Gravity waves (GWs) play a critical role in transporting momentum, heat, and constituents throughout the atmosphere. Their effects need to be parameterized in general circulation models (GCMs), because GWs are a subgrid scale process (Alexander et al., 2010). This was demonstrated as being essential to model the zonal wind reversal and the anomalous winter-to-summer temperature gradient in the mesosphere/ mesopause region by Holton (1982, 1983), who used the linear saturation theory (Lindzen, 1981) to parameterize the drag by breaking GWs. It is also shown in later studies that GW drag is important for driving the stratospheric quasi-biennial oscillation (Baldwin et al., 2001) and the mesospheric semi-annual oscillation (Dunkerton, 1982; Sassi & Garcia, 1997). GW dissipation is found to induce a net heat flux, which has been determined from a linear theory (Walterscheid, 1981). This is in addition to the diffusion by turbulence induced by GW breaking as formulated in Lindzen (1981). The same idea applies to the transport of constituents, and the wave-induced heat/constituent flux has been formulated in terms of an effective eddy diffusion coefficient (Garcia et al., 2007; Gardner & Liu, 2010; A. Z. Liu, 2009). The formulation by Garcia et al. (2007) is applied in the Whole Atmosphere Community Climate Model (WACCM). Although diffusion from GW breaking is found to play a secondary role in tracer transport in comparison to advection (Holton & Schoeberl, 1988), WACCM simulations by Garcia et al. (2014) demonstrated that the CO 2 distribution above ∼80 km depends sensitively on the