Intercontinental transport of pollution manifested in the variability and seasonal trend of springtime O3 at northern middle and high latitudes

Observations (0–8 km) from the Tropospheric Ozone Production about the Spring Equinox (TOPSE) experiment are analyzed to examine air masses contributing to the observed variability of springtime O3 and its seasonal increase at 40°–85°N over North America. Factor analysis using the positive matrix factorization and principal component analysis methods is applied to the data set with 14 chemical tracers (O3, NOy, PAN, CO, CH4, C2H2, C3H8, CH3Cl, CH3Br, C2Cl4, CFC-11, HCFC-141B, Halon-1211, and 7Be) and one dynamic tracer (potential temperature). Our analysis results are biased by the measurements at 5–8 km (70% of the data) due to the availability of 7Be measurements. The identified tracer characteristics for seven factors are generally consistent with the geographical origins derived from their 10 day back trajectories. Stratospherically influenced air accounts for 14 ppbv (35–40%) of the observed O3 variability for data with O3concentrations \u3c100 ppbv at middle and high latitudes. It accounts for about 2.5 ppbv/month (40%) of the seasonal O3 trend at midlatitudes but for only 0.8 ppbv/month (\u3c20%) at high latitudes, likely reflecting more vigorous midlatitude dynamical systems in spring. At midlatitudes, reactive nitrogen-rich air masses transported through Asia are much more significant (11 ppbv in variability and 3.5 ppbv/month in trend) than other tropospheric contributors. At high latitudes the O3 variability is significantly influenced by air masses transported from lower latitudes (11 ppbv), which are poor in reactive nitrogen. The O3 trend, in contrast, is largely defined by air masses rich in reactive nitrogen transported through Asia and Europe across the Pacific or the Arctic (3 ppbv/month). The influence from the stratospheric source is more apparent at 6–8 km, while the effect of O3 production and transport within the troposphere is more apparent at lower altitudes. The overall effect of tropospheric photochemical production, through long-range transport, on the observed O3 variability and its seasonal trend is more important at high latitudes relative to more photochemically active midlatitudes