Audiovisual Environmental Characteristics and Tourist Loyalty in Urban Waterfronts: Implications for Socially Sustainable Design

Urban waterfronts are vital public spaces that contribute to urban sustainability by providing residents with opportunities for recreation, social interaction, and nature experiences. Understanding user perceptions in these environments is essential for evidence-based design. Taking Taiyuan Fenhe Park in China as a case with local residents as respondents, this study investigated how objective audiovisual characteristics are associated with tourist loyalty through perceptual dimensions, while also examining interactive associations between visual and auditory elements. Data were collected at 539 spatial samples spaced at five-minute walking intervals. Methods included on-site acoustic measurements, panoramic imaging, computer-based visual and auditory quantification, and questionnaire surveys, yielding a total of 1768 valid responses. Visual features were quantified using semantic segmentation, object detection, and depth prediction, whereas the auditory environment was characterized by physical acoustic and psychoacoustic indicators. Three perceptual dimensions—environmental restorativeness (ERS), spatial vitality (SVS), and environmental controllability (ECS)—were extracted and tested as mediators within the stimulus–organism–response (S-O-R) framework. Results indicated that ERS, SVS, and ECS function as three parallel mediating constructs in the statistical model, with SVS showing the strongest statistical association with tourist loyalty. In addition, fluctuation strength exhibited a significant direct effect on tourist loyalty independent of these three perceptual dimensions. A total of 17 significant audiovisual interactions were identified, revealing both synergistic and antagonistic effects. These findings contribute to theoretical frameworks of multisensory integration and provide practical guidance for sustainable waterfront design. Specifically, zoning strategies and carefully selected audiovisual combinations are relevant to enhanced user experience and may contribute to long-term social well-being.