Increase in surface ozone at rural sites in the western US

We evaluated O-3 data for the period 1987-2004 from 11 rural and remote sites in the north and western US, including two sites in Alaska. All sites show a seasonal cycle with a spring or spring-summer maximum. By deseasonalizing the data, we are better able to identify seasonal and spatial patterns and long-term trends. For most of the locations in the western US that we considered, there are significant inter-site correlations in the deseasonalized monthly means. This indicates that there are large scale factors that influence the monthly mean O-3 concentrations across the western US. At seven out of nine sites in the western US, there is a statistically significant increase in O-3 with a mean trend of 0.26 ppbv year(-1) (range at the seven sites is 0.19-0.51 ppbv year(-1)). At three of the sites, we examined the data in more detail to find that the trends are present in all seasons. At the two sites in Alaska, no clear pattern was found. At the one ozonesonde site in the western US with long-term observations (Boulder, Colorado), no significant trend was identified. However, the statistical power in the ozonesonde analysis is limited due to the low frequency of ozonesonde launches. Temperature changes can explain only a fraction of the surface O-3 trend. We consider several possible explanations for these trends, including: increasing regional emissions, changes in the distribution of emissions, increasing biomass burning or increasing global background O-3. With the available data, we are not able to unambiguously identify the cause for increasing O-3 in the western US (C) 2007 Elsevier Ltd. All rights reserved.