The use of Perturbed Physics Experiments (PPEs) is growing in all aspects of Earth System Modeling (e.g., atmosphere, ocean, land surface, hydrology) to understand how model behavior is affected by incomplete representation of Earth System processes. This approach is being used in diverse contexts – in model parameter calibration (selecting “good” parts of parameter space), to aid deeper understanding of model processes, to explore model uncertainty. The PPEs are also combined with multi-model ensembles (e.g., CMIP5) to inform climate projection outcomes. Recently, there is a growing interest in multi-model PPEs to understand contributions to model differences arising from uncertain parameters and from differing model structures. We invite contributions on designing, conducting and analyzing PPEs, based on single or multiple models, to identify key parameters and processes that affect model behavior, to guide model tuning, to identify structural deficiencies in models, and to establish emergent constraints.