Summary statistics for MGWR parameter estimates.

Small towns play a crucial role in bridging urban and rural territory systems. While numerous studies have identified the characteristics and causes of small town shrinkage (STS), there remains an unexplored perspective on the reasons for their shrinkage from the perspective of the rural-urban relationship. To address this research gap, we investigated the relationship between STS and rural-urban interaction (RUI) in China. We hypothesized that a negative relationship existed between the degree of STS and the intensity of RUI. Using geo-statistical methods, such as the multi-scale geographical weighted regression (MGWR) model, the hypothesis was tested using Henan Province in China as a case study. The results indicated that the phenomenon of STS was observed extensively across the study region, with a 59% geographical overlap between the high-value area of STS and the low-value area of urban-rural interaction. Three distinct sub-types of STS regions were identified: shrinking regions along geographical borders, shrinking regions adjacent to metropolitan areas, and shrinking regions in ecologically fragile areas. The factors influencing STS demonstrated spatial heterogeneity and multi-scale characteristics. The findings will improve our understanding of urban shrinkage from a multi-level perspective and offer policy makers guidance for the sustainable development of small towns based on local conditions.

小城镇在衔接城乡地域系统中发挥着至关重要的作用。尽管诸多研究已阐明小城镇收缩（Small Town Shrinkage, STS）的特征与成因，但从城乡关系视角解析其收缩动因的研究仍有待补充。为填补这一研究空白，本文以中国为研究区域，探究了小城镇收缩与城乡互动（Rural-Urban Interaction, RUI）之间的关联，并提出研究假设：小城镇收缩程度与城乡互动强度呈负相关关系。本文采用多尺度地理加权回归（Multi-scale Geographical Weighted Regression, MGWR）等地理统计方法，以中国河南省为案例区域对该假设进行了实证检验。研究结果显示，研究区域内小城镇收缩现象广泛分布，且小城镇收缩高值区与城乡互动低值区的空间重叠率达59%。研究共识别出三类特征鲜明的小城镇收缩区域亚型：沿地理边界分布的收缩区域、毗邻大都市区的收缩区域以及生态脆弱区的收缩区域。小城镇收缩的影响因素呈现空间异质性与多尺度特征。本研究结论有助于从多维度视角深化对城市收缩的认知，并可为政策制定者因地制宜推进小城镇可持续发展提供决策参考。