AN INVESTIGATION OF THE EXTREME HYDRO-CLIMATIC
CHARACTERISTICS AND THE UNDERLYING CAUSES OF
RIVER NYANDO CATCHMENT, LAKE VICTORIA BASIN-KENYA
BY
RONALDSUNGUbG~U
156/10513/04
A THESIS SUBMITTED IN PARTIAL FULFILLMENT FOR THE AWARD OF
DEGREE OF MASTE,R OF SCIENCE (HYDROLO@Y & WATER RESOURCES) IN
THE SCHOOL OF PURE AND APPLIED SCIENCES OF KENY ATT A UNIVERSITY
DECEMBER 2008
Og~tu,Ronald Sungu
An mvestigation of
the extreme
11111111111111111111
2009/lJ9349
KENYATTA UNIVERSITY LIBRARY·
DECLARA TION
This Thesis is my original work and has not been presented for a degree in any other
university
Signatu~----
Ronald Sungu Ogutu
This Thesis has been submitted with our Approval as University Supervisors
Dr. Christopher M. Ondieki
Geography Department
Kenyatta University
,
Signature--~~ Date-----~Jl~~-~
Dr. George Lukoye Makokha
Geography Department
Kenyatta University
Signature-----~ Date----~-l-Jl-~--l--tl-~
DEDICATION
I dedicate this work to the Almighty and Everlasting God through whom everything is
possible. It is through him that the door was opened and this achievement realized.
11
ACKNOWLEDGEMENTS
I thank God for enabling me reach this far, so far. I express my gratitude to my supervisors
Drs. C.M. Ondieki, and G. L. Makokha for their continued guidance and support throughout
the project work. I acknowledge the support and encouragement I got from Dr. G. A.
Olukoye, Professors S. N. Chhabra and C. Shisanya, Drs. Joy A. Obando, Prasad and the
entire staff of Geography Department. I appreciate the cordial relationship I enjoyed from my
colleagues Messrs. Nathan Muli, Elly Ajigo, John Mwangi and Michael Maunda. I miss my
late colleague ML Julius Wanyonyi, may The Almighty God rest his soul in peace and bless
his family. Lastly I thank my wife and children for being by my side during the programme.
This victory is ours all, and my prayer is may God bless us all.
1
1ll
TABLE OF CONTENTS
DECLARATION i
DEDICATION ii
ACKNOWLEDGEMENT iii
TABLE OF CONTENTS............................................................. iv
LIST OF FIGURES vi
LIST OF TABLES vii
LIST OF PLATES viii
OPERATIONAL TERMS ix
ACRONYMS AND ABBREVIATONS xi
ABSTRACT xii
CHAPTER ONE: INTRODUCTION 1
1.1 Background 1
1.2 Statement of the Problem 3
1.3 Justification 3
1.4 Objectives 5
1.5 Significance of the Study 5
CHAPTER TWO: LITERATUREREVIEW 7
2.1 General 7
2.2 Extreme events 11
2.2.1 Floods 13
2.2.2Droughts 17
2.3 Impacts of extreme events 19
CHAPTER THREE: METHODOLOGy 29
3.1 Description of the study area 29
3.1.1 Location and topography 29
3.1.2 Geology and Soils 31
3.1.3 Climate 35
3.1.4 Drainage and Hydrological Network of the Nyando Basin 36
3.1.5 Population and Socio-Economic Activities 37
3.2 Data Collection 41
3.3 Data Analysis 41
3.3.1 Rainfall Analysis for Nyando basin 41
3.3.2 Flow Analysis for Nyando Basin : 43
3.3.3 Runoff Ratios for the Nyando basin 44
3.4 Land cover and land use in the Nyando basin 44
3.4.1 Land cover 45
3:4.2 Land use 45
3.4.3 Population Analysis 46
3. 4.4 Built -up areas 48
3.4.5 Sand Harvesting 48
3.4.6 Brick making 48
3.4.7 Infrastructure Analysis 49
3.5 Topography - 49
CHAPTER FOUR: RESULTS AND DISCUSSION ~ 50
4.1 Introduction 50
4.2 Rainfall Analysis for the Nyando basin 50
4.2.1 Data Quality Analysis for the Nyando basin 50
4.2.2 Seasonal distribution of rainfall in the Nyando basin 51
4.2.3 Annual Spatial Distribution of Rainfall in the Nyando basin 53
IV
4.2.4 Yearly Rainfall Variation in the Nyando Basin 54
4.2.5 Annual rainfall variation at Kaisugu Rainfall station 54
4.2.6 Annual Rainfall Variation at Ahero Rainfall Station No. 9034086 55
4.2.7 Annual Rainfall Variation at Kipkelion Rainfall Station No. 9035020 56
4.2.8 Annual Rainfall Station at Muhoroni Rainfall Station No. 9035315 57
4.3 Variability and Reliability of rainfall in the Nyando Basin 58
4.3.1 Kaisugu Rainfall Station 58
4.3.2 Ahero Rainfall Station 59
4.3.3 Kipkelion Rainfall station 59
4.3.4 Muhoroni Rainfall station 59
4.4 Flow Analysis in the Nyando River 61
4.4.1 Seasonal Discharge Variations in the Nyando Basin 62
4.4.2 Annual Flow Variability in the Nyando basin (1970-2000) 63
4.4.3 Annual Peak flows at 1GD03 64
4.4.4 Annual Mean Flows at 1GD03 65
4.4.5 Annual Minimum Flows at 1GD03 65
4.4.6 Runoff Ratios for the Nyando basin 65
4.5 Flood Frequency Analysis for the Nyando basin 67
4.6 Land cover and land use in the Nyando basin 69
4.6.1 Land cover 69
4.6.2 Land Use 72
4.7. Population in the Nyando basin 73
4.8 Built - up Areas and Settlements 77
4.9 Intensive Agricultural Activities 78
4.10 Livestock keeping in the Nyando basin 82
4.11 Brick making on the Nyabondo plateau 90
4.12 Deteriorating Infrastructure 94
4.13 Topography 96
CHAPTER FIVE: SUMMARY OF FINDINGS, CONCLUSSIONS AND
RECOMMENDATIONS 103
5.1' Summary and Conclusions 103
5.2 Recommendations 106
REFERENCES 112
APPENDICES 116
•..
v
ENVil II 11R.1I"'-"It'I"rU I •••nOil • __ ;;po
LIST OF FIGURES
Figure
Figure 1 The Nyando Basin
Figure2 The Nyando Basin topography
Figure 3 The Nyando Basin Hydrological Network
Figure 4a Nyando Basin Population Density, 1979 (ICRAF, 2002)
Figure 4b Nyando Basin Population Density, 1999 (ICRAF, 2002)
Figure 5a Nyando Basin Land Sat image (Unclassified, 1973)
Figure 5b Nyando Basin Land Sat image (Unclassified, 1986)
Figure 5c Nyando Basin Land Sat image (Unclassified, 2003)
Figure 6 Double Mass curve: Ahero rainfall station
Figure 7 Double Mass curve: Kaisugu rainfall station
Figure 8 Seasonal rainfall variation in the
Figure 9 Annual rainfall distribution in the Nyando Basin
Figure 10 Annual rainfall variation-Kaisugu rainfall station
Figure 11 Annual rainfall variation-Ahero rainfall station
Figure 12 Annual rainfall variation- Kipkelion rainfall station
Figure 13 Annual rainfall variation-Muhoroni rainfall station
Figure 14 Rainfall Reliability in the Nyando Basin
Figure 15 Mean monthly discharges in the Nyando Basin
Figure 16 Annual flow variability for GD03-Nyando basin
Figure 17 Nyando Basin (Classified image, 1973)
Figure 18 Nyando Basin (Classified image, 1986
Figure 19 Nyando Basin Land Cover (Classified image, 2003)
Figure 20 Nyando Basin - Food and Cash crops (Classified, 2003)
Figure 21 Nyando Basin - Subsistence crops (Classified, 2003)
Figure 22a Nyando Basin topographical map 1
Figure 22b Nyando Basin Topographical map 2
Figure 23 Nyando Basin Boundary Elevations DEM
Page
30
32
37
39
39
46
47
48
51
52
52
53
55
56
57
58
61
63
64
71
71
72
81
82
97
98
99
VI
LIST OF TABLES
Table 1 Water availability per capita in the Nyando Basin 9
Table 2 Erosion hazard rating for River Nyando catchment area 11
Table3 Prospective Dam sites in the Nyando Basin 15
Table 4 Results ofDUFLOW model 16
Table 5 Impacts of floods in the Nyando Basin 21
Table 6 Area covered by river Nyando and its tributaries 29
Table 7 Types of soils and their associated problems 34
Table 8 Population in the Nyando Basin 38
Table 9 Principal Urban Centres 38
Table 10 Population density in the Nyando Basin 38
Table 11 Rainfall variability and Reliability in the Nyando basin 59
Table 12 Runoff Ratios for the Nyando basin 66
Table 13 Flow Return periods for the Nyando river 67
Table 14 Peak Discharges for different return periods 68
Table 15 Land cover and Land use changes in the Nyando Basin 70
Table 16 Population in the Lower Nyando Basin flood prone area 76
Table 17 Population projections for the Nyando Basin 77
Table 18 Demographic and Biophysical Characteristics
of the Lake Victoria Basin 100
Table 19 Proposed Remedial measures for the Nyando Basin•.. 111
Vll
LIST OF PLATES
Plate Page
Plate 1 Fragile soils in the middle Nyando catchment 10
Plate 2 Evacuation of flood victims in the lower Nyando basin 23
Plate 3 Intensive farming in the lower Nyando catchment 79
Plate 4 Intensive farming & Human settlement Lower Nyando catchment 80
Plate 5 Livestock keeping in the lower catchment 83
Plate 6 Deforestation at Tinderet hills- ICRAF 2000 86
Plate 7 Charcoal burning in Nandi hills- ICRAF 2000 86
Plate 8 Gully erosion in the middle 88
Plate 9 Gully erosion in river Bugo 22/4/2006 88
Plate 10 Severe gully erosion at Katuk Odeyo 89
Platell Land degradation in the lower reaches - Sand harvesting at Bugo 89
Plate 12 Land Degradation around Pap Onditi 25/ 4/2006 90
Plate 13 River Nyando meanders- the lower catchment ICRAF 2000 90
Plate 14 Brick- making; Soil degradation in the Nyando basin 20/5/2007 92
Plate 15 Brick-making in the Nyando basin 30/5/2007 93
Plate 16 Burnt bricks, Nyando basin 16/6/2007 93
Plate 17 Dykes downstream of Ahero bridge 22/4/2006 95
Plate 18 Trees trapped under Ahero bridge 95
Plate 19 Flooded Nyando river at Ahero bridge 20/4/2006 96
Vlll
OPERATIONAL TERMS
Hazard: A potentially damaging physical event, phenomenon or human activity that may
cause loss of life, property damage, social and economic disruption or environmental
degradation.
Exposure: Quantification of receptors that may be influenced by an extreme event a flood or
a drought, for example, the number of people and their demographics, number and type of
properties. Improvement of dykes and proper evacuation activities would reduce the severity
of flood exposure of life and property in a flood plain.
Vulnerability: The conditions determined by physical, social, economic and environmental
factors or processes, which increase the susceptibility of a community to the impact of
hazards. Vulnerability to floods is a combination of complex dynamic and inter-related
mutually reinforcing conditions that can be divided into three major groups namely: physical
or material, constitutional or organizational and motivational or attitudinal. Social factors
contribute to, or influence these conditions to determine community vulnerability. Some of
the social factors relevant to flood management include poverty, livelihood opportunities,
cultural beliefs and special needs of weaker social groups. The level of vulnerability of a
particular community to extreme events calls for an equal magnitude of preparedness and
quick response to mitigate against the losses and damage.
Preparedness: includes preventive and precautionary measures against an event before it
occurs. It aims at minimizing the effect of development activities on accentuating the
magnitude of hazards, reducing the exposure to natural hazards, minimizing the socio-
economic vulnerability of people and material assets exposed to these hazards. Prevention
IX
deals with long-term planning and is incorporated into development process. Preparation
deals with reducing the vulnerability at local level and limiting the extent of adverse impact
of the inevitable event in the short term. The goal of emergency preparedness programme is
to achieve a satisfactory level of readiness to respond to any emergency situation through
programmes that strengthen the technical and managerial capacity of all the stakeholders; the
government, organizations and the community. These measures entail the development of
plans to save lives, minimize disaster damage and enhance disaster response operations. The
level of preparedness against an extreme event is only possible when the characteristics of the
events are known. This will enable timely and coordinated response to be meted.
Response: are measures that limit the effects of exposure to a hazard and its duration. It
focuses on alerting people, rescuing victims and providing assistance in cases of need and
measures taken to prevent further adverse impacts, provisional reconditioning of important
infrastructure and documenting events. Their effectiveness depends on the availability of
information on hazards, emergency risks and the countermeasures to be taken and on the
degree to which the government agencies, non-governmental organizations and the general
public are able to make use of the information.
x
ACRONYMS AND ABBREVIATIONS
BOD
CNA
COD
ICRAF
IRCCWSS
Biochemical Oxygen Demand
Climate Network Africa
Chemical Oxygen Demand
International Centre for Research in Agro-Forestry
International Reference Centre for Community Water
Supply and Sanitation
Inter Tropical Convergence Zone
Japanese International Cooperation Agency
Kenya Meteorological Department
Kenya Soil Survey
Lake Basin Development Authority
Lake Victoria Environmental Management Programme
Millennium Development Goals
Ministry of Water Resources and Development
National Water Master Plan
Strategy for Flood Management in Lake Victoria Basin
World Health Organization
World Meteorological Organization
ITCZ
JICA
KMD
KSS
LBDA
LVEMP
MDGs
MWRD
NWMP
SFMLVB
WHO
WMO
Xl
ABSTRACT
Extreme events, floods and droughts are natural occurrences which have caused numerous
disasters world over. On a global scale the number of people affected and economic damages
due to flooding and droughts are on the rise. Most of the deaths, post traumatic stresses and
social and economic hardships can either be tremendously reduced or avoided altogether if
the characteristics of the extreme events and the factors that cause and intensify them are
understood and accordingly managed. The study picked the Nyando river catchment because
of its vulnerability to these extreme events. Almost on an annual basis there are adverse
reports on this area due to floods and to some extent on droughts. The purpose of the study
was to come up with ways of reducing the effects of the extreme events. The objectives were
to determine the causal factors of the extreme events, flow characteristics and to establish the
effects of land cover and land use on extreme flows in the catchment. Data was collected
from MWRD, KMD, LBDA, LVEMP, ICRAF, MP&ND and on site. Flow data was
analyzed using Gumbel, Log Pearson-3 and General Extreme Value (GEV) fitting
distributions. Rainfall data analysis was done to determine variability and reliability for
Water Resources planning. Land cover and Land use analysis were done using ARCVIEW
and ARCGIS from unclassified land sat images generated from P160R60, P161R60,
P170R60 and P171R60 using GIS-IDRISI programme. The results indicate that flood
frequency is on the increase, but contrastingly rainfall which is a major contributor does not
show any appreciable corresponding rise. Rapid population increase in density of between 84
% and 300 % from 1979 to 1999 has put a lot of pressure on the resources. Land use is more
intense in terms of agriculture, deforestation for various activities, grazing and settlement.
Deforestation in the upper catchment, especially around Mau, Tinderet and the Nandi
escarpments is high thus altering the structure of the soils, and with the fragile nature of the
area soils, soil erosion and flooding are enhanced. Steep topography of the middle zone
compounded with deforestation and agricultural activities enhance both soil erosion and
flooding implying less infiltration and less base flow soon after the rains meaning early and
prolonged droughts. Gentle slopes. of the lower basin promote sediment deposition and
inundation which sometimes lasts even a month. The study provides a base from which
various measures: preventive, mitigation, emergency response and rehabilitation of the flood
management structures for both short and long term can be addressed to restore the
hydrological balance of the basin. The study recommends an integrated watershed
management to the basin, an all inclusive environmental management for sustainability. The
study recommends further research on the impacts of these extreme events on the quality of
water and the effect of global change on the extreme hydro-climatic characteristics.
Xli
CHAPTER ONE: INTRODUCTION
1.1 Background
Water is a vital component for economic, social and cultural aspects of development. More
people have over the years settled next to waterways and in flood plains because of the
advantages they offer (Ogolla et aI., 1997). Early civilization started along major rivers, for
example the Nile, Tigris and Euphrates. Water resources all over the world are diminishing
due to population pressure, climatic changes and expanding land use activities. In Africa,
most of the water sources are rapidly declining due to anthropogenic activities along the
rivers that feed them. River Nyando is discharging into Lake Victoria 100-300 tonnes of silt
per km2 per year from the basin (KSS, 2003). With the fast growing population
at a rate of more than 3% per annum, and a current population density of 1200 persons per
. km2 (ICRAF, 2000), water demand is rising. In the last two decades, agricultural activities
have increased by 40%, deforestation has been high and pollution to the water courses has
also increased causing fresh water sources to become scarce (Nyanchaga, 2004). The
scarcity of water has caused international conflicts in water resource systems and this is true
of Lakes Victoria, Turkana and other Lake basins in Africa (Daily Nation, 2005). The
conflicts arise due to mismanagement of the water resource systems. Different initiatives,
like the Nile Basin initiative, have to address the management of water systems, especially
the declining levels in the Lake Victoria. This is attributable to anthropogenic activities and
compounded by climate variability and change which result in extreme hydro climatic events
of floods and drought.
The river Nyando basin experiences extreme flows causing havoc to the population therein
and socio-economic disruptions. Floods have been causing destruction to the infrastructure,
loss of property, loss of human and animal lives and crops. The basin has experienced major
floods especially in the lower reaches in 1937, 1947, 1957-1958, 1961 and 1964 (WMO,
1
2004). In 1961, the area experienced exceptionally heavy rainfall during October and
November resulting in severe floods in the lower reaches (WMO, 2004). At first, flooding of
low-lying areas was caused by over-bank spills of the river, but the situation was worsened
by the back water effect of the rising lake level in 1964. The other floods occurred in 1964,
1985, 1988, 1997-98,2001,2002 and 2003. The' 1997-98 floods were due to El Nino rains of
October and November where the rains were 100-300% above normal. Most of the lower
reaches of the Nyando basin, the Kano plains was inundated, agricultural crops destroyed,
infrastructure in nearly the whole basin was almost entirely destroyed, land degradation was
high, siltation of dams and rivers was high, a weir on Kipchoria river, a tributary of river
Nyando was washed away (WMO,2004). In Uasin Gishu district two earthen dams were
washed away. Several river gauging stations were extensively damaged due to severe bank
erosion. Protective dykes were overtopped and suffered breaches at several places. Many
people were displaced. The breached dykes took long to repair thus making the area more.
vulnerable to floods of even smaller magnitudes. The floods of 2002 took place in April, May
and November and that of 2003 occurred in April and May. These floods were caused. by
heavy and concentrated rainfall in the upper catchment (WMO, 2004). The flood prone area
is about 30 Km stretch before Nyando enters the lake and approximately 32 km2 in area. This
area suffers almost annually from floods (WMO, 2004).
Knowledge of these extreme events is useful for planning activities in the basin. The
residents of the flood prone areas are mostly caught by floods unawares and at times in the
middle of the night thus incurring heavy losses. This is because the floods are usually
caused by rains on the highlands. Dykes have been constructed to confine the flows to the
channel a situation which no longer holds because of greater magnitudes of flows.
Knowledge of these extreme flows is useful in water resources planning activities in the
2
basin. The characteristics are however not well understood in the basin. This study therefore
seeks to establish the extreme hydro climatic characteristics of the Nyando river basin.
1.2 Statement of the Problem
Flooding is a perennial problem in the Kano plains of the Nyando basin where there are
yearly reports about devastating effects of floods (OCHA, 2002, Njogu, 2000) which include
loss of human lives, livestock, crops, stagnation of income generating activities, destruction
of infrastructure and increase in water-borne diseases.
On the other hand, droughts are also severe in the region leading to loss of livestock and crop
failure (Nyanchaga, 2004). Droughts and floods have negatively impacted heavily on various
aspects of livelihood in the Nyando catchment. Studies so far done on The Nyando basin
(Lotti and Associates, 1985, Wanjohi, 1999, MWRD, 2004a Waruru et aI., 2003 MWRD,
2004b) emphasised on flood control and not the characteristics and causes of the extreme
events.
There is, therefore, need to investigate the causes and understand the characteristics of these
extreme events to help mitigate against and manage them when they occur, as they are
increasingly becoming more frequent.
1.3 Justification
Variable, erratic and sometimes intense concentrated rainfall in the upper parts of the basin
cause destructive floods especially in the lower catchments. The magnitudes of losses both
human and economic caused by the floods are enormous. Most of the deaths, post traumatic
stresses and social and economic hardships can either be tremendously reduced or avoided
3
altogether if the characteristics of the extreme events and the factors that intensify them are
understood and accordingly managed.
Several studies have been carried out aimed at providing solutions to the flood menace in the
lower Nyando basin; the Nyando Pre- investment study (Ita IConsult, 1983) reviewed various
options of flood control measures for both short term and long term which were mostly
structural, most of which have been overtaken by the frequency and the magnitudes of the
floods experienced, the Lotti (1985) River Profile Studies, the study on the National Water
Master Plan (JICA, 1992) and the Ministry of Water Resources Management and
Development (2004) addressed flood problems in the lower Nyando basin, but fell short of
investigating the characteristics and the major underlying causes of the of the extreme events.
The studies concentrated on the flood prone area and ignored the middle and upper zones of
the basin which are the major contributors to the flooding problem. This study therefore
investigates the causes and the characteristics of the extreme events on the entire basin with
major emphasis on the upper and middle zones.
It is therefore justifiable to investigate any factors which might lead to increased flood risks.
This is a concern for individuals, private enterprises and the Government who bear the cost of
floods when they occur. Damages due to flooding are inequitable, affecting those who mostly
have no control of the situation and often without any form of compensation. There is
therefore need to find ways to minimise the losses and enhance economic growth by viably
exploiting the floods. Low flows are characterised by acute water deficiency contrary to the
belief that being near the lake, water is plentiful. Low flows also suffer high pollution which
the river system can not assimilate. This is attributable to the fact that most industries within
the upper and middle zones of the catchment discharge their untreated or partially treated
effluents into the river contributing to high point source pollution. Markets and other urban
4
centres e.g. Ahero urban council empty their fresh wastes directly into river Nyando. There is
therefore, need to undertake the study.
1.4 Objectives
The aim of the study is to investigate the extreme hydro-climatic characteristics of the
Nyando river basin. The specific objectives are:
a) To determine the characteristics of the extreme hydro-climatic variables of the Nyando
river.
b) To establish the effects of land cover and land use on the extreme flows in the Nyando
river.
1.5 Significance of the Study
Understanding the causes and characteristics of the extreme events, that is, droughts and
floods in the Nyando basin will make it easier to determine and propose the level of
preparedness, improvement and coping mechanisms against any magnitude of a flood or
drought.
The knowledge of the causal factors of the extreme events will enable reduction of 1
anthropogenically induced causes through proper resources management. Extreme flows are
destructive in terms of soil erosion; destruction of infrastructure, disruption of social set ups
and source of non-point source pollution to the water courses. The quality of the environment
often depends on the availability of low flows (Hirji, 2005). It is therefore vital to investigate
the low flow to confirm whether it is able to self purify the wastes discharged into it mostly
from industries and urban set-ups or whether additional treatment is necessary. This will also
help establish whether people are getting sufficient and clean water, if not, the extent of the
shortfall either way. The information on low flows is vital for water allocation, water
resources planning and water quality parameters especially for industrial effluent discharges
5
from middle and lower catchments. The knowledge of extreme hydro-climatic characteristics
will help in the generation of a feasible harmonised and balanced water use programme for
sustainability for all water uses. It will also help in the improvement of water use efficiency
to satisfy the various demands.
Water demand is projected to rise per year from 2,073Mm3 in 1990 to 5,817Mm3 in 2010
while annual per capita renewable fresh water supply will continue falling to 503m3 during
the same period if no measures are taken (Table 3). This makes it important to carry out this
investigative research to come up with facts on which remedial and proper management
programmes can be based.
6
CHAPTER TWO: LITERATURE REVIEW
2.1 General
The water resources of Kenya are rapidly becoming scarce. The current demand outstrips
water supplies in all sectors of the economy resulting in severe conflicts between user and
sectoral groups (Mwongera, 1996). Kenya is classified as a water scarce country with an
annual per capita consumption of 657m3 and is projected to drop to 359m3 in the next decade
(Mogaka et aI., 2002). Hydrology of the basin is characterised by the space-time distribution
of rainfall, river flow and groundwater. There are seasonal and annual variations of rainfall
and floods [hydrologic variables] within the basin. The Nyando river basin has been a
popular research area because of its hydrological challenges. The area suffers from frequent
floods, land degradation, high poverty levels and droughts as opposed to other areas where
the poverty index is influenced by the inadequacy of water (Aseto and Onganga, 2003).
In the Nyando River the magnitude of the water has not been containable within the time it
comes such that in terms of becoming a source of livelihood it becomes a disaster. The flow
can be contained and managed for the overall socio-economic development of the region, and
in this regard it is important to establish the trends and seasonal changes especially of the
extreme events. The knowledge of the extreme hydro-climatic characteristics will make it
easier to exploit, conserve, preserve, protect and restore water resources in the basin for
various needs. Water resources in the Nyando basin are under stress due to the variations in
quantity and quality. When the quantity of water is low the concentrations of pollutants,
especially the industrial based effluents, are high either raising the treatment costs or posing
threats to health. This situation is further worsened by the rapid population growth and the
associated activities. There is a progressive trend of diminishing water per capita availability
as indicated in Table 1. It is most likely that rapid population growth in the Lake Victoria
basin has induced major land use changes, which have structurally altered the hydrological
7
stability of the entire basin. High sediment load in the River Nyando could be because of soil
erosion which is enhanced by continuing deforestation and poor land management practices
that promote the generation of overland flow thus reducing infiltration. Overgrazing therefore
reduces the quality and quantity of the renewable vegetation (ICRAF, 2000). Watershed
degradation includes deforestation, soil erosion, sediment deposition, untreated or partially
treated effluent discharge into the water courses at various points from mostly urban centres,
sand harvesting along the main tributaries of the Nyando River, massive brick making around
Awasi and Nyabondo areas. These forms of land degradation are widespread in the Nyando
basin (LVEMP, 2000).
Kenya's forest loss to agriculture, settlement, wood fuel, and construction is high and
continues to rise resulting in decreased rainfall, massive soil erosion and serious floods.
Deforestation and clearing of vegetation for crop production and livestock pasturage, with
consequent heavy losses of soil, have caused serious degradation of most catchments in
Kenya. National loss of forests due to poor land use practices is estimated to 3.2% per year,
making deforestation a major threat to the successful management of natural resources in and
around catchment areas (Kipkore, 1996).
The fragile soils and improper soil cover management through poor land use practices could
be the major factors that accelerate surface runoff causing floods and high sediment
movement. ICRAF (2000), monitored sediment flow into Lake Victoria from rivers Nyando,
Nzoia, Yala and Sondu between February 2000 and July 2001 and found out that the Nyando
basin is the major contributor to sediment flow into Lake Victoria. A regional assessment by
the same confirmed that the Nyando River catchment contributes more sediments to the lake
than the other sub-basins in the Lake Victoria basin (ICRAF, 2002). This kills marine life
which is the economic backbone of the area, and because it is a function of soil erosion which
8
results from flooding, the causes have to be established. Soil erosion covers almost the entire
basin due to the fragile nature of the soil (Plate 1). Soil erosion is a function of floods which
is enhanced by impeded infiltration, slope of the area, nature of the soil, type of rainfall and
anthropogenic activities.
Table 1: Water availability per capita between 1969 and 2020
. th N d C t h tIII e ryan 0 a c men
Year Population Per capita Water Availability
m3jyear
1969 10,942,705 1853
1979 15,327,061 1320
1989 21,448,774 942
1999 28,686,607 704
2010 40,311,503 503 (Estimated)
2020 56,481,427 359(Estimated)
Source: JICA 1999a
9
Plate 1:Fragile soils in the middle Nyando catchment - ICRAF, 2000
. Surface runoff and infiltration occur simultaneously in nature during and after rainfall.
Infiltration is that part of surface runoff that gets trapped into the voids of soil. Surface runoff
occurs as a result of prolonged rainfall at intensities greater than the steady infiltration rate of
the soil, especially for the land surfaces with reduced infiltration due to compaction or
removal of vegetal cover or top soil; situations which are brought about by land use types as
cattle tracks, footpaths, degraded and grazing areas. Accelerated erosion occurs very rapidly
and is aggravated by human interventions (ICRAF, 2000). This type of erosion is the major
cause of bad land development and heavy sedimentation of water courses. There is a direct
link between bad land development and population growth and bad land development and
soil erosion (Waruru, 1992). There is therefore need to analyse population growth against
land use activities. Bad land use loosens the soil making it erodable. Soil erodibility is a
measure of the soil susceptibility to detachment and transport by the agents of soil erosion
~ich include human activities. Nyando catchment soils are fragile and therefore susceptible.
The force of erosion that causes soil detachment and transport is erosivity, which is a function
of rainfall. Nyando basin receives a lot of concentrated rainfall which results in massive
10
runoff. Once the process of erosion has started it develops from sheet erosion to rill and
finally to deep gullies which are common in the middle and lower Nyando basin.
According to Waruru et al. ( 2003), soil erosion rating, about half of the entire catchment
suffers from high soil erosion, 4% of the entire area suffers from very severe erosion e.g.
Katuk Odeyo (Plate 8b), 17% severe, 25% moderate and 4% low (Table 4). Soil degradation
has negative impacts on agricultural productivity, ecosystem and atmospheric changes, water
and habitat quality.
Table 2: Erosion hazard rating for River Nyando catchment area
Rating Designation Slope % of area covered
5 Very severe More than 30 4
4 Severe More than 25 17
3 High 16-25 50
2 Moderate 5-16 25
1 Low 0-5 4
Source: Waruru et al., 2003
The major soil degradation processes are accelerated soil erosion, depletion of soil organic
matter and soil nutrients and the deterioration of soil structure (ICRAF, 2002). The soil water
retention capacity is greatly reduced thus reducing the quantity of base flow in rivers during
low flows.
2.2 Extreme events
The frequency of droughts and storms affect soil moisture, reservoir storage, ground water
recharge/ balance, runoff, river flow, erosion, water quality, irrigation demand and water
11
supply. Soil erosion causes turbidity, silting of reservoirs and river courses thus decreasing
their life spans. The lower zone of the Nyando basin towards the lake shore lacks adequate
rainfall most of the time in the year. Rainfall here is varied and unreliable and with no water
supply systems, water demand for various utilities is not adequately met making the area be
among the poorest in rural Kenya (Odada and Otieno, 1990).
Kenya's economy is largely agro-based and with a rapidly growing industrial sector, the
demand for water of good quality and sufficient quantity is constantly rising, there is
therefore need to take stock of what there is, especially during the extremes (Mwongera,
1996). Nyanchaga, (2004) found out that floods and droughts in western Kenya, Nyando
and Nzoia river basins negate regional and national efforts towards sustainable development
in the region and therefore recommended further research on the two extreme events. Nyaga,
(1996) researching on water quality and land use management in the Lake Victoria Basin
found out that large quantities of inadequately treated effluents are discharged from
municipal, industrial, defuse agricultural, infrastructural and other indeterminate sources into
surface water bodies in the basin, these wastes, especially during low flows are a threat to the
health of the consumers. He recommended further investigation and assessment of low flows
and pollution levels. This study proposes to investigate the low flows to determine their
quantity, frequency and trends. Population growth, urbanization, agricultural expansion and
industrialization are the major challenges to water resources management in terms of
adequacy and quality (Mwongera, 1996). The study seeks to investigate the above in relation
to their effects on the extreme events.
This study seeks to determine rainfall variability in the basin to address adequacy especially
during low flows. Njogu, (2000) proposed institutional arrangements as an aspect of water
resources management and recommended further research on water quantity especially
12
during low flows to reduce uncertainty in water use planning. Matiasi, (2003) concluded that
it was necessary to understand regional occurrences of low flows including the times of the
year the regions are affected and found out that there is increasing rainfall variation in the
Lake Victoria Basin. He recommended enhanced investigation on rainfall variability and
extreme hydrological events which he said would be a basis for an integrated water resources
management during low flow period which is associated with water shortages.
Kenya is vulnerable to the extreme events because its economy highly depends on water
resources. The country has low natural water replenishment rates; its existing water resources
are far from being adequately developed and are poorly managed. The study proposes to
quantify both the extreme flows, analyse their occurrences and recommend better
management methods. It is possible to minister the impacts of such extreme climatic "events
through proper planning; predictions, forecasting and instituting measures to mitigate, and
minimize against major consequences (Mogaka et aI., 2002). The study therefore after
determining the causal factors, characteristics and predetermined magnitudes of the extreme
events, will make it easier for planners and all stake holders to be prepared with adequate
resources to mitigate and or respond to impending disasters meted by the extreme events.
2.2.1 Floods
A flood is a progressive rise in the water level of a stream or river beyond its drainage
capacity thus resulting in an over spill. Flood water spreads and inundates low lying areas
thus coming into conflict with human activities. Floods are caused by heavy rainfall,
structural failures and tsunamis and enhanced by several factors both human and
environmental. Since flood plains are desirable locations for man and some economic
activities it is important that that floods are controlled to limit the damage and reduce human
casualties.
13
Flooding in the lower Nyando basin is almost a yearly occurrence with the highest
concentration being the last 30 km before the river discharges its water into lake Victoria
(Ocharo, et. ai, 2004).
Flooding disrupts socio-economic activities, destroys infrastructure, interferes with the
seasonal patterns of agriculture, causes loss of lives to both livestock and people and
impoverishes the people. These are what the floods have caused to the Nyando people. The
basin experienced severe floods in 1961, 1964, 1985 and 1988. The 1997-1998 flood was the
consequence of EI Nino related long and intensive rainfall during the months of October and
November when the precipitation was 100 to 300 percent of the normal (WMO,2004). Other
floods were occurred in 2002, 2003 and 2004.
Several attempts have been made to control and manage the flooding problem in the basin. In
1973 the Ministry of Agriculture and Livestock Development constructed a dyke downstream
of Ahero Bridge to Apondo which has since been destroyed by floods aggravated by human
activities and livestock. In 1975 and 1977 National Irrigation Board constructed dykes
around the West Kano Irrigation scheme as a mitigation measure. The Kisumu DDC also
implemented a number of flood protection projects between 1975 and 1979 around Kabonyo
and Bwanda areas, but were destroyed by pedestrians and animals and due to lack of
maintenance were later washed away by floods. Italconsult in 1983 proposed the construction
of an embankment along the Nyando Lielango and Kibogo rivers to contain floods with a
frequency of 50 years. Lotti in 1985 proposed the construction of new dykes and
reconstruction of new bridges on Nyaidho River to offer protection against a 25- year flood.
These proposals have not been implemented over 20 years later. The lower Nyando basin
becomes flooded much more frequently now than before. What could be the cause?
14
Hydro-climatic and environmental parameters may have changed such that the designs
proposed then may not withstand the present and predicted future hydraulic magnitudes.
This study will analyse peak flows in terms of their magnitudes and frequencies against the
flood control and mitigation structures in place to ascertain their adequacy now and in future.
Pre- investment studies by Italconsult (1978, 1980 and 1983) recommended the construction
of flood mitigation structures, which a number have not been constructed to date. It identified
24 potential dam sites as a flood mitigation measure but recommended three for construction
(Table 3). Then, siltation was not a major problem because environmental degradation and
deforestation were not a major threat - most of the forested upper catchment was still intact,
now the situation is different.
Table 3: Prospective Dam sites in the Nyando basin
Dam site River Sub-Basin Catchment Purpose Live Storage Dam Dam
No. Code Area km2 (MCM) Crest Height(
Level(m) m)
05 Ainapngetun 1GB5 404 Flood 120 1,450 90
y control /
Irrigation
11 Nyando 1GC6 867 Flood 81.5 . 620 70
control /
Irrigation
51 Kibos lHA 119 Irrigation 82 540 100
Source: Ita/consult, 1983
The study also does not address the inclusion of non-structural measures, even the capacities
of structural flood control structures proposed then can no longer hold. Based on its findings,
this study proposes a review of the design parameters and the viability of the dams against the
prevailing environmental changes taking place.
Wanjohi, (1999) used DUFLOW model to simulate flow in river Nyando and proposed
possible mitigation measures against floods (Table 4). According to him if nothing was done
"\ the maximum stage corresponding to a 50 year return flood would be 7.6m and the magnitude
15
586 m3/sec. If the wetland were cleared the maximum stage would be 6.6m, if a 1.8m dyke
was constructed it would contain the 50 year flood, but if a 20m wide By-pass were
constructed it would lower the stage to less than 5m. Dyke construction as the only mitigation
measure is fast becoming unworkable; first the magnitude of a 50 year return flow is on the
rise. According to the MWRD, (2004a) the 50 year flood was 1,178 m3/sec, this figure is
more than twice what Wanjohi deduced in 1999.
Table 4: Results of DUFLOW Model for the Nyando River
Return Upstream Measures
Period Flow
Maxim urn depth Depth at Height dyke Width by-pass
at Ahero (m) Ahero(m) at Ahero(m) which lowers
depth to Sm
5 324 6.7 6.1 1.3 11
10 405 7.2 6.4 1.4 16
25 509 7.4 6.5 1.6 18
50 586 7.6 6.6 1.8 20
Source: Wanjohi, 1999
This implies that the height of the dyke would be more than double (4m) -uneconomical and
unrealistic in that subsequent silt deposition would mean increased height. Construction of a
by-pass would be hydraulically possible, but may not reduce the other impacts of the extreme
events as such may have to be subjected to environmental impact assessment. The solution
may lie in looking at the environment wholesome vis-a-vis floods, droughts, land use
activities, population growth and livestock keeping. The knowledge of these will help in
determining the most viable, sustainable, economical, structural and non-structural options to
be adopted. There is therefore need to review and update the whole design to capture the
prevailing conditions and the most likely future scenarios.
16
Two decades ago global warming was not a major issue, now it is a global threat. Global
warming due to greenhouse gases effect now compounds the problem of floods and droughts.
According to Ogolla et al. (1997), there is growing concern about the impact of climate
change on the frequency of floods. There is a likelihood of future increases in the incidences
of flooding due to increased storm activity and overall increase in depth of precipitation.
Increased climate variability can also lead to excessive floods or droughts with consequential
adverse effects.
Accelerated run-off and sheet erosion over much of the catchment especially the upper and
the middle zones have led to severe rill and gully stream bank erosion in the lower zone of
the river basin {ICRAF, 2000). The principal causes of erosion include deforestation and
overuse of extensive areas of fragile lands on both hill slopes and plains. The other
contributing factor is loss of watershed filtering functions through encroachment on wetlands
and loss of riverine vegetation. Much of the basin has and continues to lose top quality soil to
erosion, thus increasing poverty to an already poor region whose poverty index is up to 68 %
in some parts (Samez, 2005)
2.2.2 Droughts
Drought is a temporary reduction in water or moisture availability significantly below the
normal or expected amount for a specified period. The impacts of drought result from the
shortage of water, or the discrepancies between supply and demand for water.
Droughts unlike floods which are sudden creep in slowly and often very difficult to detect its
onset until some major impacts such as lack of water and food start to be discernible.
Droughts do not directly destroy food storage, shelter or infrastructure. The effects of
droughts are cumulative. Extreme climatic events such as droughts and floods are partly
17
caused by EI Nino/Southern Oscillation (ENSO) phenomena. According to Ogolla, et al.
(1997), in the recent years, and in some regions of the world, there is some evidence of
significant changes in the trend and frequency of extreme climatic events such as droughts
and floods. There are four types of drought depending on the situation namely:
Meteorological drought results from a shortfall in precipitation and can be measured against
daily, monthly, seasonal or annual timescales of rainfall amounts. Hydrological drought is a
reduction of water resources in surface, sub-surface and underground sources in relation to
the normal operation of the system.
The major impact of hydrological drought is the competition between users for the water in
the systems. Agricultural drought is the impact of meteorological and hydrological droughts
on crops and livestock production. It occurs when soil moisture is insufficient to maintain
average plant growth and yields. Socio-Economic drought correlates the supply of weather
dependent goods or services (water, hay, electric power) with demand. The physical factors
which contribute to vulnerability of droughts include moisture retention of the soil which is a
function of topography and soil cover, the timing of the rains (intervals between rains),
dependency on rain fed agriculture. Droughts may cause partial or total loss of crops
especially on rain-fed agriculture. Farmers unable to adapt to drought conditions with
repeated plantings may experience crop failure. Livestock- dependent populations without
adequate grazing area are also at risk.
Adverse effects of droughts are economic, environmental or social. The economic effects
include: Losses in production of crops, dairy and livestock, timber and fisheries, Income loss
to farmers and others directly affected, Losses to industries related to agricultural production
and decline in food production and increased food prices.
18
"r , -, - .....• ,.., •...r;n'1nr.-' 0"'~
The Environmental effects include: damage to animal and fish species and habitat, Wind and
water erosion of soils. Effects on water quality are increased concentrations of pollutants.
The social effects include: food shortage effects (malnutrition and famine), conflicts between
water uses and water users, health problems due to water shortage, decline in living
conditions, increased poverty and reduced quality of life.
For along time the traditional approach to drought mitigation remained reactive and based on
response after the occurrence of a drought. Vulnerability of the society to drought is
increasing in many parts of the world due to increasing population, changes in land use
activities, environmental degradation and other socio-economic factors. This study will
analyse hydrological drought situations and predict the future scenarios based on the
available records to recommend appropriate level of preparedness in terms of mitigation and
management of the disasters.
2.3 Impacts of extreme events
Disasters impoverish people thus increasing absolute poverty. According to Samez, (2005)
poverty level in the country stands at 50 %, Nandi district 64.2 %, Kisumu district 53.0 %,
Kericho district 60.7 % and Nyando district 68.9 %.
These poverty indices are partly attributable to the extreme hydro-climatic variables. The
response to the extreme events requires a forecast with a great degree of accuracy and hence
the need to adopt practice interventions rather than reactive responses, as seems to be the
case, to help reduce their effects. Each season flood affected families are provided with relief
food, drugs, temporary shelters, tents and blankets as short - term measures, this is because
there are no long-term guidelines to base planning on. Understanding the hydro-climatic
19
characteristics of a place will enable appropriate long-term level of preparedness to be put in
place.
Lotti in 1985 evaluated the flow in River Nyando to size and cost flood related structures and
to prepare drainage plans to contain the same, but fell short of determining the characteristics
of the extreme events.
Extreme events have impacted on all sectors of our economy negatively be it floods or
droughts. Our systems are mostly caught unprepared even where information of impending
extreme events have been relayed before hand. Kenya is affected by both local floods and
droughts and El Nifio/La Nina events of 1997-98 and 1999-2000 cost the country very highly
in all sectors of the economy. This is because the country's economic stability and growth
highly depends on water resources. The country is therefore vulnerable to the extreme events
due to the low natural water replenishment rates, low development of the water resources and
poor management of the existing resources.
The 1997/98 El Nifio rains seriously damaged water supply infrastructure, transportation
network, development setups and increased water related diseases. El Nino describes the
appearance of a large pool of warm water in the Western Equatorial pacific and La Nina the
disappearance of the same. This phenomenon alters major pacific currents and affects the
weather patterns throughout the Southern pacific and the Indian Ocean. The return period for
these phenomena is about three times every decade. El Nifio and La Nina manifest
themselves as severe flooding followed by a period of drought (Mogaka et aI., 2002). The
flooding affects some parts of the country including the Nyando basin, while the drought
component affects the whole country. The most recen't El Nifio of 1997-98 and La Nina were
the severest in 40 years (Mogaka et ai., 2002). The effects have been seen in the drying up of
20
streams and because these weather patterns are not known traditionally the affect greatly
affect agricultural production, livestock keeping and disrupt social set ups.
Most of the run off affecting the plains is generated by rainfall in the upper catchment and the
people in the plains downstream are usually caught unawares and at times at night incurring
a lot oflosses including human lives, livestock and poultry. According to the survivors of the
1963 floods, they woke up in the middle of the night to the rising water levels. They sought
safety on higher grounds, but the flood level continued to rise until they were airlifted.
Floods damage and pollute water points( Rothschild and Skills, 1992 ), for example, the 2002
floods submerged 88 shallow wells, damaged 1915 homes, swept away livestock and
increased waterborne diseases and malaria (Table 5, OellA, 2002). The average annual
maintenance cost for the flood victims was Kshs. 37 Million in 2002 and 2003 and Kshs 9.2
Million in 2004.
T hi 5 I f fl d . N do Basia e : mpacts 0 00 s m ryan 0 asm
YEAR CAMPS PEOPLE NO. OF AMOUNT OF
NO. OF PEOPLE ROLLED PER EDUCATION AGRICULTURAL
OUT CAMP INSTITUTION LAND
SAFFECTED SUBMERGED(Ha)
KILLED DISPLACED AFFECTED PRIMARY SECOND
ARY
2002 - 48,000 28,650 - - 27 3 3,09
0
2003 6 5,000 15,000 8 1,200 40 5 3,00
0
2004 6 400 10,000 19 101 34 6 4,00
0
Source:~1),2004a
-Floods interfere with seasonal patterns of agriculture. When fields are inundated or wet
cultivation is not possible forcing farmers to wait. This implies that they plant late and the
rains in most cases cease before the maturity of the crops.
21
Farmers therefore experience either total crop failure or reduced and poor quality yields.
This is almost an annual occurrence and continues to impoverish the people to the extent that
they are almost entirely depending on relief foods. Floods also limit the choice of crops to
few flood tolerant ones like sugarcane and rice. The staple food in the area is maize and
continues to do badly in the lower catchment due to floods.
During floods wells get submerged and polluted, sanitary casings washed away and the
pumping units damaged. According to the residents the floods of 2004 caused most pit
latrines to overtop, excreta overflowed to the nearest water supplies. This made the residents
resort to alternative sources of water of dubious quality which enhanced water-borne diseases
and slowed down various development activities because of the long distances covered
during the search for water. Floods destroy buildings, and damage the contents inside the
buildings. In the Kano plains most people acknowledged that annual loss to floods in terms of
household contents is enormous and the damage to the remaining great. The house contents
remain submerged until the floods subside. This is because most contents like chairs,
cupboards etc. are too heavy to be carried during emergency evacuation operations so remain
under water as long as the floods last. This also enhances their rate of deterioration and decay
due to long term rot and damp thus reducing their lifespan. From plate 14, it is evident that
after floods the contents of the house will be completely destroyed. Some people in this basin
experience total loss of possessions and have to start all over again after the rains. Floods
bring down telephone lines and electric poles thus disrupting communication services.
People from the inundated areas move to makeshift relief camps where they cluster together
(plate, 2). Such camps/homes become slums creating social problems and unhygienic
conditions which disrupt normal life leading to various social ills such as the spread of
sexually transmitted diseases. This congestion also promotes the spread of contagious
22
diseases. The displacement of people is traumatic because it is usually an indefinite torture
and a dent to the family social fabric creating indiscipline among the youth because of the
negative social exposure.
Schools remain closed for as long as the area is still submerged and repairs have not been
done to broken down infrastructure such as toilets, washed away bridges, roads, classrooms
and office facilities. When they finally reopen, they take time to attain that normalcy thus
leading to failure in the national exams which eventually translates into lower earning power.
Plate 2: Evacuation of flood victims in the lower Nyando Basin 201512005 v'-
)
During floods various economic activities come to a standstill: businesses such as shops close
down either because of non accessibility or displacement of the customers, no farm activities
"\
can take place either because the farms are submerged or are too wet to be tended. Grazing of
livestock becomes a problem because the fields are under water forcing the animals to be
grazed in small upland areas where they overgraze and enhance land degradation. These
23
factors also are responsible for the high poverty levels in the basin, especially in the lower
catchment where the flooding effects are experienced most.
According an elder around the proposed Kimira Bridge in the lower Nyando, rates of crime
generally rise; robberies and house breakings increase during floods. When floods eventually
subside and residents go back to their homes they find most household goods looted.
Sediment pollution and diffuse source pollution are high. Agricultural sediments have high
concentrations of nutrients especially phosphorous which fuel algal blooms. Algae decay and
deplete oxygen in Lake Victoria causing eutrophication which is responsible for deaths and
disappearance of various species of fish. Eutrophication has now restricted about half the
.volume of the lake to many fish species especially the Nile perch which are unable to survive
in low oxygen levels. This represents a potential annual loss of between 30-50 % of the Nile
perch biomass in the lake (Mogaka et al., 2002).
Sedimentation, nutrient run-off, urban and Industrial source pollution and biomass burning
have induced rapid eutrophication in Lake Victoria. Bullock et aI.,(1995) estimated that 50 %
of nitrogen and 56 % of the phosphorous input in the Nyando river is due to runoff from
agricultural land, 30 % of the nitrogen and 30 % of the phosphorous is due to rural domestic
waste and 10 -15 % is due to urban waste and atmospheric deposition. River Nyando is the
main source of phosphorous (the major nutrient causing Lake eutrophication) flow into Lake
Victoria (Atieno, 2004). Water pollution during floods is a major cause of various disease
outbreaks. Residents of the Kano plains depend on shallow wells for their domestic water
use. During floods these water points are submerged thus mixing with all types of pollutants.
24
Structural damage is due to the impulse of the moving water. It is a function of quantity,
velocity of the moving water and includes collapsing of buildings, washing away of bridges
and parts of roads, causing traffic congestion and disruption to communication, total loss of
crops and damage to dykes. Non-structural damage is a function of depth (maximum water
level) and the duration of inundation. For crops the harvest is not destroyed but the yields
reduce either in quality or quantity. Non-structural damage to buildings is damage caused due
to wetting of live loads; furniture, wall paper, machines, appliances and beddings. These
damages are frequent in the lower Nyando basin.
Floods weaken structures making them more damage prone in subsequent floods. Abrupt
disruption of electricity supply causing loss to industrial production and darkness in
institutions and homes that use power translate to economic drawbacks. Sudden power
disruptions also cause damage to machines that use power like computers; electronic gadgets
etc. The loss of power enhances other social problems and general insecurity in the affected
areas.
The greatest threat and contributor to the water resources problems in the Nyando River
Basin is the catchment's degradation. Severe soil erosion upstream which is due to poor land
use practices and massive deforestation deprives the upstream dwellers of top fertile soil
implying that crop yields are reduced. Maize yields have declined by 34% in nitisols
(Kipkore, 1996). The reduced agricultural output is a recipe for poverty especially for the
upstream dwellers. They are forced to use a lot of agro-chemicals increasing the cost of unit
yields for various crops. Downstream water users experience floods because of filled up
channels from erosion of catchment and river banks with consequent shifting of river
channels and meandering in the plains (plate 13, Appendix8).
25
- ,
The 1997/8 El Nino floods displaced over 300,000 people countrywide with Nyanza topping
in the number of casualties (WMO, 2004). In the Nyanza the Nyando basin was the most
affected. Water pollution and poor sanitation reached epidemic levels. Socio-Economic
activities including agricultural and industrial production declined. Tractors collecting cut
sugarcane could not access the fields because of wetness, and those which dared to go got
stuck in the mud. Because of non accessibility even agricultural activities stalled. Women and
young children especially the girl child had to trek long distances looking for cleaner less
contaminated water.
The effects of this are twofold: most of the rural economic activities which are mostly
performed by women will go unaccomplished because most of the time will be taken fetching
'water. Secondly the girl child will lose in education because as her male counterparts are in
class she is busy fetching water. Education was seriously disrupted, most schools were
inaccessible due to destroyed infrastructure so had to close down and when they eventually
reopened a lot of facilities had either been destroyed by floods or vandalised.
Most families moved to safer grounds far from the schools and took time before going back
causing academic disorientation which is very costly in terms of education continuity. This
lowered the quality of education in the area. Damage to health facilities disrupted the
delivery of quality healthcare services at a time when the epidemic of water-borne disease
was on the rise. In the Kano plains, an estimated 5,000 people are affected annually by flood
spills of the Nyando River. The average annual damage is about Kshs 5,950 0000 with an
annual relief and rehabilitation measures costing Kshs 42,000 000 (WMO, 2004). Serious
[dod and other shortages are experienced. Gross Domestic Product is reduced.
26
Positive effects of floods include preserving the wetlands, recharging of the ground water,
maintaining of the river ecosystems by providing breeding, nesting and feeding areas for fish,
birds and wild life. The people living downstream in the flood plains don't use fertilizers or
manure for agriculture because the floods bring enough fertilizers from the upper catchments.
In Budalangi where the construction of the dykes to contain the floods has been successful
the residents are saying the land is fast losing fertility because of non replenishment as was
before during floods. In the Nyando basin floods are associated with a lot of fish catch
because the fish would flow upstream of the flood waters to enjoy the warmth of water and to
feed on what the floods bring. This probably explains why those who were evacuated and
given land around Songhor during the 1963 floods abandoned their new homes and returned
after the floods. Efforts to resettle the residents in these flood prone areas on safer grounds
have not borne much fruits because they either sell the land or in the absence of a buyer they
just abandon the land and go back.
Droughts affect all sectors of human life health livelihood, production, recreation,
communication, learning and livestock. Droughts are not as sudden as floods because they
take longer to be realised and recovery from their effects take even much longer.
The 1999/2000 La Nina drought affected several development projects nationwide. Energy
and water supply sectors were the worst hit. Power and water had to be rationed to most of
the urban centres. This affected industrial production, agricultural production, human and
livestock water requirements.
The output of every major sector of the economy was reduced. Nationwide loss from
livestock deaths was Kshs 12.2 billion. Hydro-Electric power generation was reduced by over
40 % and KPLC lost Kshs 1.6 billion in income. National industrial production went down by
27
Kshs 110 billion (Mogaka et aI., 2002). Kenya has experienced major drought periods every
10 years. Droughts affect basic needs such as food, water and livelihood. In some parts of the
country the return periods of droughts have become shorter (Mogaka et al., 2002).
28
CHAPTER THREE: METHODOLOGY
3.1Description of the study area
This chapter outlines the study area in terms of location, topography, Geology and Soils,
Hydrological Network, Climate, Population and Socio-economic activities, data collected and
methods of analysing the variables.
3.1.1Location and topography
TheNyando basin (3,618 km2) is located almost on the Equator at 35.2° E longitude and
o.r's latitude. It is bounded to the West by the Lake Victoria, to the East by Tinderet Hills,
to the North by Nandi escarpment and the South by the Mau escarpment (Fig. 1). The River
Nyando basin is administratively located in the Rift Valley and Nyanza provinces covering
the following districts (Table 6).
Table 6: Area covered by River Nyando and its tributaries
Name of District Area covered km"
Nakuru 15.5
Uasin Gishu 104.29
Nandi 508.41
Kisumu 768.20
Kericho 1210.00
Nyando 1012.00
Total Area 3618.00
Source: MWRMD, 2004b
29
10 0 10 20 Ki lometers
,~,
• Townso Distrid in the BasinN River NetworkE:l Nyando Basin
_ Lake vidoria.-=
Fig. 1: The Nyando Basin
The land slopes generally in the Northeast - Southwest direction with the altitude varying
from about 3,000 m above sea level at Londiani and Tinderet forests to 1,000 m above sea
level at the shores of Lake Victoria around Winam Gulf. River Nyando with a stretch of
about 150 km starts from the Mau escarpment traversing three topographical zones to
discharge on a fairly flat swamp area at the Nyakach bay (Fig. 2). On its upper reaches from
75 to 109 km, the river flows through a V-shaped valley in the mountainous area of the
Londiani with an averag~. channel width of 20 m and gradient of 1 in 45 - vertical to
horizontal (Njogu, 2000). In its middle reaches from 40 to 75 km the Nyando meanders on a
narrow valley floor with an average channel width of 40 m and a gradient of 1 in 160. This
area is popular for sugar cane growing at both large-scale and small holder levels. It has also
small-scale tea cultivation at the higher altitudes (Onyango et aI., 2005). This is the transition
30
zone between highland and lowland and also between to ethnic communities that occupy the
basin; the Luo who are peasant farmers and fishermen live in the lowlands and the Kalenjin
mostly livestock keepers on the highlands. At this transition zone the basin lies between
1300m and 1800m above the sea level and is characterized by gentle slopes. The lowermost
45 km reach is characterized by pronounced meandering over a wide flood plain with an
average channel width of about 60 m and the gradient flattens further to 1 in 700; this is the
famous Kano plains which experience serious flooding and heavy silt deposition almost on
annual basis (WMO, 2004). The lower zone lies between 1300 and 1000m above the sea
level (Fig.2).
3.1.2 Geology and Soils
The geology of the basin consists of scarps formed by the rift faults, the latter shaping the
Kavirondo rift which branches from the main north-south Rift Valley system. The basin can
be classified under the upper zone and the lower zone, with an intermediate middle
(transition). The upper zone ranges from 1900 m to about 3000 m above mean sea level; and
the lower zone extends from 1,300m down to Lake Victoria which is about 1,000 m above
sea level (Fig.2).
31
Hyando Sub-Basin Relief Pattern
Altitude (10
HJOO To 1309
1300 To 1609
1600 To 1999
1900 To 2599
Fig. 2: The Nyando Basin Topography - Wanjohi, 1999
The basement rocks are the Nyanzian rocks on top of which lie the Bukoban rocks (Muthusi,
2004). The resulting soils have unique characteristics both physically and chemically due to
weathering and deposition. The catchment's physiography is dominated by hills such as
Nandi hills, Mau escarpments, Tinderet hills and the Londiani hills. Many rivers and streams
originate from the above hills and scarps, but face the greatest threat from human activities
'\ ,
which is attributed to rapid population growth.
Here there is high deforestation for either farming or settlement purposes, charcoal burning or
commercial timber for construction or furniture industry leaving the soils bare and making it
32
morevulnerable to degradation. Soils of the study area are derived from a variety of parent
materials such as phonolites, granites, gneisses and dolerites (Atieno, 2004). In the highlands
the soils are mainly light clay with good moisture holding capacity and high fertility rating.
The steep escarpments are dominated by stony soils which are shallow with numerous rock
out-crops making it difficult for cultivation purposes. Different types of soil cover the
highlands and lowland zones of the basin. Soils on the highlands include nitisols, luvisols and
cambisols (Atieno, 2004). Upland soils are shallow to moderately deep of moderate to low
fertility and are structurally stable. These soils are prone to sheet and rill erosions due to very
intensive and highly erosive rainstorms. Soils found on the degraded hills and volcanic foot
ridges include leptosols and cambisols. They are shallow and stony and easily eroded when
disturbed by tilling. Soil loss through erosion is greatest in the middle zone due to steep
~
III* slopes and remarkable changes in soil properties. This area is prone to gully, rill and sheetz::l
~erosions (KSS, 2003).~
~
Most of the soils in the basin are fragile and prone to high erosion because of the varied nature
of the topography and anthropogenic activities such as poor land use practices, deforestation
accompanied by high rainfall intensities and runoff. Sedimentation rates of 10-90 tonnes per
km2 per year have been reported (KSS, 2003). Lowland soils are deep to very deep, of
moderate to low fertility and are structurally unstable. They include vertisols, luvisols, glycols
and fluvisols. The vertisols have shrinking and swelling properties and in some areas have
sordic phases in sub-soil layers (LVEMP, 2000). Extensive sordicity is the cause of severe
gully erosion in some areas of the basin. Table 7 shows the locations and problems associated
with various types of soils in the catchment.
33
Table 7: Types of soils and their associated problems
Type of soil Location Associated problems
Vertisols Kano plains and Winam Gulf Poor drainage, sordicity, workability, soil
area compaction, gully erosion
Plano sols Londiani and lower Nyakach Poor drainage, soil compaction, low fertility
Glycols Swamps and Poor drainage, oxidation of peat
marshes(wetlands), Winam
Gulf area
Nit sols Tinderet Hills Leaching of nutrients
Fluvisols Riverine flood plains Variable fertility, low moisture retention
Luvisols Kipkelion Soil erosion, low moisture
Phaeozems Kipkelion and Nandi Hills Stony sub-soil layer, low moisture
and
Cambisols
Source: L VEMP, 2000
The plains with the gentle slopes are dominated by heavy grey black clay soils of poor
drainage and occasionally swampy soils. Soils are poorly drained because of the high clay
content and the low gradient (LBDA, 2003 and 1985). The river here is slow moving and with
the heavy silt load, deposition is high and the flooding frequent. The meandering is more
pronounced here with very low slope and a lot of deposition. Annual flooding near the delta
leaves rich alluvial deposits that yield good harvests. The deposits also cause the river channel
to shift which has led to inter-clan conflict in the delta region because the river which is used
as the boundary keeps on shifting; at one time the area is cultivated by one division at the
other by another. Because of its fertility the area near the delta is more valued (Ong and
Orego, 2001).
Most rocks in the Nyando basin are of volcanic origin and occur in the upper parts of the river
catchment. Mineral resources are scarce. Sand and rocks, for ballast occur in various parts of
the basin. Tuffs and conglomerates associated with tertiary volcanic and thermally
metamorphosed sediments are found near Kedowa, limestone around Koru whereas basalts
and phonolites occur near Awasi and Nyakach hills.
34
3.1.3 Climate
The climate of the Nyando basin is sub-humid inland equatorial type, modified by the effects
ofaltitude, relief and the Lake Victoria. Temperatures in the basin range from 17°C to 25° C
with an annual mean temperature of 23° C. The rainfall in the basin is controlled by the
northward and southward movement of the Inter-Tropical convergence zone (ITCZ) and
shows considerable spatial variations closely related to altitude, proximity to the highlands and
nearness to the lake (Odada and Otieno, 1990). Rainfall is the single most important element
of meteorological observations that would give an indication of climate at any place. It has a
high spatial and temporal variability and it significantly varies both annually and seasonally.
Annual fluctuation of monthly rainfall is large, varying from 1,000 mm along the shores of
Winam Gulf to 1,500 mm in the East and over 1,800 mm along the Northern boundary of the
basin. The mean annual rainfall in the upper parts varies from 1200 mm in Londiani, 1390 mm
inKipkelion to over 1600 mm in West Mau forest. The mean annual rainfall in the lower parts
ranges between 1300 mm and 1400 mm in Muhoroni and Chemelil areas, while in the Kano
plains it ranges from 1000 mm to 1200 mm (WMO, 2004).
In the middle reaches the annual average maximum temperatures range from 27° C to 31° C
and the annual average minimum temperatures range from 14° C to 19° C. The downstream
reach is the hottest and the driest region of the basin with an annual average maximum rainfall
ranging from 800 to 1200mm.The annual average maximum temperatures range from 29° C
to 31° C, while the annual average minimum temperatures range from 12° Celsius to 16° C
(WMO,2004). There is no distinct dry season in the basin; however the driest months are
usually January and September (WMO, 2004). The rainfall pattern is trimodal with peaks in
April, August and November (Njogu, 2000). Air temperature ranges from 20° C to 25° C
throughout the year except in the mountainous parts. Relative humidity varies from 55 % to 70
% throughout the year (WMO, 2004).
35
3.1.4Drainage and Hydrological Network of the Nyando Basin
The Nyando catchment is divided into three sub-catchments; river Ainamotua drains the
northern catchment flowing from the Tinderet forest. The northern catchment covers an area
of840 km2. The central catchment covering an area of 500 km2 is drained by river Tungenon.
Cherongit River drains an area of 900 km2 forming the southern catchment. The remaining
area is the swamp downstream. The rivers have a number of tributaries; for example,
Namuting, Ainapngetui, Masaita, Ainaposiwa, Awach-Kano, Asawo, Ombeyi, Nyaidho and,
others. The Nyando river basin has about 27 river gauging stations some operational and
others out of use due to various reasons (Fig.3). Data readings are taken by honoraria/meter
readers. These are volunteers who are paid honoraria for their work occasionally which partly
accounts for some of the missing data. The collected data are usually river gauge heights ;.
water elevations above some datum from which rating curves are developed to compute
discharges at each gauge height reading. The discharges can then be analysed for their
frequencies to be used in various design requirements. The river gauging station 1GD03 is the
best station for the basin analysis because most of the runoff from the basin passes through
here after satisfying most of the losses (Fig.3). The basin has 21 Rain gauge stations spread
within and just outside it.
36
•. Gauging Stations
• Rainfall Station
1\/River Network[:::J Nyando Basins
10 0 10 20 Kilometers
-=~~~::iiiiiiiiliiiii~=-==
Fig.3: The Nyando River basin Hydrological network
3.1.5 Population and Socio-Economic Activities
Based on the 1999 population census Nyando basin had 746,515 people with Kericho dis .
having 42 percent, Nyando district 35 percent and Nandi South 19 percent (Table 8).
Population in the principal urban centres (Table, 9) contribute to runoff build up due to
pavements, roofs, roads, paths and related structures.
37
1:hl 8 R I' . h N db'a e : opu ation In t e iyan 0 astn
District Area in the Population density Population (1999 census)
basin % (Persons per Km2)
1979 1989 1999 Total Proportion
Population in the
basin
Kericho 42.2 151 182 222 468,493 315,061
Nyando 34.7 181 228 284.6 299,930 258,738
Nandi 18.7 109 156 200 578,751 139,857
Nakuru 2.4 90 118 164 1,187,039 18,212
Vasin 1.1 89 138 187 622,705 7,769
Gishu
Koibatek 0.4 20 32 31 138,163 3,279
Kisumu 0.5 181 230 549 504,359 3,599
TOTAL 100.00 3,799,440 746,515
Source: Mungai et al (2002) and Republic of Kenya (1979, 1989, 1999)
The average population density in the Nyando basin is 314 persons per kro2 with some areas
having over 1,200 persons per km2 (Table 10). Figures 4a and 4b show the spatial density
distribution in the basin during the 1979 and 1999 census periods.
t: bl 9 hi . I b tra e . nctpa ur. an cen es.
Centre District Status Population (1999 Census)
Kericho Kericho Municipality 93,213
Nandi Hills Nandi South Town council 77,514
Londiani Kericho Town council 34,184
Muhoroni Nyando Town council 31,145
Ahero Nyando Town council 30,327
Kipkelion Kericho Town council 26,786
Others (5) - Others 13,777
Source: Central Bureau of statistics, 2005
Table 10: Population density-Nyando basin
Name of District Census Results for Population Density in
persons/Km'
1979 1999
Nyando 181 284.6
Kisumu 230 549
Nandi 109 200
Kericho 151 222
Average Density . 170.25 313.9
Source:~,2004a
38
POP!) lM)n d~ns'ty
13 - 100
101 - 200
_156·201
202 - 'GO_ 2 -318
_ 319-410
_ 411- ;.>90&
Fig.4a: Nyando Basin Population Density -1979, (ICRAF, 2002
Populat on den$ity
0- .,
67 - 118
_ 118-161
_ 167-282
_ 282 40
Fig. 4b: Nyando Basin Population Density - 1999, (ICRAF 2002)
39
Thisrapid population growth will exert more pressure on the available resources in the basin
worsening environmental degradation and poverty. The poverty levels in the area are already
highwith Nyando district having 66%, Kericho district 58% and Nandi district with 63%.
Despite the establishment of irrigation schemes, the lower catchment areas still experience
highincidences of poverty attributable to poor infrastructural facilities (Onyango et aI., 2005).
TheNyando river basin accommodates a large rural population. It has vast fertile land suitable
for intensive agricultural development, especially in the lowland areas.
In the upper zone especially the high potential areas of the upper midlands the area is under
fallow, natural forest and cultivation of tea, coffee and annual crops. The area under tea and
sugarcane has increased because of a substantial shift from small holder agriculture to large
scalecommercial farming.
The intermediate middle zone forms the transition zone between the highland forest areas and
theplains in the lowlands. The most conspicuous feature of this zone is the steep escarpment
of the Nyando. Here there is a rapid change of elevation promoting both erosion and
deposition. It is also in this zone that most of the streams and rivers meet to form the river sub-
systems of the Nyando. In the lower reaches of River Nyando flooding is frequent and seems
to recur much more frequently than before.
One of the major factors for inadequate human resource development in the basin is lack of
access to safe drinking water leading to an upsurge of water borne diseases especially in the
lowerreaches of the basin. With the rapid population growth more pressure is being placed on
the land and the water resources. The land is being more degraded and becoming less
productive, and with the diminishing water resources there is need to establish the extreme
hydro-climatic trends to help in the present and future management of the resource
40
IfI=AIVATTA IIl\HUFRSITV LIBRARY
3.2 Data Collection
Datawere collected both in the field and relevant offices. Primary data was collected from site
andincluded; physical environmental evaluation centred on the effects of floods and droughts
on livelihood.
Rainfall data was acquired from the Kenya Meteorological Department. Data for various
stationswithin the basin were acquired, for analysis however, the study considered Kaisugu,
Ahero,Kipkelion and Muhoroni stations.
Discharge data were acquired from the Ministry of Water Development and Irrigation. Raw
dailyflow records from 1969 to 2000 were obtained. The discharge data for stations, 1GD03,
IGD04 and 1GC06 were used.
Land cover data and satellite images were acquired from ICRAF, LVEMP and the Ministry of
Agriculture. Population data were from District Development Plans for Kericho, Kisumu and
\Nyando districts (2002) and from the Ministry of Planning and National Development. Land
use data was acquired from the Ministry of Agriculture and water quality and related
environmental data from Lake Victoria Environmental Management Programme.
3.3Data Analysis
Various data collected were analysed using different methods as shown below:
3.3.1 Rainfall Analysis for Nyando basin
Rainfall data acquired are point rainfall values taken on a daily basis over a loag period of
time therefore had to be tested for consistency. These daily rainfall values were computed into
monthly and seasonal values and then translated into annual values (Appendices 1-4). These
41
data were then subjected to various analyses including seasonal distribution and variation,
annualdistribution and variation, reliability and variability for the basin.
The annual rainfall data was subjected to quality analysis to ascertain whether they are
homogeneous and consistent. Double Mass Curve method was used for the stations chosen
(Ogallo, 1981). In this method cumulative rainfall data for one station is plotted against
cumulative rainfall data for the other stations in the region. Cumulative annual rainfall values
for Ahero were plotted against cumulative values for other stations in the region and
cumulative annual rainfall values for Kaisugu were plotted against cumulative values for other
stations in the region. The same procedure was done to the other stations.
Meanmonthly rainfall values for some selected rainfall stations over the same period of record
(1970-2000) were plotted against time. This is to show monthly and seasonal rainfall
variations around the stations in the basin.
Annualrainfall values for the various stations were also computed and isolines plotted to show
annualrainfall variation within the basin. This enables the basin rainfall distribution be seen
ata glance and assists in indicating the areas that are contributing to the basin flow most and
therefore enables environmental management decisions to be taken much more promptly and
accurately.
Annual, Long rains and Short rains data were computed and plotted against time. These will
indicate the variations and show any trends in the variations with time. This was done to
enable annual planning and management of water resources be made fairly accurately in the
basin. The same computation and plotting was done for Kaisugu, Ahero, Kipkelion and
uhoroni rainfall stations.
42
Variability and reliability analyses were done using annual and seasonal rainfalls in the basin
for the selected stations in the basin. Annual and seasonal (Long rains from March to May-
MAM and short rains from October to December-ONlf) rainfall values were computed from
dailyrecords using equations (1) and (2) below:
0'CV = -=- XI00 (1)
x
R= 100-CV (2)
Where;CV= Coefficient of variation
R = Reliability
0' = Standard deviation
x = average rainfall values
3.3.2 Flow Analysis for Nyando Basin
Flow data acquired were daily flow records over a period of thirty one years, from 1970 to
2000. Monthly flow data were computed over the same period by taking cumulative values for
eachmonth and dividing by the number of years of records.
Seasonal discharge variation in the Nyando basin. was also computed over the same period.
Cumulative seasonal discharge from 1970 to 2000 was computed and their means calculated
to determine the seasonal flow variability.
Annual Flow Variability in the Nyando basin was computed by identifying annual peak and
minimum flows and plotting them against time to show annual variability for the same period.
Mean annual discharges were also computed and plotted against time to show how the mean
varieswith time and with the extremes.
43
FloodFrequency Analysis for the Nyando River was done using Gumbel, Log Pearson-3 and
General Extreme Value (GEV) fitting distributions. Regular Gauging station 1GD03 was
chosenfor analysis because most of the flow causing havoc downstream passes through it
Annual peak flow values were identified and analysed, Flood Frequency analysis is usually
done to help in flood plain zoning and the design of flood control structures such as
dykes/embankments. Flood zoning ensures that housing and industrial developments which
arepermanent in nature are not located in high risk zones. Flood frequency analysis also helps
in predicting the magnitude of floods which are most likely to occur in a place at a
determinable time to enable a befitting response or equal mitigation measures to be put in
place.
3.3.3Runoff Ratios for the Nyando basin
Therunoff ratio is a measure of the overall basin hydrologic response. It is a function of the
basincharacteristics such as land cover, soil types, topography, population density and rainfall
andland use factors. Annual discharge and rainfall data for RGS 1GD03 were taken because
most of the discharge from the basin is registered here and the basin hydrologic response is
better assessed from this point Rainfall and discharge data from 1969 to 1996 were used
(Appendix 11).
3.4Land cover and land use in the Nyando basin
Land cover data and satellite (Unclassified, 1973 and 1986) images were acquired from
ICRAF (Figs. 5a and 5b), LVEMP and the Ministry of Agriculture. Topographical map was
acquired from the Survey of Kenya. Population data were from District Development Plans
for Kericho, Kisumu and Nyando districts and from the Ministry of Planning and Housing
(2002).
44
Land sat images from ICRAF included:
I,P160R60
2, P161R60
3, P170R60
4,P171, R60
Usingpath 160 row 60, path 161 row 60, path 170 row 60, path 171 row 60 the unclassified
land images of the Nyando basin for 1973, 1986 and 2003 were produced (Figs. 5a, 5b and
5c) .
. Supervised classification was done using Geographic Information System - IDRISI
programmewhich depicted the following:
3.4.1Land cover
Landcover was identified in the following classes:
1, Forest (indigenous)
2, Forest (planted)
3, Forest (cleared)
4, Bush land
5, Built up areas
6, Wet land
3.4.2Land use
Landuse classes were identified under four major activities namely:
1, Subsistence farming (Food crops)
2,Rice
3, Tea
4, Sugar cane plantations
45
Fig. 5a: Nyando Basin land Sat image (Unclassified 1973)
Classification and Analysis were done using ARCVIEW and ARCGIS programmes which
gave attributes to each entity or class. Each entity was analysed separately and the attributes
generated overlaid on each other to form either a land cover map or a land use map. The
topographical maps or toposheets were used for geo - referencing, that is, aligning or orienting
the image with the coordinates and in identifying the land cover classes and other features.
Other Land Use activities identified and investigated in terms of their effects on surface runoff
include sand harvesting, brick making and livestock keeping.
3.4.3 Population Analysis
Population growth from the past census (1979, 1989 and 1999) were analysed and projections
made using the growth rates for each district namely Kisumu, Nyando, Kericho and Nandi.
46
Thestudy analysed population, its growth and related it to the available resources especially
land and water.
Fig. 5b: Nyando Basin land Sat image (Unclassified 1986)
The aim was to ascertain the level of pressure exerted by the population growth which may
enhance flood build up and cause hazards in the lower catchment. Population density maps
showareas that are more densely populated in the basin with greater flood potential (Figs.4a
and 4b).
47
Fig. 5c: Nyando Basin land Sat image (Unclassified, 2003)
3.4.4 Built -up areas
Growth of urban centres was investigated with a view to relating built up environment to
magnitude of flooding in an area. The study quantified the built up area vis-a.-vis surface
water flow during storms. Figures 5a, 5b and 5c were classified using GIS-IDRISI
programme to depict built-up environment
3.4.5 Sand Harvesting
Sand harvesting especially in the lower catchment is on the rise, the study investigated their
effects on the extreme hydrological events.
3.4.6Brick making
Brick making in the basin is an increasing economic activity with wide environmental
implications, the study investigated its role in fuelling the extreme events.
48
3.4.7 Infrastructure Analysis
Aninspection was done of the water infrastructure including dykes, embankments and
culvertsto ascertain their efficiency in discharging flows, their physical appearance in terms of
maintenance,their level heights in terms of being overtopped by floods and their functionality
in termsof meeting their objectives.
3.5 Topography
From elevations and coordinates the study generated topographical and digital maps to help
analysethe topography of the basin in terms of its contribution to flooding in the basin.
49
CHAPTER FOUR: RESULTS AND DISCUSSION
4.1 Introduction
Thischapter outlines the results obtained from the analysis of the variables to the stated
objectivesin terms of rainfall, discharge, land cover, population, land use and topography.
Theanalysis of these variables and parameters and understanding their characteristics will
makeit possible to reduce the levels of hazards, exposure and vulnerability. It will enable the
stakeholdersput in place some level of preparedness and response in case the disaster strikes.
It will also help in mitigating against the impacts of the extreme events such as loss of lives,
property,crops and top soil, diseases, social disruptions and pollution to water courses.
4.2 Rainfall Analysis for the Nyando basin
Rainfalldata from the selected stations were tested for consistency, then analysed for seasonal
distribution, annual distribution and yearly variation within the basin.
4.2. 1 Data Quality Analysis for the Nyando basin
Thehomogeneity for all the rainfall stations using Double mass curve method showed that the
rainfalldata in these stations were homogeneous and consistent as shown below in (Figures 6
& 7).
50
Double Mass Curve for Ahero Irrigation Rainfall
100000.0
"" ~""
/,/ ~
"" ~""
,/,/
~
,/
./
,/ ~
,/ ~./,..V
~
~ ""..
KipKaiCum I
90000.0
80000.0
70000.0
60000.0
500.0סס
40000.0
30000.0
200.0סס
10000.0
0.0
-e '" co lO ..• ..• lO '" eoN g r;; <ri ?i ~ ~ on ~ •• 0 N rco '" N •.... '" N ~ r-, '"~ ..• r-, g •.... g -e- -e- ~ ..• '" '"'" lO ~ ~ ~ N '" ;;:;N N N
Ahero Cumulative
Fig. 6: Double Mass Curve Ahero Rainfall Station
4.2.2Seasonal distribution of rainfall in the Nyando basin
Nyandobasin ex~eriences peak rainfalls in May, August and November, however in each
rainfallstation the peaks may be experienced at different times. During the long rains areas
around Kaisugu and Kipkelion receive their peak rainfalls in the month of May while those
aroundMuhoroni and Ahero get their peaks in April (Figure 8). The whole region receives
peakrainfall during the short rains in the month of November. The third peak is experienced
inAugust.
51
Double Mass curve for Kaisugu
80000.0
....~ ~
~ ~....•.~
,/ ...."••••
,/ '"
~ ~...•~
"...••••• ••••,/
~ ~..~
I-KipAheCum I
70000.0
60000.0
50000.0
40000.0
30000.0
20000.0
10000.0
0.0 ... '" '" '" a> r-; eo '" '",..: • 6 <Xi <Xi 6 oti 6 6oo '" ... •.... eo -e- (; '" a> oo~ •.... 0; '" g: ~ ;:; ~ •....oo ~ t-, •....'" '" '" '" '"Cumulative Kalsugu
Fig. 7: Double Mass Curvefor Kaisugu Rainfall Station
It is worth noting that the August rainfall is greater than the short rains in the upper and the
middle catchments. That explains why in the upper and the middle catchments crops are
planted in March, July and October.
300.0
250.0
200.0
150.0
100.0
50.0
0.0
Seasonal Rainfall Variation in the Nyando Basin
I - 'KAlSUGU --KlPKElION - - MUHORONI - 'AHERO
r .•. ""\
/, .. \
, ,\ .. \ .•.• \
" / / .. - _ .•.• '1\.. .., I ~.
'/iV .. " # •• '-~ ; ..\ ; ..,..-- -,,~# r:-::-:-- ~ .. ; ; .; ~ ., .. .. ;~.- - ~ l/ '\ »<>; .."- •. ':- ~- .. •. \- --'" ...•.•. ~- ./-..." <,
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dee
Fig. 8: Seasonal Rainfall variation in the Nyando Basin
52
4.2.3Annual Spatial Distribution of Rainfall in the Nyando basin
Themean annual rainfall varies from north in the Nandi escarpments with 1800 mm towards
north-westwith about 1200 mm at Ahero and reduces to about 1000 mm at the lake shores. To
thesouth- west Mau escarpment receives the highest rainfall of about 1800 mm which also
reduces in a south-westerly direction to about 1200 mm at Ahero. Central areas around
Muhoroni also receive an annual rainfall of 1800 mm, but from this point rainfall reduces in
bothdirections to east to Tinderet escarpments where it is about 1260 mm and to the west to
Aherowith 1200 mm (Fig 9).
f-L...-..:..J:!l3::~ Annual Rainfall
N1140-1260
N1261-1380
N1381-1530
N.1531 -1650
N1651-1800
• Rainfall StationAI Nyando River Network'0 Nyando Basin•
Fig. 9:Annual Rainfall Distribution in the Nyando basin
53
4.2.4 Yearly Rainfall Variation in the Nyando Basin
Annualrainfall variation in the basin was analysed at some selected points within the basin to
~vea reflection on the variation over a long period. Some of the rainfall stations selected
includeAhero, Kaisugu, Kipkelion and Muhoroni. The knowledge of rainfall variation in a
placeis a water resources management tool in terms of planning for various activities:
agriculture,industrial production, power generation and water supply.
4.2.5 Annual rainfall variation at Kaisugu Rainfall station
Figure10 shows that the area around Kaisugu receives an annual rainfall of over 1500mm and
peakfalls of over 2000mm recur with a frequency of five years; such peaks were experienced
in1978,l983, 1988, 1994 and 1998. However the station registered a decreasing trend in the
generalannual rainfall over the years. 1970-2257mm,1978- 2171mrn, 1983-2093mm, 1988-
2ll2mm,1994-2059mm, and 1998-2101mm. The long rains over the same period remained
wellabove 500mm. The long rains also showed a cyclic pattern, it topped 800mm every seven
years.This is true for 1974, 1981, 1988 and 1994. Like the annual rainfall it is also showing a
generalfalling trend. The short rains depicted by OND (October, November, and December)
alsoshow a falling trend; from 418mm in 1972 to 263mm in 1996. 1997 and 1998 were EI
Nifio years and were marked with excessively high rainfall and discharge in rivers especially
riverNyando. From this analysis it is evident that the rains are generally falling yet the floods
are increasing. It is also fair to conclude that it is either the intensities of rains which have
increased, which may be remotely so, or increased human activities, which look like the major
cause.
54
4.2.6Annual Rainfall Variation at Ahero Rainfall Station No. 9034086
AroundAhero the annual rainfall is averagely 1200mm (Fig 11). The average annual rainfall
is 1200mm except for 1997 when the depth shot up to 1667mmm, attributable to the El Nino
rains. In 1998 the figure was 1445.7mm to fall to 1277.0 in 1999 and 1126.1mm in the year
2000.Theaverage long rainfall depth is 451mm.The average short rains is 277mm with most
ofthe years rainfall being over 250 mm. In 1997 the long rains was 416.2mm and the short
rains rose to 887.3mm. This El Nino rains continued to the early part of 1998 and the long
rains of the same year recorded a depth of 519mm and short rains of 208mm.
Kaisugu Rainfa" station
~~ /. ~ A~'2000.0 ~\r+--t-t-/-+~-+--+I++--+-t-+/--t-~+-t-I\r--+-+-+71-,+.\+-+-+-t--t, .A.t---+-t-~+---\1-+--I
•../' \ ,'! 1\ ~ oJ \ ••~ \
" '\ .. , 1"1-0 .•.•1~·°tl-t-t!-t-~\/Jrt-tlli4t~!-t-I4~'flt-tl-t-t!-t-Ii-tI-t-~-----.- 'Annuall
-MAM
- 'OND
\
~~~~~~~!~~~~~~~~~
Years1970-2000
Fig. 10: Kaisugu Rainfall Station
55
Ahero Rainfall Station
1800.0
800.0
I \~ , I' "\ I \ '\ /' / \ I
../ / 1/ " - 'Annual\ ~ V \...1 ~ J~ -MAM•
" - 'OND& •1\ A r r-.... ...; '" i/ It\Ir-'" r".. I\..-1-0.. I 1\ .....! ./ • "
~ fJ " ...•..~ , ,\ r-.. .J 1'. .J \ V.• I •••• ...• .••... , '" ,oJ ..J , ,
1600.0
1400.0
1200.0
1000.0
600.0
400.0
200.0
0.0
Years
Fig. 11:Ahero Rainfall Station
4.2.7 Annual Rainfall Variation at Kipkelion Rainfall Station No. 9035020
Kipkelion area gets an average annual rainfall of 1083.3 mm (Fig. 12). The annual rainfall
rose from 676.4 mm in 1971 to 1357.6 mm in 1988 and started falling to 887.2 mm in 1999.
The long rains with an average value of 370.6 mm rose from 413.3 mm in 1970 to 608 mm in
1987 then gradually fell to 145.5mm in 1999.The short rains show no much variation from
200 mm except for 1977 when the figure shot to 457.3mm, which was a wet year. From 1985
to 1997 the area has been experiencing a three year cyclic rainfall of over 1200 mm. From
1979 to 1996 long rains are seen to be on the rise. Between 1970 and 1978 both the annual and
long rains are fairly low. The short rains are averagely around 200 mm with very little
fluctuation over the same period.
56
4.2.8Annual Rainfall Station at Muhoroni Rainfall Station No. 9035315
Thisarea receives an average annual rainfall of 1576.3 mm. The graph in (Fig.l3) shows that
rainfallrises from 1087.0 mm in 1984 to 2145.0 mm in 1996 and gradually falls to 1824.00
mm in 2001 and to 670.6 mm in 2004. The year 1989 with a peak value of2825.1 mm was an
exceptionally wet year. This is also true for other stations. The long rains average 559.4 mm
andfluctuate around 500 mm except in 1989 when 1482.2 mm of rainfall was recorded. The
shortrains average 344.5 mm and generally lie below 400 mm. In 1997 however, the value
roseto 1144.5 mm and this is attributable to the El Nino rains. Annual rainfall has been
falling steadily from 2900 mm in 1989 to 1800 mm in 2001 and to 1000 mm in 2004.
Muhoronireceives more annual rainfall than Ahero area.
Kipkelion Rainfall station
1600.0
400.0
"I 1\\ " ~\\ ,\ , , ,- r I 1\•••• I V~ I ~ II 'I ~ j .•.•.,\ I ~ \ \ I~
, I j V ",'\1 \1
I~
" II' v \ .J '"iV r hIA \ " l/', If. If tvr'\f\ V ....f:\ J IV ~ ~ h, J r , "- V -- fY ," ],- •.. 1\ f\l ~ , I' -~ .. 1\# I'"J .j... ,.'
1400.0
1200.0
~ 1000.0
.5
;:J BOO.O
~
"ii 600.00:
200.0
0.0
Years
Fig. 12:Kipkelion Rainfall Station
57
Muhoroni Rainfall Station
lXlO.O
500.0
v\
I ,,
I
II 1\ I' .....
\ / \I /,
I i' .... ;~\ I \
\ . 1'/ '\/ ., , I ,\.•. •..... ./ . , A " \, , I \ ~, , ~~ I " \~ .• ~ r\ ' .... .•... - - ~ , .. .. ,-~ V" ~ - - \..~ ~.• , v~ --.. ,,- -.r- J ~ ,- ~ - r---~
2500.0
E 2000.0
E
£•j 1500.0
~
i
~ 1000.0
0.0
Years
Fig.13:Muhoroni Rainfall Station
4.3Variability and Reliability of rainfall in the Nyando Basin
Rainfall variability and reliability analysis for the Annual and seasonal rains showed the
followingresults:
4.3.1Kaisugu Rainfall Station
Annualrainfall variability around Kaisugu is l3.2% and reliability is 86.8% (Table 11).This
meansthat the fluctuation in annual rainfall is small (l3.2 %). Any well planned and correctly
timed agricultural activity will do well especially the rain-fed farming. Water supply projects
can also be reliable up to 87 %. The long rains show less variation than the short rains
therefore more reliable to the tune of 75.65 % against the short rains value of 54.4%. Any
agricultural planning for rain fed farming as long as the timing is good and the right choice of
crops is made will yield good results. In hydrological terms it means that flow, if well
managed will suffice all the water use requirements throughout the year.
58
4.3.2Ahero Rainfall Station
Annualrainfall variability around Ahero is 13.1% and the reliability is 86.9 % (Table 11). This
isan indicator of an area fairly served with rainfall. The area receives less rainfall than the
upperhighland areas but the variation is smaller than those of the other stations. Well planned
agricultureand water supply schemes should be able to do well.
4.3.3Kipkelion Rainfall station
Annualrainfall variability at Kipkelion is 16.3% and reliability is 83.7% an indication that that
therainfall varies from the mean by at most 16.3 %. Any planning in terms of water resources
usagewill be reliable to the tune of 83.7 %. That means as long as demand is within the
supplybracket then rains are adequate (Table 11). The short rains are more unreliable than the
longrains because they vary more even though with proper planning deficient periods can still
be managed.
4.3.4 Muhoroni Rainfall station
Annualrainfall variability around Muhoroni is 30.3 % and a reliability of 69.7%. This means
eventhough Muhoroni gets more rainfall than Ahero water resources planning is easier around
Aherowith lower variability than around Muhoroni where it is higher. The short rains are
themost unreliable around Muhoroni. Planning must take into account that the variability is
up to 58.3 % during the short rains (Table 11). Figure 14 shows that rainfall reliability
generally increases from the highlands to the lowlands.
Table 11: Rainfall Variability and Reliability
Annual Long Short Annual
Rains Rains
Muhoroni 30.3 43.3 58.3
86.8
Long
Rains
54.4
Variability ReliabilityStation
Kaisugu 13.2 24.4 45.6 75.6
Short
Rains
Ahero 13.1 21.5 47.6 86.9 78.5 52.4
Kipkelion 16.3 36.4 46.6 83.7 63.6 53.4
69.7 56.7 41.7
59
Towardsthe south around West Mau rainfall reliability is 86%, an indication that throughout
!be yearwith good planning and management water recourses are adequate for all utilities. In
!be East around Tinderet forest reliability is 76 %, here rainfall variability is more than around
Maucomplex therefore less reliable hence needs better management. To the North, at the
Nandiescarpments reliability is even less than at Mau and Tinderet escarpments. With a
variabilityof 34 % the areas water resources need much more care in terms of planning and
managementfor the various uses than the other parts of the basin. It shows that rainfall
fluctuationabout the mean is more towards the highlands which incidentally gets more rainfall
thanthe lowlands. Reliability is less towards the north in the Nandi and the Tinderet
escarpments. This implies that rainfall is more reliable around the Mau complex and in the
lowercatchments around Ahero in the plains.
60
N Reliability of rainfall
• Rainfall stations1\1Nyando River NetworkCJ Nyando Basin
• s
Fig. 14:Rainfall Reliability in the Nyando basin
4.4 Flow Analysis in the Nyando River (Data Analysis)
Mean monthly discharges from 1970 to 2000 were analysed. Annual peak and minimum flows
were plotted against time and their return periods determined using Extreme Value (EV1-
PMW), General Extreme Value (GEV-PMW) and LogPearsonType3 fitting distributions.
61
4.4.1Seasonal Discharge Variations in the Nyando Basin
Peak discharges in the three gauging station IGD3, IGD4 and IGC6 occur in the months of
\fay,August and November (Fig. IS), coinciding with the times peak rainfalls are experienced
inthe basin. Greater mean monthly discharges are experienced in August and not in Mayas
maybe perceived to be the case. This could be so because the long rains mostly go into June
andthe August rains start in July; this finds when the antecedent soil moisture is still high,
thusmost of rainfall does not infiltrate into the soil because of the saturation level, therefore
flowsinto the channels as flush floods. The discharges are on a downward trend from January
tomid February. This is the driest part of the year. In March the discharges start rising at the
onsetof the long rains to reach their peaks in May. The discharges are on a downward trend to
Junewhen they start rising to reach their peak in August People harvest in July so during
thistime the ground is left bare because of the just concluded harvesting. Livestock are also
left to roam the fields to feed on the stocks of crops, leaving the soil barer and given the fact
thatfields are also prepared for the next cropping thus accelerating surface run-off to the river
channels. The livestock which are left to roam the fields also dislodge the soil with their
hooves making the soil readily washable in even smaller surface runoffs. The hydrological
implications are that surface run-off is increased, soil erosion is enhanced and soil nutrients
aredepleted for plants. Economical implication is that lower crop yields are realised at higher
costs because soil fertility has to be artificially improved by use of fertilizers. This in turn
raisesthe level of poverty· in the region promoting social ills and crime.
62
River Gauging Stations 1GD3, 1GD4, & 1GC6 mean monthly discharge (1970-2000)
~o,----------------------------------------------------.
~o~------------------------------~~~--------------~
¥ 25.0 ~---------------------------+---------~------------~
••
~~ 20.0 +----------------F-----~~----____;F_--"""'''''"''''--_\__----------___j,------,-=-:-:----:--.
E -1GD3-Nyando
~ -1GD4-Nyando
B -1 GC6-Nyando
.E 15.0 t-------------;t------:;;;,;;;;;;;;;;;;;;~"--------------'------T--I
<II!!'01
J:
U15 10.0 +------------I-_~------------------------------~_\_____j
5.0 +---oo;;;;:::c---------,F--------------------:F_----~.,__------------___j
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dee
Months
Fig. 15:Mean monthly Discharges
4.4.2Annual Flow Variability in the Nyando basin (1970-2000)
Figure 16 shows annual peak, annual mean and annual minimum discharges within the basin
from 1970 to 2000. The results also indicate some cyclic occurrences and developing trends
which would help in planning and environmental management. The cyclic occurrences and the
intervals at which they occur would give an indication of roughly when to expect such events
and the magnitude and the extent of flooding expected. This time interval would serve as an
early warning signal for preparedness. The anticipated magnitude of floods will generate an
equal response if known in advance. For example, the magnitude of floods like the El Nino
required massive preparedness in terms of evacuation of people, livestock, movable property
and drainage maintenance to receive the torrential flows. The drought which followed
aggravated the already bad situation.
63
Whateverthe floods spared was finished off by the severe La Nina drought which followed.
Suchlike events if properly predicted can be met with some degree of preparedness to
mitigateon the most likely effects.
4.4.3Annual Peak flows at IGD03
Thevalue of peak discharges is seen to be on the rise from 1971 to 1997 and the trend is seen
tobe increasing. The year 1997-1998 was an El Nino year where the floods were up to 300%
abovenormal (WMO, 2004). From 1979 to 1987 the peak discharges considerably dropped.
This drop is attributable to the Government re-a forestation policy which emphasized the
plantingof at least two trees for each one cut. 1976 and 1984 were dry years while 1989 was a
wetyear (Figs 10 and 11). The discharge sharply increased from 120m3/sec in 1987 to reach
360m3/sec in 1988 and remained high till 1998 when it started falling. The sharp rise in 1997
wasdue to El Nino rains which continued to long rains of 1998.
Maximum, Minimum and Mean Disrcharges for 1GD3 In Nyando Basin
400.000.------------------------~
50.000+------.L------V---------lJ-------\r-I
350.000+---------------A.------------l
u 300.000+--------------J..-4------I\------l
III
Q)a.
gj 250.000+-----------------l----\-----I1I---1I--\-----l
li
E
u:g 200.000+--,....--------+\----....---+--\---+-++-+-----l
u
.5
<II2> 150.000+----+~"-----\-_.;,___,A_--I-+_--+-\---f---+--+--+--+----l
III.s;;;
U<II
i5 100.000f---'H------+--1------+--+------4-+----\------i
Years
Fig. 16:Annual flow variability for 1GD03-NyandoRiver.
64
The Government a forestation programme greatly improved the catchment in that the drop in
peak discharges meant less soil erosion, higher minimum flows, increased reliability of water
i1pplyprojects and safer lower catchments from floods. The La Nina period which followed is
denotedby a significant drop in discharge and confirmed by rainfall drop (Figs. 10 and 11).
4.4.4Annual Mean Flows at 1GD03
Annualmean flows remain well below 50m3isec for all the years of record. Between 1972 and
1976 it was on the rise. From 1977 it was on a falling trend to reach its lowest in 1984 which
wasan extremely dry year from rainfall records (Figs. 10 and 11), and then it rose to register
its highest record in 1989 which conversely was a wet year (Fig 13). The mean annual flow
doesnot show so much variation over the years. The implication is that emphasis should be on
seasonalflows because the annual discharges can lead to the assumption that things are okay
whilethe actual position on the ground, especially on seasonal flow is bad.
4.4.5Annual Minimum Flows at IGD03
Lowflows or droughts are temporary reductions in water or moisture availability significantly
belownormal. In the Nyando catchment low flows seem to be on a decreasing trend (Fig. 16).
Thiscould be because of deforestation and more land use activities on upper catchment and
thehill slopes which enhance surface runoff thus reducing infiltration which sustains the base
flowduring dry spells.
4.4.6 Runoff Ratios for the Nyando basin
The percentage of runoff ratio is increasing from one pentad period to the other as depicted on
table 12.
65
4.5Flood Frequency Analysis for the Nyando basin
Thelower reaches of the Nyando basin is prone to flooding causing social disruptions, loss of
humanlives, livestock, crops and top agricultural soil. It destroys infrastructure and
impoverishesthe people. The peak discharge data at 1GD03, from 1970 to 2000 was used and
theanalysis yielded the following results (Table 13).
A number of studies carried out produced varying magnitudes of flow for the same return
periods(Appendix 5).
Table:13Flow Return periods for the Nyando River
FittingProcedure Return Period in Years and Magnitude in m3/sec
I 2 5 10 25 50 100
I EVI-PMW 495.53 754.96 926.72 1143.75 1304.75 1464.56
12 GEV-PMW 495.67 755.35 927.11 1143.94 1304.66 1464.08
3 Log Pearson type 3 507.63 781.54 939.63 1111.56 1221.15 1316.89
As per this study a fifty year return period has a magnitude of about 1300m3/sec. This return
period is important for this analysis because it is the design flood used for the Nyando basin
protection works as proposed by Italconsult study of 1983.The embankments/dykes were
designed using Italconsult (1983) deduced values. Even the extensions of the dykes
constructed in 2003/2004 used the same deduced values. That is the basis for using Italconsult
values for comparison. From Table 9, peak discharge values deduced with a return period of
50 years has increased by 74% from the values deduced by Italconsult in 1983. Compared to
2003 MoWD study, the peak flow value has increased by over 57% above Italconsult values.
According to Italconsult the highest magnitude of flood (50 year flood) to be contained by
embankments was 750 m3/sec.
67
This value according to this study is even less than a 5 year flood and less than a 10 year
flood as per MWRD 2004 (Table 14), hence the increase in frequency of flooding to an almost
perennialevent. It is evident that the Italconsult values which have been and are being used for
protectiveworks can no longer hold.
Table 14: Peak Discharges (m3/sec) for different Return periods
Return This MRWD Master Italconsult Variance Percentage
Periods Study 2004 Plan1992 1983 From Variance
(Yrs) 2006 Italconsul
t
10 926.72 863.19 437 550 376.72 68.5
25 1143.75 1044.22 654 650 493.75 75.96
50 1304.75 1178.46 659 750 554.75 74
100 1464.56 1310.20 752 850 614.56 72.3
The MWRD study of 2004 gave a 50 year flood as 1178.46 M3/sec and this study gives it as
1304.25m3/sec an indication that the basin discharge continues to rise(Appendix 6 ). Two
possible reasons for this disparity could be that Italconsult used fewer data, only 11 records
which might have missed out on some major hydro-meteorological changes, or the basin has
undergone some major environmental change or both.
It is evident that floods continue to increase in frequency in the Nyando basin, rainfall
however does not show any corresponding increase, if anything, it is slightly on a decreasing
trend. It is therefore fair to conclude that the factors which increase the frequency and the
magnitude of floods are mainly anthropogenic such as: Rapid population growth, Land use
activities, Environmental degradation (Deforestation and Overgrazing) and Non maintenance
68
of theflood protection works. Other causes of floods include basin characteristics-topography,
vegetalcover, geology and infiltration characteristics, and Global climate change.
4.6 Land cover and land use in the Nyando basin
Landcover and land use in a basin have a significant effect on the quantity and quality of
wateresources available and the flow characteristics in any catchment area. Ownership and
tenureof land is vital in influencing resource use. Secure land rights in terms of ownership and
tenurewill improve land and water use and will also enhance environmental conservation and
management.
4.6.1 Land cover
Fromthe classified land sat images (Figures 17, 18 and 19) it can be deduced that in the
Nyandobasin, especially the lower zone the forest cover reduced from 5,974 ha in 1973 to
4,213 ha in 1986, a decrease of30 %. During the same period grassland increased from 3,824
hato 7, 276 ha an increase of90 %, sugar cane acreage increased from 5,282 ha to 10,895 ha
anincrease of 106 %, wetland increased by 47 % from 2,564 ha to 3,776ha and settlement
reduced 7,790 ha to 6,995 ha (Table 15, Figs. 17,18 and 19). Farming increased by 2132.37
% from 1973 to 1986.
The implication is that, the loss of forest cover translated to grassland, bare ground, farms or
wetlands. In the lower catchment with its gentle slopes water stagnation is enhanced. The area
will continue to be inundated and the depth and area affected may continue to rise. The
increase in wetland or swampy area can be attributed to sediment deposition. As stated earlier
annual rate of soil erosion in the basin is to the tune of 100-300 tonnes per square kilometre.
69
Thistranslates into annual silt deposition and pushes the wetland more inland and makes the
lake recede.
Table 15:Land cover and Land use changes in the Nyando basin
Zone Land use/ 1973 1986 Change % Remark
Cover ha
Upper Forest 3,836.07 2,499.39 35 Reduction
Zone Tea 768.87 1,323.27 72 Increase
Sugar cane 365.04 1,472.49 303 Increase
Settlement 4,707 5,300 13 Increase
Lower Forest 5,974.38 4,213 30 Reduction
Zone Grassland 3,824 7,276 90 Increase
Sugar cane 5,283 10,895 106 Increase
Wetland 2,564 3,776 47 Increase
Settlement 7,790 6,995 10 Reduction
Land cover and land use statistics
This fact is evident from the lake shore; there is a strip of very productive no man's land
which the residents are scrambling for. In the lower zone the forest cover is barely 0.3 %
implying that the zone is almost bare. The area under agriculture is 43 %; this means that the
soil is continuously disturbed making it susceptible to erosion especially that the soils here are
fragile. In the upper catchment the forest cover reduced from 3,836 ha in 1973 to 2,499 ha in
1986, a reduction of35 %. During the same period sugar cane farms increased from 365 ha to
1,473 ha, settlement s from 4, 707 ha to 5300 ha, tea from 769 ha to 1,323 ha and cultivated
farms reduced from 20,637 ha to 14, 911 ha a reduction of 28 %. Forest loss to agriculture,
bare ground, grassland or settlements implies less infiltration and percolation and with the
high intensity storms experienced in the Nyando basin this translates into flush floods.
70
Fig.17: Nyando Basin (Classified, 1973)
••- Built UP area and Bare landForest (indizenous. nlanted)Crops (cash and food crops)
_ Bush land and Grassland
_ Water
Fig.18: Nyando Basin (Classified, 1986)
71
c=::J Built_area
_Swalnl>
Grass_land
Forest...)lanted
_ Forest.shl>CIBushland
_ Cleared_forestc::::::J Nyando
s
Fig. 19:Nyando basin Land Cover (Classified, 2003)
Between 1986 and 2003 afforestation was done in the upper zone to the tune of 2640.51 ha.
Wetland increased from 1986 to 2003 from 3766 ha to 4,063 ha. This was an increase of7.6 %
and an indicator of high deposition and inundation in the lower zone.
4.6.2 Land Use
Land use affects runoff through its influence on interception, evapo-transpiration and soil
moisture movement. Removing vegetative cover on steep slopes for agricultural activities,
foraging for fuel wood and other wood products and overgrazing have enhanced massive soil
erosion and have resulted in the decline in soil fertility and soil structure. Land use activities
and changes differ from one zone to another among the three zones of the Nyando basin. In
the upper catchment large tea farms are dominant with some maize and horticultural crops.
Dairy cows are also kept. In this zone there is widespread deforestation. In the middle zone
72
~garplantations occupy vast areas and both traditional and grade livestock are kept. In the
lowereaches heavy free grazing of traditional livestock and intensive agricultural activities
aredominant. These land use activities change the structure and other soil properties mostly
~y impeding flow infiltration and making the soil loose thus being easily swept away by
surfacerunoff.
Theseland use activities continue to push more top soil to the drainage channels filling them
upsuch that they are no longer capable of containing the runoff which enter them. In the upper
catchmentbecause of high velocity, the rivers carry the sediments to the middle and lower
zoneswhere due to lower gradients deposition is high intensifying floods. In urban areas there
is little scope for infiltration and transpiration and all the rainfall immediately becomes direct
runoffproducing high discharges.
4.7 Population in the Nyando basin
Kenya has a fast growing population with an average growth rate of 3% per annum. The
country has realised a population growth from 6 Million in 1963 to about 35 Million in 2005
the land resources remaining constant. Other resources however continue on a diminishing
trend; forests are fast being destroyed that if not reversed quickly will compound the problems
of environmental conservation and flood intensification. In the Nyando basin population
growth has been rapid putting quite a lot of pressure on most of the environmental resources
which would help check the floods. The average population density has increased from 170.25
persons/ km2 in 1979 to 313.9 persons per km2 in 1999, an increase of about 84% (Table 10
andFigs. 4a and 4b).
In some areas population density is even higher than 1200 persons per km2 such as Ahero
township. The population density has increased in some places by more than 300 % (Fig. 18).
73
Theareas where the population density has increased more are areas which have experienced
graterforest loss such as the Mau, Nandi and Tinderet. This translates into increased human
activitiesin the basin which include; settlement, agriculture and livestock keeping. The lower
zoneis also densely populated especially on the flood plains and sometimes along the flood
pathsuch that even with small floods they are affected. Land is rapidly being subdivided into
smaller parcels due high population growth. Even the areas which were formerly
uninhabitable such as swamps and the lower Kimira areas are now human settlements.
Residents are. now scrambling for the strip which has been left by the receding lake water
especiallyfor farming because of its high fertility and because it is no man's land.
Rapidpopulation growth raises water demand more than the fresh water replenishment rates a
factorwhich calls for better water resources management in the basin. Because of population
increase pressure on land continues to be fragmented to smaller parcels to more people who
intensively use it. The larger tea commercial farms are slowly giving way to smaller farms for
growing subsistence crops. There is widespread deforestation to either give way for small
farms, settlements or charcoal burning. This is the source of river Nyando, and the wanton
destruction of forests, land use and general degradation is causing a lot of runoff which would
haveotherwise been trapped thus generating adverse effects downstream.
In the lower zone the population is estimated to have risen from 460,246 in 1999 to 548,438 in
2006 an increase of about 20 % (Table 16). This implies more intensive use of the available
resources; reduction of forest cover due to the clearing of wood for various utilities such as
firewood, charcoal, farming, settlements and construction. In terms of environmental
degradation it means enhanced soil erosion, more water stagnation leading to increased
flooding.
74
Assuminga growth rate of 3%, the population in the Nyando basin will almost double what it
wasin 1999 by 2020 (Table 17). In view of the fact that Nyando basin has by far surpassed its
land holding capacity and that the population is expected to double by the year in 2020, the
resourceswill be overstretched. Water demand will have increased about two fold yet the
waterresources are rapidly diminishing. It means the general quality of life will go down.
Therewill be increased pressure on land due to increased settlements; increased paved areas
whichimply reduced infiltration and percolation hence increased flush floods. The Nyando
basinpopulation is projected to increase from 748,754 in 2000 to 1,166,514 in 2015 which is
theyear Kenya is supposed to have met Millennium Development Goals (Agenda 21, 1996)
75
.-
Table 16:Population in the lower Nyando basin flood prone area
District Division Location Population Population Population
1999 census 2006 Est. Density 2006
Pers/km2
Kisumu Kadibo Bwanda 7,743 8,893 248.4
&Kanya~al
Kawino North& 12,179 13,990 279.2
South
Kochieng East & 17,244 19,808 435.3
West
Kobura &Katho 11,748 13,496 431.2
Winam Central Kolwa 19,387 22,269 623.8
East Kajulu 12,064 13,858 905.8
East Kisumu 27,626 31,733 973.4
East Kolwa 15,843 18,199 325
Kondele 69,521 79,857 16,636.9
Miwani 7,224 8,298 79
Kisumu West 17,478 20,077 925.2
Kolwa 70,402 80,869 6,628.6
Nyando Lower Asao 3,443 4,351 418.4
Nyakach Central Nyakach 5,106 6,453 645.3
East Nyakach 5,551 7,014 435.7
North East Nyakach 11,809 14,923 490.9
North Nyakach 5,093 6,436 273.9
Nyalunya 6,427 8,122 214.9
Pap-Onditi 8,173 10,328 408.2
Ranzul 3,645 4,606 158.3
Miwani North East Kano 19,509 24,654 408.2
Nyangoma 17,517 22,136 290.1
Ombeyi 21,003 26,542 298.2
Nyando Awasi 15,241 19,260 274.8
Kakola 18,634 23,547 1,010.6
Katolo 6,670 8,429 172.2
Kochogo 7,870 9,945 555.6
Onjiko 8,482 10,719 295.3
Wawidhi 7,614 9,622 181.9
TOTAL 460,246 548,434 437.0
Population computations from The Republic of Kenya, 1999 census
There will be more demand for firewood, timber for construction and increased agricultural
activities leading to decreased forest cover. This would lead to increased incidences of
flooding unless remedial measures are taken, and this can only be effectively done when the
extreme hydro-climatic characteristics are known.
76--
Tablel7: Population projections for the Nyando basin
YEAR PROJECTED POPULATION
2000 748,754
2004 842,718
2006 894,038
2008 948,485
2010 1,006,247
2012 1,067,527
2013 1,099,552
2015 1,166,514
2016 1,201,509
2017 1,237,554
2018 1,274,681
2020 1,351,162
Population computations from The Republic of Kenya, 1999 census
4.8Built - up Areas and Settlements
An increase in built up areas such as settlements and growth of urban centres means an
increasein paved areas in terms of buildings, structures, roads and footpaths. These further
impedeinfiltration and enhance runoff
Therapidly growing urban centres within the basin Ahero, Chemelil, Muhoroni, Kericho,
Songhor,Kaisugu, Masogo, Ombeyi, Awasi and Katito etc. are highly contributing to flooding
(Table 9). The built up environment of pavements, roads and buildings make surfaces
impermeable preventing infiltration such that runoff forms artificial streams which if not
checked promptly get bigger and become more destructive in terms of intensifying floods.
The network of drains in the urban areas may deliver water and fill natural channels more
rapidly than naturally occurring drainage, or, may be insufficient and spill over the banks thus
causing floods, or the natural or artificial channels become constricted due to debris, or
obstructed by river facilities, impeding drainage and overflowing the catchment areas.
77
Thebuilt up environment means the removal of vegetation to give way to buildings and other
infrastructure,an increase in paved areas and compaction of the soil which lead to less
IIlfiltrationand more surface runoff
t9lntensive Agricultural Activities
Intensiveeconomic use of flood plains for agriculture is on the increase in the basin. Land use
and tillage influence surface runoff and infiltration rates of the soils. Over-cultivation damages
thestructure of the soil by removing the vegetation cover thus alters the porosity and pore size
distributionwhich directly determine the surface and sub-surface movement of water. As is
seenin plate 3, the proposed bridge at Kimira is being protected by gunny bags from being
washedaway by floods. It is evident that even if the bridge is spared by the current floods its
safetyand stability are in serious doubt because the foundation has been undermined by
floods.
The sugar plantation up to the edge of the river further undermines the structural stability of
thesoil by loosening the soil and exposing it to soil erosion which fill up the channels thus
intensifying floods. Livestock graze up to the river edges, especially the lower zone (plate 3).
Farmers still use old traditional farming methods. Agricultural extension officers who used to
help farmers by training them on modern agricultural practices were withdrawn and the
initiative collapsed. No appreciable soil conservation measures are in place hence the increase
insoil erosion and inundation.
7~
Plate3: Intensive farming up to the river edges in the lower Nyando - Kimira
Bridge 12/5/2006
Generally disoriented, weaker and sick people could not actively participate in the economic
activities of the basin hence the rise in poverty levels. Ongoing development projects were
either destroyed by floods or stall (Plate 3). In plate 3 the foundation of the bridge structure
looks undermined by the floods that its structural stability is in doubt. There is a high
likelihood that it will be swept away by the subsequent floods. The fact that gunny or sand
bags are being used to protect the foundation can attest to this.
No buffer zone is left to help stabilise the banks. Even human settlements are put up very close
to the river banks further compounding the problems of soil erosion and flooding (Plate 4).
Lack of modern farming technology and high cost of agricultural inputs have led to poor land
use practices and catchment degradation. Such situations occur due to either weak legislation,
non existence of legislation or uncoordinated pieces of legislation as appears to be the case.
79
Plate4: Intensive farming and human settlement- lower Nyando Catchment 22/4/2006
Legislation should be such that it discourages potential permanent development or any activity
which would enhance environmental degradation as to increase floods in flood plains or flood
prone areas by planning, zoning and or removing any economic advantages or subsidies that
would encourage such development.
Where development has taken place like in some places in the lower Nyando basin, flood
proofing measures are implemented in such a way that the development does not further
enhance the flooding problem. An integrated management approach provides measures for
preventing a hazard turning into a disaster by adopting a systematic sequence of preparedness,
response and recovery. These actions are taken depending on the conditions of risk, social,
economic and physical settings; with the major focus being reducing vulnerability and
improving resilience to disasters.
80
Coffeec:=J Rice
_Tea
Sugarcanec:=J Nyando
20i!!!!!!!!!!!!!!!!Iiiiiii;;;;;;;l!!!!!!!!!!!!!!!!!!!!!!i20iiiiiiiiiiiiiiiiiiii40 Kilo mete rs
Fig. 20: Nyando Basin- Cash crops (Land use -Classified Image, 2003)
Forest loss to agriculture for both food and cash crops has aggravated the flood situation in the
basin because of increased loosened soil which is already detached and easy to wash away.
The fragile nature of the soil and the use of old traditional methods of farming soil erosion and
flooding are enhanced (Figs. 20 and 21).
Subsistence farming in the Nyando basin is the chief economic activity and the figure below
shows the extent of the activity (Figure 21). Because of the population pressure on land
resources intercropping of cash and food crops or even in the forest is being done. This
practice loosens and detaches the soil matrix and makes it vulnerable to being swept away
even by less intense rainfall, thus promoting surface runoff. The area covered by subsistence
crops as at 2003 was 219,100.64 ha.
81
_ subststenceereps
DNyando
s
20 0 20 40 Kilometers~I ~ __ ~~~ ~
Fig. 21: Nyando Basin - Subsistence crops (Land use -Classified, 2003)
4.10 Livestock keeping in the Nyando basin
In the middle and lower reaches of the Nyando river people keep many traditional cattle,
goats and sheep which consume so much vegetation and give so little in return. Overgrazing
leads to the removal of vegetation, destruction of soil structure and erosion. In the Nyando
district livestock employs about 84,000 people and in Nandi about 94,000. It was also
observed that due to land fragmentation the number of livestock kept outweigh the resources
for their sustenance at their disposal. There is a lot of overgrazing in the basin, the land is bare
and land loosened by the animal feet such that during rains there is quite a lot of surface runoff
due to low infiltration capacity and enhanced sediment flow into the channels. According to
field survey, especially in the lower catchment a lot of sheep are kept grazing on whatever
grass left by the cattle thus laying the area bare. The feet of livestock dislodge and loosen the
soil making it easier to be washed way even by small floods (Plate 5). Overstocking and
overgrazing are the main causes of sheet erosion through runoff due to compaction.
82
Plate 5:Livestock keeping in the lower catchment
Thisproblem is more pronounced in the Kano plains where free - range grazing is rampant;
livestockare left to roam about the all day only to be rounded up in the evenings. This is a
serious environmental threat especially in the lower reaches. In the middle and lower
catchment, goats prevent vegetation growth by constantly feeding on the sprouting leaves.
Overgrazingtherefore reduces the quality and quantity of the renewable vegetation.
Field survey revealed massive deforestation in the upper catchment, the Nandi hills, the Mau
forest and the Tinderet escarpments. This uncontrolled and unregulated human activity lays
the catchment bare thus increasing the velocity and the magnitude of any flood (plates 6 and
7). Rain falling on a thickly forested area may not produce any runoff because in the forests a
thick layer of mulch of leaves and grass absorb a lot of rainfall, leaving very little for runoff.
Even the little runoff encounters a lot of resistance during the overland flow enhancing further
absorption by the soil. Forests increase infiltration rates and regulate floods and help in
retaining precipitation. Under forest, soil structure is of higher grade and more stable (with
'83
lowdetachability and higher infiltration capacity) than under cultivation. Forest cover has a
variety of favourable impacts on soil conservation, hydrological balance and micro-climate. It
therefore helps in conserving and stabilizing the overall ecosystem in the watersheds. In non-
forested areas infiltration, interception and evapotranspiration losses are less and conversely
high runoff rates are experienced. High velocity flows especially like the ones experienced on
steep escarpments reduce infiltration 'and increase flood peaks such that even floods of smaller
magnitudes will assume higher proportions and inundate the lower catchments.
Deforestation also enhances soil erosion in the basin and is estimated to be between 100 and
300 tonnes per km2 per annum (Waruru et al., 2003). This implies that the average annual soil
loss in the basin is in the tune of 586 m3 per km'. This is the average sediment inflow into the
lake from the basin and is also equivalent to the average volume displaced by sediments
annually. This volume of sediments causes a backwater effect during floods to give the
impression of a very severe flood. It also impedes the rapid entry of flow into the lake causing
inundation of flood waters for sometime thus intensifying floods. When poor land use
practices occur upstream they affect homes, farmers and communities downstream with
increased flooding potential, more sedimentation, polluted and reduced water availability.
Land cover is one of the boundary conditions which directly or indirectly influence many
hydrological processes. Changes in land use occur mainly due to human activities including
urbanization, forestation, deforestation, agriculture and construction of reservoirs. Such
changes are continuously happening due to economic and population growth. There are severe
soil erosion and land degradation problems throughout the Nyando basin resulting in
accelerated run-off and sheet erosion over much of the catchment area leading to severe rill,
. gully and steam bank erosion in the lower parts of the basin. Soil erosion is severest at within
the transition zone from the highlands to the plains due to topography drop and change of soil
84
properties. The effects of this land degradation are seen in the formation of deep gullies.
Gullies are a symptom of functional disorder of the land, improper land use and are the most
visible result of severe soil erosion; they start as small drainage channels, sometimes from
animal foot tracks and grow in size and eventually form seasonal rivers (plates 8 & 9). Most of
the catchment soils are fragile and have high erodiblity. Any small human disturbance
translates into serious environmental disasters in terms of erosion gully formations (plates 10
and 12). The alluvial plains of the Kano have grey black soils with low infiltration capacities
which enhance floods.
The upper zone has experienced considerable land use changes, forests have been under
pressure from rapidly increasing population. Catchment degradation results in increased
frequency and duration of droughts and severity of floods, reduced natural water storage,
altered runoff and infiltration rates, and cost of increased man made storage requirements and
higher pollution of water courses. Some economic activities such as sand harvesting which is
highly being carried out along the main tributaries of the Nyando river undermines the
stability of the stream banks and reduces the natural bed storage of the rivers thus enhancing
and lengthening the periods of low flows (Plate 11).
Around Kericho and its environs, people are changing from large-scale commercial farms to
small-scale more intensive subsistence farming. This is partly due to diminishing sizes of land
holdings and shifts in crop production resulting from rapid population growth. Intensive land
usage means that more bare ground continues to increase in size, either in terms of farming,
settlement, pavements or urban set ups. This translates into increased and more rapid surface
runoff
85
KENYATTA UNrVERSITf LIBRARY
Plate 6:Deforestation at Tinderet hills-ICRAF 2000
Plate 7: Charcoal burning in Nandi hills-IC.R34F2000
The absence of soil cover makes the soil surface more compact, thus reducing permeability
and increasing erodiblity of the soil. Infiltration capacity is lower in cultivated areas than in
86
forests, implying more runoff and with the steep slopes of the upper and middle catchments
velocities are high, soil erosion is massive, filling up drainage channels through deposition
especially in the flatter lower reaches and also enhancing bank overflows.
The upper catchment covers West Mau, Kipkelion, Londiani, Tinderet, Nandi and Kericho. In
these areas some forest cover has given room to tea and sugarcane growing, human
settlements and livestock keeping. In the upper catchment, cultivation of stream banks and
steep slopes cause soil erosion and intensify floods by reducing the channel carrying
capacities. Forest loss to farming systems has reduced watershed forest area in the upper and
middle catchments. This implies increased velocity of flow, high sediment yields, reduced
channel carrying capacity and more frequent flooding in the lower catchment
The middle catchment is characterised by a drastic elevation change and senous soil loss
through gully and rill erosions, especially in the upper portion and considerable silt deposition
in the lower parts. The flow leaves the middle catchment with a considerably reduced velocity
to enter the lower catchment. In the lower catchment there is intensive agriculture, 43% of the
land is under agriculture. Hedges and bare ground constitute 10% of the area. The Nyando
district has a forest cover of only 0.27%, implying that there is virtually no forest in the
district. The flow enters this zone when the velocity has been greatly reduced and the land
being fairly flat, sediment deposition and river meandering is common and frequent over bank
spills occur (plate 13). The colour of water manifests high quantity of silt carried which
translates into fertile top soil loss from the catchment.
The major effects of deforestation include: decreased canopy interception of rainfall and
decreased transpiration, increased evaporation from exposed soil surface, decreased soil
infiltration due to changes in the soil structure, increased velocity of runoff after removal of
87
surface litter and roughness and decreased base flow during non-rain periods. In cloud forests
the trees intercept mist and this source of moisture is lost after logging.
Plate 8: Gully erosion in the middle catchment of the Nyando basin 20/4/200
Plate 9: Gully erosion, river Bugo 22/4 2006
88
Plate11: Land degradation in the lower reaches-Sand harvesting along
River Bugo 22/4/2006
89
Plate 12: Gulley erosion around Pap-Onditi 25/4/2006
Plate 13: River Nyando meanders -the lower catchment, ICRAF, 2000
4.11 Brick making on the Nyabondo plateau
Brick making is a thriving industry in almost the entire Nyabondo plateau. The top soil up to a
. depth of about 300 mm to 450 mm is dug and moulded to make bricks. The environmental
effects are three fold; first the top productive soil is removed leaving barren unproductive soil
90
ftv.~a,.",uSI 1I RAR
that can not sustain the growth of crops or other plant life. The area under brick making is
slightly over 50 km" Bare soil devoid of any vegetation or crops is expansive on the plateau
save for homes and a few pockets of vegetation. Secondly, the bare soil left after the removal
of the vegetation and the top soil intensifies floods by promoting free and rapid surface flow
during storms (Plates 14 and 15). As seen in the plates, ponding which is a prelude to flooding
is rampant in the area. There is need to enforce legislation on the use ofland -for sustainability.
For example, the area which is being degraded by brick making can be more economically
used for agricultural purposes especially that the area gets fairly good rainfall and the soil is
also relatively fertile. If used for agricultural purposes the area will be used sustainably
without degrading the environment and enhancing the extreme hydro climatic events. The area
can accommodate the construction of a dam without much population displacement and would
serve twin purpose of water conservation and flood control.
The dam can sustain agricultural activities through irrigation. It would also check soil erosion
which is a major problem in the area and also ensure continued water supply for other
activities such as use in homes and the institutions in the area. The dam will ensure
environmental stability against landslides that are becoming frequent in the area. Thirdly,
brick burning requires a lot of timber, therefore massive timber cutting has taken place thus
promoting high deforestation and flooding especially in the lower reaches.
The area is massively deforested to the extent that timber for the burning of bricks is being
imported from Kericho and Nandi districts. The deforestation has enhanced soil erosion,
intensified floods and also promoted landslides. Brick burning requires a lot of timber for
burning (Fig. 16). Brick makers have gone to the extent of digging out roots and tree stumps
for the burning of bricks. It is the roots of these trees which hold the soil together, as such in
their absence; the soils especially on the hills and beneath rocks give way and flow
91
downwards,which is now a common occurrence in the area. To safeguard the environment,
alternativebrick- burning methods other than the use of timber fuel to be adopted, especially
the energy saving types like Hoffman's kiln or electric kilns which would take advantage of
rural electrification scheme which fortunately is in the area. Again this area being in the
vicinity of the Sondu - Miriu hydroelectric project, special arrangements can be made to
providethis facility at a much easier cost. This alternative energy sources would push the cost
of bricks slightly higher, but will create some order in the brick making industry and also
enable the enforcement of environmental policy which should be strengthened to include
compulsoryreinstatement of the degraded areas.
Plate 14: Brick making- soil degradation in Nyando basin 201512007
92
Plate 15: Brick - making 30/5/2007
16: Burnt bricks, UpperNyando basin 16/6/2007
Reinstatement should include backfilling the pits and tree planting for environmental
protection and sustainability.
93
4.12 Deteriorating Infrastructure
The embankments / dykes show signs of neglect. They have been breached and overtopped at
various points that even floods of smaller magnitudes than the design floods will overtop
them; even the concreted portion does not seem to have been maintained for quite sometime
(plate 17). The walls which were constructed were done using shorter flood return periods so
are many times being overtopped, and because of lack of maintenance have either been
washed away or lost height due to settlement. Cattle drink water from different points along
the river further damaging the embankments. Widening footpaths, animal tracks and roads has
increased the size of paved and non vegetated areas increasing the area of potentially high
runoff producing land use type.
The dense network of roads and footpaths comprise about 1 % of the basin area and has low
infiltration capacity and contributes between 20 and 50 % of the basin sediment yield. Various
parts of the embankments are overgrown with trees and shrubs resulting in loosening and
cracking of the dykes (Plate 17). This shows lack of routine maintenance of the embankments.
Excessive encroachment of the flood plains and river reserves for agricultural activities,
settlement and livestock grazing increases vulnerability to flooding and subsequently
intensifies floods. A number of culverts are clogged by sediments that they no longer
discharge flow as would be the case. Other obstructions include trees which are dislodged by
floods through bank erosion and get trapped at various points in the channel intensifying
floods which would otherwise have been contained by the channel (plate 18). When these.,
obstacles are not removed they raise the level of flow thus causing floods (plate 19).
94
Plate 17: Dykes just downstream of Ahero bridge 22/4/2006
Plate 18: Trees trapped under Ahero bridge 10/1/2006
95
Plate: 19Flooded Nyando River at Ahero Bridge 201412006
4.13 Topography
The slope of any catchment decides the relative importance of infiltration, interflow and
overland flow. From the topographical maps 1 and 2 (Fig.22a and 22b) the land steeply slopes
from Londiani, Tinderet escarpments and the Nandi hills towards the Lake shores. The
highest portion of the Nyando basin is towards the East at Tinderet hills which is about 2400m
above the sea level. From here the land slopes in a South- westerly direction to about 1250m
above the sea level at Ahero station IGD03. This is a drop of 1150m in a distance which is
less than 150 km. To the South lies the Mau complex. The Digital Elevation Map (Fig.23)
shows the landscape and the drainage pattern in the basin. From table 18, it can be seen that
the Nyando basin has the greatest slope of 5.0 % of all the sub- basins of the Lake Victoria
basin. It has also the highest average sediment transport capacity index of 0.3 in the basin
(ICRAF, 2004). This implies that the velocity of flow builds up to very high destructive levels,
infiltration capacity is greatly reduced and very low flows are experienced shortly after the
rains because of non sustenance of the base flow. It also means that erosion is high. This is
96
made worse by the type of soils, population pressure and the rapid changing land use
activities. High sediment rates means reduced channel capacity and enhanced flooding.
10 0~~~~1~0 iiiiiiiiiiiiiiiiiiiii)20 Kilometers= iD
(!)!/)j)
/~-~l
J~~~f,~i:,~
r-:-»~ ..¢ Altitude (metres)--& r' (!) 1250 -1450
~ I N1451-1700-~....... .A.. N 1701 -1950
WyE N~~~:~~
(!) Rainfall StatioosD Nyando BasinN Rivers Netv.ork
Fig.22a Nyando Basin Topographical map 1
97
NA
10 o 10 20 Kilometers- -- -
o Besin BoundalY
Elevations in metres aslD 1100 -1300D 1300 -1500
D 1500 -1100D 1700 -1900
1900 - 2100
_2100 -2300
_2300 -2500
_2500 -2700
_2700 -2900
Fig. 22 b: Nyando Basin Topographical map 2
98
Nt
Fig. 23: Nyando Basin-Digital Elevation Map
99
"able 18: Demographic and Biophysical characteristics of the Lake Victoria Basin
'CRAF, 2004
.at terrain of the Kano plains does not allow easy drainage of water into the waterways.
ainwater stagnates for longer periods, in some instances even longer than a week. Infiltration
further impeded by the black cotton soils (vertisols) which make up almost the entire plains.
100
I
In the Nyando catchment most livestock were moved to the lakeside swamps for pasture and
water where some of them were bitten by snakes and died and others contacted diseases.
Livestock were greatly affected during this period. Drought related deaths to livestock
increased. The livestock which survived grew so thin that their market value fell so much.
The effects of droughts are usually enhanced by the poor maintenance or the non maintenance
of existing water infrastructure. Most of the water holding or conservation structures which
were swept away by the 1997/98 floods have not been rehabilitated to date in what we are
made to believe as non availability of funds. This implies that most of the water they were
meant to conserve goes downstream leading to immediate onset of droughts just after the
floods. The remaining water retaining structures not washed away by the floods got silted up
that their retaining capacities were tremendously reduced. This is an almost cyclic event in the
region with the magnitude differing from one event to the other. Good catchments
management influences the quality and reliability of water resources.
Forests experience increased deforestation during droughts due to increased charcoal burning,
forest clearing for agricultural expansion, increased logging, frequent fires, intensified
overgrazing and increased incidences of forest diseases. More than 50 % of the flooding in the
lower catchment is attributable to the deforestation of the headwaters (Mogaka et al., 2004).
Water Quality during drought or low flow is low, the concentrations of pollutants in the flows
increase thus increasing the impurity of the water and ma~ing it riskier to human, animal and
aquatic lives. The Nyando basin especially in the upper and the middle catchment there is a lot
of industrial activity which discharge their effiuents into to the river. Industrial effiuents are
mainly organic with high BOD loads. For example, sugar milling consumes large quantities of
water and the discharge has high organic load with a BOD range of 3,000 to 5,000 mg/I
101
(Hansen, 2000). The quality of the final effluent IS sub-standard and thus pollutes the
receiving river systems depressing the oxygen levels.
Nyando catchment is increasingly being deforested and degraded. Degraded catchments
enhance soil erosion and sediment transport which result in loss of top soil, nutrients,
decreased land productivity for agriculture and other uses. Massive deforestation in the
catchment is causing landslides in some parts of the basin and the neighbouring basins.
Landslides have caused human, livestock lives in the basin and much economic devastation in
terms of buried crops, homes, destroyed infrastructure and general environmental degradation.
Runoff rates increase and infiltration rates reduce implying reduction in base flow and
extended periods of droughts. These problems are experienced in the Nyando catchment
consequently reducing the economic lives of water storage facilities and the reliability of
water supply. The implication would be increased water storage requirements and higher cost
of water treatment
102
CHAPTER FIVE: SUMMARY OF FlNDINGS, CONCLUSSIONS AND
RECOMMENDATIONS
5.1 Summary and Conclusions
'Nyando a sub-basin of the Lake Victoria basin suffers from flooding almost on an annual
basis. This study investigated the hydro-climatic variables of the area to understand their
characteristics and recommend prevention, mitigation and management measures to be put in
place. The objectives were:
a) To determine the characteristics of the extreme hydro-climatic variables of the Nyando
nver.
b) To establish the effects of land cover and Land use on the extreme flows in the Nyando
nver.
The objectives were achieved by analysing rainfall in terms of its reliability and variability in
the Nyando basin at Ahero, Kipkelion, Muhoroni and Kaisugu stations. Discharge in the river
Nyando was analysed using Gumbel, Log Pearson 3 and General Extreme Value fitting
distributions. The effects of land cover and land use were analysed using G1S- IDRIS1,
ARCG1S and ARCVIEW programmes.
The study established that other than rainfall, other factors which enhance flooding in the
catchment include deforestation, population pressure due to rapid population growth, and
various land use activities such as intensive agricultural activities, charcoal burning, livestock
keeping and settlements and topography. The steep topography of the upper and the middle
zones with reduced land cover promote rapid runoff and soil erosion. In the lower catchment
the situation is worsened by the gradient which becomes gentle and the nature of vertisols
soils which impede drainage, thus allowing water to stagnate and build up over a long period.
103
It was established from the rainfall-runoff ratio and from flood frequency analysis that floods
are on an increasing trend in the Nyando river basin. The implication is that most of the
rainfall in the basin translates into surface run-off.
High population growth rate has drastically changed the environment: A lot of urban centres
have come up and the old ones grown bigger implying more paved areas in terms of roofs and
pavements. This translates into less infiltration and enhanced runoff even with small rainfall.
The design discharge with a return period of 25 years adopted by the Ministry of Water and
Irrigation, for current structural measures is equivalent to that less than 10 years, and with the
continued environmental degradation especially in the upper catchment, the return periods are
becoming even shorter. There is therefore need to redesign structural flood mitigation and
control system.
Unclear access to property rights has resulted in land use conflicts. There is massrve
encroachment on riparian and fragile ecosystems such as wetlands and river banks both for
agriculture and human settlements. Policies and institutional mechanisms for the management
and conservation of wetlands are weak. Wetlands therefore suffer the problems of
encroachment and lack of all inclusive management plans; this intensifies the increase in
sedimentation and degradation in them. Settlement, grazing and agricultural activities have
made them lose their filtering and retaining capacities thus allowing all the water to flow
downstream implying early drought. The loss of this function has seen the lake recede by over
one kilometre due to deposition and backward flow thus inundating a larger area during
floods. The declining vegetation resources and soil productivity threatens the future
livelihoods of the communities and further degrade water and nutrient cyclic functions of the
area. People around the lake are now busy cultivating the strip left by the receding water
104
because of its high fertility, implying that the lake is receiving the fertile top soil from the
basin. The remaining soil continues to become more and more barren and unproductive
implying hunger and poverty to the residents.
The bulk of the stream flow is produced in the upper and middle catchments of the Nyando
river basin. The lands capacity to hold rain water decreases due to encroachments on forest
catchments, cultivation on steep slopes and prevalence of shallow soil areas. Activities which
affect the quantity, timing and quality of water emerging from the basin such as excision of
forests, poor land use practices and encroachment into recharge areas have adversely affected
the performance of major sectors of the regional economy.
Overstocking of livestock, which has forced the owners to allow them to freely graze, has
resulted in overgrazing and the animals dislodging and loosening the soils mainly in the
middle and lower catchments resulting in sheet erosion and flooding. Land management
interventions are constrained by freely grazing livestock in the basin especially in the middle
and lower catchments. Livestock affect trees, conservation structures, vegetation on the hill
slopes and the destruction of the river bank areas.
None maintenance of the drainage of the system: weeds and trees have blocked and reduced
some parts of the river along the flow path that the river has to overflow. Some parts of the
dykes have either been washed away or have settled with time in such a way that it can not
contain even a minor flood. Dykes which were designed and partly constructed were meant for
lower discharges than being experienced now because of the unanticipated magnitudes then.
105
5.2 Recommendations
Due to watershed degradation the Nyando basin has lost hydrological balance and to restore
this balance the study recommends immediate action on the factors which enhance runoff such
as:
Encroachment on the forest catchment in the upper zone, this for whatever activity in areas
including Nandi Hills, Tinderet, Timboroa and Londiani, should be halted. A buffer zone
should be created to protect and reduce further encroachment. Buffer strips give the lands
good anchorage against erosive action which could build up due to deforestation and steep
slopes.
Cultivation on the steep slopes of the middle catchment should be stopped and land use
controlled. Any area with an excess slope of 30 % should be reforested to reduce rapid runoff
and high soil erosion. High rate of soil erosion experienced in the middle catchment translates
into huge sediment deposits at the beginning of the lower catchment thus promoting flooding.
Reforestation, adoption of modern farming technology and use of soil conservation measures
would help in restoring the lands capacity to hold rain water by enhancing infiltration and thus
reduce flooding.
The lands capacity to hold rain water should be restored so as to maximize infiltration in line
with the principle of holding rain water where it falls. According to the Agriculture Land act,
all land with slopes in excess of 30 % should be reserved for conservation as either forests or
bush lands. Failure to observe these requirements is seen in high rates of soil erosion in the
Ainapngetuny area of the upper catchment and frequent land slides in the Tinderet area. On
already settled areas, increase in woody vegetation is recommended. This will shield the land
from direct rain action while anchoring the land against shearing and incidences of mass
slides. Use of soil conservation measures and modem farming methods should be used to
106
IfJ:AlVATTA nL\u\lW:R~ITV LIBRARY
facilitate seepage without risking mass slides. Runoff detention measures should be adopted
especially in shallow soil areas which have a limited capacity to hold rain water. The
watershed suffers peak floods followed by low flow during the dry season due to the absence
of base flow recharge. Installation of small reservoirs across the rivers will be crucial in
stabilizing stream flow for the purposes of reducing flood severity downstream while storing
runoff for dry season water supply. Reclamation of wetlands for agricultural purposes should
be discouraged because in the long run they prove unviable and counterproductive.
The study recommends proper catchment management, which encompasses better land
management practices, environmental improvement, conservation and protection to retard
surface runoff, afforestation programme to be stepped up with emphasis being laid on
environmentally friendly species. Develop a land-use system that combines agriculture,
livestock and afforestation on the same unit of land on a sustainable-yield basis. This will
reduce the pressure on vegetation cover on fragile lands, which comprise the upper catchments
which are the headwaters, the escarpments and the plains. The vegetation cover will improve
the infiltration rates of surface runoff by reducing the surface velocity and delaying the
concentration time and also help in reducing soil erosion by holding together the highly fragile
soils of the upper and the middle catchments. Cutting of trees should be regulated to ensure
more trees are planted than are cut.
The removal of vegetation cover has led to loss of soil through massive erosion, land slides
and mud flows, loss of water for sustenance of river flows during droughts and eventually loss
of lives and livelihood. Cheaper alternative forms of energy should be adopted such as hydro-
based energy sources. Control land use patterns especially in the upper catchment to reduce
soil erosion. The study recommends the enforcement of the correct soil conservation practices
107
in areas adjacent to rivers and lakes and the intensification of forestation programmes in the
degraded catchment areas up to shores of the Lake.
Response to the extreme events requires a forecast with a great degree of accuracy and hence
the need to adopt practice interventions rather than reactive responses to reduce their effects.
Each season of an extreme event families are provided with relief food, drugs, temporary
shelter and blankets depending on the event for short term. Long term measures should be put
in place, but the magnitude of preparedness to that level requires that extreme hydro climatic
characteristics are well understood.
The study recommends the redesign of structural flood mitigation and control system to
contain the more frequent and greater discharge floods being experienced in the recent times.
As a short time measure, the washed away portions of the existing dykes should be
strengthened and maintained; desiltation to be undertaken especially during low flow periods
when it easier to do it. Runoff detention measures such as small reservoirs to detain water
during floods and recharge the river during low flows should be constructed at predetermined
points.
Encroachment on the wetlands should be halted and those already occupying them to be taught
on how to live with the floods, but restrained from intensive use of the same. This will slowly
restore the filtering functions of the wetlands and enhance their water retention capacities.
Encroachment into the forests should be prohibited for habitation or farming to maintain the
vegetal cover for better infiltration and base flow sustenance. The study proposes the reduction
of these conflicts through developing an integrated management plan for sustainable multiple
use of wetlands.
108
Proper land use is critical to surface water management. The promotion of surface vegetation
cover enhances higher infiltration rates by. adding organic matter to the soil and root action
which penetrates the soil and creates micro pores when roots decay. Loss of rainfall through
runoff adversely affects crop and vegetation growth by reducing soil available moisture
therefore, improved retention of surface .water enhances water availability, improved food
production and better pasture productivity.
Improper land use practices and land degradation have enhanced soil erosion which has
resulted in the formation of gullies, some of which are deep and wide to the extent of
becoming seasonal rivers. There are numerous gullies in the basin. There is therefore need to
adopt better land use practices and discard poor land management and outdated tilling
practices. There is also need to rehabilitate the numerous gullies in the basin or stop them from
becoming bigger. The measures include vegetation planting, construction of check dams,
boulder bunds, brick masonry, earth dams and sand bag plugs. Implement upstream
engineering measures for soil and water conservation for flood protection such as dams. The
study recommends proper environmental impact assessment for any development as a safety
measure against further environmental degradation in the basin.
Develop infrastructure design parameters and regulations to ensure that the structures can
sustain flooding for the designed return periods, which is not the case for Nyando catchment
as at now. Water should be treated as a strategic resource with vital economic and social
values. The characteristics of the existing Nyando river channels are that the Nyando river
basin is characterised by mountainous areas, a large amount of intense rainfall, densely
populated and a gentle slope in the lower catchment (Table 29).
109
Introduce dams and increase storage capacities to serve twin purposes of flood routing and
storage reserves 'for use during deficit times. Water Resources management to be an all
inclusive programme; all stake holders should be made to understand and play their roles in
the management of the resource.
Data recording and information management and dissemination systems, particularly of the
extreme events should be enhanced to enable adequate availability of information for use in
responding to impending disasters and in the design of the flood mitigation and protection
works.
The study recommends the development of a flood hazard map to show the most vulnerable
areas to be given more special attention in term of preparedness and response during
emergencies. The study recommends inter-basin water transfer as an option in dealing with the
excess destructive floods to water deficient areas.
Resettlement of those living in flood prone areas has never been and will never be a
permanent solution to the flood problem. This is because people who were relocated from the
flood prone areas of Budalangi and the Nyando, and given alternative lands elsewhere came
back either due to the cultural ties to their ancestral land or due to some economic advantages
the flood plains offer. A better solution would be to train the people on how to live with the
extreme events.
A much more effective communication mechanism among the related institutions in cases of
disaster occurrence should be enhanced. District Development Committees should include
disaster management in their strategies and development priorities in their district
development plans.
110
The study recommends an integrated watershed management for the Nyando basin, further
research on the water quality during the extreme events and the effect of climate change on the
extreme flows. Some of the proposed remedial measures are on table 19.
Table19· Prooosed Remedial measures
Watershed Present condition Proposed remedial measures
UPPER Increase of soil erosion due to
deterioration of forest cover
Proper land use, reforestation
MIDDLE Increase in bank erosion due to
steep slope and deforestation and
large sugar farms
Drainage improvement, use of modern
agricultural techniques, construction of
control structures and proper
maintenance of the same, Retarding /
Regulating ponds
Cut-off channel
Flood control dam
LOWER Flood damage, inundation, silt
deposition
Structural measures:
Dyke/ Embankment,
Desiltation of River channel,
Drainage Channel
Diversion Channel (Floodway)
Bank protection works
Culvert improvement
Raising the road surface
Flood control dam
Flood proofing etc.
111Ifl: '\lJi......... • IIlU '~nl'ITV I Ii
REFERENCES
Agenda 21, (1996) The earths summit agenda for change Rio De Janeiro 1996.
Aseto and Ong'ang'a, (2003) Lake Victoria (Kenya) and its Environs: Resources,
Opportunities and Challenges, published by Osienala, Kisumu.
Atieno, L. 0., (2004) Assessing the influence of geology and impacts of land use and
climate change on the river Nyando Basin. Unpublished MSc. Thesis, Nairobi University.
Bullock, A., Keya, S.O., Muthuri, F. M., Bailley-Watts,T., WilIiams,R.and Waughray,D.
(1995) Lake Victoria Environmental Management Programme Task Force 1I. Final report by
the regional consultants on Tasks 11, 16 and 17 (Water quality, land use and wetlands).
Wallingford, UK: Centre for Ecology and Hydrology (CEH) and FAO.
Central Bureau of Statistics, (2005)
Daily Nation, (2005) Water conflict in Mai Mahiu, published by Nation Media group 24th
January, Nairobi, Kenya pp 3.
Hansen, T. S., (2000) Evaluating the Environmental Impact of Participatory Soil and Water
Conservation at the Watershed level in the Nyando Watershed.Western Kenya. Paper
presented at the UNESCO-IUFRO Symposium on 'Forest, Water and People' In the Humid
Tropics; Past, Present and Future Hydrological Research. University of Kabangsaan,
Malaysia, 30th July to 4thAugust 2000.
Hirji, R. Davies, (2005) Water Quality: Assessment and Protection; Water Resources and
Environment. TechnicalNote D.1, World Bank, Washington, D.C.
ICRAF, (2002) Improved Land Management in the Lake Victoria Basin. Linking Land and .
Lake, Research and Extension, and Catchment and Lake Basin.
ICRAF, (2000) Improved Land Management in the Lake Victoria Basin. Linking Land and
Lake, Research and Extension, and Catchment and Lake Basin. Final Report, Start-up Phase,
July 1999 to June 2000: working paper 2000-2.
Italconsult (1983) Pre-Investment Study for Water Management and Development of the
Nyando and Nzoia River Basins. Ministry of Water Development. Nyando River Basin Pre-
development report, October, 1983.
Italconsult (1980) Pre-investment Study for Water Management and Development of the
Nyando and Nzoia River Basins. Ministry of Water Development, 1980.
Italconsult (1978) Pre- investment Study for Water Management and Development of the
Nyando and Nzoia River Basins. Ministry of Water Development, Technical proposal. June
1978.
Japan International Cooperation Agency (1992a) The Study on the National Water
Master Plan. Ministry of Water Development. Main Report Volume 11. Master Action Plan
towards 2000. PartI. July1992.
112
Kipkore K. W., (1996) Watershed Degradation and Soil and Water Conservation in Kenya.
Seminar on Integrated Water Resources Management in Kenya. Nanyuki, January 22-25,
1996
KSS, (2003) The soils of the Nyando catchment Area. Kenya Soil Survey, Report N021.
LBDA, (1985) Lake Basin River Catchment Development, River Profile studies, Basin
Water Balances and Inter-Basin Transfer studies vol. III, United Nations TCD Contract, CON
5/83. Annex A: Regional Hydrology prepared by WLPU consultants, October, 1985.
LBDA, (2003) Project proposal for the Nyando Flood Protection Project July, 2003.Lake
Victoria Environmental Programme (World Bank)
LVEMP,( 2000) Lake Victoria Environmental Programme, World Bank, 2000.
Lotti C. and Associates (1985) Lake Basin River catchment Development studies. River
Profile Studies, Volume l. General Summary. October 1985.
Matiasi, T. M., (2003) Modelling and comparative Analysis of low flow in Lake Victoria
Basin in Kenya, Department of Water Resources and Environmental Management, Kenya
Water Institute, Nairobi.
Mogaka, H., Gichere, S., Davis, R. and Hirji, R., (2002) Impacts and costs of climate
variability, and Water Resources Degradation in Kenya. An interim Report of the World Bank,
Nairobi. pp23-28.
Mungai D., Swallow B., Mburu J., Onyango L. and Njui A., (2002) Proceedings of a
Workshop on Reversing Environmental and Agricultural Decline in the Nyando River Basin.
December 9-11,2002. Kisumu, Kenya.
Muthusi, F. M., (2004) Evaluation of the USGSS Stream flow Model for Flood Simulation
(Nyando BasinCase Study) MSc. Thesis, Jommo Kenyatta University of Agriculture and
Technology,2004.
Mwongera, E. K., (1996) Challenges to Water Resources Management in Kenya. Paper
presented to a seminar on Integrated Water Resources Management in Kenya by Economic
Development Institute of the World Bank -Nanyuki Kenya, January 22nd_25th, 1996.
MWRMD, (2004b) A Project Proposal for Management of Flood Control in Kano Plains. An
integrated Approach. August2004.
MWRD, (2004a) Flood Control on Lower Reaches of Ny ando River. Pre-investment Study.
Final Report. May 2004.
Njogu, A. K., (2000) An Integrated River Basin Planning Approach - Nyando case study in
Kenya. MSc. Thesis, mE Delft, the Netherlands.
Nyaga, E., (1996) Water Quality and Land Use Management in the Lake Victoria Basin.
MSc. Thesis, Nairobi University, 1996.
Nyanch'aga, E., (2004) Flood and Catchment Management-Nile Basin Capacity Building
Network for River Engineering.
113
OCHA, (2002) Office for the Coordination of Humanitarian Affairs of the UN, Situation
Report No.1, Kenya -Floods.
Ocharo, R. M. and Omondi P., (2004) Floods Disaster in Western Kenya, Vulnerability and
Capacity Assessment Report. May 2004.
Odada, J. E. and Otieno, J. 0., (1990) Socio-Economic Profiles: Ministry of Planning and
National Development and UNICEF, published by the Ministry of Planning and National
Development, Republic of Kenya P.O. Box 30005 Nairobi, Kenya and The United Nations
Children Fund. Kenya country office P.O. Box 474145 Nairobi Kenya, printed by Reata
printers Ltd, P.O. Box 51247 Nairobi, June, 1980.
Ogallo, L. J., (1981) Reliability of Rainfall in East Africa, Institute for Meteorological
Training & Research. Republic of Kenya Meteorological Department report 5/81,Nairobi-
Kenya.
Ogolla, J. S., Abira, M. A., and Awuor, V. 0., (1997) Potential Impacts of Climate Change
in Kenya (IPCC). Published by Climate Network Africa, P.O.Box 76479 Nairobi, printed by
Motif Creative Arts Ltd. Nairobi-Kenya
Ong, C. and Orego, F., (2001) Link Between Land Management, Sedimentation, Nutrient
Flows and Smallholder Irrigation in the Lake Victoria Basin. Paper prepared for special issue
on the 'Changing Face of Irrigation in the Lake Victoria Basin' by the International Water
management Institute 20th March, 2001.
Onyango L., Swallow B. and Meinzen-Dick R., (2005) Hydroponics and Terranomics in the
Nyando basin of Western Kenya. International Workshop on African Laws: Plural Legislative
Frameworks for Rural Water Management in Africa, 26-28 January 200~, Gauteng, S. Africa.
Republic of Kenya, (2004) Statistical Abstract: Central Bureau of Statistics. 2004.
Rothschild, E. D. and Skills, S., (1992) The Environment and Kenya. A fact Finding report
on Environment and Kenya: preliminary investigation into a number of the Environmental
Pollution problems in Kenya, by Edmund De Rothschild and Stephen Skill. January, 27th,
1992 to February 5th 1992.
Republic of Kenya, (1979, 1989 and 1999) Ministry of Planning and National Development,
Population and Housing Census.
Samez Consultants, (2005) The Baseline Survey on the Proposed Project for Formulation of
an Integrated Watershed Management Plan for the Nyando River Basin in the Republic of
Kenya- Main Report Volume 1.
Wanjohi, D. M., (1999) Water Flow and Quality Modelling of the Nyando River in Kenya,
M.Sc. Thesis-International Institute for Infrastructural, Hydraulic and Environmental
Engineering.
Waruru B.K., Njoroge C. R. K. and Wanjogu S. N., (2003) Biophysical Baseline
Information for the Nyando Catchment Area. The Soils of the Nyando Catchment Area.
Reconnaissance Soil Survey. Report NO.2!. October, 2003.
114
..Waruru B. K., (1992) Soil physical and chemical characteristics of highly erodable soils in
Nyakach area (Kisumu district), Kenya. MSc. Thesis. University of Ghent.
WMO, (2004) World Meteorological Organisation, Strategy for Flood Management for Lake
Victoria Basin, 2004.
115
APPENDICES
Appendix 1: Kaisugu Rainfall Station
Year Annual M.A.M- Long O.N.D - Short
Rains Rains
1970 2257.4 675.7 329.1
1971 1823.4 598.4 256.1
1972 1643.3 407.7 418.0
1973 1576.2 305.4 188.6
1974 1840.0 896.5 198.9
1975 1878.3 676.5 248.4
1976 l352.7 414.5 184.1
1977 2040.6 600.6 453.6 .
1978 2171.3 643.9 378.7
1979 1668.9 487.7 204.8
1980 1436.1 663.6 194.3
1981 1905.7 888.0 167.2
1982 1855.0 763.8 622.3
1983 2093.7 660.9 450.8
1984 1468.1 383.9 369.4
1985 1692.3 776.7 183.2
1986 1432.8 568.2 261.7
1987 1942.3 705.8 427.1
1988 2112.5 851.2 248.7
1989 1945.3 650.2 434.7
1990 1614.7 627.3 300.9
1991 1576.7 527.9 193.0
1992 1733.1 525.2 338.1
1993 1720.9 680.0 164.2
1994 2059.5 883.1 387.2
1995 1749.1 564.1 349.3
1996 1784.9 577.3 263.4
1997 1901.2 507.5 821.0
1998 2101.7 658.7 266.2
1999 1666.7 529.6 423.1
2000 1546.6 440.8 453.7
Mean 1793.3 617.4 325.2
STD 236.7 150.4 148.2
116
Appendix 2: Ahero Rainfall Station
YEAR Annual MAM OND
1970 11282.1
1444.4 I 180.21
1971 1040.0 422.2 248.3
1972 1108.3 375.3 326.6
1973 1125.3 330.7 179.2
1974 1202.2 626.9 220.5
1975 1218.1 392.7 278.3
1976 1060.8 285.1 187.7
1977 1256.0 468.7 306.5
1978 1436.7 498.5 309.2
1979 1319.9 479.3 264.0
1980 974.0 378.2 211.8
1981 1086.2 544.6 111.6
1982 1326.5 380.3 284.5
1983 1029.3 383.0 250.1
1984 1010.0 304.6 340.9
1985 1148.2 594.2 174.4
1986 1309.5 546.2 359.8
1987 1180.6 491.5 256.6
1988 1310.4 620.9 161.4
1989 1428.6 582.5 342.3
1990 1187.8 473.6 180.4
1991 1032.0 399.7 275.7
1992 1012.7 358.8 174.0
1993 1050.8 305.0 227.0
1994 1236.9 627.6 345.0
1995 1131.3 457.7 292.0
1996 1427.6 420.0 340.3
1997 1616.9 416.2 887.3
1998 1445.7 519.0 208.0
1999 1277.0 399.6 344.5
2000 1126.1 455.7 322.7
MEAN 1206.4 451.1 277.1
STD 158.3 97.0 132.0
117
Appendix 3: Kipkelion rainfall Station
Year Annual MAM OND
1975 1177.4 413.3 154.5
1976 676.4 145.1 138.2
1977 1174.0 344.4 304.0
1978 836.6 174.6 40.2
1979 1041.4 370.8 156.2
1980 1123.3 417.1 132.7
1981 1021.9 310.7 175.8
1513.3 457.3 422.0
1978 1236.8 314.5 120.2
1979 969.2 118.4 104.8
1980 875.7 328.3 148.5
1981 1075.6 585.3 64.8
1982 1157.6 392.3 282.6
1983 1057.3 317.8 248.7
1984 784.1 201.9 202.7
1985 1210.3 582.4 108.1
1986 1035.0 421.8 126.7
1987 1165.7 608.5 220.4
1988 1357.6 534.9 117.9
1989 1099.5 408.7 272.5
1990 1163.9 503.9 138.7
1991 1318.5 502.2 158.1
1992 940.5 355.9 149.6
1993 1201.7 463.2 271.1
1994 1135.1 399.5 190.3
1995 1310.3 535.0 187.3
1996 955.8 272.3 129.6
1997 1030.7 312.5 376.9
1998 1100.6 333.9 123.8
1999 887.2 145.5 261.6
2000 947.9 216.2 243.1
MEAN 1083.3 370.6 186.2
STD 176.9 134.7 86.7
118
Appendix 4: Muhoroni Rainfall Station
Year Annual MAM OND
1982 1520.3 350.7 545.0
1983 1494.1 458.8 363.1
1984 1087.0 325.1 234.2
1985 1272.3 537.2 263.1
1986 1154.6 552.5 201.0
19787 1392.2 556.8 288.8
1988 2455.1 666.2 343.1
1989 2825.1 1482.4 373.2
1990 1721.2 635.6 169.3
1991 1625.9 522.3 335.0
1992 1553.8 474.2 278.8
1993 1759.7 618.6 220.5
1994 1340.0 744.9 204.2
1995 1779.4 554.7 465.7
1997 2145.0 470.2 419.0
1998 2063.8 471.7 1144.5
1999 1471.3 456.3 281.4
2000 1500.4 546.9 420.2
2001 1089.8 338.4 398.7
2002 1824.0 644.0 270.0
2003 1441.6 559.9 341.0
2001 1066.7 733.5 145.8
1992 670.6 165.5 218.1
MEAN 1576.3 559.4 344.5
STD 477.4 242.4 200.9
Appendix5: Flood Return periods from earlier studies
SOURCE RETURN PERIOD, T YEARS
10 25 50 100
1 ITALCOSULT 550 650 750 850
2 LOTTIlWLPU 345 437 506 575
3 JICA 297 362 411 459
4 LBDA(Mbu~ua) 275 344 395 444
5 IRRIG-MoWD 375 476 551 625
6 MoWD-2003 863.19 1044.22 1178.46 1310.20
7 Master Plan 1992 437 654 659 752
Source: MoWD 2004
119
Appendix 6: Comparison of peak discharges in m3/sec
Return This MoWn Master Italconsult Variance Percentage
Periods Study2006 2003 Plan1992 1983 From Variance
(Yrs) Italconsult
10 926.72 863.19 437 550 376.72 68.5
25 1143.75 1044.22 654 650 493.75 75.96
50 1304.75 1178.46 659 750 554.75 74
100 1464.56 1310.20 752 850 614.56 72.3
Appendix 7: Nyando at 1GD03-Maximum Annual Discharges
Year Gauge Q m3/sec
Height(m) Irrigation Hydro met Italconsult LOTTIMoWn Section
1969 2.80 59.461 66.00 59.777 63.59
1970 4.30 182.840 210.00 178.812 170.721
1971 3.50 111.753 120.00 111.703 110.802
1972 4.60 214.843 252.50 268.513 196.695
1973 4.00 153.799 173.00 288.840 146.666
1974 5.45 322.284 387.33 291.725 280.825
1975 4.30 182.840 210.00 202.480 170.721
1976 2.80 71.375 66.00 95.22 63.48
1977 6.14 428.608 550.80 350.117 360.709
1978 4.45 198.467 320.75 248.944 183.467
1979 5.85 381.772 478.00 428.892 325.859
1980 4.20 172.835 197.00 - 162.490
1981 3.80 136.043 150.00 - 137.689
1982 4.55 209.300 245.00 - 192.233
1983 3.40 104.268 111.50 - 104.258
1984 2.00 37.406 31.00 - -
1985 5.17 284.427 338.00 - -
120
Appendix8: Silt laden River Nyando in the lower catchment 22/5/2006
Appendix 9 : Factors that promote flood disasters
Factors Reason
Population pressure Intensive economic use of the flood plains for
agriculture and livestock farming.
Deteriorating infrastructure Lack of systematic and routine maintenance of flood
dykes. This makes them. susce\\tihle to h\:eaches even.
during floods of lower magnitudes than the design
floods.
Environmental degradation of This is caused by uncontrolled and unregulated human
the watershed activity ,especially large-scale deforestation and
cultivation practices as witnessed in the upper catchment
Floods dislodge trees at the side of the channels and
Obstructions and side wash them downstream in the course of which they get
growths along the channels. trapped and obstruct the flow thus enhancing overflow
which in turn cause destruction especially in the lower
reaches. Plate 11 shows trees trapped under Ahero
bridge.
KENYAITA UNiVERSITY LIBRARY
121
Appendix 10: Annual Rainfall and Discharge valuesfor the Nyando River
Year Annual Rainfall 1GD03 Runoff
mm mm
1969 1024.4 45.32
1970 1097.1 237.67
1971 1014.8 180.51
1972 1278.6 1438.8
1973 1301.9 152.64
1974 1044.7 182.5
1975 1297.8 238.6
1976 1393.2 97.6
1977 1314 325.5
1978 908.9 320.1
1979 1153.11 237.1
1980 1304.0 160.7
1981 1022.8 127.6
1982 986.8 201.9
1983 1200.4 63.5
1984 1321.1 170.9
1985 1206.5 75.5
1986 1310.4 92.1
1987 1415.5 379.3
1988 1199.3 246.9
1989 1032.0 287.9
1990 1012.7 148.9
1991 1950.8 183.8
1992 1236.9 95.4
1993 1122.6 125.5
1994 1427.6 290.77
1995 1616.6 149.9
1996 1445.7 330.3
122
KENYArrA UNIVERSITY LIBRARY