The hydrological discharge (HD) model of Max Planck Institute for Meteorology is forced by a variety of climate model datasets to investigate the future of discharge in the Euphrates-Tigris Basin. The data include daily time series of surface runoff and sub-surface runoff outputs of 2 global climate models (GCMs) (the SRES A1B scenario simulation of ECHAM5/MPIOM and the RCP 4.5 scenario simulation of MPI-ESM-LR) and the dynamically downscaled outputs of ECHAM5/MPIOM and NCAR-CCSM3 scenario (SRES A1FI, A2 and B1) simulations. The suite of simulations enables a comprehensive analysis of the projected river discharge, and allows a comparison between CMIP5 simulations of MPI-ESM-LR and CMIP3 results from its predecessor ECHAM5/MPIOM on a basin scale. We demonstrate that HD simulations forced with relatively low-resolution GCM outputs are not good at reproducing the seasonal cycle of discharge, which is typically characterized by less flow in the peak season and an earlier peak in annual discharge. Simulations forced with the MPI-ESM-LR yield more robust information on the annual cycle and timing of the annual peak discharge than ECHAM5-forced simulations. In contrast to GCM-forced simulations, high-resolution RCM-forced simulations reproduce the annual cycle of discharge reasonably well; however, overestimation of discharge during the cold season and bias in the timing of springtime snowmelt peaks persist in the RCM-forced simulations. Different RCM-forced scenario simulations indicate substantial decreases in mean annual discharge for the Euphrates and Tigris Rivers by the end of the century, ranging from 19-58%. Significant temporal shifts to earlier days (3-5 wk by the end of the 21st century) in the center time of the discharges are also projected for these rivers. As the basin is considered water-stressed and the region is strongly influenced by water-scarcity events, these unfavorable changes may potentially increase water disputes among the basin countries.