A recent study by Lawrence Livermore National Laboratory (LLNL) scientists and collaborators is the first to use an ensemble of global chemistry climate models to estimate death rates from air pollution caused by the impact of climate change on pollutant concentrations.
Ground-level ozone and fine particulate matter are detrimental to human health. Their future concentrations depend on changes in pollutant emissions, as well as changes in climate (e.g., temperature, humidity, rainfall and boundary layer mixing).
The new study uses a scenario extrapolating greenhouse gas and pollutant emissions into the future known as RCP8.5 (which assumes that greenhouse gas emissions continue to rise, but that air pollutant emissions decrease due to increased regulation). This is separate from the impact of changes in pollutant emissions, and the direct impact of climate change on health, such as deaths from extreme heat.
The study found that death rates due to air pollution are likely to be higher in the future due to climate change alone. This study used simulations from several modeling groups to estimate the impact of climate change alone on air quality, by keeping both pollutant emissions and population distributions constant in each epoch. The research appears in Nature Climate Change.
The average of the model estimates predicted a global increase of 43,600 deaths per year (from ozone) and 215,000 (from fine particles: PM2.5), due to climate change alone, in the year 2100 for the RCP8.5 scenario. The increased number of deaths mostly occurred in regions that are already highly populated and highly polluted. For comparison, the same group previously estimated the global death rate in 2000 due to air pollution (above the background from year 1850) to be 470,000 ozone-related deaths per year, and 2.1 million fine-particle related deaths per year.
However, there were different results among models, with seven out of 10 models predicting an increase in ozone-related deaths due to climate change in the year 2100 (the range among models is a decrease of 195,000 to an increase of 237,000 deaths per year), and five out of six models predicting an increase in fine-particle-related deaths due to climate change in the year 2100 (the range among models is a decrease of 76,100 to an increase of 595,000 deaths per year).
“Most of the models predicted increased death rates from the climate change impact on air pollution, but some did predict decreases, which shows the importance of using multiple models to quantify the uncertainty,” said LLNL atmospheric scientist Philip Cameron-Smith, one of the authors on the paper. “However, because a large majority of the models showed increases, it allowed us to conclude that actions to mitigate climate change are likely to benefit human health by improving air quality in many locations, but it also shows that further improvements in chemistry-climate models, and application to other future air-quality scenarios, are needed to better understand the feedbacks between climate and air quality. In particular, because deaths attributable to ozone and particulates are often associated with extreme events, it is critical to develop modeling systems that resolve the high-resolution details of extreme events as the climate evolves.”
Climate change can affect air quality through several pathways, including changes in: the ventilation and dilution of air pollutants, photochemical reaction rates, removal by rainfall, stratosphere-troposphere exchange of ozone, wildfires and natural biogenic and lightning emissions. Some of these processes improve air quality, and some make it worse. Since the balance between these processes, and their response to climate change, can vary from model to model, this is the origin of the differences between the models. Overall, changes in these processes are expected to increase ozone in polluted regions during the warm season, especially in urban areas, but decrease ozone in remote regions due to greater water vapor concentrations leading to greater ozone destruction.
LLNL scientist Dan Bergmann also contributed to this work.
In addition to LLNL, this research was led by the University of North Carolina, and included contributions from the National Center for Atmospheric Research, Duke University, University of Reading, NASA Goddard Institute for Space Studies, Met Office-Hadley Centre for Climate Prediction, NOAA Geophysical Fluid Dynamics Laboratory, Nagoya University, Kyushu University, University of Edinburgh, Centre National de Recherches and the National Institute of Water and Atmospheric Research.
The LLNL portion of the research was funded by the Department of Energy Office of Science, Biological and Environmental Research.