Pak. J. Engg. & Appl. Sci. Vol. 8, Jan., 2011 (p.

49-60)

Temperature Zoning of Pakistan for Asphalt Mix Design

M. Waseem Mirza1, Zahid Abbas2 and Mujasim Ali Rizvi3

1. Professor, Department of Transportation Engineering, University of Engineering and Technology, Lahore,

Pakistan, e-mail: mw.mirza@hotmail.com
2. Research Associate, Department of Transportation Engineering, University of Engineering and Technology,
Lahore, Pakistan.
3. Lecturer/Graduate Assistant, Department of Transportation Engineering, University of Engineering and
Technology, Lahore, Pakistan

Abstract
The current asphalt binder specifications in Pakistan are based on the Penetration Grade:
penetration test is performed at 25oC. Penetration is an empirical measure of the consistency that is used
as an empirical indicator of the rutting and fatigue susceptibility of asphalt binder, and is not related to
pavement performance. The new mix design methodology developed under the Strategic Highway
Research Program (SHRP), called the SUPERPAVE is a performance-based approach. The first step in
the implementation of SUPERPAVE methodology is to establish high and low pavement temperatures for
a location. The temperatures define the required Performance Grade (PG) of asphalt binder. This paper
documents the initial ground work towards implementation of SUPERPAVE mix design for establishing
high and low geographical temperature zones. The temperature zoning of Pakistan was carried out by
using temperature data obtained from 64 weather stations. The SHRP and LTPP prediction models were
utilized for predicting pavement temperatures. A significant difference was observed between the
predicted pavement temperatures from both the models. The SHRP model gives higher, high temperature
PG grade providing additional protection against rutting. Since rutting is the most common distress on
flexible pavements in Pakistan, the SHRP models at 98% level of reliability is recommended. PG 70-10
binder seems to be the most common grade that encompasses more than 70% area of Pakistan. However,
currently none of the two local refineries produce PG 70-10 binder, thus it should be a concern for the
highway agencies. The polymer modified asphalt binder produced by Attock refinery (A-PMB)
corresponds to harder PG 76-16 while A-60/70 (PG 58-22) or K-60/70 (PG 64-22) produced at Attock
and National refineries respectively are softer compared to the PG 70-10. Harder grade is more prone to
cracking, whereas softer grade of more prone to rutting. Consequently, the current construction
practices which utilize A-60/70 or K-60/70 may be prone to excessive rutting.

Key Words: Superpave, Temperature Zoning, Performance Grade, Penetration Grade

1. Introduction of the performance based binder specifications termed

as Performance Grade (PG) asphalt. The major
From October 1987 through March 1993, the
objective of the SHRP program was to relate
Strategic Highway Research Program (SHRP)
mechanical properties of asphalt binder to field
conducted a research effort to develop new ways to
performance. Therefore, the new specification tests
specify, test, and design asphalt materials [1]. The end
were developed to characterize asphalt binders at a
result of SHRP asphalt research program is a
broad range of temperatures and aging conditions.
development of a system referred to as SUPERPAVE,
The three aging conditions specified are original, short
which stands for Superior Performing Asphalt
Pavements. One of the key aspects of the term and long term aging. Original refers to virgin
SUPERPAVE is the development asphalt from the production plant; short term

49
Pak. J. Engg. & Appl. Sci. Vol. 8, Jan., 2011

aging refers to properties at the time of production and viscosity grades. The top row values in Table 1
placement of asphalt mix while long term aging refers represent relatively harder binders, whereas the lower
to properties of asphalt binder during the service life rows represent the softer binder.
of pavements. In addition to aging conditions,
SUPERPAVE characterizes the asphalt binders at the Table 1: Penetration and Viscosity Grading System
actual pavement temperatures those are likely to
experience. A brief summary of the binder grading Penetration Grading Viscosity Grading
systems for the purpose of mix design is given in the
Grade Penetration Grade Viscosity
following section. in 0.1 mm @ 60oC,
Poise
2.1 Binder Specifications
Pen 40/50 40-50 AC-40 4000±800
The development of binder testing goes back to Pen 60/70 60-70 AC-30 3000±600
1888, when H. C. Bowen invented the Bowen
Penetration Machine [2]. After several modifications Pen 85/100 85-100 AC-20 2000±400
of penetration equipment, by 1910 the penetration Pen 120/150 120-150 AC-10 1000±200
equipment became the standard for establishing the
consistency of asphalt at 25oC. In 1918 the Bureau of Pen 200/300 200-300 AC-5 500±100
Public Road (USA) introduced penetration grading AC-2.5 250±50
system and by 1931 American Association of State
Highway and Transportation Officials (AASHTO)
published the standard specification to grade asphalt Generally, softer binder grades are used in the
on penetration. The next major change in asphalt cold climates to resist cracking potential and harder
grading specification came with the introduction of binders for warmer climates to resist rutting potential.
viscosity grading system in early 1960s. Both ASTM In Pakistan, a standard grade of Pen 60/70 is used for
and AASHTO adopted the viscosity grading system construction of flexible pavement. However, the
and provided grading specification by measuring the refineries in Pakistan also produce Pen 40/50 and Pen
viscosity at 60oC. 80/100. Viscosity grading is not yet established in
Pakistan.
Penetration grading system based on penetration
of a standard needle under standard conditions in Viscosity grading system based on the
asphalt binder is empirical in nature. The empirical fundamental property is considered a step forward in
nature of test only provides the relative consistency of specifying the binder as compared to penetration
binder at specific temperatures, which can be used as grading. It require bidder to be tested at 60oC and
an indicator of susceptibility of asphalt binder to 135oC, which corresponds to typical maximum
rutting or cracking. However, it has performed quite pavement temperature and temperature at the time of
satisfactorily for many decades in overcoming the mix production and placement in the field,
major asphalt pavement distresses: permanent respectively. Viscosity specification at 60oC helps in
deformation and cracking (fatigue and thermal). minimizing rutting potential, whereas, viscosity at
ASTM D946 specified the five binder grades based 135oC aids to minimize the potential for tender mixes
upon penetration at 25oC. The greater the penetration during paving operation. With all these added
the softer is the binder. In the case of viscosity benefits, it fails to characterize the binder at low
grading, viscosity at 60oC (close to maximum temperatures to minimize the potential of thermal
pavement temperature) is specified. The specifications cracking and pavement performance prediction.
also require a minimum viscosity to be measured at Figure 1 shows the criteria used for penetration and
135oC to reduce the potential of tender mix at the time viscosity grading systems. It can be seen that two
of compaction. ASTM D3381 specified six binder asphalt binders A and B meeting the penetration and
grades based upon the viscosity measured at 60oC. viscosity specifications may behave very differently at
Table 1 provides the standard penetration and other temperatures.

50
 Temperature Zoning of Pakistan for Asphalt Mix Design

120 2500

100
B
Penetration @ 25 deg. C

2000

Viscosity @ 60 deg. C, Poise

A
80
1500
60
1000
40 A
500
20
Minimum Viscosity @ 135 deg C B
0 0
0 10 20 30 40 50 60 0 20 40 60 80 100 120 140
Temperature, deg. C Temperature, deg. C

Figure 1: Graphical Comparison of two binders meeting Penetration and Viscosity Grading Specifications

The SUPERPAVE system is unique in a sense and shear modulus requirements are constant among
that asphalt binder is specified on the basis of the all grades of binders. What differentiates the various
maximum and minimum pavement temperatures in binder grades is the temperature at which the
which the binder is expected to serve. Mechanical requirements must be met. For example a binder
properties requirement remain the same, but the classified as PG 64-22, means that the binder must
temperature at which the asphalt binders achieve the meet the required properties least up to a maximum
physical properties corresponds to the pavement pavement temperature of 64oC and down to a
minimum and maximum temperature. For example, minimum pavement temperature of -22oC. Table 2
high temperature requires binder to have G*/sin to be presents the typical binder grades as is specified in the
at least 1.0 kPa for unaged condition (G* is the shear SUPERPAVE specifications. High temperature 46oC
modulus in kPa and is the phase angle). The value have corresponding low temperatures of -34, -40, -
of 1.0 kPa remains constant but the temperature at 46oC; resulting in PG 46-34, PG 46-40 and PG 64-46.
which this value has to be achieved depends upon the It is important to understand that PG grades are not
maximum pavement temperature. Another important limited to the ones given in Table 2. It is very
feature of SUPERPAVE is that mechanical properties possible to have PG 58-10, which is not included in
are measured on the asphalt binders at three Table 2.
conditions: unaged, short-term aged and long-term
aged. The short and long-term aging is simulated in Table 2: Binder Grades as Specified in
the laboratory using Rolling Thin Film Oven (RTFO) SUPERPAVE Specifications (1)
and Pressure Aging Vessel (PAV), respectively. The
High Temperature Low Temperature
required mechanical properties at the three aging
Grade, oC Grade, oC
conditions both for high and low temperatures are
specified in the SUPERPAVE specifications (MP1: PG 46 34, 40, 46
Specification for Performance-Graded Asphalt PG 52 10, 16, 22, 28, 34, 40, 46
Binder). PG 58 16, 22, 28, 34, 40
PG 64 10, 16, 22, 28, 34, 40
1.2 Temperature Grade
PG 70 10, 16, 22, 28, 34, 40
As mentioned earlier, binders are specified based
PG 76 10, 16, 22, 28, 34
on the temperature regime in which binder is expected
to serve. The physical properties, such as the stiffness PG 82 10, 16, 22, 28, 34, 40

51
Pak. J. Engg. & Appl. Sci. Vol. 8, Jan., 2011

The temperatures given in Table 2 correspond to depth of 20 mm below the pavement surface.
the pavement temperature and can be estimated from Equation (1) represents model developed under the
the air temperature data collected over the years. SHRP program, whereas Equation (2) is the LTPP
SUPERPAVE defines the high and low temperatures model.
by 7-day average maximum air and l-day minimum air
temperature. The 7-day average maximum T pav, h (Tair 0.00618 Lat 2 0.2289 Lat
temperature is defined as the average highest air 42.4) (0.9545) 17.78 z. air (1)
temperature for a period of 7 consecutive days within
a given year. The 1-day minimum temperature is T pav, h 54.32 0.78Tair 0.0025Lat 2
defined as the lowest air temperature recorded in a 2 1/ 2
15.14 log10 (d 25) z (9 0.61 air ) (2)
given year. The data are collected over multiple years
and the design high and low pavement temperature Where
values are then estimated using the average and
standard deviations of the data collected for a desired Tpav,h = High AC pavement temperature at 20
reliability level. mm from surface, °C.
Tpav,h,d = High AC pavement temperature at
1.3 Pavement Temperature depth d from surface, °C.
Tair = High 7-day mean air temperature, °C.
Several research efforts have been carried out to Lat = Latitude of the section, degrees.
relate the air temperature to pavement temperature.
d = Pavement depth in mm,
Regression equations along with mathematical heat
flow theories have been used for the correlation. air = Standard deviation of the 7-day
maximum air temperature, Deg. C
Among these, models for the prediction of high and
low pavement temperatures based upon the air z = Standard normal dist. table, z = 2.055
temperature data was established during the Strategic for 98% reliability, and z = 0.0 for 50%
Highway Research Program (SHRP). However, later reliability.
SHRP established the Long Term Pavement
Performance (LTPP) program to support a broad
Low Temperature Models
range of pavement performance analysis leading to SHRP considers the low air temperature as the
improved engineering tools to design, construct, and design low pavement temperature [5]. The low
manage pavements. In this regards, the Seasonal pavement design temperature at the pavement surface
Monitoring Program (SMP), a task of LTPP evaluated is the same as the 1-day minimum temperature, since
the effects of temperature variations on performance the air temperature is the same as the pavement
and validated the available models [3,4]. This resulted surface temperature. This can be mathematically
in a new set of pavement temperature prediction represented by the following relationship.
models for the high and low temperature grade. Given
below are the models developed under the SHRP and Tpav,l Tair 0.051x d 0.000063x d 2 z. air
(3)
LTPP program for high and low pavement
temperature predictions. The LTPP low pavement temperature at the
surface is presented in Equation (4) below.
High Temperature Models Tpav,l 1.56 0.72Tair 0.004Lat 2
The SHRP high temperature model was 2 1/ 2
6.26 log10 (d 25) z (4.4 0.52 air ) (4)
developed from the results of theoretical heat transfer
modeling [5]. Based upon the data collected from Where Tpav,l = low AC pavement temperature in
several sites throughout the U.S. regression model was Deg. C
then developed for prediction of high pavement
temperature as a function of depth. SUPERPAVE In the above equations (Equation 1 to 4), the
defines the high pavement design temperature at a factor “z and air” is included to introduce reliability

52
 Temperature Zoning of Pakistan for Asphalt Mix Design

in the selection of binder grade. For example a high 2. Pakistan Metrological Department
pavement temperature grade of 64 at 98% reliability 3. Weather Underground Web Site
level means that 98% of times the pavement 4. Surface Water Hydrology Project, WAPDA
temperature will not exceed 64oC. Pakistan

2. Temperature Data Base Weather information for 21 stations for the years
1986-2006 was obtained from the work done by Qazi
In order to establish the Performance Grade (PG) [6,7]. The database for the 21 stations was expanded
for temperature conditions in Pakistan, temperature for additional three years (2007 to 2009) using the
zoning for high and low pavement temperatures was “Weather Underground Web Site”. The “Weather
carried as a first step. Using the above models, Underground” is a commercial weather service that
pavement temperature was estimated by making use provides weather information including temperature
of the air temperature data collected from the weather data via the Internet. Weather Underground website
stations. In this regards, air temperature data from 64 has information on major cities around the world on
weather stations across Pakistan was accumulated in a its web site [8]. Additional 43 weather stations were
weather database created in Microsoft Excel added using the information obtained from Surface
Spreadsheet. These weather station locations covered
Water Hydrology Project, WAPDA Pakistan [9]
almost all the geographical areas of the country. The
resulting in a total of 64 weather stations.
air temperature data was collected from four main
sources that included [6,7,8,9]: Table 3 summaries the weather stations used for
1. Master’s Thesis of Aurangzeb Qazi (National the development of pavement temperature zoning.
University of Science and Technology, 2008) Table has information on the weather station, latitude,

Table 3: Summary of Weather Stations

Data Availability Data Availability
Latitude Latitude
No. Station (years) No. Station (years)
(degrees) (degrees)
(Low/High)1 (Low/High)1
1 Bagh 33.83 1986/1998-2007 33 Massan 33.00 1996-2007
2 Bahawal pur 29.03 2004-2009 34 Miani Forest 25.48 1996-2007
3 Bannu 33.00 1996-2007 35 Multan 30.20 1986/1996-2009
4 Besham Qila 34.93 1996-2007 36 Munda Dam 25.56 2000-2007
5 Chillya 24.83 1996-2007 37 Murree 33.92 1997/1996-2009
6 Chitral 35.85 1986/1996-2009 38 Nabisir 25.52 1996-2007
7 Daggar 34.50 1996-2007 39 Naran 34.90 1996-2007
8 Dainyor 35.92 1996-2007 40 Nokkundi 28.82 1986/1996-2009
9 Dalbandin 28.88 1997/1996-2009 41 Oghi 34.50 1996-2007
10 Domel 34.38 2005-2007 42 Panjgur 26.97 1997/1996-2009
11 Doyian 35.55 1996-2007 43 Parachinar 33.87 1986/1996-2009
12 Faisalabad 31.41 2004-2009 44 Pasni 25.25 2004-2009
13 Fort Lock 33.55 1996-2007 45 Peshawar 34.02 1986/1996-2009
14 Gilgit 35.92 1986/1996-2005 46 Phulra 34.33 1996-2007
15 Gujar Khan 33.25 1996-2007 47 Plandri 33.72 1996-2007
16 Gungi 34.34 2002-2007 48 Quetta 30.25 1986/1996-2005
17 Hub Dam 25.25 1996-2007 49 Rehman Bridge 33.48 1996-2007
18 Hyderabad 25.38 1986/1996-2005 50 Rohri 27.70 1997/1996-2009
19 Islamabad 33.62 1986/1996-2009 51 Sakrand 26.12 1996-2007
20 Jacobabad 28.28 2004-2009 52 Sargodha 32.08 2004-2009
21 Kachura 35.45 1996-2007 53 Sehwan 26.42 1996-2007
22 Kakul 34.18 1986/1996-2009 54 Shinkiari 34.47 1996-2007
23 Kala Bagh 32.95 1996-2007 55 Sialkot 32.52 2004-2009
24 Kalam 35.53 1996-2007 56 Sibbi 29.55 1986/1996-2005
25 Kallar 33.42 1996-2007 57 Skardu 35.30 1986/1996-2005
26 Kandia 33.87 2005-2007 58 Sukkur 27.69 2005-2009
27 Karachi 24.90 1986/1996-2009 59 Tank 32.22 2003-2006
28 Karim Abad 36.30 1996-2007 60 Tarbela 34.07 2002-2007
29 Khuzdar 27.83 1997/1996-2009 61 Thana Bula 25.37 1996-2007
30 Lahore 31.55 1986/1996-2008 62 Yugo 35.18 1996-2007
31 Mangla 33.13 1996-2007 63 Zhob 31.35 1986/1996-2009
32 Mardan 34.20 1996-2007 64 Zulam Br 26.97 2000-2006
Note: 1 1986/1996-2009: low temperature data from 1986 to 2009 and high temperature data from 1996 to 2009.

53
Pak. J. Engg. & Appl. Sci. Vol. 8, Jan., 2011

Table 4: High and Low Air and Pavement Temperatures with PG Grading
PG Grade @ 50% PG Grade @ 98%
High Air Temp. oC Low Air Temp. oC SHRP Models LTPP Models
Reliability Reliability
No. Location High Low High Low
Avg. Std. Dev Avg. Std. Dev Pavement Pavement Pavement Pavement SHRP Model LTPP Model SHRP Model LTPP Model
o o o
Temp. C Temp. C Temp. C Temp. oC
1 Bagh 35.0 3.1 -1.8 3.6 63.1 -9.2 61.6 -5.5 PG 58-10 PG 54-10 PG 64-10 PG 64-10
2 Bahawalpur 43.2 1.4 3.5 1.9 68.1 -0.4 67.4 1.2 PG 70-10 PG 64-10 PG 70-10 PG 70-10
3 Bannu 43.1 1.0 -1.7 2.0 66.6 -5.9 66.6 -3.7 PG 70-10 PG 64-10 PG 70-10 PG 70-10
4 Besham Qila 42.5 1.0 2.3 1.5 65.7 -0.7 65.7 -0.8 PG 64-10 PG 64-10 PG 70-10 PG 70-10
5 Chillya 39.7 1.5 8.1 2.9 65.5 2.2 65.3 4.5 PG 64-10 PG 64-10 PG 70-10 PG 70-10
6 Chitral 39.2 1.6 -5.4 2.0 63.6 -9.5 63.3 -7.0 PG 64-10 PG 58-10 PG 64-10 PG 64-10
7 Daggar 42.2 1.4 -3.8 1.7 66.5 -7.2 65.8 -5.3 PG 64-10 PG 64-10 PG 70-10 PG 70-10
8 Dainyor 41.1 1.6 -5.6 0.8 65.4 -7.2 64.8 -6.5 PG 64-10 PG 64-10 PG 70-10 PG 70-10
9 Dalbandin 46.2 1.0 -5.1 2.4 70.4 -10.1 69.7 -5.4 PG 70-10 PG 64-10 PG 76-10 PG 70-10
10 Domel 39.1 2.8 0.0 0.6 66.3 -1.2 64.5 -2.0 PG 64-10 PG 58-10 PG 70-10 PG 70-10
11 Doyian 40.1 1.4 -1.4 1.1 64.2 -3.6 64.0 -3.5 PG 64-10 PG 58-10 PG 70-10 PG 64-10
12 Faisalabad 41.5 2.3 2.8 1.3 68.2 0.2 66.5 0.5 PG 64-10 PG 64-10 PG 70-10 PG 70-10
13 Fort Lock 29.4 1.7 -4.8 1.5 54.8 -8.0 56.1 -5.7 PG 52-10 PG 52-10 PG 58-10 PG 58-10
14 Gilgit 41.2 1.5 -7.7 1.2 65.4 -10.2 64.8 -8.2 PG 64-10 PG 64-10 PG 70-10 PG 70-10
15 Gujar Khan 41.5 2.1 -1.8 1.2 67.3 -4.3 65.9 -3.2 PG 64-10 PG 64-10 PG 70-10 PG 70-10
16 Gungi 35.3 1.6 -3.3 2.3 60.1 -7.9 60.5 -5.3 PG 58-10 PG 58-10 PG 64-10 PG 64-10
17 HubDam 40.5 2.0 8.7 1.6 67.2 5.5 66.2 6.0 PG 64-10 PG 64-10 PG 70-10 PG 70-10
18 Hyderabad 41.4 1.2 6.5 1.4 66.5 3.5 66.5 4.5 PG 64-10 PG 64-10 PG 70-10 PG 70-10
19 Islamabad 41.0 2.5 -0.2 1.0 67.7 -2.2 65.8 -2.0 PG 64-10 PG 64-10 PG 70-10 PG 70-10
20 Jacobabad 46.2 2.3 5.3 3.2 72.8 -1.3 70.4 1.4 PG 70-10 PG 64-10 PG 76-10 PG 76-10
21 Kachur 35.4 1.9 -10.2 2.1 60.7 -14.5 60.7 -10.5 PG 58-10 PG 58-10 PG 70-16 PG 64-10
22 Kakul 35.4 2.5 -2.8 2.0 62.2 -6.9 61.3 -4.7 PG 58-10 PG 58-10 PG 64-10 PG 64-10
23 Kalabagh 44.7 1.0 0.0 0.9 68.4 -1.8 67.9 -1.7 PG 70-10 PG 64-10 PG 70-10 PG 70-10
24 Kalam 30.4 2.3 -11.7 1.8 56.8 -15.3 57.0 -11.3 PG 54-16 PG 54-10 PG 58-16 PG 58-16
25 Kallar 40.9 1.0 2.9 1.2 64.4 0.4 64.7 0.2 PG 64-10 PG 58-10 PG 70-10 PG 64-10
26 Kandia 39.1 3.2 -3.2 3.1 67.2 -9.6 64.9 -6.0 PG 64-10 PG 58-10 PG 70-10 PG 70-10
27 Karachi 37.2 1.2 5.5 3.1 62.5 -0.9 63.2 2.3 PG 64-10 PG 58-10 PG 64-10 PG 64-10
28 Karimabad 34.7 1.7 -6.0 1.5 59.5 -9.2 59.8 -7.3 PG 58-10 PG 58-10 PG 64-10 PG 64-10
29 Khuzdar 40.3 1.2 -1.0 2.4 65.0 -5.8 65.2 -2.1 PG 64-10 PG 64-10 PG 70-10 PG 70-10
30 Lahore 41.4 2.0 3.2 2.4 67.3 -1.8 66.1 -0.1 PG 64-10 PG 64-10 PG 70-10 PG 70-10
31 Mangla 42.5 1.3 1.7 0.9 66.8 -0.2 66.2 -0.5 PG 64-10 PG 64-10 PG 70-10 PG 70-10
32 Mardan 42.9 0.7 -2.2 1.0 65.7 -4.3 66.1 -3.6 PG 70-10 PG 64-10 PG 70-10 PG 70-10
33 Massan 45.0 1.6 2.1 1.6 69.8 -1.2 68.3 -0.6 PG 70-10 PG 64-10 PG 70-10 PG 70-10
34 Miani Forest 44.1 3.4 5.5 3.9 73.5 -2.5 70.3 1.4 PG 70-10 PG 64-10 PG 76-10 PG 76-10
35 Multan 43.8 1.5 2.0 1.1 68.7 -0.3 67.8 0.4 PG 70-10 PG 64-10 PG 70-10 PG 70-10
36 Munda Dam 42.4 1.1 4.2 0.7 67.2 2.8 67.2 3.2 PG 70-10 PG 64-10 PG 70-10 PG 70-10
37 Murree 27.9 2.2 -6.8 3.0 54.4 -12.9 55.2 -8.5 PG 52-10 PG 52-10 PG 58-16 PG 58-10
38 Nabi Sar 43.4 1.7 4.4 1.1 69.3 2.2 68.2 3.2 PG 70-10 PG 64-10 PG 70-10 PG 70-10
39 Naran 28.2 3.4 -14.6 1.5 57.1 -17.7 56.5 -13.0 PG 52-16 PG 52-10 PG 58-22 PG 58-16
40 Nokkundi 46.0 1.1 -3.1 1.6 70.4 -6.3 69.5 -3.3 PG 70-10 PG 64-10 PG 76-10 PG 70-10
41 Oghi 36.3 2.5 -1.5 1.1 63.0 -3.7 62.0 -3.2 PG 58-10 PG 58-10 PG 64-10 PG 64-10
42 Panjgur 42.3 1.0 -2.2 2.1 66.7 -6.5 66.8 -2.6 PG 70-10 PG 64-10 PG 70-10 PG 70-10
43 Parahinar 33.8 1.5 -9.7 3.3 58.6 -16.5 59.4 -10.9 PG 58-10 PG 58-10 PG 58-16 PG 64-16
44 Pasni 37.1 1.8 9.0 2.7 63.6 3.4 63.5 5.2 PG 64-10 PG 58-10 PG 64-10 PG 64-10
45 Peshawar 42.3 2.1 1.0 1.4 68.0 -1.9 66.4 -1.5 PG 64-10 PG 64-10 PG 70-10 PG 70-10
46 Phulra 41.2 1.4 -3.4 1.0 65.5 -5.5 65.0 -4.6 PG 64-10 PG 64-10 PG 70-10 PG 70-10
47 Plandri 35.4 1.2 -0.8 0.8 59.6 -2.3 60.5 -2.4 PG 58-10 PG 58-10 PG 64-10 PG 64-10
48 Quetta 39.5 1.2 -9.3 1.8 64.1 -12.9 64.3 -8.2 PG 64-10 PG 58-10 PG 70-16 PG 70-10
49 Rehmab Br. 41.0 1.2 1.3 1.2 64.9 -1.1 64.9 -1.1 PG 64-10 PG 58-10 PG 70-10 PG 70-10
50 Rohri 45.4 1.5 4.1 1.4 70.5 1.2 69.3 2.3 PG 70-10 PG 64-10 PG 76-10 PG 70-10
51 Sakrand 44.4 3.0 7.0 3.1 73.0 0.7 70.1 3.2 PG 70-10 PG 64-10 PG 76-10 PG 76-10
52 Sargodha 42.5 3.1 4.0 3.5 70.5 -3.1 67.8 -0.7 PG 70-10 PG 64-10 PG 76-10 PG 70-10
53 Sehwan 48.2 1.0 5.8 1.4 72.3 2.9 71.5 3.8 PG 76-10 PG 70-10 PG 76-10 PG 76-10
54 Shinkiari 37.9 1.1 -1.0 0.7 61.6 -2.5 62.3 -2.7 PG 64-10 PG 58-10 PG 64-10 PG 64-10
55 Sialkot 41.4 2.6 1.8 1.0 68.3 -0.2 66.3 -0.3 PG 64-10 PG 64-10 PG 70-10 PG 70-10
56 Sibbi 47.6 1.9 2.3 1.3 73.3 -0.4 71.1 0.6 PG 70-10 PG 70-10 PG 76-10 PG 76-10
57 Skardu 36.5 2.5 -14.3 4.5 63.0 -23.5 62.0 -16.0 PG 58-16 PG 58-10 PG 64-28 PG 64-16
58 Sukkar 41.6 2.6 2.0 2.6 69.3 -3.4 67.3 -0.3 PG 64-10 PG 64-10 PG 70-10 PG 70-10
59 Tank 47.0 2.7 1.7 2.9 73.9 -4.3 70.9 -1.8 PG 70-10 PG 64-10 PG 76-10 PG 76-10
60 Tarbela 44.3 1.3 3.2 1.3 68.1 0.6 67.4 0.1 PG 70-10 PG 64-10 PG 70-10 PG 70-10
61 Thana Bulla 44.5 0.9 3.5 1.7 68.8 -0.1 68.7 2.1 PG 70-10 PG 64-10 PG 70-10 PG 70-10
62 Yugo 33.4 1.8 -9.9 1.2 58.6 -12.3 59.1 -9.6 PG 58-10 PG 58-10 PG 64-16 PG 64-10
63 Zhob 39.6 1.0 -1.0 2.4 63.7 -5.8 64.1 -3.0 PG 64-10 PG 58-10 PG 64-10 PG 70-10
64 Zulam Br. 39.6 1.7 -1.4 3.0 65.7 -7.6 65.1 -2.9 PG 64-10 PG 64-10 PG 70-10 PG 70-10

54
 Temperature Zoning of Pakistan for Asphalt Mix Design

and number of years of data used for the analysis. No significant difference was observed for high
Data collected for each station was used to estimate pavement temperature predictions using the SHRP and
the yearly high and low air temperatures values based LTPP approach at 98% level of reliability in terms of
7-day maximum average and the minimum air the average and standard deviation values. However,
temperature. The yearly high and low temperature low pavement temperature showed more variability in
data were then used for the computation of average
prediction using the SHRP and the LTPP models. The
and standard deviation values required for computing
predicted low pavement temperature using SHRP
pavement temperatures. Pavement high and low
model is lower as compared to the LTPP model for the
temperatures were estimated using both SHRP and
LTPP models (Equations 1 to 4). Table 4 gives the same air temperature data. The SHRP models provide
summary of pavement temperature data used for additional protection against low temperature cracking
estimating the PG requirements at 50% and 98% level as compared to LTPP models.
of reliability. In addition to the PG grade information,
table also summarizes the statistical parameters, Figure 2 shows the predictive high and low
average and standard deviation for each weather pavement temperature predictions as a function of air
station. temperature for the 64 sites. Figure 3 presents the
same data in the form of a histogram. Significant
It was observed that the temperatures in Pakistan
differences were observed in pavement temperature
vary significantly from one end of the country to
predictions using both the models with the SHRP
another, with northern part much cooler compared to
southern part. The lowest air temperature observed is model predicting a much lower temperature values
-14.6oC and the high 7-day average air temperature in compared to the LTPP model. A paired student t-test
excess of 48oC. From the data given in Table 4, range was carried out to compare the results of the SHRP
of air temperature at any given location may be from and the LTPP models both for 7-days high average
28oC to 51oC. However, the corresponding range of and 1-day low pavement temperatures. The results of
pavement temperature is between 60oC to 87oC at the paired t-test are summarized in Table 6. Values of
98% level of reliability using the SHRP models. t-statistics, t-critical and the probability of null
hypothesis being true (p-value) are summarized for
3. Temperature Data Analysis four cases. In all situations, the absolute value of t-
From the air temperature data, pavement temperatures statistics is greater than t-critical, i.e. because the
are predicted using the SHRP and LTPP approaches. probability of null hypothesis being true is very small
A comparison of the predicted pavement temperatures (close to zero) suggest that significant difference
from two models is presented in Table 5. The results exists between the predictions of pavement
show the statistical comparison for the low and high temperature from the two models. Therefore, the null-
pavement temperature data at 98 percent level of hypothesis that there is no statistical difference is
reliability for all the 64 sites. rejected.

Table 5: Comparison of SHRP and LTPP Predictions

SHRP Models LTPP Models

Statistical Parameter High Low High Low
Temperature Temperature Temperature Temperature
Minimum Temp. deg. C 54.4 -23.5 55.2 -16.0
Maximum Temp. deg. C 73.9 55 71.5 6.0
Average, deg. C 65.7 -4.5 65.1 -2.5
Standard Deviation, deg. C 4.5 5.8 3.7 4.6
Coefficient of Variation (%) 6.8 -128.4 5.8 -181.0

55
Pak. J. Engg. & Appl. Sci. Vol. 8, Jan., 2011

80.0
SHRP 50%
75.0
High Pavement Temp., deg C

LTPP 50%

70.0 SHRP 98%

LTPP 98%
65.0

60.0

55.0

50.0

45.0
25.0 30.0 35.0 40.0 45.0 50.0
High Air Temp., deg C
(a)

15.0
SHRP 50%
10.0 LTPP 50%
Low Pavement Temp., deg C

5.0 SHRP 98%

LTPP 98%
0.0

-5.0

-10.0

-15.0

-20.0

-25.0
-15.0 -10.0 -5.0 0.0 5.0 10.0
(b) Low Air Temp., deg C

Figure 2 High and Low Pavement Temperatures as a Function of Air Temperature

56
 Temperature Zoning of Pakistan for Asphalt Mix Design

40
SHRP 50%
35
LTPP 50%
No. of Weather Stations

30
SHRP 98%
25
LTPP 98%
20

15

10

0
52.0 58.0 64.0 70.0 76.0 82.0
(a) High Termperature PG, deg C

40
SHRP 50%
35
LTPP 50%
No. of Weather Stations

30
SHRP 98%
25 LTPP 98%

20

15

10

0
-28 -22 -16 -10 -4 2 8 14
(b) Low Termperature PG, deg C

Figure 3 Histogram of High/Low Pavement Temperatures for 64 Stations

57
Pak. J. Engg. & Appl. Sci. Vol. 8, Jan., 2011

Table 6: Paired t-Test Comparison of SHRP and LTPP Models

Average Pavement
Pavement Temp. Reliability Temperature, deg. C t Critical P(T<=t)
t Stat
Condition Level (%) two-tail two-tail
SHRP LTPP
50 61.90 58.20 37.55 2.00 0.00
High
98 65.55 69.72 35.46 2.00 0.00
50 -0.76 2.64 23.81 2.00 0.00
Low
98 -4.60 -2.60 8.99 2.00 0.00

reliability. It can be observed that most part of

4. Temperature Zoning And Pg System
Pakistan falls in PG 70 - 10 category with PG 64-10 in
the second place. PG 58-22 is only required for a
Based on the information collected from 64
limited area in northern part of Pakistan.
weather stations, 7-days average high and low
pavement temperatures were estimated for 50% and CHINA
98% level of reliability using SHRP and LTPP models
(Equations 1 to 4). Table 4 presents PG grading for
each weather station for SHRP and LTPP models for
50% and 98% reliability levels.
AFGHANISTAN Disputed
As mentioned earlier, significant differences in Territory
the prediction of pavement temperatures were
observed using both the models with SHRP models
being more severe compared to LTPP models. When
considering the conditions in Pakistan, most parts of
Pakistan experiences moderate to high temperatures. INDIA
The most common distress is rutting as a result of high
IRAN
temperature and uncontrolled traffic loads. Low
temperature cracking is not common because
temperatures below -10oC are not very common and
may be observed only on a limited scale. Because of ARABIAN
the above mentioned facts of high temperatures and SEA
uncontrolled traffic loading, it was decided to use
SHRP models for the purpose of temperature zoning. Figure 4: Temperature Zoning for PG System in
SHRP models in general require higher Performance Pakistan [10]
Grade (PG) and thus provide extra protection against
high temperatures. Since most the traffic on the The Attock and Karachi are the two major
national highways have uncontrolled axle loading, it is refineries in Pakistan each refinery producing three
important that the final required grades be based on types of binder grades which are specified based on
98% level of reliability. the penetration grading system. As a part of Pak-US
Cooperative Research Program (10), available binders
Figure 4 shows the PG temperature zoning across were graded based on the penetration, viscosity and
Pakistan based on SHRP models for 98% level of PG system and are summarized in Table 7.

58
 Temperature Zoning of Pakistan for Asphalt Mix Design

As shown in Figure 4, the entire area is divided 10. That is PG 76-16 satisfies the requirement of PG
into six grades depending upon the temperature 70-16, but may result in harder binder which may not
zoning. A summary of these grades is provided in be desirable and prone to cracking. In addition, PG
Table 8 as required PG grade (column 1). The second 76-16 (A-PMB) is polymer modified asphalt, which is
column has the grades that are available, whereas third relatively more expensive compared to neat
column list the combination of grades that fulfill (unmodified) asphalt.
requirement specified in column. That is PG 64-16, PG 70-10 which is the requirement for more than
satisfies the requirements of PG 64-10 according to 70 percent of Pakistan is not produced by any of the
AASHTO MP1. Under the remarks column, all two refineries. This is considered to be one of the
binders are considered acceptable except for PG 70- most important grade requirements.

Table 7: Summary of Binder Grading for Asphalts in Pakistan

Penetration Viscosity PG Grades of Available

Sr. No. Refinery Binder
Grade Grade Asphalts in Pakistan
1 Attock A-PMB Not Available AC-40 PG 76-16
2 Attock A-60/70 Pen 60/70 AC-20 PG 58-22
3 Attock A-80/100 Pen 60/70 AC-10 PG 58-22
4 Karachi K-40/50 Pen 40/50 AC-40 PG 64-16
5 Karachi K-60/70 Pen 60/70 AC-20 PG 64-22
6 Karachi K-80/100 Pen 85/100 AC-10 PG 58-22
Note: “A” for Attock Refinery and “K” for Karachi Refinery

Table 8: Summary of Binder Grading for Asphalts in Pakistan

Available Possible Equivalent

Required PG
Exact Grade Penetration Remarks
Grade
Grade Availability Grade
PMB, Polymer Modified may
PG 70-10 No PG 76-16 A–PMB be too hard for PG 70-10 and
may result in cracking
PG 64-10 No PG 64-16 K–40/50 Acceptable
PG 64-16 K–40/50
PG 64-16 Yes Acceptable
PG 64-22 K–60/70
K–80/100
PG 58-16 No PG 58-22 A–80/100 Acceptable
A–60/70
PG 76-10 No PG 76-16 A–PMB Acceptable
K–80/100
PG 58-22 Yes PG 58-22 A–80/100 Acceptable
A–60/70

59
Pak. J. Engg. & Appl. Sci. Vol. 8, Jan., 2011

5. Observations And Conclusions 7. References

[1] SUPERPAVETM Asphalt Mixture Design &
The main objective of the paper is to divide
Analysis, National Asphalt Training Center,
Pakistan into temperature zones according to the PG Demonstration Project 101, Federal Highway
requirements developed under SHRP. Several Administration, Office of Technology
observation and conclusions are briefly presented Applications, Washington DC and Asphalt
below: Institute, Lexington, KY. September 1994.
The two model forms, SHRP and LTPP for the [2] Welborn, J. Y. and W. J. Halstead (1974).
Testing of Asphalts and Asphalt Mixtures,
prediction of pavement temperatures resulted into
Proceeding Advanced Asphalt Pavement
significantly different predictions both for high and Technologists (AAPT), Vol. 43A.
low temperatures. However, relatively greater
[3] Diefenderfer, B. K., Al-Qadi, I. L., Reubush, S.
differences were observed for low temperature
D., and Freeman, (2002). T. E., Development and
compared to high temperature predictions. Validation of a Model to Predict Pavement
Temperature Profile, Presented at Transportation
SHRP prediction models for high and low
Research Board 82nd Annual Meeting,
temperatures were selected for the development of PG Washington, DC.
zones for Pakistan, since SHRP requires higher grade
[4] Mohseni, A. (1998). LTPP Seasonal Asphalt
requirement for the high temperature grade, which is
Concrete (AC) Pavement Temperature Models,
critical for uncontrolled axle loading.
Report No. FHWA-RD-97-103, Federal
Highway Administration, US Department of
The use of 98% level of reliability provides
Transportation.
additional safety margin against high traffic levels and
uncontrolled loadings. No additional bumping of [5] Huber, G.A., (1994). “Weather Database for the
grade is needed as is recommended by AASHTO MP1 SUPERPAVE Mix Design System”, SHRP-A-
648A.
specifications, since it will result in excessively stiff
binder. [6] Aurangzeb, Q., (2008). School of Civil and
Environmental Engineering (NIT), National
Pakistan is divided into six temperatures zones University of Science and Technology (NUST),
requiring PG 70-10 as the most important binder that Master Thesis.
covers more than 70 percent of the area. [7] Pakistan Meteorological Department (2005).
"Climatic normals of Pakistan (1971-2000)."
PG 70-10 is not produced by any of the two Climatic Data Processing Centre, Karachi,
refineries in Pakistan. The closest grade that fulfills Pakistan.
the requirements of PG 70-10 is PG 76-16 (A-PMB). [8] Website: http://en.wikipedia.org/wiki/Weather
A-PMB is relatively harder and may prone to Underground_(weather_service).
cracking. Use of PG 64-16 is relatively softer than the [9] River and Climatological Data of Pakistan,
required grade and may prone to rutting. Precipitation Evaporation, Temperatures,
Relative Humidity, Solar Radiation and Wind
At present, commonly used grade in Pakistan is Movement Data, Surface Water Hydrology
A-60/70 and K-60/70. The corresponding grades are Project, Hydrology and Water Management
PG 58-22 and PG 64-22 which is likely to rut in areas Organization, WAPDA, Lahore, Pakistan (1996–
requiring PG 70-10. This may be one of the major 2007).
reasons of premature failures especially rutting in [10] Baladi,G.Y., Haider,W., and Mirza, M.W.,
most of our pavements. “Implementation of Superpave Binder & Asphalt
Mix Specifications to Improve Pavement
6. Acknowledgement Performance in Pakistan”, Annual Report –
Third Year, Pakistan-US Science Technology
The authors acknowledge the Pakistan-United Cooperative Program, The National Academies,
States Science and Technology Cooperative Program Policy and Global Affairs, Washington DC,
for funding this research. 2011.

60