Development and assessment of the non-hydrostatic dynamical core of the University of Tehran Global Atmospheric Model (NH-UTGAM)

Articles in Press

Document Type : Research Article

Authors

1 Space Physics Department, Institute of Geophysics, University of Tehran

2 Department of Space Physics, Institute of Geophysics, University of Tehran

3 Professor, Department of Space Physics, Institute of Geophysics, University of Tehran

4 Associate professor, Department of Space Physics, Institute of Geophysics, University of Tehran

5 استادیار-دانشگاه علوم و فنون دریایی نوشهر

Abstract

The changes made to build a fully compressible global atmospheric model are presented. The non-hydrostatic dynamical core named NH-UTGAM is based on the atmospheric and hydrostatic model developed by University of Tehran, built on the DCASL algorithm for its dynamical core. The distinct feature of the DCASL algorithm is the simultaneous use of a contour and a grid representation for a potential vorticity (PV) like variable, enabling it to achieve effective resolutions for the PV-like vaiable much higher than that of conventional grid-based algorithms. With the inclusion of non-hydrostatic processes, this model is able to represent scales as small as kilometer in horizontal direction. Like the hydrostatic model, the new model uses the hybrid generalized vertical coordinate, with the definition of sigma in terms of hydrostatic pressure, which provides both the possibility of sigma-theta and sigma-pressure vertical coordinates. Therefore, while maintaining its originality, the model will be able to switch from non-hydrostatic to hydrostatic and vice versa with minimal changes. For this purpose, in the vertical direction, an implicit method is used to suppress the vertical propagation of sound waves, which is combined with an explicit method in the horizontal direction, leading to the HEVI (Horizontally Explicit – Vertically Implicit) scheme. After presenting the formulation and the changes made, as a first assessment, the way the new model works in simulating the evolution of synoptic-scale Rossby waves in mid-latitudes is discussed. This is done through implementation of the (dry) Jablonowski–Williamson baroclinic wave test. The performance of the models constructed are then investigated in the face of meso-scale waves such as the mountain wave. This is done by simulating non-hydrostatic gravity waves through the ideal test of the mountain wave presented in the reference tests of the Dynamical Core Model Intercomparison Project (DCMIP). The results obtained for this test are comparable to those by the world-famous models available and indicate the power of the dynamical core of the new model in the detection and time evolution of meso-scale and non-hydrostatic scale waves. Finally, by using a set of full physics parameterization schemes, the non-hydrostatic model has been evaluated in a five-day weather forecast. The output of the rainfall field as a clear example of the model's performance and rainfall forecast has been compared with the results obtained from the hydrostatic version of the model with similar horizontal and vertical resolution. Comparison with two global reference models has also been carried out: GFS (Global Forecast System) with horizontal resolution of 0.5 degrees and ERA5 reanalysis data from ECMWF (European Centre for Medium Range Weather Forecasts) with horizontal resolution of 0.25 degrees. In general, the performance of the developed UTGAM model is acceptable compared to the reference models given their much higher spatial resolution and more accurate settings related to the physical parametrizations of the model such as the boundary layer, surface turbulence fluxes, and cumulus convection.

Keywords

Main Subjects

Journal of the Earth and Space Physics

Articles in Press, Accepted Manuscript
Available Online from 04 January 2025

Files

Share

How to cite

Statistics