全球变化知识资源中心

Recent work has established that there is a robust upper limit to the human body's ability to adapt to heat and humidity stress¹. If people are exposed to the combination of temperature and humidity above this level over extended periods, hyperthermia and ultimately death will result, because dissipation of heat from the body becomes physically impossible. Using global climate models, Sherwood and Huber¹ had estimated that this limit will become important in the distant future, if a 7 °C increase in the global mean temperature is reached. Now, writing in Nature Climate Change, Pal and Eltahir² use a high-resolution regional climate model to provide evidence that such conditions could occur much earlier. The study pinpoints the Persian Gulf (also known as the Arabian Gulf, referred to here as the Gulf) as a region where heat waves are likely to exceed this critical threshold in the course of the current century, assuming that climate change proceeds unabated and reaches a global temperature increase of around 4 °C.

The consequences of major heat waves for human health has become apparent from the death toll of recent events such as those in Chicago in 1995, Europe in 2003 and Russia in 2010^{3, 4, 5}. During these heat waves, mostly elderly and ill individuals were at risk, as well as people who did not take appropriate precautions. The focus of the Pal and Eltahir² study concerns another category of heat waves — one that may be fatal to everybody affected, even young and fit individuals under shaded and well-ventilated outdoor conditions.

Our bodies cool by ventilation (exchange of heat) and sweating (evaporation of water). However, these processes have physical limits that are expressed by the concept of wet-bulb temperature (TW), which represents the temperature to which an object may be cooled by ventilation and evaporation. If TW approaches or exceeds body temperature (i.e. TW > 35 °C), then the body's natural cooling system is disabled and people exposed to such conditions, in the absence of external cooling aids (such as air conditioning), are at high risk.

The study of Pal and Eltahir shows that these dangerous heat waves may occur in the course of this century in the vicinity of the Arabian Peninsula. They employ an ensemble of regional climate model simulations at 25 km horizontal resolution and consider control (1976–2005) and scenario (2071–2100) conditions using two greenhouse gas emission scenarios (RCP8.5 and RCP4.5) that represent business-as-usual and mitigation pathways.

Although TW = 35 °C is never reached under current climatic conditions in their simulations, it is projected to be reached several times in the business-as-usual 30-year scenario period in Bandar Mahshahr and Bandar Abbas (Iran), Dhahran (Saudia Arabia), Doha (Qatar), Dubai and Abu Dhabi (United Arab Emirates), and probably in additional locations along the Gulf that have not been specifically investigated. Furthermore, temperatures are projected to reach unacceptable levels; for instance, in some years of the scenario period T = 60 °C will be exceeded in Kuwait City. The rise in temperature and humidity would thus be likely to constrain development along the shores of the Gulf. The study also shows that global mitigation efforts consistent with the RCP4.5 scenario would be of significant benefit in the Gulf region.

Why is this effect so severe around the Gulf (and to some lesser extent the Red Sea)? This region is well known for its extremely warm sea surface. These conditions owe their existence to the shallow depth of the Gulf, which implies that in response to the seasonal cycle it warms much more than the Arabian Sea (Fig.1a). There is also evidence of a significant warming of the Gulf during the past 60 years⁶, which was most pronounced in the period 1990–2010 with a trend of a 0.47 °C increase per decade. Again, this warming was significantly larger in the Gulf than in the Arabian Sea.

a, Sea surface temperature in the Arabian Sea and the Gulf on 31 July 2015. Note the high temperatures in the Gulf. b, Temperature and humidity conditions at the airport of Bandar Mahshahr, Iran, from 28 July to 2 August 2015. The curves show temperature (T) and dewpoint temperature (T_d) (both in green), relative humidity (RH, blue) and wet-bulb temperature (TW, red). Note how TW peaks above 34 °C on 31 July.

1. Sherwood, S. C. & Huber, B. Proc. Natl Acad. Sci. USA 107, 9552–9555 (2010).
   • PubMed
   • Article
2. Pal, J. & Eltahir, E. Nature Clim. Change 6, 197–200 (2016).
   • Article
3. Changnon, S. A., Kunkel, K. E. & Reinke, B. C. Bull. Am. Meteor. Soc. 77, 1497–1506 (1996).
   • Article
4. Robine, J.-M. et al. C. R. Biologies 331, 171–178 (2008).
   • PubMed
   • Article
5. Barriopedro, D., Fischer, E., Luterbacher, J., Trigo, R. & Garcia-Herrera, R. Science 332, 220–224 (2011).
   • CAS
   • ISI
   • PubMed
   • Article
6. Shirvani, A., Nazemosadat, S. M. J. & Kahya, E. Arab. J. Geosci. 8, 2121–2130 (2015).
   • Article
7. Xue, P. & Eltahir, E. A. B. J. Climate 28, 5041–5062 (2015).
   • Article
8. Van Oldenborgh, G. J. et al. in Climate Change 2013: The Physical Science Basis. (eds Stocker, T. F. et al.) Annex 1, 1311–1393 (IPCC, Cambridge Univ. Press, 2013).

Download references

Fig. 1a is based on the ECMWF analysis (www.ecmwf.int) and was provided by Daniel Lüthi (ETH Zurich). The meteorological surface data analysed includes hourly data from a large number of airports in the region. The author is indebted to Ralph Rickli (Meteotest, Bern) for providing the raw data for Fig. 1b and for professional support in analysing the Iranian heat wave. Useful comments from Erich Fischer, Niki Gruber and Jan Rajczak (ETH Zurich) on an earlier version of this article are also appreciated.