Вы находитесь на странице: 1из 100

ISSN - 1810 - 3979

Pakistan
Journal of Meteorology

Vol. 5; Issue 9 July, 2008

Published by
Pakistan Meteorological Department
P. O. Box No. 1214, Sector H-8/2,
Islamabad Pakistan
PAKISTAN JOURNAL OF METEOROLOGY
Chief Editor : Dr. Qamaruzzaman Chaudhry
Editor : Sarfraz Shahid

Board of Advisors

1. Prof. Dr. Qin Dahe, Academician, Chinese Academy of Sciences, Beijing, China.
2. Prof. Toshio Koike, Tokyo University, Japan.
3. Prof. Gianni Tartari, CNR-IRSA, Eu-K2-CNR Committee, Italy.
4. Dr. Zheng Guoguang, Administrator, China Meteorological Administration (CMA).
5. Dr. Bruce H. Raup, CIRES, University of Colorado, USA.
6. Prof. Dr. Shouting Gao, Laboratory of Cloud-Precipitation Physics, IAP/CAS, China.
7. Dr. Muhammad Akram Kahlown, Chairman, PCRWR.
8. Prof. Dr. Dev Raj Sikka, Chief Scientist, IMD, India.
9. Prof. Dr. Samuel Shen, University of Alberta, Canada.
10. Mr. Arif Mahmood, Chief Meteorologist, PMD.
11. Prof. S. K. Dube, Director IIT, India.
12. Dr. Muhammad Qaiser, MSSP, PINSTECH, Islamabad.
13. Dr. Madan Lal Shrestha, Academician, Academy of Sciences, Nepal.
14. Dr. Mona Lisa, Asst. Professor, Dept. of Earth Sciences, QAU, Islamabad.
15. Dr. Ghulam Rasul, Chief Meteorologist, R&D, PMD.

Note: This is a six monthly journal of the Pakistan Meteorological Department


which serves as a medium for the publication of original scientific research work
in the field of Meteorology and related sciences.
Views expressed in these papers are only those of the authors. Director General,
Pakistan Meteorological Department or the Editor does not hold himself
responsible for the statements and the views expressed by the authors in their
papers.

Price Annual Price Per Copy


Inland: Rs. 600/- Rs. 300/-
Foreign: US$ 30
ISSN – 1810 – 3979

PAKISTAN
JOURNAL OF METEOROLOGY

Vol. 5 Issue: 9 July, 2008

Published by
Pakistan Meteorological Department
P.O. Box No. 1214, Sector H-8/2,
Islamabad – Pakistan
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

E d i t o r i a l
.
After its regular publication for the last four years, we are
proudly presenting to our readers the 9th issue of the “Pakistan
P P

Journal of Meteorology,” published in July, 2008.


The Journal is becoming quite popular at home and abroad.
Copies of the current and previous issues are frequently demanded
by the individual researchers, scientific departments and other
organizations.
In order to provide incentive to new authors and research
workers, it has been approved by the Director General, that a token
honorarium of Rs. 2000/- will be awarded to the main authors of each
paper published in the Pakistan Journal of Meteorology with effect
from January, 2008. We congratulate all writers of the 8th and 9th P P P P

issue of the Journal for receiving the honorarium and we hope they
will continue this practice in future as well.
We will be exceedingly grateful, if our esteemed readers
favour us with their analytical views and comments to enable us for
further improvements.

Sarfraz Shahid
Editor

3
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

INUNDATION OF TSUNAMI WAVES AND ITS


RELATION TO THE TSUNAMI RUNUP.
Humeira Hafeez* TP PT

Abstract:
Natural hazards like tsunami cannot be avoided but efforts can be made to minimize
their demolition. The devastating tsunami of 1945 caused a severe damage of lives and
infrastructure across the coastal areas of Pakistan particularly the Makran coast. Any
kind of preemptive measures to reduce the after effects of tsunami and its destruction
needs proper assessment of risk areas and more precisely to how much threat certain
areas are likely to be exposed in case of any such incident. The previous study was
based on the amount of possible destruction that coastal areas of Pakistan are likely to
suffer in case of tsunami generated in the Arabian Sea for which a tsunamagenic
earthquake along Makran Subduction Zone is the most likely cause [3]. Present study
does not focus on any particular location of the globe instead an effort is made to
explore various factors that will affect the inundation of tsunami waves across the coast
and more precisely its relation with run-up height. An effort has also been made in this
document to highlight and point out various efforts that are needed to be done in this
regard to predetermine the possible loss in case of tsunami generated along the Makran
Subduction Zone and to establish an effective “Tsunami early warning system”.

Introduction:
Tsunami is one of the most destructive coastal hazards generating in deep oceans by
volcanic activities, submarine land slides and more usually by an earthquake particularly
generated by a subduction zone. Makran Subduction Zone that is located 100 Km from
Makran coast is a major source of tsunami threat particularly to Pakistan’s coastal cities.
Initially the intensity of tsunami depends upon the size of hump created due to the
deformation of sea floor as a result of triggering source. Than the potential energy of the
tsunami is transformed into the kinetic energy of tsunami waves that propagate in the
oceans at steady speed and as it reaches coast, the kinetic energy of tsunami waves
transforms back into potential energy creating giant waves at the coasts hence causing
enormous damage to the coastal zones. One such incident took place in December 1945
as a result of an earthquake of magnitude 8.5 at Richter scale that was responsible for
loss more than 4000 lives and immeasurable property along the coast of Makran.
Run-up height is the maximum elevation of tsunami waves above mean sea level and
the inundation distance is the amount of horizontal distance tsunami wave travels
landwards from the coastline (Figure 1.). In current study, the run up and inundation
data of Madghaskar and India has been used because of non-availability of the data of
only recorded tsunami that struck coastal areas of Pakistan in 1945.

*
TP TP Pakistan Meteorological Department.
5
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

Figure 1: Measurement of run up height and inundation distance of tsunami.

Inundation Distance:
Table 1 and table 2 is the database showing the run up height and inundation distance of
different location in Madghaskar and India respectively collected after the devastating
tsunami of December 26, 2004.
Table 1: Run up and inundation distance of different locations of Madghaskar
City Run up (m) inundation (m)

Antalaha 2.30 28
Ampahana 2.53 50
Sambava 1.77 71
Sambava 1.91 30
Vohemar 1.60 10
Vohemar 1.48 9
Vohemar 3.19 24
Benarevika 3.51 29
Ambila 2.35 30
Soanierana-Ivongo 2.00 46
Soanierana-Ivongo 1.30 12
Manafiafy (Sainte Luce) 3.10 34
Ankaramany 2.70 35
Ampasimasay 3.20 29
Betanty (Faux Cap) 4.40 34
Betanty (Faux Cap) 2.30 30
Betanty (Faux Cap) 4.80 37
Benaiky 2.90 19
Tolagnaro (Fort Dauphin) 4.10 36
Manakara Be 3.80 58
Manakara North 2.0 25
Manakara Be 3.5 59
Manafiafy (Saint Luce) 3.1 34

6
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
Study related to the tsunami inundation has revealed that tsunami inundation mainly
depends upon the geomorphologic settings of the coastal areas. The degree of
inundation was controlled by coastal geomorphologic features such as sand dunes, cliffs,
coastal vegetation, nature and configuration of the beach and most effective of all is the
angle of elevation of the coastline and the velocity of the invading tsunami surge. The
angle of elevation of a coastline plays a major role in assessing the possible inundation
distance. It has been observed that elevated coastlines are less prone to high degree of
inundation as there is difficulty for the up-rushing tsunami waves due to increasing
gradient and exposure to the rocks as compared to the low lying coasts that are prime
victims of the tsunami inundation and low lying coasts are unable to put up considerable
resistance to the inundating tsunami wave. Narayan et al. (2005a) found that the
elevated landmass played a key role in preventing inundation and minimizing damage
along the coast of Tamilnadu India. Narayan et al. (2005b) found that the degree of
inundation was strongly scattered in direct relationship to the morphology of the
seashore and run-up level on the Kerala coast. Kurian et al. (2006) investigated the
inundation characteristics and geomorphologic changes resulting from the 26December
2004 tsunami along the Kerala Coast, India. They noted that river inlets had been
conducive to inundation; the devastation was extremely severe there, as the tsunami had
coincided with the high tides. Chandrasekar et al. (2006b) reported that the extent of
inundation depends mainly upon the nature of the coastal geomorphology.
Table 2: Run up and inundation distance of different locations of India
City inundation (m) Run up (m)

Azhakappapuram 40 1–2
Kanyakumari 50 1–2
Agastheeswaram 50 1–2
Tamaraikulam 375 4–6
Thengampudur 100 3–4
Madhysoodhanapuram 50 2–3
Dharmapuram 50 2–3
Rajakkamangalam 350 3–4
Kadiapatnam 75 2–3
Manavalakurichi 250 3–4
Lakshmipuram 300 4–5
Colachel 450 5–6

Correlation between Run-Up and Inundation:


Although inundation distance is affected by the geomorphology of the coastal area,
gradient of coastline and nature of the coastal area but run up height is the main factor
that affects the inundation distance. The run-up height and inundation distance seems to
be directly proportional to each other. Narayan et al (2005a) reported the inundation
distance of 20 to 7000m with the run-up height of 1to 12m at Tamilnadu coast of India.
Rasheed et al (2006) reported the inundation distance of 60 to 600m as a result of 2 to
5m run up height at Kerala coast of India.

7
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
Figure 2, is showing the relationship between run up height and inundation distance at
different locations at Madghaskar and it can be noted that the extant of inundation
distance is directly proportional to the amount of run up at the respective coasts.

Figure 2: Run-up v/s inundation distance at Madghaskar


Figure 3, is the plot of inundation distance v/s run up height at several cities of India. It
also shows a direct relationship between run up height and inundation distance. As we
can observe that as run up height increases inundation distance also increases.
In the present study it has been observed that extant of inundation distance is directly
proportional to the run up height and substantial inundation coupled with run up resulted
in the heavy devastation. It has been found that in addition to the amount of run up
height, coastal geo morphology and elevation of the coastal area, river inlets and canals
also proved to be very conducive for substantial amount of inundation.

Figure 3: Run-up vs. inundation distance at India

8
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

Results and Discussions:


It has been observed that although inundation distance is directly proportional to the run
up height but it does not depend solely upon the run up height, it also depends upon the
type of coast encountered. Low lying coasts are often prime victims of the tsunami surge
as they are unable to present sufficient amount of resistance to the path of invading
tsunami waves. Elevated coasts especially with rockey exposure normally recieve
minimal inundation due to the resistance they are able to put up in the way of
accelerating tsunami surge because of their steep slope and rocky behaviour as it is
evident from figure 3 and table 2 that Agastheeswaram and Kadiapatnam being rockey
and steep coastal areas recived very low inundation as compared to the low laying
shallow coast of Thengampudur.
River inlets and canals also contribute heavily to the amount of inundation that a
particular area recieves. In figure 3 and table 2, the anormously inundated
Rajakkamangalam and Manavalakurichi lie at the mouth of the river Panniyar and
Valliyar respectively and hence these rivers also contributed to such a huge amount of
inundation.
In case of any damage assessment or risk mitigation efforts these two factors must be
considered side by side to grasp the complete picture of the scenario. The voulnerability
of any coast depends upon the run up height at that particular coast as well as its
location and geomorphologic setting also.

Recommendations:
For future work in the field of tsunami science and for the development of some kind of
the early warning system following steps are unavoidable and without these no
sustainable efforts in this regard can be realized.
A detailed survey in particular of the coastal cities of Pakistan may be conducted to
find out complete data of the fault lines adjacent to these cities as well as the slope
and elevation of the coastlines to carry out future research and risk mitigation efforts
in the field of tsunami.
A data bank comprising of the complete fault parameters and locations may be
established for near future “Tsunami early warning system” in Pakistan.
For the development of “tsunami early warning system”, seismic data, tidal gauge
data and tsunameter data should be made possible at the earliest for an alert signal in
shortest possible time to minimize the risk of tsunami destruction.
A tsunami warning system should be built on the foundation of international
cooperation in the principal of open, free and unrestricted exchange of data and
information.

Acknowledgement:
The author would like to thank for the gracious support of Mr. Abdur Rasheed, Director,
Institute of Meteorology and Geophysics, Karachi. The author would also like to
acknowledge Mr. Andriamiranto Ravelson, IOGA, Madghaskar and Mr. S. Jaya Kumar

9
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
of Indian Institute of Oceanography for generously providing run-up and inundation
distance for this study without which this effort could not have been materialized.

References:
M. V. RAMANAMURTHY, S. SUNDARAMOORTHY, Y. PARI, V. RANGA
RAO, P. MISHRA, M. Bhat, TUNE USHA, R. VENKATESAN, B. R.
SUBRAMANIAN1, 2005. “Inundation of sea water in Andaman and Nicobar Islands
and parts of Tamil Nadu coast during 2004 Sumatra tsunami” Current science, VOL. 88,
NO. 11.
Chadha, R. K., Latha, G., Yeh, H., Peterson, C. and Katada, T, 2005. “The tsunami
of the Great Sumatra earthquake of M 9.0 on 26 December 2004 – Impact on the east
coast of India”. Curr. Sci., 88, 1297–1301.
Humeira Hafeez, 2007. “The potential of tsunami generation along Karachi and
Makran coast of Pakistan” Pakistan Journal of Meteorology, Vol. 4, Issue 7.
Janaka J. Wijetunge, 2006. “Tsunami on 26 December 2004: Spatial distribution of
tsunami height and the extent of inundation in Sri Lanka” Science of Tsunami Hazards,
Vol. 24, No. 3, page 225.
Narayan J.P., Sharma M. L., Maheswari B.K., 2005a, Effects of medu and coastal
topography on the damage pattern during the recent Indian Ocean tsunami along the
coast of Tamilnadu, Sci. Tsunami Hazards, 23 (2), 9–18.
Narayan J.P., Sharma M. L., Maheswari B.K., 2005b, Run-up and inundation pattern
developed during the Indian Ocean tsunami of December 26, 2004 along the coast of
Tamilnadu (India), Gondwana Res., 8 (4), 611–616.
Kurian N.P., Pillai A.P., Rajith K., Krishnan B.T.M., Kalaiarasan P., 2006,
Inundation characteristics and geomorphological impacts of December 2004 tsunami on
Kerala coast, Curr. Sci. India, 90 (2), 240–249.
Chandrasekar N., Saravanan S., Immanuel J. L., RajamanickamM.,
Rajamanickam G.V., 2006b, Classification of tsunami hazard along the southern coast
of India: An initiative to safeguard the coastal environment from similar debacle, Sci.
Tsunami Hazards, 24 (1), 3–24.
Rasheed A. K.A., Das V. K., Revichandran C., Vijayan P.R., Thottam T. J., 2006,
Tsunami impacts on morphology of beaches along South Kerala Coast, West Coast of
India, Sci. Tsunami Hazards, 24 (1), 24–34.

10
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

PRELIMINARY RESULTS OF
AGROMETEOROLOGICAL SOIL MOISTURE
CONSERVATION EXPERIMENT AT QUETTA
Ghulam Rasul* TP PT

Abstract:
Quetta represents high elevation agricultural plains of Balochistan Plateau where
agriculture is dependent upon rains which are not sufficient to satisfy the water demand
of crops. Soil moisture conservation after any wet spell or season had been the top
priority of the farming community in such extremely arid climates. Making the best use
of conserved moisture for crop production may not be overemphasized. Present study is
also an effort to demonstrate soil moisture conservation through simple orientation of
sowing practices under farmer’s condition without involving any complexity of science.
Two years experiment involved the change of conventional north-south orientation of
furrows to east-west which saved considerable amount of moisture at early stages of
crop development. However, in the top soil layer conserved moisture was less than
slightly deeper depth at 10 cm. The result appeared in the form of better crop stand and
hence economic yield enhancement.

Introduction:
Quetta is located in arid elevated plains of Balochistan. The rainfall amount is not
sufficient to meet the overall crop water requirement. Erratic distribution of rainfall over
time and space is another constraint on the agriculture production due to degradation of
soil, the water holding capacity of the soil is low. The dry and windy conditions results
into loss of moisture from upper soil surface at the faster rate. Water deficiency
generally coincides with reproductive stages of wheat crop. Fulfillment of water demand
of wheat at critical stages of development determines the final yield (Razzaq et. al.
1990).
The rainfall pattern is uni-modal in Quetta. The area receives precipitation only from
western disturbance in winter and summer monsoon does not reach there (Chaudry
1992). That is why Rabi crops especially wheat is generally grown in agricultural plains
surrounding Quetta valley. Crop production in rainfed areas of Balochistan is relatively
more risky as compared to other rainfed plains due to climatic constraints and as such
farmers are conservative in use of inputs on their crops (Sheikh et. al. 1988).
Water is considered to be as valuable as gold in dry climates, therefore, search of better
and better measures of soil moisture conservation is the aim of scientific community
working in such areas. The objective of this experiment was to evaluate different
amounts of moisture conserved under two furrow orientations keeping in view the
characteristics of evaporation from the soil surface exposed to incident solar radiation
and wind irrespective of texture and structure of soil.

*
TP TP Pakistan Meteorological Department.
11
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
Present study was conducted in the experimental field of Agricultural Research Institute,
Quetta. Wheat was planted on the same date in two different furrow orientations. Both
the fields were located side by side at the same level with drill. The furrow orientation in
one field was north- south and east-west in the other. The soil moisture was insufficient
for seed germination at sowing depth therefore both the fields were irrigated before
sowing. Later on crop survived under rainfed conditions.

Data:
Phenological observations:
Routine phonological observations were carried out on both the fields at the same
time and the data on soil moisture reserves as well as appearance of different
phonological phases from emergence to maturity were recorded. Each field was split
into four replications. In each replication 10 plants were selected ten days after
emergence. They were tagged and future observations were carried out on the same
plants. In case of disease attack to a plant or mechanical damage, the plant was
substituted with another plant bearing growth and development rate of average
plants of the entire field. 10% of the plants in a particular phase considered that
phase has set in and 75 % level indicates the phase completion. Subsequently
observations on next phase started. In some cases, two phases at a time were in
progress. For example, booting and flowering have very short inter-phase duration
i.e. one or two days, therefore they coexisted in the field. The summary is given in
table-1.
Table 1: Degree Days and Inter-Phase Period for Wheat at Quetta
Period in days (50%) Cumulative
S. Occurrence of Previous Degree Days
Inter Phase Frequency
No. Phase to 50% Occurrence of (T-5°C)
Present Σ (T-5°C)

1 Sowing – Emergence 31 51.6 51.6

2 Emergence-Third Leaf 32 21.3 72.9


3 Third Leaf-Tillering 23 43.9 116.8
4 Tillering-shooting 57 319.5 436.3
5 Shooting-Heading 12 167.4 603.7
6 Heading-Flowering 13 193.9 797.6
7 Flowering-Milk Maturity 7 93.2 890.8
8 Milk -Wax Maturity 6 116.3 1007.1

9 Wax -Full Maturity 11 214.7 1221.8

12
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
Meteorological Observations:
An agrometeorological observatory is located at a distance of about 100 meters from
the experimental field. The meteorological observations were registered daily at
0800, 1400 and 1700 PST. Soil temperature at 5, 10, 20, 30, 50 and 100 cm depths
were also recorded daily at 0800, 1400 and 1700 PST. Meteorological parameters
recorded at agrometeorological observatory are presented in table-2.

Table 2: Summary of Meteorological Parameters recorded at RAMC Quetta

Temperature (°C)

Month Daily
Wind Speed Wind
Daily Mean Daily Mean Max. Mean
(km/hr) Direction
Min.

Nov. 97-98 9.5 15.0 4.1 6.9 NW

Dec. 97-98 5.4 10.8 -0.1 9.8 N

Jan. 98-99 4.5 10.3 -1.2 11.8 NW

Feb. 98-99 5.9 11.6 0.1 9.3 NW

Mar. 98-99 11.1 17.0 5.3 10.3 N

Apr. 98-99 18.4 25.6 11.1 11.4 NW

May. 98-99 21.9 29.5 14.3 10.7 NW

Jun. 98-99 25.0 32.3 17.6 11.0 NW

Soil Moisture:
Soil moisture in different layers of soil at 5, 10, 20, 30, 40 and 50 cm was measured
after an interval of 10-days. The famous Gravimetric Method was employed in soil
moisture measurements which is on one hand laborious but highly reliable on the
other hand. The soil samples were taken with the help of auger on both sides of the
furrows in both the fields. Following the standard sampling technique the samples
were weighted initially. They were then dried and weighted. Same practice was
repeated twice or thrice till the weight became unchanged. The difference of dry and
wet weights gave the moisture contents of the individual soil samples. The ratio of
moisture contents to dry weight of soil produced the percentage of moisture contents
in the soil.
W eig h t o f w et so il − W eig h t o f d ry so il
% S o il M o istu re = ×100
W eig h t o f d ry so il

13
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
Comparison of moisture contents of soil (%) in both the experimental plots was
carried out in shallow layers of soil (5cm and 10cm depth). The mean deviation
during the growing season was also calculated. The results obtained are shown in
table-3.

Analysis of Data:
The data on phenological observations were complied for 10%, 50% and 75% phase
appearance (50% level is considered to be the phase peak stage). The duration between
50% to 50% consecutive phases was taken as inter-phase period. Inter-phase period
gives the idea that how many days were taken to reach the next phase.
The recorded meteorological parameters were also averaged on monthly basis and
phase- wise. The combined analysis of meteorological and phenological data helped to
understand the rate of growth and development constraints as probabilistic precipitation
is not generally sufficient to meet crop water requirement (Chaudry, 1998). The
comparative analysis of soil moisture contents in both the furrow orientations was
carried out to probe the most feasible way to conserve the available soil moisture. The
routine soil moisture conservation practices (mulching etc.) are not acceptable at
farmer’s level.
Heat units or degree-days concept relates the crop growth and development rate with
accumulated heat. The requirement of heat accumulated for advancement of crop from
one phase to another differs according to crop/variety. In this study, biological zero for
wheat is taken as 5°c and above that summation of temperature is termed as effective
heat units.

K = Σ ( T - Tb) When T > 5°c


K=0 When T ≤ 5°c
Where,
K = Sum of heat units.

T = Mean daily temperature (°C)


Tb = Biological zero 5°C
Biological zero may be defined as the temperature below which growth of crop does not
take place. The crop in both the plot matured at the same time. The accumulated heat
units are in given in table - 1

Results and Discussion:


The results presented in Table-3 show that the soil moisture contents in east-west
orientation are slightly more than north-south furrows in identical measurements and
exposed to similar environmental conditions. Crop canopy was denser in east-west
furrows as compared to north-south furrow orientation plot. Physical health and colour
of plants was better in the former case than the latter one. The grain yield is also higher
in east-west case as compared to north-south. Rasul (1993) stated that optimum soil
14
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
moisture contents at the time of emergence gave rise to good initial establishment of
wheat crop ground water rain fed conditions.
Table 3: Soil Moisture Measurement in both the experimental plots at Quetta
Month / Decade North–South East–West Furrows EW / NS Ratio
Furrows (NS) (EW)
5 cm 10 cm 5 cm 10 cm 5 cm 10 cm
I 11.7 10.3 13.3 10.9 1.14 1.06
December II 12.0 11.7 12.9 13.5 1.08 1.15
III 9.9 7.5 9.7 10.3 1.0 1.37
I 9.5 7.5 9.3 10.0 1.0 1.33
January II 12.7 9.9 13.6 12.3 1.07 1.24
III 16.9 17.2 17.5 18.9 1.04 1.10
I 17.0 18.4 17.9 19.9 1.05 1.08
February II 13.9 15.5 14.7 16.9 1.06 1.09
III 14.9 15.3 15.8 16.4 1.06 1.07
I 14.5 16.3 14.9 17.7 1.03 1.09
March II 9.5 11.4 10.8 12.6 1.14 1.11
III 8.0 10.4 9.9 11.8 1.24 1.13
I 7.8 9.3 9.7 10.7 1.24 1.15
April II 4.7 8.7 5.5 9.9 1.17 1.14
III 5.1 6.2 7.4 9.3 1.45 1.5

It is important to mention here that furrow height was not abnormally greater than the
routine farming practice. The height of furrows after drill-seeding was not more than 8-
10 cm in both the cases which were leveled naturally due to wind and precipitation
effect later on when crop canopy was around 30-40%. The crop canopy (%) showing the
soil cover by plants and general condition of crop in successive development process is
given in table – 4. The impact of furrow change was prominent at the initial stages of
crop growth and development till crop plants did not fully cover the soil. Ultimately
good initial establishment of the crop plants resulted into better yields as the water
requirements partially fulfilled at successive development stages throughout the life
cycle of the crop.
Orientation of furrows determines the exposure of surface area to the direct sun. The
area exposed to sun in case of east-west orientation is less than the north-south because
of lower solar angle during winter. Also the duration of exposure is longer in the latter
case. Since September equinox (days and nights are equal), the solar angle decreases
gradually and reaches its minimum i.e. winter solistice, the lowest towards southeast on
22nd January. Due to freezing temperatures, the crop growth reaches the slowest stage
15
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
and northerly winds generally blow during December and January. Under such
conditions, the east west furrows remain less exposed to sunshine and wind which are
the major factors accelerating the evaporation rate of moisture from the soil surface.
Table 4: Visual Observations on Crop Condition of Wheat at Agricultural Research Institute
Quetta.
North – South Furrows East – West Furrows
Physical Physical
Date Canopy Canopy
Appearance Appearance
Emergence Completed Emergence Completed
after 10 days 5% Good 5% Good
after 10 days 10 % Good 12 % Good
after 10 days 18 % Good 25 % Better
after 10 days 25 % Fair 30 % Good
after 10 days 32 % Fair 40 % Good
after 10 days 45 % Good 55 % Better
after 10 days 55 % Good 70 % Better
after 10 days 68 % Good 75 % Better
after 10 days 75 % Good 80 % Good
after 10 days 85 % Good 87 % Good
90 % Good 90 % Good
Average number of
47 53
grains per spike

Yield (kg / hectare) 2854 3049

The evaporation process is the function of solar angle and the exposed area of the
surface subjected to evaporation as well as the dryness of adjacent environment
(Chaudhry et al 2004). The surface which is exposed to sun for longer time, warms up
faster and to higher temperature resulting into higher rate of evaporative loss of moisture
from soil surface directly and shallow layers of soil indirectly.

Conclusions:
The results of this experiment show that some more moisture may be conserved by
avoiding solar exposure and wind effect in upper layers of soil by sowing the wheat crop
in east-west furrows instead of north- south furrow; in the dry climates like Quetta and
adjoining arid plains. The analysis disclosed that moisture reserves at 10cm depth are
much better than shallow layer of soil. The economic gain of conserved moisture may

16
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
leave significant effect on future performance of crop through better early establishment
and better harvest.

Refereces:
Ahmed, M. (1989). “Prospects of Dry land Farming in Pakistan “Progressive Farming
Vol.9(3):5-9.
Chaudhary, Q.Z. and G. Rasul 2004: Climatic Classification of Pakistan. Science
Vision, Vol. 9 No. 1-2 (Jul-Dec, 2003) & No. 3-4 (Jan-Jun, 2004). Pp. 59-66.
Chaudhry, Q. Z. (1988). “Rainfall and Soil Water Averages and Probabilities & other
Pertinent Agroclimatic Data for Quetta.” PMD Report.
Chaudhry, Q. Z. (1992). “Analysis and Seasonal Prediction of Pakistan Summer
Monsoon Rainfall.” Ph.D. Dissertation. University of Phillipines.
Rasul, G. (1992). “Water Requirement of Wheat Crop in Pakistan, Journal of Engg. &
Applied Sciences. Vol.3 No.1, pp. 65-79.
Rai, L. A. (1999). “Weather and Wheat Crop Development during 1998-99 in Quetta.”
Research Report PMD/NAMC.
Razzaq, A., N. I. Hashmi, B. Khan, B. R. Khan and P. R. Hobbs. (1990). “Wheat in
Barani Areas of Punjab: A Synthesis of On-Farm Research Results 1982-88.”
PARC/CIMMYT Paper 90-2.
Sheikh, A.D., D. Byerlee and M. Aseem. (1988).” Analytic of Barani Farming Systems
of Northrern Punjab: Cropping Intensity, Crop-Livestock Interactions and Food Self
Sufficiency.” PARC/CIMMYT Paper No. 88.2, Islamabad: PARC.

17
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

PREDICTION OF SOIL TEMPERATURE BY AIR


TEMPERATURE; A CASE STUDY FOR FAISALABAD.
M. Fahim Ahmad*, Dr. Ghulam Rasul*
TP PT

Abstract:
The agriculture productivity depends upon the soil type as well as other meteorological
parameters. Soil temperature greatly affects most of the biological and physical
processes occurring underground. Soil temperature plays an important role during the
life cycle of the plants right from germination, root extension, emergence to the
reproductive stage. The measurement of soil temperature is a cumbersome business and
not as simple as air temperature be noted by single thermometer. Separate
thermometers are required to measure the temperature of the soil at different depths.
Sowing depth of seeds is commonly taken between 5 cm and 10 cm. major root density
at reproductive stage of most of the cereal crops vary from 20 cm to 30 cm keeping in
view the importance of these depths, an attempt has been made to estimate soil
temperature from air temperature of that particular location at that time. In this study
the regression equations between the seasonal daily mean air temperature and seasonal
daily mean soil temperature have been derived for the Faisalabad region. Faisalabad
soil is mainly loamy with significant proportion of silt. These soils represent vast
agricultural plains of the Indus river system in irrigated parts of Punjab, NWFP and
Sindh. These equations will be used to predict the daily mean soil temperature with the
help of air temperature. Correlation coefficients between these two variables were
derived. The value of R2 has different value for different season. Highest value (0.86) in
spring season at 10 cm depth and the lowest value (0.32) in winter season at 20 cm
depth. The least value in summer season may be due to vegetation cover. These
regression equation can also be used for others stations with some changes. These
equations can also be used for the estimation of missing data.

Introduction:
Pakistan has agriculture based economy. About 22 % of the GDP and 44.8 % of total
employment is generated in agriculture. 65.9 % of the population living in rural areas
depends upon the agriculture directly or indirectly for their livelihood. Contribution of
agriculture in the export is also substantial (Year Book, 2005-2006, Ministry of Food,
Agriculture and Live Stock, Pakistan).
The quality and quantity of the crop depends upon many factor including the soil. Soil
temperatures significantly affect the budding and growth rates of plants. For example,
with the increase in soil temperature, chemical reactions speed up and cause seeds to
germinate. Soil temperature plays an important role in the decomposition of soil. It also
regulates many processes, including the rate of plants development and their growth Soil
temperature also plays an important role for setting life cycles of small creatures which

*
TP TP Pakistan Meteorological Department.
19
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
live in the soil. For example, hibernating animals and insects come forth of the ground
according to soil temperature.
Soil temperature is also regarded as sensitive climate indicator and stimulus. Scientists
use soil temperature data in the research on variety of topics including climate change.
(Sharratt et al., 1992). Soil temperature anomalies also directly affect the growth and
yield of agricultural crops. For example cool spring season, soil temperature in shallow
layers delays corn development and on the other hand warm, spring season, soil
temperature contributes to an increase in corn yield (Bollero et al, 1996). Soil
temperature although is integral in many ecosystem processes, is costly and its
observation is difficult (Shannon E. Brown et al, 2000)
Soil temperature also determines the state of the water in the soil whether it will be in a
liquid, gaseous, or frozen state. In cold soils, the rate of decomposition of organic matter
will be slow because the microorganisms function at a slower rate, as a result the color
of soil will be dark. In tropical climates intense heating causes increased weathering and
the production of iron oxides, which results into the reddish color of soil.
(http://www.globe.gov/tctg/sectionpdf.jsp?sectionId=92#page=10)
HT TH

The objective of this study is to develop a general methodology for the prediction of soil
temperature using air temperature. A significant relationship exist between averaged
daily air temperature and observed daily soil temperature at 10 cm depth (Zeng et al.
1993) . Soil temperature also correlate with the soil respiration and soil moisture, there
is an exponential correlation between soil temperature and soil respiration (Tang Xu Li
et al, 2006). Air temperature can be used to determine soil temperature and the exchange
of gasses between soil and atmosphere.
Successful prediction of soil temperature with the help of Air temperature will lead to
minimize the time, cost, and equipment maintenance necessary for on site monitoring
and will help researcher to use data from other sources also. Instead of the fact that
Pakistan is an agricultural country, we have to import wheat and also some type of
cotton. This is because the production is not sufficient as compared to our demand. The
per acre yield is very less. The reason behind this less productivity may be inappropriate
manners of cultivation, old fashioned equipment and also the use of old species of
certain crops. With the change in climatic conditions and availability of water one must
switch to the specie of crop best suited to climate are more resistant to droughts.
According to agroclimatic classification the 2/3 area of Pakistan is under arid climate
(Chaudhry, Q.Z. and G. Rasul; 2004). This study can be used as a base to determine the
condition of soil (temperature, moisture and respiration) to determine which type of crop
will be beneficial and when it should be grown to be more productive. It can also be
used to examine the soil temperature variations indicated by the Global and Regional
Circulation or Climate Models and can be used to check the validity of models to
describe the surface and atmosphere energy interaction.

Data and Methodology:


Pakistan Meteorological Department has five Regional Agromet. Centres (RAMC's),
which record the data of different soil parameters. Seventeen (17) years data of Air
temperature and soil temperature were used in this study. These data are recorded at
RAMC Faisalabad w.e.f 1989 to 2005. Daily Mean Air temperature has been calculated
20
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
from the average of Daily Maximum and Daily Minimum air temperature. The Daily
Mean Soil temperature was calculated form the average of three observation taken daily
at 0300 hr, 0900 hr and 1200 hr (GMT). Daily Mean Air temperature is assigned the
variable name ‘X’ while the daily Mean Soil temperature is described as ‘Y’.
The correlation coefficients between Daily Air temperature and Daily Soil temperature
were calcuted using the Pearson’s Formula for correlation coefficient. Actually it
indicates the strength and direction of a linear relationship between two random
variables. In general statistical usage, correlation or co-relation refers to the departure of
two variables from independence (Faqir Muhammad)
(n∑ xy − ∑ x∑ y)
R=
n(∑ x − (∑ x)2 ) n(∑ y2 − (∑ y)2 )
2

Where n represents the number of entries.


By calculating the correlation coefficient it becomes clear that how much one variable
depends upon the other. Regression equation were formed in the form
y = mx + b
Where ‘m’ is the slope of the trend line and ‘b’ is the intercept of trend line with y-axis.
Slope ‘m’ was calculated using the formula
(n∑ xy − ∑ x∑ y )
m=
(n∑ x 2 − ∑ x∑ x)
And intercept ‘b’ is calculated with the help of the following formula

b = y − mx
Where y and x are mean of soil temperature and air temperature of respectively.
Standard errors of Mean (SEM) for each season were also calculated. This represents
the mean error in the predicted values.
Table 1: Characteristics of soil of the area under the study (RAMC, Faisalabad)
Depth (cm) Sand% Silt% Clay% PH% Colour Type
Moist dry
dark/dark
0-23 50 26 25 8.1 Pale brown silt loam
brown
Dark yellowish
23-64 52 48 28 8.2 Pale brown Silt loam
brown
64-92 45 26 29 8.3 Brown Pale brown Silt loam
Yellowish Very pale
92-150 46 31 25 8.4 Silt loam
Brown brown

21
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

Results and Discussion:


The values of Correlation coefficients (R2) calculated for the four seasons are
P P

0.366999539, 0.864590297, 0.732919235 and 0.817574878 respectively for winter,


spring, summer and autumn. The value of R2 is quiet good for all seasons except the
P P

winter season, for which its value is less then 0.5. The reason for the lower values of
correlation coefficient in autumn and winter may be that in these seasons the duration of
sunshine hour is less. The air temperature changes rapidly as compared to soil
temperature, so that coarse correlations exist during these seasons. Wheat crop is grown
on the site where the data is collected. Winter season (Dec-Feb) is the growing period of
wheat crop. In this season the plants of wheat comes out of the ground. So the area has a
quite good vegetation cover. This change in vegetation cover will also cause to lower
the R2 value, because change in vegetation cover can impact the amount of energy
P P

reaching the soil (Balisky and Burton, 1993).


Table 2: R2 values for different seasons at 10cm, 20cm and 30cm depth
P P

10cm 20cm 30 cm
DJF 0.367 0.319 0.424
MAM 0.865 0.839 0.829
JJA 0.733 0.668 0.594
SON 0.818 0.851 0.727

2
R values (Seasonal Daily Mean)

1
0.9
0.8
0.7
0.6 10cm
R2

0.5 20cm
0.4 30 cm
0.3
0.2
0.1
0
DJF MAM JJA SON
Seasons

Figure 1: Comparison of R2 Values for different Seasons and Depths.


P P

22
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

Seasonal Classification:
Winter = December, January and Feberuary
Spring = March, April and May
Summer = June, July and August
Autumn = September, October and November
Four equations calculated for the four seasons are listed below.
For 10 cm Depth:
YDJF =0.0.739x + 5.234 (1)
YMAM =1.157x + 0.9935 (2)
YJJA =1.346x – 5.738 (3)
YDJF =1.231x - 3.407 (4)
For 20 cm Depth:
YDJF =0.614x + 6.084 (5)
YMAM =0.014x + 2.15 (6)
YJJA =1.075x + 1.304 (7)
YSON =1.199x – 2.089 (8)
For 30 cm Depth:
YDJF =0.5697 x + 6.894 (9)
YMAM =0.9701x + 2.298 (10)
YJJA =0.946x + 4.723 11)
YSON =1.087x + 0.2566 (12)
For winter, spring, summer and autumn season respectively
While the values of SEM for the four seasons of year for the depth of 10 cm and 20 cm
are given in the following table.
Table 3: Standard Error of Mean

10 cm 20 cm 30 cm

DJF 1.36 1.26 0.93


MAM 1.24 1.21 1.195
JJA 1.39 1.30 1.34

SON 1.14 0.98 1.30

23
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
The standard error of a method of measurement or estimation is the estimated standard
deviation of the error in that method. Specifically, it estimates the standard deviation of
the difference between the measured or estimated values and the true values.
These general equations can be applied to find out the soil temperature of most of the
sites, but individual coefficients and constants will have to be calculated first.
The graphs of the plotted data and trend line are given below.
Seasonal(DJF/10cm)
Seasonal(MAM/10cm)

25
40

y = 0.7388x + 5.2341 y = 1.1517x + 0.9935


R2 = 0.367 R2 = 0.8646
35
20
Soil Temperature ( oC)

Soil Temperature ( oC)


30

15

25

10
20

5
15
5 10 15 20 25
15 20 25 30 35 40
Air Tem perature (oC) Air Temperature (oC)

Figure 2: Trend line of air temperature and soil temperature in winter and spring season (10 cm
depth)
In the Fig. 2 in the graph of winter season (Dec-Feb) the dispersion in the data is quite
large. Due to this large dispersion the value of R2 is very low. Winter season (DJF) is
the growing period of wheat crop, and this vegetation may prevent soil temperature to
rise as rapidly as air temperature. Because of this contrast the correlation coefficient R2
is also very weak. The reason of large dispersion seams to be the different moisture
contents and canopy cover in different years which caused the variation in thermal
conductivity patterns in the same soil. In spring season (MAM) R2 is the largest as
compared to the remaining three seasons. Data points are very close to the trend line.
This high correlation may be due to the fact that April and May are dry months as
compared to the other seasons and the dry soil is less resistant to heat penetration as
compared to wet soil. The insolation received on the soil through crop canopy if any
(wheat harvested in April-may and fields are directly exposed to sun) and heated up the
soils simultaneously. It may be noted that specific heat of dry soil less than wet soil
therefore heat exchange between air and soil is much faster.
In summer season the dispersion in the data is not very large but due to some outliers,
the value of correlation coefficient is not very high as compared to that in the spring
season. One more reason of this relatively weaker correlation may be effect of monsoon
season. In Faisalabad monsoon starts from July and last till the end of September. Due
to difference of specific heat of air and water the heat conductivity characteristics
widely vary on temporal and spatial scale. The correlation coefficient for the autumn
season is quite good. After the wet spell of monsoon season in autumn the dry soil and
relatively moderate temperature causes the correlation coefficient to become large as
compared to JJA.
24
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

Seasonal (JJA/10cm) Seasomal(SON/10cm)

45 35

y = 1.3455x - 5.7377
y = 1.2311x - 3.4074
R2 = 0.7329
R2 = 0.8176
40
30
Soil Temperature ( oC)

S oil Te m pe ra ture ( o C )
35

25

30

20
25

20 15
20 25 30 35 40 45 15 20 25 30 35

Air Tem perature (oC) Air Temperature (oC)

Figure 3: Trend line of air temperature and soil temperature in summer and autumn seasons (10
cm depth)
Seasonal (DJF/20cm) Seasonal (MAM/20 cm)

25 40
y = 0.6142x + 6.0842 y = 1.0145x + 2.1496
2
R = 0.3189 R2 = 0.8392
35
20
Soil Tem perature ( o C )
Soil Te m pe ra ture (C )

30

15 25

20
10

15

5 10
5 7 9 11 13 15 17 19 21 10 15 20 25 30 35 40
Air Temperature (C) Air Temperature (oC)

Figure 4: Trend line of air temperature and soil temperature in winter and spring seasons (20
cm depth)
Above Fig shows the trend of DJF and MAM for 20 cm depth. In the correlation
coefficient is very weak. The effect may be the vegetation cover which plays its role to
generate larger lags in air and soil temperatures. However, some data quality is also
questionable. In spring season the correlation is very strong. The reason may be that
wheat reaches its maturity in this season, allows more solar radiation to penetrate into
the soil. The value of R2 is less then for 10 cm depth for the same season, because the
thermal conductivity takes longer to deeper depths.
In summer season the land is almost bare because in these months the wheat crop is
harvested in the area. Air temperature changes rapidly as compared to soil temperature,
in the monsoon season some time the air temperature drops suddenly due to cloud
formation but the soil does not respond so quickly. That’s why R is less then the
previous season.

25
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
Seasonal (SON/20cm)
Seasonal (JJA/20cm)

35
45
y = 1.1994x - 2.0895
y = 1.0754x + 1.3091
R 2 = 0.8512
R2 = 0.6677
40
30
S o i l T e m p e r a tu r e ( o C )

Soil Te m pera ture ( o C )


35
25

30

20
25

20 15
20 25 30 35 40 45 15 17 19 21 23 25 27 29 31
o
Air Temperature ( C) Air Temperature (oC)

Figure 5: Trend line of air temperature and soil temperature in summer and autumn seasons (20
cm depth)
(DJF/30cm) (MAM/30cm)

20 35

y = 0.5697x + 6.8948 y = 0.9701x + 2.298


R2 = 0.4237 R2 = 0.8296

30
Soil Tem perature ( o C)
Soil Temperature (oC)

15

25

10
20

5 15
5 10 15 20 15 20 25 30 35

Air Temperature (oC) Air Temperature (oC)

Figure 6: Trend line of air temperature and soil temperature in winter and spring seasons (30
cm depth)
(JJA/30cm) (SON/30 cm)

40 35
y = 1.0875x + 0.2566
y = 0.9463x + 4.7231 R2 = 0.7269
R2 = 0.5937
35
Soil Temperature ( o C)

30
Soil Temperature ( oC)

30
25

25
20

20
20 22 24 26 28 30 32 34 36 38 40 15
15 20 25 30 35
Air Temperature (o C)
Air Te m perature (oC)

Figure 7: Trend line of air temperature and soil temperature in summer and autumn seasons (30
cm depth)

26
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
When comparing it with the 10 cm and 20 cm the correlation coefficient is little bit large. Also
the standard error of the mean is less then that for the less depth The value of correlation
coefficient is less as compared to that of the 10 cm and 20 cm depth. However SEM remains
almost the same.
There is a very notable change in correlation coefficient in summer season when we
compare it with the 10 cm and 20 cm depth. In autumn the correlation coefficient
decreases with increase in depth.
The measurement of soil temperature is very important for the selection of suitable
variety of crop to be grown in particular areas. Pakistan has very diverse climatic
regions ranging from burning deserts in the south to the frozen cryoshpere in the north.
Due to excessive heat, except monsoon season, the water balance over most of the
central and southern parts of the country remains negative (chaudhry, et al, 2004). There
are only five Regional Agromet Centers all over the Pakistan, which do not represent all
the areas of the country. In this connection the regression equation obtained through this
analytical study can be used to determine the soil temperature of different depths. This
would help to choose the best suitable region and specie of crops to be cultivated. The
site collection of the soil temperature is very complicated as well as costly. The
installation of thermometer correctly in the soil is another very complicated technical
matter. The use of these regression equations will save the time as well as the finances.
Here one just has to input the air temperature and the result will be in the form of soil
temperature with a sufficient accuracy to rely upon.

27
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

References:
Balisky, A.C, and P.J.Burton. 1993. ‘Distinction of soil thermal regimes under various
experimental vegetation covers. Canadian Journal of Soil Science. 73:411-420
Bollero, G. A., D. G. Bullock, and S. E. Hollinger, 1996: Soil temperature and
planting date effects on corn yield, leaf area, and plant development. Agron. J., 88, 385–
390.
Chaudhry, Q.Z. and G. Rasul; 2004: Agro-climatic Classification of Pakistan. Science
Vision, Vol. 9, No. 3-4. pp 59-66.
Sharratt, B. S.: 1992, ‘Growing Season Trends in the Alaskan Climate Record’, Arctic
45, 124–127.
Year Book, 2005-2006, Ministry of Food, Agriculture and Live Stock, Pakistan
Shannon E. Brown, Kurt S. Pregitzer, David D. Reed, and Andrew J. Burton.: ‘
Predicting Daily Mean Soil Temperature from Daily Mean Air Temperature in Four
Northern Hardwood Forest Stands’, Forest Scinece,46(2)2000.
ZHENG, D. E. R.Hunt Jr. and S.W.Running. 1993.’A daily soil temperature model
based on air temperature and precipitation for continental application. Climate Research.
2: 183-191
Faqir Muhammad, ‘Statistical Methods and Data Analysis’ Kitab Markaz, Faisalabad
http://www.globe.gov/tctg/sectionpdf.jsp?sectionId=92#page=10,
(last visited 2/11/2007)
Tang X.L., Zhou G.Y., Liu S.G., 2006, ' Dependance of soil respiration on soil
temperature and soil moisture in successional Forest in Southern China’, Journal of
Integrative soil Biology 48(6):654-663

28
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

TEMPORAL MODEL OF HYDROMETEOROLOGY IN


THE CITARUM RIVER BASIN INDONESIA
Ruminta* TP PT

Abstract:
Water resources availability for meeting reservoir storage, water supply diversion and
environmental in stream flow requirements must be assessed on various premises
regarding future water used as well as climatic and hydrometeorological conditions.
This study, therefore, investigates temporal dynamical model and detects persistence
and plausible long-term trends of the hydrometeorology in the Upper Citarum River
Basin, West Java, Indonesia. The investigations based on monthly observations data of
the rainfall, evapotranspiration, humidity, and runoff from January 1968 to December
2000 and monthly global phenomena data from National Centers for Environmental
Prediction (NCEP). Identification of the temporal dynamical model of the
hydrometeorology based on adaptive neuro-fuzzy inference system (ANFIS). The
conclusion obtained in this study shows that temporal dynamical model of the
hydrometeorology based on ANFIS can simulate the observations accurately. The
model is capable to minimize the bias and root mean squared error (RMSE). The
rainfall, evapotranspiration, and humidity had two firm persistence i.e. wet and dry
periods, while the runoff has no firm persistence. The rainfall and runoff had been
decreased by about -3.64% and -1.11% respectively. On the other hand, the
evapotranspiration and air humidity had been increased by about 3.88% and 4.21%
respectively. The results of this studies can give an information for improvement of an
integrated water management in the Citarum river basin.
Key words : rainfall, evapotranspiration, humidity, runoff, wet and dry periods,
long-term trends

Introduction:
The design, planning, and operating of river system depend largely on relevant
information derived from extreme event forecasting and estimation. Reliable runoff and
rainfall forecasts are particularly important for warning against dangerous flood or
inundation and drought as well as in the case of multi-purpose reservoirs. Variability of
hydrometeorology in Indonesian had been changed as consequences of global climate
changes and ENSO (Peel, et al., 2001, Hendon, 2002). The rainfall in South East Asian
had been changed significantly (Burn and Hag, 2002, Leon, 2002; Xua et al., 2003;
Cheng et al., 2004).
The Citarum river basin is a complex basin containing most of the feature of water
system: storage, uncontrolled sources, multi purposes, and potential for all forms of
operational management. Information of temporal models, predictions, persistence (wet
and dry periods), and long-term trends of the hydrometeorological components have the

*
TP TP Laboratorium of Climatology, Padajdjaran University, Indonesia.
29
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
potential to drastically improve the effective use of available water resources in this
basin.
The application of ANFIS to various aspects of hydrometeorological modeling has
undergone much investigation in recent years (Franc and Panigrahi, 1997; Mashudi,
2001; Ozelkan and Duckstein. 2001). ANFIS is mostly suited to the modeling of
nonlinear system. Zhu (2000) and Shapiro (2002) have indicated that ANFIS is the best
modeling to analysis numerical data, because in training process based on minimize
value of root mean square error (RMSE). The others investigation show that ANFIS can
predict better than back propagation multilayer preceptron (BPMP) or autoregression.
This computational methods offers real advantages over conventional modeling,
especially when the underlying physical relationship are not fully understood ( Nayak et
al., 2004; Cigizoglu, 2003; Tokar and Markus, 2000 ). This study aims at applying
ANFIS as a modeling tool to predict and estimate the runoff and rainfall data of the
Citarum river basin in West Java. Several ANFIS applications comprise of predictions
of runoff and rainfall 1-month to 12-months and 1-year to 6 years in advance model.
The purposes of this study are investigates temporal dynamical model and detects
persistence and plausible long-term trends of the hydrometeorology in the Upper
Citarum River Basin, West Java. The results of this studies can give an information for
improvement of an integrated water management in the Citarum river basin.

Study Area and Data:


The upper Citarum river basin is located in West Java Indonesia, with a geographical
position of about 6o43’-7o04’ southern latitude and 107o15’-107o55’ eastern latitude. It
P P P P P P P P

covers an area of approximately 4.500 km2. The investigations based on monthly


observations data of the rainfall, evapotranspiration, humidity, and runoff from January
1968 to December 2000 and monthly global temperature (GT) and central Indian
precipitation (CIP) data from National Centers for Environmental Prediction (NCEP).

-6.5

UPPER CITARUM RIVER BASIN


-6.6

-6.7

-6.8

-6.9

-7

-7.1

-7.2

107.1 107.2 107.3 107.4 107.5 107.6 107.7 107.8

Figure 1: Map of the Upper Citarum River Basin

30
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

Statistical Methods:
Identification of the temporal dynamical model of the hydrometeorology based on
adaptive neuro-fuzzy inference system (ANFIS). The cumulative sum chart analysis and
logistic linear regression are used in this study for estimating wet and dry periods and
long-term trends in hydrometeorological components, respectively.
ANFIS is a type biologically inspired computational model. The functioning of ANFIS
is based on learning process (Jang, 1993). A network of this model is made up of a
member interconnected nodes arranged into three basic layers: input, hidden, and
output. The input nodes perform no computation but are used to distribute input to
network. In this network an information passed one way through the network from the
input layers, through the hidden layers and finally to the output layer. In this study,
ANFIS trained with the standard back propagation algorithm which uses a set of input
and output pattern was applied. An input pattern is used by the system to produce an
output, which then is compared with the actual output. If these no difference, then no
learning takes place.
The application of the ANFIS to the time series data consisted of two steps. The first
step was the training of the ANFIS, which comprised the time series data describing the
input and output to the network and obtaining the inter-connection weight. Once the
training stage was complete the ANFIS were applied to the testing data. The best
training of the ANFIS is the lowest root mean square error (RMSE) and the lowest mean
absolute percentage error (MAPE) and is defined as:
N
1 ……….…….(1)
RMSE =
N
∑ (Y
t =1
t
'
− Yt ) 2

1 ⎡ N (Yt ' − Yt ) ⎤ ………….….(2)


MAPE = ⎢∑ ⎥100%
N ⎣ t =1 Yt ⎦

where Y’t is the model’s output; Yt is the observed data; and N is the number of data
B B B B

points.
The degree of the model precision (also was used to evaluate general quality of the
model) is defined as E :
σ c2 ……………..(3)
E =1−
σ o2

σ
Where o is the variation of the observation data and
σ c is the variation of the
difference between the observation data and model results.
The cumulative sum (CUSUM) chart analysis is a standard technique used to show
changes in the underlying mean of a system (Simmonds and Hope, 1997; Kadolu et al.,
1999; Whiting et al., 2003). The CUSUM were calculated from :
t
St = ∑ ( xk − x ) ……………..(4)
k =1

31
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

where St : CUSUM, xk : time series, x : average of the time series data, and n : number
B B B B

of data points.
Logistic regression is a special case of formal generalization of linear regression
concepts commonly summarized under the term generalized linear models. These
statistical models apply for trend modeling of rare events (Frei and Schar, 2000,
Sveinsson et al.,2002). The logistic trend model expresses a transformed form of the
expected value of counts (π) as a linear function of time:
⎛ π ⎞
η (π ) = α + β .t ≈ η (π ) log⎜ ⎟
⎝1− π ⎠ ……………..(5)

exp(α + β .t )
π (t ,α , β ) =
(1 + exp(α + β .t )) ……………..(6)

here t is time; α, β are the regression intercept and coefficient, respectively, to be


estimated from the data, ; π is a events probability; and η is a prescribed monotonic link
function.
The magnitude of the trend, as given by model parameter β is conveniently expressed
as the odds ratio defined as :
θ = exp[β .(t 2 − t1 )] ……………..(7)

The odds ratio represents the relative change in the ratio of the events against nonevents
during the period (t1, t2) and is an exponential function of the period length.
B B B B

Results and Discussion:


Temporal Dynamical Model of the Hydrometeorology:
The first application of the ANFIS was carried out for identification and prediction
the annually temporal dynamical model of the runoff. The simulation during
training and testing stages showed that ANFIS provides the lowest RMSE or MAPE
and the highest values E (precision) to forecasts the unique value in the output. The
prediction and scatter plot for the training period are compared with the observed
data of the runoff showed in Figure 2. Relationship between input and output of the
annually runoff (RO) and rainfall models showed in Figure 3. The statistical
measures exhibit satisfactory agreements between observed and forecasted data, i.e.
RMSE= 18.87 mm and 17.41 mm, for training and testing stages respectively. This
model has values precision (E) =66.6% (Table 1).
In identification and prediction the annually temporal dynamical model of the
rainfall show that for the simulation during training and testing stages have the
lowest RMSE and the highest values E (precision) to forecasts the rainfall. The
prediction and scatter plot for the training period are compared with the observed
data of the rainfall showed in Figure 2. The statistical measures exhibit satisfactory
agreements between observed and forecasted data, i.e. RMSE= 1.16 mm and 1.16
mm, for training and testing stages respectively. The annually temporal model of the
rainfall has values E =69.3%.

32
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
The others application of the ANFIS was carried out for identification and
prediction the monthly temporal dynamical model of the runoff and rainfall. The
simulation during training and testing stages showed that ANFIS can not provide the
RMSE and values E as well as that of the annually temporal dynamical models. The
prediction and scatter plot for the training period are compared with the observed
data of the runoff and rainfall showed in Figure 4. Relationship between input and
output of the monthly runoff and rainfall models showed in Figure 5. The statistical
measures and forecasts of the models showed in Table 1 and 2.
Table 1: Statistics of the temporal dynamical model of the rainfall and runoff.
Training Testing Precision
Model (E) χ2 test Note
RMSE MAPE RMSE
(%)
Monthly Runoff 19.087 0.145 22.041 77.6 3.91 Acceptable
Monthly Rainfall 21.471 2.039 21.713 49.9 18.20 Acceptable
Annually Runoff 18.872 0.151 17.410 66.6 4.80 Acceptable
Annually Rainfall 1.159 0.013 1.156 69.3 6.41 Acceptable
RMSE=(Root Mean Square Error),
MAPE=(Mean Absolute Percentage Error)
Table 2: Statistics of the rainfall and runoff forecasted
Forecasted Data Observed Data
Model
Minimum Maximum Minimum Maximum
Monthly Runoff 381 M m3 576 M m3 392 M m3 587 M m3
Monthly Rainfall 85 mm 181 mm 90 mm 197 mm
Annually Runoff 3641 M m3 7565 M m3 3691 M m3 8170 M m3
Annually Rainfall 2905 mm 4539 mm 2832 mm 4642 mm
M = million

Figure 2: Annually temporal dynamical models of the runoff and rainfall


33
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

Figure 3: Relationship between input and output of the annually


runoff (RO) and rainfall models

Figure 4: Monthly temporal dynamical model of the runoff and rainfall

34
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

Figure 5: Relationship between input and output of the


monthly runoff (RO) and rainfall models.

Wet and Dry Periods in Citarum River Basin:


The cumulative sum chart of the hydrometeorological component in Citarum river
basin clearly show periods during which the time series data is persistence below or
above the long-term mean (Fig.6). Positive slope on these chart indicate a period of
above average values (wet periods) and negative slope indicating a below average
values (dry periods). The rainfall, evapotranspiration, and air relative humidity had
two firm persistence i.e. wet and dry periods, while the runoff has no firm
persistence. The rainfall has wet periods for 1968-1981 and dry periods for 1987-
2000.
The absence of the firm persistence in the hydrometeorological components
indicated that temporal dynamical of the hydrometeorology in Citarum river basin
has relationship with land cover changes, deforestation, and land used changes as
well as impacts of the global climate changes.

35
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

Figure 6: Standardized CUSUM of rainfall (P), evapotranspiration (ET), relative


humidity(RH), and runoff (RO).
Long-term Trends of the Hydrometeorology:
The rainfall and runoff had been decreased by about 3.64% and 1.11% respectively
(Fig.7). The decreasing trend in the rainfall had controlled mainly by the decreasing
of the clouds and rains formation as a consequence of the global climate change.
The runoff also has decreased in Citarum river basin. The main factor determining
the change in river runoff is rainfall. In 1968 to 2000, periods of rainfall decline
have coincided with declining river runoff.

Figure 7: Standardized moving averages (MA) and coefficient of variability (CV) for rainfall
(P), evapotranspiration (ET), relative humidity(RH), and runoff (RO).
The evapotranspiration and air humidity had been increased by about 3.88% and
4.21% respectively. The increasing trend in the evapotranspiration and air humidity
had influenced by the increasing of the water vapour removal from vegetation lands
surface to the atmosphere by both evaporation and transpiration.

36
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
Variability of rainfall, evapotranspiration, relative humidity, and runoff had been
increased significantly by about 31.37 %, 24.38%, 105.48%, and 14.19%
respectively (Fig.7). This result incicated so that the events of flood and drought in
Citarum river basin are more extremes.

Conclusions:
The conclusion obtained in this study shows that temporal dynamical model of the
hydrometeorology based on ANFIS can simulate the observations accurately. The
model is capable to minimize the bias and root mean squared error (RMSE). This model
can use to predict the runoff or rainfall in the future. The rainfall, evapotranspiration,
and humidity had too firm persistence i.e. wet and dry periods, while the runoff has no
firm persistence. The rainfall and runoff had been decreased by about 3.64% and 1.11%
respectively. On the other hand, the evapotranspiration and air humidity had been
increased by about 3.88% and 4.21% respectively. Variability of rainfall,
evapotranspiration, relative humidity, and runoff had been increased significantly so that
the events of flood and drought are more extremes.

References:
Burn, D.H. and E. Hag, 2002: Detection of Hydrologic Trends and Variability.
Journal of Hydrology, 255(1-4), 107-122.
Cheng, K., H. Hsu, M. Tsa, K. Chang, and R. Lee, 2004: Test and Analysis of Trend
Existence in Rainfall Data. Journal of Hydrology, 259, 254-271.
Cigizoghu, H.K., 2003: Estimation, Forecasting, and Extrapolation of River Flows by
Artificial Neural Networks. Hydrology Science Journal, 48(3), 349-361.
Frei, C. and C. Schar., 2000: Detection Probability of Trends in Rare Events: Theory
and Application to Heavy Precipitation in the Alpine Region. Journal of Climate, 14,
1568-1584.
Franc, J.L. and S. Panigrahi, 1997: Artificial Neural Network Models of Wheat Leaf
Wetness. Journal of Agricultural and Forest Meteorology, 88(1-4), 57-65.
Hendon, H.H., 2002: Indonesian Rainfall Variability: Impacts of ENSO and Local Air-
Sea Interaction. Jour-nal of Climate, 16(11), 1775-1790.
Jang, J.S.R., 1993: ANFIS: Adaptive-Network-Based Fuzzy Inference System. IEEE
Trans. on Systems, Man and Cybernetics, 23(3), 665-685.
Kadolu, M., Z. Sen, and E. Batur, 1999: Cumulative Departures Model for Lake-
Water Fluctuations. Journal of Hydrologic Engineering, 4(3), 245-250.
Leon, D.R., 2002: Tropical Rainfall Trends and the Indirect Aerosol Effect. Journal of
Climate, 15(15), 2103-2116.
Nayak, P.C., K.P. Sudheer, D.M. Rangan, and K.S. Ramasastri, 2004: A neuro
Fuzzy Computing Techni-que for Modeling Hydrological Time Series. Journal of
Hydrology, 291, 52-66.

37
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
Ozelkan, E.C. and L. Duckstein, 2001: Fuzzy Conceptual Rainfall-Runoff Models.
Journal of Hydro-logy, 253(1-4), 41-68.
Peel, M.C., T.A. McMahon, and B.L. Finlayson, 2001: Variability of Annual
Precipitation and It’s the Relationship to the ENSO. Journal of Climate, 15(5), 545-551.
Simmonds, I. and P. Hope, 1997: Persistence Characteristics of Australian Rainfall
Anomalies, Journal of Climatology, 17, 597-613.
Sveinsson, O.G.B., J.D. Salas, D.C. Boes, and R.A. Pielke, 2002: Modeling Dynamics
of Long-term Variability of Hydroclimatic Processes. Journal of Hydrometeorology,
4(3), 489-505.
Tokar, A.S. and M. Markus, 2000: Precipitation-Runoff Modeling Using Artificial
Neural Networks and Conceptual Models. Journal of Hydrologic Engineering, 5(2),
156-161.
Whiting, J. P., M. F. Lambert, and A.V. Metcalfe, 2003: Modeling Persistence in
Annual Australian Point Rainfall, Hydrology and Earth System Science, 7(2), 197-211.
Xua, Z.X., K. Takeuchi, and H. Ishidaira, 2003: Monotonic Trend and Step Changes
in Japanese Preci-pitation. Journal of Hydrology, 279((1-4), 144-150.

38
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

THUNDERSTORM FREQUENCY OVER PAKISTAN


(1961–1990)
Zubair Ahmed Siddiqui* , Abdul Rashid* TP PT P P

Abstract:
A thunderstorm is simply a storm that generates lightning and thunder. It frequently
produces gusty winds, heavy rain and hail. A thunderstorm may be produced by a single
cumulonimbus cloud which influences a small area or it may be associated with cluster
of cumulonimbus clouds covering a large area. This study explores mean monthly and
total-annual thunderstorm frequency for thirty years over various meteorological
stations of Pakistan. From the analysis of the data, it is evident that Parachinar and
Kakul are at the top with the maximum numbers of thunderstorm days annually. July
and August are the months with maximum thunderstorm frequency at nearly all stations,
except southwestern parts of the country such as Jiwani whose maximum thunderstorm
frequency occurs during December to February.
Knowledge of thunderstorm activity over Pakistan can be of immense help to the
weather forecasters, aircraft pilots, agriculturists and also to the planners associated
with the management of natural disasters.

Introduction:
A Thunderstorm (TS) forms when warm humid air rises in an unstable environment. It
is a weather hazard from the view points of aviation, agriculture, building constructions,
electrical and strategic installations on the ground. Thunderstorms are hazards on
account of the squalls which sometimes come with them and the low clouds, heavy
rains, poor visibility and electrical discharge as well. Thunder heard (with or without
lightning seen) has been regarded as a thunderstorm phenomenon occurring at an
observing station. According to observational practices, a thunderstorm is reported
whenever a thundershower is occurring at the station, or when a thunder is heard at the
station or vicinity. At any given time, there are estimated 2000 thunderstorms in
progress on the earth. The greatest proportion occurs in the tropics, where warmth,
plentiful moisture and instability are always present. About 45000 thunderstorms take
place everyday and more than 16 million occur annually around the world. The
lightning from these thunderstorms strikes the earth 100 times each second (Lutgens and
Tarbuk, 2004).
A thunderstorm possesses great potential to become a natural hazard. The fury with
which a thunderstorm unleashes its power can cause loss of life and property of various
degrees. The objective of study of thunderstorm frequency, therefore, is to help weather
forecasters to predict occurrence of thunderstorm activity in any area of Pakistan. The
strong wind gust, heavy precipitation, lightning and hails can damage an aircraft
whether parked or in the air. Lightning and hails also damage standing crops and

*
TP TP Pakistan Meteorological Department.
39
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
sometimes cause deaths of farm animals giving economic blow to farmers and
agriculturists. The information contained in this paper regarding the frequency of
occurrence of thunderstorms in Pakistan will help the weather forecasters to predict its
occurrence. The operators of aviation industries can do rescheduling of flights in order
to avoid any possibility of accident. Agriculturists may also adopt safety measures to
prevent damage to crops and animals. This paper will also help the planners involved in
disaster planning, preparedness and prevention.

Data and Methodology:


Forty eight synoptic observing stations were selected for the period 1961-1990 to study
the thunderstorm frequency distribution over Pakistan (Map-1). The criterion for
selection of these stations is that they cover almost whole of Pakistan. The data
comprises of all types of thunderstorms, i.e. convective, orographic, and seasonal. The
study however does not show the actual preferred times of occurrence, whether
occurring in the morning, afternoon or night. Any number of occurrence in 24 hours
have been counted as “one” day. For example, if on a particular day, thunderstorm
occurs a number of times with breaks, it is considered as “one” or if the occurrence is
only once in 24 hours, the thunderstorm day is counted to be “one”. All stations have
been arranged in latitudinal ascending order. Monthly average and the total annual
thunderstorm frequency of each station is calculated and given in Table-14(Annex-A).
Thunderstorm frequencies were then plotted on the maps of Pakistan (Map-2 to Map-
14) and isolines were drawn. The interval of the isolines for each month is different and
adjusted according to the density of occurrence of thunderstorm days. These maps also
show the areas where maximum, minimum or medium numbers of thunderstorms occur.
In this way data were compiled and monthly and total annual thunderstorm frequency
charts for Pakistan were prepared. Above data were also utilized to generate bar-graphs
(Graph-1 to Graph-13) to compare station-wise distribution of each month and total-
annual thunderstorm frequency. The bar-graph (Graph-14) showing total values of
thunderstorm days occurring annually over Pakistan also provides the month-wise trend
of thunderstorm frequency. Month-wise and total annual distribution of thunderstorm
days over different regions of Pakistan have been arranged into three categories of
maximum, medium and minimum depending on the frequency of occurrence monthly or
annually (Table-2 to Table-14).
The criteria for maximum, medium and minimum numbers of thunderstorm days have
been fixed separately for each month by the authors by considering the range of values
of thunderstorm days in each month.

40
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

Map 1

Discussion:
January:
In January, it is winter season in the whole region. This is the time for passing active
western disturbances. Sometimes these western disturbances may produce a front
over the western part of Pakistan. If there is a strong trough, then the air mass may
get moisture supply from the Arabian Sea. The areas which show maximum TS
days are along the path of western disturbances. The plain areas generally have no
triggering mechanism, hence Cb clouds rarely form. If this system is consisted of
two parts i.e. primary and secondary, then more parts of the country may come
under its influence.
Table 1: January
Maximum no. of
> 1 day Sialkot, Jehlum, Islamabad, Kakul, Kotli
TS days
Medium no. of TS Muzafarabad, Garhidupata, Murree, Lahore,
0.3 to 1 day
days Faislabad, Peshawer, Karachi, Jiwani
Minimum no. of South Punjab, Sindh and Balochistan
< 0.3 days
TS days excluding coasts, extreme Northern areas

41
No of Thunderstorm Days

0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
1.10
1.20
1.30
1.40
1.50
Badin
Karachi
Jiwani
Pasni
Hyderabad
Chorr
Lasbela
Nawabshah
Padidan
Pakistan Journal of Meteorology

Panjgur
Rohri
Khuzdar
Jacobabad
Khanpur
Nokundi
Dalbandin
Kallat
Bahawalpur
Sibbi
Barkhan
Bahw'lNagar
Multan

42
Quetta
Zhob

Map 2
Faisalabad

Graph 1
Lahore

STATIONS
D.I.Khan
Sialkot

JANUARY (1961-1990)
Jehlum
Kotli
Islamabad
Cherat
Parachinar
Murree
Peshawar
Kakul
Garhidopatta
Muzaffarabad
Balakot
Dir
Skardu
Astore
Chillas
Drosh
Bunji
Chitral
Gilgit
Gupis
Vol. 5 Issue 9: July 2008
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
February:
It is winter in February and little heating starts at lower latitudes of Pakistan and
path of western disturbance slightly moves to lower latitudes (middle part of the
country) as compared to January, and the recurving to NE as it move out of
Pakistan, hence many hill stations of northern Punjab and their adjacent areas of
AJK are showing maximum TS days, while central parts of Pakistan are showing an
increasing number of TS days as compared to January. Whole of Sindh and plain
areas of Balochistan have very little or zero TS days.
Table 2: February
Maximum no. Kakul, Garhidupatta, Kotli, Islamabad, Jehlum,
> 2 days
of TS days Lahore,
Medium no. of Central & South Punjab, Hill Stns of Balochistan,
0.5 to 2 days
TS days NWFP(plain areas), Jiwani
Minimum no. Extreme North of Pakistan, plain areas of
< 0.5 days
of TS days Balochistan, whole Sindh

Map 3

43
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

FEBRUARY (1961-1990)

3.20

3.00

2.80

2.60

2.40

2.20
No of Thunderstorm Days

2.00

1.80

1.60

1.40

1.20

1.00

0.80

0.60

0.40

0.20

0.00
Barkhan

Zhob

Skardu
Nokundi

Kakul
Sialkot

Balakot
Jehlum

Islamabad
Kallat
Lasbela

Quetta

Cherat

Garhidopatta

Chillas

Gilgit
Badin

Dalbandin
Karachi

Muzaffarabad

Drosh
Pasni
Hyderabad

Padidan

Jacobabad

Multan

Faisalabad

D.I.Khan
Chorr

Panjgur
Rohri
Khuzdar

Khanpur

Sibbi

Lahore

Kotli

Chitral

Gupis
Parachinar
Murree
Peshawar

Dir

Astore

Bunji
Nawabshah
Jiwani

Bahawalpur

Bahw'lNagar

STATIONS

Graph 2
March:
The month of March is the ending of winter season in which tracks of western
disturbance lie more towards southern Pakistan. Clouds are mostly cumuli, which
give showers for small intervals. Thunderstorm may be more frequent and
accompanied by squalls and sometimes hail. The secondary Low pressure areas are
quite frequent and nearly whole of the Pakistan comes under their influence. The
western disturbance normally enters through Quetta and moves towards NE. That’s
why the TS days are more in mountainous NWFP & Punjab and less in the plains of
Balochistan and Sindh. Again extreme north of Pakistan have very few or zero TS
days.
Table 3: March
Maximum no.of Kakul, Islamabad, Murree, Muzaffarabad,
> 4 days
TS days Garhidupatta, Northern Punjab, Parachinar
Medium no.of Central & South Punjab, NWFP (plain areas), Northern
1 to 4 days
TS days Balochistan
Minimum no.of Whole Sindh, Southern Balochistan, Extreme northern
< 1 day
TS days areas except Drosh.

44
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

Map 4

MARCH (1961-1990)

7.50

7.00

6.50

6.00

5.50
No of Thunderstorm Days

5.00

4.50

4.00

3.50

3.00

2.50

2.00

1.50

1.00

0.50

0.00
Barkhan

Zhob

Skardu
Jehlum
Nokundi

Kakul
Sialkot

Islamabad

Balakot
Lasbela

Quetta

Garhidopatta
Badin

Hyderabad

Multan
Chorr

Padidan
Panjgur

Khuzdar
Jacobabad
Khanpur

Dalbandin
Kallat

Faisalabad
Lahore
D.I.Khan

Cherat
Parachinar
Murree

Muzaffarabad

Dir

Astore

Drosh

Gilgit
Karachi

Kotli
Pasni

Rohri

Sibbi

Chillas

Bunji
Chitral

Gupis
Nawabshah

Bahawalpur

Bahw'lNagar

Peshawar
Jiwani

STATIONS

Graph 3
April:
April is the month of harvesting of many crops in Punjab and Sindh and the
beginning of pre-monsoon period. A low pressure area starts developing over the
central part of Pakistan. The path of western disturbance is along the central
Pakistan. Dust raising winds fill the atmosphere when westerly winds blow over
heated sands. Afternoon TS accompanied by dust storms generally occur. The effect
45
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
of western disturbance and its secondary is continued towards the mountainous
NWFP & Punjab. Clouds are of cumuliform in type. The maximum TS activity is in
the NWFP and northern Punjab.
Table 4: April
Maximum no. of Islamabad, Northern Punjab, Parachinar, Kakul, Peshawer, Dir,
> 7 days
TS days Azad Kashmir
Medium no. of TS Northern Balochistan, Lahore, Central & South Punjab, NWFP
1.5 to 7 days
days (plain areas)
Minimum no. of Whole Sindh, Central & South Balochistan, extreme Northern
< 1 day
TS days areas.

Map 5

APRIL (1961-1990)

15.00

14.00

13.00

12.00

11.00
No of Thunderstorm Days

10.00

9.00

8.00

7.00

6.00

5.00

4.00

3.00

2.00

1.00

0.00
Jehlum
Barkhan

Zhob

Islamabad

Skardu
Hyderabad
Badin

Lasbela

Padidan

Jacobabad

Dalbandin

Multan
Quetta

Faisalabad

D.I.Khan

Garhidopatta

Drosh
Muzaffaraba
Nokundi

Sialkot

Kakul

Balakot
Nawabshah
Chorr

Parachinar
Panjgur

Khuzdar

Khanpur

Bahw'lNagar

Lahore

Murree
Peshawar

Dir

Astore
Karachi

Pasni

Rohri

Kallat
Bahawalpur
Sibbi

Kotli

Cherat

Chillas

Gupis
Bunji
Chitral
Gilgit
Jiwani

STATIONS

Graph 4

46
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
May:
The month of May falls in the pre– monsoon period. The heating is at its peak in
many plain areas. Thunderstorm is very frequent in the NWFP and northern Punjab.
Dust storms and thunderstorms are very common in the sub-mountain areas
generally as a result of local instability conditions and are accompanied by squally
weather.
Table 5: May
Maximum no. Parachinar, Dir, Murree, Islamabad, Jehlum, Kakul,
> 9 days
of TS days Muzaffarabad, Kotli, Garhidopatta.
Medium no. Peshawer, Drosh, Chitral, Bunji, Chillas, Gilgit, Balakot,
2 to 9 days
of TS days Lahore, Sialkot, Multan, Faisalabad, Quetta, Zhob,Barkhan
Minimum no. Whole Sindh, south & central Punjab except Multan &
< 2 days
of TS days Bwl’Nagar, south & central Balochistan except Khuzdar

Map 6

47
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

MAY (1961-1990)

17.00

16.00

15.00

14.00

13.00

12.00
No of Thunderstorm Days

11.00

10.00

9.00

8.00

7.00

6.00

5.00

4.00

3.00

2.00

1.00

0.00
Nokundi

Kakul
Sialkot

Balakot
Barkhan

Zhob

Skardu
Karachi

Pasni

Rohri

Sibbi

Kotli
Islamabad

Bunji
Chitral
Kallat

Cherat

Gilgit
Quetta
Jiwani

Chillas

Gupis
Badin

Hyderabad

Padidan

D.I.Khan
Chorr

Panjgur

Khuzdar
Jacobabad

Dalbandin

Multan

Faisalabad

Parachinar
Khanpur

Lahore

Murree

Drosh
Dir

Astore
Lasbela

Garhidopatta
Nawabshah

Muzaffaraba
Bahawalpur

Bahw'lNagar

Jehlum

Peshawar
STATIONS

Graph 5

June:
The temperature reaches to highest annual values in the southern and western
lowlands of Pakistan. There is a trough of low pressure over the Gangetic Plain
with its core of lowest pressure over Balochistan. Thunderstorm activity is much
more in northern Pakistan and whole NWFP. The thunderstorm or dust storm
occurring over the plains, sometimes show a frontal behavior.

Table 6: June
Maximum no. Parachinar, Kakul, Garhidopatta, Muzaffarabad, Kotli, Dir, Islamabad,
> 10 days
of TS days Murree.
Medium no. Peshawar, DI.Khan, Balakot, Drosh, Chillas, Bunji, Chitral, Gilgit, north
2 to 10 days
of TS days & central Punjab, Khuzdar, Barkhan, Zhob
Minimum no. Whole Sindh, whole Balochistan except, Barkhan, Zhob, & Khuzdar,
< 2 days
of TS days Southern Punjab, Quetta, Cherat, Astore, Gupis

48
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

Map 7

JUNE (1961-1990)

15.00

14.00

13.00

12.00

11.00
No of Thunderstorm Days

10.00

9.00

8.00

7.00

6.00

5.00

4.00

3.00

2.00

1.00

0.00
Barkhan

Zhob

Jehlum
Nokundi

Kakul

Skardu
Sialkot

Balakot
Islamabad
Badin

Chorr

Panjgur

Khuzdar

Khanpur

Lahore

Parachinar
Murree

Garhidopatta
Muzaffaraba

Dir

Astore
Multan
Quetta
Karachi

Hyderabad

Lasbela

Padidan

Dalbandin
Pasni

Rohri

Jacobabad

Faisalabad

D.I.Khan

Drosh
Kallat

Sibbi

Kotli

Bunji
Chitral
Cherat

Chillas

Gilgit
Bahawalpur

Bahw'lNagar

Gupis
Peshawar
Jiwani

Nawabshah

STATIONS

Graph 6
July:
In the month of July, onset of monsoon takes place in Pakistan. The first monsoon
rains occurs at the north-eastern and northern part. Therefore Sialkot, Jehlum, Kotli
etc are showing maximum thunderstorm frequency. Minimum thunderstorm
frequency is observed in some parts of southern Punjab, Sindh and Balochistan.
During the first half of July, at coastal areas, little thunderstorm activity is due to the
49
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
effect of mid-tropospheric cyclones, while during the second half, thunderstorm
activity is due to the start of monsoon.
Table 7: July
Maximum no. of Parachinar, Kakul, Garhidupatta, Muzaffarabad, Dir,
> 12 days
TS days Murree, Islamabad, Jehlum, Kotli.
Medium no. of TS Central parts of Punjab, Gilgit, Chillas, Drosh, Bunji,
3 to 12 days
days northern Balochistan.
Minimum no.of TS Skardu, Astore, Gupis, Whole Sindh & whole Balochistan
< 3 days
days except northern parts, southern parts of Punjab.

Map 8
August:
Highest TS frequency of more than 12 days is observed in upper Punjab, Barkhan
and northern areas. Medium TS frequency between 3 to 12 days occur in central
Punjab, Zhob and in NWFP. The moisture availability for these thunderstorms, is
from the out west monsoon that is Bay of Bengal branch. Lowest frequency of less
than 3 days occur in Sindh, Balochistan except Barkhan and Zhob, southern Punjab
and extreme north of northern areas of Pakistan. The high frequency of TS at
Barkhan and Zhob is mainly due to orographic effect. But in this case moisture
supply is from the Arabian Sea branch.

50
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

JULY (1961-1990)

20.00

18.75

17.50

16.25

15.00

13.75
No of Thunderstorm Days

12.50

11.25

10.00

8.75

7.50

6.25

5.00

3.75

2.50

1.25

0.00

Jehlum
Barkhan

Zhob

Skardu
Islamabad
Badin

Hyderabad

Padidan

Jacobabad

Dalbandin

Multan

Faisalabad

D.I.Khan

Drosh
Sialkot

Balakot
Chorr
Lasbela

Panjgur

Khuzdar

Khanpur

Quetta

Lahore

Parachinar
Murree

Astore
Nawabshah

Nokundi

Kakul
Garhidopatta
Muzaffaraba

Dir

Chillas

Gupis
Bahawalpur

Bahw'lNagar

Peshawar
Kallat

Cherat

Gilgit
Karachi

Pasni

Rohri

Sibbi

Kotli

Bunji
Chitral
Jiwani

STATIONS

Graph 7
Table 8: August
Maximum no. of TS Parachinar, Kakul, Garhidupatta, Muzaffarabad, Dir, Murree,
> 12 days
days Islamabad, Jehlum, Kotli.
Medium no. of TS Central parts of Punjab, Gilgit, Chillas, Chitral, Drosh, Bunji,
3 to 12 days
days northern Balochistan, Peshawer, D.I.Khan, Balakot
Minimum no. of TS Skardu, Astore, Gupis, Whole Sindh & whole Balochistan
< 3 days
days except northern parts, southern parts of Punjab.

Map 9

51
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

AUGUST (1961-1990)

20.00

18.75

17.50

16.25

15.00

13.75
No of Thunderstorm Days

12.50

11.25

10.00

8.75

7.50

6.25

5.00

3.75

2.50

1.25

0.00
Barkhan

Zhob

Jehlum

Skardu
Sialkot

Balakot
Nokundi

Islamabad

Kakul
Chorr

Nawabshah

Panjgur

Khuzdar

Khanpur

Parachinar
Murree
Badin

Hyderabad

Lasbela

Padidan

Jacobabad

Garhidopatta
Dalbandin

Multan
Quetta

Lahore
Faisalabad

D.I.Khan

Muzaffaraba

Dir

Astore

Drosh
Kallat

Cherat

Gilgit
Chillas

Gupis
Bahawalpur

Bahw'lNagar

Peshawar
Karachi

Pasni

Rohri

Sibbi

Kotli

Bunji
Chitral
Jiwani

STATIONS

Graph 8
September:
It is the month of ending of monsoon. The seasonal low over Balochistan is now
dissipating; therefore, there is no triggering mechanism to lift the moisture in the
plains and coastal areas of Sindh, Balochistan, central & southern Punjab. Hence
these areas do not show enough thunderstorms. NWFP, northern Punjab and AJK
including Parachinar, Dir, Murree, Kakul and Muzaffarabad are showing maximum
TS frequency due to their orography. Plains of north-east and south-west Punjab,
south NWFP, plains of AJK are showing medium TS frequency. Extreme northern
areas of Pakistan like Gilgit, Gupics, Astore, Chelas, Bunji and Skardu ,etc., are
showing very less TS frequency.
Table 9: September
Maximum no. Parachinar, Kakul, Islamabad, Murree, Garhidupatta,
> 9 days
of TS days Muzaffarabad, Dir.
Medium no. of Central parts of Punjab, Balakot, Chitral, Chillas, Drosh,
2 to 9 days
TS days Bunji, Barkhan, Zhob., Peshawer, Balakot.
Minimum no. Skardu, Astore, Gilgit, Whole Sindh & whole Balochistan
< 2 days
of TS days except Barkhan & Zhob, southern Punjab.

52
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

Map 10

SEPTEMBER (1961-1990)

14.00

13.00

12.00

11.00

10.00
No of Thunderstorm Days

9.00

8.00

7.00

6.00

5.00

4.00

3.00

2.00

1.00

0.00
Jehlum
Barkhan

Zhob

Skardu
Islamabad
Nokundi

Sialkot

Kakul

Balakot
Badin

Hyderabad

Lasbela

Padidan

Jacobabad

Dalbandin

Multan
Quetta

Faisalabad

Muzaffaraba
D.I.Khan

Garhidopatta

Drosh
Chillas
Chorr

Panjgur

Khuzdar

Khanpur

Lahore

Parachinar
Murree

Dir

Gupis
Astore
Nawabshah
Karachi

Pasni

Rohri

Kallat

Sibbi

Kotli

Cherat

Bunji
Chitral
Gilgit
Bahawalpur

Bahw'lNagar

Peshawar
Jiwani

STATIONS

Graph 9
October:
This month falls in the post-monsoon season. Western disturbances are infrequent.
In October, there is no weather system of pronounced nature. Moisture supply is
only from Arabian sea which moves northward causing higher TS frequency at hill

53
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
stations like Barkhan & Parachinar and then at the footsteps of hills in NWFP and
Punjab. Air looses its maximum moisture at this mountain belt and becomes nearly
dry on reaching the extreme north. That is why thunderstorm frequency again drops
to its minimum count in extreme north of Pakistan.
Table 10: October
Maximum no. Parachinar, Kakul, Islamabad, Murree, Garhidupatta,
> 4 days
of TS days Muzaffarabad, Dir, Drosh.
Medium no. Central parts of Punjab, Hill stations of Northern & Central
0.5 to 4 days
of TS days Balochistan, Peshawer, D.I.Khan, Balakot.
Minimum no. Skardu, Astore, Gupis, Chillas, Bunji, Gilgit, Whole sindh,
< 0.5 days
of TS days plain stations of Balochistan, southern parts of Punjab.

Map 11
November:
During this month western disturbances are not very common. If some disturbance
reach Pakistan, it affects only the NWFP, northern areas and western areas of
central/northen Punjab and to some extent the hills of northeast Balochistan.
Thunderstorm frequency is maximum at Gharidupatta and Kakul and it decreases up
to central Punjab and northeast Balochistan. The thunderstorm frequency is almost
zero in the Northern areas and northwest of NWFP. It is very rare in Sindh, almost
at all the areas of Balochistan and eastern & southern parts of the Punjab.

54
No of Thunderstorm Days

0.00
0.50
1.00
1.50
2.00
2.50
3.00
3.50
4.00
4.50
5.00
5.50
6.00
6.50
7.00
7.50
8.00
8.50
9.00
Badin
Karachi
Jiwani
Pasni
Hyderabad
Chorr
Lasbela
Nawabshah
Padidan
Panjgur
Rohri
Pakistan Journal of Meteorology

Khuzdar
Jacobabad
Khanpur
Nokundi
Dalbandin
Kallat
Bahawalpur
Sibbi
Barkhan
Bahw'lNagar
Multan

55
Quetta
Zhob

Map 12
Faisalabad

Graph 10
Lahore

STATIONS
D.I.Khan
Sialkot
OCTOBER (1961-1990)

Jehlum
Kotli
Islamabad
Cherat
Parachinar
Murree
Peshawar
Kakul
Garhidopatta
Muzaffarabad
Balakot
Dir
Skardu
Astore
Chillas
Drosh
Bunji
Chitral
Gilgit
Gupis
Vol. 5 Issue 9: July 2008
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

Table 11: November


Maximum no. Parachinar, Kakul, Muzaffarabad, Garhidupatta, Peshawer,
> 1 day
of TS days Murree, Islamabad, Jehlum, Kotli.

Medium no. Central & Southern Punjab, most of NWFP, northern


0.1 to 1 day
of TS days Balochistan, coastal areas

Minimum no. Gilgit, Skardu, Astore, Chillas, Gupis, Whole Sindh &
< 0.1 days
of TS days Balochistan except coastal & northern parts of the provinces.

NOVEMBER (1961-1990)

2.40

2.20

2.00

1.80
No of Thunderstorm Days

1.60

1.40

1.20

1.00

0.80

0.60

0.40

0.20

0.00
Barkhan

Zhob

Skardu
Jehlum

Islamabad
Nokundi

Sialkot

Kakul

Balakot
Badin

Hyderabad

Multan

Muzaffarabad
Chorr
Lasbela

Padidan
Panjgur

Khuzdar
Jacobabad
Khanpur

Dalbandin

Quetta

Faisalabad

D.I.Khan
Lahore

Parachinar
Murree

Garhidopatta

Drosh
Dir

Astore

Gupis
Nawabshah
Karachi

Pasni

Rohri

Kallat
Bahawalpur

Bahw'lNagar

Chillas
Sibbi

Kotli

Cherat

Peshawar

Chitral
Gilgit
Bunji
Jiwani

STATIONS

Graph 11

56
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
December:
December is the month of beginning of winter season in Pakistan. During this
month western disturbances reach Pakistan and cause thunderstorm and rain in
almost all over Pakistan except southern part of Sindh where they are rare.
Maximum TS is observed in northern Punjab, Garhidupatta and Islamabad. TS
frequency is 0.5 over the hills of west and central Balochistan and Jiwani in
southwest Balochistan.
Thunderstorm frequency is almost zero in Northern areas and northwestern part of
NWFP. It is less than 0.1 in southern Punjab, in almost all the Sindh except Karachi
and all the areas of Balochistan except Jiwani and west central hills of Balochistan.

Table 12: December


Maximum no. Quetta, Lahore, Kakul, Garhidupatta, Islamabad, Murree,
> 0.6 days
of TS days Jehlum, Sialkot, Kotli
Medium no. Whole NWFP, central Punjab, hill stations of Balochistan,
0.1 to 0.6 days
of TS days Coastal areas of Sindh & Balochistan,
Minimum no. Extreme northern areas, southern Punjab, whole Sindh
< 0.1 day
of TS days except coastal areas, plain stations of central Balochistan.

Map 13

57
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

DECEMBER (1961-1990)

1.40

1.30

1.20

1.10

1.00
No of Thunderstorm Days

0.90

0.80

0.70

0.60

0.50

0.40

0.30

0.20

0.10

0.00
Nokundi

Kakul
Sialkot

Balakot
Barkhan

Zhob

Skardu
Karachi

Pasni

Rohri

Islamabad
Kallat

Sibbi

Kotli

Cherat

Bunji
Chitral
Chillas

Gilgit
Chorr

Parachinar

Gupis
Murree
Badin

Hyderabad

Lasbela

Panjgur

Khuzdar

Khanpur

Lahore

Jehlum
Padidan

Jacobabad

Dalbandin

Multan
Quetta

Garhidopatta

Dir

Astore
Muzaffarabad

Drosh
Jiwani

Faisalabad

D.I.Khan
Bahawalpur

Bahw'lNagar

Peshawar
Nawabshah

STATIONS

Graph 12

Results and Conclusions:


Northern Punjab and adjacent areas of Azad Jammu and Kashmir (AJK), Islamabad and
North-West Frontier Province (NWFP), except extreme northern and southern part, are
showing highest number of TS days during the whole year. Parachinar shows maximum
frequency. Central and eastern Punjab and southern NWFP are showing medium
number of TS days. This is because of the absence of triggering mechanism. Extreme
SW coast of Balochistan i.e. Jiwani is showing comparatively a little higher number of
TS days from December to February. It is due to the influence of secondary western
disturbances which pass through the Persian Gulf and Gulf Oman. Barkhan in
Balochistan is an exception. This shows high TS days during summer, because only the
northeastern hill stations of Balochistan are influenced from summer monsoon.
Table 13: Annual
Maximum no. Northern Punjab & adjacent areas of AJK, Islamabad,
55 to 108 days
of TS days NWFP except extreme northern & southern part
Medium no.
25 to 54 days Central & NE Punjab, Southern NWFP, Barkhan
of TS days
Minimum no. Whole Sindh, Balochistan except Barkhan, Southern
1 to 24 days
of TS days Punjab, Extreme Northern Areas

58
No of Thunderstorm Days

0.00
4.00
8.00
12.00
16.00
20.00
24.00
28.00
32.00
36.00
40.00
44.00
48.00
52.00
56.00
60.00
64.00
68.00
72.00
76.00
80.00
84.00
88.00
92.00
96.00
100.00
104.00
108.00
Badin
Karachi
Jiwani
Pasni
Hyderabad
Chorr
Lasbela
Nawabshah
Padidan
Panjgur
Rohri
Khuzdar
Pakistan Journal of Meteorology

Jacobabad
Khanpur
Nokundi
Dalbandin
Kallat
Bahawalpur
Sibbi
Barkhan
Bahw'lNagar
Multan
Quetta
Zhob

59
Faisalabad

Map 14
Lahore

Graph 13
STATIONS
D.I.Khan
Sialkot
Jehlum

Total Annual (1961-1990)


Kotli
Islamabad
Cherat
Parachinar
Murree
Peshawar
Kakul
Garhidopatta
Muzaffaraba
Balakot
Dir
Skardu
Astore
Chillas
Drosh
Bunji
Chitral
Gilgit
Gupis
Vol. 5 Issue 9: July 2008
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

PAKISTAN MONTH-WISE

300
290
280
270
260
250
240
230
220
210
200
Total TS Days at 48 Stations

190
180
170
160
150
140
130
120
110
100
90
80
70
60
50
40
30
20
10
0
1 2 3 4 5 6 7 8 9 10 11 12
MONTHS

Graph 14
Months of July and August are showing the maximum TS frequency in Pakistan because
thia is the season of southwest monsoon. During this season the seasonal low lies over
Balochistan and central Pakistan and in association with the axis of monsoon trough
produce maximum of thunderstorm s in the region.
Acknowledgment:
The authors of this paper are very much thankful to Mr. Muhammad Muslehuddin, (Rtd)
Director and Mr. Syed Anzar Ali Jafri, (Rtd) Deputy Director, of Pakistan
Meteorological Department, for their valuable suggestions.

References:
Lutgens, F.K. and Edward J. Tarbuck, 2004. “The Atmosphere, an introduction to
Meteorology”, Prentice Hall, USA.
Shamshad, K. M., 1988. “Climate and Weather of Pakistan”, ,Royal Book Co.,
Karachi.

60
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
Annex –A
Table 14: Monthly Mean (1961-1990) of “No. of Thunderstorm Days” of 48 Stations of Pakistan
1961-1990 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total
Badin 0.17 0.13 0.13 0.07 0.17 0.93 1.30 1.73 0.57 0.10 0.03 0.00 5.33
Karachi 0.33 0.37 0.33 0.37 0.00 0.47 2.10 1.57 0.47 0.07 0.17 0.17 6.40
Jiwani 0.67 1.00 0.57 0.47 0.13 0.07 0.50 0.33 0.07 0.13 0.27 0.60 4.80
Pasni 0.23 0.33 0.13 0.10 0.13 0.00 0.27 0.13 0.10 0.07 0.13 0.20 1.83
Hyderabad 0.07 0.17 0.33 0.60 0.87 0.83 1.27 1.47 0.43 0.07 0.07 0.03 6.20
Chorr 0.13 0.20 0.40 0.50 0.87 1.67 3.17 3.13 1.37 0.27 0.10 0.03 11.83
Lasbela 0.00 0.07 0.18 0.18 0.57 0.32 0.43 0.43 0.04 0.18 0.04 0.00 2.43
Nawabshah 0.07 0.23 0.44 0.38 0.59 0.77 2.01 2.03 1.27 0.10 0.27 0.03 8.18
Padidan 0.03 0.30 0.50 0.20 0.43 0.63 2.33 2.00 1.00 0.23 0.10 0.10 7.87
Panjgur 0.13 0.33 0.57 0.47 0.70 0.63 1.17 0.70 0.23 0.13 0.00 0.07 5.13
Rohri 0.10 0.10 0.43 0.23 0.43 0.33 0.87 0.60 0.20 0.10 0.03 0.03 3.47
Khuzdar 0.13 0.75 1.50 2.08 2.08 3.38 4.46 5.54 1.75 0.88 0.33 0.13 23.00
Jacobabad 0.10 0.60 1.30 1.07 1.10 0.83 2.77 1.80 0.73 0.20 0.10 0.07 10.67
Khanpur 0.23 0.57 1.33 1.13 1.63 1.23 2.70 2.17 1.10 0.27 0.23 0.07 12.67
Nokundi 0.17 0.43 0.43 1.07 0.37 0.03 0.23 0.03 0.10 0.27 0.03 0.03 3.20
Dalbandin 0.30 0.70 1.20 1.70 1.47 0.33 1.53 0.47 0.17 0.53 0.10 0.47 8.97
Kallat 0.17 0.10 0.50 0.27 0.50 0.33 0.97 0.87 0.13 0.07 0.03 0.07 4.00
Bahawalpur 0.20 0.50 1.23 1.47 1.90 1.33 2.43 1.60 0.73 0.23 0.30 0.03 11.97
Sibbi 0.00 0.20 0.67 0.67 0.20 0.57 2.00 1.30 0.37 0.07 0.03 0.03 6.10
Barkhan 0.17 0.67 2.96 3.67 4.38 5.79 10.04 9.67 4.54 1.71 0.58 0.08 44.25
Bahw'lNagar 0.29 1.14 1.75 2.57 3.36 2.89 4.86 3.68 1.68 0.57 0.18 0.29 23.25
Multan 0.53 1.27 3.40 4.63 5.10 3.70 6.73 5.70 2.23 0.73 0.57 0.40 35.00
Quetta 0.87 1.07 3.40 4.50 2.37 0.93 3.77 2.37 0.33 0.53 0.27 0.63 21.03
Zhob 0.07 0.60 2.20 3.80 3.30 3.43 9.97 6.40 2.07 0.53 0.13 0.07 32.57
Faisalabad 0.40 1.10 2.77 3.63 3.60 3.60 6.87 6.13 2.63 0.73 0.23 0.33 32.03
Lahore 0.97 2.27 4.20 4.33 5.67 4.50 8.83 7.97 4.07 2.07 0.77 0.63 46.27
D.I.Khan 0.30 0.67 3.07 3.83 3.73 3.17 5.73 4.73 1.80 0.73 0.40 0.20 28.37
Sialkot 1.42 1.96 4.13 4.88 5.83 7.46 11.83 10.29 5.13 1.96 0.83 1.13 56.83
Jehlum 1.20 2.33 5.30 7.07 9.13 9.63 15.33 14.97 8.13 3.27 1.23 1.23 78.83
Kotli 1.17 2.07 4.83 6.90 9.83 10.93 13.80 13.17 8.43 3.33 1.47 0.80 76.73
Islamabad 1.10 2.70 6.20 7.80 9.03 10.07 16.50 18.33 10.53 4.40 1.53 1.27 89.47
Cherat 0.10 0.33 0.83 1.60 2.03 1.13 2.20 2.47 1.73 0.93 0.33 0.13 13.83
Parachinar 0.13 0.71 4.54 13.79 15.25 13.75 17.13 17.67 13.33 8.21 2.25 0.42 107.17
Murree 0.97 1.90 5.60 7.43 9.43 10.90 14.67 13.67 9.00 4.33 1.70 0.67 80.27
Peshawar 0.37 1.27 4.80 8.67 7.83 5.70 9.33 10.40 6.37 3.33 1.10 0.23 59.40
Kakul 1.23 2.87 7.10 10.87 13.00 13.17 18.70 18.90 11.73 5.87 2.20 0.93 106.57
Garhidopatta 0.80 2.17 5.93 9.40 12.77 13.77 17.53 16.17 10.67 5.43 2.00 1.10 97.73
Muzaffarabad0.37 1.57 4.70 8.83 11.40 12.43 16.47 14.27 9.90 4.97 1.13 0.43 86.47
Balakot 0.03 0.90 3.03 6.13 7.57 9.40 11.17 9.17 7.33 3.10 0.57 0.53 58.93
Dir 0.13 0.58 2.75 8.42 12.88 11.50 12.92 12.92 11.96 6.83 0.96 0.13 81.96
Skardu 0.00 0.00 0.00 0.13 0.75 1.21 2.29 2.67 0.71 0.04 0.08 0.00 7.88
Astore 0.00 0.00 0.03 0.28 1.55 1.93 2.21 1.72 0.31 0.03 0.00 0.00 8.07
Chillas 0.00 0.00 0.00 0.73 2.67 3.63 4.33 4.67 2.17 0.43 0.00 0.00 18.63
61
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
1961-1990 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total
Drosh 0.00 0.27 1.33 5.53 8.70 7.53 9.30 10.40 8.47 4.47 0.43 0.00 56.43
Bunji 0.00 0.00 0.07 0.87 3.53 2.70 4.77 4.43 2.40 0.40 0.00 0.00 19.17
Chitral 0.04 0.00 0.30 2.48 4.89 2.67 2.89 3.63 3.33 1.63 0.00 0.00 21.85
Gilgit 0.00 0.03 0.17 0.33 2.27 2.57 3.87 4.10 1.33 0.20 0.00 0.00 14.87
Gupis 0.00 0.00 0.00 0.03 0.33 0.47 1.27 0.77 0.40 0.03 0.03 0.00 3.33

62
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

GLOBAL WARMING AND MELTING GLACIERS


ALONG SOUTHERN SLOPES OF HKH RANGES
Ghulam Rasul*, Qin Dahe†, Q. Z. Chaudhry*
TP PT TP PT P P

Abstract:
Glaciers are the most sensitive precursors of climate change due to natural and
anthropogenic reasons. Using meteorological and hydrological data in combination
with remote sensing data from various sources, the isothermic dynamics of heat in
upslope direction was evaluated on pentad basis. It revealed that 30ºC isotherm has
creeped upward by 725m higher elevation than 28 years before. Frequency and
duration of heat waves both have increased by two fold. The rate of increase since 1990
has also been doubled resulting into frequent occurrence of severe thunderstorms and
lightening events. As an example of anthropogenic change in climate, ever fastest rate of
glacial depletion is presented on Siachen glacier which has lost about 2km of its length
and 17% of ice mass since 1989. Surface velocity of the glacier has also increased
considerably due to the interacting warmer atmosphere with frozen water reserves in
the presence of large human concentration. Projected future temperatures would further
exaggerate the ice depletion and drift related dynamic processes making the situation
more and more complex for the planners and dependants.
Key words: Global Warming, Melting Glaciers, Siachen Glacier, Heat Waves,
Anthropogenic Climate Change

Introduction:
Himalaya-Karakorum-Hindukush together makes the largest mountain chain over the
earth and they are custodian of the third largest ice reserves after the Polar Regions.
Located side by side north-south makes difficult to distinguish where one ends and other
starts. They are elongated almost east-west drawing a border between China and south
Asian nations including Pakistan, India, Nepal and Bhutan. Existence of these ranges is
a blessing for South Asia. They protect the inhabitants from the cold surges in winter
associated with northerly winds. They confine the monsoon precipitation to this region
which is the great resource of water. In addition to that they possessed a treasure of solid
water which melts with high temperature in summer and makes this precious resource
available in rivers during needy times. Several famous rivers such as the Indus, the
Ganges, and the Yangtze are fed by the runoff from the glaciers of these ranges which
serve as the lifeline for more than a billion people in Asia.
Almost all the glaciers over the globe are subjected to depletion with a few exceptions to
Andes and Alps in Europe (IPCC 4AR). Among them polar ice sheets and Greenland
glaciers are on the top followed by the Himalayan glaciated reserves. Shroder et al.,
2007 found that the loss of significant glaciers between Afghanistan and Pakistan may

*
TP TP Pakistan Meteorological Department.

TP TP Chinese Academy of Sciences, Beijing, China.
63
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
be becoming more serious progressively unless global warming ultimately generates
greater marine evaporation that augments precipitation. Concerns over worldwide loss
of ice have prompted several research groups to address the global warming issues
related to the depletion of glaciers with a special focus on interaction of atmosphere and
cryosphere. Both the systems have a complex exchange mechanism and its
understanding is utmost important to answer the issue related questions.
Global glacier depletion is the roar of the last two decades which would consequently
result into sea level rise and water scarcity over the plains and changes into mountain
landscape. The question is not only if the ice is melting, but also how fast it is melting;
how much melt water be produced and for how long. To answer these questions, the
denser network of direct field measurements along the glaciers and snowlines is
required. For this purpose remote sensing science may play an important role along with
installation of Automatic Weather Stations (AWSs). Presently this set up is insufficient
to draw authentic inference.
Lack of data collection points and their reasonable distribution over this complex terrain
is a great hurdle for predicting future trends accurately. Most of the studies made use of
satellite remote sensing data except a few in situ datasets generated by some scientific
groups during small temporal and spatial scale experiments. Chinese side of Himalaya
has, however, denser network of 40 automatic and satellite-linked weather stations
across Tibet. China has recently established the world’s highest climate station at an
altitude of 5200m above mean sea level on the Tibetan side of Mount Everest
(Qomolangma in Chinese; Sagarmata in Nepali). But due to Pakistan and Indian
tensions on the south of the Himalayan chain, researching efforts are capped in the name
of classified strategic assets.
The HKH frozen water reserves are the water tank over the roof whose operation is
regulated with temperature. Unfortunately its thermostat has been tempered with global
warming and it has started changing much of solid mass into liquid flowing
downstream. Accelerated depletion of solid mass has been exposing threat to future
generations by a paucity of this precious commodity essential for all the living creatures.
Several studies on hydrological, meteorological and glaciological aspects of the third
largest ice reserves of the globe are underway from around this pyramid by
international, governmental, non-governmental research and welfare organizations.
Indian peninsula is platform of four Asian nations namely India, Pakistan and Nepal.
Pakistan makes the western border of the peninsula bound by the Arabian Sea on its
south and covered on the north by Himalaya, Karakoram and Hindukush. These
mountains are the water tanks over the roof which provides supply of water when
required. The environment has given the operational control of this tank in the hands of
temperature after the strong buildup of greenhouse gases. Temperature is surpassing its
normal limits not only due to natural processes but mainly because of non-environment
friendly anthropogenic activities. Growing population caused accelerated deforestation
which believed to be the main cause of environmental heating among others. Snow and
ice extent is shrinking must faster than ever now, as a whole, on the HKH mountainous
terrain (Rasul et al., 2006). Many of the atmospheric and environmental processes
directly or indirectly interact with, control, or are controlled by landforms of the high
mountains. These linkages are exemplified by the great snow and ice reserves of HKH
64
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
region. They are the sources of the downstream snow- and ice-melt water which is the
lifeline for millions of people who depend upon such water for food production.
Anthropogenic and natural aspects of global change, including climatic warming due to
increased proportion of greenhouse gases and global dimming from aerosol emissions,
are opposite climate forcings that complicate interpretations. The changes occurring in
the portion of atmosphere immediately above the earth surface have direct effect on the
frozen water reserves over mountain tops. Study of changes taking place can be
perfectly understood until the interaction of atmosphere and cryosphere is properly
addressed. Simply, climate warming alone could lead to melting away of small glaciers,
with drastic decrease in vital melt water downstream. But increased heating over oceans
could lead similarly to greater evaporation and increased monsoonal precipitation.
Similarly certain aerosol absorb solar radiation and results in warming, while others
reflect them back cooling earth’s surface. Due to increased evaporation enhanced cloud
cover can increase precipitation, but aerosol-induced clouds have smaller droplets that
reduce precipitation. Thus, very serious questions about direction of future change exist
to be answered by robust research and innovative observations.
The frequency and intensity of extreme precipitation events has increased along the
foothills of Himalayan southern slopes where landslides and lightening have taken
several lives and incurred huge damage to the infrastructure (Chaudhry et al., 2007).
This paper presents the behaviour of thermal regime, the way it is stepping upward from
the foothills of southern slopes of HKH ranges impacting the melting rate of perennial
snow and ablating the centuries old glaciated ice. The heat has advanced to southern
slopes in a fluctuating way composed of heat waves with some examples of severe
snowmelt floods and occurrence of extreme precipitation events. Impact of natural and
anthropogenic climate change on some important glaciers has also been discussed with
the help of remote sensing tools.

Materials and Methods:


Due to the lack of regular data collecting points in the most difficult mountainous terrain
of HKH, various sources of data are employed to draw representative inferences linking
the past to the present state of cryosphere and its future potential of production of fresh
water to feed the rivers. However, the maximum reliance remained on meteorological
data collected at PMD stations located at the elevation ranging from 1250m to 2700m
asl. Samiraglia et al., 2007 made use of maps and photographs, and satellite images to
quantify the recent and ongoing fluctuatuations of the ice mass in the HKH region.
Bishop et al., 2007 emphasized the integration of traditional data sets with Remote
Sensing and GIS for glacier mapping, parameter estimation, and numerical modeling to
produce reliable results. The detail about various sets of data used in this study is placed
below:
Meteorological Data:
National Meteorological Service of Pakistan has been maintaining its 13
meteorological stations along the southern slopes of HKH ranges at various heights
as shown in Table 1.

65
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

Table 1: Geographical coordinates of meteorological stations maintained by Pakistan


Meteorological Department along southern slopes of HKH ranges.
S. No. Station Latitude(°N) Longitude(°E) Altitude(m asl)
1 Astore 35.34 74.90 2167
2 Bunji 35.67 74.63 1372
3 Skardu 35.30 75.68 2317
4 Gilgit 35.92 74.33 1459
5 Gupis 36.17 73.40 2155
6 Chitral 35.85 71.83 1500
7 Chilas 35.67 74.10 1250
8 Dir 3520 71.85 1369
9 Drosh 35.57 71.78 1464
10 Kakul 34.18 73.25 1308
11 Parachinar 33.87 70.08 1725
12 Murree 33.85 73.41 2167
13 Rawalakot 35.85 73.80 1676

The daily maximum temperature records from 1981 to 2005 have been incorporated
in isothermal analysis on pentad basis. The daytime maxima were also used to carry
out frequency of severe stress days and their continuous stretch on temporal scale
for heat waves.
Hydrological Data:
Water and Power Development Authority (WAPDA) of Pakistan have installed 10
hydrological stations in HKH southern slopes at heights even more than PMD.
Basic geographical information is presented in Table 2.
Table 2: Hydomet Stations maintained by WAPDA in Northern parts of Pakistan.
S. No. DCP Stations Basin Latitude Longitude Elevation(m.a.s.l.)
1 Naltar Hunza 36° 07‫ ׳‬40‫״‬ 73° 11‫ ׳‬05‫״‬ 2810
2 Yasin Gilgit 34° 22‫ ׳‬00‫״‬ 73° 18‫ ׳‬00‫״‬ 3353
3 Hushey Shyok 35° 22‫ ׳‬00‫״‬ 76° 21‫ ׳‬30‫״‬ 3010
4 Burzil Astore 34° 54‫ ׳‬20‫״‬ 75° 05‫ ׳‬30‫״‬ 4030
5 Shogran Kunhar 34° 38‫ ׳‬00‫״‬ 73° 29‫ ׳‬15‫״‬ 2895
6 Saif ul Muluk Kunhar 34° 54‫ ׳‬15‫״‬ 73° 38‫ ׳‬40‫״‬ 3240
7 Kelash Chitral 35° 39‫ ׳‬00‫״‬ 71° 37‫ ׳‬00‫״‬ 2810
8 Shangla Swat 34° 52‫ ׳‬30‫״‬ 72° 35‫ ׳‬45‫״‬ 2134
9 Besham Indus 34° 55‫ ׳‬27‫״‬ 72° 52‫ ׳‬55‫״‬ 480
10 Rehman Bridge Poonch 33° 29‫ ׳‬20‫״‬ 73° 53‫ ׳‬05‫״‬ 530

Data from above stations were used to assess stream flow and river discharge at rim
stations.

66
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
Reanalysis Data:
National Center for Environmental Prediction (NCEP) provided long records of
reanalysis data although at 2.5° x 2.5° interval but proved to be a good input for
filling up the gaps. The NCEP reanalysis data was able to capture much of the
synoptic scale variabilities in the pressure and temperature fields and the differences
between the observational and the reanalysis data decreased rapidly as the time scale
went from synoptic to monthly (Xie Aihong et al, 2007). Some JMA (Japan
Meteorological Agency) reanalysis data set of resolution 1º x 1º spanned over a
period 1988-1998 was also incorporated over this complex terrain. It was combined
with meteorological observed data and interpolation was carried out to generate
uniform grided data for further analysis.
Satellite Data:
LANDSAT Satellite imageries of Northern Hemisphere fall (cloud free) are used at
irregular intervals to draw inference on changes in Siachen glacier which draws the
highest frontier between Pakistan and India.
ASTER images are used for Batura glacier which was studied in detail by Chinese
scientists in 1974-75 and its past characteristic data are also derived from Chinese
publications.
In Situ Data:
2
Italian scientists from Ev-K -CNR committee have extensively studies Baltoro
glacier from different research aspects. Data from two Automatic Weather Stations
(AWSs) at Askole and Urdukas located in Baltoro valley were also used to assess
the interaction of atmosphere and cryosphere. Their secondary data from the
ablation experiment has been utilized to identify the characteristics of atmosphere
and cryosphere interaction
Stereo photographic Data:
Photographs taken during different expeditions by various mountaineers on glaciers,
glacial lakes and their allied features have also been incorporated to overcome the
lack of scientific data with the courtesy of photographer.

Results and Discussion:


Global warming is an inevitable fact and effects are very clear in most parts of the
world. According to IPCC 4th Assessment Report, all the 12 warmest years ever
P P

recorded over the globe occurred in 1990s and the recent ongoing decade (1990, 2005,
2003, 2002, 2004, 2006, 2001, 1997, 1995, 1999, 1990 and 2000). The strongest ENSO
event of the meteorological history was also recorded during the same period (1997-98)
which triggered 5 year long severest drought in Pakistan history. Global data show a
sharp increase in temperature from 1980s.
Global trend is also reflected by the changing thermal regime in many parts of Pakistan
and the most pronounced in northern high elevation stations located on the southern
slopes of HKH. It was an interesting investigation on how the heat is creeping upward
from lower elevations to higher elevation. It not only clarified the invasion of warmer

67
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
thermal regime to the glaciated reserves responsible for sustainable supply of fresh
water but also disclosed the characteristic features of atmospheric and cryospheric
interaction. The results include higher elevation heat dynamics, heat waves,
hydrodynamics and cryodynamics. They are discussed one by one as under:
Isothermic Dynamics:
Daily maximum temperatures from 1981-2008 for generally dry summer period
from April to June were analyzed on pentad basis except last one which is two years
less yet. Movement of isotherms with time along the elevation revealed the fact that
heat is rushing towards the peaks of this elevated complex and highly irregular
terrain. To know how fast it is moving upward, the dynamics of 30°C was
considered the reference indicator. Figure 1 shows that in 1980s the changes in
isothermic pattern were very slow and hardly a 35m upward shift of reference
isotherm was seen from 1981 to 1990. Warming trend increased in 1990s. By the
end of this decade, the 30°C isoline existed at about 300m higher than its position in
1981-85. The ever hottest year recorded over the globe was1998 which co-occurred
with the strongest El~Nino event 1997-98 of the recorded history. The ENSO event
associated with severe hygrothermal stress conditions contributed a lot to carry the
heat to new heights. Overall the decade of 1990s was believed to be the warmest
one.
Isotherms Based on Mean Summer Max Temperature (April-June)(1991-95) Isotherms Based on Mean Summer Max Temperature (April-June)(1995-2000)

Isotherms Based on Mean Summer Max Temperature (April-June)(2001-05)

Figure 1: Heat shifting northward is represented by isothermic dynamics along southern


slopes of HKH ranges

68
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
st
The first eight years of 21 century are seen surpassing all trends in the past among
P P

them 2005 believed to be the hottest for Pakistan when a historic snowmelt flood in
June created a havoc downstream Indus. Last pentad has completed 3 years so far
(i.e. 2006, 2007, 2008 April-June period) and is showing moderate advance of
warmth in the upward direction. The dynamics of 30°C isotherm along southern
slopes of HKH ranges is shown in Fig 1. Temporal isotherm’s spread show that flux
of upward creeping heat is more over the eastern part (Himalaya and Karakorum) of
the southern slopes than western part comprising Hindukush range. Isothermic
advance is not uniform rather skewed due to the complexities of terrain and
environmental degradation. On the average, the 30°C isotherm has now moved at
580m above its location in early 1980s.
Heat Waves:
Heat waves are a continuous stretch of persisting maximum temperatures above
certain threshold for a specified time period. Rising temperatures are embedded with
thermal extremes which were rare occurrence in the past but now becoming more
common year by year. They are grouped into three categories as defined below:
Severe Heat Wave=Five consecutive Days with Daily Maximum Temperature ≥ 40°C
Moderate Heat Wave= Five consecutive Days with Daily Maximum Temperature ≥ 35°C and <40°C
Mild Heat Wave= Five consecutive Days with Daily Maximum Temperature ≥ 30°C and <35°C

Figure 2: Frequency of moderate, mild and severe heat waves of 5-days duration
(bars) and their linear trend. The numbers on the top of the bars indicate the longest

69
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
Frequency analysis carried out over all the above thresholds is presented in figure 2
from 1981 to 2008 using combined data set from all sources. The height of the bar
represents the number of events when temperatures reached the mild, moderate and
severe stress level in a particular pentad. Frequency of mild stress was quite
common even in 1980s during May and June but moderate and severe stress
conditions rarely occurred. Not only a significant increase had been noticed in
occurrence of mild stress days but a sharp rise in moderate and severe stress events
was also registered during the recent decade. The longest heat wave occurred during
respective pentad is shown in Figure 2d. It can be observed that the persistence of
heat waves has become longer over the temporal scale on one hand while their
intensity has increased on the other hand during the recent years. Similarly their
areal extent has also increased significantly. The valley areas or shadow zones
appearing unaffected by increasing heat in the earlier period are dominated by the
heat sweep. However, they are following certain lag period than the exposed
elevations.
Snowmelt Flood:
Snow and ice are the most sensitive entities of the natural ecosystem to any change
in thermodynamic regime directly or indirectly taking place. Due to a general trend
of increasing warming, these frozen water reserves have also started melting at an
accelerated rate giving rise to the formation of new glacial lakes and producing local
flooding due to their outburst phenomenon. In addition to Glacial Lake Outburst
Floods (GLOFs), ice and snowmelt water runs through the streams and converge
into the major rivers. The convergence of stream flow peaks, sometimes, generates
high river floods downstream.

FlOOD IN INDUS RIVER (AT TARBELA) DURING 15TH JUNE TO 30TH JUNE
FlOOD IN KABUL RIVER DURING 15TH JUNE TO 30TH JUNE
400000
WATER DISCHARGE (in Cusecs)

180000 350000 2005


2005
160000
WATER DISCHARGE (in Cusecs)

300000
140000
250000 2003
120000

100000 200000 2002


2003
80000 150000 2001
60000 2002
2004 100000 2004
40000
2001 50000
20000
Kabul-2001 Indus-2001
0 0
Kabul-2002 15- 16- 17- 18- 19- 20- 21- 22- 23- 24- 25- 26- 27- 28- 29- 30- Indus-2002
06

06

06

06

06

06

06

06
-

06 06 06 06 06 06 06 06 06 06 06 06 06 06 06 06 Indus-2003
15

17

19

21

23

25

27

29

Kabul-2003
Kabul-2004 Indus-2004
DATES DATES
Kabul-2005 Indus-2005

Figure 3: Comparison of 5-year river discharge (in cusecs) for river Kabul and Indus
during the last half of June.

The most devastating flood of pre-monsoon season in the rivers Indus and Kabul
during 2005 was the worst example in recent era. During the last fortnight of June,
the high pressure ridge of westerly waves prevailed over the northern mountainous
regions of Pakistan and Afghanistan. Temperatures plunged quickly above 40°C at
most of the meteorological stations up to 2500m as and such conditions persisted
until the end of June. Not only this stretch of hot spell was recorded as the longest

70
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
heat wave ever occurred in these areas but also established new records of the
maximum temperatures reaching 43.6 °C at Gilgit.
The large amounts of melt water rushed through the streams and flooded the main
rivers. The river discharge of the Indus at Tarbela and the Kabul at Naushera are
compared for the last 5-years during the last fortnight of June (Figure 3). The curves
show an increasing trend of river discharge reflecting warming trend but an abrupt
increase in flow of both the rivers during 2005 represent the extreme heat wave
conditions discussed above. The river Indus and the Kabul join together off
Naushera and known as the Indus. This historic and untimely flood took hundreds of
lives, damaged thounsands of hectares of standing crops and incurred huge loss to
infrastructure downstream from NWFP to Sindh. The main flooding season
generally coincides with the monsoon season which extends from July to
September.
Melting Glaciers:
Increasing concentration of GHGs due to uncontrolled emissions has made the
situation very complex for projection of future climates. It does allow the change to
be linear rather changes in pattern on every bit of increase makes the things
complicated. The whole climate system is although nicely integrated but response to
any change differs from entity to entity. Emissions taking places in one part of the
world are not limited to that region rather immediately distributed over the globe as
earth’s climate system is dynamic machine like a centrifuge. However, the
differential distribution pattern of climate system does not allow uniform
distribution of the effects of added constituents in the solid, liquid or gaseous forms.
The global retreat of glaciers during 20th century is striking as shown in fig
5. The interaction between cryosphere and atmosphere approach is quite appropriate
to study the dynamic behaviour of glacial fluctuations (Kaser, 2001; Wagnon et al.,
2001). Trends in long time series of cumulative glacier-length and volume changes
indicate convincing evidence for fast climate change or sudden variability at global
scale. Since 1990, the IPCC has documented such changes as an evidence (Fig 5) of
global warming. Mountain Glaciers are the most sensitive precursors of global
warming and climate change, although which parameter is playing important role
and quantitative relationship between climate change and glacier fluctuations is still
ambiguous. However it corresponds to a warming of ~0.3ºC in the first half of the
20th century in the northern hemisphere. On the global scale, air temperature is
considered to be the most important factor reflecting glacier retreat, but this has not
been demonstrated for tropical glaciers (IPCC, 2001). However work carried out in
Kalimanjaro concludes that increased air temperature governs the glacier retreat in a
direct manner (Kaser, 2004).
The effect of global warming on the cryosphere in mountain areas are most visibly
manifested in the shrinkage of mountain glaciers and in reduced snow cover
duration (Barry, 2002). There is some disagreement among the scientists that all
glaciers of Himalaya-Karakorum-Hindukush region are retreating. Most of the
world’s mountain glaciers have been shrining for at least the last 30 years (World
Glacier Monitoring Service 2002), including the neighbouring Greater Himalaya
(Hasnain 1999; Mastny 2000; Shrestha and Sherestha 2004). However, Hewitt
71
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
(1998) reports the widespread expansion of larger glaciers in the centeral
Karakoram, accompanied by an exceptional number of glacier surges. This contrast
raises many questions about the standard of instruments, methods of data collection
and accuracy of collected data used in these studies.

Figure 4: Response of world Glaciers to climate change

It is, however, clear that visible changes in mass and extent of glaciers can not be
hidden from the eye of the satellite when present is related to the past. A similar
evidence of decay is presented in Fig. 6 about Siachen glacier which had been a
warfront between Pakistan and India since decades. The Siachen Glacier is located
in the eastern Karakoram Range in the Himalaya Mountains in Ladakh at about
35.5ºN and 77.0ºE. Englishman W. Moorcroft was the first outsider to reach the
region in 1820 while another traveler Henry Starchy was the first to explore the
glacier in 1848. Major F. N. Workman and his wife stayed in the area for a long
time in 1910s and named the glacier as Siachen. It ranges from an altitude of 18,875
ft. a.s.l. at its source Indira Col Pass near China Border to its snout at 11,680 ft. a.s.l.
Melting water of Siachen Glacier is the main source of Nubra River flow and it
drains into Shyok River which in turn joins the Indus River.
The visible decay of Siachen Glacier presented in satellite imageries (Fig. 5) is the
proof of anthropogenic contribution to the warming of glaciated climate. Due to the
presence of army on both sides, there is a large vehicular movement and allied
activities such as building infrastructure. Temperature data collected at Hushey over
the period 1991-2004 shows 4ºC rise in maximum temperature. Since temperature
maxima have been increasing at a greater rate in the presence of large human

72
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
concentration, the thinning of ice and retreat of glacial extent has taken place
simultaneously at an alarming rate. The decay estimates calculated by remote
sensing techniques show that Siachen Glacier has reduced by 1.9km in longitudinal
extent from 1989 to 2006. Thinning of ice mass is evaluated 17% during the same
period. It is also speculated that human presence at Siachen Glacier has also been
affecting the neighbouring glaciers such as Gangotri, Miyar, Milan and Janapa
which feed Ganges (first two glaciers), Chenab and Sutlej Rivers respectively.

Figure 5: Damage to Siachen Glacier observed from the eye of Landsat satellite at
different times.
Fowler and Archer (2006) have tried to prove that the glaciers of Upper Indus Basin
UIB (HKH region) are thickening and expanding based on meteorological data from
some PMD stations and hydrological data of WAPDA stations. They claimed that
diurnal temperature range (DTR) has increased and runoff reduced. They must
check the health of data used and also based on 3 or 4 stations data making such a
big claim for such a large basin does not look very scientific. Also cloud cover over
most of the stations (included in their study) has increased when compared
meteorological normals (1931-60) with (1961-90). It proved that logically DTR
could not increase as mostly clearer skies would be the basic requirement.
Regarding discharge data, runoff from UIB finally reaches Tarbela Dam and water
reserves showed significant increase during the last two decades. Moreover, the

73
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
interacting atmosphere with cryosphere has shown warming trend which is
gradually accelerating and lot of reduction in ice mass has been incurred.
Global research results compiled by IPCC in its four Assessment Reports are
presented below:
• FAR(1990): The observed warming is likely to have been due to natural
variability;
• SAR(1995): There are obvious evidences to detect the anthropogenic
influence on climate;
• TAR(2001): New and strong evidences indicate that the observed warming
in the past 50 years is likely to have been due to anthropogenic influence;
• For the next two decades a warming of about 0.2°C per decade is projected
for a range of SRES emission scenarios. Even if the concentrations of all
greenhouse gases and aerosols had been kept constant at year 2000 levels, a
further warming of about 0.1°C per decade would be expected.
• It is very likely that hot extremes, heat waves, and heavy precipitation
events will continue to become more frequent.
• Extra-tropical storm tracks are projected to move poleward, with
consequent changes in wind, precipitation, and temperature patterns,
continuing the broad pattern of observed trends over the last half-century.
“Very likely” denotes the possibility of above 90%, “likely” states the possibility of
at least 66%.

Conclusions:
Global warming is an undeniable fact evident from the most sensitive indicators in the
form of frozen fresh water which is the lifeline of billions of people. Himalaya-
Karakoram-Hindukush ranges interconnected with each other house world’s largest
number of Mountain Glaciers. In this way, they possessed the largest ice mass after the
polar regions. As the warmer atmosphere interacts with these ice reserves, they are
suffering from faster melting. Increasing warming rate leaves the alarming signals that
their life span is reduced to several decades only. In recent changes observed in climatic
behaviour, the anthropogenic factor is more pronounced than the natural one. The
results of isothermic dynamics, changes in frequency and duration of heat waves,
hydrodynamics and cryodynamics are discussed in this paper.
Pentad analysis of summer temperatures including data from various sources (1981-
2008) revealed that 30°C isotherm has moved to 725m above the elevation where it
existed three decades before. A study conducted in the same area on the basis of
meteorological data only showed the position of that isotherm at 350m by the end of
2005 (Rasul et al., 2006). It reflects that warming is a continuous process and it is
invading the higher elevations from the plains along the southern slopes of HKH ranges.
There was a rare occurrence of heat waves (temperature >35°C) at an elevation above
1500m in 1970s but now continuous spells of 40°C have been recorded. The frequency

74
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
of moderate as well as severe heat waves has increased to almost double now as
compared to 25 years before. These heat waves have not only become frequent but also
persistant over longer time scales. The spatial extent has also increased by about three
folds. Due to the increase in frequency, intensity and prevalence of heat waves, the
stream flow has increased resulting into downstream floods occasionally.
Anthropogenic related warming has badly damaged the Siachen Glacier which had been
the war-front between Pakitani and Indian forces since mid 1980s. Large human
presence and their related military activities caused 4°C increase in temperature on the
average in that areas. Satellite estimates show that Siachen has reduced its snout by 1.9
km from 1989 to 2006. It is worth-mentioning that not only it has contracted
longitudinally but also has lost about …….% layered ice thickness. Study on Batura,
Baltoro and Biafo is in progress and initial results show a depleting trend in general.

References:
Barry, R. G., 2002: Changes in Mountain Climate and Glacio-hydrological Responses.
Mountain Research Development. 10, PP. 161-170.
Rasul, G. and Q. Z. Chaudhry, 2006: Global Warming and Expected Snowline Shift
along Northern Mountains of Pakistan.Proc. of 1st Asiaclic Sympos. Yokohama, Japan.
Fowler, H. J. and D. R. Archer, 2006: Conflicting Signals of Climate Change in the
Upper Indus Basin. Journal of Climate vol 19. PP 4276-4292.
Samiraglia, C., C. Mayer, C. Mihalcea, G. Diolaiuti M. Belo and G. Vassena, 2007:
Ongoing Variations of Himalayan and Karakoram Glaciers as Witnesses of Global
Changes: Recent Studies on Selected Glaciers. Development in Earth Surface Processes
No. 10 “Mountains Witnesses of Global Changes”. Elsevier Publishers, pp 235-248.
Shroder Jr, J. F., M. P. Bishop, H. N. N. Bulley, U. K. Haritashya and J. A.
Olsenholler, 2007: Global Land Ice Monitoring from Space (GLIMS) Project Regional
Center for Southwest Asia (Afghanistan and Pakistan). Development in Earth Surface
Processes No. 10 “Mountains Witnesses of Global Changes”. Elsevier Publishers, pp
187-207.
Bishop, M. P., J. F. Shroder Jr, U. K. Haritashya and H. N. N. Bulley, 2007:
Remote Sensing and GIS for Alpine Glacier Change Detection in the Himalaya.
Development in Earth Surface Processes No. 10 “Mountains Witnesses of Global
Changes”. Elsevier Publishers, pp 209-234.
Hasnai, S. I., 1999: Report of Working Group on Himalayan Glaciology. International
Commission on Snow and Ice, 22.
Mastny, L., 2000: Melting of Earth’s Ice Cover Reaches new High. World Watch
Institute. Washington DC.
Shrestha, M. L. and A. B. Wake .P.A. Shrestha, 2004: Recent Trends and Potential
Climate Change Impacts on Glacier Retreat/ Glacial Lakes in Nepal and Potential
Adaptation Measures. OECD Global Forum on Sustainable Development. Development
and Climate Change Paris, France. OECD PP. 5-14.

75
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
Hewitt, K., 1998: Glaciers Receive a Surge of Attention in the Harakoram Himalaya.
Ecos, Trans. Amer. Geophys. Union. 79, 104-105.
World Glacier Monitoring Service, 2002: Glacier Mass Balance Data 2000/01.
(Available online at www. Geo.unizh.ch/wgms/)
Aihong, X., R. Jiawen, Q. Xiang and K. Shichang, 2007: Reliability of NCEP/NCAR
Reanalysis Data in the Himalayas/ Tibetan Plateau, J. Geo. Sci., DOI 10.2007/s11442-
007-0421-2.
IPCC (eds.). Climate Change 2001: The Scientific Basis. Cambridge University Press:
Cambridge, 2001.
IPCC 4th Assessment Report 2007: Available online

76
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

DIAGNOSIS OF THE IMPACT OF DEEP


DEPRESSIONAL ACTIVITY IN NORTHERN ARABIAN
SEA OVER KARACHI DURING MONSOON
Muhammad Hasan Ali Baig*, Ghulam Rasul* TP PT

Abstract:
The burgeoning usage of Numerical weather prediction techniques provide numerous
dimensions to forecasting in modern days. As reliance on these models is increasing,
hectic involvements to address the shortcomings of their applications is also in full
swing. In this study an attempt has been made to diagnose the devastating dust/wind
storm accompanied by torrential pre-monsoon rain which lashed Karachi and smashed
the city’s infrastructure. Sudden and rapid development of supercell Cumulo nimbus
(Cb) clouds over Karachi in the outskirts of strong Deep Depression over Arabian Sea
followed by severe wind storm (whole gale) which deceived the forecasters has been
analyzed here. High Resolution Regional Model (HRM) has been utilizing by PMD for
forecasting purposes. In this study simulations based upon the U.S. National Centre for
Environmental Prediction (NCEP) and Albany data under the scope of conventional
techniques have been analyzed and compared with HRM simulations to diagnose the
causes of this disastrous activity and to address the discrepancies/limitations of the
HRM. Model successfully predicted the pressure fall, and upper atmospheric conditions
at 1200 GMT. Little deviation has been observed in HRM’s simulations for humidity
and temperature to the actual data recorded at different meteorological stations of
PMD. Almost complete deviation is observed in simulations for predicting wind,
cloudiness and precipitation from actual record. This event has forced the modelers to
review its physical processes, parameterization and numerical schemes for capturing
wind, cloud cover and precipitation to realistic limits.
Key Words: Deep Depression, Pakistan Meteorological Department, Monsoon, High
Resolution Regional Model, Numerical Weather Prediction, Mesoscale
Phenomenon.

Introduction:
Pakistan is located in South Asia and extends northeast to southwest from the Arabian
Sea between latitudes 24 o N to 37 o N and longitudes 60 o E to 75 o E. It has a typical
characteristic climate by virtue of its peculiar orographic features, such as the Himalaya
and Hindu- Kush ranges in the north and northwest respectively (Chaudhry et al, 2007,
Khan, 1993). These features play an important role in establishing all four seasons in
Pakistan. In the present study, our main focus is on the dust/thunderstorm activity took
place on 23rd June, 2007 in Karachi, which devastated the metropolitan city. Inhabitants
of Karachi experienced the unforgettable event of their lives in the form of severe
thunderstorm and heavy rain with strong gusty winds, which started almost at 1145UTC

*
TP TP Pakistan Meteorological Department
77
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
(4:45 pm of PST). This mesoscale phenomenon of monsoon deceived the forecasters
and modern techniques in pursuit of its rapid development. Pakistan Meteorological
Department (PMD) issued a communiqué but that much sudden development of Cb
clouds during such a short interval of time accompanied by wind storm was never
expected. Karachi is situated at 24.86o N and 67.01o E along northern coast of Arabian
Sea. This windstorm lashed the infra-structure of the city. Carnage by weather had been
reported by media, the strong wind and rain late Saturday wreaked havoc in Karachi,
which received 17.2mm of rain within one hour. Death toll of 228 people had been
reported (Dawn, June 24, 2007). Pakistan Meteorological Department issued an
advisory report on June 22, 2007 which predicted widespread heavy rain by declaring
strong monsoon weather system (deep depression) over India as a cause of such unusual
climatic change.
In June, at least one thunder or dust storm is expected every year in Karachi.
Thunderstorm may be accompanied with strong squalls (Shamshad, 1988). The rainfall
over Karachi in summer months is generally inadequate which occurs mostly due to
local heat thunderstorms (Khan, 1993). Because during pre-monsoon season South-
Westerly flow in the presence of strong convection gives rise to occasional strong
buildups of Cumulonimbus clusters to produce intense rain and strong down drafts.
The weather we experience on our planet is caused by the complex atmospheric and
oceanic processes governed by certain physical laws. NWP relies on the fact that the
gases of the atmosphere obey a number of known physical principles like the hydrostatic
equation describes vertical motion in the atmosphere and the first law of
thermodynamics is used to predict changes in temperature that result from the addition
or subtraction of heat or from expansion and compression of air (Lutgens, 2004)
Depending on the ranges for which forecast is being made, there are different models,
like a global model for a medium range forecast (4 to 10 days), a regional model (a
limited area model) for short range forecast (few hours to few days) and a meso-scale
model for a very short range forecast (up to 12 hours) (NCAR,2002). A regional
climate model (RCM) is a downscaling tool that adds fine scale (high resolution)
information to the large-scale projections of a global general circulation model (GCM).
Normally a global model provides boundary conditions for regional model and a
regional model provides boundary values for a mesoscale model (Das,2002). There are
three types of technique for obtaining regional climate change projections: statistical,
dynamical and hybrid (statistical-dynamical) techniques. RCMs fall into the dynamical
category. Basically two types of mesoscale prediction models can be identified, the non-
hydrostatical and hydrostatical mesoscale prediction models. The non-hydrostatic
mesoscale models are of very high horizontal (<20 km) and vertical (0.5 km) resolution
(Pielke et al., 1992). However, the hydrostatic mesoscale prediction model operates with
horizontal grid resolution of roughly 20 to 100 km and vertical resolution of about 1 km.
But the hydrostatic High Resolution Regional Model (HRM) can be used upto the
resolution of ~11km.The non-hydrostatic Mesoscale model equations are formulated in
advection form, using temperature and pressure as prognostic variables, along with the
velocity components and moisture fields (Black, 1994).

78
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
When developing 10 20 km numerical weather prediction (NWP) models, hydrostatic
water loading should be evaluated in preference to adopting non-hydrostatic models,
while a non-hydrostatic model with hydrostatic water loading is thought to be
recommendable for a high-resolution NWP model (Kato, 1997).
In this paper an effort has been made to investigate the merits and shortcomings of the
model in forecasting the mesoscale phenomenon so that in future more careful reliance
and correlation between conventional forecasting techniques and Numerical Weather
Prediction can be made in a cautious manner.
The high resolution model, popularly known as HRM, is a regional numerical
hydrostatic model used for short-term forecasts. It is based on a set of equations that
describe the evolution of variables such as temperature, wind speed, humidity, and
pressure defining the weather or the state of the atmosphere. The HRM uses a hybrid
sigma pressure coordinate system. Near the surface, the model follows the terrain with
sigma coordinates while the upper atmosphere employs the pressure coordinate system.
δ-Two stream radiation scheme including long- and shortwave fluxes in the atmosphere
and at the surface; full cloud - radiation feedback; diagnostic derivation of partial cloud
cover (relative humidity and convection) is employed in the model. Grid-scale
precipitation scheme including parameterized cloud microphysics is also used. Mass
flux convection scheme differentiating between deep, shallow and mid-level convection
is utilized.Level-2 scheme of vertical diffusion in the atmosphere, similarity theory at
the surface and for soil seven-layer model including snow and interception storage is
used (Majewski, D., 2007). It comprises of 40 vertical layers while the horizontal
resolution of 22 km (0.22o ) has been utilized here.
The process starts with analyzed data using observations and previous forecast to obtain
the best estimate of the current state of the atmosphere. A global model (GME) is used
to produce a forecast by the Deutsche Wetterdienst (DWD), the Meteorological Service
of Germany. The Pakistan’s area of interest is then singled out to provide the initial and
boundary conditions of the limited area model HRM. At the receiving station the HRM
can run in parallel to the GME at DWD using the interpolated GME analysis as initial
state and GME forecasts from the same initial analysis as lateral boundary conditions.
This is a considerable step forward because up to now most regional National
Meteorological services (NMS) have to use 12h “old” lateral boundary data which
degrade the regional forecast quality by more than 10%.

Methodology:
In pursuation of the goal of analyzing the devastating extreme event in the form of
severe wind storm by the eye of HRM, firstly the data acquired from the Climatological
Data Processing Centre (CDPC), Karachi has been analyzed. Then NCEP Reanalysis
data was used for studying the synoptic situation of the atmosphere above and around
Karachi. The NCEP/NCAR Reanalysis 1 project is using a state-of-the-art
analysis/forecast system to perform data assimilation using past data from 1948 to the
present. Besides NCEP charts, manually prepared synoptic charts at Main Analysis
Centre, Karachi were also analyzed in detail for grasping the true picture.

79
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
For supporting the recorded data, Satellite pictures were also considered to grasp an
accurate picture of surface and atmosphere of the selected region. Finally, HRM model
has been used for prognosting the situation. Some analysis developed by using the
software ‘Digital Atmosphere’ based on the data of University of Albany, New York are
also presented where coarse resolution of NCEP data created hurdles. All these
simulations are visualized by using GrADS software.
Related Surface Data
In the following table the trend of fall in pressure and rise in temperature is
presented from the 00Z of the 22nd June, 2007 to the onset of devastating
thunderstorm nearly just before 12Z of 23rd June, 2007. While data given in the
table shows the high humidity just before the occurrence of event and rise in
maximum temperature on that day. From the table it is evident that almost northerly
winds have been recorded which were blown from land and cause rise in
temperature with maximum velocity at the time of onset of thunderstorm. Rain
recorded for 24 hours from 03Z of 22nd June to 23rd June has been mentioned
under 22nd June and that from 03Z of 23rd June to that of 24th June has been
mentioned under 23rd June. It is important to note here that 17.2mm rain has been
recorded for just one hour as no rain was recorded other than that heavy downpour
up to 03Z of 24th June. Sky was totally covered with clouds at the time of
occurrence of event.
Table 1: Pressure and Temperature with other data recorded at M.O., Karachi
22nd June, 2007 23rd June, 2007 Corresponding Differences

Parameters
0000 UTC

0300 UTC

1200 UTC

0000 UTC

0300 UTC

1200 UTC

0000 UTC

0300 UTC

Pressure(hPa) 997.9 998.6 994.6 995.9 996.6 992.6 -2 -2 1200 UTC


-2

Temperature(oC) 27 30.5 38.5 32.5 35 29 +5.5 +4.5 -9.5

Temperature(oC)
39.3 42 +2.7
(Maximum)

Temperature(oC)
23 31.5 +8.5
(Minimum)

Humidity (%) ----- ----- 50 ----- ----- 78 ----- ----- +28

0 0 6 4 6 60
Wind (knots) +4 +6 +54
(N-360o) (N-360o) (NW-225o) (N-360o) (N-360 o) (NE-45o)

Rain (mm)
19.4 17.2 -2.2
24-Hrs

Clouds (Oktas) 7 4 5 0 4 7 -7 0 2

Analysis of Synoptic Condition


The sagacious and discreet analysis of the NCEP simulations along with synoptic
conditions and available data at CDPC reveals that three days prior to this incident,
80
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
a depression was present over India which later intensified in the form of deep
depression. This system sucked all the moisture from Arabian Sea caused the wind
to blow from northeast direction over entire lower Sindh including Karachi. Under
the influence of this deep depression, a low was formed over northeast Arabian Sea
and adjoining areas of lower Sindh. All the winds from surface level to 300hpa level
were north easterlies. This northeast flow caused the day temperature to rise as high
as 42 °C with relative humidity 78% on that particular day, i.e. on 23rd June, 2007.
The temperature rising trend i.e. from 36.5 °C to 42 °C persisted for preceding six to
seven days with unusual high humidity having resulted in huge energy
accumulation, owing to tremendous convective instability around Karachi. And the
final result was the outburst in the form of tornado/gale winds followed by heavy
dust/thunderstorm that lashed the city in few minutes. The maximum wind speed of
60 knots was recorded at Meteorological Observatory (M.O) Karachi Airport. The
surface visibility was as low as 500m at airport with approaching to zero in some
parts of city.
Pressure & Temperature Variation

1000 45
999 40
998
35
997
996 30

995 25
994 20
993 15
992
10
991
990 5
989 0
Z

Z
00

03

12

00

03

12
7

7
-0

-0

-0

-0

-0

-0
un

un

un

un

un

un
-J

-J

-J

-J

-J

-J
22

22

22

23

23

23

Pressure hPa Temperature deg C

Figure 1: Analysis of Falling Pressure and Rising Temperature trends.

Synoptic History of this historic incident reveals that an area of convection


developed and persisted in the central Bay of Bengal on 19th of June about 165
nautical miles east of the Andaman Islands. It was observed that observed that deep
convection was increasing near a developing mid to low-level circulation, and
surface observations in the area had revealed P24 (24-hour Pressure change) of -
1.5hPa. While the upper level environment was characterized by diffluent flow, a
strong tropical easterly jet was contributing to 20-30 kts of vertical shear. The
disturbance drifted slowly to the west, and by the next day the Low Level
Circulation Centre (LLCC) had consolidated some and the potential for
development was upped to ‘fair’. Although the environment was still not
particularly favourable for cyclogenesis, surface pressures had continued to fall by
about P24 = -2.0. At 0300 UTC on 21st of June, the Pressure area was coined to a
depression with winds estimated at 25 knots by Indian Meteorological Department
(IMD).

81
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

On 21st of June, the deep convection had developed over the well-defined LLCC.
At 1200 UTC of 21st June, 2007 IMD classified the system as a deep depression
while at the same time warning issued by the Joint Typhoon Warning Centre
(JTWC) declared it as Tropical Cyclone 03B. the system was centred about 100 nm
south-southeast of Visakhapatnam and moving west-northwestward at 14 knots.
Deep convection was located primarily over the south-western quadrant of the
LLCC. TC-03B was tracking westward along the southern periphery of a mid to
upper-level ridge over northern India.

a b

c d

Figure 2: Surface Pressure patterns (a-d) before the incident from 0000 UTC of 22nd June,2007 to 1200
UTC of 23rd June,2007 with 12 hrs intervals.

82
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

Figure 3: Stream Lines Analysis at 200hPa (a-d) showing strong Anti-cyclonic Circulation from 0000 UTC of
22nd June,2007 to 1200 UTC of 23rd June,2007 with 12 hrs intervals. It shows that almost Easterly Regimes were
predominant over the Arabian Sea in the beginning but later they started to transform into Cyclonic Circulation.

Pakistan’s area of interest started when deep depression (which was later declared
by PMD as TC-Yemyin) started subsequently moved westward and inland into
coastal Andhra Pardesh around 0000 UTC of 22nd of June,2007 as evident from
the figure 2(a). At the same time bubble low or cut of low (996 hPa) was developed
at 70° E and 24° N in the trough of above discussed deep depression which itself
was centred at 82° and 17° N with pressure of 994 hPa.
North-westward movement was tracked by this deep depression while low at SE
part of Pakistan intensified further upto 994 hPa with more or less on the same
location at 1200 UTC of 22nd June. At 0000UTC of 23rd June both lows merged in
single low with troughs extending NW, SW and SE-wards. At 0600 UTC, North
Western trough converted into mesoscale low while the South Western trough
merged with the above system. A shallow low was present in the South Eastern
trough. At 1200 UTC, it further intensified upto 992 hPa and located at 70o E and
25° N. At this time severe wind storm was observed by the inhabitants of Karachi in
the periphery of this deep depression.

83
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
Upper air conditions also supported the deep depression to become more intense.
Strong anti-cyclonic circulation was pervasive at 200 hPa with centre over Iran and
extension in the form of ridge was dominant over sub-continent as shown in the
figure 3(a & b). But a clear trough was present (refer figure 3: c & d) between two
anticyclonic circulatios i.e. one over Iran and second over the Nort-Eastern India at
0000 UTC of 23rd June, 2007. These two highs supported the convergence over
land in the form of deep depression
At 500 hPa situation is also anti-cyclonic below the same anti-cyclonic centre of
200hPa. But strong cyclonic circulation is evident in figure 4 (a-d) over South
Western India and its extension over Arabian sea with second inter-linked cyclonic
circulation over the coastal regions of Gulf. This Cyclonic circulation strengthened
the deep depreession over Surface. Side by side if isotachs are also analyzed then it
is revealed from figure 7(d) that wind was blowing fast upto >30knots almost over
the surface’s deep depression.

Figure 4: Stream Lines Analysis at 500hPa (a-d) showing movement of Anti-cyclonic Circulation
from SW direction with 0000UTC of 22nd June,2007 over Iran to 1200 UTC of 23rd June, 2007 over
Afghanistan with 12hrs intervals. Cyclonic Circulation is also dominant over Arabian sea and South
Eastern India.

84
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
Analysis of wind stream charts of 700 hPa show that cyclonic circulation had been
intensifying from 22nd June, 2007 till the day of occurrence of tragic wind storm.
Two inter-linked patterns of strong cyclonic circulation can be seen in the attached
figures. But shallow ridge was present over Karachi at 23rd of June, 2007. This
ridge might become helpful in strong upward convective movement which led to
support the formation of severe storm.
Upper air patterns of Geopotential heights can also be seen in figure 6 with
isotherms. Geopotential heights of 200 hPa support parallel isotherms in normal
conditions. If they are nearly perpendicular to the contours then it would be an
indication of arrival of disturbances at surface after some duration which is
important for forecasting. One High is persistent over Iran and Balochistan at 200
hPa. This High is normally form over the seasonal low of Balochistan. Surface
systems are tilted NWard in the upper air. Analysis of figure 6 (a-d) makes it clear
the Sub Tropical high (STH) had been intensifying gradually. Its orientation shows
that it supported the cold air advection from higher latitudes of North towards South
and created instability. Convective unstable vertical currents with lack of moisture
supply usually develop to produce local dust/sand storms or simply dust devils.
Sometimes, in the case of a moisture supply, they produce intense local showers
(Rasul et al., 2005).

Figure 5: Stream Lines Analysis at 700hPa (a-d) from 0000 UTC of 22nd June, 2007 to 1200 UTC of
23rd June, 2007 with 12 hrs intervals, showing intensification of Cyclonic Circulation over SW India,
Arabian Sea and Coastal Belt of SE Gulf.

85
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

When Isotherms are also analyzed then they are observed nearly perpendicular in
some 200 hPa charts of 22nd June, 2007. 500 hPa charts reveal that core of Low
was warm and the whole pattern was shifting towards South-Eastern Pakistan from
22nd of June, 2007 to 23rd of June, 2007. Interesting features of this deep
depression unearth when isotherms of 850hPa and 700 hPa are analyzed. It is
revealed from those charts that at these levels deep depression had a characteristics
of Cold Core Low.

a b

d
c

Figure 6: Geopotential Heights (solid black line) and Isotherms (dotted black line) at 200 hPa with a & b
representing 0000 UTC & 1200 UTC of 22nd June, 2007 and c & d representing 0000 UTC & 1200 UTC of
23rd June, 2007 respectively. Intensification of STH is evident over Afghanistan, Iran and Balochistan with
the passage of time.

Jet Stream is pervasive at 200 hPa and its movement can be analyzed as Westerly
Jet Stream from North of Pakistan. Easterly Jet Stream can also be analyzed over
Bay of Bengal, Southern India and Arabian Sea. Between these two Patterns High
exists over the sub-continent at 200hPa. Isotachs at 500hPa also show the westerly
Jet Stream over the same location.

86
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

a b

c d

Figure 7: Isotachs showing movement of strong Jet Stream (a: 1200 UTC of 22nd June, 2007 at 200
hPa; b: 1200 UTC of 23rd June, 2007 at 200 hPa; c: 1200 UTC of 22nd June, 2007 at 500 hPa; and d:
1200 UTC of 23rd June, 2007 at 500 hPa; ) from West to North of Pakistan at 200hPa and weak Jet
Stream at 500 hPa in 24 hrs. Strong Easterlies are pervasive over South-eastern India and Arabian Sea.
In (e & f) Isotachs showing movement of strong currents accompanied by Depression at 850 hPa over
Central India towards Pakistan (e: 1200 UTC of 22nd June, 2007 & f: 1200 UTC of 23rd June, 2007.

87
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
But Easterlies were quite slower which helped in movement of deep depression
towards West direction. Low level Jet stream catalysed the movement of the Deep
Depression in the vicinity of Karachi. Synoptic history of cyclones reveals that
strong activities over surface occur where Jet streams persist at upper troposphere
and lower stratosphere.
Comparison b/w the Actual Pattern (NCEP real analysis) and
HRM Forecast:
An excellent coincidence between the actual pressure present at Karachi with the
prognostic pressure and the pressure recorded at Meteorological Observatory (M.O),
Karachi has been shown in figure 8. In Figure 9, upper level geopotential charts
based on NCEP data and HRM simulations are compared. These charts also show
almost complete coincidence between actual data and HRM forecast. At 850 hPa
real analysis of NCEP data shows Karachi under the contours between 1420 gpm
and 1440 gpm while HRM simulation represents 1420 gpm over Karachi. At 500
hPa, contours between 5860 gpm and 5880 gpm are present over Karachi. While
almost the same picture is represented by HRM simulation.

a b

Figure 8: (a) indicates the actual Sea Level Pressure Pattern with 992 hPa at South Western parts
of Pakistan while HRM forecasts in (b) is indicating the more intense pressure almost over the
same location with trough extending towards the coastal belt of India. HRM forecast was initialized
at 0000UTC of 22nd June,2007
P P

88
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

b
a

Figure 9: (a) is representing the actual pattern of Contours at 850hPa with blackish dotted line
representing the temperature. In (b) HRM forecast is showing the closed cell of almost equal
low with extended Trough over Arabian Sea as indicated in (a). Black dotted line of Isotherm
in (b) showing the low as Cold Core Low.

b
a

Figure 10: (a) is representing the actual pattern of Contours (solid black line) at 500hPa
1200 UTC of 23rd June, 2007 with blackish dotted line representing the temperature. In
(b) HRM forecast is showing the closed cell of almost equal low value with extended
Trough over Arabian Sea as indicated in (a). Black dotted line of Temperature in (b)
showing the centre of low at higher values in comparison to adjoining values of
temperature.
As far as temperature is concerned for comparison between the real analysis and
HRM forecast, Table 1 indicates that the recorded temperature at 1200Z was 29 °C
but HRM simulation gave temperature values between 40 ° C to 42 ° C which is
much higher than the actual value. The fall in temperature at that time is due to cold
downward advection of air at 12Z due to the presence of Cb clouds over Karachi.
Deep Depression clearly appears as Cold Core Low with SST at the core of the
Depression as 24° C in actual analysis while HRM forecasted SST as 25° C in the
Depression region.

89
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

a b

Figure 11: Vertical Profile of Rel. Humidity at 67o E longitude of Karachi in actual scenario (a) and in HRM
Forecast (b) at 1200 UTC of 23rd June, 2007.

Figure 11 represents the comparison of the vertical profiles of relative humidity


between NCEP simulations and HRM simulations longitudinally. Coincidence
between both simulations is obvious although the recorded value of relative
humidity was much higher than that presented by these simulations. As Karachi is
located at 67.01° E, so a discreet look over this longitude shows almost 70%
humidity above 760 hPa upto 680 hPa in NCEP simulation while almost 65%
humidity is represented in HRM simulation between 760 to 680 hPa. At 19° N
higher humidity in vertical profile indicates that the airmass of that region was
highly unstable due to the presence of deep Depression there. When latitude of
Karachi is fixed at 24.86° N, it was observed that 70% relative humidity was
present between 720 hPa and 680 hPa besides more than 60% relative humidity
presence at more vertical extent in NCEP simulation. Actual image shows that air
was dry near the surface and wet aloft caused convective instability. It supported the
development of Cb clouds with strong vertical updrafts which led to cause severe
wind storm. While HRM simulation also coincides with NCEP simulation by
showing more than 70% relative humidity between 760 hPa and 650hPa. It is
important to mention here that relative humidity was much high at oceanic region
near Karachi due to the presence of deep depression (which later became Tropical
Cyclone Yemyin-03B) in the North Indian Ocean. NCEP data (Kalnay et al, 1996)
which is used here provided by the NOAA/OAR/ESRL PSD, Boulder, Colorado,
USA, from their Web site at http://www.cdc.noaa.gov/
HT TH

90
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008

Figure 12: Wind Speed (Knots) representation in actual observation at different stations of
Albany data (a) is clearly indicating the 60 Knots Wind at Karachi. (b) is representing the HRM
forecast of 10 knots at the same station.

91
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
Figure 12 represents the comparison between the images of winds of Albany data
and HRM forecast. 60 knots wind can be seen in a barb form at station of Karachi in
the image of Digital Atmosphere but less than 15 knots wind is shown by HRM
simulation which shows the complete deviance from the recorded one.

Figure 13 (a): Recorded Rain for Karachi and adjoining areas of Sind and Balochistan with
more than 16mm rain at Karachi Source: Climatological Data Processing Centre(CDPC),
Pakistan Meteorological Department.

a b

Figure 13 (b & c): HRM Forecast of Rain at Karachi at 1200 UTC and 1300 UTC on 23rd June,
2007 with 0mm rain in both simulations

Table 1 is a proof with satellite picture of 14(b) that sky was almost overcast over
Karachi at the time of thunderstorm/windstorm while less than 1 oktas of clouds
were forecasted by HRM in its simulation (figure is not given here). This again
represents variance from the actual cloud coverage.

92
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
Observed rain of Karachi and adjoining areas of Sind and Balochistan has been
analyzed in the form of contours which is representing more than 16 mm of rain at
Karachi (recorded rain was 17.2mm) in figure 13(a) while HRM forecasts no rain at
Karachi which shows the complete mismatch with the actual one (figure 13: b & c).

Figure 14: (a)Met-7 (Dundee), 22-June-2007, 1200Z (b): Met-7 (Dundee), 23-June-2007, 1200Z
Satellite pictures are showing the cloud picture of clouds at different time intervals.
Dense cloud cover was present over Arabian Sea before the historic wind storm at
Karachi. At 1200 UTC strong Cb clouds can be seen in figure 14(b). This shows
that system covered almost whole Karachi with severe wind storm of Gale intensity.

Results and Discussion:


It is a crystal clear fact which from the above analysis and diagnosis of synoptic
conditions that a quagmire in a form of deep depression was present over Arabian Sea
near Karachi two to three days prior to this devastating thunderstorm. It was fait
accompli that deep depression became a peculiar cause of unpredicted and accidentally
happened thunderstorm of Karachi. When satellite pictures are being analyzed
expeditiously in the context of all synoptic charts, as a professional demeanour,
forecaster found himself between the two fire lines in predicting the exact time of
thunderstorm activity as a mesoscale phenomenon. That deep depression and the
convective movement of air over Karachi due to rise in temperature exacerbated the
situation that resulted in a form of burst which led to severe wind storm.
In monsoon season at 300 hPa or so there exists a well marked anticyclone over Iran and
another over Northeastern India and Bangladesh. As these two anticyclones are the
reflections of surface low pressure areas in the upper air, their position governs the
general circulation of monsoons over the southern Asia. The trough of low pressure
(with its core over Balochistan) over the Gangetic plain (Monsoonal Trough) plays an
important role in determining the weather conditions in Monsoon period. The axis of
this trough is a semi-permanent zone of convergence. Generally the outbreak of
monsoon showers is preceded by squalls and thunderstorms. Afternoon thunderstorms
accompanied by dust storms occur frequently over the country (Shamshad, 1988). Dust
storms of gradient and instability types caused the peak of wind speeds in squalls just
before the onset of the monsoon. Vertical Wind Shear (which was present in this
93
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
activity) where winds increase or change direction with height serves to enhance rather
than weaken the strongest storms through wind removal of precipitation from the
updraft summit, and in some cases introduction of updraft rotation as in the case of
Supercell thunderstorms. Generally the strongest and most violent tornadoes are
spawned by these supercell thunderstorms. As strong winds were present in the upper
atmosphere in the vicinity of Jet Stream so divergence aloft played an important role in
this activity and intensification of deep depression into Tropical Cyclone. Upper air
trough was present as discussed earlier, so an area of upper air divergence and surface
cyclonic circulation developed downstream from an upper-level trough. As from the
scrutiny of surface and upper atmospheric patterns it is somewhat clear that divergence
aloft exceeded convergence at ground level which caused surface pressure to fall and
intensified the cyclonic storm. As mid-level of troposphere was analyzed as cooler
especially at the centre of deep depression, so this mid-level tropospheric cooling forced
strong localized upward motions at the leading edge of the region of cooling which then
enhanced local downward motion at the time of thunderstorm. In the mature stage of Cb
clouds when strong updrafts and downdrafts are pervasive and tornado activity was also
there, severe wind storm blew up the infrastructure.
At surface a low was formed over east-central Bay of Bengal on 20th June. It later
concentrated into a depression at 0300 UTC of 21st June. It further intensified into a
deep depression at 1200 UTC of 21st June and lay centered near 16.0° N and 84.0° E. It
then moved west-northwestwards and crossed Andhra Paradesh coast around 2300 UTC
of 21st June. It continued to move in a west-northwesterly direction and on 23rd June,
2007 at 1200 UTC, it was at 70° E and 25° N during which at its outskirt, Karachi faced
severe wind storm. As the recorded wind was 60 knots, so it is classified according to
WMO standards as ‘Severe Cyclonic Storm’. It behaved like a tornado and uprooted
strong trees with devastation of infrastructure and sign boards.

Conclusions:
In this paper the devastating severe wind storm accompanied by thunderstorm and rain
(17.2mm in 1hr) produced by a mesoscale convective system over Karachi in pre-
monsoon season adjacent to inten se deep depression over Arabian sea has been studied
and an attempt is made to address the causes of its development by detailed synoptic
analysis. It is tried to diagnose its behaviour for forecasters with special emphasis on
NWP tool i.e. HRM which is being used by Research and Development Division
(R&D), Islamabad of Pakistan Meteorological Department in collaboration of Deutsche
Wetterdienst (DWD), Germany. The main results are:
• The trend of rise of temperature due to north easterly winds coming from the
more hot land and fall in pressure supported by the strong deep depression over
Arabian sea made easy for the rapid and sudden development of meso-scale
convective system to get moisture from Arabian sea.
• Pressure gradient was enough and heavy downdrafts of air resulted in creation
of severe wind storm (at least 60knots) accompanied by torrential rain lashed
the human-made installations of the metropolitan coastal city of Pakistan.

94
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
• Continuous supply of water vapours and instability due to warm surface as
compare to upper air paved a way for moist vertical currents to develop
marvelous Cb cloud.
• High Resolution Regional Model went good for prognosting fall in pressure and
trend of contours at different geo-potential levels. Model appears to some what
satisfactory in prognosting parameters of relative humidity and temperature.
• Model appears almost fail in correct simulation of wind, cloudiness and
precipitation.

Recommendations:
It is recommended that the mere reliance on model itself may lead to failure for micro-
and mesoscale phenomenon, therefore conventional techniques must also be taken into
account. Need of further case studies to address the shortcomings of the model is
realized. There appears a necessity of upper atmosphere data for plotting Tephigram
which is a strongest tool for forecasting mesoscale thunderstorm activity over a
particular station. In addition to HRM, it is advisable that side by side we should avail
the opportunities of other regional models along with further modification in HRM itself
according to our own topography.

Acknowledgments:
Authors express their deep gratitude to all those who cooperated in materializing of this
research paper especially the scientists of R&D, Islamabad and Forecasters from both
Karachi and Islamabad. We pay tribute to Dr. David J. Knight of University at Albany,
State University of New York for providing data which has been used in Digital
Atmosphere.

References:
Black T.M, 1994. NMC’s ETA model: Description and Examples. Weather and
Forecasting, 9, 265-278.
Chaudhry, Q.Z., G. Rasul., 2007. Development of a mesoscale convective system over
the foothills of Himalaya into a severe storm. Mountains witnesses of Global Changes,
ELSEVIER, Ch.33, pp 301-311.
Das, S., 2002. Real Time Mesoscale Weather Forecasting over Indian region using
MM5 modelling system. NCMRWF R.No. NMRF/RR/3/2002, NCMRWF, Noida,
UP, India.
Kalnay et al., 1996. The NCEP/NCAR 40-year reanalysis project, Bull. Amer. Meteor.
Soc., 77, 437-470,.
Kato, Y., 1997. Hydrostatic and Non-hydrostatic Simulations of Moist Convection:
Review and Further Study. Meteorology and Atmospheric Physics, 63. Springer-
Verlag, Austria. pp 39-51.
Khan, J.A., 1993. The Climate of Pakistan. Rehbar Publishers, Karachi, Pakistan.

95
Pakistan Journal of Meteorology Vol. 5 Issue 9: July 2008
Lutgens, F.K., E.J. Tarbuck,, 2004. The Atmosphere: an introduction to meteorology.
Pearson Education, Inc. Upper Saddle River, New Jersey, USA.
Majewski, D., 2007. HRM User’s Guide. DWD, Germany.
NCAR, 2002. PSU/NCAR Mesoscale Modelling System (MM5 version 3), tutorial
class, notes and user guide, Available from the National Centre for Atmospheric
Research Boulder, Colorado, USA.
Pielke, R.A., W. R. Cotton1, R. L. Walko1, C. J. Tremback1, W. A. Lyons1,
L. D. Grasso1, M. E. Nicholls1, M. D. Moran1, D. A. Wesley1, T. J. Lee1 and
J. H. Copeland1, 1992. A Comparison of Meteorological Modelling System-RAMS.
Meteorological and Atmospheric Physics. 49, pp 69-91.
Rasul, G., Q. Z. Chaudhry, Z. Sixiong, Z. Qingcum, QI. Linlin, and Z. Gaoying,
2005. A Diagnostic Study of Heavy rainfall in Karachi Due to Merging of a Mesoscale
Low and a Diffused Tropical depression during South Asian Summer Monsoon.
Advances in Atmospheric Sciences, Vol. 22, No. 3, 2005, 375-391.
Shamshad, K.M., 1988. The Meteorology of Pakistan. Royal Book Company, Karachi,
Pakistan, pp 167-169.

96
Information for Authors
Pakistan Journal of Meteorology (PJM) is an international journal on the dynamics, physics and chemistry of
the atmosphere and with papers across the full range of the atmospheric sciences, published bi-annually by
Pakistan Meteorological Department. The journal includes Articles, Notes, and Correspondence. Contributions
from all over the world are welcome.
Submission: A submission to PJM must be the original work of the author(s) and must not be published in any
language elsewhere or under consideration for another publication in a similar form. Manuscript in English
should be submitted to Mr. Sarfaraz Shahid, Editor PJM, Pakistan Meteorological Department, P.O. Box 1214,
Islamabad. Please include your email address, phone and fax numbers on the bottom of the first page of the
manuscript. Two complete copies of the manuscript should be submitted by mail along with its electronic
version in MS Word format on a CD or by email at pmisba@isb.paknet.com.pk.
Manuscript Preparation: Each manuscript should have been carefully reviewed for correctness and clarity of
expression before submission. Complete manuscript (including tables, references, and figure captions) must be
double spaced, printed on one side of the page only, with wide margins, and all pages must be numbered
consecutively.
Components of a manuscript: Each manuscript should include the following components, presented in the
order shown below:
• Title: name and affiliation of each author and email address of the corresponding author. These items
should be written on the title page.
• Abstract: a brief, concise abstract is required at the beginning of each manuscript. Authors should
summarize their conclusions and methodology in the abstract. First person construction should not be used
in the abstract, and references should be omitted since they are not available for abstracting services.
• Key words: 4-6 keywords should be provided.
• Text: The text should be divided into sections, each with a separate heading and numbered consecutively
(e.g. 1, 1.1, 1.1.1, ….).
• Acknowledgements.
• Appendix: Lengthy, mathematical analyses whose details are subordinate to the main theme of the paper
should normally appear in an appendix. Each appendix should have a title.
• References: References should be arranged alphabetically without numbering. The text citation should
consist of the author’s name and year of publication [e.g., “according to Robert (1990),” or “as shown by
an earlier study (Robert, 1990)”].When there are two or more papers by the same author in the same year,
the distinguishing suffix (a, b, etc.) should be added.
• Figure caption: Each figure must be provided with a self-explanatory caption. Authors should include
captions below the figures used for the reviewer copies.
• Illustrations and tables: Each figure and table should be numbered consecutively and cited specifically in
the text by number. All tables should have a title and legend.
Figures: Authors should send high quality hard copy originals of all figures. Authors should strive to submit
their figure at the size they will appear (but do not submit photocopy reductions in place of larger, higher
quality originals).
Mathematical formulae, symbols, units: Different typefaces should be set for different kinds of variables.
Scalar variables are set as italic, and vectors, matrices and tensors are set as bold italic (e.g., V). Units should be
SI with the exception of a few approved non-SI units of wide meteorological or oceanographic usage. Units
should be set in Roman font using exponents rather than the solidus (/) and with a space between each unit in a
compound set (e.g., m s-1).
References: References must be complete and properly formatted. Technical reports and conference
proceedings should be referenced only when no other source of the material is available. Examples:
Boville, B.A., and J.W. Hurrels, 1998: A comparison of the atmospheric circulations simulated by the CCM3
and GSM1. J. Climate, 11, 1327 – 1341.
Krishnamurti, T.N., and T. Murakami, 1981: on the onset vortex of the summer Monsoon. Mon. Wea. Rev.,
109, 344-363.
Rao, Y. P., V. Srinivasan, and S. Raman, 1970: Effect of middle latitude westerly systems on Indian
Monsoon Syp. Trop. Met. Hawaii. No. IV 1-4.
ISSN - 1810 - 3979

List of Contents:

Inundation of Tsunami Waves and its Relation to the Tsunami Runup 5


Humeira Hafeez
Preliminary Results of Agrometeorological Soil Moisture 11
Conservation Experiment at Quetta
Ghulam Rasul
Prediction of Soil Temperature by Air Temperature; 19
A Case Study for Faisalabad.
M. Fahim Ahmad, Dr. Ghulam Rasul
Temporal Model of Hydrometeorology in the Citarum River Basin, Indonesia 29
Ruminta
Thunderstorm Frequency over Pakistan (1961-1990) 39
Zubair Ahmed Siddiqui , Abdul Rashid
Global Warming and Melting Glaciers along Southern Slopes of HKH Ranges 63
Ghulam Rasul, Qin Dahe, Q. Z. Chaudhry
Diagnosis of the Impact of Deep Depressional Activity in Northern Arabian 77
Sea during Monsoon over Karachi
Muhammad Hasan Ali Baig, Ghulam Rasul

PAKISTAN JOURNAL OF METEOROLOGY


Published by
Pakistan Meteorological Department
P. O. Box No. 1214, Sector H-8/2,
Islamabad Pakistan
Designed & Composed by Shahid Mahmood

Вам также может понравиться