Changes in daily and monthly surface air temperature variability by multi-model global warming experiments

Abstract

Future changes in daily and monthly surface air temperature variability are associated with temperature extremes and thus will have a large impact on our life and various sectors. In this study, we analyze variability of surface air temperature by 14 CMIP3 coupled ocean-atmosphere general circulation model results. We first assess the 20th century climate in coupled models (20C3M) experiment for the period 1981-2000 with three reanalysis datasets and then show the changes in future climate for the period 2081-2100 under the SRES A1B scenario. The interannual variability of simulated monthly mean surface air temperature agrees with the observations based on three reanalysis datasets. Although there are large model-to-model scatters in the future changes of interannual variability, the ensemble mean of the available model results shows a large decrease in variability over the Northern Hemisphere high latitude region in winter, and an increase over the Northern Hemisphere midlatitudes and over the tropical land region in summer. For the daily temperature variability, the model ensemble mean underestimates the reanalysis data, probably related with less developed synoptic disturbances in the models than in reality. In future, the daily temperature variability is projected to increase over land in the Northern Hemisphere summer and in the tropics, and to decrease over the ocean throughout the year, consistent with the projected weakening of cyclonic disturbances. Inter-model variability is smaller than in the case for the interannual variability of monthly mean temperature. © 2009, Meteorological Society of Japan.