Tove M. Svendby

The total ozone series from Dobson spectrophotometer 56 (D56) operating at the University of Oslo (59.9°N, 10.7°E), Norway, from March 1978 to May 1998 has recently been reevaluated. Ozone trends for the time series are calculated using multiple regression models with explanatory variables for the 11 year solar cycle (lagged 22 months), the 2.5 year Quasi‐Biennial Oscillation (QBO) cycle, stratospheric aerosol loading from volcanoes, detrended temperatures at 100 hPa and 500 hPa levels, El Niño–Southern Oscillation (ENSO) events, and various teleconnection patterns. Six of the teleconnection explanatory variables examined are significant to a 95% confidence level in describing the observed total ozone variations. The year‐round total ozone trend in Oslo (±SD) is estimated to −4.2 ± 0.4 % per decade, whereas the winter and summer trends amount to −6.2 ± 0.9 % per decade and −2.4 ± 0.5 % per decade, respectively. The analyses imply that climatic and dynamical changes account for up to a 20% increase in winter ozone trend and a 40% decrease in summer ozone trend. The East Atlantic Jet is the single most dominating teleconnection pattern in Oslo and contributes to about 0.7 percentage points per decade change in summer total ozone trend. The regression analyses demonstrate that the start and termination dates of the ozone record are highly important for the trend results: The summer and winter ozone trend can change by 0.9 percentage points per decade if 1 year is included or omitted from the analyses.

Total ozone measurements from Dobson spectrometer number 8 at Dombås, Norway (62.1°N, 9.1°E), and Oslo, Norway (59.9°N, 10.7°E), from 1940 to 1949 have been examined and reanalyzed. New sets of Bass‐Paur absorption and Rayleigh scattering coefficients have been created, and total ozone values have been recalculated using the new coefficients. Approximately half of the ozone registrations at Dombås were based on direct Sun observations calculated from the CC′ wavelength pairs. The long C′ pair has provided valuable information about the effect of atmospheric aerosols on the ozone measurements. A method for determining aerosol corrections is presented which demonstrates that the monthly mean aerosol error can reach 4% total ozone, normally with a higher correction in the summer than in the winter. Also, the influence of SO₂ on the Oslo ozone measurements is estimated. The D8 ozone series from 1940 to 1949 has been compared to ozone records from other European stations and to the D56 ozone series from Oslo 1978–1998. Studies of old and new Dobson data demonstrate that the annual variation in the ozone layer has changed during the last 50–60 years. The comparison indicates that the ozone decrease is relatively small for the summer months (2.9 ± 1.8%), whereas the average winter and spring values have deceased by 6.1 ± 3.9% from the 1940s to the 1990s. If the postwar increase in tropospheric ozone is taken into account, the depletion of the ozone layer is considerably higher, probably 8–9% for winter/spring. All the reanalyzed total ozone data from D8 for the period 1940 to 1949 are available at the World Ozone and Ultraviolet Radiation Data Center.