International Journal of Disaster Risk Reduction 5 (2013) 38–48

Contents lists available at ScienceDirect

International Journal of Disaster Risk Reduction

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

Indigenous knowledge, coping strategies and resilience

to floods in Muzarabani, Zimbabwe
Emmanuel Mavhura a, Siambabala Bernard Manyena b,n, Andrew E. Collins c,
Desmond Manatsa a
a
Department of Geography, Bindura University of Science Education, Private Bag 1020, Bindura, Zimbabwe
b
Disaster and Development Centre, Faculty of Engineering and Environment, Northumbria University, Newcastle upon Tyne NE1 8ST, UK
c
Disaster and Development Centre, Faculty of Engineering and Environment, Northumbria University, Newcastle upon Tyne NE1 8ST, UK

a r t i c l e in f o abstract

Article history: The connection between indigenous knowledge systems and disaster resilience derives
Received 17 November 2012 from both theory and practice highlighting potential contributions of indigenous knowl-
Received in revised form edge to building resilient communities. Using data from interviews and focus group
15 July 2013
discussions, this paper explores people’s indigenous survival strategies and variations in
Accepted 16 July 2013
Available online 8 August 2013
people’s ability to cope with floods in two flood-prone villages of Muzarabani district,
Zimbabwe. The findings reveal that indigenous knowledge systems played a significant
Keywords: role in reducing the impact of floods in Muzarabani district. However, the extent to which
Indigenous knowledge indigenous knowledge enhanced resilience to floods was influenced by geophysical
Floods
locations, exposure to flooding and socio-economic abilities. Communities in an area
Resilience
with low flooding and with a strong socio-economic base such as education and income
Coping strategies
Muzarabani were more likely to cope with flood impacts compared to those communities in areas with
Zimbabwe high and sudden flooding and weak socio-economic base. The paper shows how
indigenous knowledge systems are an indispensable component of disaster resilience
building. This is because indigenous knowledge systems can, (i) be transferred and
adapted to other communities; (ii) encourage participation and empowerment of affected
communities, (iii) improve intervention adaptation to local contexts, and (iv) are often
beyond formal education about environmental hazards.
& 2013 Elsevier Ltd. All rights reserved.

1. Introduction to ‘bounce back’, if not ‘bounce forward’ [3] and to recover

The rejection of environmental determinism as an
inadequate account of disaster causation shifted attention
towards resilience [1]. The resilience notion emphasises
the importance of communities to ‘build back better’ [2],
n
Corresponding author. Tel.: +44 796 679 35 92.
E-mail addresses: edmavhura@gmail.com,
emavhura@buse.ac.zw (E. Mavhura),
bernard.manyena@northumbria.ac.uk,
bernmanyena03@gmail.com (S.B. Manyena),
andrew.collins@northumbria.ac.uk (A.E. Collins),
dmanatsa@gmail.com (D. Manatsa).
from disasters with little or no external assistance. Increas-
ingly, indigenous knowledge systems are among the ele-
ments implicated in disaster resilience ‘thinking’, as
reflected in sub-texts of the Hyogo Framework for Action
of 2005 as part of its call for ‘building the resilience of
nations and communities to disasters’ [4]. Indigenous
knowledge has been an inherent component of traditional
disaster management systems where over centuries peo-
ple have adjusted their lives and livelihoods to adapt to
changing contexts [5]. The interest in indigenous knowl-
edge systems has been particularly highlighted in flood
disasters, due to the likely increase of flood events result-
ing from anthropogenic climate change through heavy

2212-4209/$ - see front matter & 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.ijdrr.2013.07.001
 E. Mavhura et al. / International Journal of Disaster Risk Reduction 5 (2013) 38–48
precipitation, increased catchment wetness and sea level
rise [6]. Yet, there is insufficient empirical evidence sug-
gesting the effectiveness of the indigenous knowledge in
enhancing resilience to flood disasters. More importantly,
the studies that have been conducted on the connections
between indigenous knowledge and flooding are based on
Asia, with a relatively insignificant number of such studies
from Africa. In bringing indigenous knowledge more into
focus the rationale of this paper is that coping and
resilience are interrelated concepts in contexts of African
development. The effectiveness of people’s knowledge as
driver of coping and resilience from within their commu-
nity is however dependent on ecological and social con-
ditions within which they must reside. It is not the
intention of this paper to examine and reposition ter-
minologies already widely banded about the academic
discourse. Rather the paper contributes a case study
through which concepts might gain better meaning, by
closer understanding of what under-reported people in
conditions of flood exposure and socio-economic volatility
do to survive.
This paper presents findings from a study on the role of
indigenous knowledge in reducing the impact of floods in
Muzarabani district, Zimbabwe. The paper will make a
significant contribution to the conspicuous literature gap
on indigenous knowledge systems and flood disasters in
Africa. The results are a useful source of information for
preparation of a comprehensive disaster management
programme. By using indigenous knowledge, it is hoped
that the government’s institutional capacity will be
enhanced in order to effectively encounter future flooding.
The paper also highlights the resilience of Muzarabani as a
community, particularly during a period when Zimbabwe
faced great socio-economic challenges. It indicates the
merits of integrating indigenous knowledge systems in
flood risk reduction.
2. Indigenous knowledge and disaster resilience
The term ‘indigenous knowledge’ is presented as hav-
ing varied meanings including ‘local knowledge’ [7–9],
‘traditional knowledge’, ‘indigenous technical knowledge’,
‘peasants knowledge’, and ‘traditional environmental
knowledge’ or ‘folk knowledge’ [10]. Notwithstanding a
multiplicity of meanings the term suggests a body of
knowledge existing within or acquired by local people
over a period of time and passed through generations
[11,12]. Indigenous knowledge is thought to be distin-
guished from scientific knowledge, the latter being asso-
ciated with Western technology or techniques. However, it
is a subjective interpretation in that such differentiations
can be challenged since all knowledge can be indigenous,
cultural or ethnic. Nonetheless it may be reasonable to
consider mainstream knowledge branded as scientific to
be structured and global in nature [11]. Shaw et al. [13]
state that indigenous knowledge tends to be locally bound,
culture and context specific, non-formal and orally trans-
mitted, closely related to survival and subsistence,
dynamic and based on innovation, adaptation and experi-
mentation. This suggests indigenous knowledge is an
inherent valuable resource, and an integral coping or
39
resilience mechanism built over centuries. Although indi-
genous knowledge, in many ways, is not always apparent
[12], local communities utilise it to construct and recon-
struct themselves to prevent, mitigate, prepare, respond
and recover from disaster impacts. This can be considered
to be a coping mechanism and is the manner in which a
group of people or individuals become ‘resilient’ in the
sense used by UNISDR cited earlier in the Section 1 [4]. The
resilience terminology is adopted by a global array of
governmental and non-governmental organisations that
now apply it commonplace. However, bringing indigenous
knowledge more into focus and context deepens the
understanding of how people manage their own changing
circumstances and can bring more pertinent information
about the hazard event itself Mercer et al. [11] argue that
indigenous knowledge is invaluable in understanding
historical hazard events while scientific knowledge has
been proven to be of great importance when dealing with
rare or unprecedented events. As a result, indigenous and
scientific knowledge should be considered as two sides of
the same coin, which complement each other.
While we avoid any overly definitive model that would
explain the case of Muzarabani, the basic approach of this
study can be considered in relation to Dekens’ [2] frame-
work for analysing local knowledge in disaster prepared-
ness (Fig. 1). Dekens states that disaster preparedness at
the local level is a product of knowledge types, practices
and beliefs, which are mainly influenced by societal
structures and processes within the local and global
hazard contexts. In this model, disaster preparedness at
the local level is based on observation, anticipation,
adaptation and communication to enhance resilience with
the effects on levels of security, sustainability and com-
munity resilience building.
3. The study area
Muzarabani is one of the most flood-prone areas in
Zimbabwe. The term ‘muzarabani’ in the local Shona
language means flood plain or an area frequently flooded.
It is located in the northern lowveld of Zimbabwe within
the Zambezi Basin. The Zambezi River itself flows from the
Kalene Hills in Zambia generally in an easterly direction
into the Indian Ocean. The river marks the boundary
between Zimbabwe and Zambia as well as Zimbabwe
and Mozambique. Two types of floods have been affecting
the Muzarabani area for decades. The first and most
frequent is the seasonal flood, which frequently occurs in
January or February, at the peak of the rainfall season. The
second and not so frequent one is the cyclone-induced
flood. In 2000 and 2003 tropical cyclones brought intense
storms that caused flooding in the Zambezi Basin, includ-
ing Muzarabani [14].
The unique natural setting and human engineered context
of Muzarabani makes it more susceptible to flooding. This is
because the area is downstream of Kariba Dam in the west,
but upstream of the Cabora-Basa Dam in the eastern direction
in Mozambique. To the south of Muzarabani is a range of
mountains known as Mavhuradonha which separate the
Zambezi valley from the highveld of Zimbabwe. All the
tributaries of the Zambezi River from the Zimbabwean side
40 E. Mavhura et al. / International Journal of Disaster Risk Reduction 5 (2013) 38–48
Composed of Influenced by
Knowledge
Types
Structures
Practices
Processes
Beliefs Values
World Views
Fig. 1. Framework for local knowledge on disaster preparedness.
Source: Dekens [5].
have their sources in this high-veld which is characterised by
normal to above normal annual rainfall of 700–1000 mm.
Tributaries of Zambezi River which pass through Muzarabani
are Musengezi, Musingwa, Hoya and Nzou-Mvunda.
When the water levels in Kariba Dam rise, it is released
to avoid dam failure. Most releases are conducted at the
peak of the rain season, between December and February
resulting in substantial increase in the discharge of the
Zambezi River. Further downstream Cabora-Basa dam
levels continue to rise due to inflows from the tributaries
of Zambezi and releases from Kariba Dam. If authorities in
Mozambique fail to open in time the flood gates of the
Cabora-Basa Dam or if the releases are exceeded by the
inflows from Zambezi River and its tributaries, backflow
flooding occurs. This happens when tributaries such as
Musengezi River fail to discharge into the Zambezi. As a
result water begins to pile up at the confluence of the
Musengezi and Zambezi leading to backflow flooding in
the whole area. Other tributaries of Musengezi River such
as Hoya, Musingwa and Nzou-Mvunda also tend to experi-
ence backflow flooding. On the other hand tropical
cyclones from the Indian Ocean result in flash floods in
the area. Disease outbreaks such as malaria and cholera
have been quite common during the flooding period [15].
As Muzarabani communities are directly or indirectly
dependent on the land, their lives and livelihoods are
jeopardised by the flooding. To reduce the risk of flooding,
a National Civil Protection Coordinating Committee (NCPCC)
is in place with a multi-sectoral representation for the
execution of civil protection functions. This multi-sectoral
representation is replicated at the provincial and district
levels. However, there are marked variations in the repre-
sentation as some organisations remain centralised and,
Local Knowledge System
In context of Resulting in Based on With effects on
Observation
Level of
Natural
Security/
Hazards and
Sustainability
Other Shocks Disaster Anticipation
Preparedness
at the local
level
Community
Adaptation
Resilience
Building
Global Factors
and Trends Communication
even those that are decentralised, the grades of staff
working at provincial and district level affects the manner
in which they are represented [16]. Disease outbreaks such
as malaria and cholera have been quite common during
the flooding period [15]. Subsistence farming which seems
to be the viable livelihood in Muzarabani is usually
destroyed by the floods.
The study was conducted in two communities of
Chadereka and Dambakurima in Muzarabani District,
Zimbabwe (Fig. 2). These two were chosen basing on their
history of flooding in the district. They are the most flood-
prone communities in Muzarabani district. During the
2008 floods, the worst floods since 2000, the two com-
munities were the most hit areas that attracted rescue
operations by the Department of Civil Protection (DCP).
The DCP is the government department in charge of all
civil protection issues in the country [15]. It has a National
Civil Protection Coordinating Committee (NCPCC) in place
with a multi-sectoral representation for the execution of
civil protection functions. This committee is replicated at
the provincial and district levels. The District Administra-
tor (DA) of Muzarabani is charged with the coordinating
role of disaster risk reduction, as empowered by the
Civil Protection Act of 1989. A District Civil Protection
Coordinating Committee (DCPCC) exists in Muzarabani but
it is inactive and has limited capacity to prepare for,
respond to or mitigate the effects of flooding. This leaves
the communities with the sole responsibility to decide as
individuals or households on what action to take in the
face of rising floodwaters. However, rescue operations are
conducted by the DCP during floods of high magnitude.
In the two communities under study, the majority of the
households are subsistence farmers. They either directly or
 E. Mavhura et al. / International Journal of Disaster Risk Reduction 5 (2013) 38–48
Fig. 2. Location of the study area.
Source: Authors.
indirectly depend on agriculture for their livelihoods.
Chadereka community is located mainly between Hoya
and Nzou-Mvunda rivers. A few households are situated
slightly on higher ground. The community is flooded
nearly every year. Flooding is predominantly of flash flood
type which is sudden and rapid. A combination of factors
including a rise of river-flow levels during the rainy season
as a result of intense rainfall occurring in the Mavhura-
donha Mountains located upstream, backflow of the Hoya
and Nzou-Mvunda rivers and the breaching of the land
separating the two rivers makes the area particularly
susceptible to flash flooding. The term chadereka in the
local language means low-lying land. This denotes how
susceptible the place is to flooding when running water
from the two rivers covers the land at great speed leaving
people without much time to evacuate with their
belongings.
41
Dambakurima community is also flooded regularly but
the character of flooding is different from Chadereka’s. The
land is slightly on higher ground as compared to Chadereka.
Few households are situated in-between tributaries of
Musingwa River and the backflow water is not as intense
as that of Chadereka given the small size of the tributaries
of this river system. However, floods develop gradually as
the water level rises in the flood plain of the Musingwa
river system due to excessive torrential rains and backflow.
The term dambakurima in the local language means failed
agriculture, or simply ‘do not farm in this land due to the
resultant food insecurity from flooding’.
4. Methodology
Primary data were collected between September
2011 and June 2012 through key informant interviews,
42 E. Mavhura et al. / International Journal of Disaster Risk Reduction 5 (2013) 38–48
household surveys and focus group discussions in Chader-
eka and Dambakurima communities. The three data col-
lection methods allowed for triangulation, which verified,
corroborated, enhanced the credibility and trustworthi-
ness or validity of the data [16,17]. Ten (10) formal inter-
views were conducted involving village heads, ward
councillors, school heads and nursing staff of the clinics
in the communities, the police and staff from the DA’s
office, Muzarabani. Some of the questions asked were:
What socio-economic and environmental impact do floods
have in your area? What coping strategies did the com-
munity employ during the 2008 flooding event? What
determines/influences the choice of the coping strategies?
What preventive strategies were employed by the com-
munity prior to the flooding event of 2008? The inter-
viewees were selected purposively while households for
the survey were identified using a simple random sam-
pling procedure from a total of 120 households.
The survey provided data on flood coping strategies
that members of the two communities employ before,
during and after a flood event. Some of the questions
asked were: What strategies do you use to save human
lives and household items from flooding? How do you
protect crops, poultry and other livestock from flooding?
How do you adapt to food insecurity and water scarcity
during flooding? A five-point Likert scale of rating (ranging
from strongly agree to strongly disagree) was used during
the same survey. The respondents were asked to indicate
their level of agreement or disagreement with a set of
statements dealing with flood impacts and the commu-
nities’ coping strategies. Chadereka community had 56
households and Dambakurima had 63 households that
participated in the survey, making a total of 116 house-
holds in the two communities. Of the 116 questionnaires
that were distributed; 107 were completed and returned,
giving a participation rate of 92.2 per cent.
One focus group discussion comprising 12 participants
purposively-selected was conducted in each community.
The same interview questions were used so as to explore
the role of indigenous knowledge in reducing the impact
of floods in Muzarabani district. The participants were
drawn from the communities concerned. Each group had
six male and six female participants of between 20 and 50
years of age. The selection of a gender balanced-group was
meant to ensure a balanced view of the coping strategies
employed by both men and women when faced by flood
disasters. Both descriptive and inferential statistics were
used to analyse the data.
5. Results
The results of this study reflect several of the elements
apparent in Dekens’ (2) framework for analysing traditional
knowledge in disaster preparedness. At the community
levels, disaster preparedness is a product of practices and
beliefs that are greatly influenced by community structures
within the local context. Therefore in this section, the
reasearchers first compare the flood impacts in Chadereka
and Dambakurima communities; then present the indigen-
ous flood prevention and coping strategies. Finally they
analyse the factors leading to variations in the adoption of
coping strategies.
5.1. Comparative analysis of flooding in Chadereka
and Dambakurima communities
Both communities experience recurrent flooding, but
this is more severe in Chadereka than in Dambakurima.
Chadereka community suffered moderate to high flooding
between 2001 and 2010 while Dambakurima endured only
light to moderate flooding. Over the same period, the
average flood duration was between two and three weeks
for Chadereka community and one week for Dambakurima
community. With regard to the depth of floodwater
(average flood-water height from the base of the dwelling
unit), there was more than one and half metres of water in
most households in Chadereka community compared to
one metre in Dambakurima. Thus, evidence of flood
impacts reveal that Chadereka is more prone to flooding
than Dambakurima. This has resulted in destruction of
crops, public infrastructure, loss of livestock and water
contamination. Table 1 lists other attributes pertaining to
flood impacts in the study communities.
5.2. Indigenous flood prevention and coping strategies
The coping strategies were considered in this context to
be the approaches people employ to deal successfully with
a crisis [18]. Residents of Chadereka and Dambakurima
villages developed their own coping strategies to guard
against flooding. These can be classified as structural or
non-structural; indigenous or modern. However, these
residents are used to relying on various indigenous stra-
tegies. The adoption of a particular set of strategies
depended on people’s socio-economic circumstances and
the characteristics of the flood. The study findings suggest
that a household’s response to a flood does not involve the
adoption of all strategies but rather the sequential imple-
mentation of preventative and mitigative initiatives [19].
The sequence associated with preventive strategies includes
the placing of barriers around a house, avoidance of
construction materials susceptible to cracking during
flooding, construction of floating houses and raising the
platform of the kitchens and storerooms so that they keep
food, water, fuel and valuables during flooding. Likewise,
the sequence associated with mitigative strategies includes
reducing the number of meals, relying on inexpensive
food, collecting wild fruits and honey, taking shelter on
higher ground with one’s personal belongings; searching
for alternative sources of income, and selling assets. It is
not necessarily the case, though, that all affected house-
holds move along this continuum; rather, it depends on
their level of vulnerability and their ability to absorb the
shocks of floods. The following subsections discusses over-
all preventive and mitigative strategies.
5.2.1. Strategies to save human lives and household items
Various earlier works suggest that in flood-affected local-
ities, coping starts with efforts to save people’s lives, such as
raising a homestead before a flood [21,22]. What is interesting
from the perspective of this study is the accompanying coping
 E. Mavhura et al. / International Journal of Disaster Risk Reduction 5 (2013) 38–48
Table 1
Comparative analysis of the 2008 flood impacts in Chadereka and Dambakurima communities.
Sources: Key informant interviews, household survey and focus group discussion with members from the two villages.
Flood impacts Chadereka
Riverbank erosion About three quarters of the land area eroded and
heavy siltation of the rivers
Soil fertility Improvement in soil fertility as a result of alluvium
deposits in flood plain
Width of the river Great widening of the Hoya river compounded
by stream-bank cultivation
Drinking water pollution All shallow wells and boreholes were inundated
Surface water pollution Decomposition of human and domestic waste
in stagnant water
Loss of crops Most crops were severely destroyed
Food insecurity during floods 75 per cent of households insecure
Common diseases and conditions Cholera, diarrhoea, malaria
Impact on school attendance Very high rate of absenteeism from school
Damage to public infrastructure Bridges swept away, roads severely damaged, blocks
of classrooms cracking
Loss of livestock Goats, pigs and chicken swept away
Damage to individual property Mud houses destroyed and toilets collapsed
measures people introduce to live in greater safety. Given
previous experience of flooding, communities tend to elevate
their bed using bricks or stones. Some use raised platforms
(dara)1 outside their houses as shelter and others seek for
safety in relatively higher ground. This is true in both
Chadereka and Dambakurima communities. Our findings
suggest that the building of dara and their application during
flooding is a normal practice in both communities.
Besides attempting to save human lives, communities
also try to rescue their assets such as furniture and kitchen
utensils. Initially they keep their kitchen utensils on mud
shelves built in the kitchen or outside on a dara. Such
shelves are built at a height of about a metre or two from
the ground so that kitchen utensils are not washed away
when floodwaters rises. Depending on the size of the dara,
these can be used to store fodder, ploughs, grains and any
other properties. Techniques to save property also vary
between households of the two communities. In both
communities 81.3 per cent of all respondents said they
preferred to build dara to protect their household items.
Few people (18.7 per cent) reported that they were
prepared to move to higher ground with all their posses-
sions during a flood event.
5.2.2. Techniques to save shelter
Techniques to protect shelter depend on the risk posed
by flood and erosion. Some studies conducted in flood
prone areas such as Bangladesh emphasise how villagers
built their dwelling units on raised land or on earthen
platforms so that water cannot reach the plinth level in a
low magnitude flood [19,20]. However, this study finds
that apart from the usual practice of erecting shelter on
1
Dara is a raised platform built outside the house to keep household
utensils and equipment. The platform can be a metre or three metres
from the ground level. Depending on its size, the dara can also be used to
stock fodder and as refuge during flooding.
43
Dambakurima
A quarter of the land area eroded and siltation
of the rivers increased
Improvement in soil fertility as a result of alluvium
deposits in flood plain
Small widening of the Musingwa river compounded
by stream-bank cultivation
All shallow wells and half of the boreholes were submerged
Decomposition of human and domestic waste
in stagnant water
Three quarters of crops were destroyed
41 per cent of households insecure
Cholera, diarrhoea, malaria
High rate of absenteeism from school
Damage to roads, bridges and isolated classroom
blocks developed cracks
Few goats, pigs and chicken were lost
Mud houses destroyed and toilets collapsed
raised land and a platform, community members avoid the
use of materials that are susceptible to cracking; rather
they prefer their traditional huts and houses which float
during flooding events. In addition some community
members sow coach grass (Cynodon dactylon) around their
houses to guard against erosion. About 60 per cent of
respondents in Chadereka said that the sowing of coach
grass (C. dactylon) around the houses was a common
technique to protect it from erosion, whereas in Damba-
kurima only 33 per cent reported employing this method.
People said that they had raised primarily the floor of their
bedrooms due to changing heights of floodwaters each year.
5.2.3. Strategies to protect crops
Subsistence farming is the viable livelihood in the area
and crops grown are only drought-resistant ones with
seasonal markets in the local area. This is done to increase
food security of the households. The farmers in both
communities have adopted different kinds of indigenous
coping strategies to protect their crops from the impacts of
flooding. These range from the selection of appropriate
varieties that suit the local climate, soil conditions and the
time frame of floods. Household survey data show that in
both communities, significant numbers of farmers follow
the crop calendar of the agro-ecological region 5 (in which
they are located), though the cropping pattern is not
similar in the two study communities. This region is an
arid one which receives erratic rainfall per year, making it
difficult for subsistence farmers who rely on rain-fed
agriculture for their living. In Dambakurima, 80 per cent
of farmers grow drought resistant crops such as millet,
cotton, rapoko and early-maturity varieties of maize
during the short rain season. This is mainly done in areas
far-away from the flood plain to deliberately avoid the
washing away of crops by floods. During the long dry
season of the year the same subsistence farmers grow
maize and vegetables using residual moisture from floods
44 E. Mavhura et al. / International Journal of Disaster Risk Reduction 5 (2013) 38–48
in the plains. This practice is commonly known as mud-
zedze2 and is done along the river bed, banks and flood
plains of Musingwa River. Farmers use zero tillage
whereby they prepare holes of more than 30 cm deep
and plant their seeds. Germination occurs and the crops
mature without irrigation. In this way farmers increase
their food security during the whole year.
In Chadereka, by contrast, only 25 per cent of the
respondents indicated that they were practicing mudzedze
and this improved their food security. A large proportion
of the community members (75 per cent) said they did not
have a piece of land along the flood plains of the rivers;
hence they could not practice mudzedze. This situation
made them vulnerable to drought and food insecurity
because their reliance on rain-fed agriculture often led to
low production of food. Hence, this study confirms the
findings of Rasid and Mallik [22] and Paul and Routray
[19], regarding agricultural cropping patterns and related
strategies in the case of Bangladesh.
5.2.4. Strategies to protect poultry and livestock
Poultry and livestock are important assets for house-
holds with low incomes in both areas. During the initial
stage of a flood, households keep their poultry and live-
stock in pens on slightly higher parts of their homestead.
When this becomes unsafe, they move them to safer
places on higher ground, or they sell the poultry and
livestock to outsiders. Household survey data show that
51.4 per cent of respondents in Chadereka prefer to move
their poultry and livestock to safer places on higher
ground, while in Dambakurima 23 per cent of them prefer
to keep their livestock on a raised part of their homestead.
When all such measures fail, 51.5 per cent of respondents
in both communities reported selling their poultry and
livestock to outsiders.
5.2.5. Coping strategies to conserve food and water
The storage of food and water is a big challenge for
victims of a flood. People prefer and use polythene bags to
stockpile dry food, grain and seed. These are then kept in a
special hut called dura.3 Seventy-five per cent of the total
households in both communities prefer plastic containers
to transport water as they are easy to seal and carry.
In most houses, the storing place is higher than the normal
floor; such places remain safe until there is high flood-
water. However, when floodwater increases, communities
usually relocate to higher ground with their containers of
food and water. About 60 per cent of all the respondents
have raised the basement level of their storerooms.
Most of the study participants expressed a great need
for pure drinking water during and immediately after a
2 by the scale of the flooding. In effect, changes in the size of
Mudzedze: This is a traditional practice to grow food crops along
flood plains during the dry season. The crops are not irrigated; rather
they use residual moisture from flooding. The practice leads to heavy
siltation of rivers and dams downstream.
3
Dura: This is a traditional hut meant to store grain and other food
stuffs. The hut is raised from the ground floor by large stones and
hardwood logs. It is then built using wood or bricks. The design of the
dura is varied. Some have doors at about a metre from the ground and
others have a hole at the top that is used to access the food.
flood. A common observation is that post-flood diseases
spread through the drinking of contaminated water. Peo-
ple are aware of the risk of consuming contaminated water
but often they are unable to procure pure water as shallow
wells and boreholes are submerged and water purifying
materials are not available. It is impossible to boil water
frequently as traditional fuel resources such as cow dung
cake and fuel wood become wet during the flood. This
finding is important, as whilst we found many examples of
successful adaptation strategies, health risks had not been
overcome across this region. This reality is made highly
evident by the outbreaks of cholera particularly during the
latter parts of 2008 until 2009 during the season for
flooding [24].
5.2.6. Adaptations to food insecurity
Scarcity of food during and after a flood is a common
phenomenon in flood-affected areas. When faced with
such an insufficient food supply, the household head is
primarily responsible for feeding the family members.
He or she adopts different measures to cope with such a
situation, including reducing the number of meals per day
and relying on inexpensive food stuffs such as vegetable
leaves and wild fruits. The study confirms that in both
communities, skipping a meal is a coping strategy and that
there are variations in dependency on inexpensive food.
Approximately 38 per cent of respondents in Dambakur-
ima and 62 per cent in Chadereka had decreased their
meals and increased their dependency on inexpensive
food. The situation is more severe in Chadereka because
of the high impact of flooding and the poor economic
circumstances of most residents. By contrast, respondents
in Dambakurima said that they had reduced their number
of meals because of wet firewood during and immediately
after the flood. Consequently they prepared two light
meals per day.
Households in both communities employ many other
coping methods during a flood. Among households
exposed to shocks, for example, borrowing money after a
flood [23] and disposing of assets are considered very
important initiatives. The study found that the most
common assets sold in both communities to overcome
difficult periods and manage crisis are cattle, pigs, goats,
chicken and sheep.
5.3. Variations in the adoption of coping practices: selective
determinants
Every locality might have some established or tradi-
tional coping strategies, but the type of response adopted
by community members and its effectiveness may vary
over time [22]. Many of these coping mechanisms may fail,
and not because people’s ability to cope is overwhelmed
the population and the economy, in the local market and
the environment, and in the source of the livelihood of
each household and in flood characteristics can make the
coping mechanisms outdated [24]. The study’s findings
suggest that traditional strategies make a positive con-
tribution to improving people’s adaptability to a flood
hazard, but as a whole, it is a complex process, linked to
 E. Mavhura et al. / International Journal of Disaster Risk Reduction 5 (2013) 38–48 45

other physical and socioeconomic variables. Therefore the
discussion of variations in the adoption of coping strate-
gies that follows considers endogenous factors, such as
income, education, and occupation, and exogenous factors
such as external assistance, flood characteristics and
riverbank erosion.
relation to disposing assets. High income groups are able
to cope with a disaster without selling their assets. Low
income groups are more likely to cope with the situation
by disposing their assets, though the contrasts are less
extreme than the case of food availability and differences
in eating behaviour (Table 2).

5.3.1. Household’s income level’s influence on traditional
coping strategies
The income level of a household is closely linked to
coping strategies. Households with high income or savings
can readily help themselves during a flood event and as
such are less vulnerable to flood impacts [25]. Table 2
shows that the income level of households in the two
communities is associated with the availability of food
during and after a flood. Food is available to the high and
middle income groups during the flooding event, while
most of the low income groups do not have sufficient food
to overcome the situation. As a result, the ability to cope
with flooding in terms of the storage of food is greater
among high-income groups than it is among the low-
income groups. Key in this regard is the relatively high
agricultural production of the former groups as they
possess more landholdings and farming inputs than the
later farmers; increasing savings and reducing their vul-
nerability to a potential shortfall in income in a times of
disaster [19].
In addition, eating behaviour fluctuates among differ-
ent income groups. Households in low- and a few middle-
income groups prefer to reduce their number of meals and
to rely on inexpensive food. These groups have few resources
to buy and store food for a prolonged period of flooding.
Interestingly, the behaviour of the high income group and
a few middle income groups is quite different. They prefer
to reduce the number of daily meals than consume
inexpensive food. One of the reasons for this could be that
the group is quite aware of the susceptibility to water-
borne diseases during and after a flood. Similar contrasting
behaviour is evident among various income groups in
5.3.2. Level of education as a determinant of coping
practices
Education is one of the important determinants of
coping and adaptation measures to increase one’s resi-
lience to disasters and enhance one’s quality of life [26].
Education level is also very important in generating
awareness of flood forecasting. Flood warnings can reduce
damages experienced by flood victims [27]. However,
people’s response to a flood warning depends on accurate
dissemination and reliability of the flood-forecasting infor-
mation. Similarly, capacity to understand flood forecasting
varies among people with different education levels.
Table 3 shows that all household heads with a minimum
of diploma level education and 77.8 per cent of those with
secondary-schooling are able to capture flood forecasting
information, whereas the numbers gradually decrease
among households heads with only primary-school level
education and among those who are illiterate. Therefore,
the higher the education level, the greater one’s capacity
to understand flood forecasting and to reduce one’s
vulnerability to flooding. Whilst there are complexes
of other factors important to flood early warning
[29] the basic education indicator was found to be very
important here.
5.3.3. Occupation’s influence on coping practices
The occupation of a household head is another impor-
tant variable that influences the adoption of coping stra-
tegies. The study found that food was available to
subsistence farmers whose heads of families were formally
employed, while labourers and vendors had limited
amount of food. This is because such families are able to

Table 2
Household head’s income, food availability, eating behaviour and disposal of assets.

Level of income of household head

Low (0–100 USD Middle (101–500 USD High (above 500 USD Total
per month) per month) per month)

No. of Households (%) No. of Households (%) No. of Households (%) No. of Households (%)

Availability of food
Yes 20 31.7 19 57.6 8 72.7 47 43.9
No 43 68.3 14 42.4 3 27.3 60 56.1
Total 63 100 33 100 11 100 107 100
Changes in eating behaviour
Reduced number of meals 16 25.4 18 54.5 9 81.8 43 40.2
Reduced meals and rely on less expensive 47 74.6 15 45.5 2 18.2 64 59.8
food (both)
Total 63 100 33 100 11 100 107 100
Disposal of assets
Yes 30 47.6 18 54.5 2 18.2 50 46.7
No 33 52.4 15 45.5 9 81.8 57 53.3
Total 63 100 33 100 11 100 107 100
46 E. Mavhura et al. / International Journal of Disaster Risk Reduction 5 (2013) 38–48

Table 3
Household head’s education level and capacity for flood forecasting.

Capacity to capture flood forecasting Illiterate Primary school Secondary school Diploma level and above Total
information
HHa (%) HH (%) HH (%) HH (%) HH (%)

Yes 8 34.8 27 54.0 14 77.8 16 100.0 65 60.7

No 15 65.2 23 46.0 4 22.2 0 0.0 42 39.3
Total 23 100.0 50 100.0 18 100.0 16 100.0 107 100.0

a
HH¼Number of households.

Table 4 5.3.5. Flood characteristics as a determinant of coping

Occupation of household head and food availability. practices
Flood characteristics, such as the height of floodwater,
Farming Vending and Formal Total
menial work employment the frequency and duration of flooding, influence people’s
adoption of coping strategies [19]. A longer period of
HHa (%) HH (%) HH (%) HH (%) stagnating water has more severe ramifications than
floodwater height. If floodwater remain stagnant for a
Availability of food
Yes 30 35.7 1 12.5 12 80.0 43 40.2
long time in a locality, the level of damage rises and hence
No 54 64.3 7 87.5 3 20.0 64 59.8 people’s ability to cope decreases. Similarly, pollution
Total 84 100.0 8 100.0 15 100.0 107 100.0 increases if water remains stagnant for a long time and
Loan to meet expenditure coping strategies fail in relation to health issues. It is
Yes 53 63.1 6 75.0 6 40.0 65 60.7
evident in this study that floodwater which stays stagnant
No 31 36.9 2 25.0 9 60.0 42 39.3
Total 84 100.0 8 100.0 15 100.0 107 84 for a long time, disrupts people’s food and drinking water
supply, reducing their ability to cope and augmenting their
a
HH¼Number of household. vulnerability. Thus there is a direct link between the
effectiveness of the coping strategies, the nature of flood-
outsource the food and store it safely among others. The ing and respondent’s socioeconomic conditions, with
study also revealed that borrowing during and after a flood lower income respondents being the most vulnerable.
was highest among the labourers and vendors, and in
contrast to those in formal employment (Table 4). This is 5.3.6. Distance from riverbank’s effect on coping practices
because the labourers and vendors become jobless during Localities adjacent to a riverbank are more prone to a
flooding. flood hazard. A great volume of floodwater first enters
open land along the river, damaging cropland and home-
steads, and subsequently the interiors of houses. People far
5.3.4. Influence of external assistance in the choice away from the rivers are not as vulnerable to flood risk of a
of coping practices similar intensity. Chadereka community is mainly located
External assistance such as a food relief programme has between Hoya and Nzou-Mvunda a river, which is prone to
a significant impact on a household’s capacity to respond riverbank erosion, with the severity of erosion mounting
to shocks and the timing of relief can play an important during cyclonic flooding. People in this community are
role in determining the effectiveness of coping strategies more vulnerable to flooding than their counterparts in
[28]. This study’s findings review that in addition to Dambakurima, primarily because their proximity to river-
providing food aid, some NGOs and relatives in the bank erosion and the related nature of flood.
diaspora, are also assisting the communities with inputs
and other materials to revive agricultural- and livelihood-
related activities. Food relief is a necessity but providing 6. Conclusion
food-production assistance is more sustainable and can
enhance victims’ ability to cope [29]. The study identified This research finds that flooding has disastrous impacts on
basic needs such as food, drinking water, temporary people’s socio-economic conditions as well as on the environ-
shelter and access to health services, among other vari- ment that supports them. The impact depends not only on the
ables, as important during a flood event; whereas agricul- magnitude of the event but also on some other variables such
tural inputs, employment and reconstruction of resilient as income, lack of awareness, level of education, occupational
houses are considered vital needs after a flood. Therefore, structure and the physical location of the area. Therefore the
providing such relief to those who have experienced findings are consistent with research conducted in Bangladesh
greater loss or who are poorer in the post-disaster phase [19,30,31] showing that flood problems are not merely a
can help create an enabling environment for their coping hydraulic dynamic; rather they are linked to issues of demo-
strategies. If relief is made early enough and is based on graphy, ecology, education, settlement pattern, society, socio-
victim’s priorities, this may help households to save at economic status, culture and politics.
least some of their asset base and avoid selling them or It is evident that in response to a flood, people adopt
reaching the final stage of destitution [19]. different indigenous preventive and mitigative measures
 E. Mavhura et al. / International Journal of Disaster Risk Reduction 5 (2013) 38–48
in a sequential order. The implementation of such strate-
gies may fluctuate due to variations in the aforementioned
factors. However, the study indicates that coping techni-
ques adopted by households fall into distinct stages:
preventive measures, adaptive measures and distress
mitigation. Indigenous coping strategies tend to be effec-
tive in low magnitude floods. As Thompson and Tod [21]
warned, these strategies have limited effect in reducing
damage in more extreme cases.
The adoption of coping strategies reveals an important
pattern in how different households respond to a flood in
relation to factors such as education, flood characterises,
income, occupation and riverbank erosion. A high income
level of household is accompanied by increased access to
food and safe drinking water, leading to a greater ability to
cope with a disaster without relying on inexpensive food
items or selling of assets. Similarly, the higher the educa-
tion level the greater the scope to capture flood forecasting
information that reduces vulnerability to disaster and
enhances a household’s ability to cope with a disaster.
Occupation has a similar influence. Salaried households
who are also subsistence farmers, for instance, have a
more stable income than vendors and thus have a greater
ability to procure food and understand flood forecasting
information. Flood characteristics, meanwhile, have a
negative influence on ability to cope when the flood is
prolonged and its height increases, coupled with the
associated loss and damage that limits access to food and
drinking water. Households close to a riverbank are more
susceptible to flooding and erosion, increasing their vul-
nerability and reducing their ability to cope.
External assistance such as relief at the proper time and
appropriate distribution can play a pivotal role in deter-
mining the effectiveness of coping strategies. Haque and
Zaman [31] and Paul and Routray [19] have made a similar
argument that external support and assistance along with
social capital at the community level significantly contri-
bute to swift recovery from a flood. In fact, it is vital to
ensure assistance from government and NGOs during and
after a flood to minimise the hardships that confront
victims, and to help them regain their pre-disaster status
and move forward. External support combined with social
capital can significantly improve the capacity of people
and reduce vulnerability to floods.
Finally, this study reveals that although flooding in
Muzarabani generates socioeconomic and health related
hazards and environmental and infrastructural damage,
indigenous coping strategies can significantly reduce vul-
nerability to disaster. However, many of the adaptive
strategies employed appear to be short-term measures
which are not sustainable. This is reflected by the external
assistance that the community receives after a high
magnitude flood event. The strategies are effective only
in a low magnitude flood: when the floodwaters rise and
cross a critical threshold, people have no choice but to
relocate to higher ground. A further study in this area
would need to consider the integration of proactive set-
tlement planning and policy into the disaster management
framework there. Underlying this approach however
should be due attention to people’s indigenous knowledge
systems and flood mitigation strategies. It is not suggested
47
these present a panacea; this understanding would be
complemented by additional inputs from early warning
systems and appropriate external assistance during and
after a flood so as to further enhance victims’ ability to
cope. Ultimately people aspire to moving beyond a life
focussed on coping and being resilient.
Acknowledgements
We would like to thank the Development Partnerships
in Higher Education (DelPHE) Project funded by the British
Council at Bindura University of Science Education for
their financial assistance. We would like also to express
our gratitude to the DelPHE team for their valuable
support during the action research.
References
[1] Paton D, Johnston D. Disaster resilience: an integrated approach.
Springfield: Charles Thomas; 2006.
[2] UNISDR, Building back better for next time. Bangkok: UNISDR Asia
and Pacific; 2010.
[3] Cockcroft JD. Mexico: “failed states,” new wars, resistance. Monthly
Review—an Independent Socialist Magazine 2010;62:28–41.
[4] UNISDR. Building the resilience of nations and communities to
disasters: hyogo framework for action 2005–2015. In: World Con-
ference on Disaster Risk Reduction, UNISDR, 2005.
[5] Dekens J. Local knowledge for disaster preparedness: a literature
review. Kathmandu: International Centre for Integrated Mountain
Development; 2007.
[6] Wilby RL, Keenan R. Adapting to flood risk under climate change.
Progress in Physical Geography 2012;36:348–78.
[7] Gaillard JC, Mercer J. From knowledge to action: bridging gaps in
disaster risk reduction. Progress in Human Geography 2012;37:
93–114.
[8] Mercer J, Gaillard JC, Crowley K, Shannon R, Alexander B, Day S, et al.
Culture and disaster risk reduction: lessons and opportunities.
Environmental Hazards 2012;11:74–95.
[9] Cadag JRD, Gaillard JC. Integrating knowledge and actions in disaster
risk reduction: the contribution of participatory mapping. Area
2012;44:100–9.
[10] Sillitoe P. The development of indigenous knowledge: a new applied
anthropology. Current Anthropology 1998;39:223–52.
[11] Mercer J, Kelman I, Taranis L, Suchet-Pearson S. Framework for
integrating indigenous and scientific knowledge for disaster risk
reduction. Disasters 2010;34:214–39.
[12] Tran P, Takeuchi Y, Shaw R. Indigenous knowlegde in river basin
management. In: Shaw R, Sharma A, Takeuchi Y, editors. Indigenous
knowledge and disaster risk reduction: from practice to policy.
New York: Nova Science; 2009. p. 45–57.
[13] Shaw R, Sharma A, Takeuchi Y. Introduction: indigenous knowledge
and disaster risk reduction. In: Shaw R, Sharma A, Takeuchi Y,
editors. Indigenous knowledge and disaster risk reduction: from
ptractice to policy. New York: Nova Science; 2009. p. 1–13.
[14] Madamombe EK. Zimbabwe: flood management practices—selected
flood prone areas zambezi basin. WMO/GWP Associated Programme
on Flood Management; 2004.
[15] Department of Civil Protection. Disaster risk management: a
resource book for educational institutions in Zimbabwe. Harare:
Civil Protection Organisation in Zimbabwe; 2009.
[16] Gershberg AI. Decentralisation, recentralisation and performance
accountability: building an operationally useful framework for
analysis. Development Policy Review 1998;16:405–31.
[17] Gibson N. Action research. In: Lewis-Beck MS, Bryman A, Liao TF,
editors. The sage encyclopaedia of social science research methods,
vol. 1. Thousand Oaks: Sage; 2004. p. 4–6.
[18] Clapham C. The challenge to the state in a globalized world.
Development and Change 2002;33:775–95.
[19] Paul SK, Routray JK. Flood proneness and coping strategies: the
experiences of two villages in Bangladesh. Disasters 2010;34:
489–508.
48 E. Mavhura et al. / International Journal of Disaster Risk Reduction 5 (2013) 38–48
[20] Clark P, Lees F, Greaney M, Greene G, Riebe D. Resilience in exercise
and diet: factors predicting “getting back on track” in a community-
dwelling older adult population. Gerontologist 2008;48:585–6.
[21] Thompson P, Tod I. Mitigating flood losses in the active floodplains
of Bangladesh. Disaster Prevention and Management 1998;7:113–23.
[22] Rasid H, Mallik A. Flood adaptations in Bangladesh: Is the compart-
mentalisation scheme with indigenous adjustments of rice cropping
to flood regimes? Applied Geography 1995;15:3–17.
[23] Del Ninno C, Dorosh PA. Public policy, markets and household coping
strategies in Bangladesh: avoiding a food security crisis following the
1998 floods., World Development 31 (2003) 1221-1238.
[24] Few R. Flooding, vulnerability and coping strategies: local responses
to a global threat. Progress in Development Studies 2003;3:43–58.
[25] Green C, Veen A, Wierstra E, Penning-Rowsell E. Vulnerability refined:
analysing full flood impacts. In: Penning-Rowsell EC, Fordham M, editors.
Floods across Europe: flood hazard assessment, modelling and manage-
ment. London: Middlesex University Press; 1994. p. 32–68.
[26] Tong T, Shaw R, Takeuchi Y. Climate disaster resilience of the education
sector in Thua Thien Hue Province, Central Vietnam. Natural Hazards
2012;63:685–709.
[27] Smith K, Petley DN. Environmental hazards: assessing risks and
reducing disasters. London: Routledge; 2009.
[28] Corbett J. Famine and household coping strategies. World Develop-
ment 1988;16:1099–112.
[29] Devereux S. Livelihood insecurity and social protection: a re-
emerging issue in rural development. 2001;507–19.
[30] Kunii O, Nakamura S, Abdur R, Wakai S. The impact on health and
risk factors of the diarrhoea epidemics in the 1998 Banlgadesh
floods. Public Health 2002;116:68–74.
[31] Haque CE, Zaman MQ. Human response to riverine hazards in
Bangladesh: a proposal for sustainable floodplain development.
World Development 1993;21:93–107.