三类校园绿地暴露行为对大学生身心健康的

Effects of Three Types of Campus Green Space Exposure Behavior on Physical and Mental Health of College Students

关键词:  风景园林  绿地暴露  健康  行为轨迹实验  图像识别  结构方程模型  大学生

Key words:  landscape architecture  green space exposure  health  behavioral trajectory experiment  image recognition  structural equation modeling (SEM)  college student

摘要:

城市绿地对居民身心健康的恢复性健康效益已被普遍认可，而绿地暴露作为绿地健康效益发挥的重要环节，其评估标准尚未有效区分绿地供给与实际暴露行为。以校园绿地供给作为大学生绿地暴露行为发生的物质环境基础，通过问卷调查、轨迹实验和图像识别等方法，结合相关性分析、线性回归分析及结构方程模型，综合评估主动直接暴露、主动间接暴露和被动暴露3类绿地实际暴露行为对大学生身心健康的影响。结果显示，在绿地供给相同的情况下，3类绿地暴露特征普遍存在个体差异，其中，主动直接暴露和被动暴露特征均对健康特征有显著正影响，主动间接暴露特征也与健康特征存在关联。揭示了3类绿地实际暴露对身心健康的不同影响，丰富了绿地暴露与健康领域的研究成果，有助于进一步提出校园绿地规划建议及开展大学生健康行为指导，对健康人居环境建设具有一定借鉴价值。

Abstract:

Urban green spaces (UGS) are widely recognized for their restorative effects on human physical and mental health, yet existing assessments of green space exposure often fail to distinguish between green space supply (the physical environment) and actual exposure behaviors. This study focuses on college campuses as unique high-density communities where students' daily activities - including living, studying, dining, and commuting - are concentrated, ensuring a relatively consistent green space supply. It aims to systematically explore how three types of actual green space exposure behaviors (active direct, active indirect, and passive exposure) independently and collectively affect college students' physical and mental health. The research was conducted at the Minhang Campus of Shanghai Jiao Tong University, a 4,500-acre campus with diverse functional zones that meet most students' daily needs, housing over 45,000 full-time students (undergraduates, postgraduates, and international students). Data collection employed a mixed-methods approach: 1) Questionnaire surveys (294 valid responses) to gather demographic information (gender, age, major, residence type, etc.), active direct exposure characteristics (frequency and duration of activities such as viewing, sitting/reading, exercising - further categorized into mild, moderate, and vigorous - and socializing in green spaces), and health indicators (self-rated physical health and psychological distress measured by the Kessler Psychological Distress Scale, K6). 2) Trajectory experiments involving 89 volunteers, using the "Two Steps" app to record 7-day daily activity trajectories; combined with GIS and campus green space mapping, passive exposure was quantified as the ratio of green space within a 20-meter buffer of trajectories to the total buffer area. 3) Image recognition of 221 valid window-view photos (from questionnaire respondents) via the SceneParse150 model, which calculated visible green volume (sum of trees, shrubs, and grass) and weighted it by daily viewing duration to assess active indirect exposure. Statistical analyses, including Spearman correlation, multiple linear regression, and structural equation modeling (SEM), were applied to examine relationships between exposure behaviors and health outcomes. Key findings revealed significant individual differences in all three exposure types despite consistent green space supply. Correlation analysis showed that the frequency and duration of sitting/reading in green spaces were positively associated with selfrated physical health (R=0.190, P<0.01) and negatively associated with K6 scores (R=-0.150 to -0.180, P<0.05 to 0.01). Regression analysis indicated that longer duration of vigorous exercise positively affected self-rated physical health (B=0.122, P=0.045), while higher frequency of vigorous exercise had a negative effect (B=-0.230, P=0.035); longer sitting/reading duration significantly reduced psychological distress (B=-1.383, P=0.003). The SEM further demonstrated that active direct exposure (β=0.303) and passive exposure (β=0.280) had significant positive impacts on health, explaining 30.1% and 17.2% of health variance, respectively, while active indirect exposure showed associations with health despite no significant positive effects. Notably, perceived green space quality, though not directly affecting health, influenced active indirect and passive exposure (β=0.258 and 0.415, respectively). This study contributes to the literature by differentiating three distinct green space exposure behaviors and their unique health effects, providing empirical evidence for campus green space planning (e.g., enhancing functionality to encourage active use, optimizing roadside greenery for passive exposure) and health behavior guidance for students. Limitations include inconsistent sample sizes across exposure types and a focus on visual exposure (neglecting auditory or olfactory factors), suggesting future research could integrate multi-sensory measurements and virtual reality simulations to refine exposure assessment, thereby better supporting urban green space design and public health promotion.