Output of ECHAM with radiation code in single precision

We converted the radiation part of the atmospheric model ECHAM to single precision arithmetic. We analyzed different conversion strategies and finally used a step by step change of all modules, subroutines and functions. We found out that a small code portion still requires double precision arithmetic. We made use of Fortran interfaces to generate code that can be easily changed from double to single precision and vice versa, basically using a simple switch in one module. We compared the output of the single precision version in the CR configuration with observational data and with the original double precision code. The results of both versions are comparable. We extensively tested different parallelization options with respect to the possible performance gain, in both CR and LR configuration. The single precision radiation itself can be about 40\% faster, whereas the speed-up for the whole ECHAM model using the single precision radiation can be about 18\% in the best configuration. We further measured the energy consumption, which could also be reduced by about 17\%. This dataset contains the output data of an ECHAM AMIP experiment 1970-2010 with radiation part of the code in single precision arithmetic (in netcCDF file SPFINALDATA.nc). Corresponding output of a original double precision run is also provided (in netcCDF file DPFINALDATA.nc). The files <em>rad_dp_to_sp.sh</em> and <em>rad_sp_to_dp.sh</em> are conversion shell scripts that perform the necessary changes in the source code.