Ice clouds play an important role in regulating the Earth’s radiative budget and influencing the hydrological cycle. Accurately representing ice cloud processes in a global climate model is still a challenging task. Here we introduce the cloud microphysics and macrophysics schemes used in the U.S. Department of Energy E3SM (Energy Exascale Earth System Model) Atmosphere Model version 1 (EAMv1), with a focus on ice cloud representation. Compared to the original model (EAMv0), EAMv1 includes several new treatments of ice cloud macro- and micro- physical processes and their interactions with aerosols. For example, the heterogeneous ice nucleation scheme has been updated, the turbulent kinetic energy that determines the characteristic updraft velocity is calculated differently, and the numerical coupling of the macro- and microphysical processes is more frequent compared to the original model. Model evaluations against satellite retrievals and in-situ measurements show that the microphysical and macrophysical properties of ice clouds (e.g. ice cloud fraction, ice water content, ice crystal number, and effective radius) are generally well represented in EAMv1. We performed sensitivity simulations with incremental changes in physical parameterizations and tuning parameters to identify the cause of changes in the simulated ice cloud properties. We find that apart from the changes in ice nucleation schemes, changes in the simulated number concentration of ice-nucleating particles and supersaturation in EAMv1 and EAMv0 result in large differences in the simulated ice nucleation rate for both cirrus and mixed-phase clouds. A more frequent coupling between cloud macro- and micro- physical processes reduces the numerical error caused by time splitting and has a substantial impact on the simulated ice supersaturation and the occurrence frequency of homogeneous ice nucleation. Other changes in the model, including changes in vertical resolution and parameter tuning, also have a large impact on ice cloud properties. Our ongoing work on further improvements to the representation of ice cloud processes in E3SM will also be discussed.