Are our climate models getting better and how do we know? This is one of the most common questions asked within and outside the climate modelling community. At the heart of answering this question is the development of techniques that assess model quality in simulating the current climate and techniques to link model behaviour in current climate to that in simulations of future climate. Both are challenging tasks. An even more challenging, but equally pertinent question for model evaluation is to identify the causes for some of the model shortcomings in support of making decisions on how to best improve models in the future. 

Much of the focus in model development in recent decades has been on increasing the number of processes included in climate models, with the goal to be able to answer more complex and intricate questions about the climate system and its future evolution. While such developments significantly enhance the capabilities of climate models, the community is increasingly realizing that many of the most common and long-standing model uncertainties arise from the description of water in all its forms and its circulation through the atmosphere-ocean-land system. The best-known of these uncertainties is that in climate sensitivity, which has been shown to be strongly associated with the description of clouds in models.

This presentation will summarize progress in evaluating climate models with a particular focus on water in the atmosphere. We will show that climate models are continuously improving, but that the rate of improvement has slowed. We will analyze opportunities and challenges in improving the representation of key processes in the atmospheric energy and water cycle and highlight potential pathways for accelerating model development in these key areas, both internationally and in Australia.