Urban Highway Runoff in Nigeria 1: Heavy Metals in Sheet Flow from the Main Expressway in Lagos Metropolis
Whole-unfiltered-water samples were collected from the runoff of a major highway in Lagos metropolis at four different points along the road and at the receiving stream through which the runoff eventually discharges into the lagoon. The study was conducted to characterize the types and levels of pollutants on a typical Nigerian road and was carried out during three different rainy days between the months of February and May, 2004. The collected samples were analysed for the presence of heavy metals; Cd, Cr, Cu, Pb and Zn. Pollution-indicating indices such as dissolved oxygen, BOD5, total solids, conductivity, chlorides, hardness, alkalinity and acidity were also determined. The results of the analyses showed that Zn was most prominent in the sampled runoffs with mean concentrations of 68.7±3.0, 119.7±58.3 and 33.1±24.0 ppb in the three sampled events. Cd -at a concentration of 0.6 ppb was only detected in one of the samples. The concentrations of these metals were shown to vary with the time of sampling during the precipitation. Levels were generally higher in first flush samples.
Pollution from non-point sources such as highway runoff has continued to be a major source of concern for environmental regulatory bodies and other stakeholders the world over as they contribute to the pollutant load of the receiving environment, in most cases the water bodies and farmlands. This results to a gradual degradation of the receiving water quality and an eventual impairment of the beneficial uses of such receiving environment (Pitt et al., 1995). The Nigerian situation is further exacerbated by the reality of increasing large-scale importation of old/fairly used vehicles for use on the Nigerian highways. Lagos is the commercial capital of Nigeria and its the fastest growing metropolis in Nigeria (and probably in sub-Saharan Africa) with the nations largest population and network of highways. Hence, there is a need for a comprehensive study on the effect (s) of runoff from these highways on the proximate environment.
A wide range of pollutants have been shown to be present in highway runoff
with about 75% (by dry weight) of these pollutants derived directly or indirectly
from vehicles, road surface degradation, atmospheric sources and road maintenance
(Ellis and Revitt, 1991). Heavy metals such as copper, cadmium, chromium, lead
and zinc among others are constantly being studied and monitored in highway
runoffs because of their probable mobilization to useable water systems such
as the beaches and underground water, where at elevated concentration levels
could cause public health risks (Dwight et al., 2002). As these heavy
metals and other pollutants are continuously mobilized to such water bodies
and exposed land mass, they ultimately lead to soil and water pollution (Perdikaki
and Mason, 1999). Though, the level of the pollution depends on the nature of
civil works on the highway, the duration of the Antecedent Dry Period (ADP)
before the rainfall, size of the rainfall, volume of traffic per day of the
study site and nature of land use of the adjoining area (Driscoll et al.,
1990), various studies have emphasized on the importance of the first flush
phenomenon, which are basically samples collected within the first few minutes
of the commencement of rainfall, as being a suitable marker towards identifying
the upper limits of these pollutants (Barret et al., 1993; Soller et
al., 2004; Mangani et al., 2005).
Globally, the concentration, transport and fate of some of these pollutants
are usually difficult to generalise on especially in most runoff studies (Grenato
and Smith, 1999) because rainfall occur not at fixed and definite times, but
randomly with varying intensity, resultant volume and quality of the runoff
(Marsalek et al., 1997; Luke and Barrie, 2001). Nevertheless, persistent
efforts directed at monitoring these pollutants have shown that Zinc, Lead,
Copper, Cadmium and Nickel along with some other heavy metals could all be found
in urban and highway runoffs (Marsalek et al., 1997; Luke and Barrie,
2001; Mangani et al., 2005). Other pollutants that could be found in
highway runoff include hydrocarbons and bacteria of human origin such as faecal
coliforms (Barret et al., 1993), along side pollution-indicator indices
such as total solids, total suspended solids, dissolved oxygen (DO), Biochemical
Oxygen Demand (BOD) and Chemical Oxygen Demand (COD) (Perdikaki and Mason 1999;
Hvitved-jacobson and Yousef, 1991).
The monitoring of the composition of wet-deposition (e.g., rain, sleet, snow) on urban highways and its effects on urban pollution has been an important part of environmental studies in many countries (Mangani et al., 2005; Backstrom et al., 2003), where rich data have been generated on the composition of highway and other urban source runoff. Although, increasing urbanization in Nigeria has led to thousands of kilometers of highway in Nigerias cities, yet, no study has ever been undertaken on pollution from highway runoff, even though there are heavy tropical downpours for about 9 months of the year, thus necessitating this research especially when viewed against the backdrop of a general lack of extant vehicular maintenance and road management culture in Nigeria.
The impacts are further escalated by the rampant use of derelict or fairly used automobiles with uncertified engines imported from other countries as well as vehicle maintenance with substandard motor spare parts. Yet, residences of teeming populations border the highways. Hence, a series of studies were commenced to determined the nature and levels of some trace metals in runoffs from strategic highways in Nigeria, identify other pollutants including nutrients that may be contained there-in, establish the maximum levels of these pollutants as may be contained in the first flush analysis and propose suitable BMPs to address any anomaly that may be uncovered.
MATERIALS AND METHODS
Study location: The study site was the Ikorodu expressway in the metropolis
of Lagos, southwest Nigeria. The road is a main dual carriage 8-lane highway,
traversing the entire mainland of the metropolis and stretches from Jibowu in
Lagos Mainland to the nearby twin-city of Ikorodu in the state. It is a major
highway that links various parts of the state via a rich network of adjoining
access roads. It is also a major route for vehicles conveying different products;
commercial, petroleum and agricultural, into and from Lagos. Other features
of the road include a uniform bitumen coated asphalt surface with concrete curbs
and median, service lanes on both sides, a subsurface drainage system with runoff
inlets at intervals along the entire length of the road, a mixed land use of
mainly residential, administrative and commercial enterprises and a relative
traffic volume of about 30,000 vehicles per day (Table 1).
Sampling: For this systematic assessment, a 2.4 km strip of the road
was chosen based on it proximity to a receiving stream (location ST) and a less
probability of pollutant influence from other non-highway related activities
that may lead to their introduction from other sources such as industrial and
domestic wastes. The sampling site was then subdivided into four geo-referenced
sampling locations (Table 2) at the Onipanu (OP), Palmgrove
(PG), Obanikoro (OK), Anthony Village (AT) areas of the city along the highway.
These locations were chosen as a result of their different elevations along
the road with location OK being the lowest point of the four at 75 ft above
sea level. The receiving stream at an elevation of 64 ft drains off the runoff
via the sub-surface pipe and eventually discharges it contents into the Lagos
Sampling was conducted on 2nd February, 13th April and 19th May, 2004, all
between 10-13 h local time. Due to the unpredictability of rain even in a tropical
country like Nigeria, the sampling regime adopted was a flexible one that permits
the collection of whole-water runoff samples as soon as the sampling team could
be mobilized to the various locations on indication of possible rainfall while
bearing in mind the need for at least 72 h antecedent dry period before runoff
sampling (US EPA 1992) and the need to capture the first flush samples.
||Site description of ikorodu road, lagos, Nigeria
|Adapted from Backstrom et al. (2003)
Though one of the sampled events did not meet this requirement but sampling
was nevertheless conducted for comparison basis. Acid pre-washed 2 L polyethylene
bottles were placed in such a way that their openings were at the same level
with the edge of the drainage while their bulks were held at an angle of about
30° in the drainage inlets. The tilting angle ensures that the bottles were
filled up to the brim to exclude air. The bottles were then tightly secured
and immediately transported to the lab for analysis. One sample per location
per sampled event were obtained and the mean for each location evaluated.
Sample treatment and chemical analysis: The Temperature and pH of each
sample was determined in-situ using a mercury-in-glass thermometer and pH meter
respectively. Other analyses were carried out in the laboratory immediately
||Geo-referenced position of sampling locations
|OP = Onipanu, PG = Palmgrove, OK = Obanikoro, AT = Anthony
Village, ST = Receiving Stream, ft = feet
The Dissolved Oxygen (DO) and the Biochemical Oxygen Demand
(BOD5) were determined by the azide modification method; the conductivity
was determined by electrometrical method; the chloride was determined by the
silver nitrate method; hardness, acidity, alkalinity, total solids and dissolved
solids were determined by standard methods for water and waste water analysis
The trace metals were determined by atomic absorption spectrophotometry (AAS) using the BULK SCIENTIFIC 200 instrument. Two hundred fifty milliliter of each acidified sample were pre-concentrated to about 20 mL with nitric acid digestion and the volume made up to 50 mL with distilled water. A procedural blank of the nitric acid and distilled water was also subjected to the same treatment to correct for any tailing effect from the reagent. All parameters were determined within their holding time with necessary preservatives added were applicable.
RESULTS AND DISCUSSION
The pH values obtained for the runoff samples were predominantly neutral, 6.50-7.53,
with a mean of 6.93±0.28 (Table 3). But those of
the receiving stream, ST, were alkaline with a maximum of 10.31 and a mean of
9.80±0.69 (Table 4). This may be ascribed to the
nature of the soil in the stream. The conductivity obtained correlated with
the observed chloride levels with mean of 11.9±0.8 μS cm1;
4.7±3.3 ppm and 11.7±0.3 μS cm1; 3.7±0.6
ppm for the runoff and stream, respectively. The DO level for the runoff with
a mean of 9.0±2.9 ppm was lower than that of ST with a value of 12.7±0.6
ppm. This may be due to the increase dilution effect and agitation of the rainwater
when collecting in the stream.
||Physico-chemical parameters of sampled runoffs
OP = Onipanu, PG = Palmgrove, OK = Obanikoro, AT = Anthony
village, TS = Total Solids, TDS = Total Dissolved Solids, TSS = Total Suspended
Solids, DO = Dissolved Oxygen, BOD5 = Biochemical Oxygen Demand,
ND = Not Detected, ppm = part per million, S1 = 2nd Feb. event,
S2 = 13th April event, S3 = 19th May event
||Physico-chemical parameters of the receiving stream
|S1 = 2nd February event, S2 = 13th April
event, S3 = 9th May event, ppm = part per million, SD = Standard
||Concentration range (ppm) of some pollutant indicators in
|* The Institute of Highways and Transportation, UK, 2001,
** US FHWA, 1996
||Comparison of heavy metal concentration in highway runoff
|* = The Institute of Highways and Transportation, UK, 2001,
** = US FHWA, 1996,+ = Only detected in one sample, BDL =
Below Detection Limit
Overall, the values for the DO did not show a
significant depletion of oxygen with just 20% of the total analysed samples
having values below 5.0 ppm. Fifty percent of the runoff samples showed values exceeding the 50 ppm recommended
BOD5 levels by the FMEnv, Nigeria for discharges to surface water
(FMEnv, 1991), though, the mean were 45.9±23.7 and 41.6±23.2 for
the runoff and stream samples, respectively. The acidity levels were low with
the lowest range observed for the samples from ST, though they have higher alkalinity,
which is as a result of their high pH values. The mean values obtained for the
acidity and alkalinity were, respectively 24.4±8.7; 34.3±7.5 ppm
CaCO3 and 4.9±1.6; 65.0±6.0 ppm CaCO3, respectively
for the runoff and stream samples.
The results obtained for the analysis of the trace metals (Table
8) showed that Zinc was more prominent in all the sampled events with a
mean concentrations of 68.7±3.0 119.7±58.3 and 33.1±24.0
ppb, respectively for the S1, S2 and S3 events.
Cadmium was only detected at site OK during the S2 event. Also, the
concentrations of Lead, Copper and Chromium were higher during the S2
event with values of 54.0±28.4, 31.3±11.7 and 7.75±5.7
ppb, respectively. The higher values observed for the S2 event could
be attributed to the fact that these could be termed a first flush event as
sampling was achieved within 15-20 min of the start of rainfall unlike the other
events which sampling were carried out between 30-35 min from the start of rainfall.
Amongst the samples obtained for the first flush event, location OK generally
had higher concentrations for all the metals analysed (Table
7). This was adduced to the fact that OK is at a lower elevation to all
the other locations, thus there is a strong tendency for it to experience a
wash-down process whereby pollutants in runoffs from places of higher elevations
are by the force of gravity washed down and thus collect at this point before
draining off. Though this trend was not noticed in the other events, it does
not totally foreclose this process as it may have been masked by the dilution
effect at the time of sampling. The levels of trace metals obtained in the samples
from the receiving stream, ST (Table 7) were generally lower
than those of the runoff from the highway. This is as a result of the high dilution
factor in the stream.
Comparatively, the results from this study are consistent with the trend of
similar studies around the world as shown in Table 5 and
6 (US FHWA, 1996; Institute of Highway and Transportation, UK 2001).
||Concentrations (ppb) of heavy metals in sampled runoff at
|S1 = 2nd February event, S2 = 13th April
event, S3 = 9th May event, ppb = part per billion
||Mean concentrations (ppb) of heavy metals in sampled runoff
* First flush concentrations, ** Only detected in the S2
event at location Ok, + Data provided from the nearest meteorological
station to sampling site by the Nigerian Meteorological Agency, Oshodi,
Recently, Backstrom et
al. (2003) showed that Zn at a concentration of 89.3 and 124 ppb was more
prominent in runoff collected in summer and winter, respectively in Svaneberg,
Sweden. Also, Mangani et al. (2005) concluded that Zn was the most abundant
of the heavy metals studied from the first flush analysis of runoff from a highway
in central Italy, with variabilities in observed concentrations ascribed to
site characteristics and rainfall pattern. Soller et al. (2004) in a
study conducted in San Jose, US, showed that the concentrations of heavy metals
in first flush samples is dependent on the antecedent dry period and the size
of the rainfall with an observed range of 25-260, 40-470 and 0.7-18.5 ppb for
Zn, Cu and Pb, respectively. For the present study, the high concentrations
of the metals observed for the S2 event could be ascribed to the
antecedent dry period, ADP (Table 7) and period of sampling
while those of the S3 were due to the fact that a larger size of
rainfall of 18.9 mm witnessed a day prior to the sampling must have washed most
of any accumulated pollutant off the road surface.
This is also coupled with the fact that the period of sampling was not as early as that of S2. By and large, the range of the S2 event showed the importance of ADP and early sampling for first flush analysis.
By and large, the following were deduced from this study:
||The range obtained for the heavy metals were in tandem with
those of similar studies elsewhere as shown in Table 6.
||They were also below the permissible concentrations of trace metals in
water being discharged to surface water as established by the FMEnv., Nigeria.
||Further studies are on-going on the PAHs and nutrients
levels in the Lagos highway runoff as well as those from other states
of southern Nigeria.
||Also, the Nigerian authorities are being engaged to incorporate suitable
Best Management Practice, BMP, when designing highways as is being practiced
in developed countries to check pollution from road runoff.
APHA., AWWA. and WPCF., 1995. Standard Methods for the Examination of Water and Wastewater. 17th Edn., American Public Health Association, Washington, DC.
Backstrom, M., U. Nilsson, K. Hakansson, B. Allard and S. Karlsson, 2003. Speciation of heavy metals in road runoff and roadside total deposition. Water, Air Soil Pollut., 147: 343-366.
Direct Link |
Barrett, M., R. Zuber, E. Collins, J. Malina, Jr., R. Charbeneau and G. Ward, 1993. A review and evaluation of literature pertaining to the quantity and control of pollution form highway runoff and construction. Centre for Research in Water Resources, Bureau of Engineering Research, University of Texas, Austin, TX.
Driscoll, W.J., S. Alyssa Mueller, B.A. Eipper and G.P. Mueller, 1990. Pollutants Loadings and Impacts from Highway Stormwater Runoff. Vol. 1, Federal Highway Authority, Washington DC., USA.
Dwight, R.H., C.S. Jan, B.B. Dean and H.O. Betty, 2002. Association of Urban runoff with coastal water quality in orange county, California. Water Environ. Res., 74: 82-90.
Direct Link |
Ellis, J.B. and D.M. Revitt, 1991. Drainage from roads: Control and treatment of highway runoff. Report NRA, 43804/MID OR, Thames NRA, Reading.
FMEnv, 1991. Interim effluent limitation guidelines in Nigeria for all categories of industries. Federal Ministry of Environment, Nigeria.
Grenato, G.E. and K.P. Smith, 1999. Estimating concentrations of road-salt constituents in highway runoff from measurements of specific conductance. U.S. Geological Survey Water Resources Investigation Report, 99-4077, pp: 22.
Guidelines for the Environmental Management of Highways, 2001. The institute of highways and transportation. http://www.open.ac.uk/cpd/index.php?q=node/116.
Hvitved-Jacobson, T. and Y.A. Yousef, 1991. Highway Runoff Quality, Environmental Impacts and Control. In: Highway Pollution, Hamilton, R.S. and R.M. Harrison (Eds.). Elsevier, London, pp: 165-208.
Luke, M.M. and M.P. Barrie, 2001. Partitioning of metals (Fe, Pb, Cu, Zn) in urban runoff from the Kaikorai Valley, Dunedin, New Zealand. New Zealand J. Mar. Freshwater Res., 35: 615-624.
Direct Link |
Mangani, G., A. Berloni, F. Bellucci, F. Tatano and M. Maione, 2005. Evaluation of the pollutant content in road runoff first flush waters. Water Air Soil Pollut., 160: 213-228.
Direct Link |
Marsalek, J., B. Brownlee, T. Mayer, S. Lawal and G.A. Larkin, 1997. Heavy metals and PAHS in stormwater runoff from the Skyway Bridge, Burlington, Ontario. Water Qual. Res. J. Can., 32: 815-827.
Direct Link |
Perdikaki, K. and C.F. Mason, 1999. Impact of road runoff on receiving streams in Eastern England. Water Res., 33: 1627-1633.
Direct Link |
Pitt, R., R. Field, M. Lalor and M. Brown, 1995. Urban stormwater toxic pollutants: Assessment sources and treatability. Water Environ. Res., 67: 260-275.
Direct Link |
Soller, J. and J. Stephenson, 2004. Evaluation of first flush pollutant loading and implications for water resources and urban runoff management. American Waterworks Association (AWWA), Water Sources Conference.
US EPA, 1992. NPDES storm water sampling guidance document. US Environmental Protection Agency, Office of Water, Washington, D.C. EPA 833/B-92-001.
US FHWA, 1996. Evaluation and management of highway runoff water quality. Federal Highway Authority, FHWA-PD-96-032.