Psychological stress is a concern in highly urbanized cities, where urban greenspace may support mental well-being. Yet static measures of greenspace emphasize quantity and may miss how people encounter and perceive vegetation morphology during daily mobility. We combined GPS tracking, ecological momentary assessment (EMA), activity diaries and questionnaires from 738 participants in four Hong Kong communities, yielding 3871 assessments. Nine LiDAR-derived indicators characterized vegetation within 300-m buffers over the preceding 5 min. Mixed models examined associations with perceived green quality and momentary stress across activity contexts; participant-level models examined daytime exposure and stress symptoms. Greater vegetation amount, aggregation, edge density, shape complexity and height-class areas were associated with better perceived green quality, whereas patch density showed the opposite pattern. Better perceived green quality was associated with lower momentary stress within and between participants, with small indirect associations involving vegetation structure. Activity comparisons identified contrasting experiences at work and leisure and variation in the patch-density–quality association that depended on model specification. Comparable indirect associations were not evident for participant-level stress symptoms. These findings support combining vegetation structure with perceived quality to characterize greenspace exposure during daily mobility.
Urban expansion and soil sealing are major drivers of biodiversity loss, especially in cities where suitable habitats are scarce. Nature-based solutions (NBS), such as soil unsealing, are emerging as promising alternatives and have attracted growing interest. This study investigates whether small unsealed plots (20 cm in diameter), created through citizen-led micro-greening initiatives (MIGs) in Montpellier (France) can support ant communities. A total of 335 MIGs, ranging from 0 to 5 years old, were surveyed using baiting methods to sample ant populations. Thirteen ant species were identified, with Pheidole pallidula and Lasius group niger being the most common. Ants were detected in 24% of the MIGs, although species richness per site was generally low, likely constrained by the small plot size. While neither MIG age nor surrounding vegetation significantly influenced ant abundance, the presence of a plant within the plot significantly increased the probability of detecting ants. These findings suggest that ants can be present in very small unsealed spaces, particularly when vegetation is present, with no clear increase in ant presence beyond the first year after unsealing. Our results highlight the potential of micro-scale interventions to contribute to urban biodiversity. However, it remains unclear whether these microhabitats serve as nesting sites or are used primarily for foraging. Further research is needed to assess long-term colonization dynamics and the ecological role of these newly created habitats.
