The core task of industrial geography research is to reveal the formation, evolution, and mechanisms of industrial development patterns. By applying scale thinking, industrial geography research roughly divides the real world into three geographic scale spaces: global, national, and local, and discusses their industrial geography effects from a specific scale perspective. However, due to the complexity and diversity of industrial geography processes, it is difficult to provide a systematic explanation using only one theoretical system. To compensate for this deficiency and further promote the application of scale thinking on industrial geography research, this article proposes a scaled localization model. The scaled localization model takes localization as the theoretical starting point and systematically integrates classic theories of industrial geography research, emphasizing that global, national, and local scale forces need to be localized to realize industrial geography effects. In the localization process, various types of scale forces mutually regulate and form a joint force, promoting complex and diverse industrial geography processes. This paper further employs the scaled localization model to analyze the mechanism underlying the historical restructuring of China's industrial geography. During 1978-2008, the first restructuring occurred under the combined effects of “strong global forces”, “strong local forces” and “weak national forces”, resulting in an uneven industrial geography pattern characterized by a developed eastern region and a lagging western region. During 2008-2020, there was a fundamental equilibrium among those three scale forces, significantly enhancing the inter-regional coordination in industrial development. In recent years, with the implementation of the “dual circulation” strategy, the influence of national forces has been amplified. This is expected to drive profound industrial transformation in China, ultimately shaping a multi-polar and chain-integrated spatial development pattern. In general, the scaled localization model provides theoretical support for the construction of China's industrial geography pattern in the new era by profoundly elucidating the interactive relationship between different scale forces and how they affect the evolution of China's industrial geography pattern.

The “production-living-ecological” spaces (PLES) are an important aspect of territorial spatial research. However, the theoretical and methodological systems for PLES evaluation remain incomplete. This paper proposes a shift in the evaluation object to the “human-land spatial system”. This approach overcomes a key limitation of traditional methods—the “land element method” and “indicator system method”—which substitute land elements or functions for the actual spatial entity. Based on these, it constructs a framework, principles, and procedures for the evaluation of PLES. Furthermore, it demonstrates the application of key theories and methods for evaluating PLES by taking the Huaihe Ecological Economic Belt as an example. This research covers the following aspects: (1) Guided by the proposed theories of the human-land and urban-rural spatial systems, this study adopts space as its physical evaluation object to analyze the structure, scale, and characteristics of PLES from the perspectives of dialectical materialism and structuralism. (2) This paper constructs a comprehensive methodology for PLES evaluation by integrating multiple theoretical foundations into a cohesive framework. The core evaluation framework is built upon the geographical paradigm of “pattern-process-mechanism-effect” and the formal logic of “ontology-extension”. Guided by the systems theory concept of “elements-structure-function”, the methodology establishes evaluation principles centered on “basic elements-fundamental structure-underlying function” “qualitative-quantitative” “region-locality-plot”, and “urban-rural” dimensions. Finally, the evaluation procedure is organized according to the philosophical “entity-attribute” view, progressing through “entity evaluation, attribute evaluation, and comprehensive evaluation”. (3) The study reveals distinct distribution patterns of PLES in the Huaihe Ecological Economic Belt. Production and living spaces are more concentrated in the east and north, with lower densities in the west and south. In contrast, ecological space is predominantly located in the southeast, with less in the northwest. These findings provide a scientific basis for future research on territorial spatial planning.

Currently, China is vigorously promoting its digital transformation and the development of new quality productive forces. A key objective is to establish channels for the intelligent production of high-quality land spatial information products, which has become a normalized and pressing need. The demand for accelerating the normalization of data-driven monitoring and supervision of natural resources is becoming increasingly urgent. In recent years, significant progress has been made in the natural resource sector, marked by the development of numerous spatial information products at various levels—focusing on land use, land cover change, soil, and land resources—as well as the successful construction and specialized application of diverse land information systems. However, there are still problems such as inconsistent benchmarks, information disconnection, rough mapping forms, and lack of intelligent means to meet the new requirements of full coverage, spatiotemporal uniqueness, full content integration, high frequency update, and low-cost customization. The prominent manifestation is the lack of high quality, incomplete content, insufficient relevance, and weak updates. In view of this, this paper focuses on the agricultural production space. It is oriented towards the preparation of fine land spatial information products based on a detailed investigation of existing methods. It coordinates multi-source spatiotemporal data resources and develops an intelligent solving method for land parameters. Its connotation and preferability are systematically explained. The focus is on the integrated implementation path of geographical zoning control, object stratified extraction, feature spectrum/time division recombination, knowledge direction correlation, and parameter grading calculation. Through a case study on the production of spatial information on the agricultural land for planting, this paper illustrates the step-by-step generation process of land parcel level agricultural information from the perspectives of fine spatial expression, multi-source information fusion, and multi-level parameter calculation. It preliminarily demonstrates the style, characteristics, and potential support for improving the level of comprehensive land research and intelligent application in natural resources field. This study attempts to achieve comprehensive expression of land information through object-oriented modeling. The proposed method for land parameter estimation offers a novel perspective for refining and quantifying comprehensive geographic research in the big data era. It provides insights for applying and upgrading land spatial information, with significant reference value for both theoretical methodology and practical applications.

In recent years, the efficiency of remote sensing scene image classification has significantly improved with the application of deep learning techniques. However, most of these methods heavily rely on pre-trained weights, and their performance may degrade without them. To address this issue, this paper proposes a three-branch fusion algorithm (EMA_ConvNeXt_Swin Transformer: ECST model), which integrates the Swin Transformer and ConvNeXt while incorporating an Efficient Multi-scale Attention (EMA) module after each fusion stage to enhance global contextual features, whereas the ConvNeXt branch specializes in extracting local spatial features. A dedicated fusion module is designed to refine and integrate the outputs of these two branches. The subsequent EMA module further establishes long-range dependencies in the spatial domain, enhancing the global contextual representation of the feature maps and facilitating richer feature aggregation. Experiments were conducted on two benchmark datasets, NWPU-RESISC45 and AID, achieving classification accuracies of 91.25% and 90.9%, respectively. To comprehensively evaluate the proposed model, two comparative experiments were designed: A comparison with classical deep learning models (AlexNet, VGG, Vision Transformer, ConvNeXt, and Swin Transformer), all tested without utilizing pre-trained weights.A comparison with state-of-the-art models that leverage pre-trained weights.The results demonstrate that the ECST model significantly outperforms classical models in accuracy when no pre-trained weights are used. Furthermore, it achieves competitive performance against mainstream models that rely on pre-training. Without relying on pre-trained weights, the proposed ECST model delivers superior classification performance compared to classical deep learning models and remains highly competitive when benchmarked against state-of-the-art models that utilize pre-trained weights. These findings highlight the model's effectiveness in remote sensing scene classification and its potential to reduce dependency on pre-trained weights in future applications.

Ecosystem assessment plays a crucial role in ensuring sustainable economic and social development. However, existing ecological security indicator weighting methods often neglect temporal effects and the influence of “weak links”, while the analysis of obstacle factors contributing to regional urban ecological security disparities remains insufficient. To address these gaps, this study introduces a penalty-based Variable Weight Model integrated with the Pressure-State-Response (PSR) framework to scientifically evaluate urban ecological security. Combined with standard deviation ellipse models, and obstacle degree analysis, we systematically investigate the spatiotemporal characteristics and constraining factors of ecological security in the Beijing-Tianjin-Hebei region from 2006 to 2020. Key findings include: (1) The ecological security index of cities in this region showed continuous growth with fluctuations during 2006-2020, achieving an average annual 4.84% increase. (2) The overall contribution sequence of the urban ecological security subsystem is response (accounting for 45.83%) > state (accounting for 33.53%) > pressure (accounting for 20.64%). Cities with high security levels exhibit more significant subsystem imbalances. (3) Insufficient sewage treatment capacity, low proportion of scientific and technological expenditures, and issues related to cultural resource allocation are the key obstructive factors that have long constrained the development of urban ecological security in the Beijing-Tianjin-Hebei region.

Deepening the implementation of people-oriented new urbanization and exploring the enhancement of urban vitality through the creation of well-designed built environments are crucial for urban planning and decision-making by local governments. Addressing the Modifiable Areal Unit Problem (MAUP) and introducing a relatively optimal spatial analysis unit are essential for characterizing the spatial features of urban vitality and understanding the impact of built environments on urban vitality. Taking Guangzhou City as an example, this study utilizes mobile phone SDK-based population profiling data to represent urban vitality. Building on existing research, classical built environment variables are selected, and the technical framework for empirical analysis is optimized. The Optimal Parameter-based Geodetector (OPGD) model is employed to compare the explanatory power of multiple common spatial analysis units, thereby identifying the relatively optimal spatial unit (1 km²). Based on this, the relative importance of built environment variables on urban vitality during nighttime and daytime is clarified, and the spatial heterogeneity of these impacts is explored using the Multi-scale Geographically Weighted Regression (MGWR) model. Key findings include: (1) The influence of built environment variables on urban vitality varies with spatial analysis units. The relatively optimal spatial analysis unit is identified as 1 km². (2) The spatial structure of urban vitality in Guangzhou remains consistent across both time periods, exhibiting a finger-like radial pattern centered on the four core districts and extending along major transportation corridors. (3) The densities of commercial, residential, and public service POIs exert the most significant impacts on urban vitality. Transportation convenience and accessibility serve as crucial sustaining factors for urban vitality. Strong interactive effects are observed among variables such as commercial and residential POI densities and distance to the urban center. (4) The spatial influences of built environment variables exhibit heterogeneity across the two time periods in terms of spatial significance ranges, magnitude of impacts, and directional effects. These spatial characteristics can be summarized as a gradient distribution across the entire city, a concentric ring distribution centered on core areas of the central four districts, and localized “cluster-based” or “scattered” spatial patterns.

During the urbanization process, high-intensity urban development has contributed to the degradation of protective green spaces, resulting in fragmented patterns and weakened structural integrity. This severely compromises their essential functions, which include protecting historical and cultural buildings, conserving biodiversity, reducing dust and noise pollution, and facilitating carbon sequestration and oxygen release. Therefore, rationally coordinating ecological restoration with functional optimization, and clarifying the three core issues of “where”, “when”, and “how” to renew these spaces, has become a critical research focus in advancing the national strategy of “urban renewal”. Based on the concept of Urban Renovation and Ecological Restoration, this study proposes a “Three health assessments and One zoning” framework for the evaluation and management of urban protective green spaces. This framework systematically addresses three key challenges in green space renewal. It employs a protective system assessment mechanism for the precise spatial positioning of suitable areas, ecological restoration and urban repair assessments to define intervention priorities, and zoning principles to establish phased implementation plans. The findings reveal that: (1) Between 2013 and 2023, 96.49 km² of green space in Xi'an experienced degradation, exhibiting an overall relatively low level of both quality and connectivity; (2) The suitable network area for protective green spaces totals 585.81 km², primarily consisting of industrial buffers, historical heritage protection belts, and river/lake water source protection zones. Within these, 37.59 km² and 63.55 km² show significant potential for ecological restoration and urban regeneration, respectively; (3) According to the renewal zoning results, 61.34 km² are identified as priority areas for near-term green space renewal. These results provide both theoretical and practical insights for the assessment and renewal planning of protective green spaces in Xi'an and other cities with similar contexts.

This study constructs a dual-scale analytical framework “inter-provincial and intra-provincial” technology transfer network, integrating spatial analysis, simulated attack methods, and mixed-effects generalized linear model to systematically reveal the evolutionary patterns and influencing factors of key channels and core nodes in Jiangsu's technology transfer network. The findings indicate that: (1) Jiangsu's inter-provincial technology transfer network has expanded significantly, exhibiting core-periphery structure of “dense in the east, but sparse in the west”. The network center of gravity has shifted westward, forming a bow-and-arrow spatial pattern with Jiangsu as the hub, connecting Beijing, Shanghai, Guangdong, Sichuan. Key transfer channels have shifted from “Shenyang-Nanjing” to higher-weight routes such as “Shanghai-Suzhou” and “Nantong-Shanghai”. Concurrently, the emergence of structurally critical, though lower-weight, channels has facilitated Jiangsu's transition from a “technology recipient” to a “national organizer and radiation source of technological elements”. (2) Within Jiangsu, the network has matured into an efficient cooperative system, with spatial patterns evolving from “Southern Jiangsu agglomeration” to a balanced “north-south collaboration”, forming a “diamond-shaped” structure with Nanjing, Suzhou, Nantong, Xuzhou, and Yancheng as its core hubs. The core has expanded from a single Nanjing centered system to a stable multi-core structure of “Suzhou-Nanjing-Nantong”, marking the emergence of a diversified, functionally complementary innovation structure that fosters balanced regional technological development. (3) The evolution of key channels is primarily driven by the intensity of technology flow, the aggregated weighted degree of nodes, and industrial structure similarity, whereas excessive economic disparity hinders their development. The formation and evolution of core nodes are highly dependent on their technological absorption and diffusion capacities, with economic and technological levels serving as secondary drivers.

Digital technology innovation serves as a crucial driving force for economic development, while synergistic growth provides a key pathway for promoting the orderly advancement of economic systems. Accurately understanding the relationship between the two is essential for achieving integrated regional development. This study investigates the evolutionary characteristics of economic synergistic growth across 41 cities in the Yangtze River Delta region from 2000 to 2021 by employing a composite system synergism model, a panel vector autoregressive (PVAR) model, and spatial econometric techniques. It analyzes the interactions between economic synergy and the ambidextrous innovation of digital technology, with particular attention to spatial effects. The findings are as follows: (1) The overall level of order in the urban composite economic system has exhibited a fluctuating upward trend over time. Among its subsystems, the economic structure subsystem has shown the fastest rate of orderly development, whereas the economic quality subsystem has achieved the highest degree of order. (2) The predominant patterns of economic synergistic growth have been characterized by antagonistic synergy and low synergy. By the end of the study period, 95.12% of the cities had achieved some degree of synergy, with 66.67% having transitioned out of disorder. The analysis of synergy migration patterns reveals that Shanghai experienced upward mobility, most cities in Jiangsu and Zhejiang remained in a stable low-synergy state, and cities in Anhui exhibited more dynamic changes. (3) A significant and stable mutual reinforcement has been observed between synergistic economic growth and incremental innovation. Breakthrough innovation promotes synergistic growth through self-reinforcement and injects new momentum into incremental innovation, which, in turn, drives long-term improvements in regional economic synergy. (4) Both breakthrough and incremental digital technology innovations have positively contributed to the region's economic synergistic growth. However, incremental innovation has demonstrated certain inter-regional negative externalities. Based on the interaction mechanisms between dual innovation and synergistic growth, this study proposes optimization pathways aimed at steadily enhancing economic quality and fostering efficient innovation development in the Yangtze River Delta region.

The generation and evolution of regional new technological activities is a classic proposition in economic geography. Taking the Yangtze River Delta region as a case study, this paper explores the impact of local knowledge base, inter-regional complementary linkages and their interactions on breakthrough innovations from the perspective of technological relevance, based on 4,119,100 granted patents in China from 2000 to 2020. The study finds that: (1) The level of breakthrough innovation in the study area had been increasing from 2000 to 2020, and the geographical distribution showed a significant core-edge structure. Moreover, the focus of breakthrough technological innovation has shifted from the traditional industries that met the basic societal needs of the society in the early stage to the knowledge-intensive and high-tech industries. (2) Technological unrelatedness is the core driving force of breakthrough innovation in this region, which can significantly promote the entry of breakthrough technologies in new fields and the growth of existing fields, while technological relatedness has a significant negative effect on the entry of breakthrough technologies and has no significant effect on the growth of these technologies. (3) Although interregional complementary linkages can foster the entry and growth of breakthrough technologies, their effect is not always positive. In fact, the interaction between complementary linkages and technological relatedness can hinder the entry of such technologies. The contribution of this paper is to explore the role of complementary external linkages in the generation and development of breakthrough technologies in the region, and the study provides policy implications for the development of breakthrough technological innovations in the Yangtze River Delta and other regions.

In recent years, with the prevalence of anti-globalization and trade protectionism, global trade frictions have occurred frequently, posing a threat to global economic growth. In-depth analysis of the characteristics of the global trade frictions relations and scientific assessment of their driving factors can not only help clarify the world economic pattern and trade situation, but also provide scientific basis for China to actively respond to trade frictions and promote high-level opening up. Based on the global trade frictions data from 2009 to 2022, this paper adopts GIS and complex networks to explore the evolution of the global trade frictions, and empirically analyses their driving factors using the negative binomial regression model. The main conclusions are as follows: (1) The intensity of global trade frictions has fluctuated, initially rising before declining, with state aid and subsidy policies being the main forms. Resource-intensive products and labor-intensive products have always been the hot spots among them. (2) The initiators are more concentrated and show the trend of high in the north while low in the south, which are absolutely dominated by the high-income countries and the upper-middle-income countries and their share fluctuates dramatically. The targets have become more dispersed, which are dominated by high-income countries and their share has a “stratified effect”. (3) Global trade frictions relations are netted into a global trade frictions network, which shows a trend of first expanding and then shrinking. Its core-periphery structure is developing steadily, and the core significantly shifts to the east, resulting in the prominent position of world's major trading countries. (4) The main drivers of the evolution of global trade frictions networks include economic development of the initiators and targets, scientific and technological level of the initiators, trade proximity, political inclination proximity, and organizational proximity.

Regional development imbalance is a common phenomenon worldwide. Strengthening the competitive advantage of economic growth poles and reducing income and employment gaps between regions are important areas of research in regional economic spatial governance. As a developed country that has achieved remarkable success in its economic development model, Germany's regional economic policies and the evolution of its economic landscape are of great significance for expanding the international perspective of China's regional coordinated development strategy and accumulating valuable experience. This study consolidates Germany's 16 federal states into six major regions, covering the period from 1991, after reunification, to 2020. It traces the transformation of regional economic policies in Germany after reunification, examines the patterns of regional economic development, and classifies the economic growth types of the six major regions during different phases using the deviation-share method. The study identifies the driving forces behind the economic growth of different regions and, finally, draws insights for China's regional economic development. The findings show that: Germany's regional economic policy underwent a marked shift following reunification, moving away from a capital-centric model toward an innovation-driven approach. It went through three major stages: resource transfer to the East, deepened development in the West, and the East's efforts to catch up with change. As a result, the economic gap between East and West narrowed, while the disparity between North and South expanded. The “intervention-feedback-adjustment” spiral evolutionary model has formed between policy and regional economic development, and regions that align with national policies achieve more favorable development. For China, the study suggests that policy formulation should aim for long-term developmental impact by strengthening synergy between government leadership and regional autonomy, seizing opportunities for path creation, and enhancing regional organizational capacity to build structural and competitive advantages.

The Taiwan region of China occupies a strategic position along a vital global shipping route, where its shipping direction plays an integral role. The “New Southbound Policy” put forward by Taiwan province seeks to shift its development strategy from a “westward” to a “southward” orientation, in an attempt to reduce its economic reliance on the Chinese mainland. Against this backdrop, it is imperative to explore the region's shipping reorientation and the effects of related policies. Against the backdrop of the “New Southbound Policy”, this paper systematically examines the evolution of foreign shipping links of Taiwan from 2011 to 2022. Employing an economic model coupled with a quasi-natural experiment methodology, this study delves into the causal effects of the “New Southbound Policy” and changes in shipping of Taiwan, exploring the actual impact of the policy on the transportation volume of different types of goods. The results show that: (1) The maritime shipping structure in the study area has remained largely stable. In terms of total trade volumes, Taiwan's strongest maritime trade links are with Australia, followed by several prominent Asian economies. Australia is the largest import market for the Taiwan region of China. The Chinese mainland serves as the largest export market for Taiwan. However, among the top 20 economies ranked by the significance of their maritime trade connections with Taiwan, only eight of the 18 economies covered by the “New Southbound Policy” are represented. (2) The “New Southbound Policy” has not significantly promoted shipping exchanges between Taiwan and “New Southbound” economies. A separate regression analysis of total maritime trade and industrial shipping types reveals that the “New Southbound Policy” has yielded a statistically significant positive effect only on bulk shipping. In contrast, total maritime trade volumes and container shipping have not been affected by this policy. (3) The regression results indicate that both trade openness and transportation capacity show significant positive effects among the factors influencing the bulk shipping connections between Taiwan and other economies. In contrast, the production capacity and liner connectivity indexes exhibit the opposite trend. Overall, the “New Southbound Policy” has not played a significant role in the overall bilateral trade in shipping between Taiwan and the “New Southbound” economies.

As integral elements of both material and spiritual civilization, couplets are an integral part of the cultural landscape of traditional villages, and an important carrier for the inheritance and promotion of traditional culture. Taking two typical national-level traditional villages in Hunan as case studies, the article applies the text analysis and landscape gene identification. The study aims to identify the landscape genes of couplets and interpret their spatial expression within these villages. The study reveals that: (1) The material cultural landscape features of the couplets in the villages are manifested as a diverse and highly visible landscape mainly consisting of architectural door walls, decorative patterns and calligraphic fonts. (2) The intangible cultural landscape features include cultural connotation features and literary and artistic features. The cultural connotation features reflect the historical, spiritual and geographic influences embedded in the couplets of the villages, while the literary and artistic features are manifested in the unique metrical characteristics and the use of natural and social imagery in their expression. (3) The genes of the couplet cultural landscape in Langshi Village are clean and honest culture, auspicious culture, patriotic culture, cultivation culture, root culture, seclusion culture, and ritual culture; the genes of the couplet cultural landscape in Longjiadayuan Village are seclusion culture, ritual culture, cultivation culture, and root culture; and symmetry aesthetics is the common cultural genes of all the couplet landscapes. (4) The landscape image of couplets in traditional villages exhhibits a rich hierarchy, constructing three spaces: a cultural space for aesthetic imagination, an educational space for philosophical discussion, and a spiritual space for emotional expression. These spaces are derived from the spiritual cultural genes of traditional Chinese culture, such as “closeness to nature”, “virtue and ambition”, and “the pursuit of poetry”. This study aims to provide new perspectives for the understanding of rural cultural inheritance, enrich the theoretical study of traditional village cultural landscape, and promote the sustainable development of traditional villages and the effective inheritance of rural cultural heritage.

Based on the background of the national strategy of “rural revitalization” and the preservation and destruction of traditional villages, this study focuses on the landscape of traditional villages in Liaocheng City. Integrating geographical cultural landscape theory and the “landscape information chain” concept, we constructed a corresponding information chain for these villages. Using the AHP-FCE method, we evaluated the weight indices of various factors within the vernacular landscape space. This analysis aims to identify paths for inheriting traditional village landscape genes in the context of rural revitalization, offering practical insights for enhancing cultural cohesion and safeguarding cultural landscape genes. The results show that: (1) Extraction of landscape genes. Using the feature deconstruction method and considering practical conditions, we classified traditional village landscape genes into material and non-material categories. A total of 4 identification elements and 14 identification indexes were selected to establish a landscape gene identification system, thereby constructing the landscape information chain. (2) Landscape gene evaluation. The intangible landscape issues such as folk activities and folk skills in the cultural landscape are the most prominent, followed by the material landscape issues of water bodies and square spaces. The most important landscape factors in traditional villages are traditional drama, traditional music, folk activities, folk skills and folk spirit. (3) Landscape gene inheritance path. For the inheritance path of village protection, the spatial scope of traditional villages is clarified as “landscape information point-landscape information corridor-landscape information network”. For the inheritance path of industrial introduction, For the industrial introduction pathway, regional economic development can be stimulated by utilizing the emerging model of integrating local handicrafts with tourism, thereby enhancing the area's cultural development. For the path of cultural activation and inheritance, regular special ethnic cultural activities can enhance villagers' sense sense of identity, deepen the endogenous link between native culture and landscape space, promote the inheritance of traditional civilization, and provide new ideas for realizing the sustainable development of local traditional villages.

Investigating the fusion and variation of multi-ethnic settlement landscape genes, along with their underlying mechanisms, constitutes a cutting-edge frontier in this field of research. Taking Tuanshan Village in Jianshui County as an example, this research identifies the village's landscape genes based on landscape gene theory and analyzes their variation characteristics and mechanisms from the perspective of cultural blending. The findings reveal that: (1) Across different stages of Han-Yi cultural blending and village evolution, the planar genes of Tuanshan Village underwent variation and subsequently integrated with indigenous elements; the facade genes largely harmonized with local traits, though some conflicted with or even vanished from the indigenous context; while cultural genes exhibited a coexistence of blending and contradiction with local traditions. (2) Prior to Han-Yi contact, the village developed indigenous Yi environmental adaptive genes, rooted in natural conditions and shaped by Yi culture. During the contact stage, the migration of Han people and the formation of the Zhang clan shifted landscape gene characteristics toward “Han-Yi fusion”. In the Han-Yi blending stage, sustained acculturation from the Han culture and the accumulation of family wealth led to the formation of a landscape gene pattern towards a “Han-dominant, Yi-supplementary” model. The decline stage featured abrupt mutations in decorative elements, clan culture, and customs driven by political movements. Finally, the revival stage saw the regeneration of “Han-Yi fusion” genes, driven by urbanization and tourism.

The green transformation of traditional agricultural villages is a crucial pathway for comprehensively advancing rural revitalization. However, their socio-economic structures are deeply embedded in the intrinsic social relationship networks, production structures, and cultural traditions of rural society. Consequently, these villages face severe challenges during the transformation process. For hilly and mountainous villages with scarce arable land resources, a critical issue in their modernization process is how to develop and activate elements embedded within the existing rural structure and achieve their re-embedding within a green agricultural framework. Grounded in embeddedness theory, this study constructs a three-dimensional coordination framework of “relational-institutional-place” embeddedness and employs Daizhuang Village, a hilly rural settlement in Jiangsu Province, as a case study to explore how geographically constrained traditional villages achieve the coordinated evolution of human-land-industry factors through embeddedness mechanisms. The findings reveal the following: (1) In the early stage of the green transition, the mobilization of “pioneer forces-natural resources-individual economy” creates localized and limited coordination, effectively initiating the initial transformation of the village's economic and social structure. (2) In the mid-transition stage, structural changes drive a broader and more integrated coordination among “the majority of villagers-locational resource-collective economy”, leading to continuous adaptation of the economic and social structure to the green production model. (3) In the later stage, “multiple stakeholders-local characteristics-industrial integration” foster deep and spontaneous coordination, resulting in the re-embedding of human-land-industry factors into a new rural economic and social structure. The research indicates that in future green transition planning and implementation, village cadres and rural elites should prioritize the embeddedness characteristics of key factors, emphasize the organic integration of factor development and structural transformation, and scientifically design phased transition frameworks. These approaches provide a more systematic theoretical foundation and offer a practical guidance for the green transition of traditional agricultural villages in hilly and mountainous regions.