Premature mortality in Japan due to ozone Amin Nawahda a, * , Ken Yamashita b , Toshimasa Ohara c , Junichi Kurokawa b , Tsuyoshi Ohizumi b , Fang Chen b , Hajime Akimoto b a Environmental Research Center, Sohar University, Sohar 311, Oman b Asia Center for Air Pollution Research Center, Niigata-shi 950-2144, Japan c National Institute of Environmental Studies, Japan highlights  Premature mortality cases in Japan were estimated using modeled and monitored ozone.  Modeled ozone in 2005 in Japan caused 13,000 (95% CI: 4320e17,300) premature mortality cases.  Monitored ozone in 2005 in Japan caused 5220 (95% CI: 1740e6960) premature mortality cases.  Modeled ozone caused about 2.5 times greater premature mortality cases compare to monitored ozone.  Premature mortality estimation using monitored ozone from 1145 monitors in Japan is less uncertain than that of modeled ozone. article info Article history: Received 26 January 2013 Received in revised form 16 September 2013 Accepted 21 September 2013 Keywords: Premature mortality Ozone Exposure CMAQ/REAS Japan abstract In Japan, all 47 prefectures conduct routine air quality monitoring at 1145 stations throughout the country to assess environmental effects. This study aims to provide a better understanding of possible estimations of premature mortality in Japan caused by exposure to monitored and modeled concen- trations of tropospheric ozone during the period from January to December, 2005. The spatial distri- bution and temporal variation of ozone concentrations were modeled using the Models-3 Community Multiscale Air Quality modeling system coupled with the Regional Emission Inventory in Asia (CMAQ/ REAS). Premature mortality caused by exposure to ozone was calculated assuming a relative risk (RR) value of 1.003 [95% Confidence Interval (CI): 1.001e1.004] for concentrations above 35 ppb according to the SOMO35 index (annual Sum of daily maximum 8-h Ozone Means Over 35 ppb) recommended by WHO (2008). Based on CMAQ/REAS simulations, the estimated all-cause premature mortality in 2005 is about 13,000 (95% CI: 4320e17,300) cases. This value is 2.5 times greater than the estimated premature mortality based on monitored ozone concentrations, which is 5220 (95% CI: 1740e6960) cases. Ó 2013 Elsevier Ltd. All rights reserved. 1. Introduction In recent years, the environmental risks caused by exposure to tropospheric ozone have increased all over the world (US EPA, 2006; Anenberg et al., 2010; WHO, 2005, 2008; Nawahda et al., 2012). When ozone is formed, it remains suspended for hours to days in the lower atmosphere and can endanger local and regional receptors (Lippmann, 2009). Statistically significant relationships have been identified between elevated concentrations of ozone and premature mortality (WHO, 2008; Lippmann, 2009; Jerrett et al., 2009; Bell et al., 2005; Yang et al., 2012; Zhang et al., 2006; Ou et al., 2012; Liu et al., 2013). However, spatial variability and the lack of a known threshold value for premature mortality caused by ozone exposure complicate the process of establishing well- defined standards and exposure guidelines. Most studies on the association between short-term exposure to ozone and mortality are based on different exposure metrics such as 1-h maximums, maximum 8-h averages, and maximum 24-h averages. Jerrett et al. (2009) showed that maximum 8-h average and 1-h exposures resulted in similar associations between ozone and mortality. Chen et al. (2012) showed that mortality analysis based on the maximum 1-h metric produced similar results, but not as strong as the 8-h and 24-h metrics. Smith et al. (2009) showed that a simple scaling relationship cannot be used to convert one metric to another. When scaling has to be done, seasonality and meteorological parameters need to be considered (Anderson and Bell, 2010). Short-term exposure to maximum 8-h ozone concentrations has been posi- tively associated with all causes of mortality in many countries * Corresponding author. Tel.: þ968 26720101; fax: þ968 26720102. E-mail address: amin.nawahdah@gmail.com (A. Nawahda). Contents lists available at ScienceDirect Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv 1352-2310/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.atmosenv.2013.09.049 Atmospheric Environment 81 (2013) 538e545