Attributing heatwave-related mortality to climate change: a case study of the 2009 Victorian heatwave in Australia

Abstract

Determining the influence of climate change behind human mortality is of interest to many sectors. However, it is a fledgling field where studies have centered on northern hemisphere events. This study presents the first attribution assessment on the mortality burden of an Australian heatwave to climate change. We focus on excess heatwave- (defined by climatological definitions) related mortality in the state of Victoria that occurred during the 2009 southeast Australian heatwave. An epidemiological model derived from well-established methods defining the relationship between observed heatwave temperatures (95th, 97.5th and 99th percentiles) and mortality is applied to heatwaves in simulations that either include or omit anthropogenic climate forcing from eight climate models. Across all models, the frequency of a heatwave-related mortality event similar to the 2009 Victorian event has, on average, doubled under factual conditions relative to counterfactual conditions. Moreover, on average, around 6 +/- 3-4 extra individuals out of 31 (an increase of 20%) died as a direct result of extreme temperatures due to anthropogenic influence on the climate. Despite the small total number of attributable deaths as per the epidemiological model, six out of eight climate models predicted a statistically significant anthropogenic influence, indicating that climate change increased the heatwave-related mortality impact of this event. We make clear that, in line with previous Australian-based studies, the focus on mortality relative to the top 5% of temperatures logically infers a smaller mortality signal relative to the top 50% of temperatures, as would be defined by a more general temperature-related epidemiological model. As research, planning and policy interest in the role of climate change behind the burden health-and other adverse impacts of weather and climate extremes-continues to grow, it is vital that interdisciplinary collaborations are nurtured, so that the resulting science is of high-quality rigour, and policy relevance.