Climate analyses require one to consider the uncertainties that may arise from the internal and/or external sources such as natural internal variability, errors in observations, uncertainty in forcing or scenario, and inter-model differences. In this respect, during the past decade, Bayesian statistical approaches have been increasingly developed and applied to various climate studies as a method for systematic consideration of the uncertainties. However, although they are all based on the Bayes’ rule, actual applications appear very different depending on which variables to be analyzed, what kinds of assumptions to be made, and how to set up experiments with climate models. In this session, we would like to discuss advantages and shortcomings as well as recent advances and challenges in developing and applying Bayesian approaches for climate studying by sharing ideas and experiences from various applications. All climate studies using Bayesian methods are invited on any temporal and spatial scales, including seasonal to decadal prediction, climate change detection and projection, climate impact assessment, climate model evaluation, and paleoclimate reconstruction.