Global changes in indices describing moderate temperature extremes from the daily output of a climate model

Climate change indices derived from daily climate model temperature output are computed and analyzed to study the change of moderate climatic extremes between 1950 and 2100. We used output from the Ensemble Simulations of Extreme Weather Events Under Nonlinear Climate Change (ESSENCE) project, in which a 17-member ensemble simulation of climate change in response to the SRES A1b scenario has been carried out using the ECHAM5/MPI-OM climate model developed at the Max-Planck-Institute for Meteorology in Hamburg. The large size of the data set gives the opportunity to accurately detect the change of extreme climate indicators. We choose indices describing moderately extreme temperatures from the Expert Team on Climate Change Detection, Monitoring and Indices, focusing on percentile-based and duration indices. Additionally, we define some new indices measuring the intensity of daily temperature extremes. To study extremes within different consecutive 50 year time intervals (1950-2000, 2001-2050, and 2051-2100), we use corresponding reference periods (1961-1990, 2011-2040, and 2061-2090, respectively). Trends of the indices within each of the three 50-year periods are estimated using the Mann-Kendall slope estimator. The trends found in our model output for the period 1950-2000 compare well with those reported in the literature from observations. Future trend patterns resemble those from the 1950-2000 period, but have larger amplitudes. This suggests that the pattern of extreme temperature change might already emerge from the weather noise. Outside the tropics, the trend of indices defined from minimum daily temperatures is greater in absolute value than the trend of indicators related to maximum daily temperatures. The trend of the annual temperature range (T-max - T-min) is positive or close to zero over the tropics and negative over the extratropics, indicating that the value of the yearly maximum temperature is increasing faster than the minimum temperature in the tropics and vice versa in the extratropics. Finally, using the empirical distribution, we study the probability distribution functions (PDFs) of the occurrence of cold nights and warm days for nine regions. All PDFs shift in the direction of warming.