The gully edge line is a critical geomorphic feature that demarcates positive and negative terrains on the Loess Plateau.Its accurate and efficient extraction is fundamental for studying gully erosion and landscape evolution,and holds significant importance for guiding ecological restoration in small loess watersheds.To address the existing challenges of insufficient extraction accuracy,limited feature diversity,and the lack of a systematic evaluation framework,which hinder comprehensive and objective model assessment,this study employs machine learning models for positive/negative terrain segmentation and automatic gully edge line extraction in a small watershed of the loess tableland area.The key findings are as follows.（1）In the loess tableland area,the optimal feature subset selected by the random forest（RF）model comprises six features:elevation variation coefficient, B1,B11,B11_cont, B11_mean,and B9_mean.（2）A comparative analysis integrating classification accuracy for positive/negative terrains and gully edge line displacement showed that RF and BP neural network models achieved accuracies of 92.71% and 90.01% ,respectively.These represent improvements of 8.0% and 5.3% over the slope distortion neighborhood judgment method,with a corresponding gully edge offset of 32.67 m （within three raster pixels）.This study demonstrates that the optimal feature subset selected by the RF model is well–suited for loess tableland topography,resulting in fewer misclassifications and omissions,thereby providing a scientific basis for the accurate extraction of gully edge lines.

Slope units,defined as terrain units delineated by watersheds and drainage lines,serve as fundamental elements for characterizing the internal geomorphological structure and hydrological processes within watersheds.This study leverages the descriptive advantages of slope units for geomorphological analysis and applies them to identify six basic geomorphological types in China:large undulating mountains,medium undulating mountains,small undulating mountains,hills,platforms,and plains. Firstly,using 30 m–resolution DEM data,we selected 300 watershed sample areas across China for slope unit extraction.Secondly, weighted complex networks of slope units were innovatively constructed based on their physical structures,and corresponding network indices were calculated.Finally,machine learning algorithms（XGBoost,ET,RF,and LightGBM）were then employed for geomorphological type identification.It is found as follows.（1）All four machine learning algorithms based on network indices achieved an overall accuracy above 85%,with the LightGBM performing best（overall accuracy: 88.33%,Kappa coefficient: 0.86）.（2）For slope units at different scales,models built with the optimal parameter set exhibited the best performance in geomorphological type identification.（3）Integrating network indices with terrain indices improved the overall accuracy by 1.67% compared with using a single network index.SHAP–based feature importance analysis further confirmed that network indices play a key role in identifying all geomorphological types.By establishing an interdisciplinary framework that integrates geomorphological units,complex networks,and machine learning,this study advances slope unit–based geomorphological characterization and provides a foundation for geomorphological classification based on slope units.

The confidentiality of high–precision vector data poses increasing security challenges in data sharing.On the premise of ensuring data usability,how to achieve controllable accuracy reduction for such vector data has become a key challenge.Aiming at the lack of balance between usability and controllability in existing algorithms,a geometric accuracy reduction algorithm with controllable accuracy is proposed.Firstly,the original vector data is converted from rectangular coordinates to polar coordinates, which can control the perturbation distance and direction separately,making the accuracy control targeted and flexible.Control points are then selected using the light barrier method,ensuring alignment with the distribution density of the vector data. Secondly,a spatial coordinate offset model is established based on Hermite polynomials.The selected control points are input into this model to obtain initial parameters,which are then iteratively optimized using sample points to derive the final optimal parameters.Finally,the perturbation is added to the polar coordinates and then transformed back to the rectangular coordinates. Experiments demonstrate that the algorithm accurately achieves the target reduction level across different accuracy standards, enabling effective control.The processed data maintains high usability,showing excellent performance in topological relationship retention,shape pattern similarity,and spatial direction consistency.The algorithm exhibits outstanding controllability,safety, and reliability,facilitating the compliant and secure use of vector data,fully unlocking its value,and providing a strong guarantee for spatial data security applications.

Under global climate change,the frequency and intensity of drought events continue to increase,making the identification and understanding of abrupt changes in vegetation photosynthetic processes under drought stress crucial for ecological monitoring and climate adaptation management.However,although a variety of methods have been developed to identify change points in vegetation photosynthesis,a systematic evaluation of the performance differences among these methods and their ability to characterize drought–related responses is still lacking.Focusing on subtropical regions of China,this study utilized solar–induced chlorophyll fluorescence（SIF）data and the standardized precipitation evapotranspiration index（SPEI）data,and comprehensively applied four change point detection methods,namely BFAST（Breaks For Additive Season and Trend）,DBEST （Detecting Breakpoints and Estimating Segments in Trend）,PELT（Pruned Exact Linear Time）,and the Pettitt test,to synergistically analyze the characteristics of change points in regional vegetation photosynthetic dynamics from multiple dimensions and to characterize the spatiotemporal relationship between SIF and SPEI.It is found as follows.（1）SIF in subtropical China exhibited a upward trend from 2003 to 2022 ,reflecting an overall enhancement of regional vegetation photosynthetic activity.（2）Each detection method demonstrated specific advantages in detecting change points:BFAST performed well in capturing long–term trend variations,with monotonic increases accounting for 82.25% of its detected change types;DBEST showed strong capability in identifying gradual changes in complex terrains such as the Yunnan–Guizhou Plateau;PELT was highly sensitive to multiple breakpoints and frequent disturbances;and the Pettitt test primarily identified structural shifts caused by abrupt changes.（3）The correlation between SIF and SPEI displayed significant spatial heterogeneity,with predominantly positive correlations in hydrothermally favorable regions such as the Sichuan Basin and the Fujian–Guangdong coastal areas,and mainly negative correlations in southern Yunnan and eastern Xizang.（4）Integrated multi–method analysis further revealed that the overall correlation between SIF and SPEI increased by 112.07% after detected change points,whereas the increase associated with PELT–derived change points was only 62.94% ,indicating its greater susceptibility to noise.The complementary characteristics of BFAST and PELT improved the spatial completeness and robustness of change–point identification when jointly applied. Overall,the study provides theoretical support for assessing ecosystem resilience and climate adaptive management in subtropical regions of China.

The Heilonggang River Basin is a vital agricultural planting region in the North China Plain and a key transportation hub in Hebei Province.To investigate land subsidence in this region,this study employed the Stanford method for persistent scatterers（StaMPS）based on the interferometric synthetic aperture radar（InSAR）technique and utilized 157 scenes of Sentinel–1A ascending orbit images to derive surface subsidence information for the main upper reaches of the basin from January 2019 to June 2024,and analyzed its primary driving factors.It is found as follows.（1）There were multiple subsidence zones within the study area.Notably,a subsidence zone located between Julu County and Nangong City exceeded 1400km² ,with a maximum subsidence rate exceeding 40mm/a .（2）Several major highways,including National Highways G106,G340,G514,and the Xingde Line,exhibited subsidence–prone sections longer than 30 km （the longest affected stretch exceeded 60 km ）due to combined effects of regional land subsidence and vehicular dynamic loads.（3）Analysis indicates that the primary driving factors for the land subsidence in the study area are the underlying soil and lithology distribution,coupled with variations in groundwater depth.

Taking 40186 city pairs from 284 cities in China from 2011 to 2021 as the research samples,this study employs an improved efficiency value–added model,the social network analysis,and the dynamic GMM to examine the evolution characteristics and multi–dimensional proximity drivers of regional synergistic emission reduction networks in China.It is found as follows. （1）The overall density of regional synergistic emission reduction networks followed a dual–cycle pattern of＂high–low–high–low＂ with two distinct declines during the study period.Notably,the network densities of the three cross–regional networks（the Eastern–Central,Eastern–Western,and Central–Western networks）were significantly higher than those of the three within–region networks（the Eastern,Central,and Western networks）,highlighting the＂cross–regional＂nature of synergistic efforts.（2）The regional synergistic emission reduction network has undergone a dynamic evolution from a fan–shaped network pattern of ＂three–region grouping and sporadic coverage＂to a spindle–shaped nested network pattern of＂east–west linkage and multi–region support＂.Moreover,the majority of synergistic relationships were led by the eastern region,underscoring the＂east–led＂ characteristic of the network.（3）The cross–regional feature is primarily driven by geographical proximity,which leads to technological lock–in,homogenized industrial competition,and more severe public pollution problems.The east–led feature stems from stronger circular cumulative causality effects in eastern network,eastern–western network,and eastern–central network, together with more pronounced positive impacts of institutional,technological,and virtual proximity.

Armed conflict has become a major obstacle to the advancement of the Belt and Road Initiative（BRI）for China–Myanmar. By conducting an in–depth analysis of the complex dynamics of armed conflicts before and after the political changes in Myanmar in February 2021 and accurately quantifying the spatiotemporal evolution of these conflicts,this study effectively reveals the pathways through which armed conflicts influence the construction and development of the China–Myanmar Economic Corridor （CMEC）.Using methods such as kernel density estimation,standard deviational ellipse analysis,and spatial autocorrelation,this study examines data on armed conflict events from 2015 to 2024 ,as well as data on China's foreign direct investment（FDI）in Myanmar and related trade and economic indicators under the BRI framework,to investigate the spatiotemporal patterns and distribution trends of armed conflicts in Myanmar and their impact on the CMEC.A structural equation model is introduced to construct the mediating pathway of＂armed conflict → policy uncertainty → investment withdrawal＂.It is found as follows. （1）Prior to the political changes in February 2021,Myanmar experienced relatively few armed conflicts,but entered a period of intense conflict thereafter.（2）Spatially,the conflicts exhibit a＂multi–center,high–density＂clustering pattern.The distribution of conflicts closely aligns with the development trajectory of the CMEC,with bilateral trade ports and key nodes emerging as hotspots for armed conflicts.（3）The mediating pathway of＂armed conflict → policy uncertainty → investment withdrawal＂is validated.The key constraint on China's FDI flows to Myanmar is uncertainty in the policy environment,rather than security risks arising solely from armed conflicts.

Using data from 41 cities in the Yangtze River Delta from 2006 to 2021,based on the coupling coordination theory framework,this study employs multidisciplinary methods to investigate the spatiotemporal interaction characteristics of coupling coordination between urbanization and technological innovation at the municipal level,and identifies key obstacle factors.It is found as follows.（1）From the perspective of temporal evolution,the overall coupling coordination degree between the two systems shows a steady increasing trend,while intra–regional imbalance exhibits significant expansion.（2）From the perspective of spatial characteristics,the Yangtze River Delta as a whole has reached a basic coupling coordination level,demonstrating a gradient spatial differentiation pattern along the northwest–southeast direction.（3）In terms of spatiotemporal interaction,the coupling coordination degree of the two systems shows strong integration during spatial evolution,with inter–city collaboration being more prominent than competition.（4）Regarding obstacle factors,three of the top five obstacle factors belong to the urbanization system,while the other two are associated with the technological innovation system.The conclusions provide theoretical guidance and empirical support for optimizing urban spatial layout,enhancing urbanization quality,improving the innovation–driven environment,and strengthening technological innovation capabilities in the Yangtze River Delta.

Understanding the influence of the spatial characteristics of urban commercial centers on carbon emissions from shopping travel is crucial for alleviating traffic congestion around these centers and reducing transportation–related carbon emissions.This study combines mobile phone signaling data with social survey data to analyze shopping travel behavior and its associated carbon emission characteristics in commercial centers of the six urban districts of Beijing.A method for characterizing the spatial features of commercial centers is established,and a mediating effect model is employed to decipher the association mechanism among the spatial characteristics of commercial centers,shopping travel behavior,and carbon emissions.It is found as follows. （1）Shopping travel modes vary significantly across different travel distance intervals,with the greatest variation in destination choice occurring within the <5km range.The total carbon emissions from shopping trips exhibit a spatial pattern of higher values in the city center and lower values in the periphery,while emission intensity shows a＂high–low–high＂pattern from the center to the periphery.（2）Seven spatial characteristic indicators significantly impact the carbon emission intensity of shopping travel,namely the hierarchical level of the commercial center,the internal gravitational index within a 1 km range,the proportion of residential land use,the location entropy of population density,the shortest distance to subway stations,the degree of functional mix,and the proportion of POIs classified as commercial facilities and other functions.The influence mechanisms of these spatial indicators on travel carbon emissions are relatively similar,they primarily affect carbon emission intensity by attracting long–distance travelers and influencing the choice of green travel modes.However,the mediating effect via the pathway of green travel proportion is stronger than that via the pathway of long–distance travel proportion.The findings offer valuable insights for guiding low–carbon shopping travel and optimizing commercial spatial planning towards low–carbon objectives.

Expressway traffic flow embodies multiple socio–economic attributes and directly reflects intercity connections. Analyzing passenger and freight flows separately is fundamental to understanding the complexity of expressway traffic and effectively identifying regional interconnected structures.Based on electronic toll collection（ETC）data for passenger and freight vehicles on expressways in Jiangsu Province,this study examines the spatial distribution characteristics of passenger and freight flows from both toll station and county perspectives.The geographical detector model is employed to analyze the influence of socio–economic factors on these traffic flows.It is found as follows.（1）Both passenger and freight flows exhibit significant uneven distribution patterns,with spatial concentration centers formed in Southern Jiangsu and along the Yangtze River area,while traffic flows in other regions are relatively dispersed.（2）The distribution of passenger flow shows an agglomeration structure centered around major cities in Southern Jiangsu,primarily consisting of short–distance trips.In contrast,freight flow displays a dumbbell–shaped pattern,being dense in the north and south but sparse in the middle,and relies on cross–river bridge nodes to form a typical＂corridor–style＂flow space.（3）GDP,value–added of the tertiary industry,and total fixed–asset investment have stronger explanatory power for passenger flow distribution;whereas resident population,GDP,and total retail sales of consumer goods demonstrate greater explanatory power for freight flow distribution.The findings of this study can support the planning of expressway infrastructure and the optimization of related facilities in Jiangsu Province,while also provide a scientific basis for formulating and adjusting differentiated,location–specific expressway toll pricing policies.

Tourism corridor construction serves as a vital strategy for optimizing the regional tourism spatial structure and promoting high–quality tourism development.The spatial heterogeneity of resource types within a region constitutes the foundational condition for establishing such corridors.Based on the identification of health and wellness tourism resource types,this study investigates the spatial distribution pattern of health and wellness tourism destinations in the Bohai Rim Region by applying the nearest proximity index and kernel density analysis to identify tourism source areas.The Geodetector method is employed to examine the influencing factors,and the minimum cumulative resistance（MCR）model is used to identify and construct health and wellness tourism corridors in the region.It is found as follows.（1）Health and wellness tourism destinations in the Bohai Rim Region overall exhibit a spatial distribution pattern of＂large–scale contiguous areas with small–scale clusters＂,with high–density zones located in Beijing,central Liaoning,and central–western Shandong.（2）In terms of the type distribution of health and wellness tourism destinations,geological–geomorphological and biological health and wellness tourism destinations show an agglomerated pattern,whereas water–body types are more evenly distributed,pastoral manor types and cultural health–preserving types display a random distribution pattern.（3）The spatial distribution of these destinations is influenced by multiple factors, with socioeconomic factors collectively exerting a stronger influence than natural factors.Among these,economic foundation, transportation conditions,and tourism investment level have particularly significant impacts.（4）A total of 10 source areas （e.g.,Beijing,Tianjin,Shenyang）, 4 regional–level corridors,and 5 local–level corridors are identified.These regional and local corridors interconnect to form an integrated health and wellness tourism corridor network across the Bohai Rim Region.

As a significant growth pole for their host cities,seaports influence the high–quality development of coastal port cities through multiple mechanisms,including industrial effects,opening–up effects,crowding–out effects,risk effects,and pollution effects,which can generate positive or negative impacts.Using panel data from 43 coastal port cities in China from 2003 to 2022,this study explores the impact of seaport scale on the high–quality development of coastal port cities by means of benchmark regression model and threshold effect analysis.It is found as follows.（1）The expansion of seaport scale significantly enhances the high–quality development of coastal port cities.However,threshold effects indicate an inverted U–shaped nonlinear relationship between the two.When the seaport scale exceeds a certain threshold value,its promoting effect on high–quality development of coastal port cities will slow down.（2）There is significant heterogeneity in the improving impact of seaport scale on the high–quality development of coastal port cities of different tiers:it is pronounced in fourth–and fifth–tier cities,while relatively weaker in second–and third–tier cities.Higher–tier port cities exhibit higher threshold values for seaport scale effects. Accordingly,coastal port cities should comprehensively evaluate the role of seaport scale in high–quality development according to their respective developmental stages,and rationally manage seaport scale to foster a mutually beneficial and synergistic port–city relationship.

Regional transport integration serves as a key driver of regional coordinated development,making its scientific assessment and characteristic analysis highly significant.Taking the Shandong Peninsula Urban Agglomeration as an example,this study constructs an evaluation index system based on multi–source data from three dimensions:facility coverage,connectedness,and service systematization.Adopting an integrated measurement perspective of＂quality–equilibrium＂,it analyzes the level of transport integration and its spatio–temporal evolution characteristics in the region from 2017 to 2023.It is found as follows.（1）From a dimensional perspective,facility coverage exhibited a＂center–strip–scatter＂distribution pattern with significantly narrowed inter–county disparities;connectedness developed rapidly with concentrated high–value areas;service systematization experienced fluctuations during the pandemic but overall demonstrated an accelerating and improving trend.（2）At the urban agglomeration scale,the level of transport integration increased markedly,transitioning from rapid to stable growth,and spatially formed a point–axis diffusion pattern centered around the Jinan–Qingdao dual core and supported by key metropolitan areas and transport corridors.However,significant spatial heterogeneity remained in the degree and type of integration at the prefectural–city level. （3）The evolution of transport integration showed improved equilibrium but notable quality divergence.In 2023,both Qingdao and Jinan were classified as＂high–quality synergistic＂types,while the vast majority of prefecture–level cities were in or entering a＂low–quality equilibrium＂predicament.Heze exhibited＂dual lag＂and required comprehensive improvements across multiple aspects.（4）In terms of correlational relationships,the driving effect of facility coverage on transport integration has weakened,while connectedness has gradually become the key factor.Addressing issues such as cross–scale transport linkage imbalances,urban–rural disparities in transport services,and the＂single–point collapse＂of high–level facility coverage in peripheral areas will be critical for future development.Finally,suggestions are proposed for optimizing intercity and urban–rural transport layouts,with the aim of providing reference for promoting regional transport integration in urban agglomerations and other areas.

The digitalization development of the tourism industry serves as a key benchmark for examining the depth of integration between the digital economy and the tourism industry,as well as a pioneering force for cultivating and enhancing new quality productive forces in tourism industry.Taking the explanation of the generation mechanism of tourism industry digitalization as its logical starting point,this study constructs a comprehensive evaluation index system for tourism industry digitalization. Using the entropy weight method,spatial dynamic kernel density estimation,and the spatial Durbin model,it investigates the spatiotemporal differentiation pattern and driving mechanisms of tourism industry digitalization levels across 30 Chinese provincial regions from 2006 to 2022 .It is found as follows.（1）The level of tourism industry digitalization in China has shown continuous growth,which can be divided into three evolutionary stages:the incubation period（2006–2011）,the rapid development period （2012–2019）,and the optimization and adjustment period（2020–2022）.The development exhibits a spatially declining pattern from east to west,with increasingly apparent regional disparities.（2）A significantly positive global spatial correlation is observed in tourism industry digitalization levels.However,as the＂digital divide＂in adjacent regions widens,this positive spatial correlation gradually weakens in dynamic estimations.（3）Economic development,technological innovation,government regulation,and consumer demand have positive promoting effects on the digitalization development of local tourism industry,while human capital has a negative inhibitory effect locally.Technological innovation,industrial structure,resident consumption,and openness demonstrate positive spillover effects on tourism industry digitalization in neighboring regions,whereas the spatial spillover effects of human capital and government regulation are not significant.

The digital transformation of the tourism industry has become an important driving force for regional economic growth and industrial upgrading,and has a profound impact on carbon emissions.This study constructs a five–dimensional evaluation system to assess tourism digitalization performance,analyzes its spatiotemporal evolution patterns,and explores the spatial spillover effects,effect boundaries,and nonlinear characteristics of its carbon emission reduction effect.It is found as follows.（1）From 2011 to 2022,the national level of tourism digitalization performance demonstrated steady improvement, peaking in 2019.The eastern region significantly outperformed the central and western regions,maintaining a stable growth trajectory.（2）Provincial tourism digitalization performance exhibited characteristics of＂club convergence＂,which was divided into initial areas,development areas and leading areas,but there was a state transfer at different stages.（3）Tourism digitalization performance exerted a significant negative indirect effect on carbon emissions through spatial spillovers.The emission reduction effect exhibited nonlinear distance sensitivity and clear stage–dependent heterogeneity,with the strongest mitigation effect observed at intermediate levels of digitalization and diminishing marginal effects at higher levels.

Tourism is a typical environment–dependent industry.Coordinating ecological protection with tourism economic development through environmental regulation has become a crucial driver for enhancing tourism economic resilience and promoting high–quality development of the tourism economy in the new stage.This study reveals the spatiotemporal evolution characteristics of environmental regulation and tourism economic resilience in 78 prefecture–level cities in the Yellow River Basin from 2008 to 2022 .It constructs a two–way fixed effect model and panel threshold regression models to investigate the spatial heterogeneity and threshold effects of environmental regulation in empowering tourism economic resilience.It is found as follows.（1）During the study period,the intensity of environmental regulation and the level of tourism economic resilience in the Yellow River Basin show an overall upward trend,spatially exhibiting a distribution pattern of＂lower in the upper reaches and higher in the middle and lower reaches＂.（2）Environmental regulation empowers tourism economic resilience in the Yellow River Basin by incentivizing tourism enterprise innovation and driving industrial restructuring.（3）The empowerment effect of environmental regulation exhibits heterogeneity based on resource endowments.Compared to non–resource–based cities,environmental regulation demonstrates a stronger marginal effect in empowering tourism economic resilience in resource–based cities.（4）The empowerment of tourism economic resilience by environmental regulation displays double–threshold effects for both tourism industrial agglomeration and tourism industrial scale.The former causes the empowering effect to first strengthen then weaken, while the latter causes it to first weaken then strengthen.

In the context of the new media, big data era, increasing online attention on ecotourism is of great significance for enhancing ecotourism attractiveness, promoting sustainable tourism development, and facilitating harmonious coexistence between humans, nature.Based on Baidu Index, this study employs methods including entropy–weighted TOPSIS, global spatial autocorrelation, traditional, spatial Markov chains, and panel spatial Durbin regression to explore the spatiotemporal evolution, influencing factors of provincial–level online attention on ecotourism in China from2011to2021.It is found as follows.（1）During the study period, online attention on ecotourism in China showed a fluctuating upward trend, presenting a spatial differentiation pattern of＂eastern regionscentral regionswestern regions＂with positive spatial autocorrelation. （2）Online attention on ecotourism in China generally exhibits significant path–dependent characteristics, making short–term leapfrog growth difficult to achieve.Provinces with high online attention exert positive spatial spillover effects on those with medium or higher attention, while generating negative spatial spillover effects on provinces with low attention.（3）Tourism agglomeration, transportation development can significantly enhance both local, neighboring provinces'online attention on ecotourism;resource endowment, industrial structure, and digital technology primarily contribute to improving local online attention, whereas consumer demand has a significant promoting effect on neighboring provinces'online attention.

Accurately understanding the residents'sense of gain in rural tourism destinations is a major research topic endowed by the times,which profoundly affects the future development direction of rural tourism and the implementation of the national rural revitalization strategy.Three typical rural tourism destinations of Miaoshan Village,Yanhu Village,and Qingzhen Village in Yangzhou City were selected as the research objects.By using methods such as questionnaire surveys,factor analysis,and multiple regression analysis,this study explored the residents'sense of gain in rural tourism destinations and analyzed its influencing factors.It is found as follows.（1）Residents in rural tourism destinations have a sense of gain in five aspects:economic life,ecological environment,social culture,social relationships,and value enhancement.The overall scores of residents'sense of gain are relatively high,especially those of Miaoshan Village residents,who have the highest scores.The questionnaire designed in this study demonstrates good reliability and validity,effectively revealing the subjective psychological feelings and states that residents have actually experienced during tourism development.（2）The residents'sense of gain is influenced by multiple factors,including individual factors such as gender,age,and health status,as well as external factors such as living conditions, ecological environment,social atmosphere,and government policies in the case sites.In the future,local residents'sense of gain can be enhanced by increasing their income,optimizing the ecological environment,strengthening government supervision,and improving the cultural ecology.The research findings expand the application of the sense of gain theory in the field of tourism research and provide theoretical and practical references for rural tourism destinations to effectively safeguard and improve people's livelihoods and promote high–quality rural tourism development.