GFS Orographic Gravity Wave Drag

Parameterization developed specifically for orographic and convective source of gravity waves are documented separately. More...

Detailed Description

At present, global models must be run with horizontal resolutions that cannot typically resolve atmospheric phenomena shorter than ~10-100 km or greater for weather prediction and ~100-1000 km or greater for climate predicition. Many atmospheric processes have shorter horizontal scales than these "subgrid-scale" processes interact with and affect the larger-scale atmosphere in important ways.

Atmospheric gravity waves are one such unresolved processes. These waves are generated by lower atmospheric sources. e.g., flow over irregularities at the Earth's surface such as mountains and valleys, uneven distribution of diabatic heat sources asscociated with convective systems, and highly dynamic atmospheric processes such as jet streams and fronts. The dissipation of these waves produces synoptic-scale body forces on the atmospheric flow, known as "gravity wave drag"(GWD), which affects both short-term evolution of weather systems and long-term climate. However, the spatial scales of these waves (in the range of ~5-500 km horizontally) are too short to be fully captured in models, and so GWD must be parameterized. In addition, the role of GWD in driving the global middle atmosphere circulation and thus global mean wind/temperature structures is well established. Thus, GWD parametrizations are now critical components of virtually all large-scale atmospheric models. GFS physics includes parameterizations of gravity waves from two important sources: mountains and convection.

Atmospheric flow is significantly influenced by orography creating lift and frictional forces. The representation of orography and its influence in numerical weather prediction models are necessarily divided into the resolvable scales of motion and treated by primitive equations, the remaining sub-grid scales to be treated by parameterization. In terms of large scale NWP models, mountain blocking of wind flow around sub-grid scale orograph is a process that retards motion at various model vertical levels near or in the boundary layer. Flow around the mountain encounters larger frictional forces by being in contact with the mountain surfaces for longer time as well as the interaction of the atmospheric environment with vortex shedding which occurs in numerous observations. Lott and Miller (1997) [42], incorporated the dividing streamline and mountain blocking in conjunction with sub-grid scale vertically propagating gravity wave parameterization in the context of NWP. The dividing streamline is seen as a source of gravity waves to the atmosphere above and nonlinear subgrid low-level mountain drag effect below.

In a review paper on gravity waves in the middle atmosphere, Fritts (1984) [19] showed that a large portion of observed gravity wave momentum flux has higher frequencies than those of stationary mountain waves. This phenomenon was explained by cumulus convection, which is an additional source of tropospheric gravity waves, and is particularly important in summertime. When the surface wind and stability are weak, the magnitude of the surface drag and the resultant influence of orographically-induced gravity wave drag on the large-scale flow are relatively small compared with those in wintertime (Palmer et al. 1986 [47]). In this situation, the relative importance of cumulus convection as a source of gravity waves is larger. In addition, in the tropical regions where persistent convection exists, deep cumulus clouds impinging on the stable stratosphere can generate gravity waves that influence the large-scale flow.

GWD parameterization in GFS

• Gravity-wave drag is simulated as described by Alpert et al. (1988) [3]. The parameterization includes determination of the momentum flux due to gravity waves at the surface, as well as upper levels. The surface stress is a nonlinear function of the surface wind speed and the local Froude number, following Pierrehumbert (1987) [49]. Vertical variations in the momentum flux occur when the local Richardson number is less than 0.25 (the stress vanishes), or when wave breaking occurs (local Froude number becomes critical); in the latter case, the momentum flux is reduced according to the Lindzen(1981) [40] wave saturation hypothesis. Modifications are made to avoid instability when the critical layer is near the surface, since the time scale for gravity-wave drag is shorter than the model time step.
• The treatment of the GWD in the lower troposphere is enhanced according to Kim and Arakawa (1995) [37] . Orographic Std Dev (HPRIME), Convexity(OC), Asymmetry (OA4) and Lx (CLX4) are input topographic statistics needed (see Appendix in Kim and Arakawa (1995) [37]) .
• Mountain blocking influences are incorporated following the Lott and Miller (1997) [42] parameterization with minor changes, including their dividing streamline concept. The model subgrid scale orography is represented by four parameters, after Baines and Palmer (1990) [5], the standard deviation (HPRIME), the anisotropy (GAMMA), the slope (SIGMA) and the geographical orientation of the orography (THETA). These are calculated off-line as a function of model resolution in the fortran code ml01rg2.f, with script mlb2.sh (see Appendix: Specification of subgrid-scale orography in Lott and Miller (1997) [42]).
• The orographic GWD parameterizations automatically scales with model resolution. For example, the T574L64 version of GFS uses four times stronger mountain blocking and one half the strength of gravity wave drag than the T383L64 version.

Intraphysics Communication

• Routine GWDPS (orographic) is called from GBPHYS after call to MONINEDMF
• Routine GWDC (convective) is called from GBPHYS after call to SASCNVN

Modules
	GFS Orographic Gravity Wave Drag and Mountain Blocking
	This subroutine includes orographic gravity wave drag and mountain blocking.
	GFS Convective Gravity Wave Drag
	This subroutine is the parameterization of convective gravity wave drag based on the theory given by Chun and Baik (1998) [12] modified for implementation into the GFS/CFS by Ake Johansson(Aug 2005).