Stormwater impact on water quality of rivers subjected to point sources and urbanization - the case of Addis Ababa, Ethiopia: Stormwater impact on water quality of rivers

Abstract

Urbanization of a catchment often causes degeneration of rivers. We studied the water quality of three rivers in Addis Ababa based on the impact of stormwater and non‐point sources, and urbanization. Along these rivers several point sources were registered, with direct discharge of industrial and domestic wastes into them. To distinguish the impact of these year‐round point‐sources from stormwater, we analysed physicochemical parameters, nutrients and heavy metals sampled from upstream to downstream sections of each river in the dry and wet season. Dissolved oxygen (DO), NO2–N, NH4–N, PO4–P, (Cr(VI) and Cu) exceeded international standards, pointing to a generally poor water quality of the rivers in both seasons. NO3–N, Mn and Zn were problematic in dry season only. Although stormwater improved DO, conductivity, PO4–P, Cr(VI) and Zn, the levels were still critical, pointing to construction sites, agriculture and pit latrines, somewhat offsetting the effect of stormwater dilution. No clear impact of urbanization pressure was found.

Full text

Water and Environment Journal. Prin t ISSN 1747-6585
Water and Environment Journal 33 (2019) 98–110 © 2018 CIWEM.98
Introduction
Rivers are often at the centre of urban development due
to the many services they provide (Cohn 1998). As a city
grows bigger, the environmental pressure on its local
rivers increases, especially if the city does not have the
resources to manage solid waste and wastewater ade-
quately. This is often the case in cities in developing
countries, where wastewater from industries and house-
holds may be discharged in untreated form to rivers, and
where solid waste may accumulate at the course of the
rivers (Strandberg, 1971). In addition, stormwater is a
major cause of urban river water pollution in both devel-
oped and developing countries (Karbassi et al. 20 07;
Nazafpour et al. 2008). As it moves on, stormwater picks
up contaminant s that have accumulated on the various
surfaces in the antecedent dry period, and carries the
load of contaminants to the river. The concentration of
contaminants in stormwater entering a river is a result
of both the strength of the non-point sources passed on
the way, and the volume of stormwater.
The capital city of Ethiopia, Addis Ababa is experienc-
ing rapid urbanization and may be one of the fastest
growing cities in Africa (Mazhindu et al. 2010). The current
population is 3.434 million (CSA, 2013), corresponding to
a population density of approximately 6,359 persons per
square kilometer. The city has evolved around a number
of rivers, all starting in the mountainous areas within the
boundary of the city. Despite a mean annual rainfall of
1058 mm, most of the rivers are ephemeral due the rain
being concentrated in a four months period from June
to September (NMA, 2016). The elevation of the city pre-
sents significant slopes from 3,000 meters in the north
to 2,100 meters in the south over 30 km (GPS measure-
ment 2016).
Whilst the rivers of Addis Ababa are crucial for the
water supply of the city, they are also being used for
discharge of wastewater and dumping of solid waste.
The level of river water pollution in Addis Ababa has
increased as urbanization intensified and the economic
development remained at a low level (Alemayehu et al.
2003). According to the City of Addis Ababa Bureau of
Finance and Economic Development (2013), about 14.3%
of the total population practices open defecation, whilst
the remaining uses pit and flush latrines. The extent of
wastewater collection coverage of the city is 44.3% dis-
tributed on 7.3% by sewer lines and 37% by vacuum trucks
(AAWSA 2011). Concerning solid waste, 25% is not col-
lected but disposed off, frequently by dumping into rivers
or drains or onto open spaces (Regassa et al. 2011). The
Stormwater impact on water quality of rivers subjected to point
sources and urbanization – the case of Addis Ababa, Ethiopia
Dagnachew Adugna1,5, Brook Lemma2, Geremew Sahilu Gebrie3, Larissa Larsen4, Kumelachew Yeshitela1 &
Marina Bergen Jensen5
1Ethiopian Institute of Architecture, Building Construction and City Development, Addis Ababa University (AAU); 2College of Natural and Computational
Sciences, Addis Ababa University (AAU); 3Ethiopian Institute of Water Resources, Addis Ababa University (AAU); 4Urban and Regional Planning, University
of Michigan; and 5Department of Geosciences and Natural Resource Management, Copenhagen University
Keywords
Addis Ababa; point sources; river water
quality; stormwater; seasons; urbanization.
Correspondence
Dagnachew Adugna, Ethiopian Institute of
Architecture, Building Construction and City
Development, Addis Ababa University (AAU).
Email: dagnachew2@gmail.com
doi:10.1111/wej.12381
Abstract
Urbanization of a catchment often causes degeneration of rivers. We studied the
water quality of three rivers in Addis Ababa based on the impact of stormwater
and non-point sources, and urbanization. Along these rivers several point sources
were registered, with direct discharge of industrial and domestic wastes into
them. To distinguish the impact of these year-round point-sources from storm-
water, we analysed physicochemical parameters, nutrients and heavy metals
sampled from upstream to downstream sections of each river in the dry and
wet season. Dissolved oxygen (DO), NO2–N, NH4–N, PO4–P, (Cr(VI) and Cu) ex-
ceeded international standards, pointing to a generally poor water quality of the
rivers in both seasons. NO3–N, Mn and Zn were problematic in dry season only.
Although stormwater improved DO, conductivity, PO4–P, Cr(VI) and Zn, the levels
were still critical, pointing to construction sites, agriculture and pit latrines, some-
what offsetting the effect of stormwater dilution. No clear impact of urbanization
pressure was found.

Stormwater impact on water quality of riversD. Adugna et al.
Water and Environment Journal 33 (2019) 98–110 © 2018 CIWEM. 99
existing stormwater management systems are fragmen-
tary and mainly serving major roads with rivers being
the final receiving system (Belete, 2011). Stormwater gen-
eration is increasing with the growing area of impervious
surfaces as the city expands and densifies (Butler and
Dav ies, 2 004).
Addis Ababa is subdivided into two major river catch-
ments, namely Little Akaki and Big Akaki. For rivers in
Little Akaki catchment, the following water quality param-
eters have been found by various authors: dissolved
oxygen, DO 0.1–7.12 mg/L (Beyene et al. 2009), electrical
conductivity, EC 576–1790 µS/cm (Tolla 2006; Beyene et
al. 2009; Temesgen 2009), pH 6.59–7.8 (Tolla, 2006;
Temesgen, 2009), NO3–N 0.48–39.78 mg/l (Beyene et al.
2009; Temesgen 2009), Cu 0.02 mg/L, Zn 0.023–0.192
mg/L (Tolla, 2006), and Mn 0.64 mg/L (Temesgen, 2009).
Studies conducted in South Korea (Chang, 2005), Nepal
(Kannel et al, 2007) and Bulgaria (Astel et al. 2007) found
that the water quality of urban rivers deteriorated down-
stream due to cumulative effects coupled with waste-
water discharges into rivers. Furthermore, studies
conducted in China by Ho and Hui (2001) and Ding et
al. (2015) on rivers passing through both forest and
urbanized areas showed that urban areas are the major
contributors of nutrients (N, P) during both the dry and
wet season which could be attributed to point and non-
point source pollution such as municipal and industrial
wastewaters. Another study conducted for two urban
rivers in Japan found that the pollutant types and levels
varied inconsistently from upstream to downstream loca-
tions showing pollutant concentration of NO3-N in one
river increased by 50% moving downstream, whilst in
the other river a downstream reduction was ascribed
to purification provided by riverine wetlands and dilution
(Woli et al. 200 4).
Understanding the contribution of year round point
sources and stormwater-activated non-point sources to
the pollution of running waters will help urban planners
and city decision-makers to effectively prioritize respon-
sive strategies. Taking Addis Ababa as case, the present
study was conducted to:
• observe the water quality of three rivers and compare
findings with Ethiopian and international standards,
• investigate the impact of stormwater runoff on river
water quality by comparing dry and wet seasons, and
• establish any relations between river water quality and
the level of urbanization.
Materials and Methods
Study area
To document the water quality of rivers in Addis Ababa,
we selected three rivers, namely, Shegole, Little Akaki
and Jemo, which are all located in the Little Akaki River
catchment in the western part of the city (Fig. 1). This
catchment drains into the main Akaki River, which is a
tributary of Awash River, draining the central and eastern
part of Ethiopia into an endorheic basin near the border
of Djibouti. Their headwaters originate north of Addis
Ababa. Shegole passes through the oldest, most congested
and densely built-up parts of the city, close to the urban
core, Little Akaki passes through a moderately urbanized
areas, and Jemo passes through relatively less and recently
urbanized parts nearer the periphery of the city.
Thus, at large scale, the three rivers represent an
urbanization gradient with a population density per square
kilometre of 19,294 in Shegole catchment, 8,793 in Little
Akaki catchment and 6,490 in Jemo River catchment, as
Fig. 1. Left: Addis Ababa city with Little Akaki catchment shaded in blue. Right: The sampled rivers with the sampling points indicated with number and
black dots. [Colour figure can be viewed at wileyonlinelibrary.com]
Shegole River
1
2
11
5
3
8
4
6
7
10
9
Jemo River
Little Akaki River
Addis Ababa

D. Adugna et al.Stormwater impact on water quality of rivers
Water and Environment Journal 33 (2019) 98–110 © 2018 CIWEM. © 2018 CIWEM.100
calculated from population numbers (CSA, 2013) and area
of rivers’ catchment was delineated with ArcGIS (version
10.3.1). Within each river, attempts were made to repre-
sent the urbanization gradient by placing sampling sites
in upstream, midstream, and downstream parts, with
upstream sites representing less urbanized areas com-
pared to mid- and downstream sites.
In total, 11 sites were sampled, distributed on the three
rivers as shown in Fig. 1 and Table 1. To document the
difference between seasons, water samples were collected
twice in each season of 2015. Dry season samples were
obtained from March 10–12 (i.e. replication 1) and May
15–17 (i.e. replication 2); and wet season samples on July
18 (i.e. replication 1) after rainfall of 7.4 mm, and August
18 (i.e. replication 2) after rainfall of 12.3 mm.
Water samples were grabbed by lowering a polyethylene
bottle, rinsed in distilled water, into the river. Samples
were stored in a dark refrigerated system for a maximum
of two days until analysis.
Physicochemical parameters analysis
To describe physicochemical conditions dissolved oxygen
(DO), electrical conductivity (EC) and pH were measured
Table 1 Names and coordinates of the sampling sites, and observed main point sources
Rivers and site codes Site descriptions
UTM-coordinate and elevations
above sea level (m)
Point sources (expected corresponding
pollutants are footnoted)
Shegole 1 Nuna gamo Plc 09°03.517′N Tannerya
38o42.762′E Toilet and showers (services)b
2544 m Garagesc, recreationd
2 EiABC 09°00.695′N Liquor and soft drink factoriese
38°43.274′E, 2344 m Public college with dormitoriesf
3 Opposite Mekanisa liquor
factory
09°58.564′N Abattoirg
38°44.101′E, 2234 m Urban agricultureh
Little Akaki 4 Bero 09°03.016′N Urban agricultureh
38°41.063′E, 2470 m Toilet & showers (settlement)b
5 Kolfe High School 09°02.012′N Tannerya, fodder processingi
38°42.125′E, 2374 m Animal productionj, urban agricultureh
6 Zenebe work/Alert 08°59.39′N, 38°42.573′E, 2268 m Toilet and showers (settlement)b
7 Mekanisa liquor factory 08°58.504′N Liquor factorya
38°43.949′E Urban agricultureh
2334 m Seedling nurseryk
Jemo 8 Ayer Tena 08°58.936′N Garagec
38°41.553′E, 2274 m Toilets (recreation facility)b
9 China camp 08°58.418′N Animal productionj
38°42.322′E, 2229 m Gravel quarryingl
10 Yetebaberut fuel station 08°57.76′N Large auto-workshopc, fuel stationm
38°43.107′E Urban agricultureh
2232 m Solid waste dumping siten
11 Lebu 08°57.217′N Urban agriculturei
38°43.877′E Public sanitary facilitiesb
2216 m Demolition wasteo
aHeavy metals, organic micropollutants (e.g polychlorinated biphenyls (PCBs), formaldehyde resins, pesticides (UNEP 1994; Mwinyihija 2010)
bNutrients, detergents (Schmoll 2006)
cHeavy metals, sulphuric acid, aliphatic hydrocarbons (used oil, petroleum), organic micropollutants (automotive paints)
dNutrients, suspendable solids
eNutrients, suspendable solids (Eremektar et al. 1995)
fNutrients, organic micropollutants (cleansing agents),aliphatic hydrocarbons (used oil and grease)
gNutrients, suspendable solids, NaCl (Sangodoyin and Agbawhe 1992)
hNutrients, pesticides, suspendable solids
iNutrients, suspendable solids
jNutrients, NaCl, suspendable solids (Williams 1995)
kNutrients, pesticides, suspendable solids
lHeavy metals, suspendable solids, aliphatic hydrocarbons (used oil, grease, petroleum) (WHO 2003)
mAliphatic hydrocarbons (oil, grease, petroleum leaks), heavy metals (Raad et al. 2012)
nHeavy metals, organic micropollutants, (e.g. cleansing agents, PCB, formaldehyde, resins and pesticides), total solids
oHeavy metals, suspendable solids

Stormwater impact on water quality of riversD. Adugna et al.
Water and Environment Journal 33 (2019) 98–110 © 2018 CIWEM. 101
on-site using a portable multi-parameter digital metre
(HACH, HQ40d) equipped with probes (DO-probe LDO
10105, EC-probe CDC 40105, and pH-probe USPAT 6912050)
and turbidity was measured with portable turbidity-metre
(OAKTON T-100).
Nutrients analysis
Water samples were filtered through a 0.42 µm filter
(Whatman). Nutrients including phosphate (PO4–P), nitrate
(NO3–N), nitrite (NO2–N), and ammonium (NH3–N) were
analysed in a laboratory using a spectrophotometer
(JENWAY 6405UV/Vis) following standard methods in APHA
(2005).
Heavy metals analysis
Water samples from the three rivers were filtered using a
0.42 µm filter (Whatman) to analyse heavy metals including
Cr(VI), Cu, Mn and Zn. The analyses were conducted in a
laboratory with HACH reagents and a DR200 HACH spec-
trophotometer, using diphenylcarbohydrazide for Cr(VI),
bicinchoninate (cuver1 copper reagent) for Cu, periodate
oxidation for Mn, and zincon for Zn methods, respectively.
To the extent possible, the observed water quality of
the three rivers in Addis Ababa was compared with the
Ethiopian environmental standard set by the Ethiopian
EPA and UNIDO (2003) and with freshwater requirements
or acceptable levels for US (USEPA, 1986, updated on-line
2015), Europe (EU, 2013), and Australia-New Zealand
(ANZECC 2000), referred to as international standards,
as summarized in Table 2 in Results.
Regarding turbidity no requirements exist. For nutrient s,
the US and EU standards state that the acceptable levels
depend on local ecoregions, and accordingly should always
be adjusted hereafter. The higher Ethiopian limit for NO3–N
compared to Australia-New Zealand standard may reflect
such differences in ecoregions. Based on a comparative
study by Claussen et al. (2012) on the national require-
ments used by 17 EU member states when implementing
the EU Water Frame Directive it was possible to retrieve
insight on the PO4– P levels employed. From this study,
it is seen that 11 of the 17 member states require a total
PO4-level not exceeding 0.2 mg/L for good river water
quality to be obtained, whilst up to 1 mg total PO4–P is
the maximum allowed by one member state.
Again, the higher Ethiopian limit for PO4–P may reflect
differences in ecoregions. It should be noted that nutrient
standards, both Ethiopian and international concern total
concentrations, so when we compare with our observa-
tions for dissolved concentrations it is a conservative
Table 2 Observed river water quality parameters and allowable environmental standards of Ethiopia, US, Europe and Australia-New Zealand
Pollutant Ethiopian standard International standards
Dry season observations – average
values for river:
Wet season observations
– average values for river:
Shegole Little Akaki Jemo Shegole Little Akaki Jemo
DO (mg/L) No data 4.50a0.18 3.16 2.49 5.43 7.29 5.16
pH 6.5 – 9 6.5–9a7.76 7.94 8.01 8.18 8.27 8.47
Turbidity (NTU) No data No data 239 35 54 376 661 302
Total NO2–N
(mg/L)
No data 0.10b0.90 2.74 2.62 5.37 10.26 11.00
Total NO3–N
(mg/L)
20 3.0b2.39 1.31 1.46 1.88 2.05 2.36
Total NH4–N
(mg/L)
No data 1.0b2.56 2.24 3.16 0.97 0.75 1.76
Total PO4–P
(mg/l)
5.0 0.20c45.55 20.97 41.46 14.77 8.92 16.55
Dissolved
Cr(VI) (mg/L)
0.05 0.0034d0.021 0.075 0.063 0.010 0.055 0.060
Dissolved Mn
(mg/L)
0.5 0.150d0.370 0.320 0.590 0.040 0.043 0.058
Dissolved Cu
(mg/L)
2.0 0.0049d0.070 0.010 0.020 0.060 0.003 0.010
Dissolved Zn
(mg/L)
5.0 0.0078d0.320 0.090 0.100 0.110 0.050 0.070
aUSA
bAustralia-New Zealand
cEU
dEU, Denmark

D. Adugna et al.Stormwater impact on water quality of rivers
Water and Environment Journal 33 (2019) 98–110 © 2018 CIWEM. © 2018 CIWEM.102
Fig. 2. Water quality as observed in wet and dry seasons expressed as DO, EC, pH, turbidity (a) , and dissolved concentrations of PO4
2––P, NO3
––N, NO2
–
–N, and NH4
+–P (b) and Cr (VI), Cu, Mn, Zn (c) of three rivers in Addis Ababa.
Sampling points 1, 2, and 3 represent Shegole, 4, 5, 6 and 7 represent Little Akaki, and 8, 9, 10 and 11 represent Jemo. This order reflects an urbanization
gradient within the catchment, from highest (Shegole) to lowest (Jemo). The numbering of sampling points for each river goes from upstream to
downstream, which also reflects an urbanization gradient along each river. Black lines with circles = wet season, Grey lines with diamonds = dry season,
Non-filled symbols = sampling replication 1, Filled symbols = sampling replication 2.

Stormwater impact on water quality of riversD. Adugna et al.
Water and Environment Journal 33 (2019) 98–110 © 2018 CIWEM. 103
Fig. 2. (continued)

D. Adugna et al.Stormwater impact on water quality of rivers
Water and Environment Journal 33 (2019) 98–110 © 2018 CIWEM. © 2018 CIWEM.104
comparison. Regarding Cr(VI), Cu, Mn and Zn both the
Ethiopian limit s and the Danish long-term threshold values
for implementation of the EU Water Frame Directive
(Ministry of Environment and Food of Denmark 2015) were
used. According to the EU Water Frame Directive, the
threshold levels for Mn and Zn must take the natural
background concentrations into account, which are
unknown for the studied rivers in Addis Ababa, but may
explain the higher limits for these to metals in the Ethiopian
limit. The reason for higher levels for Cr(VI) and Cu in
the Ethiopian limits compared to the Danish is unknown.
Results
The observed water quality in the three investigated riv-
ers of Addis Ababa is presented in Fig. 2. Moreover, the
average values of the dry and wet season observations
in parallel with the Ethiopian and international standards
are presented in Table 2.
Sampling points 1, 2 and 3 represent Shegole; 4, 5, 6
and 7 represent Little Akaki; and 8, 9, 10 and 11 represent
Jemo. This order reflects an urbanization gradient within
the catchment, from highest (Shegole) to lowest (Jemo).
Fig. 2. (continued)

Stormwater impact on water quality of riversD. Adugna et al.
Water and Environment Journal 33 (2019) 98–110 © 2018 CIWEM. 105
The numbering of sampling points for each river goes
from upstream to downstream, which also reflects an
urbanization gradient along each river. Black lines with
circles = wet season, Grey lines with diamonds = dry
season, Non-filled symbols = sampling replication 1, Filled
symbols = sampling replication 2.
Despite limited number of samples, the observed levels
and patterns of the general parameters DO, EC, pH and
turbidity from replication 1 are to a large extent repeated
in replication 2 for both seasons. This indicates that the
employed sampling strategy was capable of capturing
the main factors controlling the river water quality.
When the observed river water quality is compared
with the available Ethiopian standards (Table 2), pH was
within the standard range, although with a tendency to
reaching too alkaline levels. NO3–N, Mn, Cu and Zn were
well below the standard, except Mn for Jemo river in the
dry season. In both seasons, the average values of Cr(VI)
were higher in Little Akaki and Jemo rivers, whilst in
Shegole river it was below the Ethiopian limit. Average
PO4–P was strongly exceeding the limits (Table 2), although
a few samples from both dry and wet season were below
the limit (Fig. 2b). The actual exceedances for PO4–P are
assumed to be significantly higher because we only meas-
ured dissolved PO4– P.
If the observed water quality is compared with inter-
national standards it is seen that the dry season samples
generally did not meet the standards, whilst the wet
season samples for some parameters met the standards
whilst in other cases not, or were even worse (Table 2).
For DO, the international standard of a minimum con-
centration of 4.5 mg O2/L was observed in most of the
wet seasons’ samples but not in the dr y seasons, where
the oxygenation of the water in some samples was close
to zero (Fig. 2a).
Regarding nutrients, it is seen that although nitrate
was in most measurements below the 3 mg/L level, the
associated and more toxic N-forms of nitrite and ammo-
nium severely exceeded the international standards.
Likewise, phosphate surpassed the international standards
of 0.2 mg P/L by several orders of magnitude.
When it comes to the four measured metals it is seen
that Cr(VI) exceeds the international standard in extreme
degree in both the wet and dry seasons, and that this is
also the case for Cu regarding Shegole, whilst Cu is around
the international standard in Little Akaki and exceeds the
international standard slightly in the dry season but not
in the wet season in case of Jemo. Because the natural
background level should be added to the international
standard for Zn it is not possible to conclude weather Zn
met the standard. It is noted that Zn shows a pattern
similar to Cu. Although the international standard for Mn
should also take the natural background concentration,
it can be concluded that in the wet season the Mn is
below the international standard of 0.15 mg/L. The dry
season samples showed higher levels, exceeding this level,
because no background level information is available this
level may still be within the international standard.
The influence of stormwater runoff on river water qual-
ity can be revealed by comparing dry and wet season
samples. Based on DO, EC, NH4–N, PO4–P, Cr, Mn, Cu
and Zn there may be a tendency for improved river water
quality in the wet season, corresponding to stormwater
having a diluting effect on the contaminants in the river
in wet season. Thus, considering samples from same site
it is seen that the DO was up to 86-folds higher, EC was
up to 5-folds lower, both PO4–P and Mn were up to 20
folds lower, and Cr(VI) up to 6 -folds lower in the wet
season compared to the dry season. Regarding NH4–N
and the other metals no clear differences were obser ved.
NO3–N and NO2–N tended to be highest in the wet sea-
son, indicating strong non-point sources being present.
The anticipated urbanization gradient was only reflected
to some extent. This is the case when the three rivers
are compared, where the expected order of contaminant
level would be Shegole, Little Akaki and Jemo, and when
the less urbanized upstream samples of individual rivers
are compared with more urbanized downstream samples
of the same river. Whilst dry season DO was clearly at
the lowest level in the Shegole river compared to Little
Akaki and Jemo, this river was not outstanding otherwise.
Thus, nutrients, Mn and Zn were at similar levels in all
three rivers, and although Cu was highest in samples from
Shegole, the opposite was the case regarding Cr, where
Little Akaki and Jemo presented highest concentrations.
Regarding the within-river comparison, only nutrients
expressed a tendency for increasing loads when moving
from upstream to downstream, although there were quite
some exceptions. Similar patterns were not seen for other
parameters. This suggests the direct discharge of waste-
waters into rivers from the nearby industries, toilet facili-
ties and institutions to have stronger impacts on the river
water quality than the level of urbanization.
Discussion
Comparison of observed water quality levels
with Ethiopian and International standards
As revealed from the three rivers studied, the observed
nutrients and heavy metals were below the Ethiopian
standards, except phosphate, Cr (VI) and Mn which
exceeded the standard in both the dry and wet weather
in almost all measurements. If the observed water qualit y
is compared to international standards, it is extremely
high in the dry season and despite some improvement
for most parameters in the wet season, the levels are

D. Adugna et al.Stormwater impact on water quality of rivers
Water and Environment Journal 33 (2019) 98–110 © 2018 CIWEM. © 2018 CIWEM.106
still exceeded. For PO4– P, N O 2– N, NH3–N, Cr and Cu, the
exceedance is measured in decades.
This comparison is therefore quite important to ensure
local standards gradually become compatible with inter-
national ones. Because improvement of standards and
regulations are iterative processes, environmental water
quality standards are tighten based on progress achieved
and compliance with general social and economic devel-
opment of the area. Putting in place environmental stand-
ards alone doesn’t result in effective water resources
protection. Therefore, regulators, environmental engi-
neers and scientists should monitor and evaluate the
effectiveness of the set standards through continuous
pollution level observation and compare with best inter-
national standards.
Impact of stormwater runoff on river water
quality by comparing dry and wet seasons
If the contaminant levels observed in this study are com-
pared with findings by other authors (e.g. Beyene et al.,
2009), it is seen that the DO-range in our study reached
higher to around 8 mg/L in wet season measurements
in two of the three rivers, but also that EC set record
high levels in single dr y season measurements of all three
rivers exceeding 2000 µS/cm, as did pH, reaching levels
above 8 in all wet season measurements.
For NO3–N and Mn, our observations are in similar
range as reported by Alemayehu (2001) and Temesgen
(2009), as is the case for Cu and Zn concerning the rivers
Little Akaki and Jemo, whilst Shegole River exceeds previ-
ous reporting several folds. For the other parameters,
we found no previous reporting for comparison.
The higher values of the observed river water quality
parameters signifying poor quality can be attributed to
year-round direct discharge of untreated industrial efflu-
ent and domestic wastewater. Here, the tanneries, which
were found at the headwater of two of the three rivers,
may cause permanently high concentrations of Cr and
Cu, whilst the constant discharge of municipal wastewater
from toilets, showers and other domestic sources, includ-
ing open defecation causes a continuous high load of
nutrients. In support of this, it is estimated that 56 % of
municipal wastewater is connected illegally through pipes
and drains to the rivers (AAWSA, 2011).
The fact that stormwater runoff only reduced the con-
taminant levels to a minor extent indicates that additional
contaminant sources become active in wet weather. Here,
the many stretches of agricultural practices along the
river may provide a source of nutrient s, especially P, which
may be leached from soil enriched with inorganic fertiliz-
ers. Agricultural practices may also explain the increased
levels of nitrite and ammonia in wet season, which is in
agreement with the study conducted by Carpenter et al.
(1998). Another likely source of nutrients ac tivated during
wet weather is the remaining municipal wastewater, of
which a fraction is assumed to end up in pit latrines
which may overflow during rain.
The reason for the high pH levels observed in the wet
season is unknown, but one plausible explanation may
be the rapid urbanization rate and corresponding high
number of construction sites, where concrete is being
used. The many building activities may also be the reason
for high Cr during wet weather because Cr is found in
concrete (Kayhanian et al. 2009).
The up to fivefold higher turbidity of river water in the
wet season compared to the dr y season is a clear indica-
tion of the significant contribution of stormwater runoff
in the wet season. Turbidity is assumed to increase during
storm events due to soil erosion from agricultural prac-
tices, and wash- off solids from demolished materials from
construction sites and unmanaged solid wastes (Regassa
et al. 2 011).
The improvement of DO in the wet season might be
attributed by turbulence of running water, combined with
lower content of organic matter in the stormwater runoff,
as also suggested in rivers in India and Argentina (Pesce
and Wunderlin 2000; Joshi et al. 2009). The low oxygena-
tion of the river water during dry season, in some samples
approaching zero, is ascribed to the organic loads from
direct discharge of domestic and industrial wastewater,
as also pointed out by Awoke et al. (2016) comparing
river water quality in four river basins in Ethiopia, and
Alemayehu (2001) ascribing low DO in rivers of Addis
Ababa from solid waste dumping and wastewater dis-
charge. Furthermore, our finding is consistent with study
conducted by Karn and Harada (2001) who reported that
the dry season contributed significant impact on river
pollution in three urban territories of Nepal, India, and
Bangladesh due to year round pollutants flow into the
rivers and reduced river water discharge.
Generally, the increase in mean flow rate from 1.16
m3/s in June to 19.73 m3/s in August and then, the decrease
of the flow rate from 0.95 m3/s in October to 0.21 m3/s
in May of the rivers in the study area is an indication to
the decrease in concentration and wash off pollutants
during the rainy season.
Therefore, understanding of the pollution status of riv-
ers both in the dr y and wet season is central to eviden-
tially identifying the contribution of point sources and
non-point sources, as stormwater in the wet season is
known to degenerate river water quality in urban set-
tings. Stormwater, an urbanization problem, is commonly
polluted by sediments, heavy metals, untreated domestic
and municipal wastes and is mainly discharged directly
into rivers. Thus, to prevent river water pollution from

Stormwater impact on water quality of riversD. Adugna et al.
Water and Environment Journal 33 (2019) 98–110 © 2018 CIWEM. 107
point and non-point sources, establishing clear water
management measures need to be integrated with waste
management practices.
Impact of urbanization on river water quality
The fact that only weak indications of impact of urbaniza-
tion on the obser ved rivers water qualit y suggests point
sources distributed along the rivers, like direct discharge
of indus trial wastewater and leaching from municipal wa ste
dumps in the river to overrule the impact of building
densit y and it s correlated activities including direct dis-
charge of domestic wastewater, open defecation, and
overflow of pit latrines. This obser vation differs from
studies conducted in China by Deng et al. (2015) that
reported the quality of rivers decrease with urbanization
gradient, suggesting river pollution in our study is domi-
nated by specific point sources. Another explanation may
be that the urbanization gradients in the Chinese studies
were larger than in this study, where the catchments in
all cases were to some extent urbanized.
Thus, identifying the relation between river water qual-
ity and level of urbanization is key to formulate appropri-
ate environmental management tools related to water
and the environment that accommodates the continuous
and fast rates of urbanization in developing countries.
This is a roadmap for environmentalists to consider land-
use patterns, domestic and industrial pollution loads,
agricultural practices, population density, hydrological
aspects and level of urbanization when studying river
water quality which may help to realize the need for
preventive measures.
According to UNEP (1997), water quality problems have
attracted increasing attention especially in developing coun-
tries where environmental protection are now becoming a
major obstacle for sustainable development. Consequently,
water pollution control is clearly one of the most critical
issues that need urgent and informed action to prevent the
already stressed water resources from contamination.
For example, studies conducted in UK by Groffman et
al. (2002) and in USA by Neal and Robson (2000) report
that vegetated river banks reduced the reach of pollut-
ants into rivers. Conversely, due to encroachment of river
banks, this buffer mechanism is likely to have been offset
in the case of the three rivers studied in Addis Ababa.
The present study highlights the need to address point
source pollution like industries, garages, and animal pro-
duction sites, preferably with decentralized treatment
systems to prevent direct discharge to rivers (Massoud
et al. 2009). Regarding the non-point sources originating
from households without controlled emptying latrines, or
practicing open defecation, improved sanitation should
be prioritized. Here, separation of greywater and latrine
would be recommended, e.g. using composting toilet s
(Zavala et al. 2005) because flush-based toilets put pres-
sure on water resources and pollutes the receiving water
body as already witnessed today in Addis Ababa.
For the non-point pollution sources, which mainly reach
the rivers with the stormwater runoff, less focus is needed
if greywater and latrine could be managed in closed sys-
tems, as most of the expected nutrient load will then
be removed. Buffer zones, also along agricultural land,
would be an additional recommended measure
(Anbumozhi et al. 2005). The apparent load of heavy
metals, especially Cr, could be reduced if a greening pro-
gramme could be realized, ensuring vegetation cover of
bare soil (Wise, 2008; Barbosa et al. 20 12).
The prac tical way forward concerning the application of
the present study is to inform water and environmental
managers through designing cleanup plans, and help arrest
the main pollutant sources. Obser ving the state of water
quality in rivers and identifying the degree of water pollu-
tion helps to understand the current situation and help
reduce public health, environmental and economic impacts
of water pollution. Subsequently, it enables environmentalists
to establish appropriate management approaches towards
achieving sustainable water resources protection. This can
be achieved through reduction of the proportion of water
pollution originating from diffuse sources (e.g. agricultural
use of fertilizers, stormwater) which cannot be controlled
by the end-of-pipe approaches. The knowledge of the source,
concentration and type of pollutant s help identify the need
for pollution control measures, assist environmental regula-
tors in targeting the most significant problems and to assess
the necessity for making changes to legislative provisions.
Informed and enforceable water pollution control standard
is fundamental for its effectiveness.
Conclusion
Based on wet and dry season sampling of water from
three rivers in the Little Akaki catchment of Addis Ababa,
we conclude that the rivers have a poor water status
when compared to international standards. This includes
low oxygenation, too high salt concentrations and ten-
dency to too alkaline pH, high nutrient status of both N
and especially P, and severe heavy metal contamination
with Cr and Cu. Even when compared to the less strict
Ethiopian standards phosphate is still problematic, and
in two of the three rivers also Cr and Mn. When compar-
ing with the available data from other studies conducted
previously in Addis Ababa similar ranges were found for
most parameters, although some were much higher like
EC and Zn.
Direct discharge of untreated industrial wastewater
from a number of industries stretched along the rivers,

D. Adugna et al.Stormwater impact on water quality of rivers
Water and Environment Journal 33 (2019) 98–110 © 2018 CIWEM. © 2018 CIWEM.108
including tanneries, breweries and garages, supple-
mented with direct discharge of untreated municipal
wastewater serving 56% of the households are assumed
to constitute the main year-round contamination sources.
Open defecation and leaching of solid waste dumped
directly in the rivers is also considered significant. During
the wet season the river water quality is slightly improved,
but still often exceeding the international standards and
in case of P, Cr and Mn also exceeded the Ethiopian
limit.
Although stormwater dilutes the contaminants and
brings more oxygen, there seems to be a number of
additional non-point sources activated during rain. We
point to pit latrines overflowing and leaching from agri-
culture along river banks contributing nutrients, and
construction sites in the growing city contributing heavy
metals, especially Cr.
No strong correlation with urbanization gradients on
the level of pollutant for rivers passing through densely,
moderately and relatively less urbanized parts of the city
could be found, and no strong correlation when compar-
ing upstream and downstream sections of same river.
This might be due to point-sources being located along
the full stretch of all three rivers and the fact that all
catchments show some degree of urbanization.
It is recommended to target industrial wastewater with
decentralized treatment systems, municipal wastewater
with a treatment plant, and to implement improved sani-
tation for all households not connec ted to a sewage pipe.
Re-introduction of vegetated riverbanks is also recom-
mended as buffer zones. We also recommend the Ethiopian
standard to be revised as there are huge gaps with the
international standards considered from the US, EU and
Australia-New Zealand.
Acknowledgements
This research was made possible through financial support
from DANIDA to the strategic project ‘Water Resilient Green
Cities for Africa.’ We would also like to thank Mr. Kassahun
Tessema for his assistance during collecting data and
conducting laboratory analysis at the limnology laboratory
in the college of Science, Addis Ababa University.
To submit a comment on this article please go to http://
mc.manuscriptcentral.com/wej. For further information please see the
Author Guidelines at wileyonlinelibrary.com
References
Addis Ababa Bureau of Finance and Economic Development.
(2013) Socio-economic Profile of Addis Ababa for the
year 2013, Addis Ababa.
AAWSA. (2011) Business Plan from 2011 to 2020: Final
Report. Addis Ababa: Addis Ababa Water and Sewerage
Authority.
Alemayehu, T., Waltanigus, S. and Tadesse, Y. (2003)
Groundwater Vulnerability Mapping of Addis Ababa
Water Supply Aquifers. Addis Ababa.
Alemayehu, T. (2001) The Impact of Uncontrolled Waste
Disposal on Surface Water Quality in Addis Abeba.
SINET, 24, 93–104.
Anbumozhi, V., Radhakrishnan, J. and Yamaji, E. (2005) Impact
of Riparian Buffer Zones on Water Quality and Associated
Management Considerations. Ecol Eng., 24(5), 517–523.
ANZECC. (2000) Australian and New Zealand Guidelines for
Fresh and Marine Water Quality. Canberra: Australian
and New Zealand Environment and Conservation Council
of Australia and New Zealand. p. 1–103.
APHA-American Water Works Association, the Water and
Environment Federation. (2005) Standard Methods for
Examination of Water and Wastewater, 21st edition.
Washington, DC: American Public Health Association.
Awoke, A., Beyene, A., Kloos, H. and Goethals, P.L. (2016)
River Water Pollution Status and Water Policy Scenario in
Ethiopia: Raising Awareness for Better Implementation in
Developing Countries. Envir. Mangt., 58(4), 694–706.
Astel, A., Tsakovski, S., Barbieri, P. and Simeonov, V. (2007)
Comparison of Self-Organizing Maps Classification
Approach with Cluster and Principal Components
Analysis for Large Environmental Data Sets. Wat. Res.,
41(19), 4566–4578.
Barbosa, A.E., Fernandes, J.N. and nad David, L.M., (2012)
Key Issues For Sustainable Urban Stormwater
Management. Wat. Res., 46(20), 6787–6798.
Belete, D.A. (2011) Road and Urban Stormwater Drainage
Network Integration in Addis Ababa: Addis Ketema
Sub-city. J. Eng. Technol. Res., 3(7), 217–225.
Beyene, A., Addis, T., Kifle, D., Legesse, W., Kloos, H. and
Triest, L. (2009) Comparative Study of Diatoms and
Macroinvertebrates as Indicators of Severe Water
Pollution: Case Study of the Kebena and Akaki Rivers in
Addis Ababa. Ethiopia. Ecol. Indic., 9, 381–392.
Butler, D., Davies, J. and (2nd,, s (Ed.s) (2004) Urban
drainage. London and New York: Taylor and Francis.
Carpenter, S.R., Caraco, N.F., Correll, D.L., Howarth, R.W.,
Sharpley, A.N. and Smith, V.H. (1998) Non-point Pollution
of Surface Waters with Phosphorus and Nitrogen. Ecol.
Applications, 8(3), 559–568.
Claussen, U., Müller, P. and Arle, J. (2012) WFD CIS
ECOSTAT WG: Comparison of Environmental Quality
Objectives, Threshold Values or Water Quality Targets
Set for the Demands of the European Water Framework
Directive. Internal Report.
Chang, H. (2005) Spatial and temporal variations of water
quality in the Han River and its tributaries, Seoul, Korea,
1993–2002. Water, Air, & Soil Pollution, 161(1-4),
267–284.
Cohn, N. (1998) What’s Working on Working Rivers: A
Handbook for Improving Urban Rivers: Examples from.
Chicago Area Rivers.

Stormwater impact on water quality of riversD. Adugna et al.
Water and Environment Journal 33 (2019) 98–110 © 2018 CIWEM. 109
CSA, Federal Democratic Republic of Ethiopia Central
Statistical Agency. (2013) Population Projection of
Ethiopia for All Regions from 2014–2017.
Deng, X., Xu, Y., Han, L., Song, S., Yang, L., Li, G. and
Wang, Y. (2015) Impacts of Urbanization on River
Systems in the Taihu Region. China. Water, 7(4),
1340–1358.
Ding, J., Jiang, Y., Fu, L., Liu, Q. and Peng, Q. (2015)
Impacts of Land Use on Surface Water Quality in a
Subtropical River Basin: A Case Study of the Dongjiang
River Basin. Southeastern China. Wat., 7(8), 4427–4445.
Ethiopian Environmental Protection Authority (EPA) and the
United Nation Industrial Development Organization.
(2003) Standards for Industrial Pollution Control in
Ethiopia: Prepared Under the Ecological Sustainable
Development Project – US/ETH/99/068/Ethiopia.
Eremektar, G., Tünay, O., Orhon, D. and Gönenç, E. (1995)
The Pollution Profile of Alcohol Distilleries Treating Beet
Sugar Molasses. Wat. Sci. Technol., 32(12), 181–188.
EU. (2013) Directive 2013/39/EU of the European
Parliament and of the Council of 12 August 2013
Amending Directives 2000/60/EC and 2008/105/EC as
Regards Priority Substances in the Field of Water Policy.
Brussels: European Union.
Groffman, P.M., Boulware, N.J., Zipperer, W.C., Pouyat, R.V.,
Band, L.E. and Colosimo, M.F. (2002) Soil Nitrogen Cycle
Processes in Urban Riparian Zones. Envir. Sci. Technol.,
36(21), 4547–4552.
Ho, K.C. and Hui, K.C.C. (2001) Chemical Contamination of
the East River) and its Implication on Sustainable
Development in the Pearl River Delta. Envir. International,
26(5), 303–308.
Joshi, D.M., Kumar, A. and Agrawal, N. (2009) Studies on
physicochemical parameters to assess the water quality
of river ganga for drinking purpose in Haridwar District.
Rasayan Journal of Chemistry, 2(1), 195–203.
Kannel, P.R., Lee, S., Kanel, S.R. and Khan, S.P. (2007)
Spatial-temporal variation and comparative assessment of
water qualities of urban river system: a case study of
the river Bagmati (Nepal). Environmental Monitoring and
Assessment, 129(1–3), 433–459.
Karbassi, A.R., Nouri, J. and Ayaz, G.O. (2007) Flocculation
of Cu, Zn, Pb, Ni and Mn during mixing of Talar river
water with Caspian sea water. International Journal of
Environmental Research, 1(1), 66–77.
Karn, S.K. and Harada, H. (2001) Surface water pollution in
three urban territories of Nepal, India, and Bangladesh.
Environmental Management, 28(4), 483–496.
Kayhanian, M., Vichare, A., Green, P.G. and Harvey, J.
(2009) Leachability of dissolved chromium in asphalt and
concrete surfacing materials. Journal of Environmental
Management, 90(11), 3574–3580.
Massoud, M.A., Tarhini, A. and Nasr, J.A. (2009)
Decentralized approaches to wastewater treatment and
management: applicability in developing countries.
Journal of Environmental Management, 90(1), 652–659.
Mazhindu, E., Gumbo, T. and Gondo, T. (2010) Living with
environmental health risks – the case of Addis Ababa.
Ecohydrology & Hydrobiology, 10(2), 281–286.
Ministry of Environment and Food of Denmark. (2015)
Departmental Order Concerning Establishment of
Environmental Goals for Streams, Lakes, Transition
Waters, Coastal Waters and Groundwater; Bek. Nr. 1070
af 09/09/2015, Copenhagen.
Mwinyihija, M. (2010) Main Pollutants and Environmental
Impacts of the Tanning Industry. Ecotoxicological
Diagnosis in the Tanning Industry. New York: Springer,
pp. 17–35.
NMA. (2016) Monthly Rainfall Data of Addis Abeba. Addis
Ababa: National Metrology Agency of Ethiopia.
Nazafpour, S., Alkarkhi, A.F.M., Kadir, M.O.A. and
Nazafpour, G.D. (2008) Evaluation of Spatial and
Temporal Variation in River Water Quality. International
Journal of Environmental Research, 2(4), 349–358.
Neal, C. and Robson, A.J. (2000) A summary of river water
quality data collected within the land–ocean interaction
study: core data for Eastern UK rivers draining to the
North Sea. Science of The Total Environment, 251,
585–665.
Pesce, S.F. and Wunderlin, D.A. (2000) Use of water quality
indices to verify the impact of Córdoba City on Suquı́a
river. Water Research, 34(11), 2915–2926.
Raad, R., Margane, A. and Saade, E. (2012) Environmental
Risk Assessment of the Fuel Stations in the Jeita Spring
Catchment – Guidelines from the Perspective of
Groundwater Resources Protection [Online]. Available at:
https://www.bgr.bund.de/EN/Themen/Wasser/Projekte/
abgeschlossen/TZ/Libanon/spec_rep_14.pdf [Accessed 3
July 2017].
Regassa, N., Sundaraa, R.D. and Seboka, B.B. (2011)
Challenges and Opportunities in Municipal Solid Waste
Management: The Case of Addis Ababa City.
Environmental Management, 33(3), 179–190.
Sangodoyin, A.Y. and Agbawhe, O.M. (1992) Environmental
study on surface and groundwater pollutants from
abattoir effluents. Bioresource Technology, 41(3),
193–200.
Schmoll, O. (eds) (2006) Protecting Groundwater for Health:
Managing the Quality of Drinking-water Sources. World
Health Organization.
Strandberg, C.H. (1971) Water Pollution. In: Smith, G.H. (Ed.)
Conservation of Natural Resources. New York: Wiley, pp.
189–219.
Temesgen, N. (2009) Assessment of the physicochemical
parameters of selected rivers in the city of Addis Ababa.
Thesis, Addis Ababa University (AAU).
Tolla, B. (2006) Physicochemical characteristics and pollution
levels of trace metals in the Little Akaki and Big Akaki
rivers. Thesis, AAU.
UNEP. (1994) Tanneries and the Environment – A Technical
Guide (No. 4): Technical Report.

D. Adugna et al.Stormwater impact on water quality of rivers
Water and Environment Journal 33 (2019) 98–110 © 2018 CIWEM. © 2018 CIWEM.110
UNEP. (1997) Water Pollution Control - A Guide to the Use
of Water Quality Management Principles [Online].
Available at https://www.earthprint.com [Accessed 23
August 2017].
USEPA. (1986) Quality Criteria for Water. Washington, DC:
US Environmental Protection Agency.
WHO. (2003) Ammonia in Drinking Water. Background
Document for Preparation of WHO Guidelines for
Drinking Water Quality. Geneva: World Health
Organization (WHO/SDE/WSH,03.04/1).
Wise, S. (2008) Green infrastructure rising. Planning, 74(8),
14–19.
Woli, K.P., Nagumo, T., Kuramochi, K. and Hatano, R. (2004)
Evaluating river water quality through land-use analysis
and N budget approaches in livestock farming areas.
Science of the Total Environment, 329(1), 61–74.
Williams, P.E.V. (1995) Animal production and European
pollution problems. Animal Feed Science and Technology,
53(2), 135–144.
Zavala, M.L., Funamizu, N. and Takakuwa, T. (2005)
Biological activity in the composting reactor of the
bio-toilet system. Bioresource Technology, 96(7),
805–812.