Improved Track and Intensity Predictions Using Cyclone Bogusing and Regional Assimilation - Monitoring and Prediction of Tropical Cyclones in the Indian Ocean and Climate Change - page 243

Fig. 1: Model domain used in the WRF experiments. Terrain height (km)
above sea level is shaded.

Table 2: WRF model experiments ( in all cases cold start is used )

Experiment name Experiment details

Control (CNTL) Cold start run with interpolated global T382L64 model

analysis as initial conditions (IC) without performing any
cyclone bogusing or data assimilation.

Assimilation IC is prepared by regional assimilation (WRF 3DVAR)

(VAR) without the use of cyclone bogusing.

Bogus Model is run with T382L64 global model interpolated output

(BOG) initialized with bogusing without the use of assimilation.

Bogus +Assimilation IC is prepared with cyclone bogusing followed by regional

(BOGVAR) data assimilation (WRF 3DVAR).

For each of the experiments mentioned in Table 2 forecasted TC track is
extracted. The location is determined by a weighted combination of minimum
central pressure, maximum surface winds, maximum 650 hPa vorticity and
maximum 850 hPa vorticity. Thresholds for vorticity, surface temperature and
700 hPa temperature are also used.

3. Summary of Results

3.1 Impact on Initial Conditions

Figure 2 shows the difference (BOG - CNTL) in the IC due to bogusing for
'Laila', 'Giri' and 'Phet'. Panels in top row show the difference of two