Climate sensitivity to the lower stratospheric ozone variations N.A. Kilifarska n National Institute of Geophysics, Geodesy and Geography, BAS, 3 Acad. G. Bonchev, 1113 Sofia, Bulgaria article info Article history: Received 19 August 2011 Received in revised form 5 March 2012 Accepted 8 March 2012 Keywords: Climate change Mechanisms Ozone Prediction abstract The strong sensitivity of the Earth’s radiation balance to variations in the lower stratospheric ozone—reported previously—is analysed here by the use of non-linear statistical methods. Our non- linear model of the land air temperature (T)—driven by the measured Arosa total ozone (TOZ)—ex- plains 75% of total variability of Earth’s T variations during the period 1926–2011. We have analysed also the factors which could influence the TOZ variability and found that the strongest impact belongs to the multi-decadal variations of galactic cosmic rays. Constructing a statistical model of the ozone variability, we have been able to predict the tendency in the land air T evolution till the end of the current decade. Results show that Earth is facing a weak cooling of the surface T by 0.05–0.25 K (depending on the ozone model) until the end of the current solar cycle. A new mechanism for O 3 influence on climate is proposed. & 2012 Elsevier Ltd. All rights reserved. 1. Introduction Climate is defined as an averaged state of meteorological parameters over a long time period (conventionally accepted by the WMO to be 30 years). So it is sensible to assume that the climate variability is driven by long-time periodicities of forcing parameters, instead of their shorter-term fluctuations. In the most of recent climatic studies, however, the longest estimated solar variability is its 11 year solar cycle. It seems that the solar impact on the Earth atmosphere is understood mainly through its quasi- decadal periodicity, what is illustrated by the hot scientific debate on this topic. It should be realized, however, that compared to the time scale of climate change, the 11 year solar signal is a ‘‘short- term’’ periodicity. The long-term variability of the solar forcing is formally taken into account in the IPCC periodical reports. How- ever, the divergence between multiple historical reconstructions of the solar irradiance (used as a driver of climatic models) and the more numerous paleoclimatic reconstructions (used as a benchmark for modelling accuracy) are source of great uncer- tainty for the estimated solar signal in the climate variations. On the other hand, even conventional linear statistical meth- ods reveal the stronger connection of the surface air temperature with the multi-decadal variations of solar UV radiations and galactic cosmic rays. Table 1 presents statistically significant at 95% correlation coefficients of near surface air T with two external and two internal for the climate system forcing parameters, calculated first from their annual and then—from their smoothed by 22 year running average procedure values. In both cases the land air temperature (landT) is not smoothed. The enhanced correlations coefficients of the smoothed time series, given in Table 1, are well seen. The most intriguing, however, is the high anti-correlation (  0.75) of landT with the multi-decadal varia- tions of total ozone. The strength of this relation becomes comparable with the one between surface T and CO 2 —equal to 0.82 (see Table 1). The examination of the time series of these parameters, given in Fig. 1, shows that unlike exponentially increasing concentration of CO 2 , the land air T and total ozone have a multi-decadal variations. The relation between them is obviously non-linear, which begs non-linear statistical methods for analysis. A distinct non-linearity in all European climatic time series, reported recently by Miksovsky and Raidl (2005), is generally more apparent in the longer time records and is expressed especially well in multivariate systems like the climate system (Mende and Stellmacher, 2000). In this paper we will show a non-linear statistical model of land air T driven by long and short term ozone variations. We will also briefly analyse the factors affecting ozone variations and will propose a new mechanism for solar influence on climate variability. 2. Data and methods summary 2.1. Data In this analysis we have used the annual values of Northern Hemisphere land air temperature anomalies (CRUTEM3v time series), Sunspot numbers for the period 1900–2011 and the Contents lists available at SciVerse ScienceDirect journal homepage: www.elsevier.com/locate/jastp Journal of Atmospheric and Solar-Terrestrial Physics 1364-6826/$ - see front matter & 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jastp.2012.03.002 n Tel.: þ359 2 979 33 29. E-mail addresses: nkilifarska@geophys.bas.bg, Natalya_Kilifarska@yahoo.co.uk Please cite this article as: Kilifarska, N.A., Climate sensitivity to the lower stratospheric ozone variations. Journal of Atmospheric and Solar-Terrestrial Physics (2012), http://dx.doi.org/10.1016/j.jastp.2012.03.002 Journal of Atmospheric and Solar-Terrestrial Physics ] (]]]]) ]]]–]]]