 The study found that bias correction of GCM data improved the accuracy of hydrologic simulations, particularly for rivers with a strong seasonal pattern. Bias correction reduced the spread and discharge results of GCM-based runs and made them more consistent with observations. It also showed that the bias correction method had a greater effect on mean discharge than extreme events, suggesting that future GCM data sets should be corrected before being used for hydrologic impact studies. This article was authored by F. C. Sperna Weiland, L. P. H. Van Beek, J. C. J. Quaddock and others.