To What Extent Does Discounting 'Hot' Climate Models Improve the Predictive Skill of Climate Model Ensembles?

It depends. The Intergovernmental Panel on Climate Change's (IPCC) Assessment Report Six (AR6) took a step toward ending so‐called 'model democracy' by discounting climate models that are too warm over the historical period (i.e., models that 'run hot') when making projections of global temperature change. However, the IPCC did not address whether this procedure is reliable for other quantities. Here, we explore the implications of weighting climate models according to their skill in reproducing historical global‐mean surface temperature using three other climate variables of interest: global average precipitation change, regional average temperature change, and regional average precipitation change. We find that the temperature‐based weighting scheme leads to an improved prediction of global average precipitation, though we show that this prediction could be overconfident. On regional scales, we find a heterogeneous pattern of error reduction in future regional precipitation. This stands in sharp contrast with the broad regional pattern of error reduction in future temperature projections, though we do find regions where error is not significantly reduced. Our results demonstrate that practitioners using weighted climate model ensembles for climate projections must take care when weighting by temperature alone, lest they produce unreliable climate projections that result from an inappropriate weighting procedure. Plain Language Summary: Climate model ensembles are widely used for risk assessment. However, a few of the most recent generation climate models 'run hot' in the historical period, widening the spread of future global warming. The Intergovernmental Panel on Climate Change's (IPCC) sixth assessment report presents a number of weighting schemes to address this 'hot model' problem, each of which discount models that are 'too hot' in the historical period. However, it is unclear if this procedure is reliable for other quantities of interest. Here we explore the impact of this procedure on global average precipitation change, regional temperature change, and regional precipitation change. We find that while this scheme improves the prediction of global precipitation change and generally improves the prediction of regional temperature, it does not broadly improve regional predictions of future precipitation change. We conclude that users of climate model output must be careful when applying a global temperature‐based weighting scheme in regional impact studies. Key Points: Using historical warming to weight climate models can improve global predictions of annual temperature change and precipitation changeUsing past warming to weight future climate projections has varied effects on regional error reduction depending on the metric of interestClimate model end‐users should use caution when applying a weighting scheme to avoid biased or overconfident assessments of climate impacts [ABSTRACT FROM AUTHOR]