In this paper, we introduce a scalable preconditioner within the Community Atmospheric Model (CAM) model that is designed to improve the efficiency of the linear system solves in the implicit dynamics solver. Performing accurate and efficient numerical simulation of global atmospheric climate models is challenging due to the disparate length and time scales over which physical processes interact. Implicit solvers enable the physical system to be integrated with a time step commensurate with processes being studied rather than to maintain stability. The dominant cost of an implicit time step is the ancillary linear system solves, so the preconditioner, which is based on an approximate block factorization of the linearized shallow-water equations, has been implemented within the spectral element dynamical core of CAM to minimize this expense. In this paper, we discuss the development and scalability of the preconditioner for a suite of test cases with the implicit shallow-water solver within CAM, and show how the choice of solver parameter settings affects the behavior of both the solver and preconditioner. We also present the remaining steps to gain efficiency using this solver strategy.