Different General Circulation Models (GCMs) produce different climate change projections, especially when evaluated at subcontinental (regional) scales. When it is time to try and combine their responses into a summary measure, and relative uncertainty bounds, it makes sense to weigh more the output of those GCMs that show better performance in reproducing present day climate (i.e. have smaller bias) and that agree with the majority (i.e. do not seem like outliers). After briefly explaining what GCMs are and how they are utilised in climate change experiments, we present a suite of Bayesian models that combine climate reproductions and projections from 9 different GCMs, by formalizing the weighing criteria of "discounting bias" and "rewarding convergence", in order to provide a final distributional estimate of regional climate change. We start from a simple univariate model (one region at a time), show how to improve it by making it robust, and then go on to present a multivariate version (many regions at a time). Predictive distributions of climate change and other parameters of interest are derived. Last, we point out further possible extensions. This is joint work with Prof. Richard L.Smith, of UNC-Chapel Hill.