Characterizing the Relative Importance Assigned to Physical Variables by Climate Scientists when Assessing Atmospheric Climate Model Fidelity

• Published in: Advances in Atmospheric Sciences
• Main Authors: Burrows, Susannah M., Dasgupta, Aritra, Reehl, Sarah, Bramer, Lisa, Ma, Po-Lun, Rasch, Philip J., Qian, Yun
• Language: unknown
• Published: 2022
• Subjects: Southern Ocean
• Online Access: http://www.osti.gov/servlets/purl/1510431; https://www.osti.gov/biblio/1510431; https://doi.org/10.1007/s00376-018-7300-x

Evaluating a climate model’s fidelity – its ability to simulate the observed climate – is a critical step in establishing confidence in the model’s suitability for future climate projections. Because the criteria used by climate modelers in evaluating simulation fidelity are not always documented in a manner that is accessible to model users, and the transfer and dissemination of this expertise to new scientists entering the field is inefficient. Here we report results from a broad community survey studying one aspect of the criteria used in climate model evaluation – the relative importance of different variables in evaluating a global atmospheric model’s mean climate – with respect to several different science goals. Opinions on variable importance are diverse, although there is greater consensus on some variables (e.g., short-wave cloud forcing) than others (e.g., aerosol optical depth). For most variables, consensus does not change significantly with greater climate modelling experience, demonstrating that the establishment of objective criteria for climate model evaluation is still an area of active research. For each science goal, we report community mean importance ratings of selected model variables. Experts adjust their ratings of variable importance in response to the science objective, for instance, rating surface wind stress as significantly more important for Southern Ocean climate than for the water cycle in the Asian watershed. The concise variable lists and community ratings reported here provide a snapshot of current expert understanding of certain aspects of model evaluation, and can serve as a starting point for developing more sophisticated evaluation and scoring criteria with respect to specific scientific objectives.