Role of Surface Roughness Length on Simulation of Cyclone Aila - Monitoring and Prediction of Tropical Cyclones in the Indian Ocean and Climate Change - page 255

Fig. 3: Model simulated 24-hr accumulated rainfall valid for 03UTC 25 - 26 May
2009 (landfall day) from (a) TRMM, (b) CNTL, (c) Zo-50 and (d) Zo+50.
the centre over the BoB due to strong southerly surge of the monsoon current
during end of May (onset of Monsoon). This climatological feature is recaptured
by the Zo+50 run by producing the maximum convection (wind speed) to the
right (east) sector of the system centre. Though the Zo-50 run simulates the
maximum convection in the right sector, the strength of the convection is weaker
as compared to CNTL and Zo+50 run. Figure 3 gives the 24-hr accumulated
rainfall valid for landfall day (03 UTC 25 - 03 UTC 26 May 2009) from three
experiments along with TRMM observation. The coastal region of West Bengal
receives more rainfall (showed in a rectangular box) which is well replicated
by Zo+50 experiment.
4. Conclusions
The numerical simulations of TC Aila by varying surface roughness lengths is
shown the importance of the surface roughness length values in a numerical
model. A noticeable impact is observed on the simulation of track of TC Aila
with different values of surface roughness lengths. The higher value of surface
roughness length showed relatively better performance in track simulation.
The impact is seen when the system comes close to the coast. The impact on
intensity is meagre when the system is in developing stage; however, the
intensification during landfall is well captured with higher surface roughness
length. Such findings (Wada and Kohno, 2010) are also seen for the west pacific
ocean Typhoon Fanapi (2010). The results presented here are based on only
one TC case and there is need for more case studies for drawing better
conclusion.
Acknowledgement
The Indian National Center for Ocean Information Services (INCOIS) is
gratefully acknowledged for providing financial support to carry out this
research. The authors also owe thanks to IMD for providing best track positions
of TCs for the validation of model tracks. The authors gratefully acknowledge
the NCEP/NCAR for their input data sets as well MMM division of NCAR for
making available the ARW modelling system on public domain.