Impact of interannual meteorological variability on total ozone in northern middle latitudes:: A statistical approach

The midwinter (January-February) and summer (June-July-August) means of the total O-3 derived from ground-based measurements taken at northern midlatitudes are analyzed for the period 1963-1999. We use a simple, multiple, linear regression model to examine the ozone changes related to interannual meteorological variability. The explanatory variables of the model are selected from many variables characterizing the atmospheric circulation in the troposphere, the lowermost stratosphere, and the stratospheric overworld. The modeled time series of the midwinter and summer total ozone over Europe and North America show significant year-to-year variations corresponding to observed variations. Moreover, the model reproduces correlation between the summer (June-July-August) midlatitudinal ozone and its preceding midwinter (January-February) value. The model calculates its lowest ozone amount during winter 1993. This follows the Mount Pinatubo eruption, but die ozone behavior can be explained as a result of dynamical variations only. The model-observation agreement provides support for the hypothesis that the long-term variations in total ozone over northern midlatitudes are strongly modulated by transport processes. It is estimated that interannual changes in the dynamical processes account for similar to 70-80% of the long-term ozone decline over the northern midlatitudes for the 1970-1997 period.