Limiting warming to 1.5C could prevent ‘thousands’ of heat deaths in US cities

a man shades his face from the summer heat
A man shades his face from the summer heat, New York. Credit: Kathy deWitt / Alamy Stock Photo.

Holding global temperature rise to 1.5°C above pre-industrial levels, rather than 2°C or 3°C, could help prevent thousands of deaths in US cities during heatwaves, a new study says

— The research projects heat-related deaths for 15 cities across the US under the different levels of future warming. The results suggest that major cities, such as New York City and Los Angeles, could see hundreds or thousands more deaths in extreme heat without greater ambition in global emissions cuts.

The findings provide “compelling evidence for the heat-related health benefits of limiting global warming to 1.5°C” in the US, the study concludes.

The work “breaks new ground”, a scientist not involved in the work tells Carbon Brief, and shows that “the strongest climate policies, both for mitigation and adaptation, will save lives and help us to avoid never-before-seen human suffering from extreme heat”.

Heat deaths

As global temperatures increase, the likelihood and severity of summer heatwaves is expected to rise – and so too is the number of heat-related summer deaths. For example, a recent report by the Lancet Countdown on Health and Climate Change project – an annual review of the scientific evidence of climate change’s effect on human health – found that the number of vulnerable people exposed to heatwaves is already increasing by millions.

Attribution studies are increasingly able to link rising global average temperatures with the frequency and severity of heatwaves, and even heatwave deaths. A study from 2016, for example, found that hundreds of deaths in London and Paris during the 2003 European summer heatwave could be attributed to the impact of climate change.

The new study, published in Science Advances, estimates the future heat deaths in 15 US cities under different levels of warming. The researchers compare the two warming limits enshrined in the Paris Agreement – 1.5°C and 2°C above pre-industrial levels – with the estimated 3°C of warming that existing worldwide mitigation commitments would amount to.

us states

Using data on daily mortality and temperature data for 1987-2000, the researchers identified the “exposure-response” relationship between temperature and heat-related deaths for each city. This essentially describes how the risk of death relates to temperature.

Typically, the risk of death is highest in very cold and very hot conditions. Somewhere in between will be a “minimum mortality temperature” (MMT) where the number of heat-related deaths is at its lowest. The lowest MMT of the cities studied in St Louis in Missouri at 15°C, while the highest is 34.5°C in Phoenix, Arizona.

Applying these observed relationships between temperature and heat-related deaths in climate model simulations, the researchers projected the number of deaths under each level of warming. They focus on “1-in-30-year heat-related deaths” – the annual number of people dying from heat where that year is the warmest in 30 years.

The researchers kept all other factors constant in their experiment, explains lead author Dr Eunice Lo, a research associate at the University of Bristol. She tells Carbon Brief:

“We’ve assumed no adaptation, we’ve kept global population to be the same, we’ve kept the relationship between mortality and temperature to be the same – so that the only difference between these worlds would be temperature.

“We know that there are many factors that would affect future level of mortality and temperature-related mortality, but we’re asking the question of how global mean temperature rise – the level of global mean temperature rise, or difference – would affect mortality, and only this factor.”


In the video clip below, recorded at the American Geophysical Union Fall Meeting last year, Lo explains her study to Carbon Brief.

‘Mitigate more’

The chart below shows the projected increase in heat deaths for warming of 1.5°C (dark blue), 2°C (blue) and 3°C (red) compared to present day (2006-15) for a 1-in-30-year heat event.

Additional annual heat deaths in 15 US cities for a 1-in-30-year heat event, compared to 2006-15, for a 1.5°C (dark blue), 2°C (blue) and 3°C (red) warmer world. Whiskers show 2.5-97.5% uncertainty range. Data provided by Eunice Lo. Highchart by Carbon Brief.


The results show an increasing number of heat-related deaths as the climate warms. At 1.5°C, the study estimates between 35 and 779 additional annual deaths, depending on the city. For 2°C, this increases to 70-1,515, and then at 3°C to 139-3,495.

The results highlight the benefits of meeting the limits set out in the Paris Agreement, says Lo:

“Our main finding is, basically, for most of our studied cities, if we mitigate more – so if we increase our climate action to meet the 2°C target – then fewer people will die from heat. But this level of mortality – heat-related mortality – would be substantially even lower if we mitigate to 1.5°C of warming. So meeting the Paris Agreement’s targets – especially the 1.5°C target – would be substantially beneficial to the population in the US.”


New York City stands out with the most to gain from limiting warming, the paper says:

“New York City…could see 1,980 1-in-30-year heat-related deaths avoided in the 2°C warmer world relative to the 3°C warmer world under the assumption of constant population. If the 1.5°C world is realised, 2,716 of 1-in-30-year heat-related deaths could be avoided, relative to 3°C.”


This is partly because New York City has the largest population of all the cities that the team studied, says Lo, which means it is likely to have more people who are vulnerable during heatwaves.

Similarly, the second and third most populous cities in the US – Los Angeles and Chicago – see substantially lower numbers of projected heat deaths under stricter warming limits.

The chart below shows the results on a relative scale, where the number of deaths are estimated per 100,000 of population. With the data presented in this way, the cities of Miami and Detroit see the largest reductions in heat deaths with lower warming levels.

Additional annual heat deaths per 100,000 of population in 15 US cities for a 1-in-30-year heat event, compared to 2006-15, for a 1.5°C (dark blue), 2°C (blue) and 3°C (red) warmer world. Whiskers show 2.5-97.5% uncertainty range. Data provided by Eunice Lo. Highchart by Carbon Brief.


There are some cases where the results don’t show clear differences at each warming levels the study notes. For example, in Atlanta, San Francisco and St Louis, the increases in heat deaths at 2°C and 3°C are statistically similar.

The paper also finds that some of the future 1-in-30 year maximum temperatures in these cities “may be hotter than what’s been observed in the last 30 years”, says Dr Vijay Limaye, a climate change and health science fellow at the US Natural Resources Defense Council science centre. Limaye, who wasn’t involved in the research, tells Carbon Brief:

“We may be literally off the charts in terms of unprecedented, widespread human exposure to insufferable heat if we don’t meet the challenge of limiting carbon pollution.”


The projected number of avoided deaths from limiting global temperature “may actually be underestimated”, says Limaye, because the study does not take into account future population growth.

The research also assumes that the existing relationship between temperature and heat-related deaths remains constant. This could change in the future as it has done over past decades. For example, adaptation measures could help reduce the future health burden of extreme heat. However, this is “not a given” says Limaye:

“For example, we see present-day access and affordability issues with air conditioning. We’re in new territory as public health scientists, and it’s not clear that this sobering picture even represents the full magnitude of the threat posed by deadly heat.”


The results draw attention to the raised health risks of global warming beyond the Paris limits, says Lo. This is particularly relevant as existing commitments by governments to cut emissions, known as “Nationally Determined Contributions” (NDCs), need to be ramped up in order to hold warming to 1.5°C or 2°C, she concludes:

“One of the motivations of this work is that the next round of NDC submissions for the Paris Agreement will happen in the year 2020, and we hope that our results could motivate increasing climate action internationally in these NDCs.”


Lo, Y. T. E., et al. 2019. Increasing mitigation ambition to meet the Paris Agreement’s temperature goal avoids substantial heat-related mortality in U.S. cities, Science Advances, doi:eaau4373

by Robert McSweeney | Carbon Brief


When the river runs dry: the Australian towns facing heatwave and drought
Melbourne heatwave: city expecting 44C as Victoria faces hottest day since Black Saturday
Adelaide breaks its all-time heat record, hitting 46.6C, in extreme Australia heatwave