




































In ternationa l
Scholars
Journa ls

 

African Journal of Environmental Economics and Management ISSN 2375-0707 Vol. 8 (4), pp. 001-011, April, 
2020. Available online at www.internationalscholarsjournals.org © International Scholars Journals 

 

Author(s) retain the copyright of this article. 
 
 

 

Full Length Research Paper 

  

Assessment of heavy metal pollution in soils 
along major roadside areas in Botswana 

 
K. B. Mmolawa1*, A. S. Likuku2 and G. K. Gaboutloeloe1

 
 

1
Department of Agricultural Engineering and Land Planning, Botswana College of Agriculture, Private Bag 0027, 

Gaborone, Botswana. 
2
Department of Basic Sciences, Botswana College of Agriculture, Private Bag 0027, Gaborone, Botswana. 

 
Accepted 21 January, 2020 

 
Assessment of heavy metal pollutants: Al, Co, Cu, Fe, Pb, Mn, Ni and Zn was conducted along major roadside soils of 
Botswana, lying between latitudes 18°S to 27°S and longitudes 20°E to 29°E using enrichment factor ratios (EF), 
contamination factor (CF), pollution load index (PLI) and geoaccumulation index (Igeo) methods. The studied sites were 
demarcated into five zones referred to as FN (Francistown-Nata), NM (Nata-Maun), MG (Maun-Ghanzi), GK (Ghanzi-
Kang) and TS (Tshabong-Sekoma). All the four pollution assessment methods revealed that zones FN, NM and MG are 
pollution impacted as compared to GK and TS zones. Results of multivariate analysis suggest mixed origins of 
pollution sources including human activities, vehicular emissions and lithogenic occurrences. Al, Cu, Fe, Mn, Zn and 
Co is of mixed origins of pollutants, with Fe and Mn being predominantly lithogenic, and vehicular emissions 
characterised by Pb and Ni. The findings in this study will serve to create awareness of vehicular heavy metal 
pollution to Botswana policy makers in the mitigation of vehicular pollution, as it is barely monitored. 

 
Key words: Heavy metal contamination, roadside soils, enrichment factors, contamination factor, pollution load 
index, geoaccumulation index, cluster analysis, factor analysis. 

 
INTRODUCTION 

 
Pollution of the natural environment by heavy metals is a 
universal problem because these metals are 
indestructible and most of them have toxic effects on 
living organisms, when permissible concentration levels 
are exceeded. Heavy metals frequently reported in 
literature with regards to potential hazards and 
occurrences in contaminated soils are Cd, Cr, Pb, Zn, Fe 
and Cu (Akoto et al., 2008; Alloway, 1995). Vehicle 
exhausts, as well as several industrial activities emit 
these heavy metals so that soils, plants and even 
residents along roads with heavy traffic loads are 
subjected to increasing levels of contamination with 
heavy metals (Ghrefat and Yusuf, 2006).  

Road construction has been the main activity for 
development of industrial units. This has led to the loss of 
forest cover and subsequent loss of soil fertility. Roadside  
 
 

 
*Corresponding author. E-mail: kmmolawa@temo.bca.bw. Tel: 
+267 3650195. Fax: 267 398753. 

 
 
 

 
soils often show a high degree of contamination that can 
be attributed to motor vehicles. Various researchers have 
found that the concentrations of the metals Pb, Cu, Zn, 
Cd and Ni decrease rapidly within 10 to 50 m from the 
roadsides (Joshi et al., 2010; Pagotto et al., 2001). 
According to Panek and Zawodny (1993), pollution of 
roadside soils and plants by combustion of leaded petrol 
products is localized and usually limited to a belt of 
several metres wide on either side of the road, and that 
for similar topography and vegetation, the level of 
pollution decreases with the distance from the road. Due 
to their cation exchange capacity, complexing organic 
substances, oxides and carbonates have high retention 
capacity for heavy metals. Hence contamination levels 
increase continuously as long as the nearby sources 
remain active. Nevertheless, some heavy metals 
attached to the soil particles can be removed from the soil 
surfaces and get translocated elsewhere by the action of 
water and wind (Harrison et al., 1981; Ndiokwere, 1984; 
Ghrefat and Yusuf, 2006).  

Mmolawa et al. (2010), demonstrated that heavy metal 



   
 
 
 
 
 

 

(a) (b)  
 

NM   
 

FN 
Background 

10 km 
 

MG  
 

 sample  
 

 0.5 km 
10 km  

  
 

GK   
 

Sampled sites   
 

along roads Direction of  
 

TS road travel  
  

 
 

 
Figure 1. (a) Map of Botswana locating the sampled sites (indicated by dots along major roads). Sites were zoned as follows; Francistown-
Nata (FN); Nata-Maun (NM); Maun-Ghanzi (MG); Ghanzi-Kang (GK) and Tshabong-Sekoma (TS), and (b) schematic drawing of field 
procedure showing the sampling sites (empty circles) relative to the background sampled site (closed circle). 

 

 

contamination by Al, Co, Cu, Fe, Mn, Ni, Pb and Zn was 
variable along major Botswana roadside soils with Pb 
being extremely enriched in the soils, mainly due to 
vehicular emissions. However, the authors in this study 
used world background reference values to determine 
enrichment factors due to unavailability of local 
background. Due to spatial variability in lithology and 
mineralogy, world reference has been known to be 
erractic when used to determine enrichment factors 
(Abrahim and Parker, 2008). The present study assessed 
heavy metal pollution in soils using locally determined 
background values for metal concentrations, employing 
in-depth heavy metal analysis using four different 
approaches.  

The objectives of the present work were to: (1) Assess 
heavy metal contamination by Al, Co, Cu, Fe, Mn, Ni, Pb 
and Zn using background soils obtained some 0.5-1 km 
away from sampling sites; (2) Assess roadside soil 
contamination using four approaches, namely; (a) 
Enrichment factor (EF), (b) Contamination factor (CF), (c) 
Pollution load index (PLI), and (d) Geoaccumulation index 

(Igeo), and (3) Classify heavy metals by their similarities 
and establish their probable sources using both cluster 
and factor analysis, respectively. 
 

 
MATERIALS AND METHODS 
 
Study area 
 
The study was conducted along major roadside areas of  Botswana 

 
 

 
lying in latitudes 18 to 27°S and longitudes 20 to 29°E. The country 
has a semi-arid climate, with highly variable rainfall, both spatially 
and temporally. On annual averages, rainfall ranges from 250 mm 
in the extreme southwest and 650 mm in the extreme north 
(Batisani and Yarnal, 2010). The south-eastern part and the north is 
dominated by grassland and savannah trees whereas the Ghanzi, 
Kgalagadi and the west of Southern and Kweneng districts have 
sparse trees and grasses. The soils can be generally categorised 
according to the predominant physiographic units of the sandveld 
and hardveld. The hardveld is characterised by soils which have 
been weathered and alluvial deposits. On the other hand, the 
sandveld area is mostly covered by the Kgalagadi sands (Batisani 
and Yarnal, 2010). 
 
. 
Site description and sampling techniques 

 
Soils were randomly collected along major roadsides (Figure 1a), 
avoiding areas with obvious signs of disturbance such as animal 
burrowing and landfills. The distances between sampling sites were 
chosen to be about 50 or 100 km, depending on proximity of major 
settlements. Four samples were collected at each location as 
follows: One sample at about 10 km before the 50th (or 100th) km 
stretch, the second one at the site of concern and another one 
about 10 km after the site of concern. The fourth (background or 
control) sample was collected at least 500 m away from the 
direction of sampling locations (Figure 1b). All soils were sampled 
at the surface (0 to 10 cm in depth) using hand driven stainless 
steel augers. Exact locations for all sampled sites were determined 
using a global positioning system and entered into a geographical 
information system for data processing. 

 

Sample preparation and analysis 
 
Collected soil samples were air-dried to constant weight and then 



 
 
 

 

sieved through a 500 µm stainless steel mesh wire. Samples of 0.5 

g were digested in 20 ml freshly prepared aqua regia (1:3 HNO3: 
HCl) on a hot plate for 3 h, then evaporated and analysed for metal 
concentration. Standard reference material was prepared using 
stock solution from SAARCHM and MERCH and was used to have 
a check on the accuracy of the results.  

The total concentrations of Al, Co, Cu, Fe, Pb, Mn, Ni and Zn in 
filtrate were then determined using a flame atomic absorption 
spectrometer (Varian SpectrAA 220 FS) at wavelengths, λ: Al = 
309.3 nm; Co = 240.7 nm; Cu = 324.8 nm; Fe = 372.0 nm; Pb =  
217.0 nm; Mn = 279.5; Ni = 232.0 nm and Zn = 213.9 nm, using air 
acetylene flame. 

 

Assessment of metal contamination 
 
ENRICHMENT  FACTOR  (EF) 

 

Assessment of metal and level of contamination in soils require pre-
anthropogenic knowledge of metal concentrations to act as pristine 
values. A number of different enrichment calculation methods and 
different reference material have been reported (Ogusola et al., 
1994; Gaiero et al., 1997; Sutherland et al., 2000; Kamau, 2002; 
Valdés et al., 2005; Ghrefat and Yusuf, 2006; Abrahim and Parker, 
2008; Akoto et al., 2008; Dragović et al., 2008; Charkravarty and 
Patgiri, 2009; Harikumar and Jisha, 2010; Sekabira, 2010; 
Olubunmi and Olorunsola, 2010). In this manuscript, the degree of 
anthropogenic pollution was established by adapting enrichment 
factor ratios (EF) used by Sutherland et al. (2000), as follows: 
 

EF  

C
m Sample  

(1)  
Median Cm Background  2  MAD Cm Background 

 

 
 

 

Where, Cm Sample is the concentration of a given metal along the 

roadside. Median Cm Background is median concentration of an 

element in the background soil sample and MAD Cm Background is 
the median absolute deviation from median, defined as: 

 
 

 

 

Each site was evaluated for the extent of metal pollution by 
employing the method based on the pollution load index (PLI) 
developed by Thomilson et al. (1980), as follows: 
 

1  
 

PLI   CF1  CF2  CF3  K CFn  
 

(4) 
 

n 
 

 

where n is the number of metals studied (eight in this study) and CF 
is the contamination factor calculated as described in Equation 3. 
The PLI provides simple but comparative means for assessing a 
site quality, where a value of PLI < 1 denote perfection; PLI = 1 
present that only baseline levels of pollutants are present and PLI > 
1 would indicate deterioration of site quality (Thomilson et al., 
1980). 

 
This type of measure has however been defined by some authors in 
several ways, for example, as the numerical sum of eight specific 
contamination factors (Hakanson, 1980), whereas, Abrahim (2005) 
assessed the site quality as the arithmetic mean of the analysed 
pollutants. In this study, the authors found it appropriate to express 
the PLI as the geometric mean of the studied pollutants since this 
method tends to reduce the outliers, which might bias the reported 
results. 

 

GEOACCUMULATION  INDEX (Igeo) 

 
Enrichment of metal concentration above baseline concentrations 
was calculated using the method proposed by Muller (1969), 

termed the geoaccumulation index (Igeo). This method assesses the 
metal pollution in terms of seven (0 to 6) enrichment classes 
ranging from background concentration to very heavily polluted, as 
follows: 
 

   Cm  Sample  
(5)  I

 geo  log2 
  

 

 

1.5 

 
 

   Cm  Background  
  

The factor 1.5 is introduced in this equation to minimise the effect of  
MAD  median   x1 − median j  x j     (2) possible  variations in  the  background  values, Cm Background,  

which may be attributed to lithogenic variations in soils. The seven 

proposed descriptive classes for Igeo values are given in Table 1 
(Muller, 1969). 

 

 

Statistical analysis 
 

 
 

 

 
 

 

CONTAMINATION FACTOR  (CF) 

 
The level of contamination of soil by metal is expressed in terms of 
a contamination factor (CF) calculated as: 

 
 

In order to study the characteristics of roadside soils, the 
concentrations of heavy metals content in surface soils 
were subjected to correlation analysis, Principal 
Component Factor Analysis (PCA) and Hierarchical 
Cluster analysis (CA) by SPSS PASW Statistics 17 to 
determine association as well as the differences in the 
concentration between different zones. 
 

 

RESULTS AND DISCUSSION 
 

CF   

 Cm  Sample  

(3) Heavy metal concentrations in soils 
 

C m Background  
 

       

where the contamination factor CF < 1 refers to low contamination; 
The mean heavy metal concentrations (in µg/g) along 

 

roadside soils ranged from (12.80 to 34.46) Al; (0.01 to  

1 ≤ CF < 3 means moderate contamination; 3 ≤ CF ≤ 6 indicates  

0.02) Co; (0.02 to 0.08) Cu; (25.36 to 87) Fe; (0.04 to 
 

considerable  contamination  and  CF  >  6  indicates  very  high  

0.51) Mn; (0.37 to 0.48) Ni; (0.04 to 0.21) and (0.05 to 
 

contamination.  
 

EF < 2: Deficiently to minimal enrichment 

2 ≤ EF < 5: Moderate enrichment 

5 ≤ EF < 20: Significant enrichment 

20 ≤ EF < 40: Very high enrichment 

EF ≥ 40: Extremely high enrichment 

This method is less affected by extremes in the tail often 
encountered with geochemical data, because the data in the tails 
have less influence on the calculation of the median than they do on 
the mean (Chester et al., 1985; Gaiero et al., 1997). Enrichment 
factor categories for Equation 1 are outlined as follows: 



  
 
 

 

Table 1. The Igeo classes with respect to soil quality.  
 

Igeovalue Igeoclass Designation of soil quality  

> 5 6 Extremely contaminated  

4 - 5 5 Strongly to extremely contaminated  

3 - 4 4 Strongly contaminated  

2 - 3 3 Moderately to strongly contaminated  

1 - 2 2 Moderately contaminated  

0 - 1 1 Uncontaminated to moderately contaminated  

0 0 Uncontaminated  

       
       

       
        

 
 
 

 
Figure 2. Enrichment factors for heavy metals along roadside soils for each sampled zone. 

 

 

0.14) Zn. Since this study is the first of its kind for 
Botswana major roadside soils, there is no local 
information in literature available for comparison. Data 
reported here were therefore used to examine the extent 
of contamination by Al, Co, Cu, Fe, Pb, Mn, Ni and Zn 
using comparable pristine samples obtained at least 0.5 
km from the roadside sampled sites. Concentrations of 
individual heavy metal elements and their background 
data are given in the appendix. 
 

 

ENRICHMENT  FACTOR 

 
Enrichment factors of various metals in the roadside soils 
in sampled zones are presented in Figure 2.  

According to Figure 2, EF ratios suggest that all metals 
are deficiently to minimally enriched. These results are 
contrary to those previously reported by Mmolawa et al. 
(2010). In their preliminary study, the authors reported 
moderate (Co, Cu, Fe and Ni) to extreme (Pb) enrichment 

 
 

 

in most roadside soils studied here. The dissimilarities 
may however, be ascribed to the different approaches 
used in the enrichment factor calculation methods. The 
previous study employed a normalised enrichment factor 
approach for metal concentrations using world 
uncontaminated background soils values, and iron as a 
metal of normalization, an approach which is less reliable 
since it ignores the fact that some geologic materials may 
have naturally high element concentrations and that the 
world reference values could be higher or lower 
compared to local conditions. 

 

CONTAMINATION FACTOR (CF) 
 
Contamination factors of various metals in the roadside 
soils in sampled zones are presented in Table 2.  

Using the contamination factor categories previously 
described, zones FN and MG suffered moderate 
contamination by all metals except Co and Zn, 
respectively. On the other hand, zones GK and TS 



 
 
 

 
Table 2. Contamination factors for heavy metals along roadside soils for each 
sampled zone.  

 
  Al Co Cu Fe Mn Ni Pb Zn 

 FN 1.46 0.86 1.30 1.05 1.95 1.18 1.21 1.01 

 NM 0.93 0.78 1.01 0.69 1.36 1.02 2.14 0.50 

 MG 1.10 1.07 1.56 1.52 2.60 1.16 1.51 0.93 

 GK 1.30 0.39 0.44 0.39 0.22 0.91 0.37 0.38 

 TS 0.59 0.91 0.47 0.88 0.30 1.01 0.57 0.52  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 
Figure 3. Pollution load index, PLI for the eight metals studied at the sites. 

 

 

displayed low contamination by all metals except for Al 
and Ni, respectively, which showed moderate contami-
nation. Zone NM displayed moderate contamination by 
Cu, Mn, Ni and Pb, and low contamination by Al, Co, Fe 
and Zn. 
 

 

POLLUTION LOAD  INDEX (PLI) 

 

To effectively compare whether the five stations suffer 
contamination or not, the pollution load index, PLI, 
described in Equation 4, was used. The PLI is aimed at 
providing a measure of the degree of overall 
contamination at a sampling site. Figure 3 shows results 
of the PLI for the eight metals studied at these zones.  

Based on results presented in Figure 3, the overall 
degree of contamination by the 8 metals is of the order 
MG > FN > NM > TS > GK. MG and FN show strong 
signs of pollution or deterioration of site quality, whereas 
NM is almost at baseline level. Sites GK and TS suggest 
perfection (or no overall pollution). Relatively high PLI 

 
 

 

values at MG, FN and, to some degree, NM suggest input 
from anthropogenic sources attributed to increased 
human activities and/or vehicular emissions. These sites 
are along a major highway connecting a number of 
townships and villages having higher populations and 
establishments. Furthermore, FN zone is along the 
highway which is frequently used by commercial trucks 
for transportation of goods to and from Zambia and other 
countries into central Africa. 

 

GEOACCUMULATION INDEX (Igeo) 
 

The calculated geoaccumulation (Igeo) values are 
presented in Figure 4. It is evident from Figure 4 that the 

uncontaminated to moderately contaminated Igeo value of 
‘0 to 1’ is observed at zone MG by Cu, Fe, Mn and Pb, at 
zone NM by Pb and at zone FN by Mn.  

As revealed from the four pollution assessment 

methods; Igeo, PLI, CF and to a less degree, EF, roadside 
soils of zones FN, NM and MG are pollution impacted, as 
compared to GK and TS zones. Statistical tests were 



    

     
     

     
     
      

 
 
 

 
Figure 4. Geoaccumulation indices of heavy metals along the roadside soils. 

 

 
Table 3. The Spearman’s rank correlation coefficient, ρ, between concentrations of metals in 
FN, NM, MG, GK and TS zones.  

 
 Al Co Cu Fe Mn Ni Pb Zn 

Al 1.000        

Co 0.013 1.000       

Cu 0.324* 0.291 1.000      

Fe 0.110 0.268 0.342* 1.000     

Mn 0.506** 0.361* 0.811** 0.307 1.000    

Ni 0.338* 0.091 0.484** 0.240 0.556** 1.000   

Pb 0.056 0.253 0.649** 0.385* 0.602** 0.170 1.000  

Zn 0.350* 0.323* 0.434** 0.335* 0.593** 0.421** 0.293 1.000 

 
* Correlation is significant at the 0.05 level (2-tail)..** Correlation is significant at the 0.01 level 
(2-tail). 

 

 

then performed to establish the inter-metal relationships, 
and classify metals. 
 

 

Statistical analysis 

 

Analysis of variance was employed to determine whether 
groups of variables have the same mean. Sites showed 
no significant effect on variation between group means of 
the heavy metals at different zones except for copper (P  
< 0.001), manganese (P < 0.003) and zinc (P < 0.05). 
This suggests that there is some degree of input of these 
(Cu, Mn and Zn) metals between sites. Inter-elemental 
association was also evaluated by Spearman’s rank 
correlation coefficient, ρ and the results are presented in 
Table 3. 

 
 

 

Table 3 indicates that some elemental pairs, for 
example Al/Mn (r = 0.51, df = 28, P < 0.001), Cu/Mn (r = 
0.81, df = 28, P < 0.0001), and Cu/Pb (r = 0.65, df = 28, P  
< 0.0001) etc, have strong correlations with each other. 
On the other hand, pairs such as Al/Cu (r = 0.22, df = 28, 
P < 0.05), Al/Ni (r = 0.34, df = 28, P < 0.05), and Al/Zn (r 
= 0.35, df = 28, P < 0.05) are moderately significant, 
whereas the rest of elemental pairs show no significant 
correlation with each other. Strong correlations signify 
that each paired elements have common contamination 
sources. Physico-chemical properties and metal 
associations were however not performed in the present 
study, to help in ascertaining these results.  

Agglomeration schedule of cluster analysis (CA) was 
performed on data using nearest neighbour linkage and 
Euclidean distance as a measure of proximity between 



  
 
 

 

Rescaled distance clusters combined  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Figure 5. Dendrogram derived from hierarchical 
analysed soils. 

 

 

samples. Results of CA are shown in Figure 5.  
The hierarchical cluster analysis using nearest 

neighbour method produced two clusters, between which 
the variables were significantly (P < 0.05) different. The 
first cluster contained Co, Cu, Zn, Pb, Mn, Ni and Al. 
These elements were classified as anthropogenic in 
origin, leaving Mn and Al as originating from mixed 
(anthropogenic and lithogenic) sources. The second 
cluster discriminated the lithogenic Fe. Similar studies by 
Al-Momani (2009) found Pb to be strongly associated 
with vehicular emissions and Zn to be associated with 
various industries and metal smelting processes. 
According to Fergusson and Kim (1991), Co, Mn, Al, Cu, 
Ni and Cu are associated with traffic related sources such 
as corrosion of metallic part, concrete materials, re-
entrained dust from roads and tear and wear of tyres and 
engine parts  
Principal component analysis (PCA) was performed to 
establish possible factors that contribute towards the 
metal concentrations and source apportionment. All data 
set was subjected to factor analysis (FA). The number of 
significant principal components (PC) was selected on 
the basis of Varimax orthogonal rotation with Kaiser 
normalisation with eigenvalue greater than 1. The rotated 
component matrix is given in Table 4, and illustrated in 
Figure 6.  

Only the first two components comprising of 60.87% of 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 
cluster  analysis  of  heavy  metals  content  in 
 
 

 
Table 4. Factor analysis (after Varimax rotation) showing 
contribution of statistically dominant variables measured in 
this study.  

 
 Variable PC-1 PC-2 

 Al 0.742 -0.039 

 Co 0.562 0.185 

 Cu 0.783 -0.032 

 Fe 0.701 0.128 

 Mn 0.872 -0.007 

 Ni 0.456 0.614 

 Pb 0.146 -0.917 

 Zn 0.784 0.155 

 Eigenvalue 3.574 1.295 

 % of total variance 44.680 16.188 

 Cumulative (%) 44.680 60.867 
 

 

the total cumulative variances were retained. The first 
principal component, PC-1 explains that 44.68% of the 
total variance is highly loaded by Al, Cu, Fe, Mn, Zn and 
moderately loaded by Co. This factor is a source of mixed 
sources including human activity and vehicular emis-
sions, which is evident from the fact that the soils were 
excavated alongside major highways connecting a 
number of townships and villages having higher 



   
 
 
 
 
 
 
 
 

 

PC-1 
 
 

 

PC-2 
 
 
 
 
 
 
 
 
 
 
 
 

 
Figure 6. Loading plots of PCA analysis of heavy metals concentration for roadside 
soils of Botswana. 

 

 

populations and establishments. Close association of 
these metals is supported by their significant correlation 
(Table 3) and cluster 1 from CA results. The association 
of Mn and Fe could also be due to their common 
occurrence in the basic rock, since the concentrations of 
these elements were lower than that the background 

values (Igeo < 0) except for FN and MG zones whose Igeo 

class category for Mn was ‘1’ and again, Igeo class = 1 for 
Al just for zone MG.  

The second component PC-2 accounts for 16.19% of 
the total variance and contains Ni and Pb. PC-2 is 
strongly loaded by Pb indicating that its source is from 
vehicular emissions. It has been proven that leaded 
gasoline contributes to Pb concentrations in soils. The 
moderate loading of Ni in PC-2, shared in between, to a 
lesser extent, PC-1 suggests both vehicular and industrial 
origins. 
 

 

Conclusions 

 

Anthropogenically impacted and background soils on 
major roadsides were assessed using enrichment factors, 
contamination factors, pollution load index and 
geoaccumulation index for Al, Co, Cu, Fe, Mn, Ni, Pb and 
Zn. Enrichment factor ratios showed that all elements 
were deficiently to minimally enriched.  

The contamination factor showed that generally there is 
low and moderate contamination of the heavy metals 
across the zones FN, NM, MG, GK, and TS.  
The geoaccumulation index showed that zones FN, NM, 
and MG are uncontaminated to moderately contami- 

 
 

 

nated, whereas zones GK and TS are uncontaminated. 
The measure of the degree of overall contamination  

(PLI) at a site indicated strong signs of pollution 
deterioration by the eight measured metals at zones MG 
and FN, no overall contamination at TS and GK and a 
baseline level contamination category for NM.  

Cluster analysis revealed two groups of metals having 
close similarities: firstly Co, Cu, Zn, Pb, Mn, Ni and Al, 
classified as anthropogenic and secondly lithogenic Fe.  

Factor analysis generated two sources of pollutants; 
firstly mixed origin of sources including human activities, 
vehicular emissions and lithogenic occurrences 
characterised by Al, Cu, Fe, Mn, Zn and Co, and 
secondly vehicular emissions characterised by Pb and Ni. 

 

ACKNOWLEDGEMENTS 

 

The authors are gratefully acknowledging the financial 
support of the Research and Publications Committee 
(RPC) of the Botswana College of Agriculture. The 
Department of Chemistry, University of Botswana is to be 
thanked for their assistance in metal analysis. 

 
REFERENCES 
 
Abrahim GMS (2005). Holocene sediments of Tamaki Estuary: 

characterisation and impact of recent human activity on an urban 

estuary in Aukland, Newzealand. PhD thesis, University of Aukland, 

Aukland, Newzealand, p. 36. 
 
Abrahim GMS, Parker RJ (2008). Assessment of heavy metal 

enrichment factors and the degree of contamination in marine 

sediments from Tamaki Estuary, Aukland, New Zealand. Envron. 



 
 
 

 
Monit. Assess. 136: 227–238.  

Akoto O, Ephraim JH, Darko G (2008). Heavy metal pollution in surface 
soils in the vicinity of abundant raiway servicing workshop in Kumasi, 
Ghana. Int. J. Environ. Res. 2(4): 359–364.  

Al-Momani IF (2009). Assessment of trace metal distribution and 
contamination in surface soils of Amman, Jordan. Jordan J. Chem., 
4(1); 77–87. 

Alloway JB (1995). Soil pollution and land contamination. In: Harrison 
RM (Ed). Pollution: Causes, effects and control. The Royal Society of 
Chemistry, Cambridge.  

Batisani N, Yarnal B (2010). Rainfall variability and trends in semi-arid 

Botswana: Implications for climate change adaptation policy. Appl. 

Geogr. 30(4):483–489.  
Charkravarty M, Patgiri AD (2009). Metal pollution assessment in 

sediments of the Dikrong river, N.E. India. J. Hum. Ecl. 27(1): 63–67.  
Chester R, Kudoja WM, Thomas A, Towner J (1985). Pollution 

reconnaissance in stream sediments using non-residual trace metals. 
Environ. Pollut. 23: 213–238.  

Dragović S, Mihailović N, Gajić B (2008). Heavy metals in soils: 
distribution, relationship with soil characteristics and radionuclides 
and multivariate assessment of contamination sources. 
Chemosphere 74: 491–495.  

Fergusson JE, Kim ND (1991). Trace elements in street and house 
dusts: sources and speciation. Sci. Total Environ. 100: 125–150.  

Gaiero DM, Ross GR, Depetris PJ, Kempe S (1997). Spatial and 

temporal variability of total non-residual heavy metals content in 

stream sediments from the Suquia river system, Cordaba, Argentina. 

Water Air Soil Pollut. 93: 303–319.  
Ghrefat H, Yusuf N (2006). Assessing Mn, Fe, Cu, Zn and Cd pollution 

in bottom sediments of Wadi Al-Arab Dam, Jordan. Chemosphere 65: 
2114–2121.  

Hakanson L (1980). Ecological risk index for aquatic pollution control, a 
sedimentological approach. Water Res. 14: 975–1001.  

Harikumar PS, Jisha TS (2010). Distribution pattern of trace metal 
pollutants in the sediments of an urban wetland in the southwest 
coast of India. Int. J. Eng. Sci. Tech. 2(5): 540–850.  

Harrison RM, Laxen DPH, Wilson SJ (1981). Chemical association of 
lead, cadmium, copper and zinc in street dust and roadside soils. 
Environ. Sci. Tech. 15: 1378–1383.  

Joshi SR, Kumar R, Bhagobaty RK, Thokchom S (2010). Impact of 
pollution on microbial activities in sub-tropical forest soil of north east 
India. Research Journal of Environmental Sciences 4(3):280–287.  

Kamau JN (2002). Heahy metal distribution and enrichment at Port-
Reitz creek, Mombasa. Western Indian Ocean J. Mar. Sci. 1(1): 65– 
70.  

Mmolawa KB, Likuku, AS, Gaboutloeloe GK (2010). Reconnaissance of 
heavy metal distribution and enrichment around Botswana. Fifth 
International Conference of Environmental Science & Technology, 
Houston, Texas, USA July12-16, 2010.  

Muller G (1969). Index of geoaccumulation in sediments of the Rhine 
river. Geol. J. 2(3): 108–118.  

Ndiokwere CL (1984). A study of heavy metal pollution from motor 
vehicle emission and its effect on roadside soil, vegetation and crops 
of Nigeria. Environ. Pollut. Ser. B 7: 35–42.  

Ogusola OJ, Oluwole AF, Asubiojo OI, Olaniyi HB, Akeredolu FA, 
Akanle OA, Spyrou NM, Ward NI, Ruck W (1994). Traffic pollution: 
preliminary elemental characterization of roadside dust in Lagos, 
Nigeria. Sci. Total Environ. 146/147: 175–184.  

Olubunmi FE, Olorunsola OE (2010). Evaluation of the status of heavy 
metal pollution of sediment of Agbabu bitumen deposits area, Nigeria. 
Eur. J. Sci. Res. 41(3):373–382. 

 
 
 
 

 
Pagotto C, Rèmy N, Legret M, Le Cloirec P (2001). Heavy metal 

pollution on road dust and roadside soil near a major rural highway. 
Environ. Technol. 22: 307–319.  

Panek E, Zawodny Z (1993). Trace metals in the roadside mountain 
soils of Sierra Nevada, Spain. Environ. Geochem. Health. 15(4):229– 
235.  

Sekabira K, Origa HO, Basamba TA, Mutumba G, Kakulidi E (2010). 
Assessment of heavy metal pollution in the urban stream sediments 
and its tributaries. Int. J. Sci. Tec. 7(3):435–446.  

Sutherland RA, Tolosa CA, Tack FMG, Verloo MG (2000). 
Characterization of selected element concentration and enrichment 
ratios in background and anthropogenically impacted roadside areas. 
Arch. Environ. Contam. Toxicol. 38: 428–438.  

Thomilson DC, Wilson DJ, Harris CR, Jeffrey DW (1980). Problem in 
heavy metals in estuaries and the formation of pollution index. 
Helgol. Wiss. Meeresunlter. 33(1–4): 566–575.  

Valdés J, Vargas G, Sifeddine A, Orttlieb L Guiñez M (2005). 
Distribution and enrichment evaluation of heavy metals in Mejillones 
bay (23°S), Northern Chile: geochemical and statistical approach. 
Mar. Pollut. Bull. 50: 1558–1568. 



  
 
 

 
Appendix 1. Concentrations of heavy metals in roadside and background soils (µg/g).  

 
 Zone FN Al Co Cu Fe Mn Ni Pb Zn 

 FN1A 54.70 0.01 0.08 109.98 0.43 0.51 0.13 0.10 

 FN 38.32 0.01 0.08 39.75 0.37 0.46 0.10 0.10 

 FN1B 33.92 0.04 0.15 256.04 0.41 0.55 0.13 0.12 

 FN2A 38.70 0.01 0.12 14.82 0.69 0.46 0.13 0.14 

 FN 68.39 0.05 0.04 12.19 0.71 0.62 0.06 0.29 

 FN2B 30.75 0.01 0.04 13.33 0.35 0.50 0.08 0.09 

 FN3A 24.52 0.02 0.02 78.07 0.28 0.42 0.03 0.29 

 FN 14.63 0.00 0.02 46.67 0.11 0.40 0.34 0.07 

 FN3B 6.17 0.03 0.01 39.00 0.07 0.38 0.09 0.06 

 Mean 34.46 0.02 0.06 67.76 0.38 0.48 0.12 0.14 

 S.D 19.04 0.02 0.05 77.66 0.22 0.08 0.09 0.09 

 FBb 47.99 0.02 0.11 280.52 0.97 0.50 0.01 0.34 

 Zone NM         
 NM1A 14.20 0.03 0.04 40.36 0.18 0.42 0.07 0.10 

 NM 40.76 0.01 0.05 51.78 0.33 0.31 1.30 0.06 

 NM1B 16.03 0.02 0.04 48.32 0.27 0.39 0.14 0.05 

 NM2A 14.82 0.01 0.04 32.35 0.22 0.42 0.06 0.11 

 NM 17.96 0.02 0.04 40.16 0.31 0.38 0.06 0.06 

 NM2B 50.38 0.01 0.06 90.80 0.42 0.47 0.07 0.07 

 NM3A 22.53 0.03 0.06 49.18 0.31 0.43 0.10 0.07 

 NM 8.34 0.02 0.06 22.59 0.20 0.47 0.07 0.05 

 NM3B 11.84 0.00 0.06 23.12 0.16 0.42 0.06 0.05 

 Mean 21.87 0.02 0.05 44.30 0.26 0.41 0.21 0.07 

 S.D 14.19 0.01 0.01 20.47 0.08 0.05 0.41 0.02 

 NMb 23.58 0.01 0.05 47.17 0.20 0.41 0.19 0.14 

 Zone MG         
 MG1A 15.09 0.03 0.05 34.56 0.22 0.52 0.07 0.06 

 MG 14.06 0.01 0.07 52.98 0.17 0.21 0.49 0.10 

 MG1B 0.00 0.02 0.06 19.92 0.14 0.41 0.07 0.04 

 MG2A 15.78 0.01 0.06 79.16 0.19 0.45 0.08 0.06 

 MG 74.50 0.06 0.12 274.94 0.93 0.51 0.08 0.32 

 MG2B 50.80 0.02 0.09 43.06 1.46 0.45 0.12 0.27 

 MG3A 15.63 0.03 0.08 76.62 0.42 0.45 0.23 0.08 

 MG 25.03 0.00 0.07 115.24 0.29 0.73 0.09 0.07 

 MG3B 21.53 0.04 0.09 184.16 0.75 0.51 0.13 0.16 

 Mean 25.60 0.02 0.07 92.78 0.48 0.46 0.16 0.13 

 SD 21.45 0.02 0.02 79.80 0.44 0.13 0.13 0.10 

 MGb 63.18 0.06 0.11 488.76 0.75 0.51 0.10 0.13 

 Zone GK         
 GK1A 24.34 0.01 0.02 32.62 0.03 0.38 0.04 0.11 

 GK 31.33 0.02 0.02 30.74 0.02 0.29 0.04 0.05 

 GK1B 32.06 0.00 0.03 31.89 0.10 0.30 0.03 0.03 

 GK2A 30.23 0.00 0.01 19.97 0.08 0.42 0.04 0.04 

 GK 32.06 0.01 0.02 21.08 0.01 0.38 0.03 0.04 

 GK2B 33.42 0.00 0.02 15.84 0.01 0.44 0.04 0.05 

 Mean 35.23 0.02 0.03 91.56 0.14 0.39 0.05 0.06 

 S.D 12.67 0.02 0.03 175.27 0.27 0.08 0.02 0.04 

 GKb 6.32 0.02 0.02 27.67 0.07 0.40 0.31 0.15 



         

  Appendix 1. Contd.        
           

  Zone TS         

  TS1A 20.43 0.06 0.03 92.48 0.07 0.41 0.06 0.09 

  TS 15.39 0.00 0.02 63.03 0.07 0.48 0.04 0.07 

  TS1B 15.17 0.00 0.02 58.51 0.05 0.52 0.05 0.06 

  TS2A 9.53 0.03 0.02 56.32 0.05 0.33 0.08 0.05 

  TS 13.28 0.03 0.03 43.91 0.08 0.32 0.06 0.06 

  TS2B 9.45 0.00 0.02 27.23 0.03 0.39 0.05 0.10 

  Mean 12.80 0.02 0.02 52.74 0.06 0.41 0.09 0.08 

  S.D 4.74 0.02 0.00 22.69 0.02 0.07 0.10 0.04 

  TSb 13.29 0.02 0.02 64.57 0.02 0.38 0.08 0.06 
 

SD = standard deviation; FBb, NMb, MGb, GKb and TSb are background sites for the five zones, respectively. 


