Sensitivity of Mesoscale Simulation of Aila Cyclone to the Parameterization of Physical Processes Using WRF Model - Monitoring and Prediction of Tropical Cyclones in the Indian Ocean and Climate Change - page 301

microphysics with pressure drop and surface winds varying in the range of
11-26 hPa and 5-17 m/s with the use of different CM parameterization schemes.
The peak intensity of the storm is well captured by the model with Goddard
scheme but the intensity of the storm before and after reaching the peak intensity
is over-predicted. With Goddard cloud microphysical parameterization scheme,
the model simulation shows minimum central pressure of 968 hPa which exactly
matches with the observation. The model simulated maximum surface wind is
30 m/s compared to 31 m/s in the observation.
Figure 4 presents the temporal variation of vertically averaged mixing
ratio (g/kg) (over the area within the 250-km radius from the storm centre) of
different hydrometeors obtained from model simulations. This shows that the
magnitudes and distributions of hydrometeors (cloud water, rain water, snow,
ice, graupel and vapour mixing ratio; g/kg) significantly varies with different
cloud microphysics parameterization schemes. The results indicate that except
cloud ice, the mixing ratio of other hydrometeors (particularly the cloud water
mixing ratio) is more in the Goddard scheme than in other cloud microphysical
schemes.
4. Summary
The results discussed in the previous section may be summarized as follows:
The model simulated track is equally sensitive to representation of PBL
and cumulus convection in the model. The track of the storm is slightly better
predicted by the model with BMJ-YSU combination of cumulus convection-
PBL schemes.
The model simulated intensity of the storm is more sensitive to the
representation of cumulus convection than the PBL in the model. The intensity
and track of the cyclone is better simulated by model with combination of G3
convection and MYJ PBL scheme.
The model simulated track and intensity is sensitive to the representation
of CM processes in the model. The track and peak intensity of the storm is
slightly better predicted with Goddard cloud microphysics. The mixing ratio
of cloud liquid water is significantly higher in Goddard scheme and may be
the cause of more intense cyclone.
Acknowledgements
The Council of Scientific & Industrial Research (CSIR) and Ministry of Earth
Sciences (MoES) are sincerely acknowledged for providing financial support
to carryout the work. The authors gratefully acknowledge the IMD for the
best-fit track data, NCEP for their analysis datasets and NCAR for the WRF
model.