Urban park qualities driving visitors mental well-being and wildlife conservation in a Neotropical megacity

Abstract

Green infrastructure has been widely recognized for the benefits to human health and biodiversity conservation. However, knowledge of the qualities and requirements of such spaces and structures for the effective delivery of the range of ecosystem services expected is still limited, as well as the identification of trade-offs between services. In this study, we apply the One Health approach in the context of green spaces to investigate how urban park characteristics affect human mental health and wildlife support outcomes and identify synergies and trade-offs between these dimensions. Here we show that perceived restorativeness of park users varies significantly across sites and is mainly affected by safety and naturalness perceptions. In turn, these perceptions are driven by objective indicators of quality, such as maintenance of facilities and vegetation structure, and subjective estimations of biodiversity levels. The presence of water bodies benefited both mental health and wildlife. However, high tree canopy coverage provided greater restoration potential whereas a certain level of habitat heterogeneity was important to support a wider range of bird species requirements. To reconcile human and wildlife needs in green spaces, cities should strategically implement a heterogeneous green infrastructure network that considers trade-offs and maximizes synergies between these dimensions.

Figure 1

Theoretical framework depicting the relationships investigated in this study. Modified from Felappi et al.. Colors refer to the three steps of the analytical strategy: dark blue—first, green—second, and light blue—third step. A complete description of factors and interlinkages can be found in the original paper.

Figure 2

Adjusted predictions of perceived restorativeness scores (PRS) for each park and 95% confidence intervals. The dotted line represents the mean perceived restorativeness score of the total sample. The pictures provide an idea of the different parks’ configuration and represent Alfredo Volpi (upper right), Jardim da Luz (middle), and Rio Verde (bottom).

Figure 3

Path diagram of the perceptions and control variables affecting perceived restorativeness (full model, model 1 in Supplementary Table 1) with standardized coefficients. Dotted lines depicture pathways that were statistically non-significant (p > 0.05). Regression estimates within the measurement models were omitted in this picture for simplification.

Figure 4

Standardized coefficients and 95% confidence interval of each variable included in the final models of setting perceptions (naturalness, management, and soundscape) and safety perception. The plots represent the models with objectives variables only (models 1, left side), and the models replacing biodiversity perception in the setting perceptions and including control variables in safety perception (models 2, right side). Variables with confidence intervals (bars) that do not cross 0 are statistically significant. See Table 2 for definition of variables.

Figure 5

Standardized coefficients and 95% confidence interval of each variable included in the final models of bird community support (species richness, diversity, and urbanity). Variables with confidence intervals (bars) that do not cross 0 are statistically significant. See Table 2 for definition of variables.

Figure 6

Relationship between proportion of tree canopy (Prop canopy) and urbanity index in the presence or absence of understory vegetation. Shaded area represents the 95% confidence interval.

Figure 7

Maps showing the (a) Location of the city of São Paulo (yellow triangle) within the state of São Paulo (dark gray) and Brazil. (b) Land use map of São Paulo municipality showing the distribution of vegetation and green areas in relation to the build area, with study areas and their 1 km buffers in detail (c). Land use data downloaded from GeoSampa platform (https://geosampa.prefeitura.sp.gov.br/).