Multi-domain human-oriented approach to evaluate human comfort in outdoor environments.

Human comfort outdoors is widely investigated, but most studies explore the comfort domains singularly. This paper aimed to evaluate human comfort in parks, verifying the importance of using a multi-domain (simultaneously evaluating thermal, visual, acoustic, and air quality) and multi-disciplinary (combining environmental and social fields) approach. A walk through a pre-defined path from one park to another was repeated twice per day on four consecutive days in June, with three participants per walk. The two investigated parks are in central Italy and were chosen because they differ in their design and spatial characteristics. Environmental data were recorded with an innovative wearable device during the whole walk, and surveys were used to assess people's perceptions of the parks. Despite observed differences in collected physical parameters, the survey's responses were similar, and different comfort domains showed dependence on each other in the two parks. Logistic regression models were developed for each park, and they revealed that the qualitative information predicted the overall comfort level more accurately than the environmental data. In detail, the models based on environmental data resulted in R 2 equal to 0.126 and 0.111 in Parks 1 and 2, respectively, whereas using the survey answers increased it up to 0.820 (Park 1) and 0.806 (Park 2). This study contributes to addressing the gap in multi-domain comfort studies outdoors and confirms the importance of using multi-disciplinary and multi-domain approaches for a complete comfort analysis, supporting holistic human-biometeorology-oriented models and forecasting opportunities that can promote improvements in urban environmental quality and liveability.