Assessment and economic valuation of air pollution impacts on human health over Europe and the United States as calculated by a multi-model ensemble in the framework of AQMEII3

The impact of air pollution on human health and the associated external costs in Europe and the United States (U.S.) for the year 2010 is modelled by a multi-model ensemble of regional models in the frame of the third phase of the Air Quality Modelling Evaluation International Initiative (AQMEII3). This is the first study known to use a common health assessment approach across the two continents. The modelled surface concentrations of O 3 , CO, SO 2 and PM 2.5 from each model are used as input to the Economic Valuation of Air Pollution (EVA) system to calculate the resulting health impacts and the associated external costs. Along with a base case simulation, additional runs were performed introducing 20 % emission reductions both globally and regionally in Europe, North America and East Asia. Health impacts estimated by different models can vary up to a factor of three in Europe (twelve models) and the United States (three models). In Europe, the multi-model mean number of premature deaths is calculated to be 414 000 while in the U.S., it is estimated to be 160 000, in agreement with previous global and regional studies. In order to estimate the impact of biases coming from each model, two multi-model ensembles were produced, the first attributing an equal weight to each member of the ensemble, and the second where the subset of models that produce the smallest error compared to the surface observations at each time step. The latter results in increase of health impacts by up to 30 % in Europe, thus giving significantly higher mortality estimates compared to available literature. This is mostly due to a 27 % increase in the domain mean PM 2.5 levels, along with a slight increase in O 3 by ~ 1 %. Over the U.S., the mean PM 2.5 and O 3 levels decrease by 11 % and 2 %, respectively, when the optimal ensemble mean is used, leading to a decrease in the calculated health impacts by ~ 11 %. These differences encourage the use of optimal-reduced multi-model ensembles over traditional all model-mean ensembles, in particular for policy applications. Finally, the role of domestic versus foreign emission sources on the related health impacts is investigated using the 20 % emission reduction scenarios applied over the source regions as defined in the frame of HTAP2. The differences are calculated based on the models that are common in the basic multi-model ensemble and the perturbation scenarios, resulting in five models in Europe and all three models in the U.S. A 20 % reduction of global anthropogenic emissions avoids 54 000 and 27 500 premature deaths in Europe and the U.S., respectively. A 20 % reduction of North American emissions foreign emissions avoids ~ 1000 premature deaths in Europe and 25 000 premature deaths in the U.S. A 20 % decrease of emissions within the European source region avoids 47 000 premature deaths in Europe. Reducing the East Asian emission by 20 % avoids ~ 2000 premature deaths in the U.S. These results show that the domestic emissions make the largest impacts on premature death, while foreign sources make a minor contributing to adverse impacts of air pollution.