MM5 has the capability to simulate nested domains of finer resolutions within the primary simulation domain. In MM5, the software is configured to enable up to 9 nests (10 domains) within a particular run. However, due to current hardware resources, the state of the science, numerical stability, and practicality, the number of domains in a simulation is generally limited to four or fewer.
Nesting can be accomplished by either “one-way” or “two-way” methods. In one-way nesting, the coarse-resolution domain simulation is run independently of the nest. That domain then provides the initial and boundary conditions for its nest. There are no feedbacks between the coarse domain and its nest in one-way nesting. Note that the simulated meteorology at the same grid point in the coarse-resolution domain is likely to be different (if only slightly) from the nest in a one-way nest simulation.
Two-way interactive nesting [Zhang et al., 1986; Smolarkiewicz and Grell, 1992] allows for feedback to occur between the coarse-resolution domain and the nest throughout the simulation. The two domains are run simultaneously to enable this feedback. When two-way nesting is used, the portion of the coarse-resolution domain that is simulated in the nest must be discarded. The nesting ratio between domains in MM5 is generally 3:1. (Some other mesoscale meteorology models allow for user-defined nest ratios.) For example, if the coarse domain is a 36-km resolution domain, its nest will be a 12-km resolution domain. This is strictly true for a two-way nest, but is largely held as a standard for the one-way nests in MM5. As discussed in Section 3.5.1, the nest ratio restricts the number of grid points in each dimension of the nest domains to a multiple of 3, plus 1.
Figure 5.3 illustrates a sample simulation domain and its nests. Domain A is a 108-km coarse (or “parent”) domain. Domain B, which has a smaller area coverage and a higher (or finer) resolution, is a 36-km nested (or “child”) domain. Likewise, Domain C is a 12-km child domain to Domain B.
