Global and regional changes in exposure to extreme heat and the relative contributions of climate and population change

www.nature.com/scientificreports OPEN received: 10 August 2016 accepted: 31 January 2017 Published: 07 March 2017 Global and regional changes in exposure to extreme heat and the relative contributions of climate and population change Zhao Liu1,2,3, Bruce Anderson2, Kai Yan4, Weihua Dong4, Hua Liao4 & Peijun Shi1,3 The frequency and intensity of extreme heat wave events have increased in the past several decades and are likely to continue to increase in the future under the influence of human-induced climate change. Exposure refers to people, property, systems, or other elements present in hazard zones that are thereby subject to potential losses. Exposure to extreme heat and changes therein are not just determined by climate changes but also population changes. Here we analyze output for three scenarios of greenhouse gas emissions and socio-economic growth to estimate future exposure change taking account of both climate and population factors. We find that for the higher emission scenario (RCP8.5-SSP3), the global exposure increases nearly 30-fold by 2100. The average exposure for Africa is over 118 times greater than it has been historically, while the exposure for Europe increases by only a factor of four. Importantly, in the absence of climate change, exposure is reduced by 75–95% globally and across all geographic regions, as compared with exposure under the high emission scenario. Under lower emission scenarios RCP4.5-SSP2 and RCP2.6-SSP1, the global exposure is reduced by 65% and 85% respectively, highlighting the efficacy of mitigation efforts in reducing exposure to extreme heat. Over the past decade tens of thousands of people have died from heat waves all over the world, for instance, Europe in 2003, Australia in 2008, Russia in 2010 and China in 20131–5. Although the mortality attributable to extreme heat events has actually decreased over the last century due to medical progress and economic development6–8, it is expected to increase in the future under the influence of human-induced climate change as heat wave events become more intense and more frequent with longer duration over most land areas in the 21st century9, leading to increased risk of heat-related morbidity and mortality10,11. It is important to note, however, that the changes in heat-related morbidity and mortality are not only a function of changes in physical hazards resulting from human-induced climate change but also are a function of changes in the number and vulnerability of individuals exposed to these hazards12. Accordingly, many studies have focused on future heat wave risk assessment13–16, which attempts to estimate the probable heat-related mortality and morbidity of people that an area would experience13,17 as a function of the changing characteristics of a given hazard, the number of people exposed to that hazard and their sensitivity and adaptive capacity to that hazard12,18. However, this last characteristic, which represents a population’s vulnerability to a particular hazard, can vary significantly for various factors such as age, season and geographic region19. Although a number of indices have been proposed focusing on different aspects of a population’s vulnerability, there is no consistent standard to measure vulnerability due to its inherent complexity20–22. In addition, projections of future demographic and socioeconomic factors of vulnerability are difficult to obtain, generally of coarse resolution and of short duration. As a result, most prior research uses fixed population estimates when quantifying future risk to changing climates, including those related to heat-induced mortality10,23. Here we extend this analysis by including future changes in the magnitude and spatial patterns of both extreme heat events and the 1 State Key Laboratory of Earth Surface Processes and Resource Ecology, Beijing Normal University, Beijing, 100875, China. 2Department of Earth and Environment, Boston University, Boston, 02115, USA. 3Academy of Disaster Reduction and Emergency Management, Beijing Normal University, Beijing, 100875, China. 4School of Geography, State Key Laboratory of Remote Sensing Science, Beijing Normal University, Beijing, 100875, China. Correspondence and requests for materials should be addressed to Z.L. (email: ) or W.D. (email: ) or P.S. (email: ) Scientific Reports | 7:43909 | DOI: 10.1038/srep43909 1 www.nature.com/scientificreports/ number of people impacted by these events. We recognize that such an analysis ignores both the spatially and temporally non-linear and non-uniform vulnerability of these populations to these events and as such we do not attempt to estimate changes in heat-related mortality and morbidity explicitly. Instead, following the lead of Jones et al.24, we seek to quantify population exposure to extreme heat waves (referred to throughout simply as ‘exposure’), which is an important first step towards estimating global and regional scale changes in the risk of heat wave mortality and morbidity. In particular, the primary purpose of this paper is to assess the future change of exposure to heat waves at global and regional levels under different development pathways. In addition, we also discuss the impact of three different factors: climate change, population growth and the interaction between the two. Further we select China, India and Nigeria as three typical countries to represent the different development paths and discuss their exposure change and the impact factors’ contribution. Based on this analysis of exposure and the relative contribution of different factors, we aim to characterize the future increase of heat-related exposure and clarify the importance of climate change and population growth to the change. Methods Materials. To estimate changes in the hazard we use bias-corrected and downscaled projections25 of Coupled Model Intercomparison Project Phase 5 (CMIP5) climate models to characterize future climate change. CMIP5 projections are based on general circulation model (GCM) estimates of present and future climate such as temperature and precipitation in response to various increases in radiatively active atmospheric constituents26,27. These increases are represented by four Representative Concentration Pathways (RCPs) corresponding to the atmospheric radiative forcing up to the year 210028. Each RCP represents a pathway based on simulated influences of land use and emissions of aerosols and greenhouse gases (GHG). The highest pathway is RCP8.5 and depicts a scenario with radiative forcing rising to 8.5 W/m2 by 2100, without applying any mitigation policy to GHG emissions. The lower scenarios, including RCP6.0, RCP4.5 and RCP2.6, do adopt some mitigation measures to control the GHG emission28,29. Following the lead of previous climate change impact assessments24, in this paper we subsequently use a subset of these models for which high spatial and temporal resolution downscaled data are available. In particular, we a (...truncated)