Adverse health impacts of extreme heat exposure are expected to rise globally due to a warming climate, urban-induced warming, and a growing and aging population. The concerns for human health, productivity, and well-being are greater in humid climates and for vulnerable populations, such as older adults, unhoused, and/or those with chronic diseases. Therefore, robust models to assess current heat-health impacts and project future risks must incorporate specific vulnerabilities and diverse environmental contexts. Methods to project future heat stress risk can be broadly categorized into epidemiology/econometric and physiology-based approaches, which have contrasting benefits and limitations. Epidemiology/econometric approaches are empirical in nature, analyzing time series of historical temperature paired with particular health consequences (e.g., morbidity or mortality) across populations to determine heat-health relationships. These studies often find higher rates of cardiovascular and respiratory deaths associated with high ambient temperatures. Future health burdens from heat can be estimated by applying these relationships to climate model outputs (i.e., daily temperature) under different warming scenarios. Empirical approaches are based on real-life outcomes and the range of realistic living conditions, and they can explore the cumulative effects of exposures over multiple days. However, two limitations for climate change projections include (1) assumptions needed to extrapolate results to warmer temperatures than observed in the historical sample and (2) ambiguity regarding the role of humidity in heat-health outcomes. While some epidemiological studies find a relationship between mortality in the heat and humidity, most find minimal associations between humidity and heat-health outcomes. Given that specific humidity is robustly expected to increase with global warming, this uncertainty is a key research gap for epidemiology-based projections of future heat stress. Physiology-based studies of future heat stress risk employ relationships between the thermal environment and health outcomes based on human energy balance considerations, with parameters constrained by studies of physiologic processes. In contrast to epidemiology/econometric approaches, physiology-based studies of heat health outcomes consistently find a robust role of atmospheric humidity in heat stress via its modulation of evaporative cooling from sweat. However, physiology studies are limited in not directly observing health outcomes, such as hospitalization or death, and employ idealized conditions from thermal chamber studies. A range of physiology-based metrics has been applied to project future heat stress. Sherwood & Huber introduced a 35 °C wet bulb temperature (Tw) threshold that would result in death after 6 h of exposure and applied this threshold to project future adaptability limits under varying levels of warming. Since then, numerous studies have used this approach, wherein a psychometric Tw of 35 °C assumes death. The Tw of 35 °C represents a thermodynamic limit to heat exchange, whereby the human body becomes an adiabatic system, assuming the person is indoors or shaded, unclothed, completely sedentary, fully heat acclimatized, and of average size without thermoregulatory impairments. As an example of a different metric, Dunne et al. estimated future reductions in labor capacity under different warming scenarios using established guidelines for physical labor under different wet bulb globe temperature levels.While these studies incorporate valuable information about humidity and physiology more realistically than epidemiological studies, their thermal physiology theory remains relatively unsophisticated. These approaches cannot capture complexities and personal characteristics affecting human thermoregulation (e.g., body size, activity levels, clothing, or physiological restrictions-such as sweating-to thermoregulation), which may cause substantial errors.To be useful, heat-health projections should realistically account for factors that increase health risks, such as individual physical characteristics and physiological impairments, as well as interventions that modify or decrease impacts (e.g., lowering metabolic rate; behaviors to reduce exposures). Moreover, models should incorporate ranges in environmental parameters that, together with temperature, result in specific thermoregulatory effects (e.g., dry, humid, sun/shade, windspeed). Physiological and biophysical models offer new opportunities to assess how humans might live and work in a warmer future rather than merely determining the prospects for life and death. Here, we demonstrate a unique approach using physiological principles that align with human thermal responses to heat (e.g., heat strain) to overcome simplified approaches that miss essential physiologic and behavioral factors in the heat.

Existing literature on heat stress and human health uses two main approaches: epidemiological/econometric and physiology-based. Epidemiological studies analyze historical temperature data alongside health outcomes to establish correlations but face limitations in extrapolating to future, warmer temperatures and in understanding humidity's role. Physiology-based studies utilize human energy balance principles, consistently highlighting humidity's impact through evaporative cooling, but lack direct observation of health outcomes and often use idealized conditions. The 35°C wet-bulb temperature (Tw) threshold, introduced by Sherwood & Huber, has been widely used to project future survivability limits. However, this approach simplifies human thermoregulation, omitting crucial factors like individual characteristics (body size, activity level, clothing, physiological impairments), and behavioral adaptations (seeking shade, reducing activity). Other studies, such as Dunne et al.'s work on labor capacity, offer a more nuanced understanding but still lack the physiological detail needed for accurate predictions across diverse populations and environmental conditions.

This study uses a whole-body human heat exchange model based on partitional calorimetry to estimate heat stroke deaths (survivability) and maximum safe sustained physical activity (liveability). The model incorporates established physiological principles and accounts for factors like age, body size, sweat rate, sun exposure, and humidity. It assesses survivability for 3- and 6-hour exposure windows, aligning with climate model outputs and previous survivability studies. Liveability is defined as the maximum metabolic rate (Mmax) sustainable without continuous heat storage, expressed in metabolic equivalents (METs). The model was applied to two age groups (younger, 18–30 years, and older, >65 years, female adults) under various air temperature (Tair) and relative humidity (RH) conditions, both in shade and sun. Survivability was determined by whether core body temperature (Tcore) exceeded 43°C within the exposure window. Liveability was determined by calculating Mmax under different environmental conditions. Global climate model (GCM) data (GFDL ESM4) under current (2016–2025) and projected (2091–2100) climates (SSP2-4.5 and SSP5-8.5 scenarios) were used to estimate future liveability. The model considers three factors that restrict evaporative heat loss: high environmental humidity, limited maximum skin wettedness, and maximum sweat production rate. The survivability algorithm categorizes outcomes based on whether a person survives within sweating limits, survives despite exceeding limits, or fails to survive due to environmental or physiological limitations. Liveability calculations determine the maximum sustainable activity level without unchecked core temperature increases.

The study's key findings demonstrate significant limitations in the commonly used 35°C Tw threshold for predicting survivability. The physiological model revealed substantially lower survivability limits, particularly in hot-dry conditions. For 6-hour exposures, the difference between the physiological model's critical Tw and the 35°C threshold ranged from -0.9°C (young adults, humid conditions) to -13.1°C (older adults, dry conditions). Liveability assessments showed that Mmax decreases sharply with increasing air temperature and humidity, and dramatically with age. Under shaded conditions at 25°C, young adults could sustain approximately 5.0 METs of activity in humid conditions and up to 8.4 METs in dry conditions. Older adults, however, showed significantly reduced Mmax, around 2.0–2.5 METs less than young adults. Sun exposure further reduced Mmax. Global climate projections using the liveability model suggested that under SSP2-4.5, median Mmax reductions will be modest (~0.25 METs) for young adults during warm conditions; however, under SSP5-8.5 this more than doubles (~0.64 METs). Significant declines in Mmax were projected by the end of the century, particularly in already densely populated and heat-vulnerable regions. The model revealed that aging impacts safe activity levels more profoundly than projected warming. Several locations already experience a high frequency of conditions where older adults can only perform low-intensity tasks; climate change is expected to exacerbate this, making many areas survivable but not liveable (i.e., capable of supporting human life but restricting any activity above rest).

This study's findings challenge the widely used 35°C Tw survivability threshold, highlighting its limitations in accounting for human physiological diversity and environmental complexities. The physiological model presented here provides a more accurate and nuanced assessment of survivability and liveability under extreme heat, particularly in dry conditions and for older adults. The model's projections of future liveability demonstrate that climate change, coupled with an aging global population, poses substantial challenges, especially in already densely populated, heat-vulnerable regions. The more pronounced decline in liveability with aging compared to warming underscores the urgency of developing targeted adaptation strategies for vulnerable populations. The study acknowledges limitations such as its focus on healthy female adults, the use of a single GCM, and the steady-state nature of the model. Further research is needed to incorporate these factors and improve model accuracy. Nevertheless, this work represents a significant advancement in heat stress modeling, offering a more robust and comprehensive approach to assess future human health risks.

This research introduces a novel physiological model for assessing human survivability and liveability under extreme heat, surpassing the limitations of the commonly used 35°C wet-bulb temperature threshold. The model demonstrates significant underestimation of heat risks by the 35°C threshold, particularly in dry conditions and for older adults. Future climate projections using this model highlight substantial reductions in liveability, particularly in already vulnerable regions. Aging is shown to have a greater impact than warming alone. The findings emphasize the necessity for focused adaptation strategies to manage increasing heat risks in vulnerable populations.

The current study has several limitations. First, it focuses on two specific age groups of healthy female adults, limiting the generalizability to other demographics, especially males and individuals with pre-existing conditions. Second, it utilizes only one GCM and emission scenario, potentially underrepresenting the range of possible future climates. Third, the model is based on steady-state conditions, neglecting the dynamic aspects of heat stress and human thermoregulation over time. Future research should address these limitations to enhance the robustness and predictive capability of the model.