Impact of urban green spaces on the well-being and thermal comfort of pedestrians: a qualitative assessment integrated with thermography
Name: LUCIÉLIA LACERDA DA SILVA
Publication date: 10/11/2025
Examining board:
Name |
Role |
|---|---|
| CRISTIANE COELHO DE MOURA | Examinador Interno |
| FLÁVIO CIPRIANO DE ASIS DO CARMO | Examinador Externo |
| NILTON CESAR FIEDLER | Presidente |
| REJANE TAVARES BOTREL | Examinador Externo |
| RICARDO HIDEAKI MIYAJIMA | Examinador Externo |
Summary: This study aimed to investigate the influence of urban green spaces on well-being and thermal comfort. The range of surface and air temperatures, as well as thermal comfort, were analyzed in wooded and non-wooded scenarios. Furthermore, it sought to evaluate the possibility of estimating the Wet Bulb Globe Temperature (WBGT) index from surface temperature, as well as visualizing future climate change scenarios. The methodology included defining typical winter days, characterizing the sociodemographic profile of pedestrians, and assessing well-being using a scale of positive and negative affects associated with the Momentary Ecological Assessment method. Surface temperature was obtained by aerial survey with a thermal sensor at three times (6 am, 12 pm, and 4 pm) in wooded and non-wooded areas of central streets in the municipality of Jerônimo Monteiro (ES). Simultaneously, two thermal stress meters recorded data every five minutes, under shaded and sunny conditions, between 6 am and 4 pm. For the climate change analysis, the positive (RCP 4.5) and negative (RCP 8.5) scenarios were considered for the period from 2071 to 2100. The thermal comfort analysis followed Regulatory Standard No. 15 and Occupational Hygiene Standard No. 06, using multiple linear regression models and the Random Forest machine learning algorithm. The results indicated that well-being, assessed by the average of affects, did not differ significantly between times of day or between wooded and non-wooded areas. However, a more pronounced polarization between positive and negative affects was observed in the non-wooded scenario. In general, positive emotions predominated, especially "attentive," "determined," and "alert," while "guilty," "hostile," and "terrified" were rarely mentioned. Wooded areas showed a greater thermal amplitude, reaching 13.46 °C between shade and sun conditions, confirming their importance in regulating the urban microclimate. The WBGT index could be estimated from surface temperature: in multiple linear regression, the adjusted R² ranged from 0.238 to 0.962, while in Random Forest it reached 0.835 in wooded areas and 0.890 in non-wooded areas. Multiple linear regression presented limitations, requiring segmentation by hourly periods and separation between shade and sun to meet statistical assumptions. Air temperature modeling via multiple linear regression showed fit only for the non-wooded scenario (adjusted R² between 0.844 and 0.985). With Random Forest, the adjusted R² ranged from 0.800 to 0.949 between scenarios. Thus, it was possible to spatially represent, with high resolution (1 m²), the positive and negative climate change scenarios in the studied areas. Climate projections showed that maximum temperatures, in both scenarios, will occur at 4 PM. In the wooded area, the estimated maximum temperatures ranged from 27.30 to 34.62 °C in RCP 4.5 and from 29.73 to 37.05 °C in RCP 8.5. In the non-wooded area, the maximum temperatures were also concentrated at 4 PM, ranging from 23.09 to 26.40 °C (RCP 4.5) and from 26.95 to 30.26 °C (RCP 8.5).
Keywords: Ergonomics; Climate change; Urban afforestation; Quality of life.
