Search

To retrospect the impact of urbanization on vegetation, consistent and reliable maps that provide urban vegetation estimates are required. In this study, we make the first attempt to employ deep learning for urban vegetation mapping of 2120 cities in China using a 30-years Landsat archive (UV-30). We present a multimodal deep learning (MDL) model to combine features from 5 categories, namely reflective bands, remote sensing indices, texture variables, topographical variables, and time labels. We normalized elevation data locally to remove the inter-city differences, while retaining topographic details. We used time labels to account for spectral variability over time and Landsat sensors. The performance of the MDL models was compared with a vanilla deep neural network, random forest, and support vector regression. Results indicated that the fusion of different categories of features improved the prediction accuracy. The MDL performed best in most validation cities, and our results were superior to the fractional maps aggregated from high-resolution datasets in homogeneous urban landscapes. We further analyzed urban vegetation dynamics across the urban core-new town-outskirt gradients of different geographical regions and city sizes. We found a gradually diminishing de-greening trend in urban China, especially in urban cores and new towns. Nevertheless, we observed a more pronounced degradation of urban vegetation in the south regions of China compared to the north, and smaller cities show no evident urban vegetation increase. The suggested framework is valuable for urban vegetation assessments and facilitates a thorough understanding and management of urban ecosystems. • Deep learning is utilized effectively to estimate large-scale urban vegetation for the first attempt. • Multimodal deep learning yielded promising results in urban vegetation estimation. • A 30-year time series of urban vegetation in China was reconstructed. • Dynamics of urban vegetation in China were analyzed. [ABSTRACT FROM AUTHOR]

Urban trees are important for adapting to climate change; however, the absence of fine-grained data describing the distribution of urban tree crown cover and carbon stocks hinders recognition of the contribution of urban systems to climate change mitigation. Here, we present an approach for extracting fine-grained tree crown cover by coupling the Segment Anything Model and vegetation indices using Google Earth imagery with a spatial resolution of 0.298 m. We estimated the aboveground biomass of tree-covered regions in Wuhan, representative of China's urbanization, using multi-source remote sensing data and machine-learning techniques. We show that tree crown cover accounts for 18.86 % of the study area, implying that the nationwide proportion of urban trees probably represents 1.88–2.69 % of the total forested area. Tree growth in urban regions remains reasonably stable owing to the high level of human management, with a 60–86 Mg C ha−1 carbon density of aboveground biomass in the tree-covered region in Wuhan. Street trees at specific distances from infrastructure represented an average level of aboveground biomass, whereas urban trees at relatively distant locations were the dominant contributors to aboveground biomass. Our study highlights the carbon stocks of urban trees and the various mechanisms that indirectly influence carbon emissions, representing a potentially promising nature-based solution. We recommend enhancing the socio-ecological characteristics of urban systems to address future climate change. [ABSTRACT FROM AUTHOR]

Non-grain production (NGP) represents a complex social phenomenon driven by a range of interests that involve agricultural development, livelihood security, and ecological preservation. It significantly alters the structure of agricultural production and vegetation cover, influencing regional grain supply and the multifunctional composition of agriculture. To achieve sustainable development goals related to grain supply and ecological stability, this study proposes a research framework for multifunctional evaluation and multiscenario regulation of non-grain farmlands (NGFs). First, NGFs in the Wuhan metropolitan area (WHM) were extracted based on land survey data to carry out multifunctional evaluation of NGFs, including production function, ecological functions and recreational function. Second, the SOM+K-means algorithm is used to identify and explore the multifunctional bundle characteristics of NGFs to determine the regulation priority order of NGFs. Finally, a multiscenario simulation of the rigid control scale and layout arrangement of NGFs from the grain security perspective is carried out. The research results show that (1) there are 1294.34 thousand ha of NGFs in the WHM, mainly of grain and non-grain rotation type (52.36 %). The production function is higher in the central and western Jianghan Plain areas than in the northeastern and southern hilly mountainous areas; the ecological functions are the opposite; the high-value areas of recreational function are mainly located around the built-up areas of the city. (2) Multifunctional bundles divide NGFs into seven categories and determine the order of regulation based on the principle of "production priority, functional diversity". (3) By setting different grain security scenarios, it is predicted that the rigid control scale of NGFs in the WHM in 2030 is 256.08–797.55 thousand ha. Divide and restore grain production in order of the priority zone for grain production, the reserve zone for grain production, the resilience zone for grain production, the ecological improvement zone and the agricultural improvement zone, so as to achieve a regulatory layout of strict control and adaptable regulation of NGFs. The research supports the regulation of NGFs in balancing grain security and ecological stability to promote sustainable agricultural production. • A research framework for multiscenario regulation of non-grain farmlands was proposed. • Multifunctions of non-grain farmlands and the bundled characteristics were explored. • Regulation zoning and control order of non-grain farmlands were set up. • Layout of non-grain farmlands from the grain security perspective was regulated. [ABSTRACT FROM AUTHOR]

Thermal exposure in street canyons has an essential impact on individual outdoor activities and overall health. Street view images (SVIs) have demonstrated significant value in investigating thermal exposure within street canyons. However, current SVIs-based studies cannot explicitly reflect daily thermal exposure, as their focus has been primarily on the spatial distribution of thermal exposure indicators. Few studies explored the relationship between thermal exposure and walking accessibility. In response to this situation, this study proposed a human-scale approach to quantify daily thermal exposure by integrating street thermal environment with walking accessibility. Hemispherical images generated from Baidu Street View panoramas were utilized to estimate the solar duration reaching street canyons, thereby reflecting the thermal exposure for each street canyon. Street networks were analyzed using the spatial syntax method to quantify walking accessibility for each road. By combining street-level solar duration and walking accessibility, this study measured daily thermal exposure from the human-scale perspective and identified priority areas for thermal exposure interventions. Furthermore, a comparison was performed between daily thermal exposure and land surface temperature derived from remote sensing. The results indicated that thermal exposure interventions need to be prioritized in new development areas near the Third Ring Road. Additionally, top-down thermal exposure measurements may not accurately represent the daily thermal exposure experienced by pedestrians. Incorporating human-scale thermal exposure as another essential indicator can contribute to a more comprehensive analysis of thermal exposure in urban planning practice. In summary, the combined approach of spatial syntax and SVIs provides a promising paradigm for fine-grained thermal exposure assessment, which is an initial exploration of human-centered urban planning. • Solar duration is estimated by overlaying the solar path with hemispherical images. • Space syntax is utilized to measure walking accessibility. • A human-scale method is proposed to explore pedestrian daily thermal exposure. • Priority streets are determined by overlaying solar duration and accessibility. • Relying solely on land surface temperatures may cause bias. [ABSTRACT FROM AUTHOR]

[Display omitted] • Spatiotemporal heterogeneity of NEP mechanisms was explored. • Terrestrial ecosystems in study area showed a growing trend of carbon sinks. • Effects of natural factors was weakening, opposite for socioeconomic factors. • Eco-improvement benefited NEP, opposite for urban built-up area expansion. Urban cluster carbon sinks are key parts of the terrestrial carbon cycle, and play an important role in global climate change. Based on the inversion results of net ecosystem productivity (NEP) in Middle Reaches of the Yangtze River Urban Agglomeration (MRYRUA), this study considered the influence of both natural environmental and socioeconomic factors on NEP, and investigated the driving mechanism of carbon sink emphasizing spatial and temporal heterogeneity by the geographically and temporally weighted regression model (GTWR). Results demonstrated that (1) During 2008–2020, the annual average NEP was positive in MRYRUA, and the carbon sink effect of terrestrial ecosystem increased. (2) In MRYRUA, Precipitation and DEM were positively correlated with carbon sinks, opposite for temperature, percentage of water areas and population density. (3) During the study period, the influence of natural environmental factors, such as temperature and precipitation on NEP grew gradually weaker, whereas that of socioeconomic factors increased. The influence of urbanization rate, night lights index and factors of industrial structure changed from negative to positive, while percentage of built-up area had an increasing negative effect on NEP. Based on the conclusions above, this study proposed differentiated policy implications, including improving vegetation hydrothermal environment, controlling urban expansion and accelerating industrial transformation. It would provide empirical inspiration for attaining peak carbon, carbon neutrality, and sustainable development goals. [ABSTRACT FROM AUTHOR]

• A historical-information-based PMACO model was proposed to support land planning. • Methods for quantifying and coupling historical information were designed. • Interpretable machine learning models were used to explain land use evolution. • A size-adaptive strategy is applied to depict the heterogeneity in neighborhood. • The effectiveness of the HI-PMACO model was compared and validated. Land use optimization is a crucial approach for promoting sustainable development. However, current land use optimization models usually ignore the temporal pattern of land use evolution. Research on how to quantify and incorporate historical information into land use optimization process is scarce, and worth exploring. This study proposed a historical-information-based patch-level multi-type ant colony optimization (HI-PMACO) model to examine the effect of incorporating historical information into the optimization model from both macro and local perspectives. Specifically, we (1) constructed five machine learning models to explore the transition potentials and used the SHAP algorithm to visualize the non-linear effects of spatial variables; (2) introduced a size-adaptive neighborhood strategy to quantify the local land use evolution preference; (3) designed an evolution preference-weighted roulette wheel mechanism to incorporate the extracted historical information into the biomimetic intelligent algorithm; and (4) verified the effect of the HI-PMACO model. Results demonstrated that the proposed HI-PMACO model exhibited better performance in improving economic benefit and transition potential objectives with smaller land use change costs, and the local details were highly in line with relevant planning policies. This study contributes to spatio-temporal land use optimization modeling in methodology and land use management considering temporal evolution patterns. [ABSTRACT FROM AUTHOR]

Many ecologists have demonstrated that building ecological networks (ENs) can resist habitat fragmentation. However, existing studies have not yet explored the directionality of ENs. In this study, we proposed a research framework for the construction and structural evaluation of directed ENs from the perspective of dynamic biological flows. The framework reveals the operational essence of ENs, which is that dynamic biological flows rely on static carriers for directed movement. Thus, this framework overcomes the shortcomings of previous studies that only focused on static carriers and ignored dynamic biological flows. In other words, this study will guide the research of ENs from undirected to directed. The specific research and innovations are as follows. (1) First, the MaxEnt model and InVEST model were combined to evaluate habitat suitability in Wuhan Metropolitan Area (WHM), and 180 species habitats were identified accordingly. A total of 366 potential species migration corridors were extracted based on the identified species habitats and ecological resistance surfaces. (2) Second, we constructed a new relationship function between dynamic biological flows and potential biological flows, species dispersal probability, and species dispersal attraction. The results indicate that dynamic biological flows can serve as the measurement of directionality for directed ENs, and their density and intensity exhibit significant spatial heterogeneity. (3) Finally, we introduced the Chu-Liu/Edmonds' algorithm for complex network analysis to reveal the tree-like structure of directed ENs. The results show that the backbone and branches of directed ENs in WHM are species migration pathways with maximum dynamic biological flows linking global and local habitats, respectively. Meanwhile, the implementation of hierarchical protection according to the tree-like structure of directed ENs can significantly enhance its robustness. [Display omitted] • A conceptual framework of directed ecological networks (ENs) was developed. • MaxEnt and InVEST model were combined to identify species habitats. • Directed ENs was constructed from the perspective of dynamic biological flows. • Chu-Liu/Edmonds' algorithm was applied to reveal the tree-like structure of directed ENs. • Protection priority of ENs was determined based on the structural evaluation. [ABSTRACT FROM AUTHOR]

One of the key factors affecting rural settlements distribution is intervillage social connections. However, most of the current studies ignored the effects of social connections on the spatial distribution of rural settlements. This study proposed a multi-objective spatial reconstruction model for rural settlements based on the particle swarm optimization algorithm. The model considers intervillage social connections as one of the reconstruction objectives together with land-use quantity, spatial compactness, land-use suitability and relocation distance. Based on a questionnaire survey, the social connection strength between villages was measured from spatial and nonspatial aspects. The model was calibrated for the Chengui Township of Hubei Province in Central China. Five modeling scenarios for rural settlement reconstruction were designed according to potential development modes used in practice. A comparison of the reconstruction results of the different scenarios was performed to verify the applicability of the model and to analyze the joint interaction of the spatial objectives and social connections during the restructuring of rural settlements. The results indicated that five scenarios were effectively and robustly implemented using the proposed model. Social connections had a direct effect by guiding the restructuring direction of rural settlements. This study advances our understanding of the influence of social connections on rural settlement reconstruction and provides a useful exploratory tool for policymakers and rural planners in formulating reasonable reconstruction plans. • We integrate intervillage social connections into the spatial objectives of rural settlement reconstruction. • Five modeling scenarios are designed for rural settlement reconstruction. • Intervillage social connections have a direct effect by guiding the restructuring direction of rural settlements. • This study advances our understanding of the influence of social connections on rural settlement reconstruction. [ABSTRACT FROM AUTHOR]

Ecologists and policymakers strive to develop and optimize ecological networks (ENs) to prevent landscape fragmentation. However, the role of the cluster structure for ENs is often overlooked. In this study, we established a cluster management mode of ENs in Wuhan Metropolis (WHM), with a focus on dividing ENs based on identified clusters and implementing differentiated optimization, protection, and strengthening measures. First, we developed an optimization method to solve the defects of layout (connectivity blind areas) and quality (barriers and pinch points) for the initially constructed ENs. The results showed that the connectivity blind areas of ENs were significantly shrunk after layout optimization. Meanwhile, ENs after quality optimization avoided rugged, redundant or obstructive corridors, while narrow corridors were also given priority protection. Second, the Infomap algorithm was applied to identify 33 clusters of optimized ENs in WHM. Finally, we evaluated EN clusters to provide guidance and bases for the cluster management mode of ENs. The results indicated that centrality indicators can be applied to evaluate the protection importance for ecological edges within clusters, and connectivity indicators can be used to evaluate the protection importance and strengthening potential for ecological edges between clusters. We assert that incorporating the cluster management mode of ENs into future ecological protection plans is essential for maintaining ecological integrity and connectivity. • A cluster management mode for ecological networks (ENs) was developed. • An optimization method was developed to solve defects for initial ENs. • The Infomap algorithm was applied to identify the cluster structure of ENs. • The protected and strengthened components between and within clusters were identified. [ABSTRACT FROM AUTHOR]

• UHRswere systematically defined and classified based on active and passive demands. • The GCLQ can better identify the proximity and complementarity of UHRs. • The combined weighted method effectively measure the comprehensive accessibility. • Diversity types could enhance the complementarity and promote comprehensive equity. • Great differences in the accessibility and equityofUHRsamong socioeconomic groups. Previous studies have focused more on the accessibility and equity of single-type facilities for urban health, limited in comprehensive assessment of multi-type facilities and ignored the complexity of urban system. This study takes the first attempt to explicitly defined and classified urban health resources considering the active and passive health demands by building a conceptual framework. An integrated approach framework, including the global collaborative location quotient (GCLQ), the Gaussian 2SFCA method and Gini coefficients, were constructed to evaluate the proximity, complementarity, accessibility and equity of multi-tiered healthy resources in Guangzhou. In addition, the accessibility associated with the socioeconomic characteristics were explored by the local correlation model and further explained by the Mann-Whitney U test. The results showed that significant differences in spatial access to various health resources. The diversity types could enhance the complementarity of health resources and promote comprehensive equity, but groups with low economic levels enjoyed less access to health resources. This study contributes to the international literature on health-seeking behaviors from the perspective of active and passive demands and employing advanced geospatialmodeling methods. Through accurate identification of spatial barriers for health, the findings could provide valuable suggestions for the healthy and sustainable development of high-density population cities. [ABSTRACT FROM AUTHOR]

Understanding how built environment attributes affect health remains important. While many studies have explored the objective characteristics of built environments that affect health outcomes, few have examined the role of human perceptions of built environments on physical health. Baidu Street View images and computer vision technological advances have helped researchers overcome the constraints of traditional methods of measuring human perceptions (e.g., these methods are laborious, time-consuming, and costly), allowing for large-scale measurements of human perceptions. This study estimated human perceptions of the built environment (e.g., beauty, boredom, depression, safety, vitality, and wealth) by adopting Baidu Street View images and deep learning algorithms. Negative binomial regression models were employed to analyze the relationship between human perceptions and cardiovascular disease in older adults (e.g., ischemic heart disease and cerebrovascular disease). The results indicated that wealth perception is negatively related to the risk of cardiovascular disease. However, depression and vitality perceptions are positively associated with the risk of cardiovascular disease. Furthermore, we found no relationship between beauty, boredom, safety perceptions, and the risk of cardiovascular disease. Our findings highlight the importance of human perceptions in the development of healthy city planning and facilitate a comprehensive understanding of the relationship between built environment characteristics and health outcomes in older adults. They also demonstrate that street view images have the potential to provide insights into this complicated issue, assisting in the formulation of refined interventions and health policies. • Human perceptions were evaluated using street view images and deep learning. • Links between human perceptions and cardiovascular disease (CVD) risk were explored. • Wealth perception was negatively associated with CVD risk. • Perceptions of depression and vitality were positively related to CVD risk. • Perceptions of beauty, boredom, and security were not associated with CVD risk. [ABSTRACT FROM AUTHOR]

Abstract Residential communities are the basic living units in Chinese cities. Housing prices are closely associated with the community location and surrounding support facilities. When selecting satisfactory residential accommodation, potential real estate purchasers prioritize the community location in a city at the macro-level and then consider other micro-factors (i.e., the floor, orientation, structure, etc.). This paper attempts to explore the relationship between housing prices and locational factors at the community level. We collect the current market prices of 545 residential communities built in the last decade in Ningbo, the second largest city in Zhejiang Province. Then, thirteen locational factors of five dimensions are identified to research their influences on housing prices. In the process of selecting certain locational variables, both extant features and additional features (i.e., planned ones) are considered. The geographic field model is introduced to quantify the external effects of locational factors, due to its advantages of producing more accurate results than that of traditional distance-based measure methods. Then, regression analysis is performed based on the average housing prices of residential communities and explanatory variables by the ordinary least squares model and the geographically weighted regression. The regression coefficients demonstrate that the externalities of parks, lakes, department stores, banks, secondary schools and rail transit have significant but spatially non-stationary effects on housing prices. The results provide references for local real estate planning departments and potential real estate purchasers. Highlights • This paper explores the relationship between the locational factors' externality and housing prices at the community level. • The geographic field model is introduced to quantify explanatory variables for more accurate results. • The magnitude and spatial heterogeneity are observed in the influence of different locational factors on housing prices. • Even rail transit in the planning stage can still have a significant effect on housing prices. • The natural landscape features must be of a certain size or quality to exert an economic influence on housing prices. [ABSTRACT FROM AUTHOR]

Highlights • Identifying the advantages and disadvantages of the extant RHL transfer modes. • A new framework for RHL transfer is established based on the separation of three rights and two values. • A long-term mechanism to protect farmers' interests is built under collective membership. • It increases the understanding of the relationships among farmers, collectives, and governments. Abstract Rapid urbanization in China has triggered the mass migration of rural populations to cities, resulting in a shortage of urban land and inefficient use of rural homestead land (RHL). To address these issues, there is increasing interest in allowing for RHL transfer. Although several transfer modes have been implemented in pilot areas, the long-term protection of farmers' interests has been largely neglected. One-off transfer compensation and forced transfer often occur in practice, and there is no collective profit sharing. This paper explores the advantages and disadvantages of the extant transfer modes using comparative analysis. We propose a novel RHL transfer system under collective ownership, and three highly related key aspects add layers of protection: the separation of three sets of rights (ownership, tenure, and membership), the separation of land value from the value of accessory buildings and durable goods, and profit sharing among the collective members. The advantages of the proposal are summarized as follows: (1) it provides long-term protection for farmers' interests based on membership in the collective; (2) endows the collective with actual rights of self-governance and collective autonomy; and (3) decreases institutional costs and reform risks and improves land-use efficiency of RHL. The implications of the proposal regarding topics of concern are also discussed. This study offers a clear direction for RHL transfer and increases the understanding of the essence of the collective ownership of RHL and relationships among farmers, collectives, and governments during RHL transfer. [ABSTRACT FROM AUTHOR]

Due to the rapid socio-economic development, citizens are approaching a leisure era during which they are pursuing more psychological needs towards lifting the quality of life. Leisure venues act as the spatial carriers of leisure activities. The spatiality of leisure venues is closely linked with both cities’ intelligent growth and citizens’ life quality. We highlight the need for quantifying the leisure spatial configuration to guide citizens’ healthy leisure lifestyle, to optimize urban planning and to promote people-centered urbanization transition. This article aims to quantitatively characterize the spatiality of leisure venues via a series of geospatial techniques for leisure spatial pattern metrics. The Metropolitan Development Zone (MDZ) of Wuhan City is selected as the study site. Research findings are concluded that: 1) A three-level urban polycentric structure is identified. 2) The gradient variation of leisure venues is found. That is, leisure venue gradient density decreases with the distance from urban centers increasing. 3) The gradient effects of the four sub-types of leisure venues (i.e., tourist, cultural, sportive and recreational sites) are all highly fitted to the power function. 4) A significant correlation exists between the leisure spatiality and the demographic distribution. The research facilitates political strategies-making of sustainable urban spatial development. [ABSTRACT FROM AUTHOR]

Urban land expansion is closely related to and interacts with human activities. Although extensive studies have investigated the characteristics and patterns of urban land expansion, the coupling between urban land expansion and human activities has been neglected. Therefore, this paper explores the relationship between urban land expansion and the scope of human activities in the Yangtze River Economic Belt (the YREB) of China during the period of 1995–2015 based on Landsat and nighttime light remote sensing data. Overall, a coupling exists between urban land and the scope of human activities at the scales of the metropolitan area and the urban agglomeration. On the one hand, the degree of match between urban land and the scope of human activities has an upward tendency with time. According to the results of our regression model and landscape indexes, the scope of human activities outside urban land is associated with the magnitude of urban land, and the land within the scope of human activities outside urban land changed more intensely in developed areas. On the other hand, a coupling between newly increased urban land and the scope of human activities was proven by calculating the degree of match and identifying three urban land expansion types based on the location relationship. This paper argues that although the degree of match increased and the dominant type of urban land expansion transformed from outlying to backfilling, the problems of disorderly expansion of urban land and imbalanced development of the spatial distribution and scale structure still exist. The results can be used to formulate reasonable policies and planning and to promote the regional integration and coordinated development of urban land and human activities. [ABSTRACT FROM AUTHOR]