Improvement in Track and Intensity Prediction of Indian Seas Tropical Cyclones with Vortex Assimilation - Monitoring and Prediction of Tropical Cyclones in the Indian Ocean and Climate Change - page 215

Advanced Research WRF, i.e., ARW (Skamarock, 2005) of NCAR and Non-
hydrostatic Mesoscale Model i.e. NMM (Janjic et al., 2001, 2003a,b) of NCEP/
National Oceanic and Atmospheric Administration (NOAA), and Hurricane
WRF, i.e., HWRF of NCEP/NOAA (Gopalakrishnan et al., 2011a) are used
for the simulation/prediction of tropical cyclones. These models have already
demonstrated the skill to produce the fine structures in the inner core and outer
rain bands region of tropical cyclones (Pattanayak and Mohnaty, 2008;
Gopalakrishnan et al., 2011b; Pattanayak et al., 2011; Osuri et al., 2011a etc.).
Krishnamurti et al. (2007) examined the influence of the physical initialization
for the hurricane forecast by using the WRF model and suggested that there is
a positive impact on hurricane intensity prediction. The performance of these
meso-scale models is highly dependent upon the global model analysis and
forecast product which is used as initial and lateral boundary condition to the
meso-scale models. It is being noticed that the global model has large initial
error in the vortex position of the tropical cyclones. These initial errors
deteriorate the performance of the dependable meso-scale models in forecast
mode. Hence, the improvement in track and intensity prediction of tropical
cyclones needs more appropriate depiction of initial vortex and structure of
these convective systems. Furthermore, the sparsity of observations over the
tropical oceans causes either undetected or poorly analyzed initial field. The
initial analysis can be improved with the vortex assimilation of conventional
and non-conventional observations at the initial time of the model integration.
In the present study, the improvement in track and intensity prediction of
tropical cyclones over Indian seas is evaluated based on WRF-NMM model.
Furthermore, the impact of the data assimilation is analysed at the model initial
time as well as in subsequent forecast hour. A brief description of the WRF-
NMM model with the numerical experiments is presented in section 2. The
synoptic situations for the tropical cyclones under consideration in the study
are described in section 3. The results and analysis are given in section 4 and
the conclusions are presented in section 5.
2. Model Description and Numerical Experiments
The WRF-NMM is designed to be a flexible, state-of-the-art atmospheric
simulation system that is efficient for wide range of applications. It is a fully
compressible, non-hydrostatic model with a hydrostatic option (Janjic et al.,
2001; Janjic, 2003a, Janjic 2003b). The horizontal Arakawa E-grid staggering
is used for computational efficiency. The model uses a terrain following hybrid
sigma-pressure vertical co-ordinate. The dynamics conserve a number of first
and second order quantities including energy and enstrophy (Janjic 1984).
Forward-backward schemes are used for the horizontally propagating fast-
waves and implicit scheme is used for the vertically propagating sound waves.
Adams-Bashforth scheme and crank-Nicholson scheme are used for horizontally
and vertically propagating waves. The Geophysical Fluid Dynamic Laboratory