What drives the fractal evolution of urban road networks? Evidence from the Yangtze River Delta, China

Urban road networks serve as the structural backbone linking functional subsystems within cities, and their geometric forms and topological structures play a critical role in shaping transport efficiency and spatial organization. While the fractal characteristics of road networks have been widely recognized, few studies have systematically examined the evolution of geometric and structural fractality across temporal and spatial dimensions under multiple influencing factors. This study analyzes road network data from cities in the Yangtze River Delta (YRD) between 2014 and 2024. Six geometric and structural fractal dimensions are computed to reveal patterns of spatiotemporal variation and network evolution. A Geographically and Temporally Weighted Regression (GTWR) model is applied to identify the key drivers of fractal change and to explore their spatial heterogeneity. The results indicate that: (1) road networks in the YRD exhibit persistent fractal characteristics, with geometric fractal (GF) dimensions ranging from 0.714 to 2.181 and structural fractal (SF) dimensions from 2.037 to 4.519. The more rapid increase in SF dimensions suggests a transition from planar to increasingly volumetric spatial development; (2) Both GF and SF dimensions display a clear core-periphery pattern: GF increases more rapidly in inland areas while stabilizing in coastal cores, gradually converging toward the theoretical GF dimension of an ideal urban form(1.701). In contrast, the growth of SF dimensions shows less variation across the region, indicating minimal structural differentiation between core and peripheral areas; (3) Fractal evolution is jointly driven by terrain, infrastructure investment, and population dynamics, with pronounced spatial heterogeneity. Population density has a stronger influence in the southern cities of the YRD, while infrastructure investment and traffic demand follow a core-periphery decay pattern centered on Shanghai. This study offers new insights into the self-organizing nature of urban systems as complex spatial organisms and provides theoretical guidance for human-centered urban planning.