N.A. Kilifarska Copyright © 2012 Praise Worthy Prize S.r.l. - All rights reserved International Review of PHYSICS, Vol. 6, n. 3 279 Mechanism of lower stratospheric ozone influence on climate Natalya A. Kilifarska Abstract – The aim of this paper is to present an alternative to the widely accepted concepts for factors governing the variability of the stratospheric O 3 and climate. We show that long-term variations of the lower stratospheric O 3 initiate changes in the climatic system – offering a mechanism of this impact and giving some sensitivity calculations, and experimental evidences supporting its foundation. Additionally we show that galactic cosmic rays (GCR) exert an effective influence on the lower stratospheric ozone variability. Re-examination of the efficiency of the lower stratospheric ion-molecular reactions allows us to reveal an existence of autocatalytic cycle, continuously producing O 3 into the lower stratosphere. Our model calculations show that the ozone produced via the ion chemistry could influence substantially the O 3 profile near the tropopause. The revealed causal relations between GCR, lower stratospheric ozone and humidity, and their influence on the surface temperature opens a broader horizon for improvements of our current understanding and expectations for the further evolution of climate system. . Copyright © 2012 Praise Worthy Prize S.r.l. - All rights reserved. Keywords: galactic cosmic rays, lower stratospheric ozone, climate, mechanisms I. Introduction – factors impacting climate variations Climate is defined as an averaged state of meteorological parameters over a long time period (conventionally accepted by the World Meteorological Organisation -WMO to be 30 years). So it is reasonable to assume that climate variability is driven by the long- time periodicities of the forcing factors, instead of their shorter-term fluctuations (the latter add simply a noise to the climate evolution). This aspect of attribution of factors determining climate variations seems to be not well understood by many researchers, who use anomalies (i.e. the short-term deviations from climatology) – instead of the long-term variations of forcing factors [1]. Another problem of climate research is the fact that the dominant part of detection and attribution techniques is linear. It is well known that the linear statistics favours linearly evolving forcings and underestimate the non- liner ones [2]. In Table I we have summarised results from applying non-linear statistics for analysis of the relations between land air temperature (LandT) and several forcing factors calculated in [2]. The examined explanatory variables include: carbon dioxide CO 2 , multi- decadal variations of Sun spot numbers (SSN), galactic cosmic rays (GCR) and ozone. Besides regression coefficients, Table I presents also the capability of each non-linear model to describe the total variability of dependent parameter (i.e. R 2 ) in percents. Note that there are at least two alternatives to the broadly accepted CO 2 as a main driver of climate variability – i.e. the multi- decadal variations of galactic cosmic rays (GCR) and total ozone content (TOZ) in the column with unit cross- section in the atmosphere (see Table I). TABLE I STATISTICALLY SIGNIFICANT NONLINEAR REGRESSION COEFFICIENTS OF LAND AIR T AND TOTAL OZONE WITH DIFFERET FORCING FACTORS CO 2 Ann+11yr Smoot.SSN Ann+22yr Smoot.GCR Ann+11yr Smoot.TOZ Land T 0.86 74% 0.79 62% 0.84 71% 0.88 77% EESC Ann+11yr smoot.SSN Ann+22yr smoot.GCR Ann+11yr smoot.Wcirc Total O3 0.68 46% 0.58 34% 0.74 55% 0.62 0.38% This is quite unexpected result, having in mind the unquestionable leading role of CO 2 , derived by the linear statistical methods. But even more intriguing is the strong relation between the longest ozone records in the world (from Arosa, Switzerland) and the Northern Hemisphere land air temperature from CRUTEM3v data set. Fig. 1 illustrates the capability of CO 2 and Arosa total ozone to describe the variability of LandT. The non-linear function fitting ozone data to that of the land air T is given in [2], while the CO 2 data are fitted by the use of quadratic function. Fig. 1 shows that the ozone model describes better the surface air temperature variability than CO 2 model. The lagged correlation analysis provides information that surface air T responds without time delay only to the impact of the CO 2 and the total ozone [2]. The effect of SSN and GCR is delayed by several years, what means that there is a mediator transferring their influence to the