou, Steward Pickett, Daniel Childers, Wenjuan Yu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7369333/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Rural ecosystem recovery is often attributed to population decline, but the social-economic pathways linking urbanization to ecological change remain underexplored. Grounded in the continuum of urbanity theory, we developed two alternative pathway models (exogenous-driven vs. endogenous-driven) to examine how urbanization influenced rural ecological quality in China from 2010 to 2020. Results indicate that ecological greening was most prominent in areas where urbanity increased while population declined, supported by rising forest cover, stable cropland use, and improved grain yield. Rural household livelihood and lifestyle changes had negligible direct impact on vegetation trends. However, the pathway models revealed significant indirect roles: urbanization effects were mediated by rural households through shifts from farming to non-farming livelihoods, improved quality of life, and rural-urban migration. The exogenous-driven model demonstrated stronger explanatory power and clearer pathway significance than the endogenous driven model. These findings move beyond the conventional narrative that rural depopulation alone leads to greening, offering a new social-ecological perspective on how urbanization can actively contribute to rural sustainability. Earth and environmental sciences/Ecology/Urban ecology Scientific community and society/Geography Continuum of urbanity social-ecological effects Rural transition Causality analysis Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 1. Introduction China has experienced the world’s largest rural to urban migration over the past three decades, with more than 290 million people relocating to cities between 1990 and 2019 1,2 . This demographic transition has contributed to widespread rural depopulation, agricultural land abandonment, and, in many cases, ecological recovery 3-5 . Many studies have attributed improvements in rural vegetation and ecosystem services to reduced human pressure, declining cultivation, or afforestation initiatives 6-10 . However, the dominant narrative—linking depopulation directly to ecological recovery—offers only a partial explanation 2,11-14 . It overlooks the deeper social and economic restructuring associated with urbanization that may indirectly shape ecological outcomes 15,16 . Urbanization alters more than just land use or population numbers 2,17-21 . It also transforms rural livelihoods, lifestyles, and connectivity, which may in turn influence ecological conditions beyond urban areas 22-26 . These changes include shifts in employment, material consumption, and land management, yet such indirect social-ecological pathways remain understudied 27-29 . To address this gap, we adopt the concept of the continuum of urbanity 30-32 ( Fig. 1 ). It posits that urbanization may act as a quality rather than only an entity 33 , where even remote villages may exhibit non-farm livelihoods, urban-influenced lifestyles, and integration into broader flows of people and resources 34 . Mixtures of urban and rural qualities can appear widely in human settlements 32,35 . We quantify this mixture using an urbanity index to capture the intensity of urban traits in rural areas 36 , enabling us to explore how urbanization influences rural social-ecological systems beyond its direct physical impacts 37-39 . We hypothesize that urbanity drives rural ecological change through two distinct yet interrelated pathways—exogenous and endogenous—each reflecting a different scenario of social-ecological transformation ( Fig. 2 ). In the exogenous pathway, national urbanization policies and the concentration of urban features (e.g., infrastructure and job opportunities) in urban areas encourage rural-urban migration 34,40 . This rural out-migration reduces local labor capacity, prompting households to shift from farming to relying on off-farm income or remittance 41,42 , while consumption patterns also change toward less land-dependent lifestyles (e.g., purchasing food and fuel), promoting quality of life 43 . These adjustments in livelihoods and lifestyles lead to altered land management, fields are downsized, marginal land is abandoned, and vegetation recovers 44 . In contrast, the endogenous pathway originates within rural areas, where increased urban features—such as improved roads, digital connection, and social services—gradually shift livelihoods toward wage labor and services, and encourage more urban-like consumption 34,45,46 . These changes enhance household mobility and financial capacity, enabling migration as a later step. Circulating migrants bring back capital and knowledge that shape land-use decisions, such as afforestation or cropland consolidation 21,47-49 , that improve rural ecological quality. In this case, qualities of urbanity act as an endogenous forces embedded in rural areas rather than as a product of depopulation. Our study tests whether different sequences of demographic, livelihood, and lifestyle changes—triggered by an urbanity index calculated for varying places—lead to distinct ecological outcomes within the continuum of urbanity. To evaluate these hypotheses, we integrated satellite-derived vegetation indices, including Enhanced Vegetation Index (EVI) trends and the amount of greening area, with an infrastructure-based urbanity index across China from 2010 to 2020. We examined their spatial distribution and relationship with population dynamics across the urban–rural typology. In combination with household survey data, we assessed how non-farm livelihoods (classified by income structure) 50 and quality of life (measured by the inverse of the Engel coefficient) 51 evolved with increasing urbanity. Moreover, we applied linear mixed-effects models (LMMs) and structural equation modeling (SEM) to quantify the direct effects of urbanity index on ecological quality and the indirect effects mediated through both exogenous and endogenous pathways. This study bridges macro environmental trends with micro socio-economic transformations, moving beyond the simplistic “depopulation leads to greening” narrative and highlighting urbanity beyond physical cities as a positive driver of sustainable rural social-ecological transitions. 2. Results 2.1. Interaction of urbanity, population dynamics, household shifts, and ecological outcomes From 2010 to 2020, rural areas in China exhibited a synergistic transformation marked by increased urbanity index, depopulation, and ecological greening across the urban-rural typology ( Fig. 3a ). The average urbanity index rose from 1.51 to 1.88, indicating the widespread diffusion of urban infrastructure and services into rural settlements, including villages and rural centers ( Fig. 3a and Appendix Fig. 1 ). Meanwhile, rural population density declined—villages alone lost over 25 million residents—coinciding with notable vegetation recovery, as evidenced by positive EVI trends and expanded greening areas ( Fig. 3b ). These areas also showed increases in tree cover and food production, highlighting land use transitions amid demographic and infrastructural shifts ( Appendix Fig. 2 ). To further explore these patterns, we integrated rural household-level income and consumption data ( Fig. 4a ). A significant trend in the reduction of farming (here labelled “de-farming”) occurred during this period: the share of non-agricultural-dominant households increased from 33.2% to 68.1%. Over half of the households transitioned into de-agrarianized (DA) types, typically moving from farming-dependent or mixed-income profiles to predominantly non-farming livelihoods ( Fig. 4b – c ). Alongside this shift, household quality of life improved, as reflected by changes in the inverse of Engel coefficient (IEC which varies between 0 and 1)—a widely used proxy for quality of life. Between 2010 and 2020, the average IEC increased by 0.06, indicating widespread reduction in the share of food expenditure in total household consumption. This shift suggests improved material conditions, greater access to market-based goods, and more diversified consumption patterns ( Fig. 4d ). Quality of life gains were most pronounced in households with stable non-farm livelihoods (NFS, +0.13), followed by DA households (+0.08), while farming-dependent households (FS) experienced little to no improvement (–0.01). The moderate gains among returning-agriculture (RA) households (+0.03) may reflect already elevated baseline conditions due to earlier non-farm exposure, whereas the steeper rise in DA households suggests rapid structural adjustment and catch-up in consumption-based well-being. To preliminarily assess the relationships among urbanity index, livelihoods, quality of life, land management, and ecological outcomes, we conducted partial correlation analyses while controlling for environmental covariates. Results indicated that the rural urbanity index (RUI) was positively associated with both the EVI trend (r = 0.07, p < 0.001) and greening area (r = 0.04, p < 0.01). The within-urban urbanity index (UUI) also showed a statistically significant correlation with the EVI trend (r = 0.07, p < 0.001), but not with greening area (r = –0.01, p = 0.51) ( Fig. 4e and Appendix Table 1 ). De-farming was positively associated with tree cover (r = 0.05, p < 0.001) and quality of life (r = 0.13, p < 0.001), and negatively associated with greening area (r = –0.04, p < 0.05), while its relationship with the EVI trend was negligible (r = 0.01, p = 0.60). Quality of life was marginally associated with the EVI trend (r = 0.03, p < 0.05), but not with other ecological indicators. In contrast, land management variables, particularly tree cover (0.16, p < 0.001) and food production (0.14, p < 0.001), showed the strongest positive associations with vegetation improvement, highlighting their central role in ecological recovery. The linear mixed-effects models confirmed these patterns ( Fig. 5, Appendix Fig. 3 and Appendix Table 2 ). For EVI trend, positive effects were observed for UUI (estimate = 0.004, p < 0.001), tree cover (0.0002, p < 0.001), and food production (0.00002, p < 0.001), while RUI (–0.006, p < 0.001), RuralPop (–0.0000045, p < 0.001), and cropland ratio (–0.00008, p 0.99). In the greening area model, topographic slope (–0.03, p < 0.001), temperature (–0.09, p < 0.001), and CO₂ emissions (–2.82, p < 0.001) were dominant negative predictors, while tree cover (0.005, p < 0.001) and food production (0.001, p < 0.001) showed significant positive effects. UUI had a strong negative influence on greening (–2.17, p < 0.001), whereas RUI remained non-significant (p = 0.70). These findings suggest that ecological change is more closely linked to land conditions than to household-level socioeconomic shifts. Notably, areas with both population decline and vegetation recovery tended to occur in hilly terrain, with increased tree cover and food production, indicating that depopulation may support greening where land conditions are favorable ( Appendix Fig. 4, Appendix Fig. 5, Appendix Table 3, Kruskal–Wallis test with Dunn’s post-hoc test ). 2.2. Path analysis of the exogenous-endogenous hypothesis To test the hypothesized exogenous–endogenous social-ecological mechanisms, we applied Partial Least Squares Structural Equation Modeling (PLS-SEM) to disentangle how urbanity from different sources influenced rural ecological outcomes through household and land management transitions. Following the conceptual framework, we built two distinct models. The exogenous model captures the cascading effects of UUI, while the endogenous model reflects effects of RUI. In both models, agricultural activity was defined as a formative latent variable constructed from cropland ratio and food production, whereas ecological quality was modeled reflectively using EVI trend and greening area. In both models, urbanity index influenced rural population growth, although by different mechanisms: UUI reflected return migration driven by urban saturation or policy shifts, while RUI indicated locally enabled return under improved rural conditions and diversified livelihoods. These models allowed us to trace multiple direct and indirect pathways linking urbanity, household livelihoods and quality of life, migration, and land cover to ecological change. The UUI and RUI influenced ecological quality through different pathways ( Fig. 6 ). In the exogenous model ( Fig. 6a ), higher UUI led to increased rural population (0.21, p < 0.001), which in turn promoted livelihood transition, measured as de-farming (0.15, p < 0.001). De-farming, a shift away from farming-based livelihoods to non-farming, significantly improved quality of life (0.15, p < 0.001) but had a negative direct effect on ecological quality (–0.03, p = 0.031). In contrast, the endogenous model showed that RUI directly promoted reduction in farming (here labelled “de-farming”; 0.11, p < 0.001), which further enhanced quality of life (0.15, p < 0.001) and led to rural population growth (0.15, p < 0.001) ( Fig. 6b) . However, such socio-economic shifts did not directly improve ecological conditions. Instead, ecological quality was mainly driven by environmental factors (slope, temperature, and CO₂) and land management variables, particularly tree cover (exogenous: 0.21; endogenous: 0.23; p < 0.001 for both). The indirect pathways exhibited notable differences between the exogenous and endogenous models. In the exogenous model, UUI triggered a chain of social-ecological responses that ultimately impaired ecological quality of rural areas. One significant negative pathway was UUI → RuralPop → De-farming → Ecological quality (–0.001, p < 0.05), suggesting that urban influence increased rural population, which then drove de-farming, and in turn, reduced vegetation greening ( Fig. 6a and Appendix Table 4, Specific indirect effects of the exogenous model ). Another significant path extended further to include land management. UUI → RuralPop → De-farming → Agriculture activity → Ecological quality (–0.0002, p < 0.01). These results imply that although de-farming reflect livelihood transition and reduce agricultural activity, it did not necessarily improve ecological quality in the exogenous context. By contrast, the endogenous model exhibited more coherent and positive links among livelihoods, land management, and ecological quality of rural areas. A key indirect pathway was De-farming → RuralPop → Agriculture activity → Tree cover → Ecological quality (0.0009, p < 0.01), illustrating that de-farming, when accompanied by population return, contributed to reduced agricultural pressure and enhanced forest regrowth ( Fig. 6b and Appendix Table 5, Specific indirect effects of the endogenous model ). Another notable path was RUI → De-farming → RuralPop → Agriculture activity → Tree cover → Ecological quality (0.0004, p < 0.01), indicating that rural urbanization promoted livelihood transformation and land management, ultimately facilitating ecological recovery. Additionally, De-farming → RuralPop → Agriculture activity → Ecological quality (0.0038, p < 0.01) further reinforces the role of livelihood shifts in mediating ecological outcomes through agriculture activity contraction. These findings demonstrate that indirect ecological benefits emerge when socio-economic transitions in rural areas align with land recovery trajectories. Finally, the endogenous-driven model yielded 17 significant paths out of 19, while the exogenous-driven model identified only 9 out of 21. This suggests that local rural urbanity shaped ecological quality through more diverse and consistent social-ecological linkages. In contrast, distant urbanity influenced ecological outcomes primarily through fewer pathways centered on population shifts. These differences point to stronger structural coherence in the endogenous model. 3. Discussion Improvements in ecological quality across rural areas—particularly vegetation greening—has often been attributed to climatic fluctuations, reduced human pressure, or large-scale afforestation programs 10,52,53 .Yet our findings show that greening between 2010 and 2020 was not merely due to depopulation, but also to urbanity-driven shifts in livelihoods, lifestyles, and land management. Regions with higher urbanity index exhibited a stronger tendency toward livelihood de-farming among rural households, improved quality of life, reduced agricultural activity, and expanded tree cover ( Fig. 3b, Fig. 5e, Fig. 6, Appendix Fig. 2 ). The exogenous–endogenous hypothesis revealed two contrasting mechanisms. Under the exogenous pathway, the UUI led to rural population return and de-farming, but yielded limited ecological benefits. Conversely, the endogenous pathway, shaped by the RUI, actively transformed livelihoods, enhanced quality of life, and promoted vegetation recovery through reduced land pressure. These findings challenge the depopulation-centric view and underscore the importance of rural urbanity in shaping its own social–ecological transitions 2,3,54 . The exogenous–endogenous hypothesis highlights how interactions among urbanity, livelihood transitions, and migration collectively shape rural ecological outcomes. In the exogenous scenario ( Fig. 6a ), migration to cities is commonly driven by the pursuit of higher income and improved living conditions 55 . However, our results show that UUI did not promote vegetation recovery ( Appendix Table 4 ). Instead, it stimulated population return to rural areas, where returning households increasingly engaged in non-farm activities and became less reliant on agricultural income. Paradoxically, this shift did not lead to ecological improvement. In fact, areas influenced by UUI exhibited weaker greening effects, suggesting that even if returnees disengage from farming, their resettlement may place renewed pressure on rural land systems. Although UUI reflects distant urban features, its high level may signal enhanced connectivity and infrastructure, enabling easier return migration to rural areas 36 . This suggests that rural population growth in the exogenous model could be driven not only by urban growth saturation 56 and pandemic lockdown 57 , but also by improved transport links and policy incentives facilitating urban–rural connectivity 58,59 . This finding aligns with previous research suggesting that shorter migration distances or more frequent rural returns are often associated with limited or unstable natural vegetation recovery 2 . In contrast, the RUI-driven return appears to reflect endogenous attraction, where better infrastructure, services, and in-situ urbanization enabled return migration under more favorable socio-ecological contexts 60,61 ( Fig.6b and Appendix Table 5 ). Migrants returning to more urbanized rural areas often bring with them capital, skills, and exposure to urban norms, enabling investments in sustainable land practices such as agroforestry, ecological enterprises, or non-land-intensive services 62,63 . These transitions reduce agricultural intensity while supporting rural household income, contributing to long-term ecological resilience. In areas where such livelihood diversification accompanies repopulation, land degradation is less likely to recur 64,65 . For instance, green jobs and service-based economies enable rural populations to grow sustainably without intensifying land pressure 66 . This contrast underscores that migration alone is not sufficient to drive ecological change; rather, its effects depend on whether the urbanity level of the rural area, such as infrastructure, service access, and economic opportunity, supports sustainable transitions 67-71 . In other words, population return enhances ecological outcomes only when embedded within local urbanity-led transformations that enable sustainable livelihoods 72 . The pathways from social to ecological improvement hinge critically on land management. Our analysis shows that increased tree cover and food production were the most consistent predictors of vegetation greening ( Fig. 4e, Fig. 5 ), reflecting more efficient land use rather than wholesale abandonment 73,74 . These benefits were most reliable under endogenous conditions, where institutional support, diversified livelihoods, and reduced land pressure reinforced greening outcomes. In contrast, exogenous-driven de-farming was less ecologically effective and more vulnerable to reversal when economic incentives changed 9,75 . While we found no direct impact of quality of life on vegetation, its indirect role through altered consumption and reduced resource extraction may influence long-term sustainability 76 . Therefore, policies should prioritize securing post-agricultural landscapes through regulatory or financial mechanisms 77-79 , and integrate social development metrics such as the Human development index (HDI) and Nature Relationship Index (NRI) into ecological assessments 80 . The study has several limitations. We relied on two decadal snapshots of survey and satellite data, so we may miss finer-year dynamics. Our household data are at the county level, which could obscure village-scale heterogeneity and local feedbacks. Moreover, although the SEM revealed plausible causal chains, definitive causality and the role of time lags or feedback loops would require more sophisticated panel or experimental designs. Future research should use higher-resolution and higher-frequency data (e.g. annual panel surveys, village-level studies, fine-scale remote sensing) and causal inference methods to validate these pathways. Cross-regional or cross-national comparisons would also test whether the urbanity–greening dynamics observed in China hold in different social and policy contexts. 4. Data and methods 4.1. Ecological quality and its drivers We used multiple geospatial datasets to represent ecological quality (EVI) dynamics and its potential drivers. All spatial layers were resampled to a common 1km grid, using mean values or fractional coverage as appropriate ( Table 1 ). Table 1 Data list. Variable Source/Product Spatio-temporal information Pre-processing or description Ecological quality Enhanced vegetation index (EVI) from the MOD13A2 & MYD13A2 https://modis.gsfc.nasa.gov/data/dataprod/mod13.php 2010–2020 (1,000 m) Mean growing-season EVI; QA ≤1; masked for EVI > 0.05 Urbanity index From urbanity mapping method 36 . https://doi.org/10.1016/j.geosus.2024.03.004 2010, 2020 (1,000 m) Quantifies urban influence beyond physical urban areas ( Appendix 2.1 ) Population density GHSpop + census calibration http://doi.org/10.2905/0C6B9751-A71F-4062-830B-43C9F432370F 2010, 2020 (1,000 m) Calibrated with national census data at county level ( Appendix 2.2 ) Rural household income China Family Panel Studies (CFPS) https://www.isss.pku.edu.cn/cfps/ 2010, 2020 (districts or counties) Amount of farm and non-farm income (¥) ( Appendix 3.2.1 ) Rural household consumption Expenditure on food and total daily spend (¥) ( Appendix 3.2.2 ) Tree cover fraction MOD44B v061 https://doi.org/10.5067/MODIS/MOD44B.061 2010, 2020 (250 m) Resampled to 1 km Cropland fraction CACD dataset https://doi.org/10.5281/zenodo.7936885 2010, 2020 (30 m) Calculate the ratio in a 1km grid Food production China County Statistical Yearbook+ grain yield equations 2010, 2020 (1,000 m) Spatialized food production layers ( Appendix 2.3 ) Temperature & precipitation https://doi.org/10.11888/Meteoro.tpdc.270961 https://zenodo.org/records/3114194 2010, 2020 (1,000 m) Annual mean temperature and mean precipitation CO₂ emissions CAMS global emission inventories https://doi.org/10.24381/1d158bec 2010, 2020 (10,000 m) Resampled to 1 km (unit: ton·km⁻²·yr⁻¹) Slope GMTED2010 (elevation-derived) https://doi.org/10.3133/ofr20111073 2010 (232 m) Resampled to 1 km 4.2. Methodological processes This study systematically assessed the impacts of urbanity index on rural ecological quality dynamics through a four-tier analytical framework: (1) urban–rural typology analysis, (2) comparison of social-ecological characteristics across population–EVI interaction zones, (3) household survey and quantitative modeling, and (4) building hypothetical pathways and analyzing indirect effects ( Fig. 7 ). a) Urban–rural typology analysis In this study, the urban-rural typology was defined based on basic administrative unit points to reflect the impact of urbanization on vegetation greenness under the socio-economic context of China, while ensuring spatial consistency with rural household survey data. We constructed a six-level urban–rural typology based on Points of Interest (POIs) annotated with official urban–rural statistical codes. These included: urban core (111), urban fringe (112), town core (121), town fringe (122), rural center (210), and village (220). A spatial focal analysis with a 5 km moving window 81 was applied to assign each 1000 m grid cell a dominant settlement type based on the mode of POIs within the window ( Appendix Fig. 10 ). This yielded a gridded urban–rural typology surface at 1000 m resolution. For each typology zone, changes in urbanity, population density, and vegetation indicators from 2010 to 2020 were extracted. Vegetation change was captured by two indices: the EVI trend, derived using the Sen + Slope method, and the greening area, defined as the spatial extent of significantly increasing EVI (p < 0.05) identified through Mann–Kendall (MK) tests. b) Comparison of social-ecological characteristics across rural population–EVI interaction zones Based on the urban–rural typology, we identified four types of interaction zones reflecting the joint dynamics of rural population density and EVI change: (1) Rural population decrease and EVI greening (RPopDe-G), (2) Rural population increase and EVI greening (RPopIn-G), (3) Rural population decrease and EVI browning (RPopDe-B), and (4) Rural population increase and EVI browning (RPopIn-B).We used Kruskal–Wallis tests and Dunn’s post hoc comparisons to analyze differences in key social-ecological factors across rural areas (rural center and village), including urbanity, tree cover, cropland ratio, food production, slope, annual mean temperature, annual mean precipitation, and annual mean CO₂ emissions. This allowed us to determine the dominant drivers of EVI change under varying demographic contexts ( Appendix Fig. 4 ). c) Household survey and quantitative modeling To examine how urbanity index jointly affects ecological quality through household decisions and population shifts, we employed rural household income data from the China Family Panel Studies (CFPS) for the years 2010 and 2020 (N = 9,324). Using K-means clustering, households were classified into four livelihood types: agricultural-dominant (Agri. Dom.), agricultural with secondary non-agricultural income (Agri.+Non-Agri), non-agricultural with secondary farming income (Non-Agri.+Agri), non-agricultural-dominant (Non-Agri. Dom.) ( Appendix 3.2.1 ). Transitions among these four types over the 2010–2020 period resulted in 16 livelihood change combinations. These were further categorized into four representative livelihood transition types: Farming Stable (FS), Re-agrarianized (RA), De-agrarianized (DA), and Non-Farming Stable (NFS). The sequence of FS → RA → DA → NFS was used to represent an increasing degree of de-farming, indicating a progressive shift from agriculture-dependent to non-agriculture-dominant livelihoods. The Engel coefficient is calculated as the proportion of a household’s total expenditure devoted to food consumption ( Appendix 3.2.2 ). We used its inverse—referred to as the inverse of Engel coefficient (IEC), as a proxy for household quality of life, where higher IEC values indicate lower food expenditure shares and thus more urbanized and diversified lifestyles. Controlling for natural factors such as slope, temperature, precipitation, and CO₂ emissions, we first performed Spearman partial correlation analysis to assess relationships between EVI indicators (EVI trend and greening area) and potential drivers including urbanity, rural population density, livelihood type, quality of life, and land management variables (tree cover, cropland ratio, and food production) ( Appendix Table 1 ). We then applied linear mixed-effects models (LMMs) to estimate the marginal effects of each explanatory variable on ecological quality, capturing their net direct impacts ( see the 4.3 ). d) Build hypothetical pathways and analyze indirect effects Based on the “exogenous–endogenous” hypothesis, we constructed two Partial Least Squares Structural Equation Models (PLS-SEM) to assess alternative pathways through which urbanity index affects rural ecological quality ( see the 4.4 ). The exogenous-driven model assumes that urban urbanity index (UUI) directly attracts rural populations, which in turn reshape livelihood strategies and household behaviors, thereby influencing vegetation dynamics. In contrast, the endogenous-driven model posits that the rural urbanity index (RUI) first transforms household livelihoods and lifestyles, subsequently inducing rural population outmigration and adjustments in land management. 4.3. Linear Mixed Models (LMMs) To evaluate the direct effects of UUI and RUI, rural household characteristics, and land management variables on ecological quality, we applied linear mixed-effects models (LMMs) using two outcome variables: EVI trend and greening area. These two models shared the same set of predictors and allowed random intercepts by county to account for unobserved heterogeneity across spatial units. The LMMs were specified as follows: lmer(EVI trend ∼ UUI + RUI + RuralPop + De-farming + Quality of life + Tree cover + Cropland ratio + Food production + Slope + Temperature + Precipitation + CO 2 + (1 | county)) lmer(Greening area ∼ UUI + RUI + RuralPop + De-farming + Quality of life + Tree cover + Cropland ratio + Food production + Slope + Temperature + Precipitation + CO 2 + (1 | county)) Both models included the following fixed effects: UUI, RUI, RuralPop, de-farming, quality of life (IEC), tree cover, cropland ratio, food production, slope, temperature, precipitation, and CO₂ emissions. County was included as a random intercept to control for contextual differences. We extracted marginal effects (fixed effects only) to quantify the net direct influence of each predictor. Statistical significance of the fixed effects was assessed via Type III ANOVA using Satterthwaite's approximation for degrees of freedom ( Appendix Table 2 ). Model goodness-of-fit was evaluated using Marginal R 2 m (proportion of variance explained by fixed effects) and Conditional R 2 c (variance explained by both fixed and random effects) ( Appendix Fig. 3 ). All models were implemented using the lme4 and lmerTest packages in R. 4.4. Partial Least Squares Structural Equation Modeling (PLS-SEM) To investigate the complex causal pathways through which UUI and RUI influences ecological quality, we employed Partial Least Squares Structural Equation Modeling (PLS-SEM). PLS-SEM is a variance-based approach that is particularly suitable for exploratory modeling with multiple mediating variables and relatively small sample sizes or non-normal data distributions 82 . In this study, we constructed two conceptual models reflecting alternative causal logics: (1) The exogenous-driven model, where UUI first triggers rural population change, which then drives livelihood/lifestyle transformation, land use, and ecological quality. (2) The endogenous-driven model, where RUI influences household livelihoods and quality of life, which then affect rural population change and land management, ultimately impacting ecological quality. Both models were composed of formative and reflective constructs, the ecological quality was modeled as a reflective latent variable composed of two indicators: EVI trend and greening area. The agricultural activity was specified as a formative construct using cropland ratio and food production. All other variables (e.g., UUI, RUI, rural population, de-farming, quality of life, tree cover, slope, climate, CO₂) were included as single-item observed variables. Path coefficients were standardized to allow for effect size comparisons. The significance of paths was assessed via bootstrapping with 5,000 resamples, and 95% confidence intervals were used to evaluate statistical robustness (p < 0.05). We reported both direct and total indirect effects, and decomposed multi-step mediating chains where relevant. Although PLS-SEM does not rely on global fit indices like covariance-based SEM, we evaluated model quality using the following indicators ( Appendix Table 4 and Appendix Table 5 ). R² for endogenous constructs to assess explanatory power (i.e., ecological quality). Q² (Stone–Geisser’s) for predictive relevance using blindfolding procedure. Composite reliability and outer loading values for reflective indicators (threshold > 0.7). Variance Inflation Factor (VIF) for formative indicators to assess multicollinearity (threshold < 5) 83 . All modeling was performed using SmartPLS 4.0, and results were visualized using the built-in path diagram tools. Declarations Acknowledgements We acknowledge support from the National Natural Science Foundation (Grant No. U21A2010), the Postdoctoral Fellowship Program (Grade B) of China Postdoctoral Science Foundation (Grant No. GZB20250080), the National Natural Science Fund for Distinguished Young Scholars (Grant No. 42225104), the 2024 Beijing Municipal Government’s Financial Support Program for Foreign High-Level Talents (Grant No. J2024014), and the U.S. National Science Foundation (No. DEB-2224662) to the Central Arizona-Phoenix LTER Program. References Goh, C. et al. Urban China : toward efficient, inclusive, and sustainable urbanization. (World Bank, Washington, DC, 2014). Zhang, X. et al. A large but transient carbon sink from urbanization and rural depopulation in China. Nature Sustainability 5 , 321-328, doi:10.1038/s41893-021-00843-y (2022). Hou, D., Meng, F. & Prishchepov, A. V. How is urbanization shaping agricultural land-use? Unraveling the nexus between farmland abandonment and urbanization in China. Landsc. Urban Plann. 214 , 104170, doi:10.1016/j.landurbplan.2021.104170 (2021). Wang, S. et al. Urbanization can benefit agricultural production with large-scale farming in China. Nature Food 2 , 183-191, doi:10.1038/s43016-021-00228-6 (2021). Kolecka, N. Greening trends and their relationship with agricultural land abandonment across Poland. Remote Sensing of Environment 257 , 112340, doi:10.1016/j.rse.2021.112340 (2021). Shen, H. et al. Urbanization-induced population migration has reduced ambient PM2.5 concentrations in China. Science Advances 3 , e1700300, doi:10.1126/sciadv.1700300 (2017). Gebre, T. & Gebremedhin, B. The mutual benefits of promoting rural-urban interdependence through linked ecosystem services. Global Ecology and Conservation 20 , e00707, doi:https://doi.org/10.1016/j.gecco.2019.e00707 (2019). Yang, S. et al. Assessing the impacts of rural depopulation and urbanization on vegetation cover: Based on land use and nighttime light data in China, 2000–2020. Ecological Indicators 159 , 111639, doi:https://doi.org/10.1016/j.ecolind.2024.111639 (2024). Hu, T. et al. Social-ecological heterogeneity drove contrasting tree cover restoration in South China Karst. Communications Earth & Environment 5 , 484, doi:10.1038/s43247-024-01641-y (2024). Chen, C. et al. China and India lead in greening of the world through land-use management. Nature Sustainability 2 , 122-129, doi:10.1038/s41893-019-0220-7 (2019). Daskalova, G. N. & Kamp, J. Abandoning land transforms biodiversity. Science 380 , 581-583, doi:10.1126/science.adf1099 (2023). Uchida, K., Matanle, P., Li, Y., Fujita, T. & Hiraiwa, M. K. Biodiversity change under human depopulation in Japan. Nature Sustainability , doi:10.1038/s41893-025-01578-w (2025). Brandt, M. et al. Human population growth offsets climate-driven increase in woody vegetation in sub-Saharan Africa. Nature Ecology & Evolution 1 , 0081, doi:10.1038/s41559-017-0081 (2017). Garba, S. I., Ebmeier, S. K., Bastin, J.-F., Mollicone, D. & Holden, J. Wetland fragmentation associated with large populations across Africa. Nature Communications 16 , 5065, doi:10.1038/s41467-025-59373-2 (2025). Batáry, P. et al. Biodiversity and human well-being trade-offs and synergies in villages. Nature Sustainability , doi:10.1038/s41893-025-01592-y (2025). Hu, G. et al. Mitigating the ripple effects of urbanization on farmland productivity and ecological security through inclusive urbanization strategies. npj Urban Sustainability 5 , 12, doi:10.1038/s42949-025-00199-8 (2025). Marques, A. et al. Increasing impacts of land use on biodiversity and carbon sequestration driven by population and economic growth. Nature Ecology & Evolution 3 , 628-637, doi:10.1038/s41559-019-0824-3 (2019). van Vliet, J. Direct and indirect loss of natural area from urban expansion. Nature Sustainability 2 , 755-763, doi:10.1038/s41893-019-0340-0 (2019). Mbaiwa, J. E. Changes on traditional livelihood activities and lifestyles caused by tourism development in the Okavango Delta, Botswana. Tourism Management 32 , 1050-1060, doi:10.1016/j.tourman.2010.09.002 (2011). Meyerson, F. A. B., Merino, L. & Durand, J. Migration and environment in the context of globalization. Frontiers in Ecology and the Environment 5 , 182-190, doi:Doi 10.1890/1540-9295(2007)5[182:Maeitc]2.0.Co;2 (2007). Salerno, J. et al. Rural migration under climate and land systems change. Nature Sustainability 7 , 1092-1101, doi:10.1038/s41893-024-01396-6 (2024). Liu, G., Wang, H., Cheng, Y., Zheng, B. & Lu, Z. The impact of rural out-migration on arable land use intensity: Evidence from mountain areas in Guangdong, China. Land Use Policy 59 , 569-579, doi:10.1016/j.landusepol.2016.10.005 (2016). Ye, X. & Chuai, X. Have rural settlement changes harmed ecosystem carbon in China? Applied Geography 153 , 102917, doi:10.1016/j.apgeog.2023.102917 (2023). Li, T. et al. Gauging policy-driven large-scale vegetation restoration programmes under a changing environment: Their effectiveness and socio-economic relationships. Sci. Total Environ. 607-608 , 911-919, doi:10.1016/j.scitotenv.2017.07.044 (2017). Aide, T. M. & Grau, H. R. Ecology. Globalization, migration, and Latin American ecosystems. Science 305 , 1915-1916, doi:10.1126/science.1103179 (2004). Wang, C., Zhang, Y., Yang, Y., Yang, Q. & Hong, J. What is driving the abandonment of villages in the mountains of Southeast China? Land Degradation & Development 30 , 1183-1192, doi:10.1002/ldr.3303 (2019). Kan, K. & Chen, J. Rural Urbanization in China: Administrative Restructuring and the Livelihoods of Urbanized Rural Residents. Journal of Contemporary China 31 , 626-643, doi:10.1080/10670564.2021.1985841 (2022). Chen, Y., Li, X., Liu, X., Zhang, Y. & Huang, M. Quantifying the teleconnections between local consumption and domestic land uses in China. Landsc. Urban Plann. 187 , 60-69, doi:https://doi.org/10.1016/j.landurbplan.2019.03.011 (2019). Pendrill, F. et al. Agricultural and forestry trade drives large share of tropical deforestation emissions. Global Environmental Change 56 , 1-10, doi:https://doi.org/10.1016/j.gloenvcha.2019.03.002 (2019). Boone, C. G. et al. in Rethinking Global Land Use in an Urban Era 313-330 (MIT Press Scholarship Online, 2014). Pickett, S. T. A. & Zhou, W. Global urbanization as a shifting context for applying ecological science toward the sustainable city. Ecosystem Health and Sustainability 1 , 1-15, doi:10.1890/ehs14-0014.1 (2017). Zhou, W., Pickett, S. T. A. & McPhearson, T. Conceptual frameworks facilitate integration for transdisciplinary urban science. npj Urban Sustainability 1 , 1, doi:10.1038/s42949-020-00011-9 (2021). Marcotullio, P. J. & Solecki, W. in Urbanization and Sustainability (eds Christopher G. Boone & Michail Fragkias) Ch. Chapter 2, 11-25 (Springer Netherlands, 2013). Gutierrez-Velez, V. H., Gilbert, M. R., Kinsey, D. & Behm, J. E. Beyond the ‘urban’ and the ‘rural’: conceptualizing a new generation of infrastructure systems to enable rural–urban sustainability. Current Opinion in Environmental Sustainability 56 , 101177, doi:https://doi.org/10.1016/j.cosust.2022.101177 (2022). McHale, M. et al. The New Global Urban Realm: Complex, Connected, Diffuse, and Diverse Social-Ecological Systems. Sustainability 7 , 5211-5240, doi:10.3390/su7055211 (2015). Zhaxi, D., Zhou, W., Pickett, S. T. A., Guo, C. & Yao, Y. Urbanity mapping reveals the complexity, diffuseness, diversity, and connectivity of urbanized areas. Geography and Sustainability 5 , 357-369, doi:https://doi.org/10.1016/j.geosus.2024.03.004 (2024). Hoffmann, E. M. et al. Rurbanity: a concept for the interdisciplinary study of rural–urban transformation. Sustainability Science 18 , 1739-1753, doi:10.1007/s11625-023-01331-2 (2023). Gilbert, M. R., Eakin, H. & McPhearson, T. The role of infrastructure in societal transformations. Current Opinion in Environmental Sustainability 57 , 101207, doi:https://doi.org/10.1016/j.cosust.2022.101207 (2022). Boudet, F., MacDonald, G. K., Robinson, B. E. & Samberg, L. H. Rural-urban connectivity and agricultural land management across the Global South. Global Environmental Change 60 , 101982, doi:10.1016/j.gloenvcha.2019.101982 (2020). Johnson, K. M. & Lichter, D. T. Rural Depopulation: Growth and Decline Processes over the Past Century. Rural Sociology 84 , 3-27, doi:https://doi.org/10.1111/ruso.12266 (2019). Zhou, W. et al. Beyond city expansion: multi-scale environmental impacts of urban megaregion formation in China. National Science Review 9 , nwab107, doi:10.1093/nsr/nwab107 (2022). Kapri, K. & Ghimire, S. Migration, remittance, and agricultural productivity: Evidence from the Nepal Living Standard Survey. World Development Perspectives 19 , 100198, doi:10.1016/j.wdp.2020.100198 (2020). Sheng, Y., Zhao, Y., Zhang, Q., Dong, W. & Huang, J. Boosting rural labor off-farm employment through urban expansion in China. World Development 151 , 105727, doi:10.1016/j.worlddev.2021.105727 (2022). Dawazhaxi, Zhou, W., Yu, W., Yao, Y. & Jing, C. Understanding the indirect impacts of urbanization on vegetation growth using the Continuum of Urbanity framework. Sci. Total Environ. 899 , 165693, doi:https://doi.org/10.1016/j.scitotenv.2023.165693 (2023). Lin, G., Xie, X. & Lv, Z. Taobao practices, everyday life and emerging hybrid rurality in contemporary China. Journal of Rural Studies 47 , 514-523, doi:10.1016/j.jrurstud.2016.05.012 (2016). Apergis, N. & Li, J. Population and lifestyle trend changes in China: implications for environmental quality. Applied Economics 48 , 5246-5256, doi:10.1080/00036846.2016.1173184 (2016). Salerno, J. Migrant decision-making in a frontier landscape. Environmental Research Letters 11 , 044019, doi:10.1088/1748-9326/11/4/044019 (2016). Shao, S., Li, B., Fan, M. & Yang, L. How does labor transfer affect environmental pollution in rural China? Evidence from a survey. Energy Economics 102 , 105515, doi:10.1016/j.eneco.2021.105515 (2021). Nguyen, D. L., Grote, U. & Nguyen, T. T. Migration, crop production and non-farm labor diversification in rural Vietnam. Economic Analysis and Policy 63 , 175-187, doi:https://doi.org/10.1016/j.eap.2019.06.003 (2019). Cui, H., Wang, Y., Wang, W. & Liu, C. Rural households' livelihood transitions in China: processes, drivers and outcomes. China Agricultural Economic Review 17 , 171-190, doi:10.1108/CAER-11-2023-0346 (2025). Wang, Z. et al. Escaping social–ecological traps through ecological restoration and socioeconomic development in China's Loess Plateau. People and Nature 5 , 1364-1379, doi:https://doi.org/10.1002/pan3.10513 (2023). Piao, S. et al. Characteristics, drivers and feedbacks of global greening. Nature Reviews Earth & Environment 1 , 14-27, doi:10.1038/s43017-019-0001-x (2020). Zhu, Z. et al. Greening of the Earth and its drivers. Nature Climate Change 6 , 791-795, doi:10.1038/nclimate3004 (2016). Hu, T. et al. Social-ecological impacts on forest conservation and restoration in Pan-tropical Asia Karst. Environmental Impact Assessment Review 114 , 107962, doi:https://doi.org/10.1016/j.eiar.2025.107962 (2025). Black, R. et al. The effect of environmental change on human migration. Global Environmental Change 21 , S3-S11, doi:https://doi.org/10.1016/j.gloenvcha.2011.10.001 (2011). Bai, X. & Shi, P. China’s urbanization at a turning point—challenges and opportunities. Science 388 , eadw3443, doi:10.1126/science.adw3443 (2025). Ma, L., He, Q., Long, H., Zhang, Y. & Liao, L. Rural return migration in the post COVID-19 China: Incentives and barriers. Journal of Rural Studies 107 , 103258, doi:https://doi.org/10.1016/j.jrurstud.2024.103258 (2024). Chen, Z. et al. Public transport accessibility in villages in and around major Chinese cities. Nature Cities , doi:10.1038/s44284-025-00277-z (2025). Jiang, Y. et al. How transportation infrastructure construction improves economic development and ecological efficiency?—Evidence from 160 China's assisted rural counties. Journal of Cleaner Production 520 , 146182, doi:https://doi.org/10.1016/j.jclepro.2025.146182 (2025). Cheng, M., An, M. & Tan, H. After Returning to the Rural: The (Un)Sustainable Reintegration of Internal Migrant Workers in China. Population, Space and Place 31 , e2885, doi:https://doi.org/10.1002/psp.2885 (2025). Chen, L. et al. Return migration and in-situ urbanization of 79 migrant sending counties in China: Characteristic and driving factors. Journal of Rural Studies 104 , 103155, doi:https://doi.org/10.1016/j.jrurstud.2023.103155 (2023). Aide, T. M. et al. Deforestation and Reforestation of Latin America and the Caribbean (2001–2010). Biotropica 45 , 262-271, doi:https://doi.org/10.1111/j.1744-7429.2012.00908.x (2013). Hecht, S. B. & Saatchi, S. S. Globalization and Forest Resurgence: Changes in Forest Cover in El Salvador. BioScience 57 , 663-672, doi:10.1641/B570806 (2007). Zhang, R., Jiang, G. & Zhang, Q. Does urbanization always lead to rural hollowing? Assessing the spatio-temporal variations in this relationship at the county level in China 2000–2015. Journal of Cleaner Production 220 , 9-22, doi:10.1016/j.jclepro.2019.02.148 (2019). Song, B., Robinson, G. M. & Zhou, Z. Agricultural transformation and ecosystem services: A case study from Shaanxi Province, China. Habitat International 69 , 114-125, doi:10.1016/j.habitatint.2017.09.008 (2017). Radel, C. et al. Migration as a feature of land system transitions. Current Opinion in Environmental Sustainability 38 , 103-110, doi:https://doi.org/10.1016/j.cosust.2019.05.007 (2019). Hu, Y., Wang, Y. & Zhang, P. Anti-urbanization and rural development: Evidence from return migrants in China. Journal of Rural Studies 103 , 103102, doi:10.1016/j.jrurstud.2023.103102 (2023). Zhu, Y., Wang, W. W., Lin, L., Shen, J. & Ren, Q. Return migration and in situ urbanization of migrant sending areas: Insights from a survey of seven provinces in China. Cities 115 , 103242, doi:10.1016/j.cities.2021.103242 (2021). Wang, Y., Yang, A., Li, Y. & Yang, Q. Effect of e-commerce popularization on farmland abandonment in rural China: Evidence from a large-scale household survey. Land Use Policy 135 , 106958, doi:10.1016/j.landusepol.2023.106958 (2023). Jokisch, B. D., Radel, C., Carte, L. & Schmook, B. Migration matters: How migration is critical to contemporary human–environment geography. Geography Compass 13 , e12460, doi:https://doi.org/10.1111/gec3.12460 (2019). Zhu, J., Zhu, M. & Xiao, Y. Urbanization for rural development: Spatial paradigm shifts toward inclusive urban-rural integrated development in China. Journal of Rural Studies 71 , 94-103, doi:https://doi.org/10.1016/j.jrurstud.2019.08.009 (2019). Hoffmann, R., Wiederkehr, C., Dimitrova, A. & Hermans, K. Agricultural livelihoods, adaptation, and environmental migration in sub-Saharan drylands: a meta-analytical review. Environmental Research Letters 17 , 083003, doi:10.1088/1748-9326/ac7d65 (2022). Xiang, S., Zhou, M., Huang, L., Shan, L. & Wang, K. Assessing the dynamic land utilization efficiency and relevant driving mechanism in in-situ urbanized rural areas: A case study of 1979 administrative villages in Hangzhou. Environmental Impact Assessment Review 101 , 107111, doi:https://doi.org/10.1016/j.eiar.2023.107111 (2023). Rasmussen, L. V. et al. Social-ecological outcomes of agricultural intensification. Nature Sustainability 1 , 275-282, doi:10.1038/s41893-018-0070-8 (2018). Crawford, C. L., Yin, H., Radeloff, V. C. & Wilcove, D. S. Rural land abandonment is too ephemeral to provide major benefits for biodiversity and climate. Science Advances 8 , eabm8999, doi:10.1126/sciadv.abm8999 (2022). Oldekop, J. A. et al. Forest-linked livelihoods in a globalized world. Nature Plants 6 , 1400-1407, doi:10.1038/s41477-020-00814-9 (2020). Nunes, C. A. et al. Linking land-use and land-cover transitions to their ecological impact in the Amazon. Proceedings of the National Academy of Sciences 119 , e2202310119, doi:10.1073/pnas.2202310119 (2022). Chen, F., Chen, W. & Qiu, H. Poverty alleviation resettlement in China reduces deforestation. Proceedings of the National Academy of Sciences 122 , e2421526122, doi:10.1073/pnas.2421526122 (2025). Yang, H., Yang, W., Zhang, J., Connor, T. & Liu, J. Revealing pathways from payments for ecosystem services to socioeconomic outcomes. Science Advances 4 , eaao6652, doi:10.1126/sciadv.aao6652 (2018). Ellis, E. C. et al. An aspirational approach to planetary futures. Nature , doi:10.1038/s41586-025-09080-1 (2025). Injong, R. et al. On the Levy-Walk Nature of Human Mobility. IEEE/ACM Transactions on Networking 19 , 630-643, doi:10.1109/TNET.2011.2120618 (2011). Sarstedt, M., Radomir, L., Moisescu, O. I. & Ringle, C. M. Latent class analysis in PLS-SEM: A review and recommendations for future applications. Journal of Business Research 138 , 398-407, doi:10.1016/j.jbusres.2021.08.051 (2022). Sarstedt, M., Ringle, C. M., Henseler, J. & Hair, J. F. On the Emancipation of PLS-SEM: A Commentary on Rigdon (2012). Long Range Planning 47 , 154-160, doi:10.1016/j.lrp.2014.02.007 (2014). Additional Declarations There is NO Competing Interest. Supplementary Files Appendix.docx Appendix Figures Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7369333","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":523515996,"identity":"0607dfbb-f770-4024-ae3d-86a04815b3f0","order_by":0,"name":"Zhaxi Dawa","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA40lEQVRIiWNgGAWjYDCCA0DM2MDAwMbeQLIWngOkamGQSCBSB9/xM4Yff+6wS+yTfPvww4eKOgb+9gOMnwvwaJE8k2MsIXkm2ZhNOt1YcsaZwwwSZxKYpWfg0WJwIHeDhGEbsxybdBobM28b0J03GNiYefBpOf9284/EtnoeNsljbMx//9UxyBPUciN3m8TBtsNybBJsbMyMDcxAEQJaJG+8/2bZ2HbcmI0njVmy59hhHsMzic3S+LTwnU9LvvmzrTpxfvsxxg8/aurk5I4fPvgZnxYMwAONplEwCkbBKBgFlAAAa95HylCGYP0AAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0002-8272-0938","institution":"Research Center for Eco-Environmental Sciences,Chinese Academy of Sciences","correspondingAuthor":true,"prefix":"","firstName":"Zhaxi","middleName":"","lastName":"Dawa","suffix":""},{"id":523515997,"identity":"7ba02f46-bf48-43d4-bc5b-3e3c9b653fa3","order_by":1,"name":"Weiqi Zhou","email":"","orcid":"https://orcid.org/0000-0001-7323-4906","institution":"Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Weiqi","middleName":"","lastName":"Zhou","suffix":""},{"id":523515998,"identity":"79cac9ee-4ab4-4ab9-be4d-9797422db033","order_by":2,"name":"Steward Pickett","email":"","orcid":"https://orcid.org/0000-0002-1899-976X","institution":"Cary Institute of Ecosystem Studies","correspondingAuthor":false,"prefix":"","firstName":"Steward","middleName":"","lastName":"Pickett","suffix":""},{"id":523515999,"identity":"83f93556-4594-4755-bf6f-728b25fd020a","order_by":3,"name":"Daniel Childers","email":"","orcid":"","institution":"Arizona State University","correspondingAuthor":false,"prefix":"","firstName":"Daniel","middleName":"","lastName":"Childers","suffix":""},{"id":523516000,"identity":"48f324e2-b773-41c0-a386-9547561b771b","order_by":4,"name":"Wenjuan Yu","email":"","orcid":"","institution":"Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Wenjuan","middleName":"","lastName":"Yu","suffix":""}],"badges":[],"createdAt":"2025-08-14 03:25:13","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7369333/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7369333/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":92685720,"identity":"72378a4d-ebc0-4814-9113-a6a1c4e07319","added_by":"auto","created_at":"2025-10-03 03:17:10","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":92688,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe continuum of urbanity framework.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-7369333/v1/a54337d1ee17e918eb564ad6.png"},{"id":92686002,"identity":"854d0ebf-e63b-44f1-92d2-7db4cbf9c9aa","added_by":"auto","created_at":"2025-10-03 03:25:10","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":267833,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eHypothesized exogenous–endogenous pathways based on the continuum of urbanity framework.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-7369333/v1/d06a352a5e45b2e4073f0173.png"},{"id":92685574,"identity":"4339f1b3-dfed-417c-9ad3-cc905810f223","added_by":"auto","created_at":"2025-10-03 03:09:10","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":387746,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eChanges in vegetation greenness and its drivers in urban-rural typology. a, \u003c/strong\u003eA statistical urban-rural typology was defined based on the National Statistical Office's urban-rural classification codes and points of interest. \u003cstrong\u003eb,\u003c/strong\u003e Mean changes in urbanity index, total population,EVI trend and greening area in 2010-2020.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-7369333/v1/e4c9baa63110177e8934e40b.png"},{"id":92685577,"identity":"656f4a03-1359-41cc-a06c-3e5b23cddf4f","added_by":"auto","created_at":"2025-10-03 03:09:10","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":743870,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe process of livelihood transition and quality of life upgrading in rural households. a,\u003c/strong\u003e Spatial distribution of surveyed households across counties in 2010-2020. \u003cstrong\u003eb,\u003c/strong\u003eOriginal household livelihood types by primary and secondary income: agriculture-dominant (Agri. Dom.),\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-7369333/v1/5c28537680007eb4bf22054c.png"},{"id":92685578,"identity":"5fe09311-2007-4509-900a-e6e7ddf62dcb","added_by":"auto","created_at":"2025-10-03 03:09:10","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":2210563,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMarginal effect plots from LMMs. \u003c/strong\u003eMarginal effects of \u003cstrong\u003ea,\u003c/strong\u003e UUI, \u003cstrong\u003eb, \u003c/strong\u003eRUI, \u003cstrong\u003ec,\u003c/strong\u003e rural population, \u003cstrong\u003ed,\u003c/strong\u003e de-farming, \u003cstrong\u003ee,\u003c/strong\u003e quality of life, \u003cstrong\u003ef,\u003c/strong\u003e tree cover, \u003cstrong\u003eg,\u003c/strong\u003e cropland ratio, \u003cstrong\u003eh,\u003c/strong\u003e food production, \u003cstrong\u003ei,\u003c/strong\u003e slope, \u003cstrong\u003ej,\u003c/strong\u003e temperature, \u003cstrong\u003ek,\u003c/strong\u003e precipitation, \u003cstrong\u003el,\u003c/strong\u003e CO\u003csub\u003e2\u003c/sub\u003e on EVI trend (red plots) and greening area (green plots).\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-7369333/v1/1ed508840a1a40cf2119b65e.png"},{"id":92685573,"identity":"c71acc58-0b16-4b60-9106-4219618035aa","added_by":"auto","created_at":"2025-10-03 03:09:10","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":560620,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePLS-SEM models for exogenous and endogenous driven mechanisms. a,\u003c/strong\u003e Exogenous-driven model. \u003cstrong\u003eb,\u003c/strong\u003e Endogenous-driven model. Arrows represent hypothesized causal links. Blue arrows indicate significant positive paths and red arrows indicate significant negative paths; grey dashed arrows are non-significant. Numbers on solid arrows are standardized path coefficients.\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-7369333/v1/59bc881a7331078bfddc2db1.png"},{"id":92685580,"identity":"9c9c4732-4142-4b96-a45a-2d387c0d9ef6","added_by":"auto","created_at":"2025-10-03 03:09:10","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":2042589,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eResearch framework and workflow.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"image7.png","url":"https://assets-eu.researchsquare.com/files/rs-7369333/v1/2c94e9a758a7983b518e4646.png"},{"id":99308068,"identity":"da4b4bc5-fce1-4f8e-aabf-d66abbb83839","added_by":"auto","created_at":"2025-12-31 16:07:41","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":7776828,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7369333/v1/d2c2327b-79c2-44ab-b7e2-00cf3e371259.pdf"},{"id":92685579,"identity":"c45c9879-d479-4302-bc7f-bef1e1e62ef6","added_by":"auto","created_at":"2025-10-03 03:09:10","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":7304440,"visible":true,"origin":"","legend":"Appendix Figures","description":"","filename":"Appendix.docx","url":"https://assets-eu.researchsquare.com/files/rs-7369333/v1/251a2ebea3afbdca070af544.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Urbanity in the countryside: interaction of livelihood, lifestyle, connectivity and rural greening","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eChina has experienced the world\u0026rsquo;s largest rural to urban migration over the past three decades, with more than 290 million people relocating to cities between 1990 and 2019 \u003csup\u003e1,2\u003c/sup\u003e. This demographic transition has contributed to widespread rural depopulation, agricultural land abandonment, and, in many cases, ecological recovery\u003csup\u003e3-5\u003c/sup\u003e. Many studies have attributed improvements in rural vegetation and ecosystem services to reduced human pressure, declining cultivation, or afforestation initiatives \u003csup\u003e6-10\u003c/sup\u003e. However, the dominant narrative\u0026mdash;linking depopulation directly to ecological recovery\u0026mdash;offers only a partial explanation \u003csup\u003e2,11-14\u003c/sup\u003e. It overlooks the deeper social and economic restructuring associated with urbanization that may indirectly shape ecological outcomes \u003csup\u003e15,16\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eUrbanization alters more than just land use or population numbers \u003csup\u003e2,17-21\u003c/sup\u003e. It also transforms rural livelihoods, lifestyles, and connectivity, which may in turn influence ecological conditions beyond urban areas \u003csup\u003e22-26\u003c/sup\u003e. These changes include shifts in employment, material consumption, and land management, yet such indirect social-ecological pathways remain understudied \u003csup\u003e27-29\u003c/sup\u003e. To address this gap, we adopt the concept of the continuum of urbanity \u003csup\u003e30-32\u003c/sup\u003e (\u003cstrong\u003eFig. 1\u003c/strong\u003e). It posits that urbanization may act as a quality rather than only an entity \u003csup\u003e33\u003c/sup\u003e, where even remote villages may exhibit non-farm livelihoods, urban-influenced lifestyles, and integration into broader flows of people and resources \u003csup\u003e34\u003c/sup\u003e. Mixtures of urban and rural qualities can appear widely in human settlements \u003csup\u003e32,35\u003c/sup\u003e. We quantify this mixture using an urbanity index to capture the intensity of urban traits in rural areas \u003csup\u003e36\u003c/sup\u003e, enabling us to explore how urbanization influences rural social-ecological systems beyond its direct physical impacts \u003csup\u003e37-39\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eWe hypothesize that urbanity drives rural ecological change through two distinct yet interrelated pathways\u0026mdash;exogenous and endogenous\u0026mdash;each reflecting a different scenario of social-ecological transformation (\u003cstrong\u003eFig. 2\u003c/strong\u003e). In the exogenous pathway, national urbanization policies and the concentration of urban features (e.g., infrastructure and job opportunities) in urban areas encourage rural-urban migration \u003csup\u003e34,40\u003c/sup\u003e. This rural out-migration reduces local labor capacity, prompting households to shift from farming to relying on off-farm income or remittance \u003csup\u003e41,42\u003c/sup\u003e, while consumption patterns also change toward less land-dependent lifestyles (e.g., purchasing food and fuel), promoting quality of life \u003csup\u003e43\u003c/sup\u003e. These adjustments in livelihoods and lifestyles lead to altered land management, fields are downsized, marginal land is abandoned, and vegetation recovers \u003csup\u003e44\u003c/sup\u003e. In contrast, the endogenous pathway originates within rural areas, where increased urban features\u0026mdash;such as improved roads, digital connection, and social services\u0026mdash;gradually shift livelihoods toward wage labor and services, and encourage more urban-like consumption \u003csup\u003e34,45,46\u003c/sup\u003e. These changes enhance household mobility and financial capacity, enabling migration as a later step. Circulating migrants bring back capital and knowledge that shape land-use decisions, such as afforestation or cropland consolidation \u003csup\u003e21,47-49\u003c/sup\u003e, that improve rural ecological quality. In this case, qualities of urbanity act as an endogenous forces embedded in rural areas rather than as a product of depopulation. Our study tests whether different sequences of demographic, livelihood, and lifestyle changes\u0026mdash;triggered by an urbanity index calculated for varying places\u0026mdash;lead to distinct ecological outcomes within the continuum of urbanity.\u003c/p\u003e\n\u003cp\u003eTo evaluate these hypotheses, we integrated satellite-derived vegetation indices, including Enhanced Vegetation Index (EVI) trends and the amount of greening area, with an infrastructure-based urbanity index across China from 2010 to 2020. We examined their spatial distribution and relationship with population dynamics across the urban\u0026ndash;rural typology. In combination with household survey data, we assessed how non-farm livelihoods (classified by income structure) \u003csup\u003e50\u003c/sup\u003e and quality of life (measured by the inverse of the Engel coefficient) \u003csup\u003e51\u003c/sup\u003e evolved with increasing urbanity. Moreover, we applied linear mixed-effects models (LMMs) and structural equation modeling (SEM) to quantify the direct effects of urbanity index on ecological quality and the indirect effects mediated through both exogenous and endogenous pathways. This study bridges macro environmental trends with micro socio-economic transformations, moving beyond the simplistic \u0026ldquo;depopulation leads to greening\u0026rdquo; narrative and highlighting urbanity beyond physical cities as a positive driver of sustainable rural social-ecological transitions.\u003c/p\u003e"},{"header":"2. Results","content":"\u003ch2\u003e2.1. \u0026nbsp; \u0026nbsp; Interaction of urbanity, population dynamics, household shifts, and ecological outcomes\u003c/h2\u003e\n\u003cp\u003eFrom 2010 to 2020, rural areas in China exhibited a synergistic transformation marked by increased urbanity index, depopulation, and ecological greening across the urban-rural typology (\u003cstrong\u003eFig. 3a\u003c/strong\u003e). The average urbanity index rose from 1.51 to 1.88, indicating the widespread diffusion of urban infrastructure and services into rural settlements, including villages and rural centers (\u003cstrong\u003eFig. 3a and Appendix Fig. 1\u003c/strong\u003e). Meanwhile, rural population density declined\u0026mdash;villages alone lost over 25 million residents\u0026mdash;coinciding with notable vegetation recovery, as evidenced by positive EVI trends and expanded greening areas (\u003cstrong\u003eFig. 3b\u003c/strong\u003e). These areas also showed increases in tree cover and food production, highlighting land use transitions amid demographic and infrastructural shifts (\u003cstrong\u003eAppendix Fig. 2\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eTo further explore these patterns, we integrated rural household-level income and consumption data (\u003cstrong\u003eFig. 4a\u003c/strong\u003e). A significant trend in the reduction of farming (here labelled \u0026ldquo;de-farming\u0026rdquo;) occurred during this period: the share of non-agricultural-dominant households increased from 33.2% to 68.1%. Over half of the households transitioned into de-agrarianized (DA) types, typically moving from farming-dependent or mixed-income profiles to predominantly non-farming livelihoods (\u003cstrong\u003eFig. 4b\u003c/strong\u003e\u003cstrong\u003e\u0026ndash;\u003c/strong\u003e\u003cstrong\u003ec\u003c/strong\u003e). Alongside this shift, household quality of life improved, as reflected by changes in the inverse of Engel coefficient (IEC which varies between 0 and 1)\u0026mdash;a widely used proxy for quality of life. Between 2010 and 2020, the average IEC increased by 0.06, indicating widespread reduction in the share of food expenditure in total household consumption. This shift suggests improved material conditions, greater access to market-based goods, and more diversified consumption patterns (\u003cstrong\u003eFig. 4d\u003c/strong\u003e). Quality of life gains were most pronounced in households with stable non-farm livelihoods (NFS, +0.13), followed by DA households (+0.08), while farming-dependent households (FS) experienced little to no improvement (\u0026ndash;0.01). The moderate gains among returning-agriculture (RA) households (+0.03) may reflect already elevated baseline conditions due to earlier non-farm exposure, whereas the steeper rise in DA households suggests rapid structural adjustment and catch-up in consumption-based well-being.\u003c/p\u003e\n\u003cp\u003eTo preliminarily assess the relationships among urbanity index, livelihoods, quality of life, land management, and ecological outcomes, we conducted partial correlation analyses while controlling for environmental covariates. Results indicated that the rural urbanity index (RUI) was positively associated with both the EVI trend (r = 0.07, p \u0026lt; 0.001) and greening area (r = 0.04, p \u0026lt; 0.01). The within-urban urbanity index (UUI) also showed a statistically significant correlation with the EVI trend (r = 0.07, p \u0026lt; 0.001), but not with greening area (r =\u0026nbsp;\u0026ndash;0.01, p = 0.51) (\u003cstrong\u003eFig. 4e and Appendix Table 1\u003c/strong\u003e). De-farming was positively associated with tree cover (r = 0.05, p \u0026lt; 0.001) and quality of life (r = 0.13, p \u0026lt; 0.001), and negatively associated with greening area (r =\u0026nbsp;\u0026ndash;0.04, p \u0026lt; 0.05), while its relationship with the EVI trend was negligible (r = 0.01, p = 0.60). Quality of life was marginally associated with the EVI trend (r = 0.03, p \u0026lt; 0.05), but not with other ecological indicators. In contrast, land management variables, particularly tree cover (0.16, p \u0026lt; 0.001) and food production (0.14, p \u0026lt; 0.001), showed the strongest positive associations with vegetation improvement, highlighting their central role in ecological recovery.\u003c/p\u003e\n\u003cp\u003eThe linear mixed-effects models confirmed these patterns (\u003cstrong\u003eFig. 5, Appendix Fig. 3 and Appendix Table 2\u003c/strong\u003e). For EVI trend, positive effects were observed for UUI (estimate = 0.004, p \u0026lt; 0.001), tree cover (0.0002, p \u0026lt; 0.001), and food production (0.00002, p \u0026lt; 0.001), while RUI (\u0026ndash;0.006, p \u0026lt; 0.001), RuralPop (\u0026ndash;0.0000045, p \u0026lt; 0.001), and cropland ratio (\u0026ndash;0.00008, p \u0026lt; 0.001) had negative effects. Livelihood transitions and quality of life had negligible coefficients (p \u0026gt; 0.99). In the greening area model, topographic slope (\u0026ndash;0.03, p \u0026lt; 0.001), temperature (\u0026ndash;0.09, p \u0026lt; 0.001), and CO₂ emissions (\u0026ndash;2.82, p \u0026lt; 0.001) were dominant negative predictors, while tree cover (0.005, p \u0026lt; 0.001) and food production (0.001, p \u0026lt; 0.001) showed significant positive effects. UUI had a strong negative influence on greening (\u0026ndash;2.17, p \u0026lt; 0.001), whereas RUI remained non-significant (p = 0.70). These findings suggest that ecological change is more closely linked to land conditions than to household-level socioeconomic shifts. Notably, areas with both population decline and vegetation recovery tended to occur in hilly terrain, with increased tree cover and food production, indicating that depopulation may support greening where land conditions are favorable (\u003cstrong\u003eAppendix Fig. 4, Appendix Fig. 5, Appendix Table 3, Kruskal\u0026ndash;Wallis test with Dunn\u0026rsquo;s post-hoc test\u003c/strong\u003e).\u003c/p\u003e\n\u003ch2\u003e2.2. \u0026nbsp; \u0026nbsp; Path analysis of the exogenous-endogenous hypothesis\u003c/h2\u003e\n\u003cp\u003eTo test the hypothesized exogenous\u0026ndash;endogenous social-ecological mechanisms, we applied Partial Least Squares Structural Equation Modeling (PLS-SEM) to disentangle how urbanity from different sources influenced rural ecological outcomes through household and land management transitions. Following the conceptual framework, we built two distinct models. The exogenous model captures the cascading effects of UUI, while the endogenous model reflects effects of RUI. In both models, agricultural activity was defined as a formative latent variable constructed from cropland ratio and food production, whereas ecological quality was modeled reflectively using EVI trend and greening area. In both models, urbanity index influenced rural population growth, although by different mechanisms: UUI reflected return migration driven by urban saturation or policy shifts, while RUI indicated locally enabled return under improved rural conditions and diversified livelihoods. These models allowed us to trace multiple direct and indirect pathways linking urbanity, household livelihoods and quality of life, migration, and land cover to ecological change.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe UUI and RUI influenced ecological quality through different pathways (\u003cstrong\u003eFig. 6\u003c/strong\u003e). In the exogenous model (\u003cstrong\u003eFig. 6a\u003c/strong\u003e), higher UUI led to increased rural population (0.21, p \u0026lt; 0.001), which in turn promoted livelihood transition, measured as de-farming (0.15, p \u0026lt; 0.001). De-farming, a shift away from farming-based livelihoods to non-farming, significantly improved quality of life (0.15, p \u0026lt; 0.001) but had a negative direct effect on ecological quality (\u0026ndash;0.03, p = 0.031). In contrast, the endogenous model showed that RUI directly promoted reduction in farming (here labelled \u0026ldquo;de-farming\u0026rdquo;; 0.11, p \u0026lt; 0.001), which further enhanced quality of life (0.15, p \u0026lt; 0.001) and led to rural population growth (0.15, p \u0026lt; 0.001) (\u003cstrong\u003eFig. 6b)\u003c/strong\u003e. However, such socio-economic shifts did not directly improve ecological conditions. Instead, ecological quality was mainly driven by environmental factors (slope, temperature, and CO₂) and land management variables, particularly tree cover (exogenous: 0.21; endogenous: 0.23; p \u0026lt; 0.001 for both).\u003c/p\u003e\n\u003cp\u003eThe indirect pathways exhibited notable differences between the exogenous and endogenous models. In the exogenous model, UUI triggered a chain of social-ecological responses that ultimately impaired ecological quality of rural areas. One significant negative pathway was UUI \u0026rarr; RuralPop \u0026rarr; De-farming \u0026rarr; Ecological quality (\u0026ndash;0.001, p \u0026lt; 0.05), suggesting that urban influence increased rural population, which then drove de-farming, and in turn, reduced vegetation greening (\u003cstrong\u003eFig. 6a and Appendix Table 4, Specific indirect effects of the exogenous model\u003c/strong\u003e). Another significant path extended further to include land management. UUI \u0026rarr; RuralPop \u0026rarr; De-farming \u0026rarr; Agriculture activity \u0026rarr; Ecological quality (\u0026ndash;0.0002, p \u0026lt; 0.01). These results imply that although de-farming reflect livelihood transition and reduce agricultural activity, it did not necessarily improve ecological quality in the exogenous context.\u003c/p\u003e\n\u003cp\u003eBy contrast, the endogenous model exhibited more coherent and positive links among livelihoods, land management, and ecological quality of rural areas. A key indirect pathway was De-farming \u0026rarr; RuralPop \u0026rarr; Agriculture activity \u0026rarr; Tree cover \u0026rarr; Ecological quality (0.0009, p \u0026lt; 0.01), illustrating that de-farming, when accompanied by population return, contributed to reduced agricultural pressure and enhanced forest regrowth (\u003cstrong\u003eFig. 6b and Appendix Table 5, Specific indirect effects of the endogenous model\u003c/strong\u003e). Another notable path was RUI \u0026rarr; De-farming \u0026rarr; RuralPop \u0026rarr; Agriculture activity \u0026rarr; Tree cover \u0026rarr; Ecological quality (0.0004, p \u0026lt; 0.01), indicating that rural urbanization promoted livelihood transformation and land management, ultimately facilitating ecological recovery. Additionally, De-farming \u0026rarr; RuralPop \u0026rarr; Agriculture activity \u0026rarr; Ecological quality (0.0038, p \u0026lt; 0.01) further reinforces the role of livelihood shifts in mediating ecological outcomes through agriculture activity contraction. These findings demonstrate that indirect ecological benefits emerge when socio-economic transitions in rural areas align with land recovery trajectories.\u003c/p\u003e\n\u003cp\u003eFinally, the endogenous-driven model yielded 17 significant paths out of 19, while the exogenous-driven model identified only 9 out of 21. This suggests that local rural urbanity shaped ecological quality through more diverse and consistent social-ecological linkages. In contrast, distant urbanity influenced ecological outcomes primarily through fewer pathways centered on population shifts. These differences point to stronger structural coherence in the endogenous model.\u003c/p\u003e"},{"header":"3.\tDiscussion","content":"\u003cp\u003eImprovements in ecological quality across rural areas—particularly vegetation greening—has often been attributed to climatic fluctuations, reduced human pressure, or large-scale afforestation programs \u003csup\u003e10,52,53\u003c/sup\u003e.Yet our findings show that greening between 2010 and 2020 was not merely due to depopulation, but also to urbanity-driven shifts in livelihoods, lifestyles, and land management. Regions with higher urbanity index exhibited a stronger tendency toward livelihood de-farming among rural households, improved quality of life, reduced agricultural activity, and expanded tree cover (\u003cstrong\u003eFig. 3b, Fig. 5e, Fig. 6, Appendix Fig. 2\u003c/strong\u003e). The exogenous–endogenous hypothesis revealed two contrasting mechanisms. Under the exogenous pathway, the UUI led to rural population return and de-farming, but yielded limited ecological benefits. Conversely, the endogenous pathway, shaped by the RUI, actively transformed livelihoods, enhanced quality of life, and promoted vegetation recovery through reduced land pressure. These findings challenge the depopulation-centric view and underscore the importance of rural urbanity in shaping its own social–ecological transitions \u003csup\u003e2,3,54\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThe exogenous–endogenous hypothesis highlights how interactions among urbanity, livelihood transitions, and migration collectively shape rural ecological outcomes. In the exogenous scenario (\u003cstrong\u003eFig. 6a\u003c/strong\u003e), migration to cities is commonly driven by the pursuit of higher income and improved living conditions \u003csup\u003e55\u003c/sup\u003e. However, our results show that UUI did not promote vegetation recovery (\u003cstrong\u003eAppendix Table 4\u003c/strong\u003e). Instead, it stimulated population return to rural areas, where returning households increasingly engaged in non-farm activities and became less reliant on agricultural income. Paradoxically, this shift did not lead to ecological improvement. In fact, areas influenced by UUI exhibited weaker greening effects, suggesting that even if returnees disengage from farming, their resettlement may place renewed pressure on rural land systems. Although UUI reflects distant urban features, its high level may signal enhanced connectivity and infrastructure, enabling easier return migration to rural areas\u0026nbsp;\u003csup\u003e36\u003c/sup\u003e. This suggests that rural population growth in the exogenous model could be driven not only by urban growth saturation\u003csup\u003e56\u003c/sup\u003e and pandemic lockdown\u0026nbsp;\u003csup\u003e57\u003c/sup\u003e, but also by improved transport links and policy incentives facilitating urban–rural connectivity\u0026nbsp;\u003csup\u003e58,59\u003c/sup\u003e. This finding aligns with previous research suggesting that shorter migration distances or more frequent rural returns are often associated with limited or unstable natural vegetation recovery\u0026nbsp;\u003csup\u003e2\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eIn contrast, the RUI-driven return appears to reflect endogenous attraction, where better infrastructure, services, and in-situ urbanization enabled return migration under more favorable socio-ecological contexts \u003csup\u003e60,61\u003c/sup\u003e (\u003cstrong\u003eFig.6b and Appendix Table 5\u003c/strong\u003e). Migrants returning to more urbanized rural areas often bring with them capital, skills, and exposure to urban norms, enabling investments in sustainable land practices such as agroforestry, ecological enterprises, or non-land-intensive services \u003csup\u003e62,63\u003c/sup\u003e. These transitions reduce agricultural intensity while supporting rural household income, contributing to long-term ecological resilience. In areas where such livelihood diversification accompanies repopulation, land degradation is less likely to recur \u003csup\u003e64,65\u003c/sup\u003e. For instance, green jobs and service-based economies enable rural populations to grow sustainably without intensifying land pressure \u003csup\u003e66\u003c/sup\u003e. This contrast underscores that migration alone is not sufficient to drive ecological change; rather, its effects depend on whether the urbanity level of the rural area, such as infrastructure, service access, and economic opportunity, supports sustainable transitions \u003csup\u003e67-71\u003c/sup\u003e. In other words, population return enhances ecological outcomes only when embedded within local urbanity-led transformations that enable sustainable livelihoods \u003csup\u003e72\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThe pathways from social to ecological improvement hinge critically on land management. Our analysis shows that increased tree cover and food production were the most consistent predictors of vegetation greening (\u003cstrong\u003eFig. 4e, Fig. 5\u003c/strong\u003e), reflecting more efficient land use rather than wholesale abandonment \u003csup\u003e73,74\u003c/sup\u003e. These benefits were most reliable under endogenous conditions, where institutional support, diversified livelihoods, and reduced land pressure reinforced greening outcomes. In contrast, exogenous-driven de-farming was less ecologically effective and more vulnerable to reversal when economic incentives changed \u003csup\u003e9,75\u003c/sup\u003e. While we found no direct impact of quality of life on vegetation, its indirect role through altered consumption and reduced resource extraction may influence long-term sustainability \u003csup\u003e76\u003c/sup\u003e. Therefore, policies should prioritize securing post-agricultural landscapes through regulatory or financial mechanisms \u003csup\u003e77-79\u003c/sup\u003e, and integrate social development metrics such as the Human development index (HDI) and Nature Relationship Index (NRI) into ecological assessments \u003csup\u003e80\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThe study has several limitations. We relied on two decadal snapshots of survey and satellite data, so we may miss finer-year dynamics. Our household data are at the county level, which could obscure village-scale heterogeneity and local feedbacks. Moreover, although the SEM revealed plausible causal chains, definitive causality and the role of time lags or feedback loops would require more sophisticated panel or experimental designs. Future research should use higher-resolution and higher-frequency data (e.g. annual panel surveys, village-level studies, fine-scale remote sensing) and causal inference methods to validate these pathways. Cross-regional or cross-national comparisons would also test whether the urbanity–greening dynamics observed in China hold in different social and policy contexts.\u003c/p\u003e"},{"header":"4.\tData and methods","content":"\u003ch2\u003e4.1. \u0026nbsp; \u0026nbsp; Ecological quality and its drivers\u003c/h2\u003e\n\u003cp\u003eWe used multiple geospatial datasets to represent ecological quality (EVI) dynamics and its potential drivers. All spatial layers were resampled to a common 1km grid, using mean values or fractional coverage as appropriate (\u003cstrong\u003eTable 1\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1 Data list.\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"559\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eVariable\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 219px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSource/Product\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSpatio-temporal information\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 151px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePre-processing or description\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eEcological quality\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 219px;\"\u003e\n \u003cp\u003eEnhanced vegetation index (EVI) from the MOD13A2 \u0026amp; MYD13A2\u0026nbsp;https://modis.gsfc.nasa.gov/data/dataprod/mod13.php\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e2010\u0026ndash;2020\u003c/p\u003e\n \u003cp\u003e(1,000 m)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 151px;\"\u003e\n \u003cp\u003eMean growing-season EVI; QA \u0026le;1; masked for EVI \u0026gt; 0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eUrbanity index\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 219px;\"\u003e\n \u003cp\u003eFrom urbanity mapping method \u003csup\u003e36\u003c/sup\u003e.\u003c/p\u003e\n \u003cp\u003ehttps://doi.org/10.1016/j.geosus.2024.03.004\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e2010, 2020\u003c/p\u003e\n \u003cp\u003e(1,000 m)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 151px;\"\u003e\n \u003cp\u003eQuantifies urban influence beyond physical urban areas (\u003cstrong\u003eAppendix 2.1\u003c/strong\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePopulation density\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 219px;\"\u003e\n \u003cp\u003eGHSpop + census calibration\u0026nbsp;http://doi.org/10.2905/0C6B9751-A71F-4062-830B-43C9F432370F\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e2010, 2020\u003c/p\u003e\n \u003cp\u003e(1,000 m)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 151px;\"\u003e\n \u003cp\u003eCalibrated with national census data at county level (\u003cstrong\u003eAppendix 2.2\u003c/strong\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRural household income\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" style=\"width: 219px;\"\u003e\n \u003cp\u003eChina Family Panel Studies (CFPS)\u0026nbsp;https://www.isss.pku.edu.cn/cfps/\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" style=\"width: 95px;\"\u003e\n \u003cp\u003e2010, 2020\u003c/p\u003e\n \u003cp\u003e(districts or counties)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 151px;\"\u003e\n \u003cp\u003eAmount of farm and non-farm income (\u0026yen;)\u003c/p\u003e\n \u003cp\u003e(\u003cstrong\u003eAppendix 3.2.1\u003c/strong\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRural household consumption\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 151px;\"\u003e\n \u003cp\u003eExpenditure on food and total daily spend (\u0026yen;)\u003c/p\u003e\n \u003cp\u003e(\u003cstrong\u003eAppendix 3.2.2\u003c/strong\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTree cover fraction\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 219px;\"\u003e\n \u003cp\u003eMOD44B v061\u003c/p\u003e\n \u003cp\u003ehttps://doi.org/10.5067/MODIS/MOD44B.061\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e2010, 2020\u003c/p\u003e\n \u003cp\u003e(250 m)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 151px;\"\u003e\n \u003cp\u003eResampled to 1 km\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCropland fraction\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 219px;\"\u003e\n \u003cp\u003eCACD dataset\u0026nbsp;https://doi.org/10.5281/zenodo.7936885\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e2010, 2020\u003c/p\u003e\n \u003cp\u003e(30 m)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 151px;\"\u003e\n \u003cp\u003eCalculate the ratio in a 1km grid\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFood production\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 219px;\"\u003e\n \u003cp\u003eChina County Statistical Yearbook+ grain yield equations\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e2010, 2020\u003c/p\u003e\n \u003cp\u003e(1,000 m)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 151px;\"\u003e\n \u003cp\u003eSpatialized food production layers (\u003cstrong\u003eAppendix 2.3\u003c/strong\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTemperature \u0026amp; precipitation\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 219px;\"\u003e\n \u003cp\u003ehttps://doi.org/10.11888/Meteoro.tpdc.270961\u003c/p\u003e\n \u003cp\u003ehttps://zenodo.org/records/3114194\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e2010, 2020\u003c/p\u003e\n \u003cp\u003e(1,000 m)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 151px;\"\u003e\n \u003cp\u003eAnnual mean temperature and mean precipitation\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCO₂ emissions\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 219px;\"\u003e\n \u003cp\u003eCAMS global emission inventories\u003c/p\u003e\n \u003cp\u003ehttps://doi.org/10.24381/1d158bec\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e2010, 2020\u003c/p\u003e\n \u003cp\u003e(10,000 m)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 151px;\"\u003e\n \u003cp\u003eResampled to 1 km (unit: ton\u0026middot;km⁻\u0026sup2;\u0026middot;yr⁻\u0026sup1;)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSlope\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 219px;\"\u003e\n \u003cp\u003eGMTED2010 (elevation-derived)\u0026nbsp;https://doi.org/10.3133/ofr20111073\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 95px;\"\u003e\n \u003cp\u003e2010\u003c/p\u003e\n \u003cp\u003e(232 m)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 151px;\"\u003e\n \u003cp\u003eResampled to 1 km\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003e4.2. \u0026nbsp; \u0026nbsp; Methodological processes\u003c/h2\u003e\n\u003cp\u003eThis study systematically assessed the impacts of urbanity index on rural ecological quality dynamics through a four-tier analytical framework: (1) urban\u0026ndash;rural typology analysis, (2) comparison of social-ecological characteristics across population\u0026ndash;EVI interaction zones, (3) household survey and quantitative modeling, and (4) building hypothetical pathways and analyzing indirect effects (\u003cstrong\u003eFig. 7\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea) Urban\u0026ndash;rural typology analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn this study, the urban-rural typology was defined based on basic administrative unit points to reflect the impact of urbanization on vegetation greenness under the socio-economic context of China, while ensuring spatial consistency with rural household survey data. We constructed a six-level urban\u0026ndash;rural typology based on Points of Interest (POIs) annotated with official urban\u0026ndash;rural statistical codes. These included: urban core (111), urban fringe (112), town core (121), town fringe (122), rural center (210), and village (220). A spatial focal analysis with a 5 km moving window \u003csup\u003e81\u003c/sup\u003e was applied to assign each 1000 m grid cell a dominant settlement type based on the mode of POIs within the window (\u003cstrong\u003eAppendix Fig. 10\u003c/strong\u003e). This yielded a gridded urban\u0026ndash;rural typology surface at 1000 m resolution. For each typology zone, changes in urbanity, population density, and vegetation indicators from 2010 to 2020 were extracted. Vegetation change was captured by two indices: the EVI trend, derived using the Sen + Slope method, and the greening area, defined as the spatial extent of significantly increasing EVI (p \u0026lt; 0.05) identified through Mann\u0026ndash;Kendall (MK) tests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eb) Comparison of social-ecological characteristics across rural population\u0026ndash;EVI interaction zones\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBased on the urban\u0026ndash;rural typology, we identified four types of interaction zones reflecting the joint dynamics of rural population density and EVI change: (1) Rural population decrease and EVI greening (RPopDe-G), (2) Rural population increase and EVI greening (RPopIn-G), (3) Rural population decrease and EVI browning (RPopDe-B), and (4) Rural population increase and EVI browning (RPopIn-B).We used Kruskal\u0026ndash;Wallis tests and Dunn\u0026rsquo;s post hoc comparisons to analyze differences in key social-ecological factors across rural areas (rural center and village), including urbanity, tree cover, cropland ratio, food production, slope, annual mean temperature, annual mean precipitation, and annual mean CO₂ emissions. This allowed us to determine the dominant drivers of EVI change under varying demographic contexts (\u003cstrong\u003eAppendix Fig. 4\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ec) Household survey and quantitative modeling\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo examine how urbanity index jointly affects ecological quality through household decisions and population shifts, we employed rural household income data from the China Family Panel Studies (CFPS) for the years 2010 and 2020 (N = 9,324). Using K-means clustering, households were classified into four livelihood types: agricultural-dominant (Agri. Dom.), agricultural with secondary non-agricultural income (Agri.+Non-Agri), non-agricultural with secondary farming income (Non-Agri.+Agri), non-agricultural-dominant (Non-Agri. Dom.) (\u003cstrong\u003eAppendix 3.2.1\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eTransitions among these four types over the 2010\u0026ndash;2020 period resulted in 16 livelihood change combinations. These were further categorized into four representative livelihood transition types: Farming Stable (FS), Re-agrarianized (RA), De-agrarianized (DA), and Non-Farming Stable (NFS). The sequence of FS \u0026rarr; RA \u0026rarr; DA \u0026rarr; NFS was used to represent an increasing degree of de-farming, indicating a progressive shift from agriculture-dependent to non-agriculture-dominant livelihoods.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe Engel coefficient is calculated as the proportion of a household\u0026rsquo;s total expenditure devoted to food consumption (\u003cstrong\u003eAppendix 3.2.2\u003c/strong\u003e). We used its inverse\u0026mdash;referred to as the inverse of Engel coefficient (IEC), as a proxy for household quality of life, where higher IEC values indicate lower food expenditure shares and thus more urbanized and diversified lifestyles.\u003c/p\u003e\n\u003cp\u003eControlling for natural factors such as slope, temperature, precipitation, and CO₂ emissions, we first performed Spearman partial correlation analysis to assess relationships between EVI indicators (EVI trend and greening area) and potential drivers including urbanity, rural population density, livelihood type, quality of life, and land management variables (tree cover, cropland ratio, and food production) (\u003cstrong\u003eAppendix Table 1\u003c/strong\u003e). We then applied linear mixed-effects models (LMMs) to estimate the marginal effects of each explanatory variable on ecological quality, capturing their net direct impacts (\u003cstrong\u003esee the 4.3\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ed) \u0026nbsp; Build hypothetical pathways and analyze indirect effects\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBased on the \u0026ldquo;exogenous\u0026ndash;endogenous\u0026rdquo; hypothesis, we constructed two Partial Least Squares Structural Equation Models (PLS-SEM) to assess alternative pathways through which urbanity index affects rural ecological quality (\u003cstrong\u003esee the 4.4\u003c/strong\u003e). The exogenous-driven model assumes that urban urbanity index (UUI) directly attracts rural populations, which in turn reshape livelihood strategies and household behaviors, thereby influencing vegetation dynamics. In contrast, the endogenous-driven model posits that the rural urbanity index (RUI) first transforms household livelihoods and lifestyles, subsequently inducing rural population outmigration and adjustments in land management.\u003c/p\u003e\n\u003ch2\u003e4.3. Linear Mixed Models (LMMs)\u003c/h2\u003e\n\u003cp\u003eTo evaluate the direct effects of UUI and RUI, rural household characteristics, and land management variables on ecological quality, we applied linear mixed-effects models (LMMs) using two outcome variables: EVI trend and greening area. These two models shared the same set of predictors and allowed random intercepts by county to account for unobserved heterogeneity across spatial units. The LMMs were specified as follows:\u003c/p\u003e\n\u003cp\u003elmer(EVI trend\u0026nbsp;\u0026sim;\u0026nbsp;UUI + RUI + RuralPop + De-farming + Quality of life + Tree cover + Cropland ratio + Food production + Slope + Temperature + Precipitation + CO\u003csub\u003e2\u003c/sub\u003e + (1 | county))\u003c/p\u003e\n\u003cp\u003elmer(Greening area\u0026nbsp;\u0026sim;\u0026nbsp;UUI + RUI + RuralPop + De-farming + Quality of life + Tree cover + Cropland ratio + Food production + Slope + Temperature + Precipitation + CO\u003csub\u003e2\u003c/sub\u003e+ (1 | county))\u003c/p\u003e\n\u003cp\u003eBoth models included the following fixed effects: UUI, RUI, RuralPop, de-farming, quality of life (IEC), tree cover, cropland ratio, food production, slope, temperature, precipitation, and CO₂ emissions. County was included as a random intercept to control for contextual differences.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe extracted marginal effects (fixed effects only) to quantify the net direct influence of each predictor. Statistical significance of the fixed effects was assessed via Type III ANOVA using Satterthwaite\u0026apos;s approximation for degrees of freedom (\u003cstrong\u003eAppendix Table 2\u003c/strong\u003e). Model goodness-of-fit was evaluated using Marginal R\u003csup\u003e2\u003c/sup\u003em (proportion of variance explained by fixed effects) and Conditional R\u003csup\u003e2\u003c/sup\u003ec (variance explained by both fixed and random effects) (\u003cstrong\u003eAppendix Fig. 3\u003c/strong\u003e). All models were implemented using the lme4 and lmerTest packages in R.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003e4.4. Partial Least Squares Structural Equation Modeling (PLS-SEM)\u003c/h2\u003e\n\u003cp\u003eTo investigate the complex causal pathways through which UUI and RUI influences ecological quality, we employed Partial Least Squares Structural Equation Modeling (PLS-SEM). PLS-SEM is a variance-based approach that is particularly suitable for exploratory modeling with multiple mediating variables and relatively small sample sizes or non-normal data distributions \u003csup\u003e82\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eIn this study, we constructed two conceptual models reflecting alternative causal logics: (1) The exogenous-driven model, where UUI first triggers rural population change, which then drives livelihood/lifestyle transformation, land use, and ecological quality. (2) The endogenous-driven model, where RUI influences household livelihoods and quality of life, which then affect rural population change and land management, ultimately impacting ecological quality.\u003c/p\u003e\n\u003cp\u003eBoth models were composed of formative and reflective constructs, the ecological quality was modeled as a reflective latent variable composed of two indicators: EVI trend and greening area. The agricultural activity was specified as a formative construct using cropland ratio and food production. All other variables (e.g., UUI, RUI, rural population, de-farming, quality of life, tree cover, slope, climate, CO₂) were included as single-item observed variables.\u003c/p\u003e\n\u003cp\u003ePath coefficients were standardized to allow for effect size comparisons. The significance of paths was assessed via bootstrapping with 5,000 resamples, and 95% confidence intervals were used to evaluate statistical robustness (p \u0026lt; 0.05). We reported both direct and total indirect effects, and decomposed multi-step mediating chains where relevant.\u003c/p\u003e\n\u003cp\u003eAlthough PLS-SEM does not rely on global fit indices like covariance-based SEM, we evaluated model quality using the following indicators (\u003cstrong\u003eAppendix Table 4 and Appendix Table 5\u003c/strong\u003e). R\u0026sup2; for endogenous constructs to assess explanatory power (i.e., ecological quality). Q\u0026sup2; (Stone\u0026ndash;Geisser\u0026rsquo;s) for predictive relevance using blindfolding procedure. Composite reliability and outer loading values for reflective indicators (threshold \u0026gt; 0.7). Variance Inflation Factor (VIF) for formative indicators to assess multicollinearity (threshold \u0026lt; 5) \u003csup\u003e83\u003c/sup\u003e. All modeling was performed using SmartPLS 4.0, and results were visualized using the built-in path diagram tools.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch3\u003eAcknowledgements\u003c/h3\u003e\n\u003cp\u003eWe acknowledge support from the National Natural Science Foundation (Grant No. U21A2010), the Postdoctoral Fellowship Program (Grade B) of China Postdoctoral Science Foundation (Grant No. GZB20250080), the National Natural Science Fund for Distinguished Young Scholars (Grant No. 42225104), the 2024 Beijing Municipal Government\u0026rsquo;s Financial Support Program for Foreign High-Level Talents (Grant No. J2024014), and the U.S. National Science Foundation (No. DEB-2224662) to the Central Arizona-Phoenix LTER Program.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eGoh, C.\u003cem\u003e et al.\u003c/em\u003e Urban China : toward efficient, inclusive, and sustainable urbanization. (World Bank, Washington, DC, 2014).\u003c/li\u003e\n\u003cli\u003eZhang, X.\u003cem\u003e et al.\u003c/em\u003e A large but transient carbon sink from urbanization and rural depopulation in China. \u003cem\u003eNature Sustainability\u003c/em\u003e\u003cstrong\u003e5\u003c/strong\u003e, 321-328, doi:10.1038/s41893-021-00843-y (2022).\u003c/li\u003e\n\u003cli\u003eHou, D., Meng, F. \u0026amp; Prishchepov, A. V. How is urbanization shaping agricultural land-use? Unraveling the nexus between farmland abandonment and urbanization in China. \u003cem\u003eLandsc. Urban Plann.\u003c/em\u003e\u003cstrong\u003e214\u003c/strong\u003e, 104170, doi:10.1016/j.landurbplan.2021.104170 (2021).\u003c/li\u003e\n\u003cli\u003eWang, S.\u003cem\u003e et al.\u003c/em\u003e Urbanization can benefit agricultural production with large-scale farming in China. \u003cem\u003eNature Food\u003c/em\u003e\u003cstrong\u003e2\u003c/strong\u003e, 183-191, doi:10.1038/s43016-021-00228-6 (2021).\u003c/li\u003e\n\u003cli\u003eKolecka, N. Greening trends and their relationship with agricultural land abandonment across Poland. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e\u003cstrong\u003e257\u003c/strong\u003e, 112340, doi:10.1016/j.rse.2021.112340 (2021).\u003c/li\u003e\n\u003cli\u003eShen, H.\u003cem\u003e et al.\u003c/em\u003e Urbanization-induced population migration has reduced ambient PM2.5 concentrations in China. \u003cem\u003eScience Advances\u003c/em\u003e\u003cstrong\u003e3\u003c/strong\u003e, e1700300, doi:10.1126/sciadv.1700300 (2017).\u003c/li\u003e\n\u003cli\u003eGebre, T. \u0026amp; Gebremedhin, B. The mutual benefits of promoting rural-urban interdependence through linked ecosystem services. \u003cem\u003eGlobal Ecology and Conservation\u003c/em\u003e\u003cstrong\u003e20\u003c/strong\u003e, e00707, doi:https://doi.org/10.1016/j.gecco.2019.e00707 (2019).\u003c/li\u003e\n\u003cli\u003eYang, S.\u003cem\u003e et al.\u003c/em\u003e Assessing the impacts of rural depopulation and urbanization on vegetation cover: Based on land use and nighttime light data in China, 2000\u0026ndash;2020. \u003cem\u003eEcological Indicators\u003c/em\u003e\u003cstrong\u003e159\u003c/strong\u003e, 111639, doi:https://doi.org/10.1016/j.ecolind.2024.111639 (2024).\u003c/li\u003e\n\u003cli\u003eHu, T.\u003cem\u003e et al.\u003c/em\u003e Social-ecological heterogeneity drove contrasting tree cover restoration in South China Karst. \u003cem\u003eCommunications Earth \u0026amp; Environment\u003c/em\u003e\u003cstrong\u003e5\u003c/strong\u003e, 484, doi:10.1038/s43247-024-01641-y (2024).\u003c/li\u003e\n\u003cli\u003e Chen, C.\u003cem\u003e et al.\u003c/em\u003e China and India lead in greening of the world through land-use management. \u003cem\u003eNature Sustainability\u003c/em\u003e\u003cstrong\u003e2\u003c/strong\u003e, 122-129, doi:10.1038/s41893-019-0220-7 (2019).\u003c/li\u003e\n\u003cli\u003e Daskalova, G. N. \u0026amp; Kamp, J. Abandoning land transforms biodiversity. \u003cem\u003eScience\u003c/em\u003e\u003cstrong\u003e380\u003c/strong\u003e, 581-583, doi:10.1126/science.adf1099 (2023).\u003c/li\u003e\n\u003cli\u003e Uchida, K., Matanle, P., Li, Y., Fujita, T. \u0026amp; Hiraiwa, M. K. Biodiversity change under human depopulation in Japan. \u003cem\u003eNature Sustainability\u003c/em\u003e, doi:10.1038/s41893-025-01578-w (2025).\u003c/li\u003e\n\u003cli\u003e Brandt, M.\u003cem\u003e et al.\u003c/em\u003e Human population growth offsets climate-driven increase in woody vegetation in sub-Saharan Africa. \u003cem\u003eNature Ecology \u0026amp; Evolution\u003c/em\u003e\u003cstrong\u003e1\u003c/strong\u003e, 0081, doi:10.1038/s41559-017-0081 (2017).\u003c/li\u003e\n\u003cli\u003e Garba, S. I., Ebmeier, S. K., Bastin, J.-F., Mollicone, D. \u0026amp; Holden, J. Wetland fragmentation associated with large populations across Africa. \u003cem\u003eNature Communications\u003c/em\u003e\u003cstrong\u003e16\u003c/strong\u003e, 5065, doi:10.1038/s41467-025-59373-2 (2025).\u003c/li\u003e\n\u003cli\u003e Bat\u0026aacute;ry, P.\u003cem\u003e et al.\u003c/em\u003e Biodiversity and human well-being trade-offs and synergies in villages. \u003cem\u003eNature Sustainability\u003c/em\u003e, doi:10.1038/s41893-025-01592-y (2025).\u003c/li\u003e\n\u003cli\u003e Hu, G.\u003cem\u003e et al.\u003c/em\u003e Mitigating the ripple effects of urbanization on farmland productivity and ecological security through inclusive urbanization strategies. \u003cem\u003enpj Urban Sustainability\u003c/em\u003e\u003cstrong\u003e5\u003c/strong\u003e, 12, doi:10.1038/s42949-025-00199-8 (2025).\u003c/li\u003e\n\u003cli\u003e Marques, A.\u003cem\u003e et al.\u003c/em\u003e Increasing impacts of land use on biodiversity and carbon sequestration driven by population and economic growth. \u003cem\u003eNature Ecology \u0026amp; Evolution\u003c/em\u003e\u003cstrong\u003e3\u003c/strong\u003e, 628-637, doi:10.1038/s41559-019-0824-3 (2019).\u003c/li\u003e\n\u003cli\u003e van Vliet, J. Direct and indirect loss of natural area from urban expansion. \u003cem\u003eNature Sustainability\u003c/em\u003e\u003cstrong\u003e2\u003c/strong\u003e, 755-763, doi:10.1038/s41893-019-0340-0 (2019).\u003c/li\u003e\n\u003cli\u003e Mbaiwa, J. E. Changes on traditional livelihood activities and lifestyles caused by tourism development in the Okavango Delta, Botswana. \u003cem\u003eTourism Management\u003c/em\u003e\u003cstrong\u003e32\u003c/strong\u003e, 1050-1060, doi:10.1016/j.tourman.2010.09.002 (2011).\u003c/li\u003e\n\u003cli\u003e Meyerson, F. A. B., Merino, L. \u0026amp; Durand, J. Migration and environment in the context of globalization. \u003cem\u003eFrontiers in Ecology and the Environment\u003c/em\u003e\u003cstrong\u003e5\u003c/strong\u003e, 182-190, doi:Doi 10.1890/1540-9295(2007)5[182:Maeitc]2.0.Co;2 (2007).\u003c/li\u003e\n\u003cli\u003e Salerno, J.\u003cem\u003e et al.\u003c/em\u003e Rural migration under climate and land systems change. \u003cem\u003eNature Sustainability\u003c/em\u003e\u003cstrong\u003e7\u003c/strong\u003e, 1092-1101, doi:10.1038/s41893-024-01396-6 (2024).\u003c/li\u003e\n\u003cli\u003e Liu, G., Wang, H., Cheng, Y., Zheng, B. \u0026amp; Lu, Z. The impact of rural out-migration on arable land use intensity: Evidence from mountain areas in Guangdong, China. \u003cem\u003eLand Use Policy\u003c/em\u003e\u003cstrong\u003e59\u003c/strong\u003e, 569-579, doi:10.1016/j.landusepol.2016.10.005 (2016).\u003c/li\u003e\n\u003cli\u003e Ye, X. \u0026amp; Chuai, X. Have rural settlement changes harmed ecosystem carbon in China? \u003cem\u003eApplied Geography\u003c/em\u003e\u003cstrong\u003e153\u003c/strong\u003e, 102917, doi:10.1016/j.apgeog.2023.102917 (2023).\u003c/li\u003e\n\u003cli\u003e Li, T.\u003cem\u003e et al.\u003c/em\u003e Gauging policy-driven large-scale vegetation restoration programmes under a changing environment: Their effectiveness and socio-economic relationships. \u003cem\u003eSci. Total Environ.\u003c/em\u003e\u003cstrong\u003e607-608\u003c/strong\u003e, 911-919, doi:10.1016/j.scitotenv.2017.07.044 (2017).\u003c/li\u003e\n\u003cli\u003e Aide, T. M. \u0026amp; Grau, H. R. Ecology. Globalization, migration, and Latin American ecosystems. \u003cem\u003eScience\u003c/em\u003e\u003cstrong\u003e305\u003c/strong\u003e, 1915-1916, doi:10.1126/science.1103179 (2004).\u003c/li\u003e\n\u003cli\u003e Wang, C., Zhang, Y., Yang, Y., Yang, Q. \u0026amp; Hong, J. What is driving the abandonment of villages in the mountains of Southeast China? \u003cem\u003eLand Degradation \u0026amp; Development\u003c/em\u003e\u003cstrong\u003e30\u003c/strong\u003e, 1183-1192, doi:10.1002/ldr.3303 (2019).\u003c/li\u003e\n\u003cli\u003e Kan, K. \u0026amp; Chen, J. Rural Urbanization in China: Administrative Restructuring and the Livelihoods of Urbanized Rural Residents. \u003cem\u003eJournal of Contemporary China\u003c/em\u003e\u003cstrong\u003e31\u003c/strong\u003e, 626-643, doi:10.1080/10670564.2021.1985841 (2022).\u003c/li\u003e\n\u003cli\u003e Chen, Y., Li, X., Liu, X., Zhang, Y. \u0026amp; Huang, M. Quantifying the teleconnections between local consumption and domestic land uses in China. \u003cem\u003eLandsc. Urban Plann.\u003c/em\u003e\u003cstrong\u003e187\u003c/strong\u003e, 60-69, doi:https://doi.org/10.1016/j.landurbplan.2019.03.011 (2019).\u003c/li\u003e\n\u003cli\u003e Pendrill, F.\u003cem\u003e et al.\u003c/em\u003e Agricultural and forestry trade drives large share of tropical deforestation emissions. \u003cem\u003eGlobal Environmental Change\u003c/em\u003e\u003cstrong\u003e56\u003c/strong\u003e, 1-10, doi:https://doi.org/10.1016/j.gloenvcha.2019.03.002 (2019).\u003c/li\u003e\n\u003cli\u003e Boone, C. G.\u003cem\u003e et al.\u003c/em\u003e in \u003cem\u003eRethinking Global Land Use in an Urban Era\u003c/em\u003e 313-330 (MIT Press Scholarship Online, 2014).\u003c/li\u003e\n\u003cli\u003e Pickett, S. T. A. \u0026amp; Zhou, W. Global urbanization as a shifting context for applying ecological science toward the sustainable city. \u003cem\u003eEcosystem Health and Sustainability\u003c/em\u003e\u003cstrong\u003e1\u003c/strong\u003e, 1-15, doi:10.1890/ehs14-0014.1 (2017).\u003c/li\u003e\n\u003cli\u003e Zhou, W., Pickett, S. T. A. \u0026amp; McPhearson, T. Conceptual frameworks facilitate integration for transdisciplinary urban science. \u003cem\u003enpj Urban Sustainability\u003c/em\u003e\u003cstrong\u003e1\u003c/strong\u003e, 1, doi:10.1038/s42949-020-00011-9 (2021).\u003c/li\u003e\n\u003cli\u003e Marcotullio, P. J. \u0026amp; Solecki, W. in \u003cem\u003eUrbanization and Sustainability\u003c/em\u003e (eds Christopher G. Boone \u0026amp; Michail Fragkias) Ch. Chapter 2, 11-25 (Springer Netherlands, 2013).\u003c/li\u003e\n\u003cli\u003e Gutierrez-Velez, V. H., Gilbert, M. R., Kinsey, D. \u0026amp; Behm, J. E. Beyond the \u0026lsquo;urban\u0026rsquo; and the \u0026lsquo;rural\u0026rsquo;: conceptualizing a new generation of infrastructure systems to enable rural\u0026ndash;urban sustainability. \u003cem\u003eCurrent Opinion in Environmental Sustainability\u003c/em\u003e\u003cstrong\u003e56\u003c/strong\u003e, 101177, doi:https://doi.org/10.1016/j.cosust.2022.101177 (2022).\u003c/li\u003e\n\u003cli\u003e McHale, M.\u003cem\u003e et al.\u003c/em\u003e The New Global Urban Realm: Complex, Connected, Diffuse, and Diverse Social-Ecological Systems. \u003cem\u003eSustainability\u003c/em\u003e\u003cstrong\u003e7\u003c/strong\u003e, 5211-5240, doi:10.3390/su7055211 (2015).\u003c/li\u003e\n\u003cli\u003e Zhaxi, D., Zhou, W., Pickett, S. T. A., Guo, C. \u0026amp; Yao, Y. Urbanity mapping reveals the complexity, diffuseness, diversity, and connectivity of urbanized areas. \u003cem\u003eGeography and Sustainability\u003c/em\u003e\u003cstrong\u003e5\u003c/strong\u003e, 357-369, doi:https://doi.org/10.1016/j.geosus.2024.03.004 (2024).\u003c/li\u003e\n\u003cli\u003e Hoffmann, E. M.\u003cem\u003e et al.\u003c/em\u003e Rurbanity: a concept for the interdisciplinary study of rural\u0026ndash;urban transformation. \u003cem\u003eSustainability Science\u003c/em\u003e\u003cstrong\u003e18\u003c/strong\u003e, 1739-1753, doi:10.1007/s11625-023-01331-2 (2023).\u003c/li\u003e\n\u003cli\u003e Gilbert, M. R., Eakin, H. \u0026amp; McPhearson, T. The role of infrastructure in societal transformations. \u003cem\u003eCurrent Opinion in Environmental Sustainability\u003c/em\u003e\u003cstrong\u003e57\u003c/strong\u003e, 101207, doi:https://doi.org/10.1016/j.cosust.2022.101207 (2022).\u003c/li\u003e\n\u003cli\u003e Boudet, F., MacDonald, G. K., Robinson, B. E. \u0026amp; Samberg, L. H. Rural-urban connectivity and agricultural land management across the Global South. \u003cem\u003eGlobal Environmental Change\u003c/em\u003e\u003cstrong\u003e60\u003c/strong\u003e, 101982, doi:10.1016/j.gloenvcha.2019.101982 (2020).\u003c/li\u003e\n\u003cli\u003e Johnson, K. M. \u0026amp; Lichter, D. T. Rural Depopulation: Growth and Decline Processes over the Past Century. \u003cem\u003eRural Sociology\u003c/em\u003e\u003cstrong\u003e84\u003c/strong\u003e, 3-27, doi:https://doi.org/10.1111/ruso.12266 (2019).\u003c/li\u003e\n\u003cli\u003e Zhou, W.\u003cem\u003e et al.\u003c/em\u003e Beyond city expansion: multi-scale environmental impacts of urban megaregion formation in China. \u003cem\u003eNational Science Review\u003c/em\u003e\u003cstrong\u003e9\u003c/strong\u003e, nwab107, doi:10.1093/nsr/nwab107 (2022).\u003c/li\u003e\n\u003cli\u003e Kapri, K. \u0026amp; Ghimire, S. Migration, remittance, and agricultural productivity: Evidence from the Nepal Living Standard Survey. \u003cem\u003eWorld Development Perspectives\u003c/em\u003e\u003cstrong\u003e19\u003c/strong\u003e, 100198, doi:10.1016/j.wdp.2020.100198 (2020).\u003c/li\u003e\n\u003cli\u003e Sheng, Y., Zhao, Y., Zhang, Q., Dong, W. \u0026amp; Huang, J. Boosting rural labor off-farm employment through urban expansion in China. \u003cem\u003eWorld Development\u003c/em\u003e\u003cstrong\u003e151\u003c/strong\u003e, 105727, doi:10.1016/j.worlddev.2021.105727 (2022).\u003c/li\u003e\n\u003cli\u003e Dawazhaxi, Zhou, W., Yu, W., Yao, Y. \u0026amp; Jing, C. Understanding the indirect impacts of urbanization on vegetation growth using the Continuum of Urbanity framework. \u003cem\u003eSci. Total Environ.\u003c/em\u003e\u003cstrong\u003e899\u003c/strong\u003e, 165693, doi:https://doi.org/10.1016/j.scitotenv.2023.165693 (2023).\u003c/li\u003e\n\u003cli\u003e Lin, G., Xie, X. \u0026amp; Lv, Z. Taobao practices, everyday life and emerging hybrid rurality in contemporary China. \u003cem\u003eJournal of Rural Studies\u003c/em\u003e\u003cstrong\u003e47\u003c/strong\u003e, 514-523, doi:10.1016/j.jrurstud.2016.05.012 (2016).\u003c/li\u003e\n\u003cli\u003e Apergis, N. \u0026amp; Li, J. Population and lifestyle trend changes in China: implications for environmental quality. \u003cem\u003eApplied Economics\u003c/em\u003e\u003cstrong\u003e48\u003c/strong\u003e, 5246-5256, doi:10.1080/00036846.2016.1173184 (2016).\u003c/li\u003e\n\u003cli\u003e Salerno, J. Migrant decision-making in a frontier landscape. \u003cem\u003eEnvironmental Research Letters\u003c/em\u003e\u003cstrong\u003e11\u003c/strong\u003e, 044019, doi:10.1088/1748-9326/11/4/044019 (2016).\u003c/li\u003e\n\u003cli\u003e Shao, S., Li, B., Fan, M. \u0026amp; Yang, L. How does labor transfer affect environmental pollution in rural China? Evidence from a survey. \u003cem\u003eEnergy Economics\u003c/em\u003e\u003cstrong\u003e102\u003c/strong\u003e, 105515, doi:10.1016/j.eneco.2021.105515 (2021).\u003c/li\u003e\n\u003cli\u003e Nguyen, D. L., Grote, U. \u0026amp; Nguyen, T. T. Migration, crop production and non-farm labor diversification in rural Vietnam. \u003cem\u003eEconomic Analysis and Policy\u003c/em\u003e\u003cstrong\u003e63\u003c/strong\u003e, 175-187, doi:https://doi.org/10.1016/j.eap.2019.06.003 (2019).\u003c/li\u003e\n\u003cli\u003e Cui, H., Wang, Y., Wang, W. \u0026amp; Liu, C. Rural households' livelihood transitions in China: processes, drivers and outcomes. \u003cem\u003eChina Agricultural Economic Review\u003c/em\u003e\u003cstrong\u003e17\u003c/strong\u003e, 171-190, doi:10.1108/CAER-11-2023-0346 (2025).\u003c/li\u003e\n\u003cli\u003e Wang, Z.\u003cem\u003e et al.\u003c/em\u003e Escaping social\u0026ndash;ecological traps through ecological restoration and socioeconomic development in China's Loess Plateau. \u003cem\u003ePeople and Nature\u003c/em\u003e\u003cstrong\u003e5\u003c/strong\u003e, 1364-1379, doi:https://doi.org/10.1002/pan3.10513 (2023).\u003c/li\u003e\n\u003cli\u003e Piao, S.\u003cem\u003e et al.\u003c/em\u003e Characteristics, drivers and feedbacks of global greening. \u003cem\u003eNature Reviews Earth \u0026amp; Environment\u003c/em\u003e\u003cstrong\u003e1\u003c/strong\u003e, 14-27, doi:10.1038/s43017-019-0001-x (2020).\u003c/li\u003e\n\u003cli\u003e Zhu, Z.\u003cem\u003e et al.\u003c/em\u003e Greening of the Earth and its drivers. \u003cem\u003eNature Climate Change\u003c/em\u003e\u003cstrong\u003e6\u003c/strong\u003e, 791-795, doi:10.1038/nclimate3004 (2016).\u003c/li\u003e\n\u003cli\u003e Hu, T.\u003cem\u003e et al.\u003c/em\u003e Social-ecological impacts on forest conservation and restoration in Pan-tropical Asia Karst. \u003cem\u003eEnvironmental Impact Assessment Review\u003c/em\u003e\u003cstrong\u003e114\u003c/strong\u003e, 107962, doi:https://doi.org/10.1016/j.eiar.2025.107962 (2025).\u003c/li\u003e\n\u003cli\u003e Black, R.\u003cem\u003e et al.\u003c/em\u003e The effect of environmental change on human migration. \u003cem\u003eGlobal Environmental Change\u003c/em\u003e\u003cstrong\u003e21\u003c/strong\u003e, S3-S11, doi:https://doi.org/10.1016/j.gloenvcha.2011.10.001 (2011).\u003c/li\u003e\n\u003cli\u003e Bai, X. \u0026amp; Shi, P. China\u0026rsquo;s urbanization at a turning point\u0026mdash;challenges and opportunities. \u003cem\u003eScience\u003c/em\u003e\u003cstrong\u003e388\u003c/strong\u003e, eadw3443, doi:10.1126/science.adw3443 (2025).\u003c/li\u003e\n\u003cli\u003e Ma, L., He, Q., Long, H., Zhang, Y. \u0026amp; Liao, L. Rural return migration in the post COVID-19 China: Incentives and barriers. \u003cem\u003eJournal of Rural Studies\u003c/em\u003e\u003cstrong\u003e107\u003c/strong\u003e, 103258, doi:https://doi.org/10.1016/j.jrurstud.2024.103258 (2024).\u003c/li\u003e\n\u003cli\u003e Chen, Z.\u003cem\u003e et al.\u003c/em\u003e Public transport accessibility in villages in and around major Chinese cities. \u003cem\u003eNature Cities\u003c/em\u003e, doi:10.1038/s44284-025-00277-z (2025).\u003c/li\u003e\n\u003cli\u003e Jiang, Y.\u003cem\u003e et al.\u003c/em\u003e How transportation infrastructure construction improves economic development and ecological efficiency?\u0026mdash;Evidence from 160 China's assisted rural counties. \u003cem\u003eJournal of Cleaner Production\u003c/em\u003e\u003cstrong\u003e520\u003c/strong\u003e, 146182, doi:https://doi.org/10.1016/j.jclepro.2025.146182 (2025).\u003c/li\u003e\n\u003cli\u003e Cheng, M., An, M. \u0026amp; Tan, H. After Returning to the Rural: The (Un)Sustainable Reintegration of Internal Migrant Workers in China. \u003cem\u003ePopulation, Space and Place\u003c/em\u003e\u003cstrong\u003e31\u003c/strong\u003e, e2885, doi:https://doi.org/10.1002/psp.2885 (2025).\u003c/li\u003e\n\u003cli\u003e Chen, L.\u003cem\u003e et al.\u003c/em\u003e Return migration and in-situ urbanization of 79 migrant sending counties in China: Characteristic and driving factors. \u003cem\u003eJournal of Rural Studies\u003c/em\u003e\u003cstrong\u003e104\u003c/strong\u003e, 103155, doi:https://doi.org/10.1016/j.jrurstud.2023.103155 (2023).\u003c/li\u003e\n\u003cli\u003e Aide, T. M.\u003cem\u003e et al.\u003c/em\u003e Deforestation and Reforestation of Latin America and the Caribbean (2001\u0026ndash;2010). \u003cem\u003eBiotropica\u003c/em\u003e\u003cstrong\u003e45\u003c/strong\u003e, 262-271, doi:https://doi.org/10.1111/j.1744-7429.2012.00908.x (2013).\u003c/li\u003e\n\u003cli\u003e Hecht, S. B. \u0026amp; Saatchi, S. S. Globalization and Forest Resurgence: Changes in Forest Cover in El Salvador. \u003cem\u003eBioScience\u003c/em\u003e\u003cstrong\u003e57\u003c/strong\u003e, 663-672, doi:10.1641/B570806 (2007).\u003c/li\u003e\n\u003cli\u003e Zhang, R., Jiang, G. \u0026amp; Zhang, Q. Does urbanization always lead to rural hollowing? Assessing the spatio-temporal variations in this relationship at the county level in China 2000\u0026ndash;2015. \u003cem\u003eJournal of Cleaner Production\u003c/em\u003e\u003cstrong\u003e220\u003c/strong\u003e, 9-22, doi:10.1016/j.jclepro.2019.02.148 (2019).\u003c/li\u003e\n\u003cli\u003e Song, B., Robinson, G. M. \u0026amp; Zhou, Z. Agricultural transformation and ecosystem services: A case study from Shaanxi Province, China. \u003cem\u003eHabitat International\u003c/em\u003e\u003cstrong\u003e69\u003c/strong\u003e, 114-125, doi:10.1016/j.habitatint.2017.09.008 (2017).\u003c/li\u003e\n\u003cli\u003e Radel, C.\u003cem\u003e et al.\u003c/em\u003e Migration as a feature of land system transitions. \u003cem\u003eCurrent Opinion in Environmental Sustainability\u003c/em\u003e\u003cstrong\u003e38\u003c/strong\u003e, 103-110, doi:https://doi.org/10.1016/j.cosust.2019.05.007 (2019).\u003c/li\u003e\n\u003cli\u003e Hu, Y., Wang, Y. \u0026amp; Zhang, P. Anti-urbanization and rural development: Evidence from return migrants in China. \u003cem\u003eJournal of Rural Studies\u003c/em\u003e\u003cstrong\u003e103\u003c/strong\u003e, 103102, doi:10.1016/j.jrurstud.2023.103102 (2023).\u003c/li\u003e\n\u003cli\u003e Zhu, Y., Wang, W. W., Lin, L., Shen, J. \u0026amp; Ren, Q. Return migration and in situ urbanization of migrant sending areas: Insights from a survey of seven provinces in China. \u003cem\u003eCities\u003c/em\u003e\u003cstrong\u003e115\u003c/strong\u003e, 103242, doi:10.1016/j.cities.2021.103242 (2021).\u003c/li\u003e\n\u003cli\u003e Wang, Y., Yang, A., Li, Y. \u0026amp; Yang, Q. Effect of e-commerce popularization on farmland abandonment in rural China: Evidence from a large-scale household survey. \u003cem\u003eLand Use Policy\u003c/em\u003e\u003cstrong\u003e135\u003c/strong\u003e, 106958, doi:10.1016/j.landusepol.2023.106958 (2023).\u003c/li\u003e\n\u003cli\u003e Jokisch, B. D., Radel, C., Carte, L. \u0026amp; Schmook, B. Migration matters: How migration is critical to contemporary human\u0026ndash;environment geography. \u003cem\u003eGeography Compass\u003c/em\u003e\u003cstrong\u003e13\u003c/strong\u003e, e12460, doi:https://doi.org/10.1111/gec3.12460 (2019).\u003c/li\u003e\n\u003cli\u003e Zhu, J., Zhu, M. \u0026amp; Xiao, Y. Urbanization for rural development: Spatial paradigm shifts toward inclusive urban-rural integrated development in China. \u003cem\u003eJournal of Rural Studies\u003c/em\u003e\u003cstrong\u003e71\u003c/strong\u003e, 94-103, doi:https://doi.org/10.1016/j.jrurstud.2019.08.009 (2019).\u003c/li\u003e\n\u003cli\u003e Hoffmann, R., Wiederkehr, C., Dimitrova, A. \u0026amp; Hermans, K. Agricultural livelihoods, adaptation, and environmental migration in sub-Saharan drylands: a meta-analytical review. \u003cem\u003eEnvironmental Research Letters\u003c/em\u003e\u003cstrong\u003e17\u003c/strong\u003e, 083003, doi:10.1088/1748-9326/ac7d65 (2022).\u003c/li\u003e\n\u003cli\u003e Xiang, S., Zhou, M., Huang, L., Shan, L. \u0026amp; Wang, K. Assessing the dynamic land utilization efficiency and relevant driving mechanism in in-situ urbanized rural areas: A case study of 1979 administrative villages in Hangzhou. \u003cem\u003eEnvironmental Impact Assessment Review\u003c/em\u003e\u003cstrong\u003e101\u003c/strong\u003e, 107111, doi:https://doi.org/10.1016/j.eiar.2023.107111 (2023).\u003c/li\u003e\n\u003cli\u003e Rasmussen, L. V.\u003cem\u003e et al.\u003c/em\u003e Social-ecological outcomes of agricultural intensification. \u003cem\u003eNature Sustainability\u003c/em\u003e\u003cstrong\u003e1\u003c/strong\u003e, 275-282, doi:10.1038/s41893-018-0070-8 (2018).\u003c/li\u003e\n\u003cli\u003e Crawford, C. L., Yin, H., Radeloff, V. C. \u0026amp; Wilcove, D. S. Rural land abandonment is too ephemeral to provide major benefits for biodiversity and climate. \u003cem\u003eScience Advances\u003c/em\u003e\u003cstrong\u003e8\u003c/strong\u003e, eabm8999, doi:10.1126/sciadv.abm8999 (2022).\u003c/li\u003e\n\u003cli\u003e Oldekop, J. A.\u003cem\u003e et al.\u003c/em\u003e Forest-linked livelihoods in a globalized world. \u003cem\u003eNature Plants\u003c/em\u003e\u003cstrong\u003e6\u003c/strong\u003e, 1400-1407, doi:10.1038/s41477-020-00814-9 (2020).\u003c/li\u003e\n\u003cli\u003e Nunes, C. A.\u003cem\u003e et al.\u003c/em\u003e Linking land-use and land-cover transitions to their ecological impact in the Amazon. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e\u003cstrong\u003e119\u003c/strong\u003e, e2202310119, doi:10.1073/pnas.2202310119 (2022).\u003c/li\u003e\n\u003cli\u003e Chen, F., Chen, W. \u0026amp; Qiu, H. Poverty alleviation resettlement in China reduces deforestation. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e\u003cstrong\u003e122\u003c/strong\u003e, e2421526122, doi:10.1073/pnas.2421526122 (2025).\u003c/li\u003e\n\u003cli\u003e Yang, H., Yang, W., Zhang, J., Connor, T. \u0026amp; Liu, J. Revealing pathways from payments for ecosystem services to socioeconomic outcomes. \u003cem\u003eScience Advances\u003c/em\u003e\u003cstrong\u003e4\u003c/strong\u003e, eaao6652, doi:10.1126/sciadv.aao6652 (2018).\u003c/li\u003e\n\u003cli\u003e Ellis, E. C.\u003cem\u003e et al.\u003c/em\u003e An aspirational approach to planetary futures. \u003cem\u003eNature\u003c/em\u003e, doi:10.1038/s41586-025-09080-1 (2025).\u003c/li\u003e\n\u003cli\u003e Injong, R.\u003cem\u003e et al.\u003c/em\u003e On the Levy-Walk Nature of Human Mobility. \u003cem\u003eIEEE/ACM Transactions on Networking\u003c/em\u003e\u003cstrong\u003e19\u003c/strong\u003e, 630-643, doi:10.1109/TNET.2011.2120618 (2011).\u003c/li\u003e\n\u003cli\u003e Sarstedt, M., Radomir, L., Moisescu, O. I. \u0026amp; Ringle, C. M. Latent class analysis in PLS-SEM: A review and recommendations for future applications. \u003cem\u003eJournal of Business Research\u003c/em\u003e\u003cstrong\u003e138\u003c/strong\u003e, 398-407, doi:10.1016/j.jbusres.2021.08.051 (2022).\u003c/li\u003e\n\u003cli\u003e Sarstedt, M., Ringle, C. M., Henseler, J. \u0026amp; Hair, J. F. On the Emancipation of PLS-SEM: A Commentary on Rigdon (2012). \u003cem\u003eLong Range Planning\u003c/em\u003e\u003cstrong\u003e47\u003c/strong\u003e, 154-160, doi:10.1016/j.lrp.2014.02.007 (2014).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Continuum of urbanity, social-ecological effects, Rural transition, Causality analysis","lastPublishedDoi":"10.21203/rs.3.rs-7369333/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7369333/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Rural ecosystem recovery is often attributed to population decline, but the social-economic pathways linking urbanization to ecological change remain underexplored. Grounded in the continuum of urbanity theory, we developed two alternative pathway models (exogenous-driven vs. endogenous-driven) to examine how urbanization influenced rural ecological quality in China from 2010 to 2020. Results indicate that ecological greening was most prominent in areas where urbanity increased while population declined, supported by rising forest cover, stable cropland use, and improved grain yield. Rural household livelihood and lifestyle changes had negligible direct impact on vegetation trends. However, the pathway models revealed significant indirect roles: urbanization effects were mediated by rural households through shifts from farming to non-farming livelihoods, improved quality of life, and rural-urban migration. The exogenous-driven model demonstrated stronger explanatory power and clearer pathway significance than the endogenous driven model. These findings move beyond the conventional narrative that rural depopulation alone leads to greening, offering a new social-ecological perspective on how urbanization can actively contribute to rural sustainability.","manuscriptTitle":"Urbanity in the countryside: interaction of livelihood, lifestyle, connectivity and rural greening","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-03 03:09:05","doi":"10.21203/rs.3.rs-7369333/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"f93ee1c5-bed2-40be-8b15-065084ed3a28","owner":[],"postedDate":"October 3rd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":55648027,"name":"Earth and environmental sciences/Ecology/Urban ecology"},{"id":55648028,"name":"Scientific community and society/Geography"}],"tags":[],"updatedAt":"2025-12-22T19:45:35+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-03 03:09:05","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7369333","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7369333","identity":"rs-7369333","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}