Impact of Radiance Data Assimilation on Simulation of Tropical Cyclone Thane Using WRF-3DVAR Modelling System - Monitoring and Prediction of Tropical Cyclones in the Indian Ocean and Climate Change - page 382

Table 1: Overview of the WRF model configuration used in the present study

Model configuration Configurati on

Dynamics Non-hydrostatic

Model domain 3°N-26°N, 73°E-103°E

Centre of the domain 15.5°N and 89.5°E

Horizontal grid distance 9 km

Map projection Mercator

Horizontal grid system Arakawa-C grid

Integration time step 30 seconds

Vertical co-ordinates Terrain-following hydrostatic pressure vertical

co-ordinate with 51 vertical levels

3 rd order Runga-Kutta scheme Time integration scheme

Spatial differencing scheme 6 th order centre differencing

Micro physics WSM 3-class scheme

Radiation schemes RRTM for long wave/Dudhia for shortwave

Surface layer parameterization Noaa scheme

Planetary boundary layer physics Yonsei University (YSU) scheme

Cumulus scheme Kain-Fritsch

and boundary conditions from the National Centers for Environmental
Prediction (NCEP) FiNaL (FNL) analyses (1° resolution). The CTRL initial
conditions were used as a first guess field for data assimilation experiments. In
the second experiment known as GTS, the TC is initialized by assimilating all
the available GTS (global telecommunication system) observations into model
first guess. The third experiment (RAD) uses the available radiance observations
along with GTS data. The background error covariances used in assimilation
cycle are prepared for the same 9 km grid domain using one-month WRF
forecasts. A total of four cycles (6 hrly cycle) are performed to initialize the
model prior to the start of the actual model integration. In all experiments, the
model is integrated upto 108 hrs from the initial time (26 th December, 2011).
The model physics was same in all three experiments (Table 1).

5. Result and Discussion

As is commonly known, even with an accurate and optimum data assimilation
scheme, marked improvements for every single forecast are not feasible. But it
can be expected that the forecast error, on an average is reduced with each
assimilation. Hence, assimilations such as these need to be extended to similar
and independent studies in order to draw broad conclusions on the effects of
3DVAR data assimilation. For this reason this study has investigated the effect
of assimilation of radiance observations through WRF-3DVAR on simulation
of the cyclone Thane over BoB. The cyclone Thane discussed in section 2 is
simulated using the WRF and WRF-Var modelling system with a single domain