Revisiting the evidence of increasing springtime ozone mixing ratios in the free troposphere over western North America

We present a 20 year time series of in situ free tropospheric ozone observations above western North America during springtime and interpret results using hindcast simulations (1980–2014) conducted with the Geophysical Fluid Dynamics Laboratory global chemistry-climate model (GFDL AM3). Revisiting the analysis of Cooper et al. (2010), we show that sampling biases can substantially influence calculated trends. AM3 cosampled in space and time with observations reproduces the observed ozone trend (0.65 ± 0.32 ppbv yr−1) over 1995–2008 (in simulations either with or without time-varying emissions), whereas AM3 “true median” with continuous temporal and spatial sampling indicates an insignificant trend (0.25 ± 0.32 ppbv yr−1). Extending this analysis to 1995–2014, we find a weaker ozone trend of 0.31 ± 0.21 ppbv yr−1 from observations and 0.36 ± 0.18 ppbv yr−1 from AM3 “true median.” Rising Asian emissions and global methane contribute to this increase. While interannual variability complicates the attribution of ozone trends, multidecadal hindcasts can aid in the estimation of robust confidence limits for trends based on sparse observational records.