Sensitivity of numerically simulated tropical cyclone structure and intensity to changes in horizontal resolution

Inadequate representation of important physical processes has been cited as a possible reason for the lack of improvement in the forecasting of tropical cyclone (TC) intensity (Rogers et al. 2006). Therefore, as higher resolution model runs become possible, it is vital to understand how the simulation of physical processes, especially those important to TC intensification, is impacted by changes in horizontal grid spacing. Also, as numerical models are run at progressively higher resolution, it has become increasingly common to eliminate cumulus parameterization in favor of explicit simulation of convection. This study systematically investigates the sensitivity of explicit Weather, Research and Forecasting (WRF) model simulations of Hurricane Ivan (2004) to changes in horizontal grid spacing, when grid spacings between 12 and 2 km are used.

As grid spacing decreases, the minimum central pressure of Ivan deepens, dropping by approximately 20 hPa as grid spacing decreases from 4 to 2 km. However, the 8-, 6-, and 4-km simulations have intensity differences of only around 10 hPa between them. The structure shown by model-simulated radar shows that the eyewall of the tropical cyclone (TC) in the coarser resolution simulations (12- to 6-km) is highly asymmetrical and elliptically-shaped, with two large maxima in reflectivity rotating about the TC center. The 4- and 2-km runs have more circular eyewalls relative to runs with 6-km or larger grid spacing, and the finer resolution runs are characterized by more numerous and larger maxima in reflectivity embedded within the eyewall, as well as better developed spiral bands.

In order to compare these runs at the same resolution, temporally- and spatially-averaged vertical cross-sections are examined, where all fields are interpolated to the same grid spacing. The finer resolution simulations have stronger updrafts and a larger magnitude of subsidence within the eye. However, the warming of the eye, relative to the coarser runs, is confined to the upper levels of the troposphere. The eyewall of the TC in the finer resolution runs slopes radially outward less with height, as the horizontal temperature gradient changes little with height, compared with the coarser simulations. This lack of warming in the lower- and mid-levels of the TC eye indicates a ventilation mechanism at work in the finer resolution runs, acting to mix high equivalent potential temperature (θ_e) air from the eye into the eyewall. Such air could act as a fuel source for buoyant convection within the eyewall (Persing and Montgomery 2003; Eastin et al. 2005b; Yang et al. 2007).

As expected, when shown in the horizontal cross-section at a high temporal resolution, the wind, θ_e, and potential vorticity (PV) fields in the finer resolution simulations tend to have maxima and minima that are smaller in size and larger in magnitude, especially in the 2-km simulation. However, the PV field in the 2-km simulation appears to have several wave-like features moving throughout the eyewall, suggesting that smaller-scale dynamical processes, such as vortex Rossby waves (VRWs) and buoyant convective updraft cores, are at least partially resolved at this grid spacing. VRWs, waves that propagate along a PV gradient, are further explored as a possible ventilation mechanism acting in the lower TC eye. Subjective estimates of the motion of these PV features show their movement to be consistent with the theoretical and observed properties of VRWs.

Overall, the 8- and 6-km runs are similar in their simulation of an asymmetric TC dominated by one to three distinct convective regions at opposite positions within the eyewall. This unrealistic asymmetry indicates a poorly resolved primary, azimuthal circulation. The 4-km simulation, while still asymmetrical, does not exhibit the unrealistic "dumbbell" appearance that is so persistent in the coarser simulations, indicating that the primary circulation is better resolved at this grid spacing. The 2-km run shows numerous, intense maxima in PV and magnitude of vertical motion, as well as an overall appearance of higher-wavenumber disturbances. At this grid spacing, there are indications that an ensemble of updrafts is somewhat resolved.

Mixing between the eye and eyewall has been discussed as an important process in TC intensification, as high θ_e air from the eye would provide a fuel source for buoyant convective updrafts within the eyewall (Persing and Montgomery 2003; Eastin et al. 2005b). However, the mixing of low-momentum air into the eyewall from the eye would act to slow winds within the eyewall, and could be a physical process that would spin down the vortex. Therefore, a simulation in which VRWs are well-resolved and individual convective updrafts are poorly resolved might experience more of the spin down effect, as low-momentum, high θ_e air from the eye would act to slow the primary circulation, but would be under-utilized by the poorly resolved convective updrafts. Such a condition could explain why the 4-km simulation did not experience the rapid deepening of the 2-km simulation, but did exhibit evidence of breaking VRWs. This is consistent with observational studies by Eastin et al. (2005a,b), which found the median diameter of TC buoyant updrafts to be approximately 2 km, and found that 90% of updrafts had diameters of 4 km and smaller.

Regardless of whether an ensemble of convective updrafts is the dominant mechanism responsible, the minimum central pressure of the 2-km simulation is significantly deeper than all other simulations, suggesting that small-scale physical processes important to the intensification of a TC are being resolved in this run that are not well-resolved in coarser runs. Therefore, work is currently underway to perform a 1-km WRF simulation in order to further the scope of this study to even higher resolution.