Download Some associated features of Severe Thunderstorms in Bangladesh

International Forum on Tornado Disaster Risk Reduction
for Bangladesh
-To Cope with Neglected Disasters13-14 December 2009, Dhaka, Bangladesh
Some associated features of Severe
Thunderstorms in Bangladesh extracted from
conventional method and numerical modelling
Presenter: Md. Abdul Mannan, BMD
Collaboration with: Arjumand Habib, Md. Shah Alam and
S. M. Quamrul Hassan
Contents of presentation
‰ Vulnerability of Bangladesh in context of Severe
Thunderstorm and Tornado
‰ Synoptic or conventional analysis of a Severe
Thunderstorm in Bangladesh
‰ Simulation of a Severe Thunderstorm using WRF
model
‰ Simulation of a Severe Thunderstorm using MM5
model
‰ Conclusion
Tornado occurring areas in the world
Ten deadliest tornadoes
Death Toll
Event
Location
Date
1,300
The Saturia-Manikganj
Sadar Tornado
Bangladesh
(Manikganj)
1989
923
1969 East Pakistan
Tornado
Bangladesh
1969
695
The Tri-State Tornado
United States
(Missouri–Illinois–
Indiana)
March 18, 1925
681
1973 Dhaka Tornado
Bangladesh
1973
600
The Malta Tornado
Malta
1551
500
The Sicily Tornado
Italy
1851
500
The Narail-Magura
Tornadoes
Bangladesh
(Jessore)
1964
500
The Comoro Tornado
Comoro
1951
440
The Tangail Tornado
Bangladesh
1996
400
Ivanovo, Yaroslavl
Tornado
Russia
1984
What is a tornado?
A tornado is defined as a violently rotating column of air
extending from a thunderstorm to the ground. The most
violent tornadoes are capable of tremendous destruction
with wind speeds of 250 mph or more. Damage paths
can be in excess of one mile wide and 50 miles long. 800
tornadoes are reported/Yr in USA.
Life stages: A sequence of
images showing the birth
of a tornado.
‰ First, the rotating cloud
base lowers.
‰ This lowering becomes a
funnel, which continues
descending while winds build
near the surface, kicking up
dust and other debris.
‰ Finally, the visible funnel
extends to the ground, and
the tornado begins causing
major damage.
Types of Nor’wester associated with Severe Thunderstorm
TYPE “A”
Develop over WB in the afternoon
/late evening, speed 48-64 kph. 7080% are of this type.
TYPE “B”
B
A
D
Sub mountainous districts of
North Bengal during Night/EM,
speed 16-32 kph. 10% are of this
type.
TYPE “C”
Hills of N/L, Manipur &
Mizoram, they are very rare.
TYPE “D”
Similar to “B”, but originate near
Khashi hills & moves from N to S.
5% are of this type.
C
Distribution of Wind direction related to severe
thunderstorm in March
¾In March about 42% severe
thunderstorms are
associated with
northwesterly wind
NE
0%
13%
10%
¾ 14% are associated with
westerly wind
¾ 13% are associated with
northerly or southwesterly
winds;
¾ 10% are associated with
northeasterly wind.
¾There is no record of
severe thunderstorms
associated with easterly wind
during this month.
E
2%
SE
6%
13%
S
SW
42%
W
14%
NW
N
Fig: Distribution of Wind direction in
March
Distribution of Wind direction related to severe
thunderstorm in April
¾ In April about 40% severe
thunderstorms are associated
with northwesterly wind,
NE
2%
17%
¾ 21% severe thunderstorms are
associated with westerly wind.
¾ 17% severe thunderstorms are
associated with northerly wind
and
¾ 7% and 9% are associated with
northeasterly and southwesterly
winds respectively.
¾ Severe thunderstorms
associated with easterly,
southeasterly or southwesterly
winds are very rare during this
month.
7%
E
1%
3%
9%
SE
S
SW
21%
40%
W
NW
N
Fig: Distribution of Wind direction in
April
Distribution of Wind direction related to severe
thunderstorm in May
¾ In May about 37%
thunderstorms are associated
with northwesterly
NE
17%
¾ 21% thunderstorms are
associated with westerly and
¾ 17% severe thunderstorms
are associated with northerly
wind respectively.
¾ The percentages of
contribution of northeasterly
and southwesterly winds are
8% and 7% respectively.
¾ Severe thunderstorms
associated with easterly,
southeasterly or southerly
winds are very rare wind during
this month.
E
4%
8%
4%
2%
7%
SE
S
SW
21%
37%
W
NW
N
Fig: Distribution of Wind direction in
May
TIME OF OCCURRENCE
30
25
20
15
TIMES
10
BST
5
0
06-09 09-12 12-15 15-18 18-21 21-24 00-03 03-06
HAIL
Causes of Severe thunderstorms
¾ A lower level warm & moist
layer of air extending from
surface to 5000-7000 ft.
¾ An upper level cold & Dry
layer of air.
¾ One or more of the following
triggering actions:
• Insolation
• Inflow of moist wind from
Bay
• Eastward passage of
Western Disturbance
FORMATION MECHANISM OF NOR’WESTER
¾ A low pressure over
WB & Bihar during
months of March to May.
¾ Moist air from surface
to 7,000 ft.
¾ Cold & dry air within
20 to 30,000 ft .
¾ Inversion
layer
between 10 to 15000 ft.
Cold & dry air
Inversion
Low level warm and
moist air, 65 to 75 %
30,000 ft
*Cold & dry air
20,000 ft
*65 to 75 % RH
7000 ft
*Warm & moist air
L
Surface
FORMATION OF A TORNADO
TORNADO
GETTING STRENGTHS
OVER WATER BODIES
DRAWING DEBRIS
WIND GETS TWISTED
OVER GROUND
Severe thunderstorm in
Bangladesh: 06May 2009
Severe thunderstorm in
Bangladesh: 15May 2009
Synoptic analysis of severe thunderstorm
Synoptic analysis
In this method the following meteorological parameters are
carefully analyzed in each synoptic observation time based
on the surface synoptic and upper air observation over the
area of Bangladesh and its surroundings:
‰ Surface pressure, temperature, relative humidity, dew point
temperature fields
‰ pressure change fields, temperature change fields during 24
hours and its lower time intervals;
‰ Pressure departure fields, temperature departure fields from
the normal values
‰ upper air temperature, humidity and wind field are analyzed
based on the pilot balloon observation and Rawin Sonde
observational data in each observational period.
‰ The stability indices like SI, LI etc. are calculated for
understanding the instability conditions of the troposphere.
Observed meteorological features
a. surface and lower troposphere
™ Existence of a trough of low pressure over Uttar Prodesh and
Bihar with its axis extending West-East direction across the plains.
™ The orientation of the axis and its intensity are affected by the
passage of Western Disturbances in the form of low pressure area.
In such a situation the accentuated axis of trough extends SEwards into Gangetic West Bengal and Bangladesh.
™ Around the trough SW/S’ly wind normally blows from the Bay of
Bengal upto a level of about 3000ft. (1Km) incurring moisture influx
over the region covering Bangladesh, West Bengal and Assam.
™ Quite often through northeast Assam, ENE/NE’ly dry and cooler
wind descends downwards to the West across the foot of the
Himalayas.
™ In such a situation with the existence of ENE/E’ly wind over
extreme north Assam and SW/S’ly wind from the South, East-West
line of wind discontinuity takes place over north Bangladesh and
upper Assam.
b. Upper troposphere
‰ In the upper troposphere, westerlies or northwesterlies
prevailed over Pakistan, North Indian Plains, Bangladesh
and Assam.
‰ The passage of upper air trough in the westerlies or the
jet can often be detected at 30,000ft to 40,000ft (about 9.0
to 12.0 km) levels.
‰ The superposition of upper divergence having
association with a perturbation like a jet, jet-cum-trough or
trough on the convergence having association with a lowlevel ‘low’ or even straight southerly flow upto 3000ft. (1
km) is quite a significant feature on the days of
nor’westers over this area.
Forecasting Technique for severe thunderstorm
The area over which nor’westers occur varies usually from one day to the
other mostly depending on the synoptic situation at the surface and upper
level. Identification of this area with positive threat of the storms is one of the
most difficult problems for the forecasters.
To observe the basic favourable conditions being fulfilled, a forecaster within
a short span of time has to adopt certain quick techniques to make a decision
about the occurrence or non-occurrence of nor’westers. These are:
¾ determine the stability indices;
¾ locate the dynamic heat low over the vulnerable area like north Orissa,
Bihar, Gangetic West Bengal or southwestern part of Bangladesh;
¾ find out the middle and upper-level jet shear;
¾ see whether the positive area in greater than the negative area in the
thermodynamic chart
¾ find out the 12 hour surface pressure fall;
¾ see the low-level moisture influx upto at least 3000ft from the pilot balloon
chart
¾ locate the area of low-level convergence from the pilot balloon charts and
¾ carefully interpret and follow the radar echoes observed by BMD radars
¾ interpret the satellite observation received by satellite receiving systems of
BMD
Gaps and needs
• Occurrence time of thunderstorm is not forecasted
precisely
• Occurrence area of thunderstorm is not forecasted
precisely
• duration of thunderstorm is not forecasted
• Possibility of thunderstorm is not well informed to the
public due to updated communication.
•People’s awareness is not in satisfactory level.
• Research on thunderstorms or severe thunderstorm
related to tornado in Bangladesh is very limited.
A case study: 03 May 2009
May03_0001
May03_0101
May03_0201
May03_0301
May03_0401
May03_0501
May03_0601
May03_0701
May03_0801
May03_0901
May03_1001
May03_1101
May03_1201
May03_1301
May03_1401
May03_1501
May03_1601
May03_1701
May03_1801
May03_1901
May03_2001
May03_2101
May03_2201
May03_2301
Km/hr
80
60
40
20
0
Recorded rainfall (mm)
On 03 May 2009
Recorded wind
on 03 May 2009
03.05.2009: 0000UTC
03.05.2009: 0600UTC
03.05.2009: 0300UTC
03.05.2009: 0900UTC
03.05.2009: 1200UTC
03.05.2009: 1800UTC
03.05.2009: 1500UTC
03.05.2009: 2100UTC
03.05.2009:
0000UTC- 1000ft.
03.05.2009: 0000UTC- 3000
and 5000ft.
03.05.2009:
0000UTC- 2000ft.
03.05.2009: 0000UTC7000 and 10,000ft.
May 03, 2009 at 6000UTC:
1000ft
May 03, 2009 at 6000UTC:
2000ft
May 03, 2009 at 6000UTC:
3000 and 5000ft
May 03, 2009 at 6000UTC:
7000 and 10000ft
May 03, 2009 at 1200UTC:
1000ft
May 03, 2009 at 1200UTC:
2000ft
May 03, 2009 at 1200UTC:
3000 and 5000ft
May 03, 2009 at 1200UTC:
7000 and 10,000ft
Stability indices:
01 May 2009:
a. SI:+ 8.6
b. LI :+ 7.6
02 May 2009:
SI:- 4.9
LI : -5.9
03 May 2009:
SI:-6.6
LI : -3.2
Chittagong
Barisal
Rajshahi
Khulna
Dhaka
Sylhet
Maximum temperature field:
On 03 May maximum temperature
gradient was higher
MaxT_Field: 03May2009
Minimum temperature field:
On 03 May uniform minimum
temperature gradient was across
northwest to northeast over the
country.
MinT_Field: 03May2009
MaxT_24hrsChange_03May09
MinT_24hrsChange_03May09
MaxT_Departure_03May09
MinT_Departure_03May09
WRF modelling for 03 May 2009
WRF model Set up
¾ WRF V3.1 MODEL
¾ May03_09_18hrs
¾ Point in x-direction: 219
¾ Lon: 87.68513- 94.12687ºE
¾ Points in y-direction: 243
¾ Lat: 20.74589- 27.27317ºN
Bangladesh
¾ Vertical levels: 27
¾ Considered output: 60m
¾ No. of Variables: 97
¾ Centered location:
Lat: 24.05ºN & Lon: 90.91ºE
¾ Resolution: 3 km
Bay of Bengal
Domain Area
10m_WF_May03_
09: 0100UTC
10m_WF_May03_
09: 0200UTC
10m_WF_May03_
09: 0300UTC
10m_WF_May03_
09: 0500UTC
10m_WF_May03_
09: 0600UTC
10m_WF_May03_
09: 0700UTC
10m_WF_May03_
09: 0400UTC
10m_WF_May03_
09: 0800UTC
10m_WF_May03_
09: 0900UTC
10m_WF_May03_
09: 1000UTC
10m_WF_May03_
09: 1100UTC
10m_WF_May03_
09: 1200UTC
10m_WF_May03_
09: 1300UTC
10m_WF_May03_
09: 1400UTC
10m_WF_May03_
09: 1500UTC
10m_WF_May03_
09: 1600UTC
03May_0100UTC
03May_0500UTC
03May_0200UTC
03May_0600UTC
03May_0300UTC
03May_0700UTC
03May_0400UTC
03May_0800UTC
Simulated TF(2m) using WRF model: 03 May 2009
03May_0900UTC
03May_1300UTC
03May_1000UTC
03May_1100UTC
03May_1200UTC
03May_1400UTC
03May_1500UTC
03May_1600UTC
Simulated TF(2m) using WRF model: 03 May 2009
03May_0100UTC
03May_0500UTC
03May_0200UTC
03May_0300UTC
03May_0600UTC
03May_0700UTC
03May_0400UTC
03May_0800UTC
Simulated olr (W/m2) using WRF model: 03 May 2009
03May_0900UTC 03May_1000UTC
03May_1300UTC 03May_1400UTC
03May_1100UTC
03May_1500UTC
03May_1200UTC
03May_1600UTC
Simulated olr (W/m2) using WRF model: 03 May 2009
Chittagong
Barisal
Dhaka
Rajshahi
Model Predicted rainfall: 03 May 2009
Model Description
The 5th generation PSU/NCAR mesoscale model
(MM5) is a limited-area,
‰ Non-hydrostatic and terrain-following sigmacoordinate model designed to simulate or predict
mesoscale
&
regional
scale
atmospheric
circulation.
‰ Two way nested domains i.e. coarse domain (D1)
of ~9 km resolutions and fine domain (D2) of ~3 km
resolutions with grid cells 57x57 and 93×93 and
have been prepared using MM5.
‰ These domains have covered the areas 18.67027.6721qN; 84.75-94.8875qE for coarse grid mesh
and 21.760-26.674qN; 87.420-92.901E for fine grid
mesh.
‰ The topography in the model has been obtained
from six resolutions (1q, 30, 10, 5 and 2 minutes,
and 30 seconds) USGS (United States Geological
Survey) land cover data set.
‰ At the boundaries of the coarse domain, 1x1
degree resolution NCEP data have been provided
at every 6 hrs as input.
Model Description continue
‰ The model has then been run using the Grell or Anthes-Kuo
(Anthes, 1977) option for cumulus parameterization and MRF for the
boundary layer parameterization (Hong and Pan, 1996).
‰ The simple ice scheme of Dudhia (Dudhia, 1996) is employed for
explicit treatment of moisture and water vapor. Cloud radiation
interaction is allowed for radiation anticipation. Five layer soil option
is used for soil temperature. The model is run with 22 sigma levels in
the vertical from the ground to the 100 hPa top surface.
‰ The time frames of simulation for five cases in this study are
¾ 0000 UTC of 21 April to 0000UTC of 23 April 2007;
‰ Precipitation outputs of MM5 have been generated at 30 minute
intervals and processed using Grid Analysis and Display system
(GrADS) software for visual and calculation purpose.
21 May_0300UTC
21 May_1500UTC
21 May_0600UTC
21 May_1800UTC
21 May_0900UTC
21 May_1200UTC
21 May_2100UTC 22 May_0000UTC
Simulated PF using MM5 model: 21-22 April 2007
22May_0300UTC
22May_1500UTC
22May_0600UTC
22May_1800UTC
22May_0900UTC
22May_2100UTC
22May_1200UTC
23May_0000UTC
Simulated PF using MM5 model: 21-22 April 2007
21May_0300UTC
21May_1500UTC
21May_0600UTC
21May_1800UTC
21May_0900UTC
21May_2100UTC
21May_1200UTC
22May_0000UTC
Simulated TF using MM5 model: 21-23 April 2007
22May_0300UTC
22May_1500UTC
22May_0600UTC
22May_1800UTC
22May_0900UTC
22May_2100UTC
22May_1200UTC
23May_0000UTC
Simulated TF using MM5 model: 21-23 April 2007
21May_0300UTC
21May_1500UTC
21May_0600UTC
21May_1800UTC
21May_0900UTC
21May_2100UTC
21May_1200UTC
22May_0000UTC
Simulated WF using MM5 model: 21-23 April 2007
22May_0300UTC
22May_0600UTC
22May_0900UTC
22May_1200UTC
22May_1500UTC
22May_1800UTC
22May_2100UTC
23May_0000UTC
Simulated WF using MM5 model: 21-23 April 2007
21 April 2007 (D1)
22 April 2007 (D1)
21 April 2007 (D2)
22 April 2007 (D2)
21 April 2007 -Raingauge
21 April 2007 -Raingauge
Conclusion
‰ Meteorology is a relatively young science and the study of tornadoes is
newer still. Although researched for about 140 years and intensively for around
60 years, there are still aspects of tornadoes which remain a mystery.
‰ Scientists have a fairly good understanding of the development of
thunderstorms and mesocyclones, and the meteorological conditions
conducive to their formation.
‰ Reliably predicting tornado intensity and longevity remains a problem, as do
details affecting characteristics of a tornado during its life cycle and
tornadolysis.
‰ Scientists still do not know the exact mechanisms by which most tornadoes
form, and occasional tornadoes still strike without a tornado warning being
issued.
‰ Numerical modeling also provides new insights as observations and new
discoveries are integrated into our physical understanding and then tested in
computer simulations which validate new notions as well as produce entirely
new theoretical findings, many of which are otherwise unattainable.
‰ Importantly, development of new observation technologies and installation
of finer spatial and temporal resolution observation networks have aided
increased understanding and better predictions.
‰ Therefore more active research are very much essential.