Journal of the Korean Geographical Society ISSN:1225-6633(Print) 대한지리학회지

This study aimed to identify the optimal contributing area threshold for hillslope unit delineation in the Nakdong River Basin using terrain analysis. A 5-meter DEM and the Stream Drop Analysis tool in TauDEM were employed to reflect the basin’s hydrological and topographic characteristics. The analysis indicated an optimal threshold of approximately 4 ha for the entire basin. However, substantial variability among sub-basins suggested that averaging thresholds to define a single value for the entire basin is inappropriate. Additionally, while threshold values generally increased with coarser DEM resolutions (10–100 m), they declined at resolutions above 120 m, yielding values similar to the 5 m DEM. These findings provide baseline data for mountain environment management and can inform future restoration and adaptive planning strategies.

This study investigates citizens’ perceptions of Slow Cities in South Korea and analyzes how key dimensions – local economic impact, cultural promotion, traditional value preservation, environmental protection, cultural preservation, resident relationships, intergenerational exchange, tourism industry impact, and local image improvement – influence six outcome variables: sustainable development contribution, overall attitudes, satisfaction, future prospects, intention to recommend, and youth employment support. Data were collected through an online survey of 900 respondents across six designated Slow Cities, and multiple regression analysis was employed to test the relationships. The findings indicate that local economic impact is the most consistent and influential predictor across all outcomes, underscoring the central role of economic vitality in sustaining citizen support. Local image improvement was also a strong driver, particularly for future prospects and youth-related outcomes. Intergenerational exchange significantly shaped attitudes, satisfaction, and recommendation intentions, highlighting its importance for strengthening social cohesion. By contrast, cultural promotion and tourism industry impact demonstrated limited effects, suggesting these factors alone are insufficient to enhance perceptions without broader economic or social benefits. The study provides policy, managerial, and geographical implications for advancing Slow City initiatives, emphasizing localized strategies, intergenerational engagement, environmental sustainability, and more effective communication of youth development and offers practical insights for policymakers and planners seeking to foster livable, future-oriented communities.

This study proposes a methodology for quantitatively evaluating fear of crime and crime safety based on Crime Prevention Through Environmental Design (CPTED) using street view imagery. Taking Yeongdeungpo-gu, Seoul as a case study, approximately 87,000 street view images were collected, with 30% used to construct a training dataset for deep learning. Fear of crime scores were predicted using the RSS-Swin model, which incorporates a Siamese Network, RankNet architecture, and a Swin Transformer backbone. Simultaneously, 15 quantitative CPTED indicators were developed, and semantic segmentation of street images and GIS data were used to compute crime safety design scores. Results indicated that older residential and industrial areas exhibited both higher fear of crime and lower CPTED-based design quality, while high-rise residential and commercial districts showed greater perceived and physical safety. A bivariate map integrating both metrics revealed four spatial patterns, providing policy-relevant insights. This study contributes an integrated framework combining perceptual and physical safety assessments, supporting more effective urban environment improvements and crime prevention strategies.

Rapid urban population concentration and city expansion have resulted in a progressive decline of green spaces, thereby intensifying environmental challenges and weakening urban ecological resilience. Rooftop greening, which creates conditions for vegetation growth on building rooftops, has been highlighted as an effective strategy to supplement insufficient urban greenery. In addition to mitigating the urban heat island effect, reducing air pollution, and improving stormwater management, its necessity is expanding across multiple dimensions. However, systematic information on buildings suitable for rooftop greening and their actual implementation remains limited. This study aims to detect potential rooftop greening buildings using deep learning techniques and to evaluate installation priorities at the building level. To this end, flat roofs were identified from aerial orthoimages using the Mask2Former model, and suitable buildings were extracted by linking these results with integrated GIS building information based on criteria such as age, height, and area. Subsequently, six evaluation indicators covering three functional domains—climate change mitigation, environmental and ecological functions, and social functions—were designed, and indicator-specific databases were constructed to comprehensively assess and prioritize candidate buildings. This study advances beyond the limitations of prior approaches in Korea, which primarily relied on GIS data and building registers for rooftop greening assessment, by integrating deep learning with high-resolution image analysis to provide a more precise identification of candidate buildings and a practical framework for establishing priority rankings.

This study explores how the core concepts of cultural geography—specifically landscapes and mobilities—have been transferred to and expanded within health geography. The field has consistently adopted and developed these concepts, beginning with therapeutic landscapes and evolving through therapeutic mobilities, networks, and assemblages. While recent frameworks such as therapeutic networks and assemblages have attracted scholarly attention, earlier notions such as therapeutic landscapes and mobilities have received comparatively less attention regarding their subsequent development and transformation. This study traces the motivations and trajectories behind these conceptual shifts, contributing to a deeper understanding of key ideas in health geography and refining its analytical tools for addressing real-world health challenges. Simultaneously, this study also contributes to cultural geography by uncovering the therapeutic value embedded in its core concepts, thereby enhancing their vitality and potential for further development.

Apartment prices in South Korea have experienced sharp fluctuations in recent years. However, the impact of these housing price changes has not appeared uniformly across the country, as the Seoul metropolitan area and regional cities exhibit distinct housing market structures. Moreover, heterogeneous spatial patterns of apartment prices emerge even within cities, which cannot be fully explained by simple geographic proximity. Therefore, this study employed a Functionally Weighted Regression(FWR) based on a migration-weight matrix instead of geographical proximity to explore the spatial heterogeneity of apartment price determinants in Busan, a major regional city distinct from the Seoul metropolitan area. The comparison with the geographically weighted regression (GWR) model shows that the functionally weighted regression (FWR) model achieved higher explanatory power, improved model accuracy, and resolved spatial autocorrelation of residuals. In addition, the analysis identified discontinuous local variations in apartment price determinants, which appear to reflect housing demand influenced by interregional migration flows. This study suggests the importance of considering functional connectivity between regions in housing policy and indicates that this analytical approach can be widely applied to various subfields of human geography, including studies of economic inequality, residential segregation, and accessibility disparities.

This study explores camera trapping as a visual method for animal geographies. With the “animal turn” in geography and the social sciences, human geographers have increasingly examined the agency of the animals and their capacity to reconfigure “animal spaces” into “beastly places”, viewing animals as a co-constitutive of the world alongside humans. However, their research methods largely remain conventional and anthropocentric – such as interviews and archival research – which calls for methodological innovation. This paper explores the potential of visual methods, particularly camera trapping through an empirical case study of human-crane relations. From 2020 to 2024, the research team installed thirteen trap cameras in Cheorwon, Gangwon Province, collecting approximately 190,000 photographs and 47,000 video recordings. The analysis reveals that trap cameras capture nocturnal and close-up encounters inaccessible to humans, thereby unveiling the agency of cranes and the “more- than-human” world they inhabit. These visual materials evoke affects of wonder and sympathy, while “animating” the everyday lives of animals. At the same time, the use of trap cameras exposes anthropocentric limitations and ethical concerns regarding nonhuman consent and privacy. This study argues that camera trapping - beyond being an ecological monitoring device - can serve as an embodied and affective methodology for multispecies and more-than-human geographies.