| 摘要: |
| 随着城市步入高品质发展阶段，人们对于能够带来丰富感知体验的城市景观需求日益增加，学界愈发关注人与景观空间互动中产生的主观体验。既有景
观空间评价对人的主观反馈关注不足，且对快速路沿线住区这一特殊空间关注较少，导致对于全面优化城市居住环境的指导作用有限。视觉和听觉作为人们最主要
的环境感知途径，厘清二者与景观空间体验的作用机制，可为优化快速路沿线住区景观空间及提升居民感知提供新的视角。选取上海市快速路沿线住区为研究对
象，构建包含视觉环境特征、听觉环境特征、声源类型、视觉感知和听觉感知5类共20项指标，探索视听环境要素与主观感知之间的量化关系，以及视听环境要素
对主观感知的影响强度及机制。结果显示，视觉与听觉感知受对应的视、听环境特征影响，并且不同感知之间也存在交互影响作用。在揭示快速路沿线住区视听感
知影响机制的基础上，提出利用视听交互影响提升居民景观感知体验的思路，为优化城市快速路沿线住区景观环境品质提供科学指导 |
| 关键词: 风景园林 景观感知 交通噪声 回归分析 住区景观 城市快速路 |
| DOI：10.19775/j.cla.2025.01.0055 |
| 投稿时间：2024-07-30修订日期：2024-11-12 |
| 基金项目:十四五国家重点研发计划(2023YFC3805303)；上海市自然科学基金项目(24ZR1469800)；北京大学-林肯研究院城市发展与土地政策研究中心年
度研究基金项目(FS04-20141001-JTF) |
|
| Research on the Perception and Influence Mechanism of Landscape Audio-visual EnvironmentElements in Residential Areas Along Urban Expressways: Take Shanghai as an Example |
| SHAO Yuhan,ZHANG Zhoutao,JIN Tongfei* |
| Abstract: |
| In the current phase of high-quality urban development, there is a
growing demand for urban landscapes that offer dynamic and multi-sensory
experiences. Landscape perception has consequently emerged as a key research
focus within landscape architecture. However, past studies on landscape evaluation
have often lacked sufficient attention to subjective human feedback, while also
largely overlooking unique yet impactful spaces such as residential areas bordering
expressways. As two primary channels through which people perceive their
surroundings, vision and hearing are fundamental to understanding urban landscape
experiences. Despite considerable research on these perceptual mechanisms, gaps
remain, particularly in developing quantitative analyses of the combined effects of
visual and auditory stimuli on landscape perception and in examining the interactive
and synergistic relationship between these senses. Addressing these gaps, this study
investigates the residential area along Zhangwu Road in Shanghai's Yangpu District,
a site situated immediately north of the elevated Middle Ring Road and therefore
subjected to a highly variable outdoor soundscape significantly impacted by traffic
noise. This research establishes a comprehensive framework that includes five main
indicator categories - comprising 20 individual indicators - spanning audiovisual
environmental characteristics, sound source types, and perceptual dimensions. By
setting objective audiovisual environmental elements as independent variables
and landscape perception outcomes as dependent variables, this study aims to
systematically analyze the intricate interplay between landscape perception and
environmental features, yielding insights that are specific to the expressway-adjacent
residential context. To ensure a robust data collection process, 13 measurement points
were selected across the study area. Each point was carefully chosen to represent
diverse landscape environments, capturing a range of audiovisual characteristics
essential to a comprehensive evaluation. Data collection included fixed-location
panoramic photographs and sound recordings. The panoramic images underwent
semantic segmentation analysis, and the resulting data were processed within a GIS
platform to calculate the percentage of different landscape elements within each
frame. Sound measurements were conducted using professional recording equipment
to capture continuous audio data, which was later analyzed to extract key acoustic
metrics. Additionally, 37 participants, comprising residents from Tongji Xincun
and graduate students from Tongji University, engaged in immersive soundwalk
experiments at each measurement point. After each soundwalk, participants completed
a 7-point Likert scale questionnaire assessing audiovisual perceptions. These responses
were then screened to ensure validity and the refined data was utilized for subsequent
analyses. Results from this study illuminate key features of the landscape environment.
Auditory analysis revealed that over 70% of sites exhibited sound pressure levels well
above recommended thresholds, with a marked negative correlation between sound
pressure levels and the distance from the expressway, indicating that traffic noise was
the primary source of disturbance. Visually, a high percentage of vegetation coverage
was observed, suggesting favorable greenery visibility, which likely contributes to
perceived aesthetic quality. In an integrated assessment of audiovisual perception, a
notable inverse relationship was found between auditory annoyance and pleasure; this
relationship aligned with sound pressure levels and the presence of natural sounds,
underscoring the adverse impact of high-pressure traffic noise and the beneficial
influence of natural sounds on the soundscape experience. Visual perception results
indicated consistent overall satisfaction across different points, with no clear dominant
trends among individual visual indicators, suggesting that various visual landscape
elements may not exhibit inherent superiority or inferiority in terms of visual
appeal. Further correlation analysis between environmental factors and audiovisual
perception highlighted significant associations within auditory perception indicators
and between these indicators and various environmental elements. Visual features,
notably vegetation, were also found to correlate with auditory perception levels.
Strong internal correlations were observed within the realm of visual perception,
with each visual perception indicator significantly associated with specific visual
environmental factors. Through audiovisual perception modeling, the study developed
regression models for visual satisfaction and its sub-indicators, as well as auditory
satisfaction and associated sub-indicators. These models were validated for robustness
and revealed that vegetation had the most substantial positive effect on visual
satisfaction, followed by the sky, while buildings and ground surfaces contributed
negatively. Natural sounds were shown to have a significant positive effect on auditory
perception, while traffic noise exerted a negative influence mainly due to high sound
pressure levels; vegetation additionally contributed positively to auditory satisfaction.
Cross-modal perception analysis unrevealed that natural sounds and sounds associated
with human activities had a positive influence on visual perception, while traffic noise
and other disruptive sounds had detrimental effects. Auditory and visual perceptions -
except for annoyance - tended to reinforce one another, with pleasure and naturalness
showing the highest correlation. Spaciousness, however, was not significantly
impacted by auditory perceptions. In summary, this study provides a detailed
exploration of the various factors influencing audiovisual perception and offers an indepth
analysis of the cross-modal interactions at play. Despite its limitations, such
as a relatively homogeneous sample and the challenges of controlling environmental
variables, this research offers valuable scientific insights and practical guidelines for
enhancing landscape design in residential areas adjacent to expressways. The findings
underscore the importance of integrating both visual and auditory considerations in
urban planning to create more harmonious and health-promoting environments for
residents in these distinctive urban contexts. |
| Key words: landscape architecture landscape perception traffic noise regression
analysis residential landscape urban expressway |
