slidesharetrianglebookfile-text2file-picturefile-musicfile-playfile-videolocationcalendarsearchwrenchcogs stats-dotshammer2menudownload2questioncrossentergoogle-plusfacebookinstagramtwitter

Is air conditioning alleviating heat stress for everyone?

Air conditioning is a double-edged sword. It is responsible for a significant reduction of heat-stress-related mortality in developed countries, but it brings along its own set of challenges.

This story is illustrated with global Universal Thermal Climate Index (UTCI) maps. UTCI is an indicator similar to “perceived temperature”, that expresses the thermal well-being of a person in relation to the surrounding environment.

Lobelia Earth has created the UTCI projections dataset, based on Copernicus Climate Change Service (C3S) data from the Climate Data Store.

As temperatures rise globally and heat stress becomes more prevalent, energy consumption due to the use of air conditioning (AC) appliances is expected to skyrocket.

Hong Kong Apartment

A study on 25 million Mexican households in 2009-2012 showed that energy consumption is highly dependent on air temperature1, or more accurately on thermal stress.

Ciudad Juarez Sunset

The link between electricity consumption and climate is stronger in regions where AC appliances are more abundant. In the Sonora and Sinaloa states, where AC penetration exceeds 50%, a single additional day above 32 °C (90 °F) increases the monthly electricity bill by 4.2%.

Sonora Hot




By the end of this century, residential electricity consumption in Mexico is expected to rise between 64% and 83% with respect to 2010, as a consequence of higher thermal stress and more widespread AC adoption and usage.

In AC-saturated California, electricity expenditure is expected to increase by 39% in 2100 relative to 1980-2000, in a scenario with zero population growth and only moderate price increments2.

Horizon sky sun sunrise sunset

AC has played a key role in climate adaptation. In the US, where states like Texas and Florida already had 90% AC penetration in 19803, studies have shown that the significant reduction in heat-related mortality during the XXth century, especially after 1960, is fully attributable to the advent of air conditioning4.

Air Conditioners

But the benefits of AC adoption have been elusive until now for low-GDP countries, even those located in relatively warmer climates such as Brazil (30% market penetration), India (20%) or again Mexico (16%)5.

Even in relatively wealthy countries in Europe, low-income families not only have limited access to AC, but are bound to be hit by overall increased electricity consumption, feeding the loop leading to energy poverty.





In Spain, already 19% of families spend more than 5% of their income on electricity6. In future heat stress periods, the combination of demand spikes from AC-equipped households (60%) and numerous energy-poor families may lead to scenarios of deeper inequality.

Senior Woman

There is little doubt that air conditioning will become a key resilience strategy in a global changing climate. AC adoption will surge, not only ensuring thermal comfort but also reducing heat-related mortality.

But AC will also introduce additional stresses in the system, causing a regressive impact on lower-income households and poorer countries due to increased electricity consumption, and also leading to additional emissions (up to 30 millions tons of CO₂ every year, from AC usage in Mexico alone1).

Suitable policies should address the societal and ecological downsides of air conditioning, targeting a decarbonised power mix, more efficient appliances, and a resolute fight against energy poverty.

Scroll down. Drag on the globe to rotate

About the data

Shown on the map above is the monthly average of daily maximum of Universal Thermal Climate Index (UTCI)9, a variable that captures the physiological capability of the human body to maintain its own temperature, in a process known as thermoregulation. UTCI is expressed in degrees and depends on air temperature, humidity, wind and radiation.

Past values of UTCI can be obtained from ECMWF's Climate Data Store. The ERA5-HEAT dataset7 provides hourly values of UTCI since January 1979, with 0.25° resolution and coverage from 60° S to 90° N, based on the ERA5 reanalysis dataset.

Climate projections of UTCI included in this story correspond to monthly means of daily maximum fields. For the calculation of UTCI, 4 physical variables evaluated at the Erath surface are used: 2-metres temperature, 2-metres humidity, 10-metres wind speed and mean radiant temperature (whose estimation involves several components of the solar and thermal radiation10). In order to compute UTCI fields with the data available on the CDS, we have implemented the following approach: Daily fields of humidity, wind and daily maximum temperature have been used from the CIMP5 daily data on single levels dataset (1971-2100). For the computation of the mean radiant temperature (MRT), which requires hourly resolution fields not available on the CDS for climate projections, climatological monthly means of daily maximum MRT have been used directly computed from the hourly ERA5-HEAT dataset. A validation of this approach have been carefully performed showing that the corresponding error is well below the inner uncertainty of the official UTCI dataset tending to zero when monthly averaging. Currently, UTCI projections are calculated for the RCP8.5 scenario using the ACCESS1-0 model after being bias corrected.


1 Davis, L., Gertler, P. (2015): Contribution of air conditioning adoption to future energy use under global warming. Proceedings of the National Academy of Sciences, 112: 5962-5967. DOI: 10.1073/pnas.1423558112.

2 Auffhammer, M., Aroonruengsawat, A. (2011): Simulating the impacts of climate change, prices and population on California’s residential electricity consumption. Climatic Change, 109, 191–210. DOI: 10.1007/s10584-011-0299-y.

3 Auffhammer, M., Mansur, E. (2014): Measuring climatic impacts on energy consumption: A review of the empirical literature. Energy Economics, 46, 522-530. DOI: 10.1016/j.eneco.2014.04.017.

4 Barreca, A., Clay, K., Deschenes, Ol, Greenstone, M., Shapiro, J.S. (2016): Adapting to Climate Change: The Remarkable Decline in the US Temperature-Mortality Relationship over the Twentieth Century. Journal of Political Economy, 124:1, 105-159. DOI: 10.1086/684582.

5 Lapillonne, B. (2019): The Future of Air-Conditioning: How will the demand for household air-conditioning evolve in the coming years? (Executive Brief)

6 Randazzo, T., De Cian, E., Mistry, M. (2020): Air conditioning and electricity expenditure: The role of climate in temperate countries. Economic Modelling, 90. DOI: 10.1016/j.econmod.2020.05.001.

7 Di Napoli, C., Barnard, C., Prudhomme, C., Cloke, H.L., Pappenberger, F. (2020): ERA5‐HEAT: A global gridded historical dataset of human thermal comfort indices from climate reanalysis. Geoscience Data Journal, 2020, 00:1-9. DOI: 10.1002/gdj3.102

8 Asadoorian, M.O., R.S. Eckaus and C.A. Schlosser (2008): Modeling climate feedbacks to energy demand: The case of China. Energy Economics, 30(4): 1577-1602. DOI: 10.1016/j.eneco.2007.02.003.

9 Bröde, P., Fiala, D., Błażejczyk, K., Holmér, I., Jendritzky, G., Kampmann, B., Tinz, B., Havenith, G. (2012): Deriving the operational procedure for the Universal Thermal Climate Index (UTCI). International Journal of Biometeorology, 56, 481–494. DOI: 10.1007/s00484-011-0454-1.

10 Di Napoli, C., Hogan, R.J., Pappenberger, F. (2020): Mean radiant temperature from global-scale numerical weather prediction models. International Journal of Biometeorology, 64, 1233–1245. DOI: 10.1007/s00484-020-01900-5.