P.C.Sinha,
P.Chittibabu and
S.K.Dube
Centre for Atmospheric Sciences
Indian Institute of Technology
New Delhi ¨C110 016, India
Address for correspondence : Prof P.C.Sinha,Centre for Atmospheric Sciences,
Indian Institute of Technology, New Delhi-110 016, India .
Tel:91-11-6591319,
Fax: 91-11-6862037, E-mail: pcsinha@cas.iitd.ernet.in
Abstract: Storm surges
associated with severe tropical cyclonic storms are of common occurrences along
the east and west coast of India. The coastal states of Orissa, Andhra Pradesh
and Gujarat have experienced major surges in the past. The recent 1999 Paradip
super cyclone and 1998 Kandla cyclone are severe most cyclones that caused
extensive damage to the property and loss of lives. Extreme storm surges are
major causes for coastal flooding in these regions.
Damages can be minimized if the surges are forecasted well in advance. The
accuracy of the forecast depends upon the grid resolution and offshore
topography of the region of interest. Keeping this in view we developed fine
resolution location specific models for three coastal states of India which are
vulnerable to storm surges and associated inundation. An attempt has been made
to simulate extreme sea levels along the Andhra, Orissa and Gujarat coasts using
the data of past severe cyclones. The model results reported in the present
study are in good agreement with available observations/estimates.
Keywords: storm surge, numerical model, tropical cyclone, indian coast
Coastal flooding associated with storm surges is a serious concern along 7500 km of Indian coastline (Figure 1). About 60 % of all deaths due to storm surges have occurred in the low-lying arable coastal areas of the countries bordering Bay of Bengal and the Arabian Sea (Murty, 1986; Dube et al., 1997). The frequency of cyclones and severe cyclones in the Bay of Bengal and the Arabian Sea during the 109-year period from 1891 to 1999 is shown in Figures 2a and 2b. The coastal districts of Andhra Pradesh, Orissa and Gujarat (Figure 1) have experienced major surges in the past. The recent 1996 Kakinada cyclone, 1998 Gujarat cyclone and 1999 Orissa super cyclone have caused extensive damage to life and property.
Using nomograms of storm surges Rao (1968) classified the Indian coasts into three types of vulnerability zones to storm surges (Figure 3). For type A coastline, the maximum total water level is less than or equal to 2 m during storm surge events, for type B the amplitude is between 2m and 5m, and for type C the amplitude is greater than 5 m. It can be seen from the figure that certain coastal regions of Andhra Pradesh, Orissa and Gujarat experienced surges greater than 5m.
The
steady increase of population along the coastal belt continuously increases the
potential of the disaster. Even though few cyclones form over the north Indian
ocean the destruction potential is very high because of the presence of shallow
basins in this region. Keeping the above factors in view location specific storm
surge models are developed for
Fig. 1 Map showing India and three costal regions of interest
The model is fully non‑linear and is forced by a dynamic storm model of Jelesnianski and Taylor (1973). The treatment of the coastal boundaries in the model involve a procedure leading to a realistic curvilinear representation for Andhra and Orissa coast models. This coastal representation has added advantage of taking automatically into account the finer resolutions in the shallow regions. This has been achieved by using a variable grid, which leads to a substantial refinement of resolution near the coastline and a coarser resolution in the deeper waters. To represent the two gulfs of Gujarat coast more realistically the coastline boundaries are represented by orthogonal stair steps.
Fig. 2a Landfall of cyclones on the east coast of India (1891-1999)
Fig. 2b Landfall of cyclones on the west coast of India (1891-1999)
Fig. 3 Vulnerability of coastal regions to storm surges
The Andhra coast model covers an analysis area from 12.2 N to 18.2 N and 79.8 E to 83.7 E (Figure 4). The computational domain is divided into a rectangular mesh of 49x20 grid-points having a grid distance of 2 to 15 km along the x-axis and 12.8 km along the y-axis. Using this model the storm surge associated with 1977 Andhra cyclone is simulated. The track of the cyclone is also shown in Figure 4. Numerical experiments are performed with 80 hPa pressure drop and 40 km radius of maximum winds.
Fig. 4 Peak surge contours(m) for 1977 Andhra cyclone
The peak surge contours associated with this cyclone are shown in Figure 4. It can be seen that a maximum surge of about 5m occurred close to the landfall point near Divi which agrees well with the post - storm survey report. The coastal region between Ongole and Kakinada is inundated with a surge more than 2 m.
The analysis
region extends from 18.25 N to 22 N and 84 E to 90 E ( Figure 5). The
model is integrated with a pressure drop of 98 hPa and 40 km radius of maximum
winds. Figure 5 shows the peak surge contours. The model predicted a maximum
surge of 7.8m close to the landfall point. The coastal region between Konark and
Chandbali is affected by a surge of more than 5m. Post- storm survey reports
also show that the surge is more than 7-8 m, near Paradip.
The analysis
region of this model extends from 19.2 N to 23.2 N and 67.8 E to 73.0 E (Figure
6). The grid distance is 5km along both x and y directions. Using this model the
surge generated by 1998 Kandla cyclone is computed with a pressure drop of 40
hPa and 30 km radius of maximum winds. The model is integrated ahead in time
upto 48 hours. Figure 6 shows the model computed peak surge contours and a
maximum surge of about 5 m is predicted to the south of the landfall point. It
may also be seen that at Porbandar and Okha the computed surges are
about 3.5 m and 2 m, respectively. The maximum computed surge at Kandla was
2.1m, which is in
agreement with the reported surge (IMD Report, Feb. 1999).
Fig. 5 Peak Surge contours (m) for 1999 Orissa Cyclone
Fig. 6 Peak surge contours (m) for 1998 Kandla Cyclone
Fine resolution storm surge models have been described for Andhra, Orissa and Gujarat coasts of India. The model results reported in this study are in good agreement with the available peak surge observations/estimates. The results emphasise the suitability of a fine resolution location specific model for a reasonable prediction of surges along the above-mentioned coastal states.
References
[1]
Chittibabu. P. 1999. Development of storm surge
prediction models for the bay of Bengal and the Arabia Sea. Ph.D Thesis. Indian
Institute of Technology, Delhi, India. 263 pp.
[2]
Dube, S.K., A.D. Rao, P.C. Sinha ,T.S.Murty and N. Bahulayan. 1997. Storm surge
in the Bay of Bengal and Arabian Sea : The problem and its Prediction. Mausam
48: 283-304.
[3]
IMD Report. 1999. Report on cyclonic disturbances over
North Indian Ocean during 1998. Abridged report for circulation during the
meeting of WMO/ESCAP panel on Tropical cyclones. RSMC-Tropical cyclones. New
Delhi. 71 pp.
[4]
Jelesnianski, C.P. and A.D. Taylor. 1973. NOAA Technical Memorandum, ERL, WMPO‑3,
33 pp.
[5]
Murty, T.S., R.A. Flather and R.F. Henry. 1986. The storm surge problem in the
Bay of Bengal. Prog. Oceanog. 16: 195‑233.
[6] Rao, N. S. B. 1968. On some aspects of local and tropical storms in the Indian area. Ph.D. Thesis, University of Jadavapur, India.