This paper presents a characteristic-based flux partitioning for the semi-implicit time integration of atmospheric flows. Nonhydrostatic models require the solution of the compressible Euler equations. The acoustic time scale is significantly faster than the advective scale, yet it is typically not relevant to atmospheric and weather phenomena. The acoustic and advective components of the hyperbolic flux are separated in the characteristic space. High-order, conservative additive RungeKutta methods are applied to the partitioned equations so that the acoustic component is integrated in time implicitly with an unconditionally stable method, while the advective component is integrated explicitly. The time step of the overall algorithm is thus determined by the advective scale. Benchmark flow problems are used to demonstrate the accuracy, stability, and convergence of the proposed algorithm. The computational cost of the partitioned semi-implicit approach is compared with that of explicit time integration.1. Introduction. The simulation of mesoscale and limited-area atmospheric flows requires the solution to the compressible Euler equations, of which several formulations are used by operational weather prediction codes [28,29]. Expressing the governing equations in terms of the Exner pressure and potential temperature [18,30,32,33,67] does not conserve mass, momentum, and energy. Alternatively, the equations are expressed as the conservation of mass, momentum, and potential temperature [3,27,59,62,68] by assuming adiabatic flows [15]. Recent efforts [2,10,24,28,55] proposed solving the conservation laws for mass, momentum, and energy [41]. If discretized by a conservative numerical method, this approach yields a truly conservative algorithm and allows for the specification of the true viscous terms. The Euler equations are characterized by two temporal scales-the acoustic and the advective scales. Atmospheric flows are often low-Mach flows where the acoustic scale is significantly faster than the advective scale [11]. The fluid velocities vary from stationary to ∼ 30 m/s within the troposphere [64], resulting in Mach numbers lower than ∼ 0.1. In addition, the acoustic modes do not affect weather phenomena significantly.Explicit time integration methods are inefficient because the largest stable time step is restricted by the physically inconsequential acoustic time scale. Implicit time integration methods can be unconditionally stable; however, they have rarely been