A partial correlation analysis of the stratospheric ozone response to 27-day solar UV variations with temperature effect removed

Observational detection of upper stratospheric ozone responses to 27-day so lar ultraviolet (UV) variations is often inhibited by larger, dynamically i nduced ozone variations, which result mainly from the temperature dependenc e of reaction rates controlling the ozone balance. Here we show that partia l correlation coefficients of solar UV and tropical upper stratospheric ozo ne (1-5 hPa) with the temperature effect removed are larger (07-0.8) than a re total correlation coefficients of ozone and solar UV (0.4-0.6). The phas e lag of ozone relative to solar UV is also increased, and the maximum ozon e-UV correlation is obtained at higher altitudes, as compared with correlat ion analyses using ozone and solar UV data alone. Assuming that temperature variations are not forced by solar UV variations, the ozone sensitivity to solar UV and temperature can be calculated using a linear multiple regress ion model. The ozone sensitivity to solar UV is generally independent of ti me periods used for the analysis. However, the magnitude of the ozone sensi tivity to temperature at 1-2 hPa increased significantly from solar cycle 2 1 to solar cycle 22, possibly reflecting long-term changes in the compositi on of the upper stratosphere.