UC Irvine

Existing studies regarding green space and mental health were mainly with general population and relied on satellite-based imagery, without considering eye-level green space and vegetation types, which is important to elucidate the underlying mechanisms linking green space and health. To improve green space exposure assessment, a machine learning model was evaluated and applied to investigate the associations between street green space and socioeconomic factors. Microsoft Bing Maps images in conjunction with deep learning was used to measure street green space, which were compared to normalized difference vegetation index (NDVI) as commonly-used satellite-based green space measure. The results show that street view imagery coupled with deep learning can accurately and efficiently measure eye-level street green space and distinguish vegetation types (i.e., tree, low-lying vegetation, grass); street view data reflect different aspects of natural environments compared to satellite imagery. In Los Angeles County, lower socioeconomic status and racial/ethnic minority communities had substantively less street green space.

Relationships between green space and postpartum depression (PPD) has not been studied. I investigated the relationships between PPD and green space and examined the mediation effect of physical activity during pregnancy. Clinical data were obtained for 415,020 pregnancies in southern California (2008-2018) from Kaiser Permanente Southern California. PPD was based on both diagnostic codes and prescription medications. Multiple indicators were used to characterized green space exposure, including street view-based green space and vegetation types, satellite-based measures (i.e., NDVI, land-cover green space, and tree canopy cover), and proximity to the nearest park. The results indicate that street green space and tree coverage were associated with a decreased risk of PPD. Protection and restoration of trees may translate into a more pronounced reduction of PPD in southern California. Physical activity could be considered as one of the plausible pathways linking green space to depression (mediation effect: 9.6% -15.6%).

To further explore the underlying mechanism, an experimental study was conducted to examine physiological and affective responses to green space on stress recovery among pregnant women, using simulated green space exposure through virtual reality (VR). Participants (n=63) were randomly assigned to view one of three, 5-min, VR videos with different green space levels (i.e., low, moderate, and high) after a laboratory stressor, the Trier Social Stress Test. Physiological stress responses and self-reported affect were measured during the experiment. This study demonstrated that even a short immersion in VR green space environments, especially park-like setting, could effectively ease both physiological and affective stress and improve mental health during pregnancy.

This study contributes to the improvement of green space exposure assessment methodology for health studies, and provide evidence of the relationship between green space and maternal mental health during postpartum period, and potential pathways.