Qinqin Kong

← All research projects

Where/When Heat Outpaces the Human Body:
From Lab Experiments to Global Exposure Maps

The human body can tolerate only a limited range of heat. As the planet warms, understanding where, when, and how frequently human heat-tolerance limits will be breached becomes increasingly important.

Imagine being in a room with a comfortable temperature and humidity—you feel fine. Now imagine the room gradually becoming hotter and more humid. At first, your body can still maintain a stable, normal core temperature. But eventually, you reach a point where the heat and humidity overwhelm your body's ability to regulate its temperature, and your core temperature begins to rise. This marks the onset of so-called uncompensable heat. Prolonged exposure to uncompensable heat can eventually lead to heatstroke.

Our collaborators at Penn State (the PSU HEAT Project) use this same idea to experimentally identify the critical limits of uncompensable heat by placing people of different ages in climate chambers.

We applied these experimentally measured limits to future climate projections and asked: where, when, and how frequently will people be exposed to uncompensable heat as the climate warms?

In our PNAS paper, we started with the most optimistic assumption: that everyone can tolerate heat as well as healthy young adults. Even under this optimistic benchmark, uncompensable heat emerges across increasingly large parts of the world as warming increases, placing billions of people at risk of exceeding human physiological limits.

However, heat tolerance is not the same for everyone. Older adults sweat less efficiently and experience greater cardiovascular strain, so their physiological heat-tolerance limits are substantially lower than those of young adults. In our Lancet Planetary Health paper, we combined laboratory-measured thresholds for young, middle-aged, and older adults with bias-corrected climate projections and age-stratified population projections to build the first age-resolved map of future uncompensable heat exposure. Accounting for age increases the estimated number of people facing repeated uncompensable heat exposure by 2 to 8 times compared with estimates based on the conventional, one-size-fits-all young-adult threshold. Strikingly, older adults face more widespread and frequent uncompensable heat at just 1.5°C of warming than young adults would at 4°C.

Together, these studies reveal a collision between a warming climate and the physiological limits of the human body—and that collision will arrive much sooner for some people than for others. Population aging compounds the threat: the people with the lowest heat tolerance are becoming a larger share of the population just as extreme heat is becoming more intense and widespread.

Related publications

1. Kong, Q., Vecellio, D. J., Huber, M., Cottle, R. M., Wolf, S. T., Leach, O. K., Fisher, K. G., Bendavid, E. & Kenney, W. L. (2026). Exceeding human heat tolerance in a warming, ageing world: a global projection modelling study. The Lancet Planetary Health.
2. Vecellio, D. J., Kong, Q. (co-leading author), Kenney, W. L. & Huber, M. (2023). Greatly enhanced risk to humans as a consequence of empirically determined lower moist heat stress tolerance. Proceedings of the National Academy of Sciences.

← All research projects

All About Wet-Bulb Globe Temperature

Wet-bulb globe temperature (WBGT) is one of the most widely used measures of heat stress. It combines the effects of temperature, humidity, wind, and radiation, and underlies heat-safety guidelines for workers, athletes, and military personnel.

Despite its long history and widespread use, WBGT is surprisingly often miscalculated in the academic literature. Many studies rely on simplified approximations that can introduce substantial biases—some of which arguably should not even be called WBGT because they omit solar radiation, wind speed, or both.

Even when WBGT is calculated properly, it remains something of a black box. Its meteorological inputs interact in complex, nonlinear ways, making it difficult to understand what actually drives a change in WBGT. We have analytic theories for understanding the physical drivers of temperature changes; no equivalent theory exists for WBGT.

Together with my PhD advisor, Matthew Huber, I have worked on three papers that tackle these problems from calculation to physical understanding.

In the Earth's Future paper, we developed an accurate, efficient, open-source Python implementation of the Liljegren method— the physics-based approach widely regarded as the gold standard for calculating WBGT—and used it to show how strongly commonly used simplified approximations can bias WBGT estimates.

After that paper came out, we noticed that simplified WBGT formulas continued to be widely used despite their known biases. One practical reason may be that even our efficient implementation of the Liljegren method requires iterative numerical calculations, and people don't like iterations. So, in the GeoHealth paper, we derived a zero-iteration, analytic implementation of WBGT that is much simpler and faster to calculate while remaining reasonably accurate relative to the full Liljegren method.

The success of this analytic approximation led us to a more fundamental question: to what extent do the nonlinearities and interactions in the WBGT formulation actually matter? If changes in WBGT can be approximated as a linear combination of changes in its meteorological drivers, WBGT becomes much easier to understand—and, importantly, amenable to a quantitative theory of what drives its changes.

In the JGR: Atmospheres paper, we developed a linearized sensitivity framework that decomposes changes in WBGT into contributions from changes in its individual meteorological drivers, such as temperature, humidity, wind, and radiation. We are not proposing a new physical theory of WBGT itself. Instead, we derive simple, mathematically tractable relationships that connect WBGT to the meteorological variables that control it, allowing existing physical theories for those variables to be carried over to WBGT.

Together, these three papers aim to make WBGT easier to calculate correctly—and easier to understand physically.

Related publications

1. Kong, Q. & Huber, M. (2022). Explicit calculations of wet-bulb globe temperature compared with approximations and why it matters for labor productivity. Earth's Future.
2. Kong, Q. & Huber, M. (2024). A new, zero-iteration analytic implementation of wet-bulb globe temperature: development, validation, and comparison with other methods. GeoHealth.
3. Kong, Q. & Huber, M. (2025). A linear sensitivity framework to understand the drivers of the wet-bulb globe temperature changes. Journal of Geophysical Research: Atmospheres.

← All research projects

Does Wet Soil Reduce or Amplify Human Heat Stress?

By adding more water to the land (e.g., irrigation), will we feel hotter or cooler? The air temperature (at least in the daytime) will drop because of the “evaporative cooling” effect (evaporation uses energy that would otherwise heat the air). At the same time, evaporation pumps moisture into the air, making the air more humid.

So we get cooler but more humid air. Which effect wins? Do we actually experience less heat stress?

This question matters because evaporative cooling is widely used as a heat-mitigation strategy, from irrigation to urban green and blue infrastructure. Yet whether and to what extent these strategies actually reduce human heat stress remains unclear.

We found that the answer depends strongly on where you are. Using ERA5 reanalysis data, we showed that the effect of wetter soil on heat stress depends on the local evaporative regime—in simple terms, whether evaporation is limited mainly by a lack of water in the soil or by a lack of available energy such as sunlight. In water-limited regions, adding soil moisture can actually increase heat stress, despite lowering air temperature. In energy-limited regions, wetter soil tends to be associated with lower heat stress.

A pot of water on a stove provides a useful analogy. Think of the water in the pot as the heat and moisture accumulating in the lower atmosphere. In a moisture-limited regime, wetter soil shifts more surface energy into evaporation and suppresses the growth of the atmospheric boundary layer— the layer of air near the ground in which that heat and moisture are mixed. It is as if the pot becomes smaller: temperature and humidity are easier to build up in a smaller volume.

In an energy-limited regime, by contrast, wetter soil often comes with less incoming solar energy. Here, it is more like turning down the fire beneath the pot: there is less energy available to drive evaporation and heat the lower atmosphere, so heat stress tends to decrease.

There is, however, an important caveat: the answer depends on how we measure heat stress. Different heat-stress indices give different weight to temperature and humidity. Our study used wet-bulb temperature, which is particularly sensitive to humidity; an index that gives humidity less weight can produce a different answer. This leads to a more fundamental question that we explore in another project: How “humid” should humid heat be?

The takeaway is that “cooler” does not necessarily mean “less heat stress.” The effectiveness of irrigation and other evaporative cooling strategies depends on the local land–atmosphere regime—and on how strongly humidity actually contributes to human heat stress.

Related publications

1. Kong, Q. & Huber, M. (2023). Regimes of soil moisture–wet-bulb temperature coupling with relevance to moist heat stress. Journal of Climate.

← All research projects

Building Open Data Infrastructure for Heat Research

Earth science fundamentally depends on data. I believe that building reliable, accessible datasets is an important part of doing good science—not just a by-product of individual research projects.

Over the course of my research, I have accumulated a growing collection of datasets that may be useful to the broader community. These include hourly heat-stress metrics derived from ERA5 reanalysis; raw and bias-corrected CMIP6 projections of heat stress under future warming and different SSP scenarios; and datasets translating heat exposure into impacts such as labor productivity loss. I am working to organize and document these datasets and am happy to make them publicly available whenever possible.

One part of this effort has been formally published as a Scientific Data descriptor: a global, high-resolution, bias-corrected dataset of future heat stress. We calculated wet-bulb temperature, and WBGT from 16 CMIP6 climate models and bias-corrected the projections against ERA5 reanalysis. The dataset spans 1°C to 4°C of global warming in 0.5°C increments, at 0.25°×0.25° spatial resolution and 3-hourly temporal resolution.

The high temporal resolution is intentional. Heat impacts often depend on short-lived extremes and nonlinear thresholds that can be obscured by daily or monthly averages. Likewise, systematic climate-model biases can matter greatly when asking whether a physiological or occupational threshold is crossed. Our bias-correction approach combines the fine-scale baseline climate represented by ERA5 with the climate-change signals projected by individual CMIP6 models, and substantially reduces biases in both mean and extreme heat-stress conditions.

The resulting dataset is about 57 TB and is freely available for reuse. It has already supported several of our studies on the consequences of future heat, including projections of uncompensable heat exposure and heat-related labor productivity loss.

My broader goal is to make high-quality heat data easier to use, so that researchers can spend less time rebuilding the same data infrastructure and more time asking new questions. If you are working on heat and need a dataset that I have developed—even if it is not yet publicly archived—please feel free to contact me.

Related publications

1. Kong, Q. & Huber, M. (2025). A global high-resolution and bias-corrected dataset of CMIP6 projected heat stress metrics. Scientific Data.
2. Malik, A., Masabathini, S., Shaikh, M. A., Kong, Q., Usman, M., Hari Prasad, D. & Hoteit, I. (2026). A global high-resolution comprehensive heat indices dataset from 1950 to 2024. Scientific Data.
3. Rahai, R., Kong, Q., Dogan, T., Evans, G. W. & Wells, N. M. (2026). Heat stress metrics for US census tracts 1998–2020. Scientific Data.

← All research projects

How “Humid” Should Humid Heat Be?

How hot we feel depends not only on temperature but also on humidity — muggy heat feels worse than dry heat at the same temperature.
Imagine that one wants to develop a heat stress index: both temperature and humidity would be important ingredients. But how much weight should each receive is far from clear (see the nice synthesis in this Commentary by Dr. Jane Baldwin). Existing heat stress indices vary widely in the relative weight they assign to humidity versus temperature.

How “humid” should humid heat be? We care about this question because how much humidity is incorporated into a heat stress measure can shape real-world decisions and our understanding of fundamental questions, such as:

• Whether a heat warning should be issued on a given day
• Which part of a city, country, or the world experiences the greatest heat stress
• How heat stress will change across regions as the climate warms
• How effective certain heat-mitigation strategies (e.g., irrigation and green cities) actually are

— and many others.

We propose a biophysics-informed, data-driven framework to answer this question. We ask what humidity weight best predicts real-world heat-related health outcomes for a given population and region.

The framework has been applied to identify the optimal humidity weight for predicting heatstroke in Japan and can be readily applied to other health outcomes, regions, and demographic subgroups. We believe this framework can both sharpen our mechanistic understanding of how much humidity matters and inform practical decisions about predicting and managing heat stress.

Related publications

1. Kong, Q., Guo, Q., Heft-Neal, S., Jing, R., Huang, X., Wang, Z., Wagner, Z., Huber, M., Hashizume, M. & Bendavid, E. (2026). Humidity's role in population heat-health risk. Nature Communications. (accepted)