Temporal Variability of the Atmospheric Zero Trend Level

Variability of many climate forcings (e.g., water vapor, stratospheric ozone, solar radiation, methane, regional land change/land use) has occurred over time periods shorter than the entire period of available atmospheric data, making it difficult, in some cases, to determine forcing effects on atmospheric temperature. Variability in atmospheric temperature trend profiles therefore need to be analyzed over shorter or limited time periods (LTP) to understand the complex feedback processes between atmospheric forcing and temperature change that may not be seen using one long term linear trend. This study initiates this need and analyzes global and tropical averaged radiosonde temperature profiles by determining least square linear running trends over 15, 20 and 25-year LTP. Results include data analyzed from a reduced Radiosonde Atmospheric Temperature Products for Assessing Climate (RATPAC) radiosonde dataset from January 1958 to December 2006. This subset was created to alleviate long term cooling bias. The vertical distribution of temperature trends indicates mainly warming (variable magnitude) in the troposphere and strong cooling (variable magnitude) in the stratosphere over most LTP. In addition, results indicate a variability of the height of the boundary between atmospheric warming and cooling over the defined LTP even in the longer 25-year LTP trends. This level of zero trend or Zero Trend Level (ZTL) has a tendency to decrease over the temporal change of trends, yet it doesn't appear to depend on the magnitude of the cooling in the trends above the ZTL or the magnitude of the warming in the trends below the ZTL. The reduced RATPAC dataset shows global averaged ZTL varies from above 30 hPa (upper limit in radiosonde data) to ~300 hPa for 15-year trends. For 20-year globally averaged LTP trends the ZTL varies over a range of ~80hPa-250hPa, and for the 25-year globally averaged LTP over a range of ~100hPa-200hPa. This indicates the variability of the ZTL is independent of the tropopause. It is possible that causes of ZTL variability may be related to ozone depletion and recovery, water vapor variability or variability in the solar spectral output to name a few. However, a comprehensive study into the effects of these parameters over LTP would be necessary to confirm this hypothesis and in some cases vertical profiles of these forcings are not available. Results from the most current Radiosonde datasets will be presented and discussion on whether the ZTL can be used as a metric for climate studies and/or validating current climate models will be provided.