Analysis of monitoring data of ground-level ozone in Japan for long-term trend during 1990–2010: Causes of temporal and spatial variation

The puzzling increase of annual average mixing ratio of ozone (oxidant) in spite of the decrease of those of the ambient NOx and NMHC was analyzed using the twenty-one years monitoring data during 1990–2010 focusing on four high O3 areas in Japan, Tokyo Metropolitan Area (TMA), Nagoya Area (NA), Osaka/Kyoto Area (OKA) and Fukuoka Area (FA). During the period, the NOx and NMHC mixing ratios have decreased by 40–50 % and 51–54 %, respectively, in these areas. Nevertheless, the annual averaged increasing trends of O3 by linear regressions during the whole period are all positive, 1.23 ± 0.09, 0.85 ± 0.17, 1.41 ± 0.12 and 1.42 ± 0.17(1σ)% yr−1 for TMA, NA, OKA and FA, respectively. Three causes of long-term trends of O3 have been discussed: (1) the decrease of NO titration effect, (2) the increase of transboundary transport, and (3) the decrease of in situ photochemical production. The total ozone (TO) was defined by [TO] = [O3] + [NO2] – 0.1[NOx] as an index to indicate the net O3 removing the perturbation by in situ NO titration. TO shows statistically significant annual increasing trends, 0.20 ± 0.04 and 0.56 ± 0.08% yr−1 only in OKA and FA, and the increase of TO is not discernible in TMA and NA. In springtime (March–May), TO shows larger increasing rates of 0.44 ± 0.08, 0.59 ± 0.09 and 0.96 ± 0.18% yr−1 for TMA, OKA, and FA. The statistically significant increase of TO at higher rates in spring and western part of Japan suggests the increase of transboundary transport during the period. The reduction of domestic emissions of precursors was found to result the decreasing trend of the high mixing ratio range of 98-percentile most clearly in TMA, but is not enough yet to result in the decrease of annual mean mixing ratio of O3, which is compensated by the increase of transboundary transport and the decrease of the NO titration effect. The similar situation has been reported in Taiwan (Chou et al., 2006), and would apply more or less to other outflow region of Pacific rim, Korea and Hoang Kong, where the emission control started to reduce the O3 precursors. Further reduction of NOx emissions over a certain limit will negate the NO titration effect, and the decrease of net O3 due to the reduction of in situ photochemical production is expected.