To better understand the influence of urban green infrastructure (UGI) on outdoor human thermal comfort, a survey and physical measurements were performed at the campus of the University of Groningen, The Netherlands, in spring and summer 2015. Three hundred eighty-nine respondents were interviewed in five different green spaces. We aimed to analyze people’s thermal comfort perception and preference in outdoor urban green spaces, and to specify the combined effects between the thermal environmental and personal factors. The results imply that non-physical environmental and subjective factors (e.g., natural view, quiet environment, and emotional background) were more important in perceiving comfort than the actual thermal conditions. By applying a linear regression and probit analysis, the comfort temperature was found to be 22.2 °C and the preferred temperature was at a surprisingly high 35.7 °C. This can be explained by the observation that most respondents, who live in temperate regions, have a natural tendency to describe their preferred state as “warmer” even when feeling “warm” already. Using the Kruskal-Wallis H test, the four significant factors influencing thermal comfort were people’s exposure time in green spaces, previous thermal environment and activity, and their thermal history. However, the effect of thermal history needs further investigation due to the unequal sample sizes of respondents from different climate regions. By providing evidence for the role of the objective and subjective factors on human thermal comfort, the relationship between UGI, microclimate, and thermal comfort can assist urban planning to make better use of green spaces for microclimate regulation.

In this study, we investigated the effects of trees on the local urban microclimate and human thermal comfort under different local weather conditions, in a small urban area in Assen, the Netherlands. In both summer and winter, continuous air temperature and relative humidity measurements were conducted at five selected sites having obviously different environmental characteristics in tree cover. Measurements demonstrated that in summer the microclimatic conditions at each observation site showed significant differences. The cooling effects of trees on clear and hot days were two times higher than on cloudy and cold days. In winter, air temperature was slightly reduced by the evergreen trees, and weather conditions did not cause a notable change on performance of trees on the microclimate. ENVI-met, a three-dimensional microclimate model was used to simulate the spatial distribution of temperature and humidity. After selecting representative days, we simulated the study site as it currently is and for a situation without trees. Spatial differences of trees’ effects were found to vary strongly with weather conditions. Furthermore, human thermal comfort is indicated by the Predicted Mean Vote model. During the hottest hours, trees improved the thermal comfort level via reducing ‘very hot’ and ‘hot’ thermal perception by about 16 % on clear days and 11 % on cloudy days. Generally, our findings demonstrate that urban microclimate and human thermal comfort convincingly varies in close geographical proximity. Both are strongly affected by the presence of local trees. Weather conditions play an important role on the trees’ performance on the summer-time microclimate.

This study analyzed how the variations of plant area index (PAI) and weather conditions alter the influence of urban green infrastructure (UGI) on microclimate. To observe how diverse UGIs affect the ambient microclimate through the seasons, microclimatic data were measured during the growing season at five sites in a local urban area in The Netherlands. Site A was located in an open space; sites B, C, and D were covered by different types and configurations of green infrastructure (grove, a single deciduous tree, and street trees, respectively); and site E was adjacent to buildings to study the effects of their façades on microclimate. Hemispherical photography and globe thermometers were used to quantify PAI and thermal comfort at both shaded and unshaded locations. The results showed that groves with high tree density (site B) have the strongest effect on microclimate conditions. Monthly variations in the differences of mean radiant temperature (∆Tmrt) between shaded and unshaded areas followed the same pattern as the PAI. Linear regression showed a significant positive correlation between PAI and ∆Tmrt. The difference of daily average air temperature (∆T _a ) between shaded and unshaded areas was also positively correlated to PAI, but with a slope coefficient below the measurement accuracy (±0.5 °C). This study showed that weather conditions can significantly impact the effectiveness of UGI in regulating microclimate. The results of this study can support the development of appropriate UGI measures to enhance thermal comfort in urban areas.