With the rapid advancement of science and technology, artificial intelligence (AI) has become a significant force driving scientific development and social progress. In the field of geographical sciences, the application of AI technology is deepening, bringing revolutionary changes to the collection, analysis, and application of big data and spatio-temporal information, and demonstrating innovative and application potential in multiple aspects. This paper systematically reviews the development and application of AI in geographical sciences, providing a detailed introduction to the development trajectories of various AI fields such as machine learning, computer vision, natural language processing, planning systems, and large AI models, as well as their applications in geography. It discusses the problems and challenges of AI applications in geography and provides an outlook on the future development of interdisciplinary research between AI and geographical sciences.

Geographic environment has exerted profound effect on the origin and evolution of world civilizations. Chinese civilization budded and evolved on a vast and varied territory between Yellow and Yangtze rivers, and has been thus deeply affected by the local geographic conditions. But it has been hardly seen to explore the origin of Chinese civilization from the perspective of geography. On the basis of integrated scientific investigation in China's north-south transitional zone, geographic analysis of Neolithic culture distribution and interpretation of pre-Qin and Qin-Han ancient literature, the conclusions can be drawn as follows: (1) The early agriculture pattern of "Rice in the south and millet in the north" and the ancient astronomy formed before about 8000 years were the background for Chinese civilization. The geographic distribution of Neolithic Dadiwan, Yangshao, Majiayao and Longshan culture sites showed that the earliest civilization elements appeared in the upper reaches of West-Hanshui and Weihe rivers, with a spatial trend of spreading toward east. (2) The West Qinling Mts. region, located between the Tibetan Plateau and the Jialing River, especially its inner Chenghui and Xili basins, being characterized by superior natural conditions and resources, is closely related to the three major mysteries concerning the origin of Chinese civilization, i.e., the main areas of the ancient Di and Qiang ethnic groups, the location of ancient Kunlun Mts., and the site of Dayu water control. (3) The Qin ethnic group stepped onto the stage of history by assisting Dayu in water control, and in their history of multiple ups and downs, built the grand water control projects in ancient China, such as the Dujiangyan Irrigation Project, Zhengguo Canal, Lingqu Canal, etc., and pioneered the time of "Books with the same text" and "County system", forming the main line of the origin and early evolution of Chinese civilization. (4) The West Qinling areas are still basically a "blind zone" in archaeological and historical research. It is highly recommended to conduct systematic and in-depth archaeological and historical research in this region so as to realize the breakthrough in the exploration of the origin of Chinese civilization as soon as possible.

The statistical definition of migration in China is attached with its unique household registration (hukou) system, the migrants so defined are also called the floating population. A new analytical paradigm is needed to deal with the complexity of sub-flows in the floating population. The paper classifies five types of the floating population in census context by referring the UN migration criteria, and constructs a new paradigm for analyzing the floating population in China. As a particular provincial-level region (hereafter province) is concerned, the inflow and outflow people belong to different hukou groups balanced by their own counter-flow, namely, inflow vs back-inflow (both have no hukou of the province), outflow vs back-outflow (both have hukou of the province). With the clue of inter-census migration cohort, a sub-flows model is constructed to identify the inter-provincial migration based on the retention rate. The annual retention rate of the inter-provincial migration cohort from 2010 to 2020 is 88.7%. Based on the provincial retention rates, an all-increment table of population change by province in China is made by modelling simulation. The paper surfaces new characteristics of population growth and inter-provincial migration: (1) Provincial population changes are divided into five types, among which the inflow-leading increase type is mainly found in municipalities and the eastern coastal areas, while the fertility-leading increase type and the fertility-overriding increase type are mainly in the western provinces and the agricultural provinces in the middle, and the outflow-overriding decrease type and the outflow-leading decrease type in Shanxi, Inner Mongolia, Gansu and the northeastern provinces. (2) The mechanical growth of population can be divided into four types: rapid increase, equilibrium, outflow-overriding decrease and dual decrease (both hukou and non-hukou migration are negative). A "W"-shaped mechanical growth rate profile from northwest to southeast is found with the equilibrium belt standing in its middle. (3) Population floating is divided into three types. The counter-flows are highly-correlated: inflow rate and outflow rate are negatively correlated while the inflow-back-inflow and outflow-back-outflow are significant positively correlated. The analytical paradigm and model of floating population in China can be further extended to the study of "citizenship seeking migration" including international migration, and even further to identity migration including migrations with status changing such as enrollment, employment and marriage.

Thermokarst lakes, as prominent thermokarst landscapes in permafrost regions, significantly influence ecological vegetation, hydrological processes, and carbon cycling in the Arctic. However, the current understanding of the distribution characteristics and change mechanisms of Arctic thermokarst lakes remains limited. To address this gap, this study employs meta-analysis and mathematical statistical methods to investigate the distribution patterns and dynamics of thermokarst lakes. The results reveal a pronounced spatio-temporal heterogeneity in the distribution and changes of Arctic thermokarst lakes. These variations are closely associated with permafrost conditions, lithology, soil types, subterranean ice content, and soil temperatures. Most Arctic thermokarst lakes are located in continuous permafrost area, where subterranean ice content exceeds 10%, average soil temperatures are above -4°C, and there are specific vertical soil temperature gradients. The change trends of thermokarst lakes differ under various environmental conditions. Generally, the development of thermokarst lakes follows three stages: initial formation, mid-term expansion, and late-stage contraction, all influenced by hydrological and thermohydrological balances. As critical indicators of permafrost degradation and climatic environmental changes, thermokarst lakes profoundly impact carbon cycling, hydrological processes, and ecological environmental changes within the Arctic ecosystem.

Sustainable development is a significant scientific issue of global concern. Geography, as a comprehensive discipline focusing on the coupled relationship between human activities and the natural environment, provides systematic research and solutions for achieving the United Nations Sustainable Development Goals (SDGs). However, there is currently a lack of comprehensive reviews. This paper summarizes the theoretical framework and research progress of Geography supporting the SDGs and explores its future key research areas. This article indicates that: (1) Geography, in conducting integrated research on human-nature systems and serving regional and global sustainable development processes, has innovatively proposed and developed theoretical frameworks such as social-ecological systems, pattern-process-service-sustainability, metacoupling, and Classification-Coordination-Collaboration. These research frameworks include elements of human-environment system interconnections, process coupling, spatial coupling, and systematic regulation oriented towards SDGs, forming a comprehensive theoretical framework supporting sustainable development research in Geography, also referred to as "sustainable geography theoretical framework". (2) Geography has made positive progress in supporting the United Nations SDGs research in areas such as multi-source data acquisition, localization of indicator systems and multi-scale progress assessment, analysis of inter-target linkage mechanisms, and SDG achievement pathways. Geography provides important theoretical and methodological support for SDG research. (3) Geography and sustainable development-related research mainly focus on climate-ecological crisis response, sustainable utilization of food-energy-water resources, regional development and planning, human well-being and social governance, and the construction of SDG assessment indicators and databases. (4) In future research, there is a need to innovate and develop sub-disciplines of Sustainable Geography, optimize the construction of SDGs indicator systems, develop SDGs assessment and decision-making models, strengthen artificial intelligence geography, deepen research on human-nature system coupling, and promote regional and global sustainable development in the process of advancing innovation in the discipline of Geography.

In the wake of unprecedented global urbanization, an alarming trend of shrinking cities has emerged worldwide, presenting a profound challenge to conventional urban-regional planning approaches, primarily centered on growth scenarios. This trend has also bestowed upon Chinese geographers and urban and rural planning practitioners a new mandate in this evolving era. One region that exemplifies this predicament is Northeast China, experiencing the most severe population decline and the highest concentration of shrinking cities in the country. How to rationally recognize and properly deal with this phenomenon (or problem) of regional urban shrinkage has become the core scientific issue to be focused on in the comprehensive implementation of regional coordinated high-quality development, Chinese-style modernization of common prosperity for all and all-round revitalization of Northeast China in the new era. This paper delves into a logical speculation and practical discussion to understand the underlying causes and implications of regional urban shrinkage in Northeast China: Initially, it examines the original concept of regional shrinking cities within the context of urbanization, critically evaluating the rationality of such shrinkage in light of the broader national strategic objectives. Subsequently, the generation logic of regional urban shrinkage in Northeast China is explored through three key factors: the "pull" effect of regional spatial correlation, the "extrusion" effect of urban shrinkage at a regional level, and the "catalyst" effect of transportation and communication technology development in reducing the cost of factor flow. Moreover, this study draws on the market-led shrewd contraction theory and the government-led vitality regeneration governance logic. It takes into account the governance objectives of achieving "regional coordinated high-quality development, people-oriented new urbanization, efficiency and fairness, and comprehensive security concept". With a systematic perspective that considers scale heterogeneity, development context relevance, factor differentiation, and multidisciplinary integration, the paper outlines a governance model and strategy for addressing regional urban shrinkage in Northeast China. By emphasizing the alignment of governance approaches with national development goals, this research underscores that regional urban shrinkage goes beyond a mere outcome of market-driven population migration. The presence of underlying "market failures" severely impacts the vision of the five major securities (national defense, food, ecology, energy, industry) advocated by Chinese President Xi Jinping for the Northeast China region. The findings of this study offer valuable insights to guide Northeast China's pursuit of high-quality development and contribute to the region's revitalization efforts.

Understanding the spatial segregation and social integration of immigrants and local residents is highly significant in summarizing the historical experience of Chinese modernization. Additionally, it serves as a crucial foundation for promoting a people-centered new urbanization, fostering spatial fairness and justice, and achieving common prosperity. However, current research on residential spatial differentiation primarily focuses on analyzing the spatial distribution characteristics, differentiation patterns, and indicator calculations of various groups, and has limited investigation into social integration issues. This paper explores this social and spatial differentiation phenomenon in Hangzhou, usingthe initial residence registration information and mobile phone signalling data to analyze the characteristics and mechanisms of living space differentiation. Additionally, this article combines a small sample of questionnaires with a bottom-up exploration of social integration under spatial differentiation. From the research, four key findings emerge: (1) The majority of residents in the city are migrants. There is a clear difference in the distribution of living spaces between immigrants and local residents, with local residents mainly concentrated in the old city area, while the distribution of immigrants is more scattered. Compared to the typical "immigrant city" in the United States, Hangzhou has less spatial differentiation between immigrants and local residents. (2) The residential space differentiation between immigrants and local residents in the city presents a combination pattern of " multi-nuclei, sectoral, concentric zone". However, unlike the centrifugal diffusion trend of the local people in the United States, the local residents in Hangzhou tends to gather towards the center, while immigrants choose centrifugal diffusion and peripheral aggregation. (3) The spatial differentiation of living spaces is mainly caused by the competition for spatial resources between immigrants and local residents under urban expansion. However, behind the phenomenon of "local residents live in the old city", immigrants are no longer a disadvantaged group. Some intelligent new immigrants actively choose to live in high-quality peripheral areas. (4) The integration between immigrants and local residents is satisfactory, but it is more cautious about the issue of intermarriage. Additionally, residential spatial segregation has not had a negative impact on social integration. There are four main reasons for this. To begin with, the majority of the population consists of immigrants, and urban residents no longer pay attention to identity labels in their daily interactions. Besides, there is relatively little cultural difference within ethnic groups, and the assimilation between immigrants and local residents is increasing. And then, interactions between individuals in workplace and third place can alleviate the negative impacts caused by residential spatial segregation. Finally, China's coordinated development and people-oriented social system can also create favorable conditions for social integration.

The new wave of technological revolution driven by artificial intelligence (AI) has become a competitive advantage pursued by countries worldwide, and it is also expected to bring about significant innovations in urban geography. This paper reviews the development trajectory of AI and proposes a conceptual model of the urban human-environment system oriented towards AI. It systematically outlines and summarizes the involvement of AI in urban geography research from four perspectives: data representation, scenario applications, spatial transformations, and urban development. Additionally, it identifies the challenges, contexts, exploration paths, and future prospects faced in the intelligent transformation of academic disciplines. The study finds that: (1) The interactions between elements within the urban human-environment system oriented towards AI are becoming increasingly complex. (2) The trend of AI becoming a primary tool in urban geography research is increasingly evident, offering high efficiency, low cost, and strong learning capabilities in data processing. This has significant implications for spatial perception and intelligent decision-making. AI has sparked spatial transformations, not only creating complex virtual spaces but also reconstructing social spaces. Additionally, AI supports the development of smart cities and the establishment of cutting-edge urban application platforms. (3) Urban geography research in the context of intelligent transformation faces challenges related to data and technology, as well as the broader contexts of global and local changes, technological ethics, and the development of humanistic values. Future development paths could explore overcoming technical barriers, focusing on urban spatial construction and governance, and emphasizing the research on the effects of multiple intelligence shifts. The discipline urgently needs comprehensive transformation and upgrading.

The physical geography and hydroclimatic conditions in the arid region of in northwestern China leading to diverse flood-generation mechanisms. Under the influence of global and regional climate change, the spatiotemporal variation of floods and flood-generation mechanism in this region is still unclear and restricts flood prevention and mitigation and the implementation of the Belt and Road Initiative in the major regions. Based on the series of the annual maximum flood peak discharge in 58 river basins in the study area from 1961 to 2017, we analyzed and revealed the mechanisms, spatial distribution and interannual variation characteristics of flood in the basin in the past 60 years based on statistical tests and machine learning approaches. The results show that the frequency of extreme floods and the annual maximum flood peak discharge magnitude are increasing, with the maximum increase in the frequency of extreme floods at about 0.84 times/10 years, and the maximum increase in the annual maximum flood peak discharge magnitude at about 29%/10 years compared with the multi-year average, and the largest increase is mainly observed in the eastern Tianshan Mountains and the Qilian Mountains. There are three main flood generation mechanisms, i.e., rain (R), snow (S) and mix (M), the frequency of R and M floods increased significantly, while the frequency of S floods decreased. The contribution of flood mechanisms transformation to the increase of annual maximum flood peak discharge magnitude can reach up to 38%, which is significantly higher than the contribution of a single hydrometeorological factor such as precipitation. The results of this study emphasize the importance of attributing and predicting the changes of flood characteristics in geographically complex region from the perspective of flood mechanisms. Engineering hydrological design in the changing environment also needs to consider the influence of the heterogeneity of flood samples caused by different flood mechanisms on the flood frequency analysis, so as to provide scientific support for flood risk management and response in the basin.

Cultural ecosystem services (CESs), as the non-material benefits and well-being provided by ecosystems to humans, can bridge and interpret the complex relations between nature and human society and can be organised around "spatial needs-spatial planning-spatial production" to achieve sustainable development and human well-being through top-down and bottom-up interventions. Accurately grasping the practical path of CESs involved in the spatial reconstruction of the human-land relationship is not only a scientific problem that needs to be solved to realise the effective enhancement of human well-being but also an important theoretical basis for realising the harmonious coexistence of human beings and nature. For this reason, based on the deconstruction model of "theoretical analysis-associated mechanism-practical application-development path", this study explores the inherent logical system and feasibility of CESs guiding the spatial reconstruction of the human-land relationship and discusses the value assessment-value trade-offs in the process. Based on the evolution of research on CESs in the "flow" era, this study constructs a framework for CESs to guide the spatial reconstruction of the human-land relationship and use it as a theoretical basis for exploring its application in guiding the spatial reconstruction of the ecological-production-living spaces. This framework is then used to develop a program of data collection, methods system, and practical application to realise the transformation of research on CESs from theory to practice and provide new perspectives for the spatial reconstruction of the human-land relationship in the new period.

Metropolitan areas are pivotal in driving national economic growth, advancing harmonious regional development, and participating in competitive international collaboration. Consistent with international consensus, China's planning policies define metropolitan areas as a one-hour commuting circle. Simultaneously, these planning protocols particularly highlight the importance of industrial specialization and intercity collaboration, with the objective of developing modern urban agglomerations characterized by a well-defined spatial structure, complementary urban functions, and an integrated industrial division. This study uses the 2017 intercity input-output table analysis to examine patterns of industrial division and functional synergy in Chinese metropolitan areas. The findings are as follows: (1) From the perspective of industrial division and functional synergy, Chinese metropolitan areas can be categorized into four distinct spatial structures: the single-center dispersed structure, the core-periphery structure, the peripheral expansion structure, and the multi-center network structure representing an advanced phase in the spatial evolution of metropolitan areas. (2) The spatial network of industrial chains in Chinese metropolitan areas is predominantly governed by the flow of processing and manufacturing links. There is a substantial gap in the processing and manufacturing links between the Beijing and Tianjin metropolitan areas compared to similar types of metropolitan areas, and there is considerable potential for enhancing the productive service links in the Guangzhou, Chengdu, and Changsha-Zhuzhou-Xiangtan metropolitan areas. (3) The essence of transforming Chinese metropolitan areas into functionally complementary industrial cooperation areas is to further refine the spatial network of the industrial chains, strengthen functional synergy between cities, and encourage spatial integration in the metropolitan areas. (4) Single-center dispersed structure metropolitan areas, including Shijiazhuang, Chengdu, Wuhan, and Changsha-Zhuzhou-Xiangtan should initially transition toward a core-periphery structure. In contrast, peripheral expansion structure metropolitan areas such as Nanjing and Hangzhou metropolitan areas, and core-periphery structure urban agglomerations such as Tianjin, Shenzhen, Guangzhou and Beijing metropolitan areas should progress toward a multi-center network structure.

Human settlements geography leverages the comprehensive strengths of geography that are recognized both in China and internationally. Therefore, human settlements geography represents the mature evolution of the discipline of geography through interdisciplinary integration. Human settlements geography directly responds to the public's aspirations for superior human settlements and fulfills critical national priorities. Compared to urban, economic, and behavioral geography, human settlements geography's theoretical frameworks, practical applications, and integrative development remain under-explored. To advance this field, delineating its theoretical underpinnings across the following five key dimensions is essential: foundational theories, theoretical frameworks, geographical scales, research paradigms, and thematic narratives. Central to the discipline is the exploration of human-settlements relationships, which are approached through both "three dimensions" and "three elements" perspectives. By capitalizing on the unique interplay of "three states", the discipline advocates for an open integration model, proposing a novel pathway for human settlements geography, which is poised to navigate three pivotal avenues: engaging with new scientific inquiries, adopting innovative data and methodologies, and embracing progressive development concepts. This strategy aims to transition human settlements geography from a specialized intersection of interdisciplinary study to a more encompassing and influential field, marking its significance as a major branch of geography and a substantial discipline. The synthesis of theoretical depth, practical relevance, collaborative integration, and strategic development in human settlements geography seeks to catalyze cross-disciplinary synergy and paradigmatic shifts. It aims to meet residents' diverse needs, support evidence-based national development policies, and enhance China's role and authority in shaping global human settlement policies, highlighting the country's dedication to fostering sustainable and habitable human settlements.

Waterlogging, water pollution, black and odorous water, and riverine and lacustrine ecological degradation are the main water problems faced by China's cities, which have become one of the key obstacles to the green development of the Yangtze River Economic Belt. In-depth understanding of the evolution of the water system in the process of urban development and the mutual influences between the elements, clarifying the formation mechanism of the urban water problems, and constructing an integrated treatment technology system of the urban water system are the fundamental ways to solve the complex urban water problems. Aiming at the abovementioned urban water problems, this paper summarizes the concepts and characteristics of the urban water system from a systematic perspective of the multiple processes coupled with each other in the water cycle and proposes a key technology system for a systematic solution to the urban water problems, which are demonstrated with the treatment examples of the urban agglomeration in the Yangtze River Economic Belt in 2016-2022. The demonstrative applications show that the pumping stations and sponge measures can effectively reduce the severely inundated area of Wuhan city in extreme rainfall events and eliminate the waterlogging points in the Chengdu-Chongqing urban agglomeration; the proposed microbial-botanical-hydrodynamic integrated synergistic regulation and control technology can significantly enhance the removal efficiency of the nitrogen and phosphorus level in typical black and odorous water bodies of Wuhan city; the proposed source control-path dissipation-terminal regulation comprehensive ecological treatment system for urban rivers and lakes decreased the water eutrophication level significantly in the Lianghe River basin of Chongqing city; with the implementation of the abovementioned treatment technology system, the green development index of the urban agglomeration in the middle reaches of the Yangtze River increased by an average of 4.8%. This study provides theoretical and technical support of urban water system for the major strategies of the green development of the Yangtze River Economic Belt.

In recent years, the tidal limit of the lower reaches of the Yangtze River has shifted upward, the construction of water-related projects has increased, and the intensity of river bank failure disasters has been on the rise. Previous researchers have found the existence of strong vertical-axial backflow in the bank failure section through field surveys, physical modeling, and numerical simulation, and that there is a strong vertical-axial backflow in the bank failure section; however, the hydrodynamic zone characteristics of the bank-channel interactive zone and the microgeomorphic adaptation mechanism have not been explored. Based on the Doppler acoustic current profiler (ADCP) in the frequent bank failure sections, multibeam system, combined with previous high-resolution underwater topographic data, it is found that the flow velocity and flow direction in the interactive zone of the bank-channel have significant zone characteristics, and according to the backflow strength \alpha, i.e., the longitudinal flow velocity u and its ratio to the transverse flow velocity v, |u/v|, the zone from the channel to the bank is identified into the main stream zone (α<1, u>0), the mixing zone (α≥1), and the near-bank backflow zone (α<1, u<0). The mainstream zone develops large and medium-sized dunes, scour grooves, and scour pits, the mixing zone develops accumulators and small dunes, and the backflow zone develops spines and small dunes; the main stream direct flushing in the mainstream zone and the backflow lateral erosion in the backflow zone are the main driving forces for the development of scouring microgeomorphic type and the development of bank failures, and the high flow velocity gradient in the mixing zone near the main stream zone generates scouring geomorphology, and the remaining part of mixing zone remains a low flow-velocity to make the sediment fall and silt to form the accumulation body. The findings can provide important references for the disaster remediation and near-shore engineering design and monitoring.

As an important aspect of the notion that "spatial is special", spatial heterogeneity has been a central topic of geospatial analytics. It is also closely related to the methodological tradition of geography and replicability of geographic research. The emergence of fine-grained big geospatial data and the development of Geospatial Artificial Intelligence (GeoAI) have brought new opportunities and challenges to spatial heterogeneity modeling. Spatial heterogeneity may refer to (1) values of geographic variables; and (2) associations, or the generation process of geographic variables, which correspond to data and process heterogeneity, respectively. Moreover, the specification of spatial heterogeneity may be categorized as continuous and discrete. Based on the dichotomies above, we summarize the main scenarios of spatial heterogeneity modeling and review corresponding methods: (1) homogeneity-based regionalization; (2) local spatial regression; and (3) spatial regime regression. In particular, we discuss approaches to delineate spatial regimes in an endogenous manner. We also review related methodological advances in GeoAI, where the principle of spatial heterogeneity is reflected in the design of neural network models. Finally, we point out several potential directions for future research. As emerging directions in spatial heterogeneity modeling, spatial regime regression and GeoAI methods need more attention from researchers.

Quantifying the contributions of climate change and human activities to changes in vegetation is important in terms of regional ecological protection and future research on the Loess Plateau. However, in the case of areas with naturally regenerated vegetation, where there are no clearly visible indications of human activities and ecological processes are not significantly disturbed, not all relevant forms of human activity can be remotely detected. Therefore, modelling the intensity of human modification of natural ecosystems could provide a pathway for a better understanding of the questions concerning where and how much vegetation change occurs in areas with evidence of human activities or land-use change. In this study, we utilized the Normalized Difference Vegetation Index (NDVI) between 2000 and 2022 and land-use datasets (a spatial resolution of 30 m) collected in 2000, 2010, and 2020 to distinguish areas with strong and weak anthropogenic interference on the Loess Plateau. In those areas with clearly visible human activity, we explored the relative contributions of climate change and human activity to vegetation dynamics. Taking the lagged duration variation between climate factors and vegetation growth into consideration, climatic factors influencing vegetation decadal dynamics were identified in the southern part of the Loess Plateau. The results are as follows: (1) With the implementation of the Three-North Shelter Forest Program, Grain-to-Green Program, and other ecological projects, the rate of vegetation restoration on the Loess Plateau slowed by 3.4%/10a during the period 2012-2022; however, the overall trend was an increase in vegetation. (2) In sensitive areas of vegetation restoration in the Loess Plateau, human activities with a relative contribution exceeding 65% were the dominant factors of vegetation variability in the areas with both intensive and weak human interference, which is approximately 2.0-fold higher than the contribution of climate change. (3) The interdecadal anomalies of precipitation in June were effective indicators of opposite west-east vegetation anomalies in the southern part of the Loess Plateau. Specifically, when precipitation increased by 10 mm in June, this southeastern part witnessed a decrease of 1.4% in NDVI in contrast with a 1.7% increase in the southwestern part. By coupling temporal and spatial information we have clarified the spatial distribution of trends and decadal variations in NDVI and its influencing factors. These observations from the Loess Plateau provide useful insights to help understand the relationship between vegetation change, climate change, and human activities on vegetation restoration globally or in other regions of China.

With the continuous development and application of information technology, online shopping has emerged as a significant force, bringing immense economic potential. The convenience offered by online shopping, such as the ability to shop anytime, anywhere, and access to a vast array of products, is reshaping consumer habits and expectations. Despite the continued growth of online shopping, the role of physical stores remains significant. Physical stores offer tangible benefits that online shopping often struggle to replicate. These include the ability to see, touch, and try products before purchasing, as well as the immediacy of taking items home without having to wait for shipping. The relationship between online shopping and physical shopping remains a hot topic in academia. Although there have been relevant studies in China, most of them are based on the survey data from individual communities or cities, lacking large-scale and long-term systematic research from a national perspective. Furthermore, there is a lack of heterogeneous comparisons between different regions. Drawing on large-scale time utilization survey data of 29 provincial-level regions in 2017 and 2021, this paper analyzes the spatiotemporal evolution of online shopping behavior and physical shopping behavior of China's residents, and examines the influence of online shopping on residents' physical shopping behavior. The results show that: (1) There are significant spatial differences in the characteristics of residents' online and physical shopping behaviors. The standard deviation ellipse of online shopping and physical shopping behavior shows an obvious "northeast-southwest" distribution pattern, with the gravity center gradually shifting northward during the study period. (2) In 2017, online shopping had a significant complementary effect on physical shopping. But in 2021, both complementary and neutral effects coexisted. (3) There are significant differences in the impact of online shopping on physical shopping between urban and rural residents, and residents among different regions. There is also a coexistence of complementary effects and neutral effects. This paper clarifies the relationship between online shopping and physical shopping in China, provides an in-depth understanding of the changes of urban space in the internet era, and helps promote the spatial transformation of physical commerce.

Energy is pivotal for the operation of the global economy. It is thus critical for us to elucidate the characteristics and dynamics of the global energy industry during the process of transformation in the 21st century. Using the input-output data by the United Nations from 1990 to 2022, this study gauges the added value of the energy industry of 184 countries and regions in the world and seeks to show its distribution, asymmetric dependency, and key drivers across different industrial stages. (1) We find the value of the global energy industry grew slowly at first, then rapidly, followed by slowly again, and ended up a recession during the period covered in our analysis. This temporal trend corresponds to the cycle of the global economy for each decade. (2) Energy small countries exhibited growing dependence on energy superpowers, and likewise interdependence among energy superpowers also increased, especially in the stage of prospecting and exploitation. This signifies that our world has gradually become an interconnected community of economic coexistence and co-prosperity. (3)We saw the rise of China, Russia, Saudi Arabia, and other countries as they gradually became the chain leader and hubs of the global energy industrial network. Their rise also has made the boundaries of cooperatives blur and the hierarchy of power iterate. The post Cold-War world is evolving from a unipolar one dominated by the United States into a multipolar one, and the global energy order is experiencing a profound reorganization. (4) Interdependence on the energy industry is a result of differences in attributes among countries (regions), their relative advantages, and their multidimensional proximity. Trade gaps, disparities among coporates, shared languages, colonial histories, and global organizations strengthened their interdependence, whereas geographic distance acted as a significant negative factor. Both economic gaps and institutional environments did not make any significant differences in shaping energy dependence. The role of these factors vary across different industrial stages over time.

The nonlinear relationship and spatially heterogeneous relationship between environmental factors and criminal activities are the main reasons for both the theoretical and empirical divergence, but the relevant analysis remains fragmented and faces limitations such as linear relationship hypothesis, collinearity problems and omitted variable bias. This study uses Gradient Boosting Decision Tree (GBDT) algorithm and Shapley Additive Explanation (SHAP) interpreter in machine learning to systematically reveal the nonlinear and spatially heterogeneous relationships between 48 built and social environmental factors on violent crime in Beijing. Our research has revealed the existence of seven distinct types of nonlinear relationships between environmental factors and violent crime, each exhibiting unique trends in the direction of influence and marginal effects. Furthermore, we have found that the association between environmental factors and violent crime exhibits varying degrees of spatial heterogeneity. By utilizing K-means clustering analysis, the entire area can be segmented into six distinct regions, each characterized by different critical criminogenic factors. These findings suggest that the applicability of crime geography theories, such as the classification of crime generators, attractors, and inhibitors based on crime pattern theory, the validity of street eye theory and defensible space theory, and the impact of social attributes as proposed by social disorganization theory, may depend on the value range of environmental factors and differ across locations. In light of these findings, it is recommended that crime prevention strategies shift from universal to targeted approaches, wherein public resources are allocated to specific value ranges of environmental variables and prioritized regions.

This research aims to decipher the patterns and mechanisms of surface water recharge during the rainy season in plateau mountain-lake region. Based on the stable hydrogen and oxygen isotope data of multiple water bodies collected in the Erhai Lake Basin from July to August 2022, and using Bayesian mixing models and remote sensing technology, the study quantifies the sources and proportions of surface water recharge and explores water body transformation mechanisms. The results indicate that: (1) Precipitation in the Erhai Lake Basin during the rainy season is influenced by evaporation and monsoon climate, resulting in hydrogen and oxygen isotope values that are more negative compared to groundwater and surface water. Groundwater δD and δ¹⁸O indicate that in areas of high altitude and high water richness, the hydrogen and oxygen isotope values of groundwater are more negative, while d-excess values reveal evaporation differences among different aquifer groups. The hydrogen and oxygen isotope values of surface water gradually become more positive as the water flows from inflowing rivers through Erhai Lake to outflowing rivers, with the isotope values on the eastern slope and southern region of Cangshan Mountain being more positive compared to the northern region. For the eastern slope of Cangshan Mountain, the midstream area shows the most significant evaporation. (2) The spatial pattern of surface water recharge sources between Cangshan Mountain and Erhai Lake shows that surface water contributes the most to its downstream mixed water bodies. The contribution rate of precipitation and surface water in each river section between Cangshan Mountain and Erhai Lake is shown as follows: the contribution rate of precipitation and surface water in the river section above the mountain outlets is smaller than that in the river section below the mountain outlets; while the contribution rate of groundwater is vice versa. (3) Along the river flow direction, the spatial pattern of surface water recharge sources is jointly influenced by topography, geological conditions, surface cover, and water vapor characteristics. Areas with good vegetation cover, high surface temperatures, and high actual evapotranspiration receive more precipitation recharge; areas with relatively poor vegetation cover receive more surface water recharge and the contribution rate of groundwater to surface water depends on the type of aquifer groups.

In the post-Cold War era, the global landscape has evolved towards multipolarity, and the geostrategic behavior of states has gradually shifted away from the traditional binary model of "balancing-bandwagoning" to a more flexible and diverse one called hedging strategy, and it is the current strategy that the countries surrounding the South China Sea have adopted in response to China's actions. By introducing the hedging theory of international relations, we seek to understand, from a geopolitical perspective, how these geopolitical entities employ hedging strategies. First, an analytical framework is constructed to theorize hedging strategy, consisting of driving factors, mediating factors, and strategic choices. Second, the hedging intensity is assessed quantitatively via the geopolitical risk and the relative power index of the countries surrounding the South China Sea. The evolution of it is presented through a bivariate visualization method, and the various hedging patterns are classified by utilizing the quadrant diagram. Finally, both the theoretical framework and quantitative outcomes are validated by analyzing the strategic practices of the study region. We conclude that: (1) the geopolitical risks of the countries surrounding the South China Sea have been "tending to be stable but difficult to achieve" since the beginning of the 21st century, with significant heterogeneity among these countries. The evolution of state relative power exhibits a fluctuating trend, closely linked to U.S. intervention. (2) The hedging intensity of the countries surrounding the South China Sea results from the interaction between geopolitical risk and state relative power, leading to three distinct forms, represented by Vietnam and the Philippines, Indonesia, and Malaysia and Brunei, respectively. (3) The intrinsic logic of the hedging strategies of these countries can be seen as a strategic cognition shaped by their threat awareness and national strategic orientation. This cognition then orientates diverse hedging approaches that guide these countries in applying either cooperative or competitive tools. By understanding the intensity and nature of the strategic hedging by the countries surrounding the South China Sea, we aim to provide a unique vantage point for China's regional governance in the South China Sea.

This paper reviews the social background of "pole-axis system" theory and the "T-shaped" land development and economic layout framework in China, especially the objective conditions that the country cannot implement another "strategic shift" of national development priorities, and the academic contribution of economic geography serving to national strategic development, as well as some concepts, ideas, knowledge and methods to achieve this goal. According to the significance and the scholars' comments of this theory and model in national practice, this paper describes the growth course and academic responsibility of an economic geographer to inspire the academic community to uphold the rigorous attitude of seeking truth from facts to make greater contributions to the construction of socialism with Chinese characteristics in the new era.

Alluvial fans are common fan-shaped depositional landforms that develop at the outlets of mountain rivers or gullies. Mature and stable alluvial fans are important areas for both human habitation and production in mountainous regions, but they also pose potential hazards associated with flash floods and debris flows. Research on alluvial fans enhances our understanding of regional environmental dynamics and geomorphic evolution, as well as contributes to the mitigation of flood and debris-flow hazards. Therefore, it holds significant scientific value and practical importance. Although considerable research has been conducted on alluvial fans, both domestically and internationally, in recent decades, much of it has focused on geomorphology (morphometry), sedimentary history and characteristics, and historical environmental reconstruction (or inversion). Investigations into the mechanisms of fan development and their geomorphic effects remain relatively underexplored. This review systematically summarizes the key advancements in the research on the dynamic processes, mechanisms, and morphodynamics of alluvial fan development. We first provide an overview of current technical approaches applied in the study of alluvial fans, including field investigations and model experiments. Then, we summarize four critical aspects of fan dynamics processes and development mechanisms: primary and secondary processes; mechanisms of flow channel avulsion; interactions between tributary and main rivers; and the impact of alluvial fan development on sediment production, transport, and geomorphic processes. Finally, we discuss several areas that require further attention in future research. Currently, field observations and monitoring of the dynamic processes of alluvial fan development are inadequate. As an essential complement to post-event field surveys and experimental model research, there is an urgent need to enhance field observations in order to expand and deepen our understanding of alluvial fan development mechanisms. This will promote scientific insights into sediment dynamics and geomorphic processes within regional river systems.

Soil erosion is influenced by various factors, such as land use and climate change. The Wangmaogou watershed, as a typical area for soil and water conservation in the hilly and gully regions of the Loess Plateau, has implemented a series of measures since the 20th century, including the Grain for Green Project. This study evaluated the spatial and temporal evolution characteristics of soil erosion intensity in the watershed in the years of 2010, 2015, and 2020 using the CSLE model. It also studied the situation of land use/cover change (LUCC) to analyze the spatial distribution patterns of land use and the responses of soil erosion in different time periods, thereby verifying the effectiveness of the soil and water conservation measures. The results revealed that from 2010 to 2020, the annual average soil erosion modulus in the study area decreased by 0.11 t hm^-2 a^-1, indicating a slight improvement in overall soil erosion conditions. However, the proportion of slight erosion decreased by 5.56%, while severe erosion increased by 4.02%, with the higher erosion zone mainly distributed in the northern, central, and northwestern parts of the watershed. Compared to the year 2010, soil erosion conditions in the watershed were greatly relieved in 2015 due to the decrease of rainfall and restoration of vegetation, but rebounded in 2020 resulting from the extreme rainfall events and declining vegetation cover quality. From 2010 to 2020, there were significant conversions between grassland and farmland in the northern and northeastern parts of the watershed. The decline in grassland quality resulted in a higher average soil erosion modulus compared to farmland, at 13.69 t hm^-2 a^-1 and 12.99 t hm^-2 a^-1, respectively. This study would contribute to figuring out the relationship between soil erosion in typical small watersheds of the Loess Plateau, extreme climatic events, and land use changes, providing scientific data support for future efforts to improve soil and water conservation benefits and mitigate soil erosion risks.

Yangtze River Delta, as a pivotal region where the coastal axis and the riverside axis intersect, is characterized by complexity, diversity and typicality of the spatial structural evolution. Therefore, understanding its evolutionary pattern and constructing a theoretical model has important theoretical significance and application value. According to the core-periphery structure theory, the Yangtze River Delta can be divided into a core area that is centered on the Taihu Lake Basin and remaining periphery areas. However, due to its location at the junction of the river and the sea, a gateway area has emerged within the periphery area, thus forming a spatial structure that is composed of the core area and the gateway area. In the early period, the core area was centered on Suzhou, and a five-tier central place structure became well established since ancient times. However, the gateway area kept evolving and underwent three main changes: in ancient times, the gateway area was centered on Yangzhou, forming the canal gateway cluster; in the modern age, the gateway area became centered on Shanghai, forming an offshore gateway cluster; and in the contemporary era, the gateway area became centered on Ningbo, forming an oceanic gateway cluster. Their corresponding navigation capacities were 500 t, 10,000 t, and 200,000 t, respectively. Therefore, in addition to the existing central place theory and seaport spatial structure theory, the spatial structure evolution of the Yangtze River Delta presents a new evolutionary model: namely, the fusion evolutionary model of central places and port gateways. According to this model, in the early period, it was an endogenous evolution of the core area's spatial structure, which was in line with Christaller's hexagonal structure; while in modern times, the evolution of the spatial structure of the Yangtze River Delta was no longer dominated by central places, but rather, it became dominated by the gateway areas, making the k = 3 market principle turn into the k = 4 transportation principle. In this way, the Yangtze River Delta provides a globally exemplary empirical case for validating the process test of the central place theory, analyzing functional attributes of urban centrality and gateway, and refining the relevant theoretical model.

Evolutionary economic geography has proposed the principle of relatedness. According to the degree of relatedness, the regional industrial development paths can be divided into related diversification and unrelated diversification. However, the current studies neglect the symmetrical nature of relatedness. According to the degree of symmetry between two related industries, related diversification can be divided into symmetrically related diversification and asymmetrically related diversification. Using Chinese industrial enterprises database from 1998 to 2012, this paper investigates symmetrically related diversification and asymmetrically related diversification's relationship with regional economic complexity. The result shows that: (1) The regional industrial development paths are dominated by related diversification rather than unrelated diversification. The number of related new industries in China accounted for about 60%-70% of the total number of new industries from 1998 to 2012. In terms of related diversification, positively asymmetrically related diversification takes the largest part, while symmetrically related diversification takes the second largest part. The number of positively asymmetrically related new industries accounts for about 60% of the total number of related new industries, and the number of symmetrically related new industries accounts for about 35% of the total number of related new industries over the time period from 1998 to 2012. (2) Industries with higher technological complexity show stronger symmetry in their related diversification, while industries with lower technological complexity show stronger positive asymmetry in their related diversification. (3) Regions with higher economic complexity usually have more unrelated diversification and its related diversification shows stronger symmetry. Regions with lower economic complexity usually have more related diversification and this related diversification shows stronger positive asymmetry. This study can contribute to the development of the principle of relatedness in evolutionary economic geography and provide suggestions for policymakers in terms of the industrial upgrading of regions.

Rural attraction is an important indicator of the functions of the rural regional system during flow situations. This theoretical topic in rural development has been discussed within multiple disciplines, but lacks systematic research. Nevertheless, it has become a cutting-edge theoretical research topic in the field of rural geography. Following the research progress and theoretical foundations of rural attraction, this paper constructs a theoretical framework for rural attraction from a geographical perspective, and explains five of its most important issues. (1) The research literature on rural attraction has shifted from fragmented elaborations on the qualities of attractive villages and the unidimensional analysis of rural attraction to the exploration of its concept, mechanism, and evaluation system in terms of urban-rural interactions. The theories underlying the rural regional system of human-environment interactions, gravity, population migration, and attraction property rights, can provide a theoretical foundation and inspiration for the research on rural attraction. (2) Focusing on scientific issues in rural attraction, such as "what is-how-why-how to", and moving toward the framework goal of "theoretical innovation, serving practice, and methodological guidance," a theoretical framework for rural attraction is constructed with the support of spatiotemporal considerations, a comprehensive theoretical paradigm, and the methods used in the field of rural geography. (3) Five important issues experienced during efforts to improve rural attraction were analyzed. First, we analyzed the conceptual connotations of rural attraction based on the background of rural regional systems and spatial interactions from the perspective of spatiotemporal processes. Second, we constructed a measurement framework for rural attraction based on these conceptual connotations and evaluation principles. Third, we identified the regional types and spatial structures of rural attraction from the perspectives of spatial correlations and scale transformation. Fourth, following the theories of human-regional system of human-environment relationship and scale transformation, we explored the mechanisms of rural attraction's endogenous development and exogenous embedding processes. Fifth, considering historical processes and spatial heterogeneity, we explored the pathways and policy insights for enhancing rural attraction. Overall, this research attempts to construct a "what is-how-why-how to" research framework for rural attraction based on a comprehensive "spatiotemporal human-environment" perspective. This study aims to deepen and expand the theoretical research on rural attraction, in addition to providing theoretical references and policy insights for the implementation of rural revitalization strategies in the construction of harmonious and beautiful rural areas.

Training samples play an important role in machine learning-based regional landslide susceptibility evaluation. These samples consist of both landslide (positive) and nonlandslide (negative) samples collected through various sampling methods. However, existing methods for positive sample collection do not measure the reliability of the collected samples, leading to uncertainty in terms of reliability. To address this issue, this paper presents a landslide prototype sampling method (PBS). This method uses the geographical similarity and dissimilarity between a certain point and the landslide positive sample prototype to measure the reliability of positive and negative samples, respectively. A reliability threshold is set based on a mutual exclusion method to collect training samples. The Youfanggou Basin in Gansu province was chosen as the research area. The PBS and existing representative sampling methods were used to construct landslide susceptibility prediction models based on logistic regression, support vector machines, and random forests for the Youfanggou Basin. The evaluation effects of landslide susceptibility were compared between the reliable and nonreliable samples. The reliability of the positive and negative samples exhibited a "fluctuating increase" and "positive correlation", respectively, in the evaluation of landslide susceptibility. The PBS method improved the accuracy and area under the receiver operating characteristic curve (AUC) of the landslide susceptibility evaluation based on the three machine learning models by at least 14.7% and 14%, respectively, compared to the existing representative sampling methods, and the standard deviation was small, which indicates that the method proposed in this article is effective.

Amid the increasing global uncertainties, studying the risk structure and resilience level of global supply chains for agricultural products highlights their importance. It not only provides China with a scientific foundation for better utilizing international markets and resources and ensuring the safety of agricultural product imports, but also promotes the development of risk resilience management theories and methods for typical products across geographical spaces, thus expanding research in resource geography. This study builds a framework that consists of three sectors (export sectors, logistics and transportation sectors, and import sectors) and four stages (supply, procurement, transportation, and demand). Using two-stage DEA and CoDEA models, it evaluates the supply chain risks and resilience of soybeans, which is the typical import-dependent agricultural product in China. The findings indicate that: (1) The risk of China's imported soybean supply chain increased from 2000 to 2020, with risks predominantly from the demand and procurement stages. By contrast, the supply and transport stages are less risky but significantly increased. Trade relations with exporting countries and economic policy uncertainty are the main long-term risk factors that threaten the security of soybean import supply chains. (2) China's soybean supply chain from Brazil is more resilient than that from the US and Argentina, suggesting great potential for improving supply chain cooperation with Brazil. (3) Making the distinction between short- and long-term threats is necessary for managing the risk resilience of the global soybean supply chain. Improving the capacity to react to sudden hazards represented by trade restrictions, maritime obstructions, and trade sanctions should be the priority in the short term. Long-term exploration of system construction and strategic layout optimization is necessary to improve the discursive power of the supply chain.

Urban housing prices are influenced by various factors, encompassing macroeconomic conditions, urban planning strategies, and the specific characteristics of housing. These elements play a crucial role in shaping urban planning and development. Nonetheless, the regression analysis depicting the interplay between urban housing prices and their influencing factors reveals significant spatial non-stationarity and intricate nonlinear characteristics. Addressing the limitations of Euclidean distance in delineating spatial proximity for housing price modeling and the challenges encountered by the geographically weighted regression model (GWR) in capturing complex nonlinear features, this study introduces travel duration (TD) as a spatial distance metric and integrates it with a spatially weighted neural network to establish a geographically neural network weighted regression model with travel duration (TD-GNNWR) to estimate housing prices. In an empirical experiment using 2019 second-hand house data in Wuhan, the TD-GNNWR model demonstrates a 16% enhancement in fitting accuracy compared to the GWR model. The TD-GNNWR model notably enhances accuracy within sparsely sampled regions and better mimics their spatial distribution. Moreover, it adeptly captures spatial non-stationarity, offering a more precise elucidation of factors influencing housing prices in Wuhan and the resultant spatial discrepancies stemming from urban zoning. Our findings underscore the comprehensive impact of various factors on housing prices in Wuhan, such as building characteristics, neighborhood attributes, and transportation accessibility. Factors like greening rates, property fees, proximity to primary schools, universities, and public transportation exert substantial influence on housing prices in Wuhan, with varying directions and strengths across different areas, signifying clear spatial differentiation. The TD-GNNWR model clearly elucidates the mechanisms underlying housing price determinants while illustrating the inherent spatial non-stationarity, which is beneficial for urban planning departments and real estate managers in policy formulation, macro-control, urban planning, and investment decision-making. This work can also serve as a valuable reference for tackling challenges in urban analysis and modeling, thereby enriching methodologies within real estate research.