Tourism development in ecologically fragile areas faces the dual challenge of promoting economic growth while safeguarding environmental sustainability. Taking Inner Mongolia as a case study, this paper examines the spatiotemporal evolution, spatial resilience, and driving factors of A-level tourist attractions from 2005 to 2023. By integrating the Average Nearest Neighbor (ANN), Kernel Density Estimation (KDE), Standard Deviational Ellipse (SDE), and Geodetector methods, the study reveals three main findings. First, the tourism system has shifted from extensive quantitative expansion to intensive quality improvement, with the grade structure evolving from a pyramid-shaped distribution toward a more olive-shaped pattern. During 2019–2023, the spatial structure exhibited resilient stability and maintained a relatively mature polycentric pattern characterized by “one core and two sub-centers”. Second, the overall distribution of tourist attractions consistently followed a northeast–southwest orientation shaped by the regional geographical framework, while also showing increasing lateral diffusion into surrounding hinterland areas. Third, the driving factors displayed marked regional heterogeneity. Western Inner Mongolia followed an oasis-based development pattern, mainly associated with tertiary-industry GDP and public cultural facilities; eastern Inner Mongolia exhibited a resource-based, accessibility-constrained pattern, shaped primarily by topography and transport conditions; and central Inner Mongolia showed a coordinated support pattern supported by infrastructure, service capacity, and population concentration. Interaction analysis further suggests a landscape-embedded culture–nature coupling effect, in which the combined influence of topography and cultural facilities is more strongly associated with the spatial differentiation of tourist attractions than single factors alone. These findings provide a useful reference for differentiated spatial governance and sustainable tourism planning in ecologically fragile border regions.
Baohui et al. (Sat,) studied this question.