International Journal of Disaster Risk Reduction 27 (2018) 105–117

Contents lists available at ScienceDirect

International Journal of Disaster Risk Reduction

journal homepage: www.elsevier.com/locate/ijdrr

A multi-faceted analysis of annual flood incidences in Kumasi, Ghana T

a,b,⁎ b a,b a,b
Paul Amoateng , C. Max Finlayson , Jonathon Howard , Ben Wilson
a
School of Environmental Sciences, Charles Sturt University, P O Box 789, Albury, NSW 2640, Australia
b
Institute for Land, Water and Society, Charles Sturt University, P O Box 789, Albury, NSW 2640, Australia

A R T I C L E I N F O A B S T R A C T

Keywords: Flood disasters remain a serious development challenge in cities in developing countries. In many cities, the
Flooding frequency and intensity of flood disasters have increased spatially and temporally and cause more casualties and
Land use planning economic losses than any other natural catastrophe. This study used multiple qualitative research approaches to
Waterway explore the nature and causes of, as well as proposes options for addressing future disasters in Kumasi, the fastest
Rainfall
growing city in Ghana. The findings show that disastrous floods now occurred on an annual basis with at least 4
Riverine flood
per year, and are becoming increasingly severe. The floods now more regularly following moderate to high
rainfall events as a consequence of uncontrolled occupation of inland water areas especially rivers and flood-
plains by urban physical developments. Despite a well-structured policy framework, interventions tended to be
reactionary and failed to address the underlying causes of the floods. Indeed, they sometimes aggravated the
impacts. The study suggests that it is critical for city managers and planners to enforce a “no development” zone
along rivers and floodplains through integrated land use planning and relocation of settled encroachers. The
enforcement could rely on participatory processes and institutional collaborations that focus on positive out-
comes such as reducing and/or avoiding flood disasters and sustaining and/or improving liveability of the city.

1. Introduction interventions assessments ([16,31,49]). Most authors (e.g., [10,16,29])

Flooding continues to pose enormous development challenges to
both developed and developing countries [19,31]. Flooding is generally
described to occur when water overflows its defined channel or course
due to heavy rainfall events, sea level rise, storm surges, coastal de-
gradation, glacial melt, snow melt and reduced ground infiltration
[16,31,36]. Recent studies indicate that the frequency and intensity of
flooding and its associated devastating impacts on the wellbeing of
exposed and vulnerable populations has been on the rise in urban areas,
especially in developing countries [18,27,28]. Such flood disasters
disrupt the prosperity, safety and amenity of people in these countries.
For example, flood events in Maputo the capital of Mozambique in 2000
caused hundreds of deaths, destroyed 150,000 homes, rendered
500,000 people homeless and jobless and resulted in over 17,000
cholera cases [38]. Consequently, there is a general consensus among
researchers that flood disasters cause more casualties and economic
losses in the form of loss of lives and livelihoods, damage to property
and infrastructure, spread of diseases and insecurity among others, than
many other similar disasters [31,50].
Given flooding is caused by a diversity of factors, researchers have
broadly grouped floods into four different types as riverine flood, urban
flood, flash flood and coastal flood for the purposes of flood risk and
identify frequent and high magnitude flood disasters in urban areas of
developing countries as riverine floods. That is these floods occur when
rivers overflow their banks and cover built-up areas for extended per-
iods. These floods, which were hitherto predominantly considered
natural events, have increasingly become a human-induced phenomena
with their frequency and extent also increasing [27,37]. It is asserted
that the cause of this change is the development of human settlements
in rivers and floodplains which leads to the eventual loss and frag-
mentation of the natural system and, creates permanent and/or inter-
mittent flood damage zones [10,17,29]. In such situations, the occur-
rence of riverine flood disasters becomes imminent. Unfortunately,
these claims have not been studied in-depth in many urban situations,
especially in developing countries, thus undermining understanding of
and/or attempts at addressing the problem.
Traditional flood control methods entail structural measures such as
storm drains, embankments or levées, floodways, and dams [26,34,44].
These structural approaches, which aim to eliminate floods, do not only
result in changes in flow regimes and fragmentation of surface water
networks, but also result in increased flood incidences in many cases,
particularly in downstream locations [26]. As a result, integrated flood
management, which has the objective of flood avoidance, flood toler-
ance and flood resilience has become the preferred management

⁎
Corresponding author at: School of Environmental Sciences, Charles Sturt University, P O Box 789, Albury, NSW 2640, Australia.
E-mail address: pamoateng@csu.edu.au (P. Amoateng).

http://dx.doi.org/10.1016/j.ijdrr.2017.09.044
Received 18 December 2016; Received in revised form 26 June 2017; Accepted 23 September 2017
Available online 27 September 2017
2212-4209/ © 2017 Elsevier Ltd. All rights reserved.
P. Amoateng et al.
approach in recent times [10,21,31]. This holistic approach combines
structural and non-structural measures (e.g., land use planning and
zoning, natural inland water systems conservation, institutional and
policy frameworks, forecasting and early warning systems and flood
insurance) in an iterative manner to reduce susceptibility to flood da-
mage, flood impacts and the recovery time [19,26,31]. While integrated
flood management may seem both logical and ideal, in developing
countries there is often a lack of insight and information on relevant
issues such as the causes, types, nature, characteristics of victims and
impact areas of the floods, that prevents it from being implemented.
Ghana ranks high among the African countries that experience flood
disasters, which threaten economic growth and development [8,19,41].
Floods of different types and degrees of intensity occur, but widespread
flooding in large cities is both increasing in frequency and scale in re-
cent times [12,32]. This is particularly true in major cities such as
Accra, Kumasi, Sekondi-Takoradi and Tamale, which have become the
points of concentration of the country's burgeoning urban population –
urbanisation level exceeded 50% for the first time in history in 2010
[22]. In these cities, flood disasters occur annually during the rainy
seasons, and are more pronounced in the informal settlements, which
often lack adequate economic and physical capacity to deal with the
effects [19,41]. Unfortunately, the actual causes of the flooding are
largely unknown as they are linked to multiple unsubstantiated factors
that in certain cases combine to aggravate the consequences. Frequently
attributable causes include natural factors such as climate change,
physiographic conditions, sea tides as well as human factors such as
land cover/use change and poor urban planning ([2,19,32,40];
[12,24,41]). As a result, while the annual flood disasters are widely
acknowledged and expected as a consequence of extensive reporting in
the news media, the underlying causes are poorly supported by evi-
dence and, hence improperly understood and /or addressed by both
academics and practitioners.
2. Study area
Kumasi, the second largest city in Ghana, is located within the
moist, semi-deciduous south-east ecological zone in south-central
Ghana between latitudes 6°38′N and 6°45′N and longitudes 1°32′W and
1°41′W (Fig. 1). The city has a bimodal rainfall regime of major and
minor rainy seasons, which occur between the months of March and
July and September and October respectively with an overall mean
annual rainfall of 1382 mm, with June experiencing the highest mean
monthly rainfall of 214 mm. As one of the fastest growing cities in
Ghana, Kumasi has an estimated population of 2,035,064, which is
growing at a rate of 5.5% per annum, more than double the national
average of 2.5% [22]. The city has a total land area of 254 km2, over
80% of which is developed for various urban land uses. Given its geo-
graphic location, the city once had an abundance of “greenery” (ve-
getation cover) and inland water features such as rivers, wetlands, ri-
parian zones, open space and parks, which enhanced its hydrologic
conditions and scenic beauty. Unfortunately, pressures from the city's
recent dramatic and spontaneous population growth and physical ex-
pansion have adversely affected these natural land covers, especially
rivers and floodplains, and created hydrologic problems such as large
impervious surfaces and increased runoff volumes. As a result, de-
structive flood disasters affect many suburbs of the city after torrential
rains, especially in the rainy seasons.
3. Methods
This study employed a combination of qualitative data collection
methods including semi-structured interviews, focus group discussions
and direct observations for in-depth examination of the flood incidence
and management in Kumasi [6]. These methods were used to collect
data from local people and relevant institutions (Table 1), through
fieldwork conducted in June–September 2013 in Kumasi, Ghana. The
106
International Journal of Disaster Risk Reduction 27 (2018) 105–117
data collection activities followed a systematic process with the activ-
ities at each stage being partially informed by that of the preceding
stage [47].
First, semi-structured interviews were conducted with re-
presentatives of five flood and urban management related institutions
in Kumasi, including the National Disaster Management Organisation
(NADMO), Hydrological Services Department (HSD), Town and
Country Planning Department (TCPD), Ghana Meteorological Agency
(GMA), and Friends of Rivers and Water Bodies (FRWB). The institu-
tional officials interviewed were mainly institutional heads and/or
deputies and were purposefully sampled based on their knowledge and
interest in the research subject, which ensure acquisition of accurate
and reliable information.
Second, at the community level, semi-structured interviews were
conducted with forty property owners within the 100 m zone of rivers
and floodplains in eight purposefully selected flood-prone suburbs of
Kumasi. The suburbs were Kronum, Asuoyeboa, Dichemso, Moshie
Zongo, Atonsu, Ahinsan, Kokode and Adiembra and were selected
based on the high intensity of physical development in and along rivers
and floodplains, and frequent flood incidences that characterise them,
as informed by the institutional interviewees, as well as observed
during preliminary visits to the numerous flood-prone suburbs in the
city (Fig. 2). Further, the purposive sampling method was used to select
five residents from each of the eight suburbs based on the proximity of
their properties to the rivers, as well as the type of properties or ac-
tivities, which ensured the use of an information-rich sample. The
property owners interviewed also had different socio-economic back-
grounds and status, which enhanced the diversity and quality of the
responses obtained, and ultimately the study findings. In addition,
physical evidence of the nature and causes of the floods, such as
floodwater inundation marks and houses in waterways and were per-
sonally observed and photographed.
Finally, focus group discussions were organised in two of the sub-
urbs used for data collection – Moshie-Zongo and Asuoyeboah – to elicit
collective views on the flood situation and management in the city.
There were ten participants per focus group discussion, comprising
seven participants from the property owners’ semi-structured inter-
views and three purposefully selected community leaders. The focus
group discussions and semi-structured interviews provided detailed
information on the local people's perspectives and understanding of the
frequency, intensity, causes, and control of flood disasters in Kumasi in
a flexible manner [6,47]. The different qualitative data collection
methods enabled the collection of expansive, valid and reliable data for
examining flood problems and interventions in the city. The data from
each of the methods were cross-referenced to improve the usefulness of
the study findings.
In addition to these qualitative data, secondary quantitative and
spatial data were obtained from the relevant agencies and sources.
Quantitative data on reported flood incidences in Kumasi from 2009 to
2013 were obtained from the National Disaster Management
Organisation while mean daily rainfall records for Kumasi from 1961 to
2013 were acquired from the Ghana Meteorological Agency (GMA).
Data description provided by GMA indicates that the rainfall data were
derived from two weather stations located in the northern and southern
parts of the study area. For the spatial data, Landsat Thematic Mapper
(LTM) images with 30 m resolution and Advanced Spaceborne Thermal
Emission and Reflection Radiometer (ASTER) images with 15 m re-
solution of Kumasi for the years 1985, 2000 and 2013 were obtained
from the Forestry Commission of Ghana (FCG). The satellite images
were supplemented with land use and topographical maps of Kumasi, as
well as ground-truthed spatial information collected with a handheld
Global Position System (GPS) to improve the accuracy of the research
results [20,29].
The transcripts of the semi-structured interviews and focus group
discussions were analysed using the qualitative data analysis software
package, NVivo 10 [46]. The data were coded and categorised to
P. Amoateng et al. International Journal of Disaster Risk Reduction 27 (2018) 105–117

Fig. 1. The study location.

Table 1 derive land use/cover maps of Kumasi as well as detect the losses and
Research participants. gains in the land cover/use types with particular focus on the rivers and
floodplains.
Research activity Male Female Total
participants
Number % Number % 4. Results and discussion

Institutional semi-interviews 3 60 2 40 5 4.1. Flood incidence and intensity in Kumasi

Property owners’ semi- 27 78 13 32 40
structured interviews
Traditional leaders semi- 2 100 – – 2 Analysis of flood occurrence records from 2009 to 2013 docu-
structured interviews
Focus Group Discussions 12 60 8 40 20
Overall total 44 66 23 34 67
identify patterns, themes and ideas that point to understanding river
pollution and sanitation in Kumasi. In addition, axial coding was done
in order to establish relationships between the information categories
and sub-categories. The thematic categories and sub-categories were
oriented towards the main research aspects, namely, frequency and
intensity of flood events, factors influencing flood incidences and flood
management efforts. The final identified thematic categories and sub-
categories, as well as established relationships, were presented as nar-
ratives (storylines) while maintaining the perspective of the research
participants. Also, the quantitative data such as rainfall and flood re-
cords were analysed using various descriptive statistics to reveal the
trends and patterns of these variables. In particular, the Mann-Kendall
test (Z) and Sen's slope (Q) were used to detect and evaluate trends in
rainfall amounts and heavy rainfall events (determined as rainfall ≥
27 mm/day, based on the minimum rainfall magnitude that triggered
flooding in Kumasi between 1961 and 2013) in Kumasi at the standard
confidence level of α = p = 0.05. The quantitative data analyses were
performed with the aid of the Statistical Package for Social Scientists
(SPSS) 16 and Microsoft Excel 2010, and were also used to generate
tables and charts. The ArcGIS 10.1 Desktop software was used to ana-
lyse, process and classify the satellite images into three land use/cover
categories as urban land, vegetation and rivers and floodplains, to
mented by NADMO indicates Kumasi experienced 28 flood events over
the period. The data reveals that flooding has become a perennial
phenomenon in the city with at least five incidences per year, a number
that is gradually increasing over the years (Fig. 3). The floods mostly
occurred at the onset and peak of the rainy seasons. As presented in
Fig. 4, June the peak rainfall month has recorded the highest floods
incidences, although considerable flood events are being experienced in
March, at the onset of the rainy season.
Interviews and direct observations found that these flood occur-
rences were predominantly confined to suburbs with extensive devel-
opments along rivers and in floodplains in the city (Fig. 5). Amoako and
Cobbinah [7] described most of these suburbs as the informal settle-
ments or slum areas of the city; given widespread unauthorised sub-
stantive structures are common and there are woefully inadequate
urban amenities. Both institutional and property owner interviewees
indicated that floods occur in these suburbs whenever there is moderate
to heavy rainfall or a storm. Some of the interviewees said that:
"In fact there are a lot of flood prone areas in Kumasi but they are mainly
the slum communities where developments have encroached on rivers
and floodplains. I can mention communities like Aboabo, Dichemso,
Atonsu, Ahinsan and Moshie-Zongo among others which are char-
acterised by encroachment on rivers and floodplains and as such ex-
perience flooding every rainy season…In these communities, flooding is
an annual disaster….” HSD, June 2013.
"Floods are very pronounced in certain parts of the city and they [floods]

107
P. Amoateng et al. International Journal of Disaster Risk Reduction 27 (2018) 105–117

Fig. 2. Map of Kumasi showing the selected data collection sites.

8 January
September
4%
7 11%
6 March
17%
Flood Incidences

5
July
4 11%

0 April
2009 2010 2011 2012 2013 11%
Year

Fig. 3. Annual flood incidence in Kumasi (2009–2013).

Source: NADMO, Kumasi [39]

May
occur many times in the year during the rainy season” NADMO, June June 11%
2013. 35%

Fig. 4. Proportion of monthly flood incidence in Kumasi (2009–2013).

“We experience flooding in this community with the least rainfall. It Source: NADMO, Kumasi [39]
occurs several times in a year” Property Owner Interviewee [POI] 1,
July 2013.
take as long as seven days to completely recede and the floodplain to
dry, and for daily activities to return to normalcy (Fig. 6b), suggesting
In particular, the property owners, who have all had their homes
that the flood victims live under flood conditions for approximately a
flooded, described the floods experienced in the city as severe and
month in a year. Some of the interviewees explained the flood condi-
disastrous. They explained that floodwaters usually rise as high as the
tions as follows:
rafters of their houses, with some houses being completely submerged
(Fig. 6a). "The floodwater occupies the whole house and enters the rooms so we are
The interviews and direct observations further revealed that the unable to come here until it recedes…It can take more than a week before
time taken for the floodwater to recede depended on the location of the this place becomes completely dry " POI 24, August 2013.
houses. Property owners who resided far from the river channels
"Whenever the floods occur we visit the communities and we see that the
claimed the floodwaters last for only 1–5 h(s) while those whose houses
water level is up to the windows or roofs of some houses…. When it
are very close to the river channels reported that the floodwaters re-
happens like that it takes almost a week before the water recedes but it
main for days. On average it was established that the floodwaters can

108
P. Amoateng et al. International Journal of Disaster Risk Reduction 27 (2018) 105–117

Fig. 5. Flood prone map of Kumasi.

Source: Researcher's Diary, (June–September 2013)

can take shorter period to recede in some places" HSD, June 2013. 4.2. Rainfall events and flood incidences

Rainfall data for Kumasi between 1961 and 2013 revealed a de-
Thus, it appears that Kumasi is experiencing a rising incidence of
creasing trend for the total annual rainfall at rate of 0.81 mm y−1, and
disastrous riverine floods, which are characterised by prolonged
culminating with an overall decrease of 42 mm over the period
floodwater inundation with increased problems of homelessness and its
(Fig. 7a), which is about 68% higher than the national decrease of
associated vulnerabilities by rendering the affected houses unin-
about 25 mm during 1951–2010 [28]. In addition, the rainfall pattern
habitable [1,10,29]. According to Campion and Venzke [12] and Islam
has a high inter-annual rainfall variability of 0.20 compared to the
et al. [29], riverine floods mostly affect settlements/developments that
global mean coefficient of variation of 0.27 [14]. Similarly, monthly
occupy water areas such as rivers and floodplains where water over-
rainfall for June, the peak rainfall month, revealed a decreasing trend of
flows are expected during rainy seasons. Regrettably, structures de-
0.28 mm y−1, and inter-annual variability of 0.40 (Fig. 7b). However,
veloped in such locations are usually substandard and unfit to with-
heavy rainfall events are increasing annually (Z = − 0.66, Q = 0), as
stand flood disasters, while the inhabitants often lack adequate
well as for the month of June (Z = 0.62, Q = 0) (Figs. 7c and d). This
economic and physical resources and capacity to combat the threats of
suggests an increasing occurrence of high intensity precipitations that
floods, as in the case of Kumasi [7,19,41]. This situation makes it im-
are usually concentrated in the short peak rainfall periods that trigger
perative to ascertain and possibly intervene in the underlying causes of
critical processes such as flooding in Kumasi [28]. As a result,
the flood problems in the city.

Fig. 6. Flood conditions in Kumasi.

Mark of floodwater level

a. Height of floodwater b. House inundated with floodwater days after

flooding

109
P. Amoateng et al. International Journal of Disaster Risk Reduction 27 (2018) 105–117

2500 400

350
2000
300
Rainfall (mm)

Rainfall (mm)
1500 250

200
1000
150

100
500
50
0 0
61

65

69

73

77

81

85

89

93

97

01

05

09

13

61

65

69

73

77

81

85

89

93

97

01

05

09

13
19

19

19

19

19

19

19

19

19

19

20

20

20

20

19

19

19

19

19

19

19

19

19

19

20

20

20

20
Year Year
a. Annual rainfall for Kumasi) b. Rainfall for the month of June in Kumasi

30 7

25 6
Rainfall Days

Rainfall Days
20

4
15
3
10
2
5
1

0 0
61

65

69

73

77

81

85

89

93

97

01

05

09

13

61

65

69

73

77

81

85

89

93

97

01

05

09

13
19

19

19

19

19

19

19

19

19

19

20

20

20

20

19

19

19

19

19

19

19

19

19

19

20

20

20

20
Year Year
c. Annual heavy rainfall events in Kumasi d. Heavy rainfall events for the month of June in Kumasi

Fig. 7. Historical analysis of rainfall pattern and trends in Kumasi (1961–2013).

individual rainfall events on the flood days were used to ascertain the Table 2
influences of rainfall on flood incidences in the city, especially given Rainfall and flood events in Kumasi (2009–2013).
Source: Adapted from NADMO [39], GMA [23].
that annual and monthly rainfall amounts are decreasing and char-
acterised by high variability. Flood day Rainfall on flood day (mm) Rainfall on day before flood day (mm)
Floods experienced in Kumasi between 2009 and 2013 were pre-
ceded by rainfall of varying magnitudes (Table 2). Floods occurred 09-Jun-09 27.8 3.1
during and/or after moderate to heavy rainfall events, with at least 11-Jun-09 60.4 0.1
15-Jun-09 43.3 0.0
27 mm of rainfall needed to trigger a flood in Kumasi. Based on this, 26-Jun-09 90.3 17.3
analysis of the daily rainfall data for Kumasi from 2009 to 2013 in- 04-Jul-09 51.4 12.8
dicated that 81 out of the 610 rainfall events that occurred during the 09-Jul-09 72.7 27.6
period were potential flooding rainfalls (i.e., ≥ 27 mm). In other words, 15-Jun-10 83.6 0.0
06-Jun-10 48.4 0.0
81 rainfall events should have resulted in floods in Kumasi between
12-Sept-10 28.3 9.9
2009 and 2013 if rainfall was the key cause of flooding in the city. 22-Sept-10 46.6 0.0
However, only twenty-four instead of eighty-one flood events oc- 23-Jan-11 33.6 2.0
curred in the city over the 5 year period. This smaller than expected 03-Mar-11 75.5 0.0
number of flood incidences suggests a diminishing and/or less sig- 07-Mar-11 45.0 0.0
19-Mar-11 40.2 0.0
nificant role of rainfall intensity in the occurrence of the recent floods in 03-Jun-11 30.9 0.0
the city. For example, the flood that was triggered by a 27 mm rainfall, 25-Mar-12 42.9 0.0
the least rainfall to have triggered flooding in the city, occurred in 19-Apr-12 38.4 0.0
2013, suggesting that intensity of a rainfall event is not necessarily a 08-May-12 31.1 5.6
02-Jun-12 57.6 0.8
critical factor in the floods experienced in the city. Nevertheless, this
20-Jun-12 76.1 0.0
situation indicates the crucial roles of other factors (e.g., development 28-Jun-12 35.6 0.0
in water areas) in causing flooding in the city, as was echoed by some of 15-Mar-13 27.0 0.0
the interviewees as follows: 13-Apr-13 27.6 0.0
17-Apr-13 30.4 0.2
"The least rainfall and we hear reports of flooding in the communities 17-May-13 43.7 0.0
along the rivers…and these occur several times in the year, especially at 27-May-13 33.5 0.0
04-Jul-13 36.8 0.0
the start and middle of the rainy season. The duration of the downpour
08-Sept-13 28.0 1.6
influences the floods, but I thinks there are other factors that are more
critical because in many cases the floods occur/affect communities where
it has not even rained there” GMA, June 2013. "There are times we don't expect floods because it has not rained heavily
and yet we experience floods in this area. Sometimes the rainfall is

110
P. Amoateng et al.
Fig. 8. Topographic and soil map of Kumasi. The soil associations are A: Bekwai-Akomadan-Oda Compound Association, B: Bekawi-Oda Compound Association, C: Bomso-Offin
Compound Association, D: Bomso-Suko Simple Association, E: Kumasi-Offin Compound Association, and F: Nhyanao-Yinkong Compound Association.
upstream, yet we suffer from floods, so the real causes of our problem is
something else than rainfall” POI 31, September 2013.
This suggests that recent flood events in Kumasi are complex phe-
nomena that are not just the result of unusual heavy rainfall events.
This is consistent with argument in the literature that although intense
rainfall events trigger and hasten flow of large runoff, the flood dis-
asters that follow are often strongly influenced by many different in-
teracting factors such as the condition of natural and engineered drai-
nage systems, as well as basin physiographic and edaphic conditions
[12,19,43]. Thus, findings from this research and other previous studies
(e.g., [2,32,43]) indicate that total rainfall and intense rainstorm events
cannot explain the recent flood disasters in Kumasi.
4.3. Topographic and soil conditions and flood incidences
Topography and soil conditions can influence the exposure to and
incidence of flash floods in an area because they affect the quantity and
extent of overland flow [41]. Analysis of Kumasi's topography and soil
using the Digital Elevation Model (DEM) reveals that the topography is
undulating, with lowlands and occasional hills interspersed across the
landscape (Fig. 8). Elevations range from 250 m to 300 m above sea
level, and slopes are generally gentle, mostly less than 5%. There are
flat topped interfluvial ridges with widths varying from 1500 m to
2500 m, which predominantly run in a north-southward direction,
causing the greater majority of the city's rivers and surface runoff to
flow in that direction. This irregular terrain is explained by Kumasi's
location within a forest dissected plateau physiographic region of
Ghana that is prone to advanced tertiary erosion, hence dominated by
dissected landforms [25,30].
The predominant soil types in Kumasi belong to the Forest Ochrosol
soil group, which has nearly impermeable sub-horizons near the surface
[3]. The soils are divided into six soil associations (Fig. 8), and are
deeply weathered with high clay accumulation and silty clay loam
texture. As a result, the soils have high base saturation and are mod-
erately slow to slowly permeable, which leads to fairly high runoff yield
111
International Journal of Disaster Risk Reduction 27 (2018) 105–117
[3,13]. The high runoff potential of the soils is attributed to their for-
mation from weathered lower Birimian system metamorphosed sedi-
ments of Middle Precambrian Rock origin, which has little or no pri-
mary porosity [3,33].
Nonetheless, findings from previous studies (e.g., [30,43]), and in-
terview results for this study indicate that the undulating landform of
Kumasi makes the city generally free from the danger of flooding in the
topographical sense. The interview results, in particular, emphasised
that the undulating terrain is even helpful in mitigating the flood dis-
asters that occur in the city. The institutional and some property owners
interviewed opined that while runoff cannot flow upward to the rela-
tively higher grounds to cause floods, the runoff generated on those
higher grounds and other lowland areas in the city is decelerated by,
and accumulated in the valleys, thus reducing the possibilities of floods.
Accordingly, Owusu-Ansah [43] using cross section analysis found near
complete absence of flood incidences in areas of the city that are
characterised by the steepest slopes. The TCPD interviewee affirmed
this finding by stating that:
“As I was saying…Kumasi has a good topography as far as flooding is
concerned. The landform is not a flat but undulating and that is why you
hardly see the whole city flooding. We have uphill and valleys here and
there, so you see the runoff flowing in various directions in relatively
small proportions according to the terrain of Kumasi and that makes it
difficult for floods to occur as a result of excessive runoff” TCPD, June
2013.
Additionally, although the soft and gritty topsoil, as well as the low
absorptive capacity of the soils in Kumasi makes the city susceptible to
flash floods [30], results from this study indicate that the recent flood
occurrences do not necessarily correspond with soil conditions. For
example, a closer examination of Table 2 revealed that none of the flood
days was preceded by even a potential rainfall event that could have
created antecedent conditions such as saturated soil and risen water
table that favour the occurrence of floods associated with excessive soil
moisture [12,41]. Further, many of the interviewees argued that the
soil conditions have remained unchanged over the years and as such
P. Amoateng et al.
could not be the cause of the recent floods if they did not cause such
disasters in the past. The official of HSD interviewed said that:
“I agree that the impermeable nature of the soil in Kumasi generates a lot
of runoff but I don’t think that is the reason why we experience floods
here. The soil has been like that for long time but there were no floods,
the runoff were flowing away freely but these days people have developed
in water ways so the runoff enters their houses and they complain they
are suffering from floods. No runoff would’ve entered those houses had
they not been constructed in the low-lying areas” HSD, June 2013.
These findings are consistent with reports by previous studies (e.g.,
[41,43]) that physiographic features do not significantly influence the
recent flood frequency in Ghanaian cities, particularly in Kumasi and
Accra. However, in Kumasi, the low soil porosity and undulating terrain
can partly explain seasonal water logging in valleys, as well as intensive
erosion of top soil in many parts of the city [5]. In addition, the flat
terrain which slows movement of runoff could promote deposition of
silts and waste materials in rivers and storm drains and raise their beds
above floodplains, which eventually induce flooding as discussed in the
next section. This situation therefore suggests the significant influence
of the conditions of natural (e.g., rivers and floodplains) and artificial
(storm drains) drainage systems on flooding in Kumasi as discussed in
the subsequent sections.
4.4. Loss of rivers and floodplains and flood incidences
Kumasi has witnessed rapid but unregulated physical expansion in
recent years, with the built-up areas increasing by approximately
162 km2 from 43 km2 in 1985 to 205 km2 in 2013 at an annual rate of
5.8 km2 y−1 (Fig. 9). Besides creating large expanse of impervious
surfaces (developed and bare lands), which have resulted in large
runoff generation (e.g., average annual runoff is 683 and increasing at
10 mm y−1 [9] in the city, the physical expansion has also involved
extensive replacement of water areas (rivers, floodplains, and riparian
zones) by urban land uses. An overlay analysis revealed that about 9%
(18 km2) of the city's built-up area in 2013 was converted from rivers
and adjoining floodplains leading to a drastic reduction in the extent of
these inland water systems by 83% from 38 km2 in 1985 to 6 km2 in
2013. The rivers and floodplains were illegally gradually encroached
and/or instantaneously land filled for various physical developments
including residential, commercial, industrial and civic and cultural land
uses. As a result, personal observation found land use activities such as
houses, retail shops, churches, schools, fuel filling stations and wood
and metal processing factories within 100 m of every river and flood-
plains in Kumasi. This widespread unauthorised utilisation of the water
areas urban land uses accounted for the virtual loss of rivers and
floodplains in the city with the few existing ones characterised by un-
dermined hydrologic vitality.
A major underlying driver of the expansion of urban land uses into
the rivers is the rapidly increasing population of Kumasi. The popula-
tion of the city nearly doubled in a decade from 1.2 million in 2000 to
over 2 million in 2010 at an annual growth rate of 5.5% as the fastest
urbanising city in Ghana [22]. The largely unregulated rapid popula-
tion growth has resulted in a the high price and shortage of developable
lands, and a high housing deficit that encourage the city's dwellers to
encroach and infill rivers and floodplains in order to satisfy land and
housing needs. In addition, the city's population growth is driven by
high influx of poor migrants (in-migration rate = 49%) that are usually
priced out of the formal land and housing markets and as such com-
pelled to encroach and develop substandard houses in the rivers and
floodplains, which are considered cheap and wastelands [9]. Thus, the
Kumasi's spontaneous urbanisation that is strongly caused by high in-
migration significantly contributes to the drastic loss of its water areas
to physical developments with concomitant flooding problems.
Consequently, interactions with institutional officials and property
owners indicated that the extensive degradation and disappearance of
112
International Journal of Disaster Risk Reduction 27 (2018) 105–117
the rivers and floodplains have caused almost complete loss of their
flood mitigation functions such as runoff conveyance and infiltration,
hence increasing flood risk in the city. The interviewees emphasised
that the floods experienced in Kumasi are largely consequences of the
spatial deterioration and fragmentation of the inland water network in
the city. The interviewees explained that flood disasters were rare in the
city in the past because the numerous rivers and floodplains facilitated
safe conveyance and infiltration of runoff after rainfall, by serving as
natural drains and reservoirs. Unfortunately, following the loss of many
of the rivers and floodplains, flooding has become an annual occurrence
in many suburbs of the city. Some of the interviewees said that:
“….. There were many rivers here [Kumasi]……and served as natural
drainage channels that collected the rainwater/runoff for proper dis-
posal. But many of the rivers have been and still being used for physical
development, and as garbage disposal sites. Because of these, many of the
rivers and floodplains have disappeared while the channels of existing
rivers have become narrow, so a little rain just overflows the banks to
enter the houses around them…I can say that about 90% of floods in
Kumasi are due to the encroachment and loss of the rivers and flood-
plains” HSD, June 2013.
“.. I joined NADMO in 1998, and during that time Kumasi was not
experiencing much flooding. Flooding was not one of the disasters we
were battling with. But for the past five years, flooding has become one of
the major disasters we experience here because of the encroachment and
loss of rivers and floodplains that served as natural drainage channels…”
NADMO, June 2013.
“I have lived here for over 30 years but we never experienced any floods
until recently. The runoff flowed freely and no floodwater ever entered
my house until people started building in the waterways. At the moment
almost all the river channels and floodplains have been utilised to put up
structures, and we are suffering from floods…..”POI 12, July 2013.
The interview results further indicated that the development of
physical structures (Fig. 10) and dumping of garbage in and along the
rivers and floodplains, coupled with the eventual disappearance of
these water areas, causes the obstruction and/or loss of pathways of the
runoff. Also, due to encroachment, it was observed that the few existing
rivers have small and disjointed channels which cannot support the
smooth flow of runoff. As consequence, any excessive runoff on rainy
days overflows the river banks, and flows through any available land, as
well as inundating houses. Thus, flooding in Kumasi is mainly an out-
come of developments occupying waterways and/or blocking the flow
of rivers and/or runoff as indicated by the TCPD interviewee:
“We usually experience floods at the peak of the rainy season because a
lot of buildings are located in waterways that have low infiltration rate,
and also get saturated at that period. So the large quantities of runoff
generated during rainfall spreads into homes and other structures that
block its flow” TCPD, June 2013.
Additionally, some of the property owners interviewed described
the flood disasters as punishments from the “river gods” because of the
degradation of the rivers. This perception is explained by the belief in
Ghana that rivers are “gods” with “spirits and powers”, which can be
helpful if appeased, or harmful if degraded by human activities like
farming, physical development and dumping of refuse [42,48]. Based
on this traditional belief, the interviewees observed that the recent
widespread encroachment and associated pollution of the rivers and
floodplains has incurred the wrath of the river gods, who punish them
with frequent destructive floods. An interviewee said that:
“…I think the floods we experience here [Kumasi] are because the [river]
gods are angry with us for contaminating the rivers. You know these
rivers are spirits and as such do not like filth but now we dump waste into
them…. So I think the rivers are angry and punishing us for the pollution
because we didn't have any flood problems when the rivers were
P. Amoateng et al.
clean…”POI 15, July 2013
Many property owner interviewees shared similar views, with an
emphasis on the fact that the floods are expressions of the wrath of the
river deities due to the degradation of the rivers and floodplains by
human activities. Thus in a traditional sense, the spatially degraded
rivers in Kumasi now cause harmful flood disasters, instead of being
sources of blessings like carrying runoff. Botchie et al. [11], trace this
113
International Journal of Disaster Risk Reduction 27 (2018) 105–117
Fig. 9. Land use/cover maps of Kumasi for 1985, 2000 and 2013.
problem to the apparent downgrading of traditional beliefs and norms
such as delineation of the areas adjoining rivers as sacred groves, where
physical development and other human activities were prohibited as
measure to protect the rivers to stimulate and/or guarantee the pur-
ported spiritual blessings and protections from the river gods.
The above analysis has revealed that the loss of the rivers and
floodplains and their associated flood control functions is the main
Fig. 10. Developments in waterways in Kumasi.
P. Amoateng et al.
cause of annual flood disasters in Kumasi. This finding affirms wide-
spread claims that reduction in water conveyance and infiltration ca-
pacity, and channel obstruction and fragmentation of the river channels
and floodplains by physical structures and waste materials cause runoff
from any rainstorm overwhelms these water areas and overflow or
spread over any open area to inundate developments within their flood
zone [17,29]. In Kumasi, the enormous influences of urban expansion
into rivers and floodplains on flood incidences is further underscored by
the extended retention of floodwater in affected houses, as this in-
dicates that the physical developments are located in low-lying areas
with high clay soil contents and high water tables where water logging
is expected in rainy seasons [3,12,33].
Nevertheless, the large scale replacement of water areas with urban
development and the concomitant flood disasters reflects the shortfalls
in the city's urban planning practices and policy frameworks that aim at
creating safe built environment, and conserving natural environments
particularly rivers and floodplains to provide beneficial ecosystem
system services including flood mitigation. Specifically, the National
Building Regulations, 1996 (LI 1630) and Riparian Buffer Zone Policy
jointly provide that developments must be 30–100 m away from river
banks and wetlands, while the Local Government Act (Act 462) au-
thorises the planning authority (Kumasi Metropolitan Assembly [KMA])
to prohibit, alter, abate, remove or demolish physical developments
within the zone, for the purposes of avoiding flooding and controlling
river pollution. Unfortunately, the city managers’ inability to enforce
these laws and policies has resulted in the occupation of the rivers and
floodplains by various physical developments that are flooded by an-
nual overflows of the effluent streams. For example, the poor enforce-
ment capacity of the development control institutions, such as their
inability to acquire water areas for public interest as stipulated on land
use plans, has created a market for commercial sale of water areas by
traditional land owners to desperate private developers to undertake
physical developments, contrary to the provisions of existing land use
plans [9]. The urban land use/activity intensification in the water areas
and/or floodways therefore relates to several factors including admin-
istrative inefficiencies, low capacity of enforcing institutions, un-
scrupulous sale of lands by local chiefs, widespread urban poverty, land
and housing shortage, desire to live in proximity to city centre and poor
public environmental attitude [9,15]. The multiplicity of the factors
connected with the loss of the water areas further reveals the com-
plexity of the flood disasters that result from it, suggesting the need for
integrated and holistic approaches to address the problems.
4.5. Efforts at (mis)managing flooding
The efforts for managing floods and similar disasters in Kumasi are
encapsulated in the Kumasi Disaster Management Plan (KDMP), which
was formulated in 2011 and implemented through the National Disaster
Management Organisation (NADMO). Among others, the KDMP has the
central objective of promoting a more scientific and effective manage-
ment of flood disasters in the city through the application of integrated
disaster management approaches, as well as identifying measures for
managing the pre, emergency and post disaster phases of flood in-
cidences. The policy actions of the plan include flood hazard mapping
(identifying hazard situations, identifying safe havens, vulnerability
and risk assessment); education and training (creating public awareness
of hazards, skills acquisition by emergency response workers); emer-
gency response and relief management (distributing relief items,
evacuating and sheltering victims); as well as rehabilitation and reset-
tlement (reconstructing infrastructure and utility services, promoting
self-reliance and income generating activities). However, analysis of
interview results and direct observations indicate that the plan is poorly
implemented due to a plethora of challenges including poor institu-
tional collaboration, lack of integration of flood disaster management
issues into sectoral plans and lack of human and logistical resources. As
a result, flood management in the city mainly involves structural
114
International Journal of Disaster Risk Reduction 27 (2018) 105–117
engineering measures like storm drain construction and river channel
modifications, as well as reactionary flood emergency phase interven-
tions.
Storm drains – primary, secondary and tertiary – have been con-
structed to receive and carry storm water, runoff and sullage water in
various parts of the city. In addition, sections of some of rivers including
Aboabo, Subin and Sisa Rivers, totalling about 23 km, have been con-
creted to strengthen the channel walls and increase flow capacity, for
the purposes of controlling floods. Unfortunately, interviews indicated
that these structural measures have not only been ineffective in redu-
cing floods but have also contributed in many cases to the occurrence of
floods. The interviewees emphasised that inadequate provision of
drains, lack of proper maintenance and low capacities of the existing
drains are key contributors to flooding in many suburbs in the city.
Available data that indicate that the just 10% of the city is properly
served by drainage and sewerage network [35], and as such storm
drains are non-existent in many of the parts of the city, especially the
flood prone suburbs. As a result, unguided storm water and runoff have
to find their own flow paths within the communities and eventually
spread to inundate nearby houses, as was noted by one interviewee:
“The flooding we suffer in here [Moshie-Zongo] is a result of lack of
drains…The floodwaters wouldn’t have entered our homes if there were
drains for them to flow through. But because there are no drains, the
runoff passes and enters anywhere it finds a way” POI 27, August
2013.
In addition, personal observations found that many of the drains
have construction defects that inhibit their functionality and subse-
quently contribute to flooding. For example, it was observed that ma-
jority of storm drains in the city are undersized and/or connected at
almost 90°, which affect runoff conveyance. This was confirmed by the
research participants, who explained that such engineering limitations
have undermined the flow and discharge capacities of the drains, re-
sulting in a backwater effect that causes flooding with even small
rainfall event. The HSD interviewee and a focus group participant said
that:
“Yes, many of the drains in Kumasi are undersized and have engineering
deficiencies especially at the confluences and these result in overflows
and backwater effect that contributes to flooding” HSD, June 2013.
“The flood that occurred in this community this year was caused by the
small size of the drain they [city authorities] came to construct across the
river…The drain is very narrow so the storm water was not able to flow
through it and that made the water to spread and enter into the houses
around the area ….” FGD 2 Participant, September 2013.
These problems are compounded by the deplorable conditions of
drainage systems which have resulted from the lack of proper routine
maintenance, development of unauthorised structures on and along the
drains, dumping of refuse and human excreta in the drains, as well as
laying of service and utility lines within some of the drains. Interviews
with the institutional officials and property owners revealed that no
routine dredging and desilting is undertaken due to financial con-
straints, while the residents have turned the drains into garbage
dumping sites leading to accumulation of sediments and refuse in and
along the storm drains. For example, field inspection reports and hy-
draulic calculations obtained from HSD indicated that the Aboabo
storm drain, which was originally 2.5 m deep, had reduced to 0.7 m in
2013, indicating that more than three quarters of the drain is filled with
sediment and other waste materials. Similar situation was observed
with many of the drains in the city as they were heavily choked with
solid wastes such as plastic bags, cans and bottles, especially at the
entrances, which were sometimes completely blocked as shown in
Fig. 11. Consequently, the drains are unable to carry storm water
during downpours due to blockage of flow and reduced capacity, hence
leading to floods; as was narrated by interviewees as follows:
P. Amoateng et al.
“…People have turned the drains into refuse dump sites. They throw
garbage, especially plastic bags into the drains and these heavily choke
the drains so when it rains the storm water that enters them [drains] is
unable to flow and as such spreads to destroy properties. This is one
reason for the numerous flood incidences at the onset of the rainy season
because the wastes dumped into the drains severely obstruct free flow of
runoff generated by the initial rains” NADMO, June 2013.
“If you go to places like Adum and Kejetia [the city centre] among others,
we have small drains that are supposed to flow into primary drains but
these open drains are almost always choked with garbage, while people
have also built on them. So the slightest rain results in flood because it's
either there are no ways for runoff to pass or the drains have been
blocked at the entrance….. Besides, we don't have funds to maintain
these drains. For example, at the moment we need to desilt the Aboabo
storm drain and many others but we are still waiting for funds from the
Ministry of Water Resources, Works and Housing, which we don’t know
when we will receive it” HSD, June 2013.
“The city authorities don’t manage the drains well while people dump all
forms of waste into them [drains]…Most of the drains in this community
have not been dredged for more than five years and because of that the
gutters are highly silted, so with the little rain, then the whole of this area
get flooded” POI 32, September 2013.
Given the failure of the engineered drains, the cities authorities
resort to reactionary emergency and post flood phases mitigation
measures such as distributing relief items (e.g., food and non-food
items) and demolishing structures within waterways; when the floods
occur. The NADMO interviewee explained that:
“We normally give them [flood victims] food, clothes, plastic utensils,
roofing sheets, mattresses and blankets among others. We also collabo-
rate with institutions like HSD, MBID, and KMA to demolish structures in
waterways in the communities affected by the floods” NADMO, June
2013.
Nevertheless, interviews and personal observations revealed that
many of the interventions such as demolition of structures within the
rivers and floodplains hardly become a reality due to fierce opposition
from the occupants, lack of political will on the part of the KMA, as well
as corruption and connivance among staff of institutions tasked to un-
dertake such activities. For example, it was observed that the due date
for demolition of many unauthorised residential buildings within the
water areas had passed yet no actions have been taken, with people still
occupying the structures. Also, the institutional interviewees indicated
that emergency responses such as supply of relief items are often lim-
ited due to lack of human and financial resources. This was underscored
by many of the property owners, who indicated that they have never
received any assistance from the city authorities before, during or after
flood disasters. A property owner and the NADMO interviewee re-
marked that:
115
International Journal of Disaster Risk Reduction 27 (2018) 105–117
Fig. 11. Choked and silted storm drains in Kumasi.
“I don’t receive any help from the KMA or NADMO; I have been dealing
with the problem myself. Sometimes they come here to write names of
those affected by the floods, but I have never received anything from
them [KMA/NADMO]” POI 9, July 2013.
“We are unable to provide relief items to everyone [flood victims] be-
cause we usually do not to have enough funds to purchase large quan-
tities. But we have been doing our best to help them [flood victims]"
NADMO, June 2013.
The foregoing clearly indicates that despite the increasing flood
disasters in Kumasi, interventions tend to be reactionary, and some-
times aggravate the incidence the disaster. While the well-structured
disaster management plan has not been duly implemented due to sev-
eral challenges; the reliance on storm drains has been largely in-
effective, with emergency responses being mostly inadequate. In par-
ticular, the drainage networks remain woefully inadequate and
inefficient because of the general low level of economic development,
negligence by authorities, ignorance and non-compliance of main-
tenance requirements by the public. This finding underscores the claims
in the literature that traditional structural flood control measures,
which usually aim to prevent floods rather than minimise the effects of
their occurrences tend to worsen these disasters if they are not well
implemented and maintained ([1,4,26,34,44]). Further, the situation
reflects earlier observations that flood disaster management practices in
developing countries are ad hoc, one dimensional, unscientific, un-
coordinated and not largely integrated in land use plans [41,45]. Given
this situation, the underlying causes of floods – development in flood
prone areas – are not well addressed, thus exacerbating the damages
and inconveniences associated with the disaster.
5. Conclusion and recommendations
Findings from this research suggest that the recent recurrent flood
disasters in Kumasi can be described as riverine floods that adversely
affect built-up areas because of increased human occupation and loss of
rivers and floodplains in the city. As indicated in the analysis, the flood
disasters which occur during the rainy seasons, with an average of five
events per year are largely concentrated in the low income informal
communities that are characterised extensive encroachment of rivers
and floodplains by physical developments. This excessive intrusion of
urban development into the waterways beyond acceptable limits has
undermined the aquatic systems’ capability to serve as natural drainage
systems. Indeed, the rivers and floodplains now have reduced channel
capacity and land area that are incapable to safely convey and absorb
the high volume of runoff generated during storms. This leads to in-
undation of the houses and other physical developments within these
water areas, which naturally have high water table and slow absorption
capacity and as such reach saturation points quickly, by excess dis-
charge or runoff that overflows the riverbanks and floodplains.
Regrettably, the research finding further show that the effectiveness
of engineered drainage systems, which are main structural approaches
P. Amoateng et al.
for mitigating and controlling floods disasters in Kumasi, leaves much
to be desired. It was realised that the inadequacy, construction defects
and poor state of storm drains contribute to flood incidences in the city.
The research found that while storm drains are non-existent in many
suburbs, the few existing ones are either undersized, incorrectly
aligned, encroached by houses and other developments or silted and
blocked by waste materials which cumulatively result in absence of
appropriate drainage systems to convey runoff in the city. Nevertheless,
in Kumasi, the frequency and magnitude of flood induced by improper
and inadequate drainage network is strongly related to the fact that the
urban developments are increasingly locating within and/or replacing
the flow paths of storm waters and effluent rivers, as well as wetlands.
The intrusion of urban land uses into water areas is the root cause of
flood disasters in Kumasi, indicating that eliminating the disaster re-
quires efforts targeted at removing existing and preventing new urban
land use developments in these areas. There are a number of possible
solutions.
Firstly, it is important for city managers and planners to pursue an
integrated urban land use planning system that recognises rivers and
other wetlands as indispensable components of the urbanscape that
provide crucial hydrological services such as mitigating storm damage
and flood to ensure safe built environments. Urban planning and water
resources management, which have for a long time been developed in
different disciplines and practised in different institutions of national
and local governments, should be integrated at both the strategic and
local action levels. Urban planning practices would then become more
water sensitive, with land use plans in particular, clearly recognising
and incorporating water related lands by carefully delineating and re-
serving water areas, determining land uses within the watersheds/
catchments and (re)-conceiving wetlands and riparian buffer zones as
urban drainage infrastructure. These measures would help to protect
the rivers and floodplains, as well as sustain their flood mitigation
functions such as safe conveyance and infiltration of the runoff, and
thus reduce flood risk in the city.
Secondly, there is the need for the city authorities to diligently
pursue participatory relocation of settled encroachers along the rivers
and floodplains. Given, the complexity of the factors involved in the
invasion of rivers and floodplains by urban development in the city, it is
necessary that the interventions are not solely based on actions by
public institutions but with active involvement of the encroachers and
the general public. Such a participatory relocation program that entails
actively involving targeted beneficiaries in the design, planning and
implementation of relocation programmes and/or areas should involve
collaborative efforts among a wide range of stakeholders especially the
encroachers in order to produce a sense of ownership for the new lo-
cation. This may lead to voluntary relocation of the property owners to
the new location because of their active involvement in choice and
development of the location. This can ultimately help restore degraded
rivers and floodplains and preserve existing ones, as well as, which
reduce and/or avoid flood disasters.
Moreover, this research recommends intensification of public edu-
cation and campaigns on the values and protection of rivers and
floodplains. These environmental education programmes are essential
for understanding the complexity and uncertainty of hydrological dis-
asters such as floods that become pronounced with the disappearance of
rivers and floodplains from the urban fabric. Consequently, public en-
vironmental education should provide information on the negative
consequences of loss of rivers and floodplains on the entire cities,
sanctions for undertaking unapproved activities within water areas and
around drainage systems, as well as how communities and individuals
themselves can impose sanctions on and/or challenge disapproved ac-
tivities in flow paths of rivers and storm drains. These environmental
awareness campaigns can build public opinion and attitudes based on
sound and relevant information, as well as encourage the public to act
responsibly towards safeguarding water areas and drainage networks
by: avoiding developing in waterways and on drains, and disposing of
116
International Journal of Disaster Risk Reduction 27 (2018) 105–117
waste in rubbish bins instead of rivers and storm drains. These will help
keep the water areas and storm drains in good state to effectively
convey and/or absorb runoff within the city and therefore reduce and/
or prevent flood disasters.
Overall this research suggests that the various institutions and sta-
keholders involved in flood disaster and water resources management
in the city act collaboratively to ensure more successful and sustainable
outcomes. More efforts should be made to break down institutional
segmentation, remove barriers to cooperation and collaboration, as well
as promote effective stakeholder consultation and negotiation. Strong
institutional collaboration and coordination is very much needed given
the low capacity and limited resources of the institutions. The colla-
boration will help to effectively operationalise earlier recommendations
such as integrated land use planning, participatory relocation and
public environmental education. This could ultimately end the wide-
spread reliance on reactive interventions to flood disasters, and ensure
better protection of rivers and floodplains, which in turn can help mi-
tigate flood disasters by serving as natural drains. Thus, addressing
flood disasters in, and enhancing liveability of Ghanaian cities revolves
efforts that ensure ‘no development’ zone beyond certain limits of water
areas within the cities.
Acknowledgement
This paper was supported by the Charles Sturt University through
funding from the Higher Degree by Research Writing Up Award
(SN:11488876).
References
[1] D. Aboagye, Living with familiar hazards: flood experiences and human vulner-
ability in Accra, Ghana (Online), J. Urban Res. (2010), http://dx.doi.org/10.4000/
articulo.2110.
[2] A.K. Addo, L.L. Larbi, B. Amisigo, P.K. Ofori-Danson, Impacts of coastal inundation
due to climate change in a cluster of urban coastal communities in Ghana, West
Africa, Remote Sens. 3 (9) (2011) 2029–2050.
[3] T. Adjei-Gyapong, R.D. Asiamah, The Interim Ghana Soil Classification System and
its Relation with the World Reference Base for Soil Resources. Rapport sur les
Ressources en Sols du Monde, FAO, Rome, 2002.
[4] K, Afeku, Urbanization and flooding in Accra, Ghana (Master's Thesis), Miami
University, Miami, USA, 2005.
[5] B.P. Aldrin, The slow erosion of Ayigya: a community's struggle to deal with the
impact of floods in Kumasi, Ghana, TAO Shelter Online 7 (2010) 1–8.
[6] D. Amaratunga, D. Baldry, M. Sarshar, R. Newton, Quantitative and qualitative
research in the built environment: application of “mixed” research approach, Work
Study 51 (1) (2002) 17–31.
[7] C. Amoako, P.B. Cobbinah, Slum improvement in the Kumasi metropolis, Ghana: a
review of approaches and results, J. Sustain. Dev. Afr. 13 (8) (2011) 150–170.
[8] P.Y.O. Amoako, S.T. Ampofo (Eds.), Hazard Mapping in Ghana: Report to NADMO,
Accra, 2009, NADMO, Accra, Ghana. Retrieved from: 〈http://www.unisdr.org/
files/14757_6thCGDRMProgramsforPriorityCountrie.pdf〉.
[9] P. Amoateng, The Changing Spatial Extent of Rivers and Floodplains and its
Implications for Flooding: The Case of Kumasi (Ph.D. Thesis), School of
Environmental Sciences, Charles Sturt University, Albury, Australia, 2016.
[10] K. Anil, A.K. Gupta, S.S. Nair, Flood risk and context of land-uses: Chennai city case,
J. Geogr. Reg. Plan. 3 (12) (2010) 365–372.
[11] G. Botchie, T. Akabzaa, E. Gyasi, D.B. Sarpong, Land Tenure, Land Use and
Environment in Ghana, Institute of Statistical, Social & Economic Research,
University of Ghana, Accra, Legon, 2007.
[12] B.B. Campion, J.F. Venzke, Rainfall variability, floods and adaptations of the urban
poor to flooding in Kumasi, Ghana, Nat. Hazards 65 (2013) 1895–1911.
[13] B.B. Campion, Urban wetland ecology and floods in Kumasi, Ghana (Ph.D. Thesis),
Institut für Geographie an der Universität Bremen, Bremen, Germany, 2012.
[14] J. Chen, X. Wu, B.L. Finlayson, W. Michael, T. Wei, M. Li, Z. Chen, Variability and
trend in the hydrology of the Yangtze River, China: annual precipitation and runoff,
J. Hydrol. 513 (2014) (2014) 403–412.
[15] P.B. Cobbinah, R.M. Darkwah, Urban planning and politics in Ghana, GeoJournal
(2016), http://dx.doi.org/10.1007/s10708-016-9750-y.
[16] W.P. Cunningham, Principles of Environmental Science – Inquiry and Application,
McGraw Hill, New York, 2002.
[17] O.N. Dhar, S. Nandargi, Hydrometeorological aspects of floods in India, Nat.
Hazards 28 (1) (2003) 1–33.
[18] I. Douglas, K. Alam, Climate Change, Urban Flooding and the Rights of the Urban
Poor in Africa: Key Findings from Six African Cities, Action Aid International,
London, 2006.
[19] I. Douglas, K. Alam, M. Maghenda, Y. Mcdonnell, L. Mclean, J. Campbell, Unjust
P. Amoateng et al.
waters: climate change, flooding and the urban poor in Africa, Environ. Urban. 20
(1) (2008) 187–205.
[20] R. D’Souza, H. Nagendra, Changes in public commons as a consequence of urba-
nization: the Agara Lake in Bangalore, India, Environ. Manag. 47 (5) (2011)
840–850.
[21] D. Dutta, S. Herath, S. Wijesekera, Development and application of an urban flood
risk analysis system, in: Proceedings of the First International Conference of Asia-
Pacific Hydrology and Water Resources Association, Kyoto, Japan, 2003, pp.
821–826.
[22] Ghana Statistical Services (GSS), 2010 Population & Housing Census National
Analytical Report, GSS, Accra, 2013.
[23] Ghana Meteorological Agency (GMA), Daily rainfall records for Kumasi. GMA,
Accra, Ghana, 2013.
[24] A.K. Gyekye, Environmental change and flooding in Accra, Ghana, Sacha J.
Environ. Stud. 3 (1) (2013) 65–80.
[25] M. Holland, R. Kasanga, C. Lewcock, H. Warburton, Peri-urban Baseline Studies,
Kumasi, Ashanti Region, Ghana-Volume 1: Executive Summary and Main Report,
Final Technical Report. Natural Resources Institute, UK, University of Science and
Technology, Kumasi, Ghana, 1996.
[26] T. Ikeda, J. Yoshitani, A. Terakawa, Flood management under climatic variability
and its future perspective in Japan, Water Sci. Technol. 51 (2005) 133–140.
[27] Intergovernmental Panel on Climate Change (IPCC), Climate Change 2007 – The
Physical Science Basis: Working Group I Contribution to the Fourth Assessment
Report of the IPCC, Cambridge University Press, Melbourne, Australia, 2007.
[28] Intergovernmental Panel on Climate Change (IPCC), Climate Change 2013: The
Physical Science Basis. Contribution of Working Group I to the Fifth Assessment
Report of the Intergovernmental Panel on Climate Change, Cambridge University
Press, Cambridge, 2013.
[29] M.S. Islam, M.R. Rahman, A. Shahabuddin, R. Ahmed, Changes in wetlands in
Dhaka City: trends and physico-environmental consequences, J. Life Earth Sci. 5
(2010) 37–42.
[30] Japan International Cooperation Agency (JICA), The Study on the Comprehensive
Urban Development Plan for Greater Kumasi in the Republic of Ghana. Draft Final
Report, 2 Ministry of Environment, S., Technology and Innovation (MESTI), Town
and Country Planning Department (TCPD), Japan International Cooperation
Agency (JICA), Kumasi, 2013.
[31] K.A. Jha, R. Bloch, J. Lamond, Cities and Flooding: A Guide to Integrated Urban
Flood Risk Management for the 21st Century, The World Bank, Washington DC,
2011.
[32] K.N. Karley, Flooding and physical planning in urban areas in West Africa: situa-
tional analysis of Accra, Ghana, Theor. Empir. Res. Urban Manag. 4 (13) (2009)
25–41.
117
International Journal of Disaster Risk Reduction 27 (2018) 105–117
[33] G.O. Kesse, The Mineral and Rock Resources of Ghana, A. A. Balkema Publication,
Boston, 1985.
[34] D. Kounsana, T. Takahashi, Study on various measures for flood in specific region: a
case study of the 2008 Lao Flood, World Acad. Sci. Eng. Technol. 70 (2012)
948–955.
[35] Kumasi Metropolitan Assembly (KMA), Development Plan for Kumasi Metropolitan
Area (2010–2013), Ministry of Local Government and Rural Development, Accra,
2010.
[36] L.B. Leopold, M.G. Wolman, J.P. Miller, Fluvial processes in Geomorphology, W.H.
Freeman, San Francisco, California, 1964.
[37] Millennium Ecosystem Assessment (MEA), Ecosystems and Human Well-being:
Current State and Trends, Island Press, Washington, DC, 2005.
[38] A. Naidoo, K. Patric, Cholera: a continuous epidemic in Africa, J. R. Soc. Promot.
Health 122 (2) (2002) 89–94.
[39] National Disaster Management Organisation (NADMO), Annual disaster report for
Kumasi, NADMO, Kumasi, Ghana, 2013.
[40] B. Nyarko, Flood risk zoning of Ghana: Accra experience, Int. Arch. Photogr.
Remote Sens. 33 (B7/3; PART7) (2000) 1039–1050.
[41] C.Y. Okyere, Y. Yacouba, D. Gilgenbach, The Problem of Annual Occurrences of
Floods in Accra: An Integration of Hydrological, Economic and Political
Perspectives. Interdisplinary Term Paper, ZEF Doctoral Studies Programme,
Boon:Centre for Development Research, Boon University, Boon, Germany, 2012.
[42] M. Opoku-Agyemang, The role of the district assemblies in the management of
trans-district water basins in Ghana, in: Proceedings of the Paper Presented at the
International Workshop on African Water Laws: Plural Legislative Frameworks for
Rural Water Management in Africa, Johannesburg, South Africa, 2005, pp. 1–7.
[43] J.K. Owusu-Ansah, The influences of land use and sanitation infrastructure on
flooding in Kumasi, Ghana, GeoJournal 81 (2015) 555–570.
[44] H. Petroski, Levees and other raised ground, Am. Sci. 94 (1) (2006) 7–11.
[45] D. Rain, R. Engstrom, C. Ludlow, S. Antos, Accra, Ghana: A City Vulnerable to
Flooding and Drought-induced Migration, UN-Habitat, Nairobi, Kenya, 2011.
[46] G. Rossman, S. Rallis, Learning in the Field: An Introduction to Qualitative
Research, 2nd ed., Sage, Thousand Oaks, CA, USA, 2003.
[47] M. Sandelowski, Focus on research methods – whatever happened to qualitative
description? Res. Nurs. Health 23 (4) (2000) 334–340.
[48] G.A. Sarpong, Customary Water Law and Practices: Ghana, Food and Agriculture
Organization of the United Nations, Rome, 2005.
[49] D. Satterthwaite, Adapting to Climate Change in Urban Areas: The Possibilities and
Constraints in Low-and Middle-income Nations, 1 IIED, London, UK, 2007.
[50] United Nations Water, The world is thirsty because we are hungry: the world water
day, United Nation Water, 2012. 〈http://www.unwater.org/worldwaterday/〉.