Geoscience Reference
In-Depth Information
Improved Track and Intensity
Predictions Using Cyclone Bogusing
and Regional Assimilation
R.G. Ashrit*, Manjusha Chourasia,
C.J. Johny and John P. George
National Centre for Medium Range Weather Forecasting
(NCMRWF), A-50, Institutional Area, Phase II
Sector 62, NOIDA, 201 309, U.P. (India)
*e-mail: raghu@ncmrwf.gov.in
1. Introduction
Tropical cyclones (TC) originate and intensify over the oceans where data
coverage is sparse. This leads to inaccurate representation of location and
intensity of tropical cyclones in the initial condition (IC) of the NWP models;
one of the reasons for large errors in the forecast track and intensity. These
errors are reduced by use of bogus vortex in the initial condition (Trinh and
Krishnamurti, 1992; Kurihara et al., 1993; Leslie and Holland, 1995). Kurihara
et al. (1993) proposed a scheme to improve the representation of a TC in the IC
of a high-resolution hurricane model. Satellite data coverage over the ocean
along with high resolution data assimilation tools (3DVAR or 4DVAR) also
provides an opportunity to improve the IC. This study concentrates on the
impact of bogus vortex and regional assimilation in WRF model on track
predictions of some of the 2010 TCs of North Indian Ocean.
The TC bogus initialization method available in pre-processing system of
WRF is designed for cold start using interpolated IC and boundary conditions
from global models (Davis and Low-Nam, 2001). In the global model
(T382L64) analysis, used in this study, the vortex is often broad, weak and
misplaced. WRF model experiments are carried out to improve the
representation of cyclone in the initial conditions. In the next section, details
of the WRF model configuration, different experiments, cyclone bogusing
technique adopted and different sets of experiments performed are discussed.
Section 3 describes the results from the model experiments.
Search WWH ::
Custom Search