Most climate models predict an excess of precipitation of many regions near the equatorial tropical Pacific Ocean, a problem known as the “double-intertropical convergence zone (ITCZ)”. The excessive precipitation is thought to be due, in part, to deficiencies in the deep convection schemes and other physical parameterizations in the atmospheric component of the models. Here we analyze several AMIP-style large perturbed parameter ensembles of the Community Atmosphere Model version 5 (CAM5) to explore the role of input parameter values on the double ITCZ bias. We identify weaknesses in the model that could be improved by changing parameter values within their allowable range of uncertainty.