Documentation for the Global/Regional Assimilation and PrEdiction System (GRAPES-TCM) Model

Note: The following information applies to the 2007 operational version of GRAPES-TCM.

Summary

Model type:	Full physics, three dimensional dynamical model
Model timeliness:	Late
Model status:	Operational or Experimental
Basins run in:	WP
ATCF TECH identifier:	GRPS
Forecast period:	0 to 72 hours
Included in TCGP:	yes
Domain:	321 x 201 points, 80 x 50 deg, 0.25° grid spacing Domain Top: 32 km
Vertical coordinate:	Height-based terrain following coordinate (31 levels)
Grid:	Arakawa C-grid on latitude and longitude projection
Cumulus parameterization:	Kain-Fritch (KF) ETA scheme
Microphysics parameterization:	WSM5 scheme
Boundary layer parameterization:	MRF (non-local mixing)
Radiation parameterization:	RRTM long wave radiation scheme and Dudhia shortwave radiation scheme
Ocean coupling:	no ocean coupling
Initialization method:	MC-3dvar with vortex bogusing method
Initial and boundary conditions:	Analysis and forecast data of NCEP Global Forecast System
Primary contact:	Dr. Huang (huangw@mail.typhoon.gov.cn, Shanghai Typhoon Institute), Chinese Meteorological Agency
Model website:	none
Full documentation:	not available

Brief Technical Description

The GRAPES-TCM is one of the operational regional typhoon models of Shanghai Typhoon Institute, which covers whole WP and SCS domain with grid space of 0.25° resolution. The GRAPES-TCM consists of a set of pre-processors that include vortex initialization, a numerical forecast model (GRAPES-meso), and a set of post-processors. The pre-processors of the GRAPES-TCM include procedures of interpolating GFS spectral data to GRAPES grids, constructing an initial vortex consistent to observation, and assimilating observation data. The initial and lateral boundary data for the GRAPES-meso are the final product of the pre-processors. Using these data, the GRAPES-TCM modeling system predicts the location and intensity of a storm up to three days into the future. The dynamic core of the GRAPES-TCM is originated from GRAPES-meso while some of physical process parameterization schemes are different from the operational GRAPES-meso in CMA, such as cumulus parameterization scheme and air-sea exchange coefficients inside boundary layer. The post-processors of GRAPES-TCM are built upon forecaster’s requests. The major products are hurricane track and intensity at 6-hrly interval, high temporal frequency data for storm location, maximum wind and minimum pressure and other graphical products generated from the model output.

Initialization Method

GRAPES-TCM uses the MC (model constrained)-3dVar data assimilation. This method is described in the following references:

• Liang, X., B. Wang, et al., 2007. Tropical cyclone forecasting with model-constrained 3D-Var. II: Improved cyclone track forecasting using AMSU-A, QuikSCAT and cloud-drift wind data. Quarterly Journal of the Royal Meteorological Society, 133(622): 155-165.
• Liang, X., B. Wang, et al., 2007. Tropical cyclone forecasting with model-constrained 3D-Var. I: Description. Quarterly Journal of the Royal Meteorological Society, 133(622): 147-153.

References

• Chen, D. H., J. S. Xue, et al., 2008. New generation of multi-scale NWP system (GRAPES): General scientific design. Chinese Science Bulletin, 53(22): 3433-3445.

 

 

The information on this page was provided by Dr. Wei Huang on 27 June 2013. It has been reviewed by Dr. Yu Hui.