CYCLONE WIND HAZARD ASSESSMENT IN 
COASTAL REGIONS OF BANGLADESH  
Shakil Akther 
Bangladesh University of Engineering and Technology (BUET)  
Telephone: 9665650-80; Fax: (880-2)9665634  
E-mail: shakil@urp.buet.ac.bd  
 
Shakila Kayum  
Bangladesh University of Engineering and Technology (BUET)  
Telephone: 01922856316; Fax: 0541-61486; E-mail: shakila.kayum@gmail.com  
 
Kazi Nusrat Jahan  
Bangladesh University of Engineering and Technology (BUET)  
Telephone: 01674203072; Fax: (880-2)9665634; E-mail: jahannusrat_03@yahoo.com  
 
Sayeed Rokanuzzaman  
Bangladesh University of Engineering and Technology  
E-mail: sayeed_rokan@yahoo.com  
Abstract 
One of the most dangerous cyclone basins of the world is located in the Bay of Bengal and the 
population most affected lives in coastal areas in Bangladesh. Bangladesh often suffers from 
many climate induced disasters such as flood, drought, cyclone etc among which the cyclone is 
the most catastrophic one. The coastal morphology of Bangladesh influences the impact of 
cyclone hazards on the area. Especially in the south western area, cyclone hazards increase the 
vulnerability of the coastal dwellers and slow down the process of social and economic 
development. This includes districts like Chittagong, Noakhali, Patuakhali, Barisal, and Khulna 
where the cyclones strike most in Bangladesh. Cyclones continue to pose a dangerous threat to 
the coastal populations of Bangladesh, despite improvements in disaster control procedures. 
After 138,000 persons died in the April 1991 cyclone, a rapid epidemiological assessment was 
carried out to determine factors associated with cyclone-related mortality and to identify 
prevention strategies. Wind hazard assessment of the cyclones that make landfall in the coastal 
regions of Bangladesh is of great significance. To understand the land falling tropical cyclones 
of Bangladesh and the associated risk and vulnerability in coastal areas is also important and 
accurate results and probability of hazard assessment can be done through the application of 
GIS in the wind speed analysis of cyclones for the purpose. It is hoped that this study will 
contribute to taking proper disaster planning efforts in Bangladesh especially in the mitigation 
phase for the reduction of damage from the cyclone hazard. Future cyclone-associated mortality 
in Bangladesh could be prevented by more effective warnings leading to an earlier response, 
better access to designated cyclone shelters, and improved preparedness in high-risk 
communities.  
Keywords: cyclones, assessment, wind, hazard, vulnerability 
1. INTRODUCTION 
Bangladesh often suffers from many climate induced disasters such as flood, drought, 
cyclone etc and among those natural hazards, cyclone occurs in Bangladesh almost 
every year. About one-tenth of the global total of tropical cyclones occurs in the Bay of 
Bengal and about one-sixth of tropical cyclones born in the Bay of Bengal had landfall 
on the Bangladesh coast (BUET-BIDS, 1993). The coastal morphology of Bangladesh 
influences the impact of natural hazards in these areas. In Bangladesh, even cyclones 
with low intensity can be very deadly at landfall because of the shallow bathymetry of 
the Bay of Bengal, funnelling shape of the coastline with low-lying flat terrain, and very 
high population density (Ali, 1979).  
For this, the study aims to understand the land falling tropical cyclones of Bangladesh 
and the associated risk and vulnerability in coastal areas with the following objectives: 
 To determine the probability of cyclone occurrence in different regions of 
Bangladesh coast. 
 To compute the maximum wind speed for cyclone with of particular return 
period. 
 To simulate maximum wind speed in different regions of Bangladesh coast. 
This study would helps to understand the characteristics of potential cyclones that may 
make land fall in the coastal areas and thereby assessing hazards in these regions.  
2. STUDY AREA  
The coast of Bangladesh has been selected as the study site. For assigning land falling 
locations in this study, the coast of Bangladesh is divided into five zones. These are: 
Khulna, Barisal, Noakhali, Chittagong and Cox’s Bazar. For the study 295 km has been 
considered as depth and 384 km in length.  
 
 
 
Map 2.1: Coast of Bangladesh with subdivisions 
 
3. ORGANIZATION OF DATA 
Tropical cyclones in the Global Tropical Cyclone Climate Atlas (GTCCA) database are 
classified by the following criteria (Table 3.1).  
Table 3.1: GTCCA Classification 
Type Category Wind Speed (knots) 
Tropical Depression TD <34 
Tropical Storm TS 34-63 
Hurricane H >=64 
In this study, the GTCCA classification is used to develop the historical storm dataset 
and climatology for Bangladesh. Occurred hurricanes are also categorized according to 
the Saffir-Simpson hurricane Scale which is described below.   
Table 3.2: Saffir-Simpson Hurricane Scale 
Category Wind speed (Knot) 
1 64-83 
2 84-96 
3 97-113 
4 1144-135 
5 >135 
 
4. METHODOLOGY OF THE STUDY 
There are three objectives in this study and for each objective separate methodology has 
been followed for the successful completion of task.  
4.1 Data collection 
This study is based on secondary data. For the wind field model, primary data are 
collected for maximum wind speed, track speed, and radius of maximum wind. The data 
are collected from different sources such as Bangladesh Meteorological Department 
(BMD), doctoral thesis of Maniruzzaman (1997), statistical year book (2000), doctoral 
thesis of Islam (2006). Angle of landfall of the cyclones has been evaluated from the 
cyclone tracks using the Arc GIS and Corel Draw software. 
4.2 Data Analysis Procedure 
For the systematic analysis of the collected data (collected from secondary sources), all 
the data are compiled to sought out and analyzed through the help of different software 
such as MS Excel, SPSS and GIS etc. The data analysis procedure is stated below- 
 Determining probability of occurrence 
 Probable wind speed for particular return period 
 Grid preparation   
 Developing the model 
 Operational procedure of the model  
 Simulation of cyclone parameters 
 Result analysis 
4.2.1 Determining probability of occurrence: The number of cyclone strikes in all 
five zones during the period 1877-2003 (127 years) is analyzed and frequency data were 
obtained by counting the numbers of tracks that carried cyclone-force winds across the 
coast in each segment. Then for each area recurrence or return period (years) is 
calculated. At last the probability of a cyclone striking in each zone in every year is 
calculated on the basis of it.    
4.2.2 Probable Wind Speed for Particular Return Period: To fulfil the second 
objective, the equation Vmax = 6.3 (Pp - Po)
1/2
, Where Vmax is the maximum wind speed 
and Pp and Po are the atmospheric pressures at the storm periphery and centre 
respectively (Simpson & Riehl, 1981) has been used to estimate central pressure from 
the maximum wind speed.  
4.2.3 Grid Preparation: A digital map of coastal areas was overlaid with a grid of 
approximately 5 Km in size. As wind speed over the sea is not needed for the analysis 
and damage by the cyclones occur due to the wind speed above the land.  
 
 
 
                               
 
 
Figure 4.1: The wind speed grid laid over      Figure 4.2: Coastal regions divided into cells       
  the coastal region map  
4.2.4 Developing the Model: In this study Gradient Wind Model has been used to 
determine the maximum wind speed for the simulated storm. The Arc Macro Language 
(AML) was used to implement the wind model in each cell to calculate the maximum 
wind speed in workstation version of the ARC INFO GIS software.  
4.2.5 Operational procedure of the model: In order to run the module first main menu 
is called by using the Arc option of ARC INFO GIS software. All the inputs of the 
parameters are then entered. Then “Enter landfall point” button is clicked which shows 
the screens of the grids of coastal segment and gives a curser button to enter landfall 
point. After identifying the landfall point a track is drawn on the point from the given 
angle of cyclone track. Then the model is returned to the main menu. Then “Wind 
speed” button is clicked which commands the module to calculate maximum wind 
speed in each cell. After that output result (Photograph 4.5) is saved and the model is 
operates for another simulation (Photograph 4.6). In this way maximum wind speed are 
calculated for 100 simulated cyclones.  
 
 
 
 
                                                                  
                Photograph 4.5                                         Photograph 4.6 
4.2.6 Simulation of Cyclone Parameters: To compute the maximum wind speed in 
each cell, 100 cyclones are simulated. Gradient wind field model required the data of 
pressure drop, radius of maximum winds, the track speed, the angle of track (measured 
clockwise in degrees from the north) and the maximum wind speed.  
4.2.7 Result analysis: All the inputs required for the simulation procedure are given in 
the “arc” command of the Arc Info Workstation software and an output is obtained for 
each simulated cyclone which is stored in the computer. Then the estimated maximum 
wind speeds are analyzed in the Arc Map tool of Arc GIS software. 
 
 
 
 
 
 
 
 
5. PROBABILISTIC ANALYSIS 
5.1 FREQUENCY OF OCCURRENCE 
0 0 0
2
20
17
10 10
7
6
21
27
0
5
10
15
20
25
30
J
a
n
u
a
ry
F
e
b
ru
a
ry
M
a
rc
h
A
p
ri
l
M
a
y
J
u
n
e
J
u
ly
A
u
g
u
s
t
S
e
p
te
m
b
e
r
O
c
to
b
e
r
N
o
v
e
m
b
e
r
D
e
c
e
m
b
e
r
F
re
q
u
e
n
c
y
 
Figure 5.1: Monthly distribution of land falling cyclones from 1877 to 2007 
5.2 NUMBER OF LAND FALLING STORMS  
Table 5.1: Number of land falling storms in different coastal segments during the period from 
1877 to 2003 
Coastal Subdivision Tropical Depression (TD) Tropical Storm (TS) Hurricane (H) 
Barisal 15 9 7 
Noakhali 4 4 1 
Chittagong 2 12 7 
Cox’s Bazar 2 12 6 
Khulna 16 15 5 
Bangladesh coast (Total) 39 52 26 
5.3 RETURN PERIOD  
Table 5.3: Return periods of cyclones in different coastal segments  
Coastal Subdivision Return period (in years) 
Wind speed < 33 mps Wind speed >= 34 mps 
Barisal 5.3 18.1 
Noakhali 15.9 127.0 
Chittagong 9.1 18.1 
Cox’s Bazar 9.1 21.2 
Khulna 4.1 25.4 
Bangladesh coast 1.4 4.9 
 
 
5.4 PROBABILITY OF CYCLONE OCCURRENCE  
Khulna is the most (28%) cyclone prone area and Noakhali is the least (7%) cyclone 
prone area and there is also 92% probability that cyclone will make landfall in this 
country in every year.  
Table 5.4: Probabilities for a cyclone landfall in any one year for each of the coastal segment  
Coastal Subdivision Wind speed < 33 mps Wind speed >= 34 mps Total probability 
Barisal 19 6 25 
Noakhali 6 1 7 
Chittagong 11 6 17 
Cox’s Bazar 11 5 16 
Khulna 24 4 28 
Bangladesh coast (Total) 72 20 92 
5.5 PROBABLE WIND SPEED FOR PARTICULAR RETURN PERIOD 
The probable extreme wind speed for cyclone with of particular return period which is 
the second objective of the study is the basis for developing and adopting cyclone-
resistant building standards (Simpson & Riehl, 1981). 
890
900
910
920
930
940
950
960
970
980
990
1000
0 5 10 15 20 25 30 35 40 45 50 55
Recurrence (Years)
C
e
n
tr
a
l 
P
re
s
s
u
re
 (
m
b
)
   
880
900
920
940
960
980
1000
20 30 40 50 60 70
Vmax (mps)
C
e
n
tr
a
l 
p
re
s
s
u
re
 (
m
b
)
  
Figure 5.5: Recurrence interval (years) for            Figure 5.6: Maximum sustained wind speed 
cyclones when strength is measured by     of the cyclones as a function of central                                                                              
pressure              pressure 
From the figure the central pressures for cyclones that will recur in different period such 
as 5, 10, 20, 30, 40 years etc are obtained. Then the corresponding maximum winds are 
obtained for the above mentioned return periods from the next figure. 
 Table 5.5: Maximum wind speed in respect to Recurrence year 
Recurrence year or 
return period 
Pressure Max wind speed (mps) Max wind speed 
(mph) 
5 973 40 89 
10 949 50 112 
20 926 59 131 
30 912 63 141 
40 903 66 147 
6. CYCLONE WIND ANALYSIS 
6.1 SIMULATION OF WIND FIELD PARAMETERS 
In order to achieve the third objective of the study, values of all the parameters have to 
be simulated. Data of cyclone track have been analyzed according to the coastal 
subdivisions and from this number of simulated cyclones in each of the five zones have 
been determined (Table 6.1). The different percentile range depending on their 
concentration pattern for all the parameters are given in Table 6.2 to 6.5. 
Table 6.1: Number of simulated cyclones in each of the five zones 
Coastal subdivision Frequency in the study period Simulated Cyclone 
Barisal 31 26 
Chittagong 21 18 
Cox’s Bazar 20 17 
Khulna 36 31 
Noakhali 9 8 
Total 117 100 
 
Table 6.2: Percentile value of angle of landfall (Units are in degrees) 
Coastal 
subdivision Min Value 
Percentile 
Max Value 10 25 50 75 90 
Barisal 8 17.4 33 47 69 86 88 
Chittagong 3 7.2 32 45 72.5 81.4 83 
Cox’s Bazar 3 29.1 34.75 45 75 80.6 85 
Khulna 5 11.2 27.5 45 59.75 77.6 345 
Noakhali 15 15 19 45 49.5 57 57 
 
 
 
Table 6.3: Percentile value of track speed (Units are in km/hr) 
Min Value 10 
 
Percentiles 
25 10 
75 15 
90 20 
Max Value 20 
 
 
Table 6.4: Percentile value of maximum wind speed (Units are in km / hr) 
Minimum value 56 
 
 
 
Percentiles 
5 70 
25 85 
35 100 
50 150 
75 163 
80 193 
90 223 
Maximum value 232 
 
Table 6.5: Percentile value of maximum wind speed (Units are in km) 
Minimum value 30 
Percentiles 
10 42 
30 55.2 
40 62.4 
70 65 
85 70 
Maximum value 74 
 
6.2 RESULT AND ANALYSIS  
Using the distributions of historical data of storm parameters given in the above table 
values of all the parameters for 100 cyclones has been simulated and used in the model 
as input. Maximum wind speed for each of the cell for the each 100 simulated storm is 
then estimated with the model. The resulting maximum wind speed of the 100 simulated 
storms are then analyzed to determine the highest, lowest and average value of the 
maximum wind speed.  
 
Highest Maximum Wind Speed (mps)
43 - 47
48 - 50
51 - 52
53 - 54
55 - 57
Lowest Maximum Wind Speed (mps)
3
4
5
6
7 - 9
Average Maximum Wind Speed (mps)
17 - 19
20 - 21
22 - 23
24 - 25
26 - 28
26.36
23.60
24.86
25.77 25.85
22
23
24
25
26
27
Khulna CoxsBazar Chittagong Noakhali Barisal
Zones
A
v
e
ra
g
e
 W
in
d
 S
p
e
e
d
 (
m
p
s
)
2
12
18
27
41
0
5
10
15
20
25
30
35
40
45
Less than 20 21-30 31-40 41-50 Above 51
Wind speed (mps)
F
re
q
u
e
n
c
y
6.2.1 HIGHEST, LOWEST AND AVERAGE OF MAXIMUM WIND SPEED  
Highest, lowest and average value among all the maximum wind speed in each cell 
estimated by the model for the 100 simulated cyclones are given below in figure 6.1,6.2 
and 6.3 respectively. 
 
 
 
 
 
 
Figure 6.1:Simulated highest maximum wind speed       Figure 6.2: Simulated lowest wind                                                 
                                                                                         speed  
 
 
 
 
 
 
       Figure 6.3: Simulated average maximum wind speed 
The average maximum wind speeds of the 100 simulated cyclones at each zone are 
shown in the following Figure 6.4. Figure 6.5 to Figure 6.9 shows the frequency of 
cyclones in different average maximum wind speed categories for a total of 100 
simulated storms at each of the five zones. 
        
 
 
 
 
 
 
 
 
Figure 6.4: The average maximum wind speed                       Figure 6.5: Noakhali                                                     
215
20
23
40
0
5
10
15
20
25
30
35
40
45
Less than 20 21-30 31-40 41-50 Above 51
Wind speed (mps)
F
re
q
u
e
n
c
y
5
15
18
3131
0
5
10
15
20
25
30
35
Less than 15 16-25 26-35 36-45 Above 46
Wind speed (mps)
F
re
q
u
e
n
c
y
1
16
18
22
43
0
5
10
15
20
25
30
35
40
45
50
Less than 20 21-30 31-40 41-50 Above 51
Wind speed (mps)
F
re
q
u
e
n
c
y
3
10
2021
46
0
5
10
15
20
25
30
35
40
45
50
Less than 20 21-30 31-40 41-50 Above 51
Wind speed (mps)
F
re
q
u
e
n
c
y
obtained from the simulation at each zone 
 
 
 
 
 
 
 
Figure 6.6: Khulna                                                  Figure 6.7: Barisal 
 
 
 
 
 
 
 
Figure 6.8: Chittagong                Figure 6.9: Cox’s Bazar 
 
 
CONCLUSION: 
The trend of tropical cyclones hitting the Bangladesh coast is not steady. Among the 
coastal segments, Sundarban (southern part of Khulna), southern part of Bhola, Hatiya 
(Noakhali) and three other pocket region of Cox’s Bazar district is the most vulnerable 
in terms of highest maximum wind speed of cyclone. The results of the analysis  are 
directly applicable to planning, financing and developing a national programme for 
disaster prevention or mitigation. This will contribute to taking proper disaster planning 
efforts in Bangladesh especially in the mitigation phase for the reduction of damage 
from the cyclone hazard.   
 
References 
 Ali, A., (1979) “Storm surges in the Bay of Bengal and some related problems”, PhD 
Dissertation, University of Reading, England. 
 BUET-BIDS, (1993) “Multipurpose Cyclone Shelter Programme (MCPS)”, Final Reports 
vols. i-v, World Bank/UNDP/GoB/Project (BDG/91/025), Dhaka,p.1. 
 Islam, T. (2008) “Cyclone Wind Analysis and Disaster Planning” VDM Verlag Dr. Muller 
Aktiengesellsschaft & Co, Department of Meteorology, University of Texas 
 Maniruzzaman, K.M., (1997) “GIS-Based Disaster Management System for Cyclone 
Disaster Response in Bangladesh.” Department of Urban Engineering, University of 
Tokyo 
 Simpson, R.H. and Riehl, H., (1981) “The Hurricane and its Impact” Louisiana State 
University Press, Baton Rouge and London 
 Bangladesh Meteorological Department (2008) “Cyclone2.doc” Unpublished 
report, Dhaka 
 Global Tropical Cyclone Climatic Atlas (GTCCA), Version 1.0. Fleet Numerical 
Meteorology and Oceanography Detachment, Asheville, North Carolina. Internet 
version was available at http://navy.ncdc.noaa.gov/products/gtcca/gtccamain.html 
(now closed). Cited 15 Sept 2003 
 Bangladesh Bureau of Statistics (2000) “Statistical year book of Bangladesh”, 
Dhaka 
 Holland, G. J. (1980) “An Analytical Model of the Wind and Pressure Profiles in 
Hurricanes” Monthly Weather Review 108 1212 – 1218.  
 Taha, H.A. (2002) “Operation Research” Thomsom Press Ltd., India 
 Talukdar, J., Roy, G. D. and Ahmad, M., (1992) “Living with Cyclone” 
Community Development Library, Dhaka 
 WMO (1986) “TCP-21: Tropical Cyclone Operational Plan for the Bay of Bengal 
and the Arabian Sea” WMO/ TD-No.84, Geneva.