Spatial Association of Traditional Timber Covered Bridges with the Northern Tea-Horse Ancient Road: Spatial Distribution and Natural Influencing Factors in Longnan, Northwest China

Longnan, located in Gansu Province, China, at the junction of Shaanxi, Gansu, and Sichuan provinces, represents one of the key corridors of the Northern Tea-Horse Ancient Road. This region preserves abundant traditional timber covered bridges with distinct local characteristics. This study employs ArcGIS spatial analysis and documentary research methods to explore the spatial distribution, spatiotemporal evolution, and influencing factors of these bridges. Spatial analyses (nearest neighbor index, kernel density, and standard deviational ellipse) are based on 71 bridges with traceable coordinates, while the temporal evolution analysis incorporates 80 bridges (64 with definite construction periods and 16 with unknown dates; the latter are handled through a sensitivity analysis as described later in this paper The results indicate that the timber covered bridges in Longnan exhibit a significantly clustered distribution, presenting a pattern of “dense in the southwest and sparse in the northeast”, with Wen County and Kang County as the core clustering areas. Temporally, they follow a unimodal evolution pattern: initiation in the Ming Dynasty, peak in the Qing Dynasty, decline in the Republic of China period, and near stagnation in modern times. The location and distribution of the covered bridges show a strong statistical association with natural conditions (e.g., topography, hydrology) and exhibit spatial coincidence with modern vegetation coverage—the latter treated solely as a contemporary context variable rather than a historical driver. Spatial coincidence with the ancient road is quantified (60.56% within a 2000 m buffer), while settlement proximity is only qualitatively noted as background. Socio-economic factors (e.g., population, transportation, and settlements) are examined qualitatively and display spatial coincidence rather than quantitatively measured influence; these factors cannot be directly compared with natural factors.