Environ Monit Assess (2007) 133:417425

DOI 10.1007/s10661-006-9596-8

Sources of PM10 and PM2.5 in Cairos ambient air

M. Abu-Allaban & D. H. Lowenthal &
A. W. Gertler & M. Labib

Received: 12 September 2006 / Accepted: 30 November 2006 / Published online: 1 February 2007
# Springer Science + Business Media B.V. 2007

Abstract A source attribution study was performed

to assess the contributions of specific pollutant
source types to the observed particulate matter
(PM) levels in the greater Cairo Area using the
chemical mass balance (CMB) receptor model.
Three intensive ambient monitoring studies were
carried out during the period of February 21March
3, 1999, October 27November 27, 1999, and June
8June 26, 2002. PM10, PM2.5, and polycyclic
aromatic hydrocarbons (PAHs) were measured on a
24-h basis at six sampling stations during each of the
intensive periods. The six intensive measurement
sites represented background levels, mobile source
impacts, industrial impacts, and residential exposure.
Major contributors to PM10 included geological

M. Abu-Allaban
Department of Water Management & Environment,
Faculty of Natural Resources & Environment,
The Hashemite University,
Zarqa, Jordan
D. H. Lowenthal (*) : A. W. Gertler
Desert Research Institute,
2215 Raggio Parkway,
Reno, NV 89512, USA
e-mail: dougl@dri.edu
M. Labib
Egyptian Environmental Affairs Agency,
Misr Helwan Road Bldg. 30, Maadi,
Cairo, Egypt

material, mobile source emissions, and open burning.

PM2.5 tended to be dominated by mobile source
emissions, open burning, and secondary species. This
paper presents the results of the PM10 and PM2.5,
source contribution estimates.
Keywords Urban air pollution . Source
apportionment . Particulate matter . Ambient lead

Introduction
Cairo, Egypt suffers from high ambient concentrations of atmospheric pollutants (Nasralla 1994;
Sturchio et al. 1997), including particulate matter
(PM), carbon monoxide (CO), oxides of nitrogen
(NOx), ozone (O3), and sulfur dioxide (SO2). Nasralla
(1994) reported particulate lead concentrations ranging from 0.5 g/m3 in a residential area to 3.0 g/m3
at the city center. Sturchio et al. (1997) measured total
suspended particulate (TSP) and lead concentrations
using stable isotopic ratios ( 207 Pb/ 204 Pb and
208
Pb/204Pb) at eleven sites in Cairo. Lead and TSP
concentrations ranged from 0.08 and 25 g/m3,
respectively, at Helwan to over 3 and 1,100 g/m3,
respectively, at the city center.
Rodes et al. (1996) measured fine (PM2.5) and
coarse (PM10PM2.5) concentrations as a part of a
source apportionment study in Cairo from December
1994 through November 1995. The annual average

418

Environ Monit Assess (2007) 133:417425

PM10 (sum of PM2.5 and coarse) concentrations

exceeded the 24-h average US standard of 150 g/
m3 at all sites except Maadi and the background site.
An attempt was made to attribute the high PM levels
to specific sources using the Chemical Mass Balance
(CMB) source apportionment model (Watson et al.
1990). PM10 mass was dominated by the coarse
fraction, suggesting a strong influence of fugitive dust
sources. Emissions from mobile sources, oil combustion, and open/trash burning dominated the PM2.5
apportionments.
In order to develop and implement a pollutioncontrol strategy and to reduce the health impact of
air pollution in Cairo the Cairo Air Improvement
Project (CAIP) was established. As part of the CAIP,
source attribution studies were performed to assess
the impact of various sources (e.g., lead smelters,
motor vehicles, oil combustion, open burning,
geological material, etc.) to ambient pollutant levels.
In this paper we report the PM source attribution
results of the ambient monitoring study performed
during the periods of February 21March 3, 1999,
October 27November 27, 1999, and June 8June
26, 2002.

Experimental methods
Sampling sites
Six sites were selected from CAIP network (Gertler et
al. 1999). Sites included:
Background: Kaha, a Nile delta site with significant agricultural activity. During most of the
year, the prevailing winds come from this
direction.
Industrial/Residential: Shobra El-Khaima and ElMaasara were chosen to represent residential
areas that exhibit potential industrial activities.
The Shobra site is located in a heavily industrialized area and is downwind from numerous Pb
smelters and other industrial sources. This is one
of the most highly polluted areas in the city. The
El-Maasara site is near a number of cement
plants and other industrial sources.
Traffic: El-Qualaly Square, a site located downtown. The site is close to the road and has high
light- and heavy-duty (bus) traffic.

Residential: Helwan and El-Zamalek were chosen. Helwan is impacted by emissions from
nearby cement plants and has higher PM levels
than some of the other residential areas. ElZamalek is located on one of the islands in the
Nile and represents a residential area with limited
nearby sources.
Ambient measurements
Ambient PM2.5 and PM10 samples were collected at
the six sites using the sampling protocol described by
Watson et al. (1994). All samples were of 24-h
duration. During the February/March, 1999 study,
samples were collected daily, while in the October/
November, 1999 and June, 2002 studies samples were
collected every other day. Two medium-volume
samplers designed to collect samples for chemical
analyses were utilized. This type of sampler employs
a Sierra-Andersen 254 PM10 inlet or Bendix PM2.5
cyclone to determine the size fractions collected. The
ambient air is transmitted through the size-selective
inlet and into a plenum. The inlet is located at the top
of the plenum. Maintaining a constant pressure across
a valve with a differential pressure regulator controls
the flow rate in the sampler. For the size-selective
inlet to work properly, a flow rate of 113 lpm must be
maintained through the sampler. Flow rates of 20 lpm
through each Savillex filter holder were used to
collect adequate samples for gravimetric and chemical
analyses. This flow rate was drawn simultaneously
through two parallel filter packs, one with a ringed
47 mm Teflon-membrane filter (Gelman Scientific,
Ann Arbor, MI) and one with a 47 mm quartz-fiber
filter (Pallflex Corp., Putnam, CT). The remaining
73 lpm was drawn through a makeup air port. The
flow rates were each set with a calibrated rotometer
and were monitored with the same rotometer at each
sample change.
In addition to those samples recommended by
Watson et al. (1994), we measured polycyclic
aromatic hydrocarbons (PAHs). The PAHs were
critical to apportion the carbon components of the
PM based on the uniqueness of PAH compounds
associated with light-duty spark ignition (LDSI) and
heavy-duty diesel (HDD) vehicles (Fujita et al.
1998). PAH samples were collected on Teflonimpregnated glass fiber filters followed by an adsor-

Environ Monit Assess (2007) 133:417425

bent cartridge of polyurethane foam and XAD-4 resin

(TIGF/PUF/XAD-4) using the Sequential Fine Particulate/Semi-Volatile Organic Compounds Sampler
(PSVOC sampler). The PSVOC sampler inlet is
selected for PM2.5 with a Bendix 240 cyclone
operating at 113 lpm. Prior to sampling, all sampling
media were cleaned in the laboratory. The PUF plugs
and XAD-4 resins were assembled into glass cartridges (10 g of XAD between two PUF plugs),
wrapped in aluminum foil and stored in a clean
freezer prior to shipment to the field. All filter packs
and cartridges were stored cold and were shipped to
and from the field by overnight carrier in coolers with
blue ice packets.
Source measurements
Source emissions samples were collected using
methods similar to those used in the ambient sampling
program. Bulk soil and road dust samples were
collected at each of the ambient-sampling sites.
Emissions from various sources including brick
manufacturing, cast iron foundry, copper foundry,
lead smelting, refuse burning, Mazot oil combustion,
refuse burning, and restaurants were sampled. Individual motor vehicle emissions were sampled from
heavy- and light-duty diesel vehicles, spark ignition
automobiles, and motorcycles.

419

Atomic Absorption (AA) Spectrometer. Organic and

elemental carbon are measured by thermal/optical
reflectance (TOR) on 0.5 cm2 punches taken from the
remaining half of the quartz-fiber filter (Chow et al.
1993). The PAHs samples were analyzed for PAHs
following the protocol described by Fujita et al.
(1998).
Receptor modeling methods
The Chemical Mass Balance (CMB) receptor model
was used to apportion PM and its chemical constituents to their sources (Watson et al. 1990). The CMB
solution is based on effective variance weighting,
which assigns more importance to species with lower
relative uncertainties (Watson et al. 1984). The CMB
procedure requires several steps. First, the contributing sources must be identified and their chemical
profiles must be entered. Then the chemical species to
be included in the model must be selected. The next
step is the estimation of the fractions of each chemical
species contained in each source type and the
estimation of the uncertainties in both the ambient
concentrations and source contributions. The final
step is the solution of the set of chemical mass
balance equations. These procedures are described in
detail in an application and validation protocol (Pace
and Watson 1987).

Analytical methods

Results

Teflon-membrane filters were weighed on a Cahn 31

Electro-microbalance before and after sampling to
determine mass concentrations. Chemical analyses
were performed on Teflon-membrane and quartz-fiber
filters following the methodology described by
Watson and Chow (1993). Briefly, the Teflon-membrane filters are analyzed for elements (Mg, Al, Si, P,
S, Cl, K, Ca, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Br,
Rb, Sr, Ba, U and Pb) by X-ray fluorescence (XRF)
using a Kevex 700/800 analyzer. One-half of the
quartz filter is extracted with distilled-deionized
water. The extract was analyzed for chloride, nitrate,
and sulfate ions by ion chromatography (IC) using a
Dionex 4000I ion chromatograph, for ammonium by
automated colorimetry (AC) using a TRAACS 800
Technicon auto analyzer, and for sodium and potassium by a Perkin-Elmer Model 2380 Double Beam

Mass and inorganic chemical species

Upon analyzing the filters, 80 chemical species were
detected. Average PM10, PM2.5 concentrations and
major chemical species are presented in Tables 1 and
2. Other species are not shown because they were
found in small amounts, but they were used in the
CMB modeling. Most of the major species concentrations were orders of magnitude above their minimum detectable limits (MDLs). As noted above, the
effective variance solution in the CMB assigns more
importance to species with low relative uncertainties.
Mass concentrations in both size fractions were
higher at all sites (except for PM2.5 at Shobra) during
the fall 1999 sampling period. Shobra exhibited the
highest average PM10 and PM2.5 mass concentrations.
The lowest values in both size fractions were

420

Environ Monit Assess (2007) 133:417425

Table 1 Statistical summary of PM10 mass and selected chemical species for the six intensive sites (average standard deviation, g/m3)
Species

Size

Sampling period

El-Zamalek

El-Qualaly

Helwan

Kaha

El-Maasara

Shobra

Mass

PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10
PM10

Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002

127.26.5
248.513.4
99.25.0
18.31.2
51.24.0
2.10.2
4.50.3
5.50.4
4.10.2
8.70.5
17.20.9
9.80.6
9.30.6
8.60.5
2.60.2
16.11.4
63.04.3
21.21.6
16.71.9
14.21.0
7.80.9
1.80.6
5.61.9
2.50.8
5.51.8
16.05.6
9.23.0
7.01.2
13.02.8
6.21.1
2.10.1
4.40.2
2.40.1
1.60.1
1.00.1
0.20.0

219.911.1
251.613.5
136.47.1
19.61.3
132.48.6
3.20.3
5.60.4
4.60.4
5.10.3
13.00.8
18.41.8
10.80.7
7.70.5
9.60.6
2.40.2
48.54.0
73.14.9
34.32.5
20.32.4
18.21.3
17.21.9
3.00.9
4.71.6
2.81.0
9.63.1
13.54.7
10.33.6
18.73.2
13.42.9
8.71.5
4.30.2
4.20.2
3.40.2
4.80.3
1.80.1
0.50.0

88.14.6
146.39.4
141.97.4
5.10.4
69.45.3
3.70.3
3.60.3
2.90.3
6.90.4
6.10.4
13.81.5
9.80.6
1.80.1
3.10.3
1.80.2
15.01.3
39.93.0
28.62.2
6.90.9
7.80.6
7.80.9
1.50.5
3.41.1
3.21.1
4.91.6
10.43.6
12.04.2
9.81.8
15.83.2
15.82.8
1.60.1
3.60.2
3.00.2
0.20.0
0.30.0
0.20.0

93.04.8
204.711.1
100.05.3
15.11.0
43.73.5
2.10.2
4.90.3
5.50.4
3.90.2
5.90.4
13.01.1
8.20.5
9.40.6
9.20.6
1.80.2
14.61.3
55.54.0
21.41.7
7.90.9
9.60.7
5.30.6
1.60.5
5.81.9
3.41.1
4.21.4
16.85.8
11.03.7
2.50.5
6.41.3
4.10.7
1.20.1
4.00.2
2.90.2
0.10.0
0.10.0
0.00.0

186.19.4
317.417.4
175.39.1
10.70.7
70.85.9
4.90.3
4.90.3
4.00.3
7.20.4
10.40.6
20.51.4
11.60.7
1.40.1
4.00.3
2.00.2
22.42.0
68.74.8
28.12.2
7.50.9
8.30.6
5.80.7
2.80.9
6.72.2
3.31.1
9.23.0
21.27.3
13.74.6
30.15.2
41.57.4
27.84.9
3.00.2
6.60.4
3.60.2
0.70.0
0.40.0
0.10.0

265.113.6
360.319.2
153.78.0
26.41.7
118.18.3
4.60.3
5.40.4
4.90.4
4.50.2
10.60.6
21.61.9
12.40.8
7.70.5
7.90.5
2.40.2
42.23.5
86.55.8
30.22.3
10.01.2
13.41.0
8.71.0
3.31.0
7.72.5
3.71.2
13.64.4
26.39.0
15.75.3
10.41.9
21.44.4
8.81.5
6.00.3
8.30.4
4.60.2
33.71.9
12.71.0
7.20.6

Chlorine

Nitrate

Sulfate

Ammonium

Organic Carbon

Elemental Carbon

Aluminum

Silicon

Calcium

Iron

Lead

observed at Helwan, a residential location. The

background site, Kaha, generally had the second
lowest mass levels. The correlations between measured and reconstructed (sum of species) were very
high, which indicates that the data quality for the
CAIP particulate measurements was quite good. The
sum of the species consistently accounted for 7080%
of the measured mass. The difference is accounted for
by the fact that the sum of species does not contain
oxygen associated with geological species (e.g., Al,
Fe, Si) or hydrogen, oxygen, nitrogen, and sulfur
associated with organic carbon.

The PM2.5/PM10 ratio varied from 0.3 at ElMaasara to 0.8 at Shobra. El-Maasara is an industrial
location impacted by emissions from nearby cement
plants. The ratio of 0.3 at El-Maasara is consistent
with coarse particle emissions from those activities.
Shobra is a highly industrialized site with a number of
lead smelters in the vicinity. The ratio of 0.8 is likely
due to the impact of fresh combustion emissions,
although it is still unusually high. One might have
also expected very high ratios at El-Qualaly, the
mobile source site; however, the observed ratio was
0.4. This site also had high levels of crustal species in

Environ Monit Assess (2007) 133:417425

421

Table 2 Statistical summary of PM2.5 mass and selected chemical species for the six intensive sites (average standard deviation, g/m3)
Species

Size

Sampling period

El-Zamalek

El-Qualaly

Helwan

Kaha

El-Maasara

Shobra

Mass

PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5
PM2.5

Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002
Winter 1999
Fall 1999
Summer 2002

61.93.2
131.87.2
39.62.1
12.50.8
35.73.3
0.30.1
2.70.2
5.50.4
0.50.1
6.00.4
12.90.7
7.70.5
8.70.5
10.70.6
2.60.2
11.11.0
44.33.2
16.31.3
7.91.0
14.71.1
7.30.8
0.10.0
0.40.0
0.10.0
0.30.0
1.20.1
0.50.0
0.20.0
1.00.1
0.30.0
0.20.0
0.70.0
0.20.0
0.70.0
0.70.1
0.10.0

84.64.3
135.17.3
59.33.1
9.20.6
44.43.4
0.30.1
3.00.2
4.60.4
0.60.1
6.70.4
11.71.1
8.10.5
6.80.4
9.40.6
2.60.2
23.22.0
44.33.1
26.52.0
13.01.5
22.11.6
16.31.8
0.10.0
0.30.0
0.20.0
0.50.0
1.10.1
0.60.0
0.80.0
0.90.1
0.40.0
0.50.0
0.60.0
0.40.0
1.60.1
1.40.1
0.30.0

29.41.7
99.96.0
47.92.6
1.40.2
14.61.2
0.60.1
1.70.2
2.90.3
1.70.1
4.00.3
8.80.5
6.70.4
2.10.1
6.10.5
2.00.2
7.30.7
22.41.9
19.31.5
5.90.7
5.60.4
7.40.9
0.00.0
0.20.0
0.30.0
0.20.0
0.70.0
1.30.1
0.30.0
1.00.1
1.40.1
0.10.0
0.60.0
0.50.0
0.10.0
0.20.0
0.10.0

49.72.7
111.46.2
34.71.9
10.30.7
44.93.4
0.30.1
3.20.3
5.50.4
0.40.1
4.80.3
8.60.5
5.50.4
7.90.5
9.10.6
1.80.2
10.21.0
45.63.5
14.91.2
5.90.7
8.50.7
4.50.6
0.10.0
0.30.0
0.20.0
0.10.0
1.20.1
0.70.0
0.10.0
0.50.0
0.30.0
0.10.0
0.50.0
0.30.0
0.00.0
0.10.0
0.00.0

60.93.3
107.56.4
48.32.6
5.40.4
131.614.7
0.80.1
2.80.2
4.00.3
1.80.1
5.90.4
10.70.6
7.70.5
3.00.2
8.90.6
2.20.2
9.40.9
37.22.9
18.21.4
7.60.9
6.70.5
5.80.7
0.20.0
0.50.0
0.20.0
0.90.1
1.60.1
0.90.1
3.90.4
3.60.2
2.00.1
0.50.0
0.90.1
0.40.0
0.40.0
0.40.0
0.10.0

216.111.0
173.59.3
60.73.2
22.41.4
37.92.8
1.60.2
4.50.3
4.90.4
0.60.1
9.10.6
14.00.8
9.60.6
7.60.5
9.80.6
2.70.2
32.72.7
55.63.9
22.21.7
12.41.5
16.11.3
7.40.8
0.90.1
0.50.0
0.10.2
5.40.3
3.30.2
1.90.1
5.50.5
1.50.1
0.40.0
4.20.2
1.60.1
0.70.0
26.81.4
9.20.8
5.10.4

Chlorine

Nitrate

Sulfate

Ammonium

Organic Carbon

Elemental Carbon

Aluminum

Silicon

Calcium

Iron

Lead

the PM10 fraction, likely due to resuspended road

dust, leading to the reduced ratio.
Crustal components (Si, Ca, Fe, and Al) were
significant at all sites. The majority of crustal material
was in the coarse (PM10PM2.5) fraction. The highest
concentrations of PM10 crustal species, e.g., Si, were
found at Shobra and El-Maasara, probably as a result
of fugitive dust emissions from industrial operations
at these sites.
Organic carbon (OC), and elemental carbon (EC)
were major components of PM at all sites. Potential

sources include mobile emissions, open burning,

and fossil fuel combustion. The highest average
PM10 OC levels were observed at El-Qualaly and
Shobra.
Shobra exhibited the highest lead (Pb) concentrations. The average PM10 and PM2.5 ambient lead
concentrations were 734 and 527 g/m3, respectively. Note that average winter 1999 Pb concentrations at Shobra were five times higher than those in
summer 2002. Because other mass and species
concentrations were generally higher during 1999,

422

we suspect that industrial lead operations may have

been shut down or diminished at Shobra during the
summer 2002 period. Lead concentrations at Shobra
and Cairo in general are in excess of PM10 and PM2.5
mass concentrations observed in many cities in the
US. It should be noted that leaded fuel was phased out
in 1996 and so most of this Pb probably comes from
other sources.
Source profiles
Fine and PM10 particulate source profiles used in the
CMB modeling are presented in Table 3. Some
profiles were developed as part of this study (CAIP
profiles), while other profiles were compiled from
previous studies (Chow and Watson 1999; Kuykendal
1990; and Rodes et al. 1996). Tests were done to
determine which profiles best explained the ambient
data. Geological CAIP profiles were used in all cases.
These included soil (SOIL), unpaved road dust
(UPRD), and paved road dust (PVRD) collected at
the six sites. We relied on the AUTO and DIESEL
profiles determined during the Rodes et al. (1996)
study. These profiles fit the data better than did the
CAIP motor vehicle profiles, probably because the
OC/total carbon ratio in most of the CAIP motor
vehicle profiles was greater than 90% while this ratio
in the ambient fine samples averaged only 60%.
The CAIP copper foundry and lead smelter profiles
were needed to account for Zn and Pb, respectively.
Although we used the Rodes et al. (1996) cement
profile, we found that samples in both size fractions at
El-Maasara were enriched in calcium, which could
not be explained by the geological nor by cement
profiles. To account for the excess calcium in Cairo,
we used a pure CaCO3 (limestone) profile. We
assumed that this was associated with cement production and report it as such. However, it is possible
that the limestone contribution represents construction
activities or other geological source contributions.
The restaurant and gas power plant profiles were
essentially pure OC. The Mazut power plant profile
was enriched in vanadium (1.8%) and nickel (1.2%),
which distinguishes oil combustion emissions. It was
also enriched in sulfate (42%) and OC (15%), which
made it collinear with vehicle and open burning as
well as secondary sulfate profiles. The addition of
PAHs to the CMB may help resolve this profile from
other combustion profiles.

Environ Monit Assess (2007) 133:417425

Soluble potassium (K+) accounted for over 90%,

on average, of the fine potassium in the ambient
samples. This is consistent with emissions from refuse
and open burning, which are prevalent in Cairo.
While soluble K+ is enriched in open burning
emissions, it was essentially absent in all of the CAIP
refuse burning profiles. For this reason, we used a
open burning profile (VEGB1) from a US study
(Chow and Watson 1999). Despite the utility of
soluble K+ for distinguishing open burning contributions, there was still collinearity between the open
burning and mobile source (DIESEL) profiles. Secondary ammonium sulfate (AMSUL), ammonium
bisulfate (AMBSUL), sulfuric acid (H2SO4), and
ammonium chloride (AMCL) profiles were needed
to account for particulate sulfate, ammonium, and
chloride. Pure sea salt profiles, needed to account for
soluble sodium in the ambient samples, were used to
account for marine aerosols. The ammonium chloride
profile was used to account for high chlorine concentrations in excess of the marine contribution.
PM CMB results
The CMB results are presented in Table 4. The CMB
model diagnostics for PM are quite good. For all
average PM samples R2 was between 0.94 and 1.00.
R2 greater than 0.8 is considered acceptable (Pace and
Watson 1987). The calculated mass was within 10%
of the measured mass. Chi-square, another measure of
the goodness of fit, varied from 0.19 to 1.59, with the
lowest values indicating the best fit. A chi-square less
than two is considered acceptable (Pace and Watson
1987). A summary, by site, of the CMB results is
presented below:
Al El-Zamalek: The major sources of PM10 were
geological material, Mazut oil, mobile sources,
and open burning. Most of the secondary sulfate,
ammonium chloride, and combustion (motor
vehicle, open burning, and oil) were in the fine
fraction. Lead and copper smelter contributions
were small at this site.
El-Qualaly: This site was dominated by mobile
source emissions. PM10 was dominated by geological, mobile source, and open burning. About half
of the mobile and open burning emissions was in
the PM2.5 fraction. The lead and copper smelter
contributions to PM10 were significant at this site.

Environ Monit Assess (2007) 133:417425

423

Table 3 Source profiles used in the CMB source apportionment

Number

Source ID

Description

Reference

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42

BSSOIL
HESOIL
KASOIL
ELMAUPRD
QUAPVRD
SHOBUPRD
KAHAUPRD
BRICK
CASTFE
CEMENT1
CUFOUNDR
GASPP
HDDBUS
LDDMBUS
LDSI13
LDSI45
LDSILATE
MAZUTPP
MOTO
PBSMELT
REFUSEBA
REFUSED1
REFUSED2
REST
AMSUL
AMBSUL
H2SO4
AMNIT
NH4CL
MAR100
27454
18323
37705
36586
AUTO
CEMENT
DIESEL
OIL
LEAD
VEGB1
OFPP
LIME

British School Soil

Helwan soil
Kaha soil
El-Maasara unpaved road
El-Qualaly paved road
Shobra unpaved road
Kaha unpaved road
Brick foundry
Cast iron foundry
Cement plant
Copper foundry
Gas power plant
Heavy duty diesel bus
Light duty diesel microbus
Light duty spark ignition
Light duty spark ignition
Late model light duty spark ignition
Mazut oil power plant
Motorcycle
Lead smelter
Refuse burning basatin
Refuse burning desert
Refuse burning desert
Restaurant
Secondary ammonium sulfate
Secondary ammonium bisulfate
Secondary sulfuric acid
Secondary ammonium nitrate
Secondary ammonium chloride
Pure sea salt
Sea salt 0.25 nitrate replacement
Sea salt 0.5 nitrate replacement
Sea salt 0.75 nitrate replacement
Sea salt 1 nitrate replacement
Light duty spark ignition
Cement plant
Heavy duty diesel
Mazut oil
Lead smelter
Vegetative burning
Oil fired power plant
Limestone

CAIP
CAIP
CAIP
CAIP
CAIP
CAIP
CAIP
CAIP
CAIP
CAIP
CAIP
CAIP
CAIP
CAIP
CAIP
CAIP
CAIP
CAIP
CAIP
CAIP
CAIP
CAIP
CAIP
CAIP
DRI PURE COMPOUND
DRI PURE COMPOUND
DRI PURE COMPOUND
DRI PURE COMPOUND
DRI PURE COMPOUND
DRI Watson et al. (1994)
DRI Watson et al. (1994)
DRI Watson et al. (1994)
DRI Watson et al. (1994)
DRI Watson et al. (1994)
Rodes et al. (1996)
Rodes et al. (1996)
Rodes et al. (1996)
Rodes et al. (1996)
Rodes et al. (1996)
Chow and Watson (1999)
Kuykendal (1990)
PURE CACO3

Helwan: PM10 at this site was dominated by

geological material. While the marine contribution was almost entirely in the coarse (PM10
PM2.5) fraction, mobile, Mazut, and open burning
contributions at this site were almost entirely in
the fine fraction. Copper smelter contributions

were detected at low levels (12%) in both size

fractions.
Kaha: The largest contributors to PM10 were open
burning and geological dust. Mobile source and
Mazut oil contributions were small compared with
other sites. PM2.5 was dominated by open burning.

Year

Winter 1999
Winter 1999
Fall 1999
Fall 1999
Summer 2002
Summer 2002
Winter 1999
Winter 1999
Fall 1999
Fall 1999
Summer 2002
Summer 2002
Winter 1999
Winter 1999
Fall 1999
Fall 1999
Summer 2002
Summer 2002
Winter 1999
Winter 1999
Fall 1999
Fall 1999
Summer 2002
Summer 2002
Winter 1999
Winter 1999
Fall 1999
Fall 1999
Summer 2002
Summer 2002
Winter 1999
Winter 1999
Fall 1999
Fall 1999
Summer 2002
Summer 2002

Location

El-Maasra
El-Maasra
El-Maasra
El-Maasra
El-Maasra
El-Maasra
El-Qualaly
El-Qualaly
El-Qualaly
El-Qualaly
El-Qualaly
El-Qualaly
El-Zamalik
El-Zamalik
El-Zamalik
El-Zamalik
El-Zamalik
El-Zamalik
Helwan
Helwan
Helwan
Helwan
Helwan
Helwan
Kaha
Kaha
Kaha
Kaha
Kaha
Kaha
Shobra
Shobra
Shobra
Shobra
Shobra
Shobra

PM10
PM2.5
PM10
PM2.5
PM10
PM2.5
PM10
PM2.5
PM10
PM2.5
PM10
PM2.5
PM10
PM2.5
PM10
PM2.5
PM10
PM2.5
PM10
PM2.5
PM10
PM2.5
PM10
PM2.5
PM10
PM2.5
PM10
PM2.5
PM10
PM2.5
PM10
PM2.5
PM10
PM2.5
PM10
PM2.5

Size

1869
613
31717
1076
1759
483
22011
854
25213
1357
1367
593
1276
623
24913
1327
995
402
885
292
1469
1006
1427
483
935
503
20511
1116
1005
352
26514
21611
36019
1749
1548
613

18718
673
34833
1166
20217
613
19111
723
26917
1497
1378
563
1127
583
26217
1437
987
392
858
302
19015
845
1429
543
976
512
23714
1308
975
332
21910
1717
34925
20311
1437
603

656
71
14515
121
868
61
779
30
7010
31
686
10
294
10
809
51
415
20
274
10
7811
51
737
101
123
00
526
51
393
30
515
363
9010
202
525
20

10
10
00
00
00
00
71
20
30
20
10
00
20
10
10
10
00
00
00
00
00
00
00
00
00
00
00
00
00
00
535
434
173
122
112
71

10
00
10
10
00
00
30
20
30
30
00
00
10
10
20
20
00
00
10
00
20
20
00
00
00
00
00
00
00
00
101
91
71
61
30
20

11
10
41
20
11
00
51
10
41
10
20
00
30
10
51
10
20
00
20
00
31
20
31
10
10
00
11
00
10
00
111
141
142
20
91
40

20
30
10
41
10
10
30
61
20
52
10
21
30
60
20
52
10
20
10
10
10
21
00
10
20
30
10
31
10
10
40
93
40
93
30
61

123
132
113
122
82
142
334
212
436
464
242
292
162
91
244
294
193
142
133
81
133
92
223
152
133
41
143
113
91
91
214
264
235
265
152
132

444 111
192
20
1039
51
485
00
545
61
252
10
344
31
141
00
10410 21
565
00
253
31
111
10
222
41
121
00
1089
61
605
10
192
41
101
10
182
51
101
00
637
65
394
10
254
51
162
10
353
31
191
00
12310 94
828
00
343
41
121
00
405
61
51
31
14713 47
899
00
344
51
152
10

37
41
135
91
82
81
13
51
88
92
62
91
12
51
79
112
82
91
23
41
84
81
52
61
22
51
66
81
52
71
00
22
511
122
52
101

00
30
132
51
81
20
61
40
111
61
61
10
61
30
111
71
51
10
40
20
71
41
81
20
61
41
111
71
51
00
61
51
122
71
61
10

30
51
1613
171
00
00
182
131
202
191
00
00
232
181
154
201
00
00
21
30
51
101
00
00
232
151
182
141
00
00
182
161
142
202
00
00

4414
71
3623
51
2915
30
00
00
00
00
00
00
34
00
00
10
00
00
115
00
45
10
00
00
00
00
00
00
00
00
00
21
1111
00
00
00

Measured Predicted Geological Lead

Copper Steel
Heavy Motor Open
Marine Ammonium Ammonium Ammonium Cement
mass
mass
material
smelters smelters industry oil
vehicles burning
sulfate
nitrate
chloride
plants

Table 4 Summary of PM2.5 and PM10 source attribution results for the six intensive sites (averagestandard deviation, g/m3)

424
Environ Monit Assess (2007) 133:417425

Environ Monit Assess (2007) 133:417425

El-Maasara: The PM10 fraction was dominated

by geological material, cement, mobile sources,
and open burning. The fine fraction was dominated by open burning and mobile emissions.
The lead and copper smelter contributions were
found almost entirely in the fine fraction.
Shobra: The most unusual aspect of this location
is the high PM Pb level. Eighty percent of the
lead contribution was in the PM2.5 fraction. Most
of this Pb is in the form of fresh emissions from
secondary Pb smelters in the vicinity. Contributors to PM10 included geological material and
mobile source emissions. The PM2.5 apportionment shows a similar distribution of source
contributions, which is consistent with dominance of fine particles at this site.

Summary
An intensive PM10 and PM2.5 sampling program was
carried out at six sites in the greater Cairo area during a
winter period from February 18 to March 4, 1999,
during a fall period from October 29 to November 27,
1999, and during summer period from June 8 to June
26, 2002. Medium volume samplers were used to
collect PM2.5, PM10, and PAH samples for subsequent
chemical analysis and source apportionment modeling.
The CMB receptor model coupled with source
profiles measured during the CAIP and from previous
studies was used to estimate source contributions to
PM2.5 and PM10 mass. Depending on the sites, major
contributors to PM10 included geological material,
mobile source emissions, and open burning. PM2.5
tended to be dominated by mobile source emissions,
open burning, and secondary species. Aside from the
extremely high mass levels, two unusual features
emerged. First, most sites had high levels of ammonium chloride during the two 1999 sampling periods.
Second, lead concentrations were very high during
winter 1999 at Shobra.

Acknowledgments The authors wish to acknowledge the

support of USAID and the Egyptian Environmental Affairs
Agency under contract no. 263-C-00-97-00090-00 for providing the funding for this work.

425

References
Chow, J. C., & Watson, J. G. (1999). Cross border transport and
spatial variability in Mexicali and Californias Imperial
Valley. Atmospheric Environment, 34, 18331843.
Chow, J. C., Watson, J. G., Pritchett, L., Pierson, W., Frazier,
C., & Purcell, R. (1993). The DRI thermal/optical carbon
analysis system: Description, evaluation and applications
in U.S. air quality studies. Atmospheric Environment,27A,
11851202.
Fujita, E., Watson, J. G., Chow, J. C., Robinson, N. F.,
Richards, L. W., & Kumar, N. (1998). Northern front
range air quality study volume C: Source apportionment
and simulation methods and evaluation. Final Report
Prepared for Colorado State University.
Gertler, A. W., Lowenthal, D. H., Howes, J. E., Jr., Sagebiel, J.
C., Labib, M., Abu-Allaban, M., et al. (1999, June). Cairo
source attribution study. (Paper presented at the AWMA
92nd National Meeting, St. Louis).
Kuykendal, W. B. (1990). Air Emissions Special Manual.
Volume II: Particulate Matter Species Profiles, U.S.
Environmental Protection Agency (EPA), Research Triangle Park, N.C., EPA-450/2-90-0001b.
Nasralla, M. M. (1994, October). Air pollution in greater Cairo.
(Proceeding of the ItalianEgyptian study-days on the
environment).
Pace, T. G. & Watson, J. G. (1987). Protocol for Applying and
Validating the CMB Model. EPA 450/4-87-010, U.S.
Environmental Protection Agency, Research Triangle
Park, NC.
Rodes, C. E., Nasralla, M. M., & Lawless, P.A. (1996). An
assessment and source apportionment of airborne particulate matter in Cairo, Egypt. Activity report no. 22,
prepared for the USAID Mission to Egypt under EHP
activity no. 133-RCm delivery order no. 7.
Sturchio, N., Sultan, M., Sharkaway, M. E., Maghraby, A. E.,
& Taher, A. (1997). Concentration and isotopic composition of lead in urban particulate air, Cairo, Egypt, 1966.
Argonne National laboratory, Argonne, IL, and Center for
Environmental Hazard Mitigation, Cairo University, Cairo,
Egypt.
Watson, J. G., & Chow, J. C. (1993). Ambient air sampling. In
K. Willeke, & P. Baron (Eds.), Aerosol measurement:
Principles, techniques and applications. New York: Van
Nostrand Reinhold, pp. 622639.
Watson, J. G., Chow, J. C., Lu, Z., Fujita, E. M., Lowenthal, D.
H., Lawson, D. R., et al. (1994). Chemical mass balance
source apportionment of PM10 during the Southern
California Air Quality Study. Aerosol Science and Technology, 21, 136.
Watson, J. G., Cooper, J. A., & Huntzicker, J. J. (1984). The
effective variance weighting for least squares calculations
applied to the mass balance receptor model. Atmospheric
Environment,18, 13471355.
Watson, J. G., Robinson, N. F., Chow, J. C., Henry, R. C., Kim,
B. M., Pace, T. G., et al. (1990). The USEPA/DRI
chemical mass balance receptor model, CMB 7.0. Environmental Software, 5, 3849.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.