Whenever people spend time outdoors during hot weather, they are putting themselves in potentially stressful situations. Being able to predict whether a person is overheating can be critical in preventing heat-health issues. There is a clear relationship between body core temperature and heat health. However, measuring body core temperature is expensive. Identifying a non-invasive measure that could indicate a person's thermal strain would be valuable. This study investigated five physiological measures as possible surrogates: finger mean skin temperature (FSKT), finger maximum skin temperature (FMSKT), skin conductance level (SCL), heart rate (HR), and heart rate variability (HRV). Furthermore, they were compared against the results of participants' subjective thermal sensation and thermal comfort in a range of hot microclimatic conditions in a hot and humid climate. Results showed that except for SCL, each of the other four physiological measures had a positive significant relationship with thermal sensation, but a negative relationship with thermal comfort. Furthermore, through testing by cumulative link mixed models, HRV was found to be the most suitable surrogate for predicting thermal sensation and thermal comfort through a simple, non-invasive measure in outdoor environment in summer in a hot and humid area. This study highlights the method for predicting human thermal strain and contributes to improve the public health and well-being of urban dwellers in outdoor environments.
Keywords: cumulative link mixed model; heart rate variability; hot and humid climate area; non-invasive measure; outdoor environment; thermal strain prediction.