Modulation of natural variability on a trend analysis of the updated cohesive SBUV(/2) total ozone

Abstract

Utilizing the series of Solar Backscattered Ultraviolet (SBUV) instruments, total ozone measurements from the NASA/Nimbus-7 and the operational NOAA polar orbiting satellites, the NOAA/Climate Prediction Center has compiled a long term ‘cohesive’ SBUV(/2) total ozone dataset. The compilations are aimed at achieving a level of quality suitable for trend analysis. The zonal dataset starts in 1979, and extends to 2007. This paper provides an update of the statistical analysis of the above dataset to examine the impact of the Montreal Protocol and its later revisions. The traditional statistical time series trend, or hockey-stick algorithm, is applied to this cohesive total ozone dataset. The algorithm allows for a change in trend in January 1996. Specifically, we examine the net slope post 1996 to see if it is both positive and statistically significant at the 95% confidence level. Our results indicate the following: (1) For 45°N the net slope after 1996, while positive, is not statistically significant at the 95% level of confidence. Thus, we cannot state that the ozone is actually increasing in this area. (2) For 45°S the net slope remains negative, but not statistically significant at the 95% level of confidence. For this region it appears that the best we can state is that the ozone may be levelling off. Our results do indicate, however, that the ozone variability in the southern hemisphere is impacted by the Antarctic oscillation much like the northern hemisphere is impacted by the Arctic oscillation, suggesting a real linkage of these terms. (3) For the integrated domain of 50°N–50°S, the averaging process diminishes greatly the standard errors of the terms. The result is that the net slope for the integrated domain is +1.24% per decade and is statistically significant at the 95% confidence level. Also, the solar coefficient is statistically significant indicating a tropical dominance of this effect.

Acknowledgements

The authors would like to thank Dr Airong Cai for assisting with the algorithm, and Dr Shuntai Zhou and Mr Roger Lin for providing ancillary data. The comments from the reviewers are appreciated. SKY is partially supported by NASA CERES-II Project-L16260. XH, YY, and DJW are partially supported by NSF Grant DMS-0724752.