NG23A-1773
A New Dynamical Core Based on the Prediction of the Curl of the Horizontal Vorticity

Tuesday, 15 December 2015
Poster Hall (Moscone South)
Celal S Konor1, David A Randall2 and Ross P Heikes2, (1)Colorada State University, Atmospheric Science, Fort Collins, CO, United States, (2)Colorado State University, Atmospheric Science, Fort Collins, CO, United States
Abstract:
The Vector-Vorticity Dynamical core (VVM) developed by Jung and Arakawa (2008) has important advantages for the use with the anelastic and unified systems of equations. The VVM predicts the horizontal vorticity vector (HVV) at each interface and the vertical vorticity at the top layer of the model. To guarantee that the three-dimensional vorticity is nondivergent, the vertical vorticity at the interior layers is diagnosed from the horizontal divergence of the HVV through a vertical integral from the top to down. To our knowledge, this is the only dynamical core that guarantees the nondivergence of the three-dimensional vorticity.

The VVM uses a C-type horizontal grid, which allows a computational mode. While the computational mode does not seem to be serious in the Cartesian grid applications, it may be serious in the icosahedral grid applications because of the extra degree of freedom in such grids. Although there are special filters to minimize the effects of this computational mode, we prefer to eliminate it altogether.

We have developed a new dynamical core, which uses a Z-grid to avoid the computational mode mentioned above. The dynamical core predicts the curl of the HVV and diagnoses the horizontal divergence of the HVV from the predicted vertical vorticity. The three-dimensional vorticity is guaranteed to be nondivergent as in the VVM.

In this presentation, we will introduce the new dynamical core and show results obtained by using Cartesian and hexagonal grids. We will also compare the solutions to that obtained by the VVM.