Application of potential vorticity tendency diagnosis method to high-resolution simulation of tropical cyclones

Abstract

As the grid spacing in the numerical simulation decreases to ∼1 km, the potential vorticity (PV) structure of the simulated tropical cyclone (TC) is an annular tower of high PV with low PV in the eye and the resulting reversal of the radial PV gradient in the inner core is subject to dynamic instability, leading to complicated small-scale features in the PV field. While the PV tendency (PVT) method has been successfully used to diagnose TC motion in numerical simulations with relatively coarse resolution (∼10 km), it has been found that the PVT diagnosis method fails in the TC simulation with grid spacing of ∼1 km. This study reveals that the failure of the PVT diagnosis method in the high-resolution simulation with grid spacing of ∼1 km arises from the induced small-scale features in the PV field. The high localized PV features do not affect TC motion, but make it difficult to calculate the gradient of azimuthal mean PV when the time interval of the model output is ∼1 h. It is suggested that the PVT method can be applied to high-resolution simulations by increasing the time interval of the model output and/or smoothing the model output to reduce the influence of small-scale PV features.