Comparative analysis of thermoregulation models to assess heat strain in moderate to extreme heat

Abstract

As global temperatures rise due to climate change, the frequency and intensity of heatwaves are increasing, posing significant threats to human health, productivity, and well-being. Thermoregulation models are important tools for quantifying the risk of extreme heat, providing insights into physiological strain indicators such as core and skin temperatures, sweat rates, and thermal comfort levels. This study evaluated four thermoregulation models of varying complexity, differentiated by the geometry and underlying thermoregulatory mechanisms. The models assessed include the Gagge two-node model, the Stolwijk-1971 model, the JOS3 model, and the UTCI-Fiala model. Additionally, we introduce the Stolwijk-2024 model, a modified version of the original Stolwijk model, which incorporates updated empirical coefficients derived from recent studies while retaining the original framework. The models were tested against human trial data across a wide range of extreme heat exposures, including transient extreme heat, humid heat, various physical activity levels, and clothing insulation scenarios. Our findings demonstrate that multi-node and multi-segment models, such as JOS3, UTCI-Fiala, and Stolwijk-2024, reliably predict core (average RMSD: <0.3 °C) and skin (average root-mean-square deviation, RMSD: <0.6 °C) temperatures, making them suitable for assessing heat strain and thermal comfort in moderate to extreme environmental conditions. In contrast, simpler models like the single-segment, two-node Gagge's model performed poorly in predicting core temperature under conditions involving high metabolic rates (>3.75 met) in moderate to hot environments (>35 °C), with an average RMSD of 1.2 °C. Similarly, the Stolwijk-1971 model showed a systematic bias (∼0.45 °C), underpredicting core temperatures during high metabolic rates. This study underscores the robustness and applicability of open-source models like JOS3 and Stolwijk-2024 in public health, urban design, and climate impact research, highlighting their potential to improve our understanding of heat strain and thermal comfort in the context of a warming climate.