Evaluation of individual thermal sensation at raised indoor temperatures based on skin temperature

In this study, we further investigate whether skin temperature can also assess individual thermal sensation as an effective physiological index while previous studies did not draw broadly generalizable conclusions at high temperature. The climate-controlled experiments include five temperature levels of 26, 30, 33, 37 and 39 °C, two relative humidity levels of 50% and 70%, and two activity levels with light and moderate and in which the estimated average metabolic rate were respectively 58 W/m2 and 165 W/m2. A total of 136 healthy subjects who lived in hot-humid climate for a long time participated in these experiments. During the experiments, the ratings of thermal sensation, comfort and acceptability were documented, local skin temperatures were continuously monitored, and five mean skin temperatures were calculated. Using Fisher discriminant analysis to model thermal sensation from skin temperature, then obtaining the thresholds of local and mean skin temperatures corresponding to different thermal sensation categories, and finally verifying the accuracy of the models assessment. The results show that the skin temperature can also predict thermal sensation at high temperatures of 30–39 °C beyond the range of normal temperatures. The proposed evaluation models show a high accuracy to predict the “hot/very hot” category, and up to 93% accuracy of predicting thermal sensation from wrist skin temperature. Additionally, thermal adaptation to high temperature and humidity can affect the prediction accuracy of “hot/very hot” category from skin temperature. The results of this study provide an important basis for early warning when human's safety risk occurs under high temperatures.